Annual Meeting-1
119th TMSANNUAL MEETING TECHNICAL SESSIONS
Anaheim, California • February 18 - 22, 1990 The numbers in bold type on the right of each session title indicate the page on which that session is listed in the Annual Meeting Technical Program. Sessions will be in the following: I - Inn at the Park M - Marriott
SESSION TITLE
Q - The Quality Hotel
DAY
Advanced Aerospace Materials from User & Supplier Perspectives ....................................................................................... Tues. AM Advanced Metal & Ceramics Matrix Composites I: Processing Ceramic Matrix Composites ........................................... Mon. PM Advanced Metal & Ceramics Matrix Composites II: Processing Metal Matrix Composites ............................................... Tues. AM ~dvanced Metal & Ceramics Matrix Composites III: Fracture and Fatigue ........................................................................ Tues. PM Advanced Metal & Ceramics Matrix Composites IV: Interface and High Temperature Performance ............................... Wed.AM Advanced Metal & Ceramics Matrix Composites V: Process Modeling and Deformation Processing ............................... Wed. PM Advanced Metal & Ceramics Matrix Composites VI: Emerging Technologies and Applications ........................................ Thurs. AM Advanced Metal & Ceramics Matrix Composites VII: Damping, Tribology, and Non-Destructive Characterization ................................................................................. Thurs. PM Aging & Life Extension of Nuclear Reactor Materials I: Pressure Vessels ................................................................................ Thurs. AM ..,Aging & Life Extension of Nuclear Reactor Materials II: Other Components .............................................................................. Thurs. PM Alumina & Bauxite I: Bauxite and Equipment .......................................................................................... Mon. PM Alumina & Bauxite II: Bayer Process ................................................ Tues. AM Alumina & Bauxite III: Simulation, Modeling, Control ............................................................................... Tues. PM Alumina & Bauxite IV: Alumina and Chemicals ............................................................................................ Wed. AM Alumina & Bauxite V: Alternate or Modified Processes .............................................................................. Wed. PM Carbon Technology I ............................................................................ Mon. AM Carbon Technology II .......................................................................... Mon. PM Carbon Technology III ......................................................................... Tues. AM Carbon Technology IV ......................................................................... Tues. PM Carbon Technology V .......................................................................... Wed. AM Carbon Technology VI ......................................................................... Wed. PM Cast Shop Technology I: Recycling Special Session .................................................................................... Mon. AM Cast Shop Technology II: Melting Technology ................................... Mon. PM Cast Shop Technology III: Metal Treatment Fluxing ................................................................................................ Tues. AM Cast Shop Technology IV: Metal TreatmentFiltration and Recycling Al-Li ........................................................... Tues. PM Cast Shop Technology V: Grain Refining & Modification ........................................................................................ Wed. AM
ROOM
LOC.PAGE
Garden Room 1,2,3
I
44
ParkD
I
26
ParkD
I
44
ParkD
I
65
ParkD
I
85
ParkD
I
108
ParkD
I
126
ParkD
I
139
Garden Room 1,2,3
I
128
Garden Room 1,2,3
I
140
Salon 3 Salon 3
M M
27 46
Salon 3
M
67
Salon 3
M
86
Salon 3 Salon 2 Salon 2 Salon 2 Salon 2 Salon 2 Salon 2
M M M M M M M
109 11 28 47 67 87 110
Salon 4&5 Salon 4&5
M M
11 29
Salon 4&5
M
47
Salon 4&5
M
68
Salon 4&5
M
88
2--Annual~eeting
Cast Shop Technology VI: Casting Automation & Control ....................................................................... Wed. PM Cast Shop Technology VII: Casting 2xxx, 6XXX & 7XXX Alloys ............................................................. Thurs. AM Cast Shop Technology VIII: Advances in Casting Technology ........................................................................... Thurs. PM Creep .................................................................................................... Mon. PM Defects & Defect Reduction Processing in Semiconductor Heterostructures I: Dislocations and Dislocation Interactions ....................................... Mon. AM Defects & Defect Reduction Processing in Semiconductor Heterostructures II: Epitaxial Stabilization a-Sn and Sn1."Ge" Alloys ............................ Mon. PM Defects & Defect Reduction Processing in Semiconductor Heterostructures III: Defect Reduction in Misfitted Heterostructures .............................. Tues. AM Deformation & Mechanical Behavior 1............................................... Wed. PM Deformation & Mechanical Behavior II ............................................. Thurs. AM Electronic & Magnetic Materials ........................................................ Wed. PM Environmental Effects ........................................................................ Wed. AM Expert System Applications in Materials Processing & Manufacturing I .......................................................... Tues. AM Expert System Applications in Materials Processing & Manufacturing II ........................................................ Tues. PM Expert System Applications in Materials Processing & Manufacturing III ....................................................... Wed. AM Expert System Applications in Materials Processing & Manufacturing IV ....................................................... Wed. PM Extractive Metallurgy ......................................................................... Tues. AM Fatigue ................................................................................................. Mon. AM Fracture I ............................................................................................. Mon. AM Fracture II ........................................................................................... Mon. PM High Performance Copper-Base Alloys I ............................................ Mon. AM High Performance Copper-Base Alloys II .......................................... Mon. PM High Performance Copper-Base Alloys III ......................................... Tues. AM High Temperature Deformation, Fracture & Application of Intermetallic Compounds I: Intermetallic With Li2-Structure ...................................................... Wed. AM High Temperature Deformation, Fracture & Application of Intermetallic Compounds II: Nickel & Titanium Aluminides .......................................................... Wed. PM High Temperature Deformation, Fracture & Application of Intermetallic Compounds III: Selected High Temperature Intermetallics ...................................... Thurs. AM High Temperature Deformation, Fracture & Application of Intermetallic Compounds IV: Engineering Applications of Alternate Materials ............................................................................................. Thurs. PM High Temperature Materials I: Fracture Mechanics of High Temperature Materials ......................................................... Tues. AM High Temperature Materials II: Modeling of High Temperature Materials ............................................................ Tues. PM High Temperature Superconducting Compounds I: Plenary Session ................................................................................. Tues. AM High Temperature Superconducting Compounds II: Bulk A: Synthesis/Oxygen Dissolution ............................................. Tues. PM High Temperature Superconducting Compounds III: Thin Films A ...................................................................................... Tues. PM High Temperature Superconducting Compounds IV: Bulk B: Melt Texturing/Substitution ............................................... Wed. AM
Salon 4&5
M
111
Salon 4&5
M
128
Salon 4&5 Plaza B
M I
141 30
Orangewood II
Q
12
Orangewood II
Q
31
Orangewood II ParkC ParkC Conf. Theatre Plaza A
Q I I Q I
48 111 129 112 89
ParkB
I
49
ParkB
I
69
ParkB
I
90
ParkB Salon G Plaza B Plaza C Plaza C ParkA ParkA ParkA
I M I I I I I I
114 50 13 15 32 16 33 51
Plaza B
I
91
Plaza B
I
114
Plaza B
I
130
Plaza B
I
141
Plaza B
I
52
Plaza B
I
70
Orangewood I
Q
53
Orangewood I
Q
71
Orangewood II
Q
72
Orangewood I
Q
92
Annual Meeting-3
High Temperature Superconducting Compounds V: Thin Films B: Microstructure ........................................................... Wed. AM High Temperature Superconducting Compounds VI: DeformationlMechanical Properties ................................................. Wed. PM High Temperature Superconducting Compounds VII: WirelI'ape/Filaments ......................................................................... Wed. PM Hume-Rothery Memorial Symposium I: High Tc Superconductors & Related Metallic Oxides (A) ............................................................................ Mon. AM .,Hume-Rothery Memorial Symposium II: High Tc Superconductors & Related Metallic Oxides (B) ............................................................................ Mon. PM Hydrometallurgy ................................................................................. Thurs. AM Hydrometallurgy & Aqueous Processing I: Leaching of Primary and Secondary Materials ............................................................................................ Mon. AM Hydrometallurgy & Aqueous Processing II: Physical Chemistry of Precipitation and Effluent Treatment Processes ........................................................... Mon. PM Hydrometallurgy & Aqueous Processing III: Electrolytic and Separations Processes ............................................ Tues. AM Hydrometallurgy & Aqueous Processing IV: Biotechnological Treatment of Minerals and Effluents ..................................................................................... Tues. PM Interfaces in Metal-Ceramic Composites I: Thermodynamics and Kinetics A ...................................................... Mon. PM Interfaces in Metal-Ceramic Composites II: Theoretical Modeling oflnterface Properties .................................. Tues. AM Interfaces in Metal-Ceramic Composites III: Thermodynamics and Kinetics B ...................................................... Tues. PM Interfaces in Metal-Ceramic Composites IV: Interface Reactions ............................................................................ Wed. AM Interfaces in Metal-Ceramic Composites V: Thermodynamics and Kinetics C ...................................................... Wed. PM Interfaces in Metal-Ceramic Composites VI: Mechanical Properties ....................................................................... Thurs. AM 1990 Lead-Zinc and Related Metals Symposium I: Keynote Session ................................................................................. Sun. PM 1990 Lead-Zinc and Related Metals Symposium II: Lead Smelting Practice ..................................................................... Mon. A 1990 Lead-Zinc and Related Metals Symposium III: Precious Metals Recovery in the Zinc Industry A ........................... Mon. AM 1990 Lead-Zinc and Related Metals Symposium IV: Physical Chemistry of Pb-Zn-Sn ....................................................... Mon. PM 1990 Lead-Zinc and Related Metals Symposium V: Precious Metals Recovery in the Zinc Industry B ............................ Mon. PM 1990 Lead-Zinc and Related Metals Symposium VI: Lead Recycling .................................................................................... Tues. AM 1990 Lead-Zinc and Related Metals Symposium VII: Zinc Pressure Leaching ...................................................................... Tues. AM 1990 Lead-Zinc and Related Metals Symposium VIII: Zinc Practices ..................................................................................... Tues. PM 1990 Lead-Zinc and Related Metals Symposium IX: Electric Furnace Smelting of Lead, Zinc & Tin ................................ Tues. PM 1990 Lead-Zinc and Related Metals Symposium X: Application of Lead, Zinc and Tin ..................................................... Wed. AM 1990 Lead-Zinc and Related Metals Symposium XI: Processing of Zinc-Lead Bearing Steelmaking Dust .................................................................................................... Wed. AM
Orangewood II
Q
93
Orangewood I
Q
115
Orangewood II
Q
116
Orangewood I
Q
17
Orangewood I Salon G
M
Q
34 131
Salon 1
M
17
Salon 1
M
34
Salon 1
M
54
Salon 1
M
73
Plaza D
I
35
Plaza D
I
55
Plaza D
I
73
Plaza D
I
95
Plaza D
I
117
Plaza D
I
132
Salon 1,2,3
M
10
SalonA&B
M
18
Salon C&D
M
19
Salon A&B
M
36
Salon C&D
M
37
SalonA&B
M
56
Salon C&D
M
56
Salon C&D
M
74
SalonA&B
M
75
SalonA&B
M
96
Salon C&D
M
96
4.-Annual Meeting
1990 Lead-Zinc and Related Metals Symposium XII: Lead, Zinc and Tin Minerals ............................................................. Wed. PM Low Temperature Si & Si-Based Epitaxial Structures: Electrical Properties & Defects I: Low Temperature CVD Si Epitaxy ................................................... Tues. PM Low Temperature Si & Si-Based Epitaxial Structures: Electrical Properties & Defects II: Si-Ge Epitaxy by CVD ....................................................................... Wed. AM Low Temperature Si & Si-Based Epitaxial Structures: Electrical Properties & Defects III Si-Based Heterostructures ................................................................ Wed. PM Low Temperature Si & Si-Based Epitaxial Structures: Electrical Properties & Defects IV: Si-Ge Epitaxy by MBE ...................................................................... Thurs. AM Low Temperature Si & Si-Based Epitaxial Structures: Electrical Properties & Defects V: Other Si Heteroepitaxy Techniques ................................................. Thurs. PM Magnetism Superconductivity & Their Interplay I ........................... Thurs. AM Magnetism Superconductivity & Their Interplay II ......................... Thurs. PM Materials Degradation in Low Earth Orbit 1.. ................................... Mon. AM Materials Degradation in Low Earth Orbit II ................................... Mon. PM Materials Degradation in Low Earth Orbit III .................................. Tues. AM Materials Issues In Electronic & Optoelectronic Connectors I: Mechanical Issues ...................................................... Mon. AM Materials Issues In Electronic & Optoelectronic Connectors II: Electrical Issues ........................................................ Mon. PM Materials Issues In Electronic & Optoelectronic Connectors III: Environmental Issues ............................................. Tues. AM Materials Issues In Electronic & Optoelectronic Connectors IV: Optical and Opto-Electronic Connectors & Splices ......................................................................... Tues. PM Materials Science of Magnetic Materials I ........................................ Mon. AM Materials Science of Magnetic Materials II ....................................... Mon. PM Metallization for Electronics Applications I: Thin Film Phenomena A - Basic Processes, Microstructure Development ............................................................ Tues. AM Metallization fo. Electronics Applications II: Thin Film Phenomena B - Interfacial Diffusion and Reactions ..................................................................................... Tues. PM Metallization for Electronics Applications III: Metallization in Silicon Based Technologies Electromigration, Microstructure, Processing and Properties .................................................................................. Wed. AM Metallization for Electronics Applications IV: Advanced Technology for Contacts to InP ....................................... Wed. PM Metallization for Electronics Applications V: Epitaxial Heteostructures of Metal/Compound Semiconductors .................................................................................. Thurs. AM Metallization for Electronics Applications VI: Advanced Technology for Contacts to GaAs .................................... Thurs. PM Modeling of Microstructure Evolution During Solidification I: Fundamentals ......................................................... Mon. AM Modelling of Microstructure Evolution During Solidifications II: Cellular and Dendritic Growth ........................... Mon. PM Modelling of Microstructure Evolution During Solidification III: MicrolMacro Modeling ......................................... Tues. AM Novel Processing Methods & Properties of Diamond & Diamondlike Thin Films I: Nucleation & Growth Fundamentals .................................................................................... Mon. AM
Salon C&D
M
118
Orangewood III
Q
76
Orangewood III
Q
97
Orangewood III
Q
119
Orangewood III
Q
133
Orangewood III Orangewood I Orangewood I ParkC ParkC ParkC
Q Q Q I I I
142 134 143 20 37 57
California IIIIIV
Q
21
California IIIIIV
Q
38
California IIIIIV
Q
58
California IIIIIV California I-II California I-II
Q Q Q
77 22 39
California I-II
Q
58
California I-II
Q
77
California I-II
Q
98
California I-II
Q
120
California I-II
Q
135
California I-II
Q
143
Plaza A
I
22
Plaza A
I
39
Plaza A
I
60
Orangewood III
Q
23
Novel Processing Methods & Properties of Diamond & Diamondlike Thin Films II: Characterization and Properties .................................................................................... Mon. PM _Novel Processing Methods & Properties of Diamond & Diamondlike Thin Films III: Advanced Processing ..................... Tues. AM Phase Stability & Diffusion Phenomena ............................................ Tues. PM Phase Transformations ....................................................................... Wed. AM Powder Production & Thermal Mechanical Processing .................... Tues. PM Process Fundamentals I: Fluid Flow Phenomena in Materials Operations ........................................................................ Mon. AM Process Fundamentals II: Thermodynamics and Reaction Kinetics in Metallurgical Processes A ........................................................................................ Tues. AM Process Fundamentals III: Thermodynamics and Reaction Kinetics in Metallurgical Processes B ........................................................................................ Wed. AM Process Fundamentals & Materials Processing I: Chemical Vapor Deposition ............................................................... Mon. PM Process Fundamentals & Materials Processing II: Welding & Solidification ................................................................... Tues. PM Process Fundamentals & Materials Processing III: Thermal & Mechanical Manufacturing Processes ........................... Wed. PM Process Mineralogy I: Applications to Hydrometallurgy, Flotation & Liberation ........................................ Thurs. AM Process Mineralogy II: Applications to Pyrometallurgy .................................................................................. Thurs. PM Reactive Metals I: Ti Production and Use of Emulsion and Precursors ................................................................... Mon. PM Reactive Metals II: Electrolytic Processes and Mg Production ............................................................................ Tues. PM Reduction Technology I: Cell Modelling & Magnetics ........................................................................................... Mon. AM Reduction Technology II: Cell Efficiency ........................................... Mon. PM Reduction Technology III: Bath Chemistry ....................................... Tues. AM Reduction Technology IV: Environmental Matters ........................... Tues. PM ,Reduction Technology V: New Anode/Cathode Materials ............................................................................................ Wed. AM Reduction Technology VI: Plant Matters and Alumina .................... Wed. PM Refractory Gold, Silver & Precious Metals Processing I: Plant Design .................................................... Wed. AM Refractory Gold, Silver & Precious Metals Processing II: New Process Methods A ................................ Wed. PM Refractory Gold, Silver & Precious Metals Processing III: New Process Methods B ............................... Thurs. AM Refractory Gold, Silver & Precious Metals Processing IV: Alternate Lixivants ...................................... Thurs. PM Solidification Behavior of Optical & Electronic Materials I ........................................................................ Tues. AM Solidification Behavior of Optical & Electronic Materials II ...................................................................... Tues. PM Solidification Phenomena I ................................................................. Wed. AM Solidification Phenomena II ................................................................ Wed. PM Superplasticity in Aerospace I, 2nd IntI. Symposium ......................................................................................... Mon. AM Superplasticity in Aerospace II, 2nd IntI. Symposium ......................................................................................... Mon. PM Superplasticity in Aerospace III, 2nd IntI. Symposium ......................................................................................... Tues. PM Superplasticity in Aerospace IV, 2nd IntI. Symposium ......................................................................................... Wed. AM
Annual Meeting-5 Orangewood III
Q
40
Orangewood III ParkA ParkA ParkC
Q I I I
61 78 100 79
Salon H
M
24
Salon H
M
62
Salon H
M
101
Salon H
M
41
Salon H
M
80
Salon H
M
121
Salon H
M
136
Salon H
M
144
Salon G
M
41
Salon G
M
81
SalonJ&K Salon J&K Salon J&K Salon J&K
M M M M
24 42 63 82
Salon J&K Salon J&K
M M
102 121
Salon 1
M
102
Salon 1
M
122
Salon 1
M
137
Salon 1
M
145
Conference Theater
Q
63
Conference Theater Salon G Salon G
M M
Q
82 103 123
Garden Rooms 1,2,3
I
25
Garden Rooms 1,2,3
I
43
Garden Rooms 1,2,3
I
83
Garden Rooms 1,2,3
I
104
6-Annual Meeting
Superplasticity in Aerospace V, 2nd IntI. Symposium ........................................................................................ Wed. PM Surface Interactions & Corrosion of Refractory Metals & Alloys in Liquid MetaWapor Metal Systems .................. Tues. PM Surfaces & Interfaces .......................................................................... Mon. AM Thermal Analysis of Metallurgical Systems I: Principles & Techniques ................................................................... Wed. AM Thermal Analysis of Metallurgical Systems II: Phase Transformations ..................................................................... Wed. PM Thermal Analysis of Metallurgical Systems III: Processing and Microstructural Correlations .................................. Thurs. AM Thermal Analysis of Metallurgical Systems IV: . Surface and Interface Phenomena ................................................... Thurs. PM Thermal Processing of Aluminum Alloys ........................................... Wed. AM Thermodynamics & Phase Relations of Exotic Materials I: Superconducting Materials ........................................... Tues. AM Thermodynamics & Phase Relations of Exotic Materials II: Intermetallics .............................................................. Tues. PM Thermodynamics & Phase Relations of Exotic Materials III: Theory .......................................................................... Wed. AM Thermodynamics & Phase Relations of Exotic Materials IV: Experimental Studies ................................................. Wed. PM Titanium I: Advances in Alloys/Processes ......................................... Wed. PM Titanium II: Advances in Melting Practice ........................................ Thurs.AM Titanium III: Computer Modeling of Fabrication Titanium Alloys ................................................................................. Thurs. PM
Garden Rooms 1,2,3
I
123
Plaza A Plaza D
I I
84 26
California III-IV
Q
105
California III-IV
Q
124
California III-IV
Q
137
California III-IV ParkC
Q I
146 105
Plaza C
I
64 ,
Plaza C
I
84
Plaza C
I
106
Plaza C ParkA ParkA
I I I
125 126 138
ParkA
I
147
Annual Meeting-7
SUNDAY
NOTE: The 1990 LEAD-ZINC AND TIN
Februar) 18,1990 AM
...
RELATED METALS SYMPOSIUM I: KEYNOTE SESSION BEGINS ON SUNDAY, FEBRUARY 18, 1990.
PM 1990 Lead-Zinc and
N
ReiatedMe~
~
Sympoailll1l:
I
Ker1n0le SeIIIon
i
TMS
119th TMS ANNUAL MEEETING TECHNICAL SESSIONS Anaheim, California • February 18 - 22, 1990 MONDAY
February 19, 1990
1 ~
f =
J
t IS
1= I
t IS
.J• ~ -. -;
j
~
. • ~
•
S ~ ~
t ~
t
AM
PM
HIII1H101l<
HlII14I-Rol1«y Memorial Sympo&iun I: High T, Supen:ondUCD'S & Relaled MeIaIIc Oxides (B) TMS
Deleell & Deled Reduction PrOC8S8ing In Semlconducb' ~ostuctLr.. 1 TMS
Reduction Proceulng In Semlconducb' ~ostucU.. 1I TMS
Nowl PrOC8S8ing """'ods & Propnes rJ Diamond & Diamondike Ttil Alms I: NucleUon & Growlh Fundanlentals TMS/ASM-MSO
Deleell & Deled
NMlPrOO&88ing Melhods & Propdes rJ Diamond &!liamondIk8 Ttil Alms II: Characterization and Properl.. TMS/ASM-MSO
TUESDAY
AM HIgh T8fnpelallJ'e Sl4*ccnclJctng Canpa.nds I: Pien«y SeIIIon TMS
loIaIeriais Sdence rJ Magnate Malaials I TMS
Materials 1_11 Eledroric & OpIoeIedronic Conneclotsl: loIechanlcl1lseu.. TMS
Matertals168U8I11 Electronic & OpIoeIedronic Conneclotsll: Eledrtcallau.. TMS
HIgh Temperaue condu
tt
Sl4* II: a, A: Canpa.nds S)!IIhe&i&lOxygen
lliIIoIutIon
TMS
PM
HIgh Tamperaue Sl4*conduclng Canpa.nds IV: !!Uk B: ""'ITexutngJ StbIllUIIon TMS
HIgh Temperaue Sl4*ccnclJctng Canpa.nds VI:
HIgh T8fnpelalU'e Sl4*ccnclJctng Compa.nds V: TtilRmsB: t.laOlWCUe TMS
TMS/~MSO
Low Tamperaue Si & Si-BaIed EpbxlaI Stucues: BecH:iI ProperIes & Delects t. Low Ten1pelalU'e CVO Si Epitaxy TMS
LowTtmpelalU'eSi& 51-Baled Epilaxial Stucues: Bac~caI Proper'" & DeIecIs U: Si-Ge Epilaxy ~ CVO TMS
Sdidilicalion Behavior or Opical & EIoctoric loIaMIiaisl TMS
Sdidlllcallon Behavior or OpCIcai & EIocroric loIaMIiaisn TMS
Semklon~cb' ~OIt1JCU.. 1II
TMS
Nowl Proc:easlng
MeMds & Properlies rJ
DIamond & OIarnondIke Thin Alms III: Advanced
Processi
BaIic PrCC88l88, Miaosrucue DMIopnenI TMS
loIeIaIizalion lot E1ecrorics ~ II: Thil FIm PIIanomena B- Int«racial DlIlIIIion and Reaclona TMS
Materlalsl_11 Eledroric. OpIoeIedronic Conneclots HI: ErwirarmentailsaJes TMS
Malaials 168U8I 11 Electronic • OpIoeIedrcnc Conneclots IV: Oplcar and OpIo-Eledronic Conneclots • SplIces TMS
E1ecrorics AAicaIIon& I: 1'1*1 ~ Phenanena A -
February 22, 1990
AM
High T8fnpelallJ'e Sl4*ccnclJctng Canpa.nds lit. Ttil FlhsA TMS
Deleell & Deled Reducton Proc:easlng In
THURSDAY
February 21, 1990
PM
Malallzallon lot
loIaIaftais Sdence rJ Magnate IoIaMIiaIs I TMS
WEDNESDAY
February 20, 1990
DelormatlorVMedlri:ll ProperI..
TMS
AM
PM
IoIagneIIl11 Sl4*conducl'lily & TheIr lnIerpIayl ASM-MSO
IoIagneIll11 & Sl4*conducl'lily & Their inlerpay II ASM-MSO
High TemperalU'e Sl4*conduclng Canpa.nds VII:
MelTapeIRIamenll
TMS
Low TampII'" Si & Si-BaIed EjlIIaxiII Stucues: Bacrlcal ProperIes & DeIecIs II
Si-Baeed Helarosrucuw TMS
Low Tarnperaue Si & Si-BaIed EjlIIaxiII Stucues: Bacrlcal Properl.. & Delects rv: Si-Ge Epilaxy ~ lISE TMS
LowTampllau.Si& Si-BaIed Epitaxial Stucues: Bacrical Proper" &DeIecIs V: Other Si Heleroepllaxy T~
E1ecroric • Maple loIoIorIais GA
Malalizalion lot EIectrorics A!IPICaIIonI II: ....lIIIzalion il SIkcn BaledT~
E1 a
0rn9a1on,
loIaosruclJle,
an~TMS Tlwmll Analyals rJ MeIaIwgicai Systanll:
PIi~'Tec:lrliques
TMS
loIeIaIizalon lot
IoIeIaIizab lot
E1ecrorics ~
Malalizalon lot
IV: Advned Tedw1dogy fa Contxlllo InP TMS
~.. oI""laII
VI: Adv-...d Tedw1dogy fa' ConIacta ID GaAs TMS
E1ecrorics AjlpicaIonI
Tlwmal Analyail rJ MeIaIwt:: Systansll: Phase ranlfonnalons TMS
V: EpiIaXial
Canpa.nd SemklonducIots TMS
E1ecrorics~
Tlwmal Analyals rJ MeIaIwgicai SjIIIemIID: ProceuIng and t.laOllruclJraI
Tlwmal Analyals rJ MeIaIwgical SySlans IV: Sl.rface and InIa'race
TMS
TMS
Contlalionl
PI1encmena
I
AUTHORS' COFFEE
I
I
SUDE PREVIEW AREA
I
~
I
TMS
8--Annual~eeting
119th TMS ANNUAL MEETING TECHNICAL SESSIONS Anaheim, California • February 18 - 22, 1990
IIb ~.;!
.. j
.a ;J
MONDAY
AM
H~ometAJrgy a Aqueous P\'OC88SiIg II: Physical Chemisry 01 PredpilaiCII and EllluentTreaDnent
H~omeIJIlJrgya
1oIaIeri.
TIAS
cnon TedrloIcgy I TIAS
.
a
;J c
• U
a
i
:z:
a
;Z
..-.
Relractcry Gold, SIver & Predous MeIJIs Procesaing IV: Alternate UxivanlB TIAS
cnonTec:I1noIogyV TIAS
cnon TedrloIcgy VI TIAS
Contol
ChemistryolPb-Zn-5n
TIAS
1990 Lead·ZInc: and Related Metals SympoIiunV: PreciOUll MeIaIs Recowry In fie ZInc: Inc1JUy B TIAS
Process Fundamenlals & Malarials PrOC88li1g I: ChemIcal VapaDepceI~
TIAS
AlJmila & Bauxite V: Alternate cr Modlfted ~
TIAS
IV: MeIJI Treatnent • Cast Shop Technology Racyding Cast Shop Technology V: Grain ReIIniIg a vt ~Automalcrl & Al-U Modllcalkrl tol TIAS TIAS lIAS
ElIne. . MIIII1urgy lecture
Pr-.Lea~IAS
Ex~act.r. MeIJIlJrgy
GA
Process Fundamenlals II: Thermodynamics and Reaction I
TIAS
AlJmila & Bauxite IV: AlJmila and Olamicals TIAS
Cast Shop Technology
1990 Lead·ZInc and Related Metals Sympoelun VII: 2Jnc
Readlw Metals I: 11 ProclJctIon and lise 01 Emulelon and I'reo.fsors TIAS
Plant DeIIIJI
Flta~ and
1990 lead·Zlnc and Related Metals Symposlun VI: Lead Racydng TIAS
1990 Lead·:OOC and Related Metals Symposlun IV: Alysical
PrOC88l Fundamenlals I: Fkid Flow PhenomMa In Mated. OperalCIIs TIAS
Relractcry Gold, Siver & P\'edous MeIJIs I'IooeI8ing III: New P\'oc:ees MeIhods B TIAS
AlJmila & Bauxite III: SGutaICII, ModeIilg,
I
J
TIAS
Relractcry Gold, Siver & P\'edous Melals ProcessIng II: New P\'oc:ees MeIhods A TIAS
AlJmila & Bauxite II: Bayer P\'ocesa TIAS
Meting Technology TIAS
(,7
Relrac:tay Gold, Siver & P\'edous Melals Procesang I:
AlJmila & Bauxite I:
Cast Shop Technology It MeIJI TrNlment · ~
1990 Lead·ZInc and Reialed Metals Symposiun HI: Predous MeIaIs Recowry In the 2Jnc Inc1JUy A TIAS
H~omeIJIlJrgy&
cnon TedrloIcgy IV TIAS
Cast Shop Technology II:
1990 lead·&C and Related Metals S~poIk.m II: Lead elqP\'acb TIAS
PM
Trealmentol ~na'. and EIlJenIB TIAS
ID
c
AM
cnon TedrloIcgy II TIAS
1ft
Aeqdng • SpeciII SeuIon TIAS
PM
cnonTec:I1noIogyll TIAS
TIAS
Cast Shop Technology I:
AM
Biol9chnol .
TIAS
February 22, 1990
PM Aqueous P\'oces:r IV:
Bauxite and EquiJrnent
j
a ;J
H~omeIJIlJrgy&
THURSDAY
February 21,1990
Aqueous P\'OC8fI&iIg HI: Electrolylc and Separallonl P\'0C8SS86 TIAS
~
.a
••
AM
PM
Aqueous P\'OC88SiIg I: l.eadlilg 01 Primaoy and Second;ry
WEDNESDAY
TUESDAY
February 20, 1990
February 19, 1990
I
1990 Lead·ZInc: and Related Metals Symposiun IX: 88C1r1c FIIIlIC8 Smellng of lead, ZInc: & 11n TIAS
1990 Lead·ZInc: and Related Metals Symposlun X: A/lIIIIcalion 01 lead, Zinc and lin TIAS
1990 Lead·Zlnc and Related Metals S)'IIIposIun VIII: ZInc: f'IadIceI TIAS
1990 Lead·ZInc and Related Metals Sympoeiun XI: PrOC8l8lng oI2Jnc·Lead a..;ng SteetnaI
Rea. IoIeIaIs I : Electroiylc P\'cx:es. ... d lAg PnxlJctIon TIAS
SoIIdIIc*n Phencmena SoIId1tIcatIon Alenomena II I GA GA
Process Fundamenlals & MaterIals PrOC88Ii1g II: Weldilg a SofidiIcation TIAS
PrOC88l Fundamenlals III: Thermodynamics and Procesa Fundamenlals & Materials PrOC88Ii1g III: Reaction I
TIAS
Cast Shop TecIrIoIogy VI:Cu*Ig· 2XXX, 6XXX a 7XXX ~
lIAS
CaslShopT~ VUI: Advarae il TedIndogy TIAS
1990 Lead·ZInc: and Related Metals Sympoelun XII: Lead, ZInc: and rn ....als TIAS
~ometalUgy
GA
Process Mil«alogy I: AAlIIcatIon6 to ~omelJllJrgy, FIata~ & LJbaoa~
TIAS
Procesa ....alogy II: Appications to PyromelalU'gy TIAS
~
a
1
Reduction TecflnoIogy t Cell Modeling & t.IagneIcs TIAS
Reduction TecflnoIogy I : Cell Ellciency TIAS
Reduction Technology V:
ReductIon TecflnoIogy UI: ReductIon Tec:hnoiogy IV: Ball ChamIsty ErMrCllmentailAatt«s TIAS TIAS
NewAn~.
MaIeriaIs
TIAS
Reduction TecflnoIogy VI: Plant Matl8rs and AlJrnina TIAS
II.
~
j
1
I
~ w
!i
l
II
~
•
~ 0
j'
c
c
ell
I
EIIne.on , Proc.elng Dlvllicn
LuncIIeon
lIAS Din_ a A_de "-I"on
II
I
light lIIIIIIe Dl¥llion
Llllc:heon
J
AUlHORS' COFFEE
I
SUDE PREVIEW AREA
I
I
)
Annual Meeting-9
119th TMS ANNUAL MEETING TECHNICAL SESSIONS Anaheim, California • February 18 - 22, 1990 February 20, 1990
~ ~
f c! c
i!
l
AM
PM
AM Superpaslidty il Aerospace. 2nd Ind. Symposhln I TMS
High Perlormance
~·Base Alloys I
TMS
Advanced Aeroepace Malerlals hom User a Suppler Penpedlves TMSiACers
Superpaslldty il Aerospace. 2nd Ind. Sympoaun II TMS
High Performance Alloys III TMS
High Performance
~·Base Alloys
II TMS
~·Base
Expert System AjIpIlcalions in Materials
CD
i (,)
i!
l
Processl;£I ManufaclJ TMSi ·1.150
Materials Dl9adation In Low Earth Orbit I TMSiASM-MSD
c i!
l
~
i 1&1
c
S
it
...:z:
=S
Modelilg 01 Miao6truclJre Evolulion Dlring Soitilicalon I: Fllldamentais TMS
Materials Dl9adation In low Earth Orbit II TMSiASM-MSD
Malerlals Dl9adation In Low Earth Orbit III TMSiASM-MSD
Advanced Metal & Ceramics Marix Compoaites I: Processing C«amlc Matrix Composites TMSiASM-MSD
Advanced Metal & Ceramics Marlx Compoaites II: ProcessIng Melal Matrix Compoaites TMSiASM-MSD
Modeling of Miao6truclJre Evolulion Olri SoIidificalion II: CeI:t and Denliltic
Modeling of MiaOSlrUClJreEvoIulion Dlrlng SoIidIIicaIion III: MiaclMacro ModeIng TMS
Growth
TMS
High Ternperallre MaterIals I: FraclLre Mechanics 01 High Tempera..e Malerlals TMS
CD
Z Z
it
Fatigue
GA
Cleep
GA
Thermodynamics & Phase Relallons 01 Exotic Malerlalal;
(,)
a
Fracuel
it
GA
c
S it •8
l,.. i
= 1=
Su-faces & Interfaces
GA
I I
FraCllrell
WEDNESDAY
TUESDAY
MONDAY
February 19, 1990
GA
Superconduclng
Malerlals ASM-MSD
Interfacesil MeIaI-Ceramlc Compoaites I: Thermodynamics and KInetics A TMSiASM·MSD
Interfaces il Meta~Ceramic
Compoaites II: n-etical Modeling 01 Interface Properties TMSiASM·MSD
II
THURSDAY
February 21, 1990
PM
AM
Superpaslidty il Aerospace. 2nd Ind. SympoIillO nl TMS
February 22, 1990
PM
Superpaslidty il Aerospace. 2nd Ind. Sympoaun rI TMS
Superpaslidty il Aerospace. 2nd 101. SyrnpoeillO V TMS
Tllanlum I: Advanceeln
Phase Slabllty & DiIIuaIon Phenomena
Phase Translormalons
Expert System Appllcallonsln Materials Processing & ManufaclJ~ TMSi 1.150
Expert System AjIpIicalions in Materials Processing & ManufaclJrilg III TMSiASM-MSD
Expert System AjIpIicallon6 in Materials
Thermal Processilg 01 Alumilllrl Alloys
Delamalon a MechanIcal Bellavia I
GA
GA
Powder Proructlon & Thermal Mechanical Processing
GA
GA
Advanced Metal & Ceramics Matix Compoaites III: FraCllre and Fai9Je TMSiASM-MSD
Advanced Metal a Ceramics Marlx Compoeltes IV: Interlace and High Ternperallre Performance TMSiASM·MSD
Su-face Interactions & Corr06ion 01 Retacklry Metals & Alloys il Lqt.id MetaWapor Melal Systems TMS
High Temperalu'e Materials II: Modelng 01 Hgh TernperalJre Malerlals TMS
Thermodynamics &
Phase Retations 01 Exotic
Materials II: ntermetalics ASM-MSD
Interfaces il MetaI·Ceramlc Compoaites III: Thermodynamics and Kiletics B TMSiASM-MSD
ELECTRONC MATERIALS LUNCHEON
AloysA'r~
TMS
AM
PM
~ a lie Extenalon 01
AgiluclearReacICr a UIe Exlenlion 01
uclearReacICr Malerlals I: Pre.lreVeseeIs TMSiASM-MSD
TIIriIm II: Advances il Melting Practice TMS
Malerlals II:
0Iher Components
TMSiASM-MSD
TIIlinlum III: Computer Modelilg 01 Falricallon TIIlinlumAiloys TMS
ProcessIng a
ManufaclJrilg rI TMSiASM-MSD
GA
Delormalon & MechanIcal Bellavia II
GA
Advanced Metal & Ceramics Matix Compoell8l V: ProceIa Modllilg and DeIormalon Proceai~ TMSiASM-MS
Advanced Metal & C«amics Matix CompoeItes VI: Emergl~ecI1noIogI_ and !cations TMSiASM-MSD
Advanced MeIaI & Ceramics Matix Compoates VII: Dampilg, TrIloI:,' and Non-Destru e Characterization TMSiASM-MSD
High Temperallre DeIormalon, FraCllre & AjIpIlcalion 01 Intermelallc Compounds I: Intermelallc WI1h U.. SruCllre TMSiASM-MSD
High Temperau. Delormalon, FraCllre & AjIpIlcalion 01 Intermelallc Compounds II:NI&TlAlllrlInid. TMSiASM-MSD
High Temperallre DeIormalon, FraCllre & ApplIcation 01 Intermelallc Compounds III: Selected High Ternperalu'e Intermelallcs TMSiASM-MSD
High Ternperalu'e Delamalon, FraCllre & AjIpIicallon 01 Interm8lallc Compounds IV:Engileering AjIpIlcations of Alternate MaterIals TMSiASM-MSD
Thermodynamics & Phase Retallono 01 Exotic
Phase Ret.... 01 Exotic
EnvIronmental Efleds
GA
Materials III: Theory ASM-MSD
I
SUDE PREVIEW AREA
Malerlals IV: ExperInentai Stucles ASM-MSD
Interlaces il Melaf..Ceramic Compoaites V: Thermodynamics and KInetics C TMSiASM-MSD fnt'lUII of Metala Lec\ure I
Interlaces il MetaI-Ceramic Compoail8lIV: Interface ReacIIon6 TMSiASM·MSD
AUTHORS' COFFEE
Thermodynamics &
Interfaces n Melaf..Ceramic Compoaites VI: Mechanical Properties TMSiASM-MSD
10-Annual Meeting
TMS
Minerals· Metals· Materials
1990 TMS ANNUAL MEETING Technical Program with Abstracts February 18-22, 1990 Anaheim, California
SUNDAY PM
world economy. If we fail to demonstrate the industry's benefits to the public, the 1990s could mark the beginning of the end of our industry.
(3:00 p.m.) ZINC INDUSTRY IN THE 1990s: G. A. Wickham, Zinc Corporation of America, 300 Frankfort Road, Monaca, PA 15061.
1990 LEAD-ZINC AND RELATED METALS SYMPOSIUM I: Keynote Session Sponsored by the TMS-EPD Lead-Zinc-Tin Committee Sunday PM February 18, 1990
Room: Salon 1-2-3
Marriott
Session Chainnen: T.S. Mackey, Key Metals and Minerals Engr. Corp., 1201 Logan, P.O. Box 3523, Texas City, TX 77590; R.D. Prengaman, RSR Corporation, 1111 W. Mockingbird Lane, Dallas, TX 75247
In the zinc industry, the past decade has seen both prosperity and depression. Significant changes have occurred throughout the world in mining, smelting, refining, and marketing. This paper will review many of the industrywide changes of recent years and the potential impact for 1990s. In addition to an historical review of supply and demand are discussion of technology considerations for zinc and zinc based products, effects of geographical shifting of markets, and projections for the 1990s. (3:30 p.m.)
BREAK
• (3:45 p.m.) TIN INDUSTRY IN THE 1990's: M.R. Sujatmiko, P. T. Tambang Timah, JL Jend Jatot, Suproto Jakarta Selatan, Indonesia. The paper describes the current tin markets, major smelting processes, and present and future applications of tin and its alloys.
(2:00 p.m.)
(4: t5 p.m.)
METALEUROP AND THE NON-FERROUS INDUSTRIES IN THE 1990's: J. P. Rodier, Metaleurop, Peripole 118, 58, rue Roger Salenqro, 94126 Fontenaysons-Bois, Cadex, France.
LEAD SMELTING AND REFINING, ITS CURRENT STATUS AND FUTURE: K. Moriya, Mitsubishi Caminco Smelting Co. Ltd, No 6-1, Ohte-Machi 1-Chrame Chiyoda-Ku, Tokyo, Japan.
Despite the good years that the non-ferrous business is now enjoying, one should not forget that in this business growth rates are low, fixed assets are high and prices are more cyclical
In the copper industry blast and reverberatory furnaces are disappearing in favor of direct smelting processes. These processes are highly mechanized, have closed vessels, use oxygen or enriched air, minimize gas volume, reduce energy consumption, reduce emissions, and improve hygiene of the workplace. Two decades later than the copper industry, new furnace technology is being applied to the lead industry. KIVCET (flash smelting) and QSL (bath smelting) are intensive smelting technologies which should remain environmentally and economically viable in the future.
than in other businesses.
Increasing environmental concern puts
the industry· under pressure. Lead and zinc companies are now profitable and profitability in the next ten years will depend on
size of units, location, and choice between new technology and upgrading current ones. Implementation of Metaleurope was a strategic move. It produced a company of sufficient size to achieve new steps in cost cutting and yet remain flexible enough to deal with rising variety of primary and secondary raw materials.
(2:30 p.m.) LEAD IN THE 1990s; CONTINUING GROWTH OR BEGINNING OF THE END?: J. L. Zelma, The Doe Run Company, Suite 400, 11885 Lackland Road, St. Louis, MO 63146. Environmental issues and how they are addressed, or not addressed by the lead industry in the 1990s will determine the-eventual fate of lead mining in both the U.S. and worldwide. Regulations in the West will become harsher. Industry has failed to educate the public regarding lead. Instead of public support for a critical mineral, we face widespread public ignorance and paranoia. If the benefits of the industry can be demonstrated, the 1990s will be a decade of continuing growth at least comparable to the
(4:45 p.m.) TOWARDS AD 2000: T.R.A. Davey, Metacon Consulting Services, 5 Rhodes Drive, Glen Waverley, Victoria 3150, Australia. The technologies of lead, zinc, and tin extraction have made varying degrees of progress over the last twenty years. Public concern about pollution and environmental degradation require that new processes be environmentally as well as economically feasible. New processes utilize intense reactors, utilize enclosed space obviating fume emissions, accept low grade or complex materials and wastes. Pyrametallurgical refinery needs to accept modern techniques, oxygen anode overvoltage in zinc EW needs to be investigated to reduce energy consumption, and vacuum refinery techniques need to be further developed. The paper describes processes which should be developed to overcame current problems.
Annual Meeting-ll observed in practice. The influenoe of the conditions under which the anodes are set and oCthe properties of the anodes on the mechanical stresses which arise, is reported on.
MONDAY AM CARBON TECHNOLOGY I Sponsored by TMS Aluminum Committee (formerly Light Metals) Monday AM February 19, 1990
Room: Salon 2 Marriott
(10:30 a.m . )
STUDY OF ANODE MANUFACTURING PARAM£I'ERS INFLUENCING THE TIlERMAL SHOCK RESISTANCE, Jean Bigot, Aluminium Pechiney, A1uval 38340, Voreppe, France. Witb an apparatus measuring tbe anode resistanoe to thermal shock, we have been able to study manufactured anodes and bencb scale anodes. The results obtained characterize tbe manufacturing parameters' impact on tbe anode tbermal shock resistanoe. This indicates bow to adjust some of tbe parameters or bow to change the process, in order to reduce tbermal shock if the cracking phenomenon occurs in tbe pot.
CAST SHOP TECHNOLOGY I: RecyclingSpecial Session
Session Chainnen: Euel R. Cutshall, Manufacturing Sponsored by the TMS Aluminum Committee Technology Laboratory, Reynolds Metals Company, Sheffield, Alabama 35660; Dennis G. Brooks, ManufacturMonday, AM Room: Salon 4 & 5 ing Technology Laboratory, Reynolds Metals Company, February 19, 1990 Marriott Sheffield, Alabama 35660. (8 : 30 a.m.)
SOME PRACTICAL CONSEQUENCE<; OF ANALYSES OF THE CARBOXY AND MRBURN RE.4CT10NS OF ANODE CARBONS, N. Bird, B. A. Sadler, Comako R~rcb Centre, IS Edpn Road, Thomastown, Victoria 3074, Australia. B. McEnaney, School of Materials Science, University or Balb, Batb SA2 7AY, United Kingdom.
Theoretical analyses are presented of carboxy reactivity and airburn reactivity of anode carbons. The carboxy reaction rate occurs in the transitional region between chemical control (Zone I) and in·pore dilIusion control (Zone II). The airburn reaction rate occurs in tbe transitional region between Zone II and control by diffusion tbrough a gaseous boundary layer (Zone III). Overall carboxy reactivity can be decreased somewhat by reducing porosity, but to a greater extent by reducing reactivity. Airburn reactivity is less sensitive to modifications of the carbon, but may be re4uced by control of the porosity of the cover. (9:00 a.m.) THERMAL CONDUCTIVITY AND STRUCTIJllE OF CARBON ANODES, ~ T.
1.0& IlIStitute or IDorpnic Cbemistry, Trolldllelm, Norway.
~
iDltitute or TecIInolo&Y, N·7Il34
Thermal conductivity or anode carbon up to 7OO'C is reported. The studies are carried out by • recently de\oeloped transient method. Current Is passed through a strip of iron insulated OB bolb sides by mica and the temperature increue Is determined. The thermal COnductivity Is correlated 10 structure and baking temperature. (9 : 30 a.m.)
CHARACTERIZING ANODE PROPERTIES BY QUALITY FIGURES, F. Keller, Dr. W. Schmidt·Halling, A. Kooijman, Aluminium & Cbemie, Rollerdam BV, Oude Massweg SO, NL-3197 KJ Botlek·RollerdamfThe Netherlands; W. Fischer, R. Perruchoud, Research & Development Carbon, Le Cbable, P. O. Box 157, CH·3960 Sierre/ Switzerland. The anode behaviour in the reduction pot is generally deduced from the test values for the physical and cbemical properties of tbe anodes, sinoe tOOse factors can be measured directly after production of the electrodes, allowing conclusions to be drawn about the quality. Good quality is largely characterized by Ibree properties: low CO2 and airburn and high resistanoe to thermal shock. All of the tbree factors mentioned are influenced by more than one physical property. II has proved expedient, therefore, 10 express sucb properties in quality figures. Three relevant quality figures are described, and it is shown that with their help the anode quality can be presented more clearly than with values for each property. (10 : 00 a.m.)
TIlERMAL SHOCK IN ANODES FOR TIlE ELECTROLYTIC PRODUCTION OF ALUMINIUM, Dr. W. Schmidt·Halling, Aluminium & Cbemie, ROllerdam BY, Oude Maasweg SO, NL-3197 KJ BoUek·Rollerdam!The Netherlands; E. Kummer, Helbling Technik AG, Neumattstrasse 30, CH·SOOO Aarau,lSwitzerland. There are many publications describing the resistance to cracks caused by thermal shock in oeramic bodies by using quality figures. In LIGHT METALS 1988 these figures were applied to anodes for the aluminium industry. Such quality figures allow a simple and Cast appraisal of an anode's tendency to thermal shock. The mechanical stresses, which occur when a heat wave penetrates the cold anodes after they are set in the hot electrolyte, were calculated three·dimensionally. The results confirm the validity of the already mentioned quality figures on the one hand and correspond very well to the shape of cracks in anodes
Session Chainnen: G.J. Hildeman, Molten Metal Processing Division, Alcoa Laboratories, Alcoa Center, PA 15069; W.S. Peterson, 2113 East 37 Avenue, Spokane, WA 99203. (8 : 30a.m.) "Recycling, A Primary/Secondary Aluminum Producer's Perspective," James M. Creel II, Reynolds Metals Company, Recycling and Reclamation Division, P. O. Box 27003, Richmond, VA 23261
Recycled aluminum is a key source of aluminum and will remain so in the future . Scrap will become more important as aluminum demand exceeds supply. Beverage can recycling will continue to grow driven by EPA's goal of 25% reduction in MSW. An automotive casting expansion of 30% in the next 10 years will keep scrap demand high. This will promote opportunities for scrap processing improvements, alloy development, studies on trace element effects and problem solving on metallic contaminants. (9 : 00 a.m.)
A SECONDARY'S VIEW OF RECYCLING: ]. S. Viland, Wabash Alloys, P. O. Box 466, Wabash, Indiana 46992. This paper will present an overview of the secondary aluminum industry which produces foundry alloys. The secondary aluminum industry has been "mining" an aluminum resource since the advent of the primary aluminum industry but really came into its own during World War IT. Processing equipment and improved furnace technology has increased the efficiency of the process of recovery significantly. With the stronger emphasis on recycling, the future looks bright; however, the handling of solid waste and other environmental controls provide a challenge to the industry. (9 : 30 a.m.) ALUMINUM RECYCLING· EVERYBODY'S BUSINESS, TECHNOLOGICAL CHALLENGES AND OPPORTUNITIES: J H L van Linden, Alcoa Laboratories, Alcoa Center, PA 15069.
Less than 20 years after the introduction of the Hall·Heroult process in 1888, the first recycling efforts to recover the valuable metal started in the fledgling automobile industry. As the aluminum industry grew, the business of recycling was increasingly left to entrepreneurs who applied low risk, low cost technology in their scavenging operations. After World War IT the aluminum industry exploded and with it a secondary industry which started to utilize increasingly more sophisticated technology. When the primary industry joined the quest for old scrap in the mid '60s, it had to develop advanced scrap collection, preparation and remelt methods in order to be competitive with the low cost secondaries. While the primaries presently are "closing the loop" by recycling more than half the beverage cans sold in the USA straight back into container sheet, the secondaries are "integrating forward" by producing finished products on minimills, thus erasing the once distinct boundaries between the two industries. This realignment of the aluminum industry's infra structure finds its origin in a set of converging forces like rising energy costs, shrinking resources and mounting environmental concerns. The technological challenges and opportunities for everyone involved will be discussed against this background.
(10 :00 a.m.)
BREAK
12-Annual Meeting (10 :20 a.m.) ALUMINlliM RECYCLING· A EUROPEAN PERSPECTIVE, I Ronwjnlcel. Alcan Deutschland Gmbh, Postfach 1241, Hannoversche Str. I, 0-3400 Gottingen, Gennany.
In public opinion, waste of energy and resoUICeS and "hannful" to the environment are labels which are still linked to aluminum. This paper is to prove that the contrary
should apply, comparing the accumulated specific primary energy consumption in the manufacture of competing beverage containers for soft drinks and beer. Recycling takes over a decisive role in this context. Aluminum recycling has been well established in Europe for many years. With more than 100 secondary smelters and foundries, an overaJl recycling rate of 30-35% vs. aluminium consumption is already achieved. The updated aluminium recycling rates will be shown for the fields of transpon, building, electrical and general engineering. For four years successful recycling of used aluminium beverage cans is practiced in Europe. (The scenarios of the different countries will be shown.) Our industry is confident that the aluminium beverage can is well armed for 1993 when trade barriers will fall within the European market. It is believed to be the preferred container of the future. Consequently, our industry needs to expand its recycling activities in all product applications where this is economically and ecologically feasible. (10:50 a.m.) THE CONTINUOUS CASTING ANU DIRECT ROLLI NG OF WIDE, THIN SLAB INTO HOT BANU BY BARKET ALUMINUM CORPORATION USING SECONDARY OPERATIONS ANU HAZELETT TECHNOLOGY. Stanley Platek, Barmet Aluminum Corporation, P.O. Box 29, UhrichSVille, OH 44683, S. Jerry Desautels, Hazelett Strip-Casting Corporation, P.O. Box 600 , Colchester, VT 05446. Barmet Aluminum has recently perfected a proce ss for continuousl y converting recycled aluminum into commercial hot band at high tonnage rates. This paper will r eview the development of the process and will explain its various parts. Particular emphasis will be given to the use of Hazelett twin-belt technology.
(11 : 20 a.m.) UBC RECYCLING - TIIE WIN/WIN PROPOSmON: R. N. Caruso, Golden Aluminum Company, 3000 Youngfield, Suite 230, Lakewood, Colorado 80215.
The U.S. aluminum beverage can market grew to an estimated 79.1 billion containers in 1988 compared with 73 billion cans in 1987. Aluminum beverage cans account for 96% of the beer and soft drink cans and 53% of the total beverage packaging market Coors first set the stage for aluminum can recycling as an early user and developer of the aluminum beverage can in 1959. Aluminum beverage can recycling has gained consumer acceptance at an accelerated pace. The national recycling rate has reached an all time high of nearly 55% from 30% in 1979. Increasing the preservation of our natural resources, putting money back into the consumers' pockets and providing an economical source of raw material to sheet suppliers is truly a win-win situation. At Golden Aluminum Company, we consider the continuing emphasis on advanced technology, can stock quality and recycling to be key issues in assuring the long term competitiveness of the all aluminum beverage container.
The effects of unstrained caps on the motion and multiplication of dislocations in underlying strained films are also studied. It is shown that the "threading" dislocation cuts through th~ cap and is dragged along with the moving dislocation if the cap is sufficiently thin, but is constrained to remain entirely within the film if the cap exceeds a cenain thickness. These predictions are compared with recent observations of misfit di slocations in capped heteroepitaxial films of Si-Ge grown onto Si substrates. (9 :00 a . m.)
Invited
DEFECT FORMAIION AND ORIENTATION RELATIONSHIPS FOR EPITAXIAL CRYSTAL GROWIH: Jack W!!Shbum, ZUZllnna Lillienlhal·Weber, Eric P. Kvam, Materials and Chemical Sciences Division, Lawrence Berkeley Laboratories, I Cyclotron Road, Berkeley, CA 94720 Factors affecting the nucleation or prpagation of dislocations, stacking faults, microtwins, and inversion domain boundaries in epitaxially grown semiconductor layers are reviewed, with examples for homoepitaxiaJ and heteroepitaxial MBE-grown layers on substrates having varying degrees of mismatch or ()Ifferent crystal symmetry. M~h~lI i ,"1S CUI ~''''''''.~'''' of d~&;.;t. AI the heterointerface and in the epilayer are diliCussed. For epilayers with bulk mismatch from 0 to 4%, methods for reducing defect density in the epitllxiallayer are considered. Examples of structural details in epilayer~ and at heterointerfaces, particularly those which may betevealed by transmis~ion electron microscopy, arc given. (9:30 a . m.)
DEFECTS AND DEFECT REDUCTION PROCESSlNG IN SEMICONDUCTOR HETEROSTRUCTURES: .s....sI!aran and J. Narayan. Dept. of Materials Science and Engineering, Nonh Carolina State University, Raleigh, NC 27695-7916. We have studied the nature and fonnation of defects in various semiconductor heterostructures like GaAs/Si, GaAs/SOI and InP/GaAs/Si. Transmission electron microscopy was used to study the details of nucleation and glide of misfit dislocations. The misfit dislocations at the heterointerface are of both 60° and 90° type. In GaAs/Si heterostructures the nature of mi sfit dislocations changes from 60° to 90° as the thickness of the epilayer is increased. The elastic strain in the film has been correlated to the misfit dislocation spacing and a model has been developed to detennine the strain relaxation as a function of epilayer thickness. We have also studied the role of strained layer superlattices on dislocation density reduction in the epilayers. The effect of supcrlanice layer thickness, composition and period in reducing threading dislocation density has been studied and a model has been developed to lead to an understanding of the interaction of dislocation with the strained layers. (9 :50 a.m.)
MOLECULAR DYNAMICS STUDIES OF DISLOCATION CORES IN SI AND SI-GE M. S. Duesbery, D.J . Michel, Naval Research Laboratory, Washington D. C. 20375 and B. Joos , Ottawa-Carleton Institute for Physics , Ottawa, Ontario, Canada KIN 9B4
DEFECTS AND DEFECT REDUCTION PROCESSING IN SEMICONDUCTOR HETEROSTRUCTURES I: Dislocations and Dislocation Interactions
The structure and mobility of dislocation cores in StillingerWeber' Si and Tersoff2 Si and Si - Ge model s are exami ned at an atomistic level using molecular dynamics. Straight and doublekinked 30° and 90° partial dislocation glide dipoles are modelled in a strain - free environment, in the presence of applied shear strains and at temperatures up to the melting point. The results are compared with experimental observations .
Sponsored by the TMS Electronic and Photonic Device Materials and Electronic Materials Committee
I . f . H. Stillinger and T.A . Weber, Phys. Rev B31 , 5262 (1985) . 2. J. Tersoff , Phys. Rev. Letters ~ , 632 (1986).
Monday,AM February 19, 1990
Room: Orangewood II
Quality Hotel
Session Chainnan: J. Narayan, North Carolina State University, Raleigh, NC 27695-7916 (8 :30 a .m.)
Invited
MECHANISMS AND KINETICS OF MISFIT DISLOCATION FORMATION IN HEIEROEPITAXlAL SIRUcruRES ~,Department of Materials Science and Engineering, Stanford University, Stanford, CA 94305 The formation of misfit dislocations in heteroepitaxial structures is reviewed with panicular reference to the mechanisms by which dislocations in strained films move and multiply. Matthews' equilibrium expression for the critical thickness can be derived by considering the incremental extension of a misfit dislocation associated with the movement of a "threading" dislocation within the film. This model can also be used to describe the kinetics of misfit dislocation formation. It is shown that kinetic factors are responsible for the commonly observed discrepancy between the observed misfit spacing and the equilibrium spacing
(10 : 10 a.m.)
ATOMIC STRUCTURES AND ENERGETICS OF DISLOCATIONS IN Ge/Si, GaAs/Si AND C/Si HEIEROSTRUCIURES : A.S,Nandedkar and J. Narayan, Dept. of Materials Science and Engineering, Nonh Carolina State University, Raleigh, NC 27695 - 7916. Atomic structures of 60° and 90° dislocations in germanium and diamond films on silicon substrate were calculated by energy minimization technique. Suitable interatomic potentials were used for calculating energies. A realistic model of cross-grid of dislocations was simulated to include interactions of neighboring misfit dislocations in the film. A transformation of 60° dislocations to a 90° dislocation was studied on the basis of energetics and atomic structures of intermediate configurations. Other defect configurations commonly observed in diamond film on silicon substrate were studied and the results were correlated with the formation of the misfit dislocations. Three steps were followed during' determination of atomic structures : (I) calculation of atomic positions by energy' minimization, (2) computer simulation of calculated structures to obtain high, resolution lattice images, and (3) comparison of high resolution transmission electron micrographs with the calculated images. (10 :30 a.m.)
BREAK
Annual Meeting-13 (10:45 a.m.) EPITAXIAL LAlERAL OVERGROWTH OF SILICON FILMS ON SILICON DIOXIDE SURFACES: Se WoonL: Oh and A.S. Vue, Department of Materials Science and Engineering, University of California, Los Angeles, California 90024 Single-crystalline silicon films have been grown over Si02 surfaces by the epitaxiallatera1 overgrowth (ELO) technique. They were grown at a temperature from lOOO to lIOO·C utilizing a Si!4-HCI-H2 gas system. Thicknesses of the overgrown films varied from less than 111m to 30l1m and the horizontal to vertical growth ratio varied from I to 2.5. Surface morphology and growth mechanism of these films were studied in terms of deposition temperature, gas composition and Sil4/HCI ratio. The success in growing these films is attributed to the suppression of nucleation and growth of polysilicon on Si~ surfaces and an enhancement of growth of epi-silicon over the Si02 surfaces. The defect density of epitaxial silicon films was revealed by a preferential etching technique. The single crystallinity and quality of these films were revealed by electron channeling patterns. Sample cross-sections were studied using high resolution electron microscopy (HREM) and SEM. The potential for device applications and tIre problems in preparing these films are discussed. (11 :05 a.m.) MATERIALS STUDY OF Eric P. Kvam, Jack DIVISion, Lawrence and Lisa P. Allen, Taunton, MA 027BO.
SILICON-ON-INSULATOR MATERIAL BY TEM: Washburn, Materials and Chemical Sciences Berkeley Laboratories, Berkeley, CA 94720 Paul M. Zavracky, Kopin Corporation,
Silicon-on-insulator (SOl) technology addresses the need for radiation tolerant devices, high voltage and three-dimensional circuitry applications. SOl produced by Isolated Silicon Epitaxy (ISE) is a commercialized method which results in excellent SOl material quality with proven device results. The remaining, isolated dislocations have been examined in detail by transmission electron microscopy (TEM). These have been diagnosed as normal lattice dislocations with no faults or twins in the material. The nature, source, and possible elimination of the remaining dislocations will be discussed. (11 :25 a.m.) COMPARISON OF Si/Si02 INlERFACES IN SIMOX AND ZMR SOl MATERIALS: A.R.Srivatsa, K.D.Stonnington and J.Narayan, Dept. of Materials Science and Engineering, N.C. State Univ., Raleigh, N.C. 276957916; L.P.Allen, Kopin Corporation, Tauton, MA 73019. The characteristics of interfaces in SOl materials has been studied using transmission electron microscopy. High resolution electron microscopy is employed to study the details of the evolution of silicon/oxide interfaces in buried oxides. The mechanism of formation of silicon islands adjacent to the silicon/oxide interfaces and the details of the orientation of islands with respect to the silicon top layer and substrate are discussed. It is shown that extremely sharp interfaces on an atomic scale can be produced in both ZMR and SIMOX SOl materials under suitable processing conditions. Comparison is also drawn to silicon/oxide interfaces formed by other methods such as plasma oxidation and wet oxidation of silicon with germanium rich layers at the interface. Finally, attention is drawn to the fact that crystalline oxides were not observed at any of the silicon/oxide interfaces. (11 :45 a.m.) THE EFFECT OF Ge IMPLANTATION ON THE THREADING DISLOCATIONS IN SIMOX. K.S. Jones, D. Venables and F. Namavar*, Dept. of Mati's. Sci. & Engrg., Oniv. of Florida, Gainesville, FL 32511. *SPIRE Corp., Patriots Park, Bedford, MA 01730. W~ have studied the effects of Ge+ implants into single ~gh dose o implanted and annealed SIMOX wafers (160 keV ,1.5 x 10' if; annealed 1300 C, 6 hrs., dry N4 ) with cross-sectional and plan view TEM. The Ge implant energies and doses were adjusted to obtain a constant amorphous/crystalline interface depth in the surface silicon layer. Subsequent solid phase epitaxial (SPE) regrowth of this continuous amorphous layer was obtained with an 850 C, 0.5 hr. anneal. This anneal resulted in a high density of end-of-range damage (category II dislocation loops) just beyond the original position of the amorphous/crystalline interface \"200 A above the SiOk/superficial silicon interface.) We observe a reduction in dislocation density (~100 plan view TEM micrographs examined) upon Ge implantation and second, the remaining threading dislocations are deflected by the Ge implants. The dislocation density reduction is a factor of 2-3 at lower Ge doses (~1 x 10,s/cm"-) and the density appears to continue decreasing with increasing Ge dose. At present the threading dislocation density after a 1 x 10"/cm-'Ge implant is 1 x 10 8 / cm J.. The density reduction process may be associated with reverse climb back to the Si/SiO~interface induced by the high concentration of point defects below the amorphous/crystalline interface. Additional results on higher dose Ge implants will be presented.
(12 :05 a.m.) MECHANISMS OF DEFECf FORMATION AND REDUCTION IN OXYGEN IMPLANTED SOl: S Visjuemgtrakul S.J. Krause. Dept. Chemical. Bio &. Materials Engineering, Arizona State University. Tempe AZ 85287, B.F. Cordts. Ibis Technology Corp., Danvers. MA 01923, P. Roitman. National Institute of Standards and Technology, Gailhersburg, MD 20899 Optimum perfonnance and a high yield of integrated circuits built on high-dose oxygen implanted silicon-on-insulator (SIMDX) material require a high quality top Si layer wilh a minimum number of defects. We are reporting on mechanisms of defect formation and reduction in single high dose implanted SIMOX. Samples were implanted to a dose of 1.8xI018cm -2, at 600°C, at 200 keY, and 1 mA/cm2 . The general structure is a 0.3 J.l.m top Si layer, a 0.4 J.l.m buried oxide layer, and damage zone below Ihe oxide layer. In the top Si layer, Ihere are uails of bubbles at Ihe surface and large 0.1 ~m mUltiply faulted defects extending from the buried oxide interface. When temperature was varied from 550 to 7oo D C, a large increase in final defect density occurred for samples implanted at 550°C. Annealing temperatures from 1250 to 1300°C for 2 hours in a nitrogen ambient produced a "precipitate free" region in the upper half of the Si layer. Dislocations frequently ran laterally between adjacent precipitates. but some ran upward to the wafer surface. Increasing time at a given temperature increased precipitate size only slightly. Increasing temperature to 1325°C eliminated all precipitates and lateral dislocations. However, threading dislocations could not be removed. Annealing with an argon ambient produced some structural changes in comparison with the nitrogen ambient. But no threading dislocations were observed indicating a low defect density of 105 cm- 2 . The results indicate that implantation temperature is an important parameter in defect formation, and that high temperature annealing is a necessary, but not sufficient, condition for defecl density reduction. Detail mechanisms of defect reduction and of the effects of imvlantation will be discussed further.
FATIGUE TMS General Abstract Session Monday AM February 19, 1990
Room: Plaza B Inn at the Park
Session Chairman: C.P. You, Research Staff, Ford Motor Company, P.O. Box 2053, Dearborn, MI 48121 (8:30 a.m.) FOOTE ELEIIENT PREDICTION OF HIGH CYCLE FATIGUE LIFE OF ALUliiNUIl ALLOYS, William J. Baxter and Pei-Chung Wang, Physics Department, General Motors Research Laboratories, Warren, Michigan 48090-9055
A new method is described for calculating the long fatigue life (>105 cycles) portion of the stress-life (S-N) fatigue curve for precipitation-hardened aluminum alloys. It is. based upon a finite element model of the deformation of a persistent slip band (psb) and the only material parameter required is the ultimate tensile strength of the alloy. The stress dependence of the plastic strain at the tip of a psb is shown to be very pronounced and to closely match that of a S-N fatigue curve. Very good agreement is obtained for 606l-T6, 20l4-T6, 2024-T4 and 7075-T6 8 aluminum, and the fatigue strength (at 10 cycles) is calculated to be 26~ of the tensile strength of each alloy, in agreement By contrast, the plastic strain at a with experimental data. crack tip has a much weaker stress dependence. Thus, these calculations also confirm that the elongation of a psb, and not crack growth, is the rate-controlling process in high cycle fatigue. (8:50a.m.)
EFFECT OF SCRATCHES ON THE FATIGUE LIFE OF AERONAUTICAL MATERIALS: ~ Inchekel, and J.E. Talia, Department of Mechanical Engineering, The Institute for Aviation Research, The Wichita state University, Wichita, KS 67208. A comparative study was performed on the general fatigue behavior of scratched lightweight aeronautical materials when subjected to tensiontension cyclic loading. Relevant information is provided on the crack formation, propagation, and the number of cycles to failure, in relationship to the parameters of the scratch. Furthermore, optical and scanning electron microscopy were used to identify the possible failure mechanisms. This work is sponsored by the U. S. Army Research Office under contract number DAAL03-87-G-0003 and performed at the Institute for Aviation Research. (9:10 a.m.)
FATIGUE BEHAVIOR IN SCRATCHED AND SHOT PEENED ALUMINUM LITHIUM ALLOYS: A. Eftekhari and J.E. Talia, Department of Mechanical Engineering, The Institute for Aviation Research, The Wichita state University, Wichita, KS 67208. It has been observed that surface scratches drastically decrease the fatigue life of Al-Li alloys. To improve the the fatigue life of scratched
14-Annual Meeting samples, shot peening, chemical and electrochemical polishing were used. At certain stresses a detrimental effect on life of the Al-Li alloy is observed. This behavior is discusses in relation with the residual stresses available near the scratched surface.
the surface roughness and the effective stress intensity range. In 1 ight of these experimental observations. the role of the test environment in fatigue crack propagation behavior of a 2090 Al-Li alloy will be discussed.
This work is sponsored by the U. S. Army Research Office under contract number DAAL03-87-G-0003 and performed at the Institute for Aviation Research.
FATIGUE-CRACK GROWTH BEHAVIOR IN AL-LJ ALLOYS: EFFECTS OF TENSILE AND COMPRESSION OVERLOADS:· K T venkateswara Rao, W. Yu and R. O. Ritchie, Center for Advanced Materials, Lawrence Berkeley Laboratory, University of California, Berkeley, CA 94720.
(9:30 a.m.) SPECIMEN HEATING BEllAVIOR OF AUSTENITIC STEELS DURING FATIGUE TESTING IN LIQUID HELIUM": T. Ogata and O. Umezawa. National Research Institute for Metals. Tsukuba Labs .• 1·2·1. Sengen. Tsukuba. Ibaraki 305. Japan Significant temperature rise occurs during mechanical tests at liquid helium temperature. We carried out measurements and an analysis of temperature rise behavior of specimen during constant-amplitude stress and strain cycling tests in liquid helium. Materials used in this study were typical austenitic stainless steels for cryogenic applications. SUS 304L. 310S. and 316LN. The reduced section diameter of the specimen was 6 rnm. A 2 mm diameter hole was drilled along the IIlds to measure the interior temperature of the specimen and two pairs of thermocouple were inserted into the hole. The magnitude of temperature rise increased with the increase of the test frequency and the strain range. The average temperature rise at 0.1 Hz at 3 % total straiD range was about 10K. Further discussion will be made on the elastic and plastic deformation behavior and the analysis of temperature rise with physical properties. such as specific heat. thermal conductivity. elastic constants. and so on at the temperature. "This work was supported by Special Coordination of Science and Technology Agency of the Japanese Government.
(9:50 a.m.) Effect of Aging on the Corrosion Fatigue Crack Initiation and Propagation in AF1410 Steel: f.S....fsm. S.J. Gill. L.A. Cooley, Naval Research Laboratory. Washington, D.C .• 20375-5000, and G.A. Yoder. Office of Naval Research, Arlington, VA 22217. The influence of aging treatment on the corrosion fatigue crack initiation and propagation characteristics of AF1410 steel in salt water are stUdied. While aging treatment has only limited effect on the corrosion fatigue crack propagation behavior. it significantly influences the corrosion fatigue initiation life of AF141 0 steel. Overaged AF1410 exhibits fatigue life an order-of-magnitude longer than that of underaged alloy. Fatigue crack propagation rates are not influenced by the test environments. However, fatigue crack initiation life in salt water is found to be significantly lower than that in air. These observations will be discussed in terms of the microstructural feature and their interaction with the environment. (10: 10 a.m.) LOAD RATIO EFFECTS ON FATIGUE CRACK PROPAGATION IN METASTABLE AUS1ENITIC STEELS, z...Md. and J. W. Morris, Jr.• Center for Advanced Materials. Lawrence Berkeley Laboratory, 1 Cyclotron Road. Berkeley, CA 94720 The 304L austenitic stainless steel is metastable at cryogenic temperature; it undergoes mechanically induced martensitic transformation. The fatigue crack propagation rate in 304L austenitic stainless steel has been found to depend very much on the ratio of minimum load to maximum load that indicates the mean tensile load of the cyclic loading. The phenomenon is explained in terms of the crack tip stress field that results from the austenite to martensite transformation. The associated volume expansion is constrained by the surrounding elastic materials thus creates a compressive stress field at the crack tip. The compressive stress field retards the fatigue crack propagation rate. but this effect diminishes as the mean tensile load increases. This work is supported by the Director. Office of Energy Research, Office of Development and Technology. Office of Magnetic Systems. Base Technology Division of the U. S. Department of Energy under Contract No. DE-AC0376SFOOO98. (10:30 a.m.)
EFFECT OF LOAD RATIO ON FATIGUE CRACK PROPAGATION BEHAVIOR OF A 2090 Al-Li ALLOY IN VARIOUS ENVIRONMENTS: K. S. Shin and S. S. Kim. Dept. of Chemical, Bio and Materials Engineering. Arizona State University, Tempe, AZ 852B7; E. W. Lee. Naval Air Development Center. Warminster. PA 18974 Aluminum-lithium alloys have excellent resistance to fatigue crack propagation in air compared to other high-strength Al alloys. In the present study. the effect of load ratio on the fatigue crack propagation behaVior of a 2090 Al-L i alloy was investigated in various environments. Crack propagation tests were carri ed out at load rat i os of 0.1 and 0.75 in laboratory air. 3.5% NaCl solution. vacuum and dry hydrogen. The fatigue crack propagation rates varied significantly depending on the test environment. being highest in 3.5% NaCl and lowest in vacuum. The test environment also had a significant effect on
(10:50 a.m.)
Studies on the growth of long (>10 mm) fatigue cracks in AI-Li alloys have shown that behavior is markedly influenced by crack-tip shielding primarily from crack-path deflection and resulting crack closure. Consequently, where such shielding is restricted, it may be expected that properties are compromised. In the present study, crack-growth behavior in an AI-Li alloy 2090 under the influence of single tensile and compression overloads is examined, and results compared with traditional aluminum alloys. It is found that the 2090 alloy shows greater degree of retardation following a tensile overload, but small compression underloads are sufficient to reinitiate growth of cracks arrested at threshold. These effects are attributed to the promotion and supression of shielding under respective load excursions. ·Sponsored by the Department of Energy, Office of Basic Energy Sciences. (1l:10 a.m.) FATIGUE-CRACK PROPAGATION IN CERAMICS: LONG vs. SMALL CRACK BEHAVIOR:· R H Dauskardt. A. A. Steffen and R. O. Ritchie, Center for Advanced Materials, Lawrence Berkeley Laboratory, University of California, Berkeley, CA 94720. Recent studies in engineering ceramics have provided persuasive evidence of degradation and premature failure under cyclic loading. . In this study, fatiguecrack propagation behavior is investigated in Mg-PSZ, Ce-TZP and carboncarbon composites. Where behavior is examined in compact tension specimens containing preexisting through-thickness long (>2mm) cracks, crack-growth rates are found to be power-law dependent on the stress-intensity range, AK, and to show evidence of a threshold stress-intensity range, AKTH, approximately equal to SO% KIc' Conversely, for naturally-occurring small (-1 to 100 I'm) surface cracks in unnotched specimens, crack-growth behavior is observed at AK levels some 2 to 3 times smaller than the long-crack threshold. similar to metals. Mechanisms for such behavior are examined. ·Supported by the U.S. Department of Energy, Basic Energy Sciences.
(ll :30 a.m.) THE EFFECT OF CoCrAlY COATING ON THE THERlD!ECHANICAL FATIGUE BEHAVIOR OF RENE 80 , H. N. Jones and K. Sadananda, Materials Science and Technology Division, Naval Research Laboratory Washington, DC 20375-5000. When a component of a gas turbine is given a coating to increase its resistance to hot corrosion its mechanical behavior under cyclic thermal and load histories is altered. The coating can. depending on the operating environaent of the part, be considered to contribute nothing to the load carrying capacity of the component, to increase it due to the additional material added to the part, or to decrease it due to the tendency of the coating to crack. The thermomechanical fatigue behavior of coated and uncoated parts, where both the load and the temperature are cycled, can be compared under either an equiVAlent peak stress or an equivalent peak load basis. Additionally, the particular combination of load and tellperature cycles :imposed on the part will alter the effect of the coating on its fatigue life. The fatigue life of coated and uncoated Rene 80 under full reverse load, in phase and out of phase load-temperature cycles, and other combinations is presented. Also, the effect of the heating rate to peak temperature is considered.
Annual Meeting-I5
FRACTURE I TMS General Abstract Session Monday AM February 19, 1990
Room: Plaza C Inn at the Park
Session Chairman: J.G. Cowie, U.S. Army Materials Technology Laboratory, Watertown, MA 02172.
304, AISI 310 and AlSI 304L are examined. Thennomechanical processing of 304L, the least stable of the materials tested, was used to pre-transfonn and to induce texture which can affect the subsequent transformation products and thus fracture behavior. The transformation which occurs around the crack tip during the he tests was found to have a significant effect on the fractu~ toughness of the material. *This work is supported by the Director, Office of Energy Research, Office of Fusion Energy, Development and Technology Division of the U. S. Depamnent of Energy under Contract No. DE-AC03-76SFOOO98. (9:50 a.m . )
CRACK ARREST IN ULTRAHIGH STRENGTH STEELS,
M Azrin. J .G. Cowie and E.B. Kula, U .S. Army Materials Technology Laboratory, Watertown, Massachusetts 021720001.
(8 : 30 a.m. ) A NOVEL METHOD TO ESTIMATE PLANE·STRAIN FRACTURE TOUGHNESS OF METALS, E. I. Stromswold and P J Quesnel Mechanical Engineering Department, University of Rochester, Rochester, New York 14627 Chevron notches have been introduced in both short rod and bend specimens by prior investigators to determine the plane· strain fracture toughness, KIC • of engineering materials. Their work has shown that toughness estimates can be made with maximum load tests on these samples. Recently. most work has centered on short rod samples, perhaps due to the smaller volume of material required to make the test samples. There are several benefits to bend specimens. however. including reduced machining time. particularly when "L· T" measurements are needed. A stand·alone tool has been developed to implement this method of testing. The tool is economical to build and simple to use. It should lead to an increased use of fracture mechanics concepts in engineering design by making the numerical values of fracture toughness easier to obtain in the commercial/engineering/quality assurance environment. Operating principles and design considerations for the toughness tool will be presented and discussed.
(8:50 a.m.)
COMPARISON OF MODE I (TENSION) AND MODE II (SHEAR) FRAcruRE OF BOTH ULTRAHIGH STRENGTH STEEL AND HIGH STRENGTH ALUMINUM ALLOYS, J.G. Cowie. U.S. Army Materials Technology Laboratory, Watertown, MA 02172, and F.R. Tuler, Worcester Polytechnic Institute, Worcester, MA 01609.
n
Experimental results of modes I and fracture toughness tests are compared and contrasted for both ultrahigh strength 4340 steel and high strength 7075 and 6061 aluminum alloys. Hydrostatic stress field variations between fracture modes allowed for differences in the critical stress intensity factors (Kc). The aluminum alloys experienced the least amount of differences in Kc between fracture modes compared with that of the steel (i.e. Kllc/Klc ratio equaled 1.4 for the aluminum alloys and 2.5 for the steel). In mode I fracture, the high hydrostatic tensile stress field induced early microvoid nucleation leading to fracture, whereas, hydrostatic compression helps delay microvoid nucleation during mode fraCture permitting greater applied loads. These facts explain why an unrestrained shear crack quickly orients itself into a tensile crack. the preferred mode of crack propagation. In the case of the 4340 steel, both Kic and Kllc could quantitatively be related to the microstructure through the secondphase particle size and distribution.
n
( 9 : 10 a.m. )
FRACTURE TOUGHNESS BERAVIOR OF D6AC STEEL. P . K. Liaw, II. G. Peck, and G. E. Rudd, Westinghouse Science i Technology Center, Pittsburgh, PA 15235; and D. W. Sutherland, Thiokol Corporation, Brigham City, UT 84302. Fracture toughness properties (Jrc and J R curves) were developed for the D6AC steel used in the case of . tHe Space Shuttle booster rockets. Tests were performed over the temperature range of -20'F to 130'F. At temperatures below room temperature, J R curve specimens exhibited some stable crack extension before failing by unstable crack extension. The extent of stable crack extension varies with the test temperature, but the form of the J curve was found to be independent of temperature. Noticeable R to heat variations in fracture toughness were observed heat despite close control of chemistries and heat treatment parameters. The data from different heats can be correlated using two heat dependent parameters, a characteristic temperature and the fracture toughness measured at the characteristic temperature. ( 9:30 a.m.)
FRACTURE OF METASTABLE AUSTENmC STEELS IN A CRYOGENIC HIGH MAGNETIC FIELD ENVIRONMENT"': J W Chan and 1. W. Morris, Jr. Center for Advanced Materials, Lawrence Berkeley Laboratory, 1 Cyclotron Rd., Berkeley, CA 94720. Metastable austenitic stainless steels have been used as structural materials for high-field superconducting magnets. The effects of a 4.2 K high magnetic field environment on the fracture behavior of metastable austenite has not been well documented. The fracture behavior, at 4.2 K in magnetic fields up to 8 T, of AISI
Preliminary results are presented on a test procedure to assess the plate cracking sensitivity of high strength steel plate. The procedure involves the introduction of a large flaw, or "crack starter," at the center of a plate. and then impacting the center of the opposite face with a soft. bluntnosed projectile. The impact tests are performed over a range of temperatures to yield a plate shatter transition temperature. The transition temperature is simply the highest test temperature at which extensive plate cracking occurs. The influence of test parameters and correlation with crack arrest fracture toughness over the same temperature range are currently being studied. (10 :10
a.m.)
SIGMA PHASE EMBRI1TI..EMENT OF 310 L STAINLESS STEEL : A.V. Kington, F W Noble. Department of Materials Science and Engineering, University of Liverpool, P.O. Box 147, Liverpool, L69 3BX, U .K. A study has been made of the effect of systematic variations in a-phase content on the ductility and toughness of a wrought 310 stainless steel. The steel initially contained 1.1 % /i-ferrite in the form of elongated stringers. On holding at 700 or 750·C individual stringers of /i-ferrite rapidly transfonned to a-phase with no apparent alteration in their size Of geometry. The number of fully transfonned stringers increased progressively with time at temperature until, after only I h, virtually every stringer had transfonned. The variation of mechanical properties with overall degree of transformation has been studied using tensile and Charpy tests backed up by post-test metallography on defonned specimens. The presence of a-phase results in large. specimenorientation dependent reductions in Charpy energy and ducti1ity. The results are discussed in terms of the fracture characteristics of the a -phase stringers and the associated void nucleation and growth process which lead to ductile fracture. 00:30 a . m.)
DAMAGE TOLERANCE ASSESSMENTS OF AUSTENITIC FE-AL-MN ALLOYS : Shen-Chih Lee & Yin-Bean Chang, Dept. of Materials Eng'g, latung Inst. of Technology, Taipei, TAIWAN 10451 This research investigated the effects of Carbon, Aluminum, and Manganese contents on the fracture toughness (KIC) and crack growth rate (da/dN) of austenitic Fe-Al-Mn alloys. The valid KIC values obtained were in the range of 49 . 6 to 148.0 MpaJm at ambient temperature. At 150'C, most of the alloys were so tough that valid KIC can not be obtained by using compact-tension specin~ns of 25.4 mm in thickness. The crack growth exponents n of Paris' Equation were small in values (n=2.4 to 6.6) indicating excellent crack growth resistance of these alloys. The effect of temperature on n values was found to varied according to the type of alloys analyzed. A Z-value (Z=Mn/Al ratio) concept was proposed to clearify the influences of Mn &Al on the properties. When Z value is small, the tendency to have ferrite in matrix increases and the material becomes embrittled . When Z value is large, fully austenitic matri x can be obtained and the material becomes Inore ductil e, the abi 1ity of crack growth resi stance becomes better, and the fracture toughness also improved.
l~Annual
Meeting
mGH PERFORMANCE COPPER-BASE ALLOYS I
tor. consolidation of the layered powders by CIP and hot extru-
sion, and deformation processing of the bars by swaging and wire
drawing.
The Cu-19.6 w/o Nb alloy powder was produced by a high
pressure gas atomization (HPGA) approach using a novel melting practice to permit a 2000°C pouring temperature. Consolidation
of the Cu-Nb powder utilized CIP and HIP.
Co-sponsored by the TMS Structural Materials and Powder Metallurgy Committees MondayAM February 19, 1990
Room: ParkA Inn at the Park
The consolidation and
deformation processing temperatures were controlled to minimize
solid solubility and coarsening effects and. thereby. to retain elevated strength and conductivity levels. Results of the characterization of these properties will be related to the alloy microstructures as a function of processing history. *Operated for the U.S. Department of Energy by Iowa State University under contract No. W-7405-Eng-82. (10 :30 a.m.)
Session Chainnen: R.E. Lewis, Lockheed Missiles & Space Co., Inc., Dept. 93-10, Bldg. 204, 3251 Hanover Street, Palo Alto, CA 94304; I.E. Anderson, 122 Metals Development Bldg., Ames Laboratory, Iowa State University, Ames, IA 50011.
PROCESSING AND CHARACTERIZATION OF ELEMENTALLY BLENDED POWDER METALLURGY Cu- Fe AND Cu-Nb ALLOYS: W. A. Spitzig, Ames Laboratory*, Iowa State University, Ames, IA 50010, Y. S. Go, llro-yeup Muan-kun Chunnam, South Korea and C. L. Trybus. Idaho National Engineering Laboratory EG&G Idaho, Idaho Falls, ID 83415 The feasibility of producing high strength/high conductivity deformation processed eu-base composites using powder metallurgy
(P/M) processing procedures is explored because P/M offers a
(8:30 a.m.)
much wider range of compositions for the composites as compared
to traditional ingot metallurgy. In this study Cu-20vol% Fe and Cu-20vol% Nb composites were produced by P/M processing followed
AEROSPACE STRUCTURAL REQUIREMENTS FOR HIGH PERFORMANCE COPPER BASE ALLOYS, Dr. Neil E. Paton, Rockwell International Corporation, Rocketdyne Division, 6633 Canoga Avenue, Canoga Park, CA 91303.
by axisymmetric deformation processing to true strains up to ahout 9. The strengths of both composites increased with
Copper base alloys have been used in high performance rocket propulsion systems for a number of years, on account of their good combination of strength and thermal conductivity. Examples of such applications will be given, and the key structural and physical requirements enumerated. Recent advances in the properties of copper base alloys and composites have provided the opportunity for significant improvements in both mechanical and thermal properties. The potential for these app~¢aches will be summarized, and some of the new applications being opened up by these advanced copper base systems discussed.
conductivity of Cu-20% Fe was lower than that of Cu-20% Nb because of the greater solubility of Fe. as compared to Nb. in
(9:00 a.m.)
FABRICATION OF COMPOSITE WIRES FROM GAS ATOMIZED Cu-Cr POWDER, J C Malzahn Kampe, K.P. Cooper*, and J.D. Ayers, Materials Science and Technology Division, Naval Research Laboratory, Washington, DC 20375 * Geo-Centers, Inc., Fort Washington, MD 20744 Efforts to enhance the refinement of composite microstructures generated by extensive deformation processing have been directed toward refinement of starting microstructures. Along these lines, gas atomized Cu-Cr powder has been consolidated, swaged, and drawn into fibrous composite wires. In this presentation, the Cr morphological evolution from fine dispersoids within powder particles to aligned fibers within the wire will be followed through the various stages of processing. The microstructural information will be supplemented with room temperature strengths and resistivities for both "as processed" and annealed wires. (9:30 a.m.)
Cu·15v/oNb MICROCOMPOSITE MICROSTRUCTURE·PROPERTY RELATIONSHIPS AFTER THERMAL AGING: f.Jl....IS.rlll. J. A. Fint. C. Rhodes'. J. L. Yuen. Rocketdyne Division of Rockwell. 6633 Canoga Ave .. Canoga Park. CA 91303; 'Rockwell Science Center. 1049 Camino Dos Rios. Thousand Oaks. CA 91360. The microcomposne studied in this program is a ductile two phase mixture consisting of a copper matrix containing high aspect ratio second phase filaments of niobium formed by extensive deformation. This material is of interest for high·strength. high thermal-conductivny applications at elevated temperatures. For elevated temperature service. the thermal stability of the in-s~u formed filaments against coarsening requires evaluation. The purpose of this work is to evaluate the effect of thermal aging on the mechanical properties and correlate these wnh microstructural changes that occurred. The material used for this study was Cu-15v/oNb prepared by conventional casting methods and cold rolled to 0.5 mm sheet. Tensile specimens were fabricated and aged at five different temperatures (150. 315. 480. 650 & 927°C) for various lengths of time. The microstructural changes were quantified using scanning electron microscopy. (10:00 a.m.)
NOVEL TECHNIQUES FOR PRODUCTION OF FINE PARTICULATE MICROSTRUCTURES IN COPPER-REFRACTORY METAL ALLOYS: J.D. Verhoeven. B.K. Lograsso. P.M. Berge. W.A. Spitzig. ~ Anderson. and E.D. Gibson. Ames Laboratory*. Iowa State University, Ames, IA 50011. High performance powder metallurgy alloys of Cu. including Cu-Fe
and Cu-Nb. have been produced by innovative procedures involving either solid state processing of elemental alloy constituents or atomization processing of molten alloys. The eu-Fe alloys were produced by mechanical layering of eu and Fe powders in an attri-
increasing degree of deformation processing. Comparison of the two composites showed that Cu-20% Fe was stronger after an equivalent degree of deformation processing. However. the
the
eu
matrix.
*Operated for the U. S. Department of Energy by Iowa State University under contract No. W-7405-Eng-82. (11:00 a.m.) MICROSTRUCTURAL DEVELOPMENT IN MECHANICALLY ALLOYED CU-NB POWDER ~ ~ ~ and T. H. Courtney, Department of Materials Science, University of Virginia, Charlottesville, Va. 22901 Copper and niobium powders of differing volume fractions (ca. 15 and 50 vol.' Nb) have been mechanically alloyed using an attritor. We have been able to deduce a time constant for the process by determining the relative fractions of the powder particles that are elemental and "alloyed" as these depend on the time of milling. Knowledge of this time constant combined with evaluation of interphase spacing with processing time enables us to determine the strain imparted to a particle in a typical collision. These parameters will be compared to those expected from recently developed models of the HA process. We will also report on densification and structural changes that take place during subsequent hot isostatic pressing of the alloyed powder. Particular attention is given to the microstructural coarsening that takes place during this consolidation treatment.
supported by The Army Research Office.
This work was
(11 :30 a.m.)
MICROSTRUcruRAL EVOLUTION IN RAPIDLY SOLIDIFIED Cu-Cr-Ag ALLOYS: K.P. Cooper, Geo-Centers, Inc., Fort Washington, MD 20744, J.D. Ayers, J.C. Malzahn Kampe, J. Feng, Materials Science and Technology Division, Naval Research Lab., Washington, DC 20375, and I.E. Locci, NASA Lewis Research Center, Cleveland, OH 44135. Rapidly solidified samples of Cu-Cr-Ag alloys were prepared by high pressure gas atomization and by melt spinning. Both quenching processes yielded microstructures which showed evidence of the fonnation of two inuniscible liquids prior to solidification. To evaluate the effect of high temperatures on the stability of the as-solidified microstructures, samples prepared by HIP consolidation of the atomized powder and heat treated samples of the melt-spun ribbons were evaluated in the SEM. Results showed that the shape of the Crrich phase in all cases was essentially spheroidal, while the particle size depended upon the processing technique. 1be CHich phase in the melt-spun ribbon readily coarsened at.higher temperatures, but the coarser Cr particles in the atomized powders ripened less quickly when HIPed at comparable temperatures. Implications of these results for possible high temperature applications will be discussed.
Annual Meeting-17
HUME-ROTHERY MEMORIAL SYMPOSIUM I: High-Tc Superconductors and Related Metallic Oxides (A)
Room: Orangewood I Quality Hotel
Northeastern University I
Barnett
The sensitivity to composition,
the absence of a predictive theory. the complex crystal chemistry and absence of phase diagrams are obstacles in the search. A rational approach involves appropriate sections composition
and
treatment
parameter
rev iew
the
striking
superconducting
space,
and
SUPERCONDUCTORS?
Doping La2Cu04 wi th Sr. Ba or Ca produces a metal- insulator transition. wi th the insulating phase being antiferromagnetic and the metallic phase a high-temperature superconductor. Using a model similar to the Friedel model of Hume-Rothery alloy phase formation. I find that this transition is not smooth. but is character i zed by a miscibility gap. The
High-Tc superconductivity appears in oxides with compositions precariously perched near regions of structural instability
through
we
Northeastern University , Boston . MA 02115 .
CRYSTAL CHEMICAL. EMPIRICAL AND SYSTEMATIC APPROACHES IN THE SEARCH FOR HIGH-T c SUPERCONDUCTORS. B C Giessen, Department
and electronic transitions .
talk ,
R.S . Markiewi cz , Department of Physics and Barnett Institute ,
(8 : 35 a. m.)
and R.S. Markiewicz. Department of Physics and Institute. Northeastern University , Boston. MA 02115.
this
(11 :00 a.m.) ELECTRONIC PHASE SEPARATION IN THE HIGH-T c
INTRODUCTION OF HUME-ROTHERY AWARDEE
of Chemistry and Barnett Institute,
The new elec tron-
magnetic properties of the new elec tron-doped materials . Research suppor t ed by US DOE Grant No. DE-FG03 - 86ER45 230 and US NSF-Low Temperature Physics - Grant No. DMR84-ll839.
Session Chainnen: C.C. Koch, Materials Science and Engineering Dept., North Carolina State University, Raleigh, NC 27695; R.J. Blaugher, Intennagnetics General Corp., Guilderland, NY 12084. (8 : 30 a.m.)
for finding new high-Tc superconductors .
doped materials have the chemical formula Ln2_x~Cu04_ (LnPro Nd, Sm. Eu. M - Ce; and Ln - Pro Nd, Sm . M - Th) a6d have Tc as high as -25 K for x - 0.15 and y - 0.02. A related compound with the same crystal structure . Nd2Cu04_x_ Fx' has also been found to display superconductivity with Tc y- 25 K. In
Sponsored by the TMS Alloy Phases Committee MondayAM February 19, 1990
important implications regarding theory as we ll as strategies
trace
maximum Tc is associated with a pure phase at a fixed Sr concentration . Away from th i s doping , the material is two· phase, and the superconducting properties are degraded .
Evidenc e for this effect . bo t h in La2_xSrxCu04 and in other high-Tc superconductors. will be presented; a simplified picture of the phase separation will be given, showing the connection with the formation of Hwne-Rothery phases.
Research supported by DOD-SOlO Intermagnetics General Corp . (11 : 30 a.m , ) RARE EARTH NICKELATES Honig. Department of Lafayette . IN 47907.
under
subcontract
A NEW CLASS OF SUPERCONDUCTORS. Chemistry, Purdue University.
from
~
West
quantity or surface layer phase detection, a search for metastability and realistic economic considerations. Useful concepts include crystal chemical analogy , modeling HTSC's as intercalation compounds, a unifying model based on the DOS and interlayer coupling parameters, and, experimentally. magnetic field alignment for phase selection. Research supported by DOD- SDIO under subcontract from Intermagnetics General Corp. and by E.I. DuPont de Nemours and Company. Inc .
Compounds belonging to the generic class La 2 _xSrxNi04+6 have recently been reported l , 2,3 as exhibiting superconductivity characteristics . The problems of synthesizing single crystal materials of controlled oxygen stoichiometry. and the
(9 : 10 a.m.) NEW METAL OXIDES WITH COPPER. M. Sabat . J. Wiley. J. Vaughey
1. Z. Kakol . J. Spalek, and J.M. Honig, J . Solid State Chem . li . 288 (1989).
and K.
PoepRelmeier.
Department
University. Evanston. IL 60208 . The
number
of verified oxide
of
Chemistry I
superconductors
Northwestern and
increased dramatically over the last three years.
Te
have
Yet the
synthesis of new compounds that support the superconducting state remains one of the most challenging problems . Studies on the partial replacement of copper with post transition
problems of carrying out the necessary measurements will be
discussed . Both the normal and superconducting properties of these materials will be described and compared to the corresponding characteristi c s of the analogous cuprates.
2. Z. Kakol. J . Spa1ek . and J .M . Honig, Solid State Conunun . 283 (1989).
n.
3. J. Spalek. Z. Kakol. and J.M. Honig, Solid State Commun. 511 ( 1989) .
n.
possible to retain pl~nar copper-oxygen (Cu04/2 )" structures with both aluminum (AI +) and gallium (Ga 3+).
HYDROMETALLURGY AND AQUEOUS PROCESSING I: Leaching of Primary and Secondary Materials
(9:40 a.m.) NEW AND MODIFIED CRYSTAL STRUCTURES IN SYSTEMS RELA.TED TO HIGH-T c MATERIALS. H. Steinfink. J . S. Swinnea. J.S. Kim and
Sponsored by the TMSlEPD Aqueous Processing Committee
metals will be di scussed.
Single crystals have been grown and
their structures determined.
These studies reveal that it is
J . Y. Lee, Center for Materials Sc ience and Engineering and Department of Chemical Engineering . The University of Texas at
Austin. Austin . TX 78712. The attempts
to modify the phys i cal properties
of high-Tc
compounds by partial or complete replacement of elements in the basic structure t y pes leads to complex phase relations
among the products of the reac t ions.
Changes i n the basic
perovskite motif occur to produce related structures such as
brownmillerite. We have studied by single crystal X-ray diffraction procedures the structural changes that occur upon replacing Tl by Pb in TISr2YCu207(12l2 type) and in TlSr2Cu05 (121 type) . The replacement of Y by Lnl_xC~ in Pb 1212 occurs to a limited ex tent and La8_xSrxCu8020 and Sr5_xPb3+xCllv012_6 phases form .
The crystal structures and phase relation~ will
(1 0: 10 a.m. )
BREAK
be discussed .
to
1989 .
the
Room: Salon 1 Marriott
Session Chainnen: Garry W. Warren, University of Alabama, Department of Metallurgical Engineering, P.O. Box G, Tuscaloosa, AL 35487-9644; David B. Dreisinger, University of British Columbia, Metallurgical Engineering Department, 309-6350 Stores Road, Vancouver, V6T 1W5, British Columbia, Canada (8:30 a.m.)
(10:30 a.m.) ELECTRON-DOPED HIGH-T c SUPERCONDUCTORS . Cal i fornia. San Diego . La Jolla . CA 92093. Prior
Monday AM February 19, 1990
prevail i ng view
was
SELECTIVE OXIDATION OF URANIUM IN IN SITU LEACHING,
...I:IM~BI!.o.._M!:!a!!.p~l~e
that
the
.
U.
charge
c arriers involved in high-Tc coppe r oxide superconductors were
holes moving through conducting Cu0 2 pl anes. the basic building blocks in all presently known oxides with Tc > -30 K. In new superconductors recently discovered in Japan and the USA the charge carriers inv olved in the s u perconductivity appear to be electrons in the Cu02 planes. Th i s could hav e
T. Y. Yan, Kobil Research and Development Corporation, Central Research Laboratory, P. O. Box 1025, Princeton, NJ 085431025. 02' NaCI0 3 , B202, NaOCI and Ce+4 have been exaained as oxidants in an alkaline carbonate system for uranium in in situ leaching. The South Texas and New Kexico ores contained up to 2,6 meqfg of reducing compounds which consume oxidants in the leading operation. Leaching rates did not increase linearly with the oxidation potential of the oxidants. A mild oxidant
IS-Annual Meeting can be used to maximise uranium oxidation selectively and oxidant efficiency. 0 can oxidize most of the uranium mineral with an adequate rate and high oxidant efficiency.
(10:30 a.m.) HYDROTHERMAL SULFIDATION OF PLUMBOJAROSITE FOR RECOVERY OF LEAD AND SILVER: K.P.V. Lei and N.P. Gallagher, Reno Research Center, U.S. Bureau of Mines, 1605 Evans Ave, Reno, NV 89512-2295.
"'(8:50 a.m.)
KINETICS AND MECHANISM OF THE REDUCTIVE LEACHING OF MANGANESE ORES BY IRON PYRITES IN MILD ACIDIC SOLUTION ~Parida, B.B. Nayak, K.K. Rao & S.B. Rao Regional Research Laboratory,Bhubaneswar-751013
The kinetics of dissolution of inert manganese dioxide in dilute sulphuric acid was investigated using iron pyrites as reductant. The dissolution rate was dependent on rate of agitation, temperature, amounts of pyrite, concentration of sulphuric
acid,
particle
size,
and
specific
surface
area
of
manganese ore. The order of reaction with respect to the above mentioned factors were also calculated to tind out the rate
expression. The activation energy values 31. 56 KJ lmole, 30.61 KJ lmole as determined by the diffusion model and chemical controlled processes respectively indicate that the reaction is diffusion and chemical controlled. The data were found in good agreement with a mixed model where the rate of diffusion through the Nernst boundary layer is of the same order of magnitude
as the rate of chemical reaction at the interface.
(9: 10 a.m.)
AQUEOUS OXIDATION OF CHALCOPYRITE USING MANGANESE DIOXIDE IN DILUTE SULPHURIC ACID SOULTION: K.M.PARIDA AND S.B.RAO. Regional Research Laboratory,Bhubanesvar751013, India. The dissolution of chalcopyrite concentrate in sulphuric acid solution using manganese dioxide as oxidising agent vas studied. The dissolution of copper vas measured varying the parameters like:time,temperature, manganese dioxide concentration, particle size,surface area, pulp density vith respect to chalcopyrite,agitation rate, acid concentration etc. Extraction of copper increased vith decreasing pH, decreasing pulp density and particle size. The optimum temperature range is 75-100 C. Further it was observed that the copper dissolution is directly proportional to specific surface area of chalcopyrite. Details of the reactare discussed and a redox mechanism proposed. (9:30 a.m.) SILVER ION CATALYSIS IN LEACHING OF NICKEL MATTE: Jln Zuomel, Wen Wen, Dept. of Metal Materials, Chengdu University of Science and Technology, Chengdu, Sichuan, P. R. China
'!he kinetics of dissoluti c of niei-. -·1 mat--- in the presence of dioxide with s i l ver ions was inve:>tigated. '!he c,c .alyCj , silver ions enl12 :.:ed the leaching rate l.lp to a {;'I ctor of 17 . The effects of stirring, terrperature, particle size, concentrations of Mno:! ' silver and ferric iOrB, anri the ways of feeding the raw ma cerials were studied. Solid residues at various levels of nickel extraction \\ere e:-:"'nined by SEM, X-ray diffraction, electron microprobe and chemical analysis. TItese resul ' indicated that the dissolution reaction was topochernical ''lI'-'1 fit:, -, surface ,'eaction control !rodel. The activation energy wa ~ calcUl.c.\..~d to t:e 39.15 KJ/IlDl. '!he ~ eaching rate of Ni S was 1 2 indepencent with stirring rate, but the reaction l'ate cOnStant W,L; inversely proportional to the initial particle diarreter. The order~ of the reacti~ race are 0.43, 0.69 anQ 0.33 with respect to lVJ, Mn0 and Fe concentration respectively . Thus a kinetic 2 rate equation was develo;:ed. Moreover, ".'8 prc;-,se a rrechcr.ism of silver catalyzed leaching of Ni S2 ·
rna- ganese
3
(9:50 a.m.) Break
(10: 10 a.m.)
PRESSURE LEACHING OF GALENA CONCENTRATES TO RECOVER LEAD METAL AND ELEMENTAL SULFUR: Agnes Y. Lee, Ann M. Wethington, and Ernest R. Cole, Jr., Bureau of Mines, Rolla Research Center, P.O. Box 2BO, Rolla, MO 65401 As part of the Bureau of Mines program to improve base metal leach systems, pressure leaching of galena (PbS) concentrates in fluosilicic acid (H2SiFsl solution was investigated. Lead extraction was 93 pct when PbS was leached in H2SiFs at 95' C with 200-psig O2 pressure for 40 min, using Fe(N0:V3 as a catalyst. The effects of several catalysts, O2 pressure, time, and acid concentration on leaching were determined. The purified leach solution was electrowon to recover high·purity Pb metal. The H2SiF s acid used in leaching and electrowinning can be recycled continuously. A flow diagram was proposed based on this investigation. Elemental sulfur was recovered from the leach residue by solvent extraction. A method was also developed for treating the final residue to segregate the important byproduct elements such as Ag, Cu, and Zn and enable their recovery by standard commercial processes.
The U.S. Bureau of Mines investigated a hydrothermal sulfidation (HTS) procedure to recover the lead and silver from plumbojarosite. The plumbojarosite, generated during pressureoxidative leaching of zinc concentrate in a commercial operation, contained, in percent, 25.4 Pb, 18.8 Fe, 1.4 Zn, 29.5 50 42-, 3.2 SO, 13.4 total sulfur, and 7.4 oz/ton Ag. ~e HTS was conducted in an autoclave and converted the plumboJarosite into amorphous iron oxide and liberated the lead and silver as sulfides. The best operating conditions were using 50 g of plumbojarosite and 7% SO (2.04 g SO added) in 500 ml of 2.0 molll NH3' lS0 0 C, and 1 h. The sulfidation product contained, in percent, 37 Pb, 28 Fe, 25042-, 7 totalS, and 11.2 ozlton Ag. Chloride leaching of the sulfidation product extracted 98 to 99% Pb and 97 to 98% Ag from the sulfidation product. (10:50 a.m.) COPPER TELLURIDE RECOVERY FROM RAW SUME AT COPPER REFINERY : Tatsuya Mutoh, Junji Konishi, Shosaku Hayashi, Onahama Smelting and Refining Co., Ltd., Nagisa, Onahama, Iwaki, Fukushima, Japan Te and Cu in raw slime have been recovered as CU2Te and electrowon copper. The slime is decopperised with sulfuric acid solution in conventional leaching tanks. This loaded solution containing 451/t of Cu and lilt of Te is introduced to CU2Te recovering reactor. The tail solution from the reactor containing less than 20ppm of Te is recycled directly to the circulation tanks of copper electrolysis. Se or other elements contamination into CU2Te is less than 0.1% without any purification. (11: 10 a.m.) DISSOLUTION OF PHOTOCONDUCTIVE Se-As LAYERS FROM ALUMINIUM PH01OCOPIER DRUMS WITH HYDRCGEN PEROXIDE 5OLIITIONS: Luiz A. Teixeira*, Rayma'ld E. Reber**, and Lidia Y. 'Tavares* - - - ~ic University of Rio de Janeiro, Dept. Materials Science & Metallurgy, CP.38008 - Rio de Janeiro - 22453 - RJ Brazil ** Peroxidos do Brasil Ltda. - INTEROX, Av. Paulista, 2001-14 S. Paulo - 01311 - SP -Brazil A process is described whereby worn, defective or crystallized photoconductive Se-As layers can be stripped off from aluminium photocopier drums, using hydrogen peroxide solutions under conditions such that the aluminium substrate is preserved from chemical attack. The process allows the stripped drum to be recoated with a new photoconductive layer, and to be put into use again.
1990 LEAD·ZINC AND RELATED METALS SYMPOSIUM II: Lead Smelting Practice Sponsored by the TMS-EPD Lead-Zinc-Tin Committee Monday AM February 19, 1990
Room: Salon A & B Marriott
Session Chainnen: V. Ramachandran, Asarco Technical Service Center, 3422 South 700 West, Salt Lake City, UT 84119; J. Fewings, Mount Iss Mines Limited, Mt. IsB, Queensland 4825, Australia (8:30 a.m.) THE FIRST COKKERCIAL-SCALE QSL - PLANT - RESULTS OF THE COKKISSIOWIKG, Kanfred Scbaidt and P. Fischer, Lurgi GabR, LufgiAllee 5, 6000 Frankfurt/Kain, Geraany. Lurgi GmbH bal developed a radically new lead smelting technology, whicb was deaonstrated on a 30,000 tpy scale at Berzelius Ketallhutten GabH between 1981 aDd 1986. DesigDed to treat all grade. of lead conceDtrates as well as coaplex secondaries, the Dovel QSL-Proeess applied one bottom-blown, refractory-lined 10Dg reactor aDd comprises two basic steps; Continuous oxygen .aelting of tbe green cbarge is followed by continuous carbotberaic slag raductioD. Tbe QSL-tecbnology has inherent proees. advantages; low investaent aDd operating costs, flexibil-
Annual Meeting-19 ity, and environ.ental acceptance. In 1986 Cominco Ltd. decided to erect a 120,000 tpy QSL plant. The designed QSL reactor has a total length of approx. 41 m, and working diameters of 3.74 a for the oxidation section and 3.24 m for the reduction section. Separate gas off-takes on both sections will enable to reduce the sulfur dioxide gas volume, and to recover a aixed zinc-lead oxide fume. Operation of the QSL reactor will start end of September 1989. (9:00 a.m.)
THE CONTINUOUS ISASKELT LEAD PROCESS; S.P. Katthew, G.R. KcKean, R.L. Player and K.E. Ramus, Kount Isa Kines Limited, Kount Isa, Queensland, Australia. The ISASKELT Lead Process is a continuous two stage process for the production of crude lead from concentrate. It is based on the use of a top entry submerged Sirosmelt land to produce turbulent baths in which high intensity smelting or reduction reactions can occur. The process involves oxidizing the concentrates in the first stage to produce a high lead slag. This slag is tapped continuously and transferred down a launder to a second furnace where it is reduced with coal. Crude lead and discard slag are tapped continuously from this furnace through a single taphole and separated in a conventional forehearth. Process and engineering design work for a commercial scale Lead ISASKELT plant designed to produce 60000 tpa crude lead has been completed. This plant is planned to be operational in 1991. (9:30 a.m.)
THE KIVCET FURNACE CONSTRUCTION AT PORTOVESKE, A. Perillo, Techmetal - Ketallurgical Technologies, Joint Task Force Snamprogetti/N. Samim, P. zza L, Cerva, 7, 00143 Roma, Italy. The Kivcet furnace can convert continuously lead concentrates residues , secondary lead bearing materials into a lead bullio~ free of sulphur' a discardable slag. The lead recovery can be up to 99\. The experience of about 3 years at Portovesme demonstrates the capability of the furnace to smelt a very high range of lead materials without interruption of the operation. The operating costs are very competitive in comparison to the traditional technology. The work place & environment are clean and new tight regulations for the hygiene control are met. The Kivcet furnace can be considered as consisting of 3 parts (reaction shaft, waste gas shaft & electric furnace) connected with a common hearth. Refractory wear in the reaction shaft with typical pattern from watercooling soon has been stabilized & since maintained. No deformation of the furnace shell & no leakage of metal or slag through the bottom has appeared after 3 years' operation. (10: 00
a.m.)
BREAK
(11 :20 a.m.)
KODERN PRACTICE IN THE OPERATION OF THE PARKES PROCESS AND THE USE OF A COKPUTER KODEL AS AN AID TO PROCESS OPTIKIZATION, R.P. Knight, KIK Technology Karketing Ltd., Botany Road, Northfleet:" Kent DAll 9BG, England. The Parkes Process is an important step in the pyrometallurgical refining of lead bullion. The metallurgical efficiency of the process can have a significant impact on the overall refining costs, particularly with regard to zinc consumption and the costs of working up the silver-rich crusts. Several authors have used mathematical or graphical methods to calculate the minimum required zinc additions and to study the merits of alternative operating procedures but these methods are of necessity very cumbersome and are only able to deal with relatively si.ple systems. The paper explains how a computer model was developed which simulates the Parkes Process accurately and can predict the effects of changes made in operating methods. Examples are given of some of the findings and their relevance to the practical conduct of the process is discussed.
1990 LEAD-ZINC AND RELATED METALS
SYMPOSIUM llI: Precious Metals Recovery in the Zinc Industry A Sponsored by the TMS-EPD Lead-Zinc-Tin Committee Monday AM February 19, 1990
Room: Salon C & D Marriott
Session Chairmen: E. Van Den Neste, Metallurgie Hoboken-Overpelt, A. Greinerstraat 14, B-2710 Hoboken, Belgium; J.E. Dutrizac, CANMET, 555 Booth St., Ottawa, Ontario KIA OGI Canada (8:30 a.m.)
(10:20 a.m.)
THE NEW COKINCO LEAD SKELTER AT TRAIL, K.J. Walker and D.R. Reynolds, Cominco Ketals, Trail, B.C., Canada VIR 4L8; and G. W. Lee, Cominco Engineering Services Ltd. The lead smelter at Cominco's operations at Trail, B.C. has been undergoing a modernization program since 1986. Lurgi GmbH agreed to construct a lead smelter using a QSL process. Cominco Engineering Services Ltd. was contracted to provide overall
project management.
preheat was installed. Ventilation capacity has also been drastically increased. A 7 KVA electric furnace cleans the slag from both blast furnaces to improve metal recovery.
The new slDel ter replaces the existing sin-
terblast furnace plant and will have a capacity of 160,000 tonnes per year of refined lead. This new process results in significant operating cost reductions and environmental improvements. Start-up of the new plant is scheduled for the fall of 1989. The metallurgy and operating parameters, important to the design of the new smelter, are described in this paper. A plant description for the following major areas are given: pressure filter operation for zinc leach residues, the QSL reactor and boiler, process gas treatment, oxygen softening and thallium removal. Slag from the QSL reactor will be treated in existing slag furnaces until a second phase of the modernization is completed in the 1990's. (10:50 a.m.)
THE HOBOKEN COKPLEX LEAD-COPPER SKELTER OF KETALLURGIE HOBOKENOVERPELT, K.P. Van Negen, R. Kaes and C. Coekelberas, Ketallurgie Hoboken-Overpelt, A. Greinerstraat 14, B-2710 Hoboken, Belgium. During the last decade, KHO has diversified the feed of its Hoboken smel ter. Larger proportions of precious metals has been extended and the production of minor metals has been developed. The smelter has focused on increasing productivity, complying with emission standards, and achieving the high recoveries of valuable elements. Assaying and sampling have been expanded •. Each step of the smelter is controlled and managed by a simulation model. The charge preparation of the sinter plant, returns handling, sinter conditioning and corresponding ventilation systems have been completely renewed. The annual smelting requirement is now achieved with two blast furnaces instead of three. One of the furnaces was rebuilt and enlarged, a new system of charging, oxygen enrichment, and blast air
AN OVERVIEW OF SILVER AND GOLD RECOVERY STRATEGIES IN THE ZINC INDUSTRY: D.D. Rodier, Canadian Electrolytic Zinc, 860 Cadieux Boulevard, Valleyfield, Quebec, Canada J6S 4W2 Most zinc concentrates contain some silver and gold, and the relative economic importance of the precious metals with respect to zinc and other metal values is presented. Although
the precious metals can be recovered by ISF smelting or by the processing of hydrometallurgical leach residues, none of the methods is without its limitations. The economics of precious
metals recovery are a direct function of the relative prices of Ag, Au and Zo; in some instances, the most economical approach
is not to recover the precious metals.
The different problems
faced by custom and dedicated zinc plants are outlined with
reference to typical smelter contracts. The advantages of a precious metals recovery circuit to a custom smelter in times
of concentrate shortage are discussed. Lastly, general trends related to the recovery of silver and gold are pre.se.nted, and
it is concluded that the trend to enhanced silver and gold recovery in the zinc industry will continue.
(9:00 a.m.)
SILVER LEACHING FUNDAMENTALS: D.H. Wyslouzil and R.S. Salter, Lakefield Research, P.O. Box 430, Lakefield, Ontario, Canada KOL 2HO
The fundamentals of dissolving silver and silver minerals are discussed under the following four sections: (1) Basic chemical reactions in the silver system, (2) Leaching of pure silver minerals, (3) Leaching of silver-bearing ores, (4) Leaching of reaction products from recovery processes. The fundamental chemical reactions of silver are an important guide to the mineral processor or the extractive metallurgist, particularly for systems in which complex formations are balanced against solubilities. The competition of other cations for free ligands can shift the equilibrium to the detriment of silver.
Solubilities of pure silver minerals in
specific solvents may not be a good indication of the solubility of silver minerals in ores. Recovery of silver from host minerals usually requires a destructive pretreatment. Frequently, silver minerals are transformed into an other insoluble form, and this requires further treatment before the metal can be recovered.
20-Annual Meeting (9:30 a.m.) PRECIOUS METAL RECOVERY FROM ZINC PLANT RESIDUE BY THIOUREA LEACHING : L.I. Rosato*. M. Berube*. D. Shink*. M. Blais*. G. Leroux •• and R.W. Stanley •• *Noranda Technology Centre,
240 Hymus Boulevard. Pointe Claire. Quebec. Canada
H9R 1G5;
.*Canadian Electrolytic Zinc, Valleyfield. Quebec, Canada
J6S 4W2
A novel hydrometallurgical process for the recovery of silver and gold from neutral leach residue. that is based upon leaching with thiourea. is described. Labortory and pilot plant studies have shown that 80-90% of the silver and 60-70% of the gold can be recovered by: (i) leaching zinc calcine with spent electrolyte at pH 2.0-3.0 to generate a precious metal-bearing zinc ferrite residue. (ii) leaching of the thickened zinc ferrite with 3-5 giL thiourea solution in a sixstage carbon-in-pulp circuit. (iii) recovery of the carbon by screening following by elution of silver and gold with a thiourea-butanol solution. (iv) thermal regeneration of the stripped carbon and (v) recovery of the precious metals from the pregnant strip solution by cementation with zinc dust. (10:00 a.m.)
BREAK
: 10:20 a.m.) SILVER RECOVERY OPTIONS AND DEPORTMENT IN THE KIDD CREEK ZINC PLANT': J.D. Scott and U. Dienstbach. Kidd Creek Division. Falconbridge Limited. Box 2002. Timmins. Ontario. Canada p4N 7Kl
The counter-current jarosite acid wash circuit presently in operation produces a combined jarosite plus Pb-Ag residue which is discarded. Extensive pilot-scale testwork. and some past operational experience. has shown that the plant can easily be reconfigured to produce two residues. Pilot~ptimized circuit configurations are available for silver recovery to a Pb-Ag residue. for jarosite destruction in the high~cid leach and for the prevention of Pb-Ag-jarosite formation in the leaching circuits. Silver in the zinc plant undergoes several changes in state as it proceeds through these circuits. The roaster calcine is 3:1 silver sulfate:Ag metal. In the neutral leach. the majority is as silver metal or as Pb-Ag sulfide coatings on unroasted sphalerite grains. In the high_acid leach. the silver occurs as Se-rich silver sulfide particles. as silver sulfate impregnations in silica_gel masses and as a solid solution in Pb-K-Zn-jarosite. (10:50 a.m.) SILVER AND GOLD RECOVERY FROM ZINC PRESSURE LEACH RESIDUE: R,M .G.S. Berezowsky. J. Stiksma and D.G. Kerfoot. Sherritt Gordon Limited. Fort Saskatchewan. Alberta T8L 2P2
Treatment of Hudson Bay Mining and Smelting Co. Ltd. (HBM&S) zinc concentrate in the Sherritt zinc pressure leach process
produces a residue containing silver and gold. Flotation of the residue rejects most of the silver and some gold to the
tailings. This paper reports testwork to define a process for a high recovery of the precious metals from the tailings. Digestion of the residue with lime at 90 0 C to render the silver amenable to enhanced recovery by subsequent cyanidation, was selected. Batch silver enhancement and cyanidation tests
showed that treatment of the tailings with 80 kg/t CaO at 90 0 C for 1 to 2 h increased the silver extractability in subsequent
cyanidation from <20% to 96%. A 96 h continuous miniplant test (1 kg/h solids feed rate). of a circuit including a silver enhancement treatment step followed by conventional cyanidation and carbon in pulp silver and gold recovery. confirmed metal recoveries to the carbon of 90% Au and 96% Ag. Reagent consumption rates were less than 100 kg/t CaO and 5 kg/t NaCN. : 11 :20 a.m.) RECOVERY OF SPECIALTY METALS. MAINLY GERMANIUM AND INDIUM FROM ZINC PRIMARY SMELTING: A REVIEW: J.M . Demarthe. A.M. Rousseau and F.L. Fernandez. Minemet Recherche. 1 Avenue Einstein. 78191 Trappes. France. The recovery of minor metals from zinc concentrates is
dependent on the process used for their treatment. The behaviour of germanium and indium in pyrometal1urgical
processes such as the Imperial Smelting furnace or the cupola furnace is of practical importance. The Anaconda process for the treatment of zinc leach residues is not well suited to the recovery of valuable minor metals. and new technologies must be
studied. In this regard. a solvent extraction process, using Kelex 100. developed by Minemet Recherche for the recovery of germanium from dilute sulphate leach solutions is described. ~lso. new hydrometallurgical techniques in chloride media can De applied to the purification and concentration of minor netals depending on the stability of their chloro-complexes and
the valency of associated impurities. A synthesis of methods is proposed to address the problems related to the recovery of ,aluable minor metals in the zinc industry .
MATERIALS DEGRADATION IN LOW EARTH ORBIT I Sponsored by the Jt. TMS/ASM Corrosion & Environmental Effects Committee Monday AM February 19, 1990
Room: Parke Inn at the Park
Session Chairmen: Steven L. Koontz, Johnson Space Center, ES531, Houston, TX 77058; George E. Caledonia, Physical Sciences, Inc., Andover, MA 01810 (8:30 a. m.)
Welcome:
V. Srinivasan. UES . Dayton. OH 45432.
(8:35 a.m.) Introduction to the Symposium: Bruce A. Banks, NASA Lewis Research Center. Cleveland. OH 44135.
(8:45 a.m.) LABORATORY INVESTIGATIONS of MATERIALS DEGRADATION by the LOW EARTH ORBIT ENVIRONMENT: Jon B Cross, Los Alamos National Laboratory. MS G-738. Los Alamos. NM 87545; Steven L. Koontz, NASA/Johnson Space Center. ES-5. Houston. TX 77058; John C Gregory, University of Alabama. Huntsville. Alabama 35899. The high concentration of atomic oxygen (AO) in the low earth orbit (LEO) and the high orbital velocity of spacecraft (8 km/s) places severe constraints on the types of materials which are being considered for long term (10-30 years) use in LEO. Organic based materials react with = 10% efficiency toward AO and even fluorinated hydrocarbons which are usually inert toward oxidation react with 1% efficiency. Lubricants as well as some metals which are resistant to oxidation under normal conditions are oxidized in the LEO environment. A cw laser sus.tained plasma source which has been developed to simulate the LEO AO environment is being used to develop accelerated testing techniques and determine operational life times of materials for long term use in LEO. A discussion of the comparison between our laboratory results and flight experiments will be presented. (9 : 15 a.m.) THE GROwrn OF OXIDE FILMS ON MErALS UNDER THE INA..UENCE OF
HYPERTHERMAL ATOMIC OXYGEN: J C Gregory and M. J. Edgell. The University of Alabama in Huntsville. Huntsville, AL 35899; J. B, Cross. Los Alamos National Laboratory, CLS·2, Los Alamos. NM 87545; S. L. Koontz. NASA. Johnson Space Center, ~ouston, TX 77058
Many metals. both representative and transitional. have thin protective oxide films of effectively unchanging thickness in air at ambient temperatures. Thin films of AI. Nb , Ni. Ir and other materials deposited on optical flats were shown (Peters and Gregory, ~ lliUill2S, 1290 (1986» to be modified by exposure to a flux of fast atomic oxygen in low Earth orbit. Similar thin films have been exposed at the LANL hyperthermal oxygen atom facility to beams of atoms with energies of several electron· volts per atom. Changes in film properties have been measured with XPS. Auger spectroscopy and optical techniques and are compared with the measurements made in orbit.
(9 :45 a.m.) ATOHIC OXYGEN UNDERCUTTING OF DEFECTS ON SiD2 PROTECTED POLYIMIDE SOLAR ARRAY BLANKETS: Bruce A, Banks. Sharon K. Rutledge, and Bruce M, Auer, NASA Lewis Research Center, Cleveland, OH 44135; Frank DiFilippo, Case Western Reserve University, Cleveland, OH 44106. Low Earth Orbital (LEO) atomic oxygen can oxidize polyimide
Kapton solar array blanket material which is unprotected as a
result of pinholes or scratches in the protective coatings. The probability of atomic oxygen reaction upon initial impact is low, thus inviting oxidation by secondary impacts . The secondary
impacts can produce atomic oxygen undercutting which may lead to coating mechanical failure and ever increasing loss rates of Kapeon . Comparison of undercutting effects in isotropic plasma asher and directed beam tests are reported . These experimental results are compared with computational cross sections based on
Monte Carlo methods and their implications on LEO performance of protected polymers.
(10: 15 a.m.) EFFECTS OF SIMULATED SPACE ENVIRONMENTS ON PROPERTIES OF SELECTED MATERIALS H. W. Dursch and H, G. pippin Mechanical Systems Technology Boeing Aerospace and Electronics Seattle. Washington 98124-2499 This paper covers results of recent experiments conducted at Boeing Aerospace and Electronics space environment screening
Annual Meeting-21 laboratory on the effects of atomic oxygen and ultraviolet radiation on materials. Results are reported for wear and friction measurements on thin film solid lubricants before and after exposure to atomic oxygen. Mass loss rates of several organic composites as a function of atomic oxygen exposure have been determined. An ultraviolet radiation source, including vacuum ultraviolet wavelengths, has been added to the materials screening chamber. Results of preliminary experiments on materials degradation under simultaneous AO and UV exposure and characterization of the UV output are also presented. (10:45 a.m.)
THE EFFECT OF ATOMIC OXYGEN ON ALTERED AND COATED KAPTON SURFACES FOR SPACECRAFT APPLICATIONS IN LOW EARTH ORBIT: Sharon K. Rutledge NASA Lewis Research Center, 21000 Brookpark Rd., Cleveland, OH, 44135; Judith Mihelcic, Cleveland State University, Euclid Ave. at East 24th St., Cleveland, OH, 44115.
Kapton is a material of great interest as a structural support for solar cells on Space Station Freedom because of its lightweight, flexibility, and IR transparency. However, a solar array blanket made of unprotected Kapton would undergo premature mechanical failure due to attack by atomic oxygen in the Space Station Freedom orbit. Commercially applied silicon dioxide coatings on Kapton and an alternate Kapton material with Si groups on the surface were evaluated for durability to atomic oxygen in both a random plasma and directed atomic oxygen beam system. Preliminary results indicate that the alternate Kapton surface provides good protection against attack by atomic oxygen. Although the coated surface remains protecting in an atomic oxygen plasma environment, overall failure of the coated Kapton occurred due to the number of defects in the coating which allowed undercutting to take place.
(11: 15 a.m.) SILICON DroXI DE SPACE COATINGS STUDIED ELLIPSor~ETRICALL Y: Bhola N. De, Yong Zhao, Jane Hruska, Jane Peterkin, and John A. Woollam, Department of Electrical Engineering, University of Nebraska, Lincoln, NE 68588-0511 Silicon dioxide films were prepared by electron beam evaporation and by exposure of a polished silicon wafer in an asher. In both cases film thicknesses were monitored (with a resolution of a few Angstroms) after exposure to a series of ashings using a pure oxygen plasma. Substrates were maintained at a series of temperatures from room temperature and up. These experiments allowed us to investigate fundamental mechanisms of oxygen diffusion through the film. For example, the mechanisms of initial (very short time) oxidation of silicon to form silicon dioxide have been investigated.
MATERIALS ISSUES IN ELECTRONIC & OPTOELECTRONIC CONNECTORS I: Mechanical Issues Sponsored by the TMS Non-Ferrous Metals Committee and the TMS Electronic Packaging and Interconnection Materials Committee MondayAM February 19, 1990
Room: California llIIIV Quality Hotel
Session Chairman: J. Crane, Olin Metals Research Labs., New Haven, CT 06511 (8: 30 a.m.)
MATERIALS CHALLENGES/MECHANICAL REQUIREMENTS FOR FUTURE CONNECTORS, R. S. Pokrzywa, IBM Corporation, Endicott. NY The increasing circuit densities and speeds of VLSI, VHSIC and GaAs integrated circuits require new computer packaging technologies which can provide high interconnect densities. controlled electrical characteristics and high reliability. These criteria challenge the materials and establish mechanical requir~m~nts for.future connectors. In addition, greater circuit densltles requlre larger nu~bers of int~rconnections per system. Therefore. the cost of the lnterconnectlon becomes a larger part of the total system cost. This paper will examine how the materials challenges and new mechanical requirements can be fulfilled, and at lowest cost. The increased interconnection density magnifies the importance of mechanical tolerances. insertion forces and normal force containment. Increased circuit density requires more power per chip which generates more heat so that the connectors are now required to operate at higher temperatures or be compatible with a cooling medium ot~er than air.
(9:00 a.m.)
Mechanical Hysteresis Considerations for Connector Design, W. R. Cribb and J. O. Ratka, Brush Wellman, Inc., 17876 St. Clair Ave. Cleveland, OH Although mechanical hysteresis is generally considered a metallurgical characteristic of secondary importance, it may exert substantial influence on interconnection performance. Particularly in low cycle duty (less than 100 insertions) meltalS or alloys subjected to cyclic loading may display hysteresis loop, behavior. This hysteresis can be caused by predeforming the metal elastically or plastically in either a reversed or unidirectional manner. The result of mechanical hysteresis is a measurable permanent strain or "set." Common connector alloys with varying strain histories were cyclically tested in uniaxial and bend loading to demonstrate their sensitivity to hysteresis associated with forming history. In connectors, the base metal processing history, the subsequent forming stresses and the nature of the service stress can interact to produce unexpected behavior. For example, unexpected reductions in normal force can be a manifestation of this hysteresis effect. (9:30 a.m.) STRESS RELAXATION OF SOME COPPER ALLOYS FOR CONNECTORS - COMPARISON BETWEEN NI AND FE IN COMBINATION WITH SN AND INFLUENCE OF CONDUCTIVITY. C. Andersson. A. Kamf and R. Sundberi. Meta11verken. 72188 Vasteras. Sweden Miniaturization required for future connectors demands a combination of good relaxation resistance and high conductivity. Brass and rhosphor bronze--widely used as connector materia s--have marginal resistance to stress relaxation at elevated temperatures. Alloys with low Ni/Sn or Fe/Sn additions have mechanical properties similar to phosphor bronze but have higher electrical conductivity and better relaxation properties at high temperatures. Relaxation up to 17~C will be compared between these a110is and t~fiJehnOcre borfoNi e i~n~o~~~e;oi~~i~cntoarnda i~Y:; precipitates in combination with Sn will be compared. In an actual connector the temperature rise depends amo~g other things on electrical conductivity of the material. Theoretical calculations of the temperature rise in materials can be shown to account for differences in relaxation for alloys with different conductivity.
(10:00 a.m.)
11ETALLURGICAL PHENOI·IENA AFFECTING ELASTIC SPRINGBACK AND STRESS RELAXATION IN PHOSPHOR BRONZE (CA510): Roger N. Wri.9!!! and Robert Hilty, Rensselaer Polytechnic Institute, Troy, NY 12180, Claudia Dudack, General Electric Co., Aircraft Engine, Evendale, OH 45215. George Galyon and P.J. Singh, IBM Data Systems Division, Poughkeepsie, NY 12602 Recent research at Rensselaer Polytechnic Institute has clarified the roles of many of the metallurgical variables influencing elastic springback and stress relaxation in cold worked CA510. Chief among these are degree of cold work, grain size, preferred grain orientation (texture), residual stress, and state of second phase precipitation. r~ajor questions have existed concerning the effects of "stress relief" anneals, the interpretation of "negative relaxation" phenomena, and the sources of anisotropic response. Recent research has produced plausible metallurgical models, and a number of processing recommendations can be set forth. (10: 30 a.m.) FACTORS AFFECTING FORMABILITY OF COPPER ALLOYS Alex J. Morris, Olin Metals Research Laboratories, New Haven, CT 06515 Miniaturization of connectors requires maximizing spring properties--deflection, relaxation, and conductivity--without compromising formability. This paper reviews available strengthening mechanisms in copper alloys and assesses their relevance to spring performance and impact on bend formability. Selected alloy systems representing solid solution alloys (brasses and phosphor-bronze), age-hardening alloys (beryllium coppers, Cu-Ni-Al, Cu-Ni-Si, Cu-Ni-Sn and dispersed second-phase alloys (Cu-FeP) are compared for formability versus appropriate spring properties.
22-Annual Meeting
MATERIALS SCIENCE OF MAGNETIC MATERIALS I
MODELING OF MICROSTRUCTURE EVOLUTION DURING SOLIDIFICATION I: Fundamentals
Sponsored by the TMS Magnetic Materials Committee (EMPMD)
Sponsored by the TMS Solidification Committee
MondayAM February 19, 1990
Room: California I-II Quality Hotel
Session Chainnan: W.A. Soffa, Department of Materials Science and Engineering, University of Pittsburgh, Pittsburgh, PA 15261 (8:30 a.m.) OPENING REMARKS: W.A. Soffa, Department of Materials Science and Engineering, University of Pittsburgh, Pittsburgh, PA
Monday AM February 19, 1990
Room: Plaza A Inn at the Park
Session Chainnen: P.W. Voorhees, Department of Materials Science & Engineering, Northwestern University, 2145 Sheridan Road, Evanston, IL 60208-3108; J.H. Perepezko, Department of Metallurgy & Mineral Engineering, University of Wisconsin, 1509 University Avenue, Madison, WI 53201
15261.
(8:45 a.m.) SCANNING ELECTRON MICROSCOPY WITH SPIN POLARIZATION ANALYSIS: PRINCIPLES AND APPLICATIONS: R C O'Handley. J. P. Woods, Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 Low energy secondary electrons emitted from a magnetic material show a spin polarization proportional to the magnetization in the outermost nanometer or so of the sample. When the secondaries are created by a narrow, scanning beam of primary electrons, the secondary polarization can be analyzed to give a high resolution image of the vector magnetization distribution across the surface. The principles of this technique are described, including necessary primary electron current, sample cleanliness, secondary electron optics and polarimeter design. A number of applications of the technique to problems of scientific and technical interest are outlined.
(9:30 a.m.) SEMPA INVESTIGATIONS OF MAGNETIC MICROSTRUCTURE: L Unguris, National Institute of Standards and Technology, Gaitersburg, MD 20899
Scanning Electron Microscopy with Polarization Analysis (SEMPA) provides a new means of imaging surface magnetic microstructure with high spatial resolution. When a ferromagnetic sample is probed by the highly focussed electron beam of an SEM, secondary electrons are generated whose spin polarization is directly proportional to the magnetization in the region sampled by the beam. By measuring this spin polarization, the direction and magnitude of the sample magnetization can be completely and quantitatively imaged with IO nm spatial resolution. We have successfully used SEMPA to study domains and domain walls in ferromagnetic crystals, glasses and thin films. Interest in these materials ranges from the physics of domain wall structure to understanding the noise characteristics of magnetic recording media. A description of the SEMPA apparatus and various examples of SEMPA measurements will be presented.
(8:30 a.m.) MICROSTRUCTURE EVOLUTION DURING SOLIDIFICATION: M E Glicksman, Materials Engineering Department, Rensselaer Polytechnic Institute, Troy, New York 12180-3590. Predicting the evolution of microstructures during solidification is basic to controlling cast properties. Progress has been made in analyzing microscopic (atomic scale) aspects of solid-liquid transformations, including modelling of interfaces and molecular attachment processes, and determination of the kinetic dependence of segregation coefficients and associated metastable modifications to the phase diagram. Mesoscopic features (microstructural scale) of solidification processes have been examined at the early, or nascent, stages through the application of morphological stability theory which forms the baSis for the scaling laws decribing cellular, dendritic, and eutectic growth. Phase coarsening is an additional kinetic effect added to assist our understanding of cast structures in the late stages of solidification. Macroscopic aspects of the freezing process, including the overall heat transfer and fluid flows, must now be merged with smaller scale modelling efforts to provide a more comprehensive view of cast microstructures. (9:15 a.m.) THERMODYNAMIC MODELING AND PHASE DIAGRAM CALCULATION OF MULTICOMPONENT SYSTEMS: YA Chang, Department of Materials Science and Engineering, University of Wisconsin-Madison, Madison, WI 53706. Essential to the design of desired microstructure from solidification is the need for phase diagram information. Although it is possible in principle to determine the solid-liquid equilibrium compositions experimentally as a function of gross alloy composition and temperature of a ternary and higher order systems, this approach is not practical if not impossible. The objective of this presentation is to show that we can obtain the needed information using a combined computational/experimental approach.
(10: 15 a.m.)
(9:45a.m.)
K. Mishra, Physics Department, General Motors Research Laboratories, Warren,
SOLIDIFICATION PATHS IN MULTI COMPONENT SYSTEMS. R. Mehrabian, College of Engineering, University of California at Santa Barbara, Santa Barbara, CA 93106
LORENTZ MICROSCOPY AND STUDY OF MAGNETIC DOMAINS: Rail! MI 48090-9055
Lorentz microscopy includes a number of electron imaging techniques which take advantage of the Lorentz deflection of a beam of electrons by the magnetic field of a domain in a magnetic material. Fresnel and Foucault images obtained in a transmission electron microscope are widely used for qualitative study of domains and domain walls. Lorentz images are also taken in scanning electron microscopes, scanning transmission electron microscopes, electron interference microscopes and magnetic force microscopes. In this paper, we shall discuss the advantages and limitations of Lorentz microscopy in a TEM and SEM for study of magnetization distribution inside domains and domain walls with special emphasis on quantitative analysis. New developments like combining Lorentz microscopy with spin polarization analysis or with micro magnetic computer simulation will be discussed. We will specifically discuss how to combine Lorentz microscopy results with microstructural results to completely characterize ferromagnetic materials at very high resolutions. (11 :00 a.m.) CHARACTERIZATION OF MAGNETIC THIN FILMS USING DIFFERENTIAL CONTRAST TRANSMISSION ELECTRON MICROSCOPY: I. R McFadyen, IBM Almaden Research Lab., San Jose, CA 95120
The transmission electron microscope (TEM) provides a powerful technique for studying the microstructure of thin films. However, in-focus bright-field imaging does not show magnetic contrast. The Fresnel and Foucault techniques result in non-linear imaging of the magnetic structure and difficulty in correlating the magnetic information with the structural information. In the STEM it is possible to produce phase contrast through a linear imaging technique by means of a quadrant detector. The magnetic structure is revealed by differential phase contrast (DPEC) which can brought into registry with an in-focus bright field image. This talk will be a tutorial on the basic principles and implementation of DPC on a conventional TEM with a scanning attachment or on a dedicated STEM. A number of applications in recording media and soft magnetic materials will be shown.
(Abstract not available) (10:15 a.m.)
BREAK
(10:30 a.m.) INTERFACE ATOMISTICS: MODELS OF SOLIDIFICATION. George H. Gilme~ Marcia H. Grabow, AT&T Bell Laboratories, Murray Hill, NJ CJ7974, and Aloysius F. Bakker, Applied Physics Department, Delft Technical University, 2600 G A Delft, The Netherlands. We have simulated the motion of silicon crystal-melt interfaces. Molecular dynamics code was implemented on a special purpose computer ATOMS which we constructed. In most cases the Stillinger· Weber interatomic potential was used. The measured crystallization rates on the (100) face are in good agreement with laser annealing results. The crystalline material generated during (tOO) solidification is essentially perfect for temperatures greater than O.6Tm' but below this point twins and stacking faults appear. The (100) interface speed vs. temperature is described by a diffusion-limited rate equation. Defects occur frequently during (111) growth. The trapping of impurities has been examined, and the pinning of the interface by slowly diffusing impurity atoms is found to influence the process. We observe the spontaneous nucleation of crystalline clusters only in systems containing some amorphous material. A computer-generated videotape of the model illustrates many of the phenomena described in the talk.
Annual Meeting-23 (11:00 a.m.) NONEQUILIBRIUM INTERFACE KINETICS DURING RAPID SOLIDIFICATION
Aziz, Division of Applied Sciences, Harvard University, Cambridge, MA 02138.
diamond synthesis reviewed. A discussion of the future developoment of this technology including issues of scaling will conclude the presentation.
~ ~
During rapid solidification following pulsed laser melting, the solid/liquid interface can be highly undercooled while
moving at velocities of up to tens of meters per second, and nonequilibriurn solute partitioning results in supersaturated solid solutions. Equilibrium partition coefficients (the ratio of the solid composition to that of the liquid at the 3 interface) can be exceeded by factors of 10 and equilibrium 5 solubilities can be exceeded by factors of 10 in this manner. A simple model using chemical rate theory will be presented that predicts the interfacial undercooling and partition
coefficient as functions of the interface velocity and the composition of the liquid at the interface. High-resolution measurements of the velocity and orientation dependence of the partition coefficient during rapid solidification will be presented and used to test several models.
(11:30 a.m.) THEORY FOR THE TRAPPING OF DISORDER AND SOLUTE IN INTERMETALLIC PHASES BY RAPID SOLIDIFICATION: William J. Boettinger, Metallurgy Div., Nat'l. Inst. of Standards and Technology, Gaithersburg, MD 20899 and Michael J. Aziz, Div. of Applied Sciences, Harvard University, Cambridge, MA 02138 A theory is developed to predict the long range order, parameter, composition, and temperature of an ordered solid as a function of interface velocity and liquid composition during rapid crystal growth. The engulfment of atoms randomly on two sub lattices by the rapidly moving liquid-solid interface is balanced against the interdiffusion across the interface that attempts to restore local equilibrium. With increasing interface velocity the theory predicts a progression from the solidification of a phase with equilibrium long range order parameter and with equilibrium solute partitioning to the solidification of a disordered crystalline phase with the same composition as the liquid. Solids with free energy functions in which the order-disorder transition (assumed to occur above the liquidus) is first or second order exhibit a decrease of order parameter to zero with increasing velocity which is discontinuous or continuous respectively.
NOVEL PROCESSING METHODS AND PROPERTIES OF DIAMOND AND DIAMONDLIKE THIN Fll...MS I: Nucleation and Growth Fundamentals Sponsored by the TMS-SMD and Electronic and Photonic Device Materials and Electronic Materials Committees and ASM and TMS-EMPMD Monday AM February 19, 1990
Room: Orangewood ill Quality Hotel
Session Chainnen: C. McHargue, Oak Ridge National Laboratory, Metals & Ceramics Div. B 5500, P.O. Box X, Oak Ridge, TN 37831; B. Rath, Naval Research Laboratory, Code 6000, Bldg. 43, Room 212, Washington, DC 20375-5000 OPENING REMARKS: (8:30 a.m.)
(9:00 a.m.)
(Invited) tHE CHEMISTRY OF DIAMOND CVD:
J E. But1or, Gaa/Surface
Dynamic. Section, Chemiatry Divi.ion, Code 6174, Naval Re •• arch Wa.hington, DC 20375-5000
~boratQry,
The chemical mechanil",. of diamond lI"ovth at lov pr ...ur.. (one atmo.phero or le •• ) will be di.eua.ed and comp.red with the r •• ult. of in .ieu optical di'ano.tic. (IR diode la.er, R.aonanc. Enhanoed Ioniz.tion, and IR Ref1ectanoe Abaorption Speotrosoopiea) of diamond growth environment.. A simple .adel derived from the in .ieu observ.tiona vill b. pre.ented whioh rationalize. mAny of the ob.erved oorrelation. in the quality of the material depo.ited with the diamond depo.ition oondition.. The grow~h of diamond in atmoapheric combu.tion flame8 will be pre.ented and discuI •• d.
(9:30 a.m.) (Invited) Pla.ma Enhancad Chamical Vapor Oapo.ition of Thin Diamond 1ilml, R.P.H. Chang, Dept. of Mat. Sci. ~ Eng. and Hat. Ree. Ctr .• North.e.tern Univ.reity, Evan.ton. IL.
60208
Thin diamond films formed by plasma-assisted deposition have .hown great promi,e of potential technological applications, In this talk, ve vill disease in details the naeleatlon and growth of diamond polycry.talline films on various substrate materials. In particalar • • e will present result. on how the cry.tallite .ize of the film can be varied by careful adjastment of the depo.ition parameter.. It is shown that the machanical and optical properti.. of the.e film. ara directly correlated to the microstractare propertie. of the film •. In addition. we will di.eas. the importance of the interfacial properties of the diamond tilms and the sub.trate •. By controlling the interfacial propertie., it i. po •• ible to alter the adha.ion and wear properties of the diamond film. Specific examples w111 be i1ven. (10:00 a.m.)
(10:30 a.m.)
BREAK
Invited
FUNDAMENTAL ASPECTS OF EPITAXIAL GROWTH OF LASER AND PLASMA PROCESSED DIAMOND FILMS: J NaraYan, Materials Sci. and Engr. Dept., North Carolina State Univ., Raleigh, NC 276957916 The purpose of this talk is to review fundamental aspects of epitaxial growth in diamond cubic lattices. Thin films grow pseudomorphically up to a certain thickness (critical thickness) and then dislocations are generated to relax the strains and reduce the strain energy of the system. Various forms of epitaxial and textured growth mechanisms (lattice, plane, and shape controlled) including unseeded crystallization will be covered. Atomic structure of dislocations and interfaces, which hold the key to the formation of high-quality thin films, will be described in detail. The atomic structure and mechanisms of fonnation of fi ve-fold twins are also discussed.
(11:00 a.m.) Invited THERMODYNAMICAL ASPECTS OF DIAMOND THIN FILM GROWTH: Walter Yarborough, Materials Research Laboratories, Pennsylvania State University, University Park, PA 16802. (Abstract not available)
(11:30 a.m.) (Invited) J. Narayan, North Carolina State University, Dept. Mats. Sci. & Engr. ,- Raleiqh, NC 27695 INVESTIGATION OF THE ROLE OF CHARGED SPECIES IN HOT FILAMENT
Invited
SCIENCE AND TECHNOLOGY OF CHEMICAL VAPOR DEPOSmON PROCESSES: K.V.Ravi, Lockheed Missiles and Space Company, Research and Development Division, 3251 Hanover Street, Palo Alto. CA. 94304-1191
The most commonly utilized approach for the synthesis of diamond films has been the plasma enhanced chemical vapor deposition (PECVD) process", The PECVD technique for diamond synt)iesis can overcome the thermodynamic barrier to the synthesis of the metastable form of carbon, diamond, under relatively low pressures (-10 to 200 Torr) and temperatures ( - 600 to 900 oc). The use of high concentj'ations of atomic hydrogen suppresses the formation of graphite as well as preferentially dissolving any graphite that might form during the diamond synthesis process. Activation or enhancement can be achieved by a variety of means including the use of DC glow discharges, RF and microwave excitation. In this presentation the current status of the PECVD technology for the synthesis of diamond films will be reviewed. The advantages and disadvantages of the various approaches to the activation of hydrocarbon! hydrogen palsmas will be discussed and the current understanding of the thermochemistry and nucleation and growth processes as they penain to
ASSISTED CVD OF DIAMOND: F.P. Doty and W.A. Jesser, Dept. Mats. Sci., University of Virginia, Charlottesville, VA 22903-2442.
A number of researchers have reported that positive charge on the substrate increases the growth rate of diamond in low pressure CVo systems. Since positive charge would attract electrons emitted from the filament, and since ion concentrations are assumed negligible, this method has been called Electron Assisted CVD. Recent experiments on the thermal activation of conduction in the hot filament CVD apparatus indicate that thermionic emission may not be the process creating charge carriers, as has been assumed. The activation energy for conduction in vacuum or pure hydrogen was found to be 3-4 eV (consistent with reported work functions for tungsten carbides). The addition of 0.1 torr of methane increased the conduction, and reduced the activation energy to 1.5 eV. These results, and the results of growth experiments will be presented, and ~~e possible role of surface ionization will be discussed.
24---Annual Meeting
PROCESS FUNDAMENTALS I: Fluid Flow Phenomena in Materials Operations
REDUCTION TECHNOLOGY I: Cell Modelling and Magnetics
Sponsored by the TMS Process Fundamentals Committee
Sponsored by TMS Aluminum Committee
MondayAM February 19, 1990
Room: Salon H Marriott
Session Chairmen: J. Szekely, Department of Materials Science and Engineering, MIT, Cambridge, MA 02139; D.G.C. Robertson, Department of Metallurgical Engineering, University of Missouri-Rolla, Rolla, MO 65401 (8:30 a.m.) FLUID FLOW IN COUNTERCURRENT REACTORS: A. Agrawal and D.G.C. Robertson, Department of Metallurgical Engineering, University of Missouri-Rolla, Rolla, Missouri 65401 (Abstract Not Available)
(9:00 a.m.) A MATHEMATICAL REPRESENTATION GAS PLUME DRIVEN RECIRCULATION SYSTEMS: O. J. Ilegbusi and J. Szekely, Dept. of Materials Science and Engineering, MIT, Cambridge, MA 02139. Gas plume driven circulation systems play a very important role in metals processing. Steelmaking, ladle metallurgy operations and copper converting are good examples. We shall present a general mathematical description of these systems, such that the plume shape is actually predicted using the process parameters.
As part of the overall computational
scheme we can then calculate the velocity fields, turbulence levels and mixing in a broad range of systems of metallurgical interest. The theoretical predictions are compared with measurements.
(9:30 a.m.) A MODEL STUDY OF STRATIFIED FLOWS IN SUBMERGED ARC SLAG CLEANING: S. Kang and D.G.C. Robertson, Department of Metallurgical Engineering, University of Missouri-Rolla, Rolla, Missouri 65401. (Abstract Not Available) (10:00 a.m.)
BREAK
(10:15 a.m.) A mathematical model for laree scale filtratjon of alumjnjmn Stein Tore Jobansen and Nigel M. Anderson
SINTEF, Division of Metallurgy, N· 7034 Trondheim·NTH, Norway The flow of a liquid through a large ball filter has been considered. An improved form of the fluid dynamic equations describing flows through porous media is proposed. The theory is applicable to all Reynolds numbers. Particulate inclusions are desribed by passive scalar
fields where only very simple mechanisms for particle removal are included. However, the
deposition of particles are allowed to change the "porousity" of the filter. Hence, the flow
pattern thrOUgh the ruter will change as the filter ages. The theory is employed to the filtration of aluminium in an industrial ball filter. The model shows that the ruter is not effectively designed and there is room for considerable improvement. The dynamic "aging" prosess of the filter will be discussed. Finally, the predicted
inclusion collection efficiencies are compared to efficiencies measured in an industrial filter. The discrepansies between theory and experiments can both be attributed to the filter-particle adhesion as well as the micro mechanisms for particle deposition at the filter balls.
Monday AM February 19, 1990
Room: Salon J & K Marriott
Session Chairman: Kerry Farmer, Alurnax of South Carolina, P.O. Box 1000, Goose Creek, South Carolina 29445 (8:30 a.m.) EXPERT SYSTEMS APPLIED TO CONTROL OF ALUMINIUM SMELTERS: L. Tikasz and R.T. Bui, Universite du Quebec a Chicoutimi, Chicoutimi, Que. Canada G7H 2B1 and V. Potocnik, Alcan International Limitee, Jonquiere, Que., Canada G7S 4K8 The application of Expert Systems is probably the next major step in the process control and process supervising of aluminium electrolysis: the present-day systems should be kept as base, be reorganized to a new structure. After outlining the present fundamental Expert System tools, then proposes a strategy and structure for the control and supervision of aluminium smelters. Finally, demonstration is shown in connection with the reasoning strategies on the field of alumina feeding. (8: 55 a.m.)
1WO-DIMENSIONAL lHERMAL MODELING OF 1HE HALL CELL, I N BruL!L!eman and D. I. Danka, Alcoa Technical Center, Alcoa Center, PA 15069 A two dimensional thermal model of a HaIl cell was developed to aid in cell thennal analysis. The model is a pure-conduction model, using enhanced effective thermal conductivities for bath and metal phases to account for bulk convection and heat transfer coefficients at fluid-solid interfaces. Model predictions were compared to measured ledge profiles and heat fluxes for several pots. The model fit measured data well after only minor adjusttnents to lining component conductivities. The model was then used to explore the effect of pot design and operating parameters on the ledge contour, average bath temperature, and heat loss distribution. Pot performance was particularly sensitive to ambient heat transfer coefficient, small changes in the bath voltage, top losses, and bath and metal depths. Performance was surprisingly insensitive to collector bar heat loss, and spatial distribution of heat sources in the anode and cathode. In general, the extent of ledge fonnation beneath the metal pad was much more sensitive than the sidewaIlledge thickness to any of the pot operating or design parameters. These conclusions are valid only for the particular lining design considered, and not necessarily applicable to other designs. (9:20a.m.)
1HE INFLUENCE OF GAS-DRIVEN CIRCULATION ON ALUMINA DISTRIBUTION AND INTERFACE MOTION IN A HALL CELL D. C. Chesonis and A F LaCamera Aluminum Company of America Alcoa Technical Center, Alcoa Center, PA 15069 A physical model with multiple anodes was used to investigate the influence of the gas-driven circulation on alumina distribution and interface motion in a 90 KA HaIl cell. Much of the gas leaving the anode evolves into the gap fonned by the sides of adjacent anodes. This gas release was periodic with a frequency of about one hertz. Alumina distribution depended on feed location, bath depth, and the size of the gap; it was dominated by gas-driven flow. The periodic release of gas creates a region of low pressure that causes the interface to deform; the amplitude is influenced by the gap size, anode-cathode spacing, bath depth and anode current The optimum gap size for effective alumina distribution gives the maximum deformation of the interface.
(10:45 a.m.) A MATHEMATICAL REPRESENTATION OF ELECTROMAGNETIC, HEAT AND FLUID FLOW PHENOMENA INSIDE PLASMA TORCHES: J. Szekely, A. H. Dilawari, and R. Westhoff, Dept. of Materials Science and Engineering, MIT, Cambridge, MA 02139.
(9: 45 a.m.)
A mathematical representation is being developed to describe the development of the arc and the resultant heat flow and flow pheno~ena inside plasma torches. It is shown that, notwithstanding the approximations made, this approach can provide quite reliable results in predicting torch behavior and also for representing the behavior of plasma plumes.
Stable and efficient operation of Hall Heroult cells for the electrolytic production of aluminium involves a number of heat and mass transport processes. A major driving force for these transport processes is the bubble driven circulation in the molten electrolyte due to the release of carbon dioxide gas from underneath the anodes. This Circulation provides a physical mass transport mechanism to aid in alumina dissolution and distribution throughout the electrolyte. Also, the bubble driven circulation provides a convective heat transfer mechanism for dissipalion of the ohmic heat generated by the passage of current through the electrolyte in the interpolar gap. This paper presents results of mathemalical modelling of the bubble Induced flow in the electrolyte and examines the implications of these flow fields for heat and mass transport in the electrolyte. In particular these results are used to model the heat transfer from the molten electrolyte to the sidewall and hence provide for the accurate calculation of the frozen ledge profile at the sidewall.
Experimental results are cited in support of this contention.
(11: 15 a.m.) SUPERSONIC FLOW IN NOZZLES FOR LIQUID METAL ATOMIZATION: K. Hamann and D.G.C. Robertson, Department of Metallurgical Engineering, University of Missouri-Rolla, Rolla; Missouri 65401.
(Abstract Not Available)
ELECTROLYTE HEAT AND MASS TRANSPORT PROCESSES IN HALL HEROULT ELECTROLYSIS CELLS: K.J. Fraser, M P Taylor and A.M. Jenkin, Comalco Research Centre, 15 Edgars Road, Thomastown. Vic. 3074
Annual Meeting-25 (l0: 10 a.m.) TOPICS IN SIGNAL CONDITIONING AND DIGITAL PROCESSING TECHNIQUES FOR MONITORING ALUMINUM REDUCTION CELLS: C. Simard and A. Weatherely, RealTime Engineering, Goldendale, Wa. 98620 USA., and I. Jamieson, Industrial Software P/L., 628 Canterbury Rd., Vermont, Vic., 3133 AUST. General topics concerning the acquisition and mathematical processing of cell voltage and amperage signals are discussed, including acquisition of primary voltage and amperage signals, calculation and smoothing of pot resistance and processing of second-order resistance statistics. Graphical examples of the behaviors of mathematical and electrical filters are offered. Trends for the future are discussed.
(10:35 a.m.) EFFECT OF AN INSTABILITY OF THE METAL SURFACE ON THE MAGNETIC FIELD INSIDE A CELL: R. Boivin and S. Martel Alcan International Ltd, Arvida Research and Development Centre P.O. Box 1250 Jonquiere, Quebec, Canada, G7S 4K8. ' , The detrimental impact large-scale instabilities of the bathmetal interface have on the Hal I-Heroult process is reviewed. A meth~d i~ described for calculating the changes in the cel I magnetIc f,eld caused by such instabi I ities. Applying the method to a real wave observed on a high-amperage side-by-side cel I shows that the transverse and vertical field components are substantially affected. This suggests that the magnetostatic approximation frequently made in modelling metal-pad instabilities may have to be revised. (11:00 a.m.) MOOEUNG STUDIES OF ELECTAOLYTE FLOW AND BUBBlE BEHAVIOR IN ADVANCED HALL CELLS: R. Shekhar and J.W. Evans, Department of Materials Science and Mineral Engineering, University of California. Berkeley, CA 94720
SUPERPLASTICITY IN AEROSPACE I, 2nd International Symposium Sponsored by the TMS Shaping and Forming Committee Monday AM February 19, 1990
Room: Garden Rooms 1·2·3 Inn at the Park
Session Chairman: T.R. McNelley, Department of Mechanical Engineering, Naval Postgraduate School, Monterey, CA 93943 (8:30 a.m.) SUPERPLASTIC CERAMICS - AN OVERVIEW: Terence G. Langdon, Departments of Materials Science and Mechanical Engineering, University of Southern California, Los Angeles, CA 90089-1453. Although ceramic materials are generally considered inherently brittle, several examples are now available of true superplastic behavior in ceramic systems. This paper examines the reports of superplasticity in ceramics, both in materials with an intergranular glassy phase and in materials where the glassy phase is either very limited or isolated primarily in pockets at the triple junctions. The results are discussed with particular reference to recent tensile experiments on an yttria-stabilized tetragonal zirconia. (9:15 a.m.)
Much research has been carried out in recent years by corporations and university Igovernment labs. on materials for use in advanced Hall-Heroult cells. Attention has focused on materials for use as wettable cathodes and inert anodes and much has been achieved in terms of materials development. Comparatively less attention has been devoted to how these materials might be incorporated in new or existing cells, i.e., to how the cell should be designed, or redesigned, to take full advantage of these materials. The present paper is an effort, supported by the U.S. Department of Energy, to address this issue. The primary objectives are cell designs where electrolyte flow can be managed so as to promote both the removal of anode gas bubble and the convection of dissolved alumina into the inter-cathode distance is small. The principal experimental tool has been a "wale" model consisting of a large tank in which simulated anodes configurations. Gas "generation" was by forcing compressed air through porous graphite and the fine bubbles characteristic of inert anodes were produced by adding butanol to the water. Velocities were measured using a laser Doppler velocimeter and local gas fractions by an electrical resistivity probe. An interpretalion of the results in terms of a general mathematical model, aimed at predicting electrolyte flow and bubble behaviour for real cells is presented.
IN BI~ SYSTEMS: Richard C. Bradt, of ChemIcal and Metallurgical Engineering, School of Mines, University of Nevada-Reno, NV 89557-0047. phase transition in pur-e Bi~3 from the low tem~erature m?n?clinic ph~se to the high temperature CUbIC one exhIbIts extenSIve deformation on heating. The structural and microstructure aspects of that superplasticity are reviewed and compared with similar behavior in zirconia and some metallic systems. Extensive transitional deformation in Bi~3-eutectoid systems is also addressed where the role of microstructural stabilization by the second phase is shown to be critical to multiple cycling through the transformation. The role of secondary phase content is also considered.
(11:25 a.m.)
(10:00 a.m.) MECHANISMS OF SUPERPLASTICITY IN CERAMICS: Rishi Raj, Dept. of Materials Science & Engineering, Cornell University, 229 Bard Hall, Ithaca, NY 14853
MAGNETO-HYDRODYNAMIC PHENOMENA IN PRODUCTION OF ALUMINIUM BY ELECTROLYSIS: V Almukhametoy, V. Kryukocvsky, V. Kolesnichenko and Khripchenko. This paper covers the magneto-hydrodynamic phenomena which takes place in high-amperage cells and have a considerable impact on cell operation economics and efficiency. A pinch mechanism of generation of circular flows in the cell was studied as well as a mechanism caused by a flow of electrical current in various current-carrying elements. It was shown that a magnetic field established by currents in anode risers situated along an end of the cell enhances circulation of melt caused by the pinch mechanism the anode risers situated along a side of the cell establish the induction of a magnetic field reducing the circulation. The mechanisms are considered of originating the two four and eight contour melt circulation patterns. Mathematical relations were checked on a single layer physical model and were confirmed qualitatively. A wave formation mechanism in the cell was studied and the limits of an MHD-stability were established.
SUPERPLASTICIT~
Dept. Mackay Reno, The
Ceramics can deform superplastically, that is with a strain rate sensitivity of greater than 0.5 and at a strain rate faster than 10-4 g-I, not only by the classical Coble creep mechanism, but also by two other processes. In one the presence of a fluid phase at grain interfaces promotes diffusional transport at temperatures that are much lower than those where Coble creep Would be expected. In the other mechanism it appears that deformation occurs by dislocation cellular structures where the cell size is about equal to the grain size. Theories and their application to engineering materials such as silicon nitride will be presented. (10:30 a.m.)
SUPERPLASTIC BEHAVIOR OF TITANIUM ALUMINIDES: C. Gandhi and Clifford C. Bampton, Rockwell International Science Center, Thousand Oaks, CA 91360 Both super alpha two and gamma titanium aluminides are potential str~ctural materials in hypersonic aircraft and high performance englnes. Superplastic forming and diffusion bonding characterlStlCS of these alloys are evaluated for applications in these systems. Both alloys show high flow stresses and relatively low strain rate sensitivities at the superplastic temperatures. Superplastic behavior of super alpha two is very sensitive to grain size and sh~pe, phase distribution, texture and compositlon. At the optlmum superplastic deformation conditions, super alpha two exhibits nearly 600% elongation, but gamma alloys show about 300%. Recent experimental results and theoretical modelling of superplastic behavior in these alloys will be discussed.
26-Annual Meeting (11:00 a.m.)
SUPERPLASTICITY AND CAVITATION OF BORON DOPED Ni AI: J. Mukhopadhyay, G. Kaschner and A.K. Mukherjei, Division of Materi a1s Sci ence and Engineering, Department of Mechanical Engineering, University of California, Davi.s, CA 95616. High temperature tensile properties of boron doped Ni Al alloy containing Cr and Zr have been studie~5at temper!tu~ys from 1273K to I373K and strain rates of 5xl0 to 2xIO S . Parameters such as strain rate sensitivity index (m), elongation to failure and activation energy have been evaluated. The m values lie in the range of 0.5 - 0.9 for the range of strain rates and temperatures used in this investigation. The m.aximum_4elo!,~ation (575%) is found to occur at 1323K and £=6xl0 S. The recrystallization kinetics appear to be very sluggish. The grain size of the alloy is 91'm and the amount of stat i c gra ingrowth observed for this graiR size is negligible. Microstructural features such as grain boundary sliding, cavitation, strain enhanced grain growth and fracture morphology will be discussed in the context of superplasticity in intermetallic compounds. This investigation is supported by the NSF Grant DMR-8700583.
SURFACES & INTERFACES TMS General Abstract Session MondayAM February 19, 1990
Room: Plaza D Inn at the Park
Session Chairman: John R. Bradley, General Motors Research Labs, Warren, MI 48090 (8:30 a.m.) CHARACTERIZATION OF ZINC GRAIN STRUCTURES IN ELECTROPLATED ZINC· COATED STEEL: *w Najjma, S.J. Shaffer, lW. Morris, Jr., Center for Advanced Materials, Lawrence Berkeley Laboratory, I Cyclotron Rd., Berkeley, CA 94720 Laboratory prepared samples of electroplated zinc-coated steel were polished and chemically etched in cross-section to reveal the grain structures in the 10 IJ.III thick zinc layer. Special polishing techniques were developed due to the reactive nature of zinc, and the difference in hardness between steel and zinc. Equiaxed and columnar structures were observed with variations in the electroplating conditions. Characterization of these structures was done with respect to surface morphology and crystallography. *This work was supported by the Office of Basic Energy Science, the U.S. Department of Energy, and by the Ford Motor Company, Rouge Steel Company and LTV Steel Company.
(8:50 a.m.)
AUGER ANALYSIS OF L12 TYPE INTERMETALLIC COMPOUND BORON -DOPED Ni,AI N.Masahashi, K.Tatsumi and M.Yamamoto R&D LaboratoriesI, Nippon Steel Corp. Nakahara-ku Kawasaki 211 JAPAN
Grain boundary segregation behavior of boron in Ni3AI has been studied. As already reported, boron segregates to grain boundary strongly. But even at Al rich offstoichiometry alloy, which does not show any ductility, a sufficient amount of boron is observed in segregate for increased boron doping. This result suggests that stoichiometry is important for ductility Improvement and Ni-rich composition has a potential to show some deformability even without boron. The enrichment factor of boron is calculated to be from 4 to 7, without any colleration with stoichiometry. Although SACP-ECP experiment have failed to confirm above hypothesis, ECP have proved effective to estimate the grain-boundary stress through comparing the clearness between boron doped and free alloys. (9: 10 a.m.) AUGER ANALYSIS OF THE FRACTURE SURFACES OF A HIGH
CHROMIUM STEEL EMBRITTLED AFTER 30 YEARS OF SERVICE Ashok Choudhyry, Oak Ridge National Laboratory, Oak Ridge, TN 37831; C.R. Brooks, U. of Tenn., Knoxville, TN 37996 and R. A. Padgett, Oak Ridge National Laboratory, Oak Ridge, TN 37831. A broken bolt made of high - temperature steel containing primarily 12 wt % Cr and 0.2 wt %C was analysed after 30 years of service in a fossil fuel power plant. It was extremely brittle and its fracture surface showed predominantly intergranular fracture. A heat treatment was devised which restored the toughness, and the level of intergranular fracture was found to be markedly less. A detailed Auger
analysis was conducted on the fracture surfaces of both as - received and heat- treated samples. This paper reports the results and discusses Ipossible causes for the embrittlement. 'Research sponsored by the USDOE, Asst. Sec. for Cons. & Ren. Energy, Off. of Trans. Sys., as part of the HTML User Program, under contract DE-AC05-840R21400 with Martin Marietta Energy Systems, Inc.
(9:30 a.m.)
CHARACTERISTICS OF OXIDE SCALES IN Ce DOPED Fe-25%Cr BINARY ALLLOY: V. Srinivasan, universal Energy Systems, Dayton-Xenia Road, Dayton, Ohio and Raghavan Ayer, STEM, Inc., 305 Bic Drive, Milford, CT 06460 The influence of rare earth additions on the kinetics of scale growth in cr 2 0 3 forming systems has been a subject of continuing l.nterest. The effect of Ce incorporation in the chromia scale was investigated in Fe-25%Cr alloys in which Ce was introduced by surface implantation or as a solute addition to the bulk alloy. Analytical electron microscopy studies of the chromia scale oxidized at 973K revealed a duplex grain size and an gradient in Fe concentration. In addition, a new structure of the oxide was observed. The details of the analysis will be discussed. *The research was sponsored by Martin Marietta Energy Systems, Inc., which operates Oak Ridge National Laboratory for DOE. Program Manager: Dr. R.R. Judkins.
(9:50 a.m.)
ENHANCED DIFFUSION DURING CONSTANT STRAIN-RATE DEFORMA· TION OF NaCI SINGLE CRYSTALS USING NMR Klaus Detemple, O. Kanert, and K. Linga Murty', University of Dortmund, Dortmund, FRG, 'North Carolina State University, Raleigh, NC 27695· 7909. NMR pulse techniques are used to investigate the effect of deformation on diffusion in single crystals of NaCI by monitoring the temperature dependence of spin-lattice relaxation rates in the rotating frame [l/T,p]'" a function of the test temperature from ambient to 900K. High purity [lppm] single crystals of NaCI were studied in compression at constant strain·rates using a closed-loop servo hydraulic set up in a 4.2T superconducting magnet along with a Brucker NMR spectrometer. At low temperatures, below about 473K, two phonon Raman process dominated the relaxation behavior while the motional narrowing was noted at higher temperatures with a peak in the relaxation rate at about 725K. The excess vacancy concentration during deformation ~tracted from these results indicated a linear increase with the applied strain-rate with a slope decreasing with increased test temperature due probably to in-oitu vacancy annihilation. We acknowledge the financial support from the NC Board of Science and Tech· nology and the Government of Noerdrhein·westfalen.
MONDAY PM ADVANCED METAL & CERAMIC MATRIX COMPOSITES I: Processing Ceramic Matrix Composites Sponsored by: TMS Powder Metallurgy Committee, TMS Physical Metallurgy Committee, and Joint TMS/ASM-MSD Composite Committee Monday PM February 19, 1990
Room: ParkD Inn at the Park
Session Chairmen: Ram B. Bhagat, Penn State University, Applied Research Laboratory, P.O. Box 30, State College, PA 16804; Maurice F. Amateau, Penn State University, Applied Research Laboratory, P.O. Box 30, State College, PA 16804 (2 :00 p.m.) WELCOME AND INTRODUCTORY REMARKS: Gunther E. Petzow, Max-PlanckInstitut f. Metallf., Pulvermetallurglsches laboratorium, Heisenbergstr 5, D-7000 Stuttgart 80, FRG; Ashok K. Dhingra, E. I. Du Pont De Nemours & Co., Pioneering Research Laboratory, Experimental Station, Wilmington, DE 19898.
Annual Meeting-27 (2:10 p.m.) INTERMETALLIC AND CERAMIC MATRIX COMPOSITES FOR 1500°F - 2500°F GAS TURBINE ENGINE APPLICATIONS. Joseph R. Stephens, NASA Lewis Research Center, Cleveland, OH 44135. Lightweight and potential high temperature capability of intermetallic compounds, such as the aluminides, and structural ceramics J such as the carbides and nitrides, make these materials attractive for gas turbine engine applications. In terms of specific thrust and specific fuel consumption, revolutionary improvements over current technology are being sought by realizing the potential of these materials through their use as
matrices combined with high strength, high temperature fibers. This presentation will summarize the status of NASA's Advanced High Temperature Engine Materials Technology Program (HITEMP) and describe the potential benefits to be gained in the 21st century transport aircraft by utilizing intermetallic and ceramic matrix composite materials. (2:40 p.m.) PROCESSING, MICROSTRUCTURES, PERFORMANCE AND ECONOMICS OF CERAMIC COMPOSITES. Roy W. Rice, W. R. Grace & Co., 7379 Route 32, Columbia, MD 21044. Status
and needs
of ceramic particulate, whisker,
and fiber
composites are reviewed from not only the processing microstructure performance perspective, but also from the standpoint of economics, i.e., materials and process costs vs markets. For example, beyond the basic compositional and homogeneity requirements, performance of ceramic particulate composites is often better with finer dispersed particles. Hot pressing or sinterHIPing fairly readily meet these requirements for high valueadded components (e.g., cutting tools), but are much less economically viable for many broader applications. Major needs/ opportunities are to develop much lower cost whiskers or fibers (or to use existing lower cost fibers). For whiskers, an important challenge is to achieve non-catastrophic failure.
(]:OO p.m.) MULLITE-CORDIERITE COMPOSITES BY SOL-GEL ROUTE. V. Srikanth, P. Ravindranathan, L. Rani and R. Roy, Materials Research Laboratory, The Pennsylvania State University, University Park, PA 16802. Mullite-Cordierite composites with various mullite contents were prepared by the sol-gel route. Sinterings of the composites were carried out for different lengths of time at different sintering temperatures to .understand the densification behavior of the composites. Sintering atmosphere was also varied to have better control on the microstructure. Elastic moduli, fracture toughness, thermal expansion of the composites were determined as a function of the composition. Mechanical strength of the composite increases linearly with mullite content in the composite.
(]:20 p.m.)
Sintering and Mechanical Properties of SiC-Reinforced Mullite- (Zr02)-Composites, M.J. Hoffmann and G. Petzov, Max-Planck-Institute for Metal Research, Institute for Material Science, Povdermetallurgical Laboratory, Heisenbergstr. 5, D-7000 Stuttgart 80, FRG.
.
Mullite composites containing up to 15 vol.% Zr02 and 20 vol.% sic were cold isostatic pressed as well as slip cast. The rheology of the aqueous slips was characterized by viscosimetry. The influence of the SiC-inclusions with different morphologies (particles, platelets and whiskers) on the flow behaviour was measured. Densification was achieved by pressureless sintering as well as hot-pressin~ in the temperature range between 1550 and 1650 C. The shrinkage behaviour of the composites was studied by dynamic high temperature dilatometry. Mechanical property measurements confirm that Zr02-toughening and crack deflection at the SiC-inclusions can work simultaneously. (]:40 p.m.)
4.5 - 7.3 MPalID(RT). Observations of metallographic structures with SEM and optical micrographs and identification of crystal phases in the composites were carried out. (4:00 p.m.) SHOCK. -COMPRESSION PROCESSING OF lflGH-PERFORMANCE CERAMICS"
R. Ward, N N Thadhani, and P.A. Persson, CETR, New Mexico Tech, Socorro, NM 8780 I.
Shock-induced reaction synthesis is used to assist in the consolidation of IiB 2, B4C, SiC and AI2~+B4C, ceramic powders. These inert ceramics are mixed with elemental Ii and B or Ti and C powders. The passage of explosively generated shock waves initiates an exothermic chemical reaction between the elemental powders, The resulting in the formation of a hard ceramic binder phase. ceramic binder in addition to the reaction heat aid in the consolidation of the otherwise difficult-to-consolidate ceramics. Almost crack-free compacts of TiB2, B4C, SiC and AI20 3+B4C, with Microstructural better than 96% density have been produced. analysis of that reaction products reveals that the elemental Ti, B and C also react with the inert ceramics. The shock-compression processing technique, structural and mechanical properties of reaction products, and the effects of processing conditions on these properties will be discussed. ·Supported by DARPA under Grant #DE-AC04-87AL43261. (4:20 p.m.) EFFECT OF WHISKER GRAIN ORIENTATION ON THE MECHANICAL PROPERTIES OF SILICON CARBIDE/ALUMINA COMPOSITES. A. Srinivasa Rao* and Om P. Arora, David Taylor Research Center, Department of the Navy, Annapolis, MD 21402. Silicon carbide (whisker) and alumina composites were processed with and without concurrent application of an external electric field in order to orient the whiskers in the composites. Both the flexural strength and the fracture toughness of the composites were measured. The results suggest that the application of electric field improves the whisker orientation of the sintered composites. However, the improvement in the mechanical properties of the composites due to the orientation of the dispersed whiskers appears to be very small. Samples processed from flocculated suspensions produce poorly dispersed microstructure. *Permanent Address:
University of Cincinnati, Cincinnati, OH.
(4:40 p.m.) SILICON CARBIDE FIBER REINFORCED SILICON NITRIDE MATRIX COMPOSITES, R. T. BHATT, NASA Lewis Research Center, Cleveland, OH 44135. For advanced heat engine applications, materials are required that are strong, tough, and able to withstand aggressive environments at high temperatures and high heat fluxes. Fiber reinforced ceramic matrix composites have the potential for these applications because of their material tailorability and structural reliability. One fiber-reinforced ceramic system which shows these features is SiC!Si3N4 developed at NASA Lewis. These composites typically contain-30 vol% of high strength, high modulus CVD SiC fiber (Textron SCS-6) in a relatively porous Si3N4 matrix. In the as-fabricated condition, this composite system displays strain capability beyond matrix fracture~ graceful failure, notch insensitivity, and gauge length independent strength behavior. In this paper, the deformation behavior in tension and in compression, and the influence of matrix density and interfacial shear strength on the mechanical properties of unidirectionally and bidirectionally reinforced composites will be discussed.
ALUMINA AND BAUXITE I: Bauxite and
Equipment
Sponsored by TMS Aluminum Committee
STRUCTURE AND MECHANICAL PROPERTIES OF FIBER REINFORCED SiC AND B C COMPOSITES : K. Nakano and A. Kamiya, Government Industrial 4 Research Institute, Nagoya, 1 - I, JIil'ate-cho, Kita-ku, Nap,oya, Japan
Monday PM February 19, 1990
For the sake of imprcvement of fracture toughness in the S;C and B C ceramics, the composi.tes of these materials have been deveioped with incorporating of continuous carbon fibers. The unidirectional carbon fiber reinforced composites were fabricated by employing the filament winding method in which slurry consisting of matrix material powder, organosilicon resin and solvent was impregnated, followed by the hot pressing method. The composites had open porosities of 8 - 9 %(SiC) and 4 - 6 % (B C). Vf of the composites were 0.3 - 0.4. SiC composites ha~ 4 flexural strength of 400 - 600 MPa(RT) and 390 - 520 MPa(1200 C , Ar) and Kl of 10 - 16:4 MPaliil(RT). On the other hand B C com4 posites had flexural strength of 210 - 400 MPa(RT) and K of lc
Session Chainnan: Don Donaldson, APTEL, Walnut Creek, CA 94595
Room: Salon 3 Marriott
(2:00 p.m.) VENEZUELAN BAUXITE PROJECT: Esteban Quintero and Noel Marino, C.V.G. Bauxiven C.A., Avda La Estancia Torre Diamen, Piso 2, Chuao, Caracas 1060, Venezuela. This presentation sh~several aspects of the Venezuelan Bauxite Project to start operations by September 1, 1992.
28-Annual Meeting Some of the aspects to be presented in this paper are among others geographical location, environment, geological, production targets till maximum capacity, relevant engineering aspects and a description of an early production squeme that has permitted to start mining and sales possible before the overall project is finished. (2:30 p.m.) SOURCES OF BAUXITE AND ALUMINA IN NORTHERN BRAZIL-PRESENT SITUATION AND PERSPECTIVES: Euclides Penedo Borges, CVRD Companhia Vale do rio Doce, Av. Graca Aranha 26, 20-005 Rio de Janeiro, Brazil, South America. This paper presents an explanation on the existence and exploration of important deposits of bauxite ores in the Brazilian northern region and their main technical characteristics. It shows also the present and prospective production of alumina in the same area. The northern part of Brazil, specially the Amazon region, finds the requirements to become one of the world's main producers of bauxite and alumina. The availability of natural resources and industrial inputs, added to the continuous support represented by the Brazilian domestic market provide it with the capacity to produce and export substantial quantities of these products and to remain in the market , even at times when adverse conditions prevail. (3:00 p.m.) ELEVATED-TEMPERATURE HYDROGEN REDUCTION OF BAUXITE BEPORE BAYER LEACHING, ~. R. s.odgr.n , D. J. MacDonald, M. J. Zamzow, and D. E. Shanks, U.S. Bureau of Hine., Reno, NY
In Bureau of Min.. research r.lated to the Bayer process, it WIS found that treatment of bauxite with hydrogen under pre •• ure at temperatures from 250 0 to 450 0 C partially reduced ferric iron to magnetite, F. 30 4 , Magnetic susceptibility of reduced bauxite va. a maximum for a treatment temperature of 400 0 C, and the r •• idue after Bayer leaching was magnetic and amenable to magnetic filtration. This would facilitate .ore complete clarification of Bayer liquor. Bauxite'. organic matter waa partially converted to cO 2 by thermal decompoaition in the hydrogen atmosphere. The 0.2-pct original organic carbon content was decreased to 0.05 pet by treatment with hydrogen at 450' C. Treatment with hydrogen under pre.aure for 6 h at 380 0 C decreased aluminum leachability to 70 pet, compared to 97-pct leachability for untreated bauxite. But hydrogen treatment diminished the leachability of organic carbon more than it diminished the leachability of aluminum. For example, the ratio of organic carbon to aluminum in leach liquor waa cut by half aa a reault of treatment in hydrogen at 300' C.
(3:30 p.m.) EXPERIENCES FOR THE RECONSTRUCTION OF TRADITIONAL DIGESTION LINES WITH THE APPLICATION OF TUBE DIGESTION ELEMENTS IN THE HUNGARIAN ALUMINA PLANTS: E. Singhoffer Ede, Budapest, XIV, Kassai Ter 13, H-1142, Hungary; J. Steiner, Budapest, XI, Orlay U3, H-l1l7, Hungary. The experiment series with the tube-digestion elements, developed by ALUTERV-FKI (Hungary) carried out in the ALUMINA PLANT in ZHENG-ZHOU (CHINA) and ALUMINA PLANT AJKA (Hungary). Behave of diasporie bauxite-slurry by different temperature ranges, optimum velocity, scaling, erosion. Possibilities of the step by step reconstruction traditional digestion limes into tube - or combined digestion limes. New energy-saving methods for the control .of the recuperation. Results in the Alumina Plant in Ajka.
(4:00 p.m.) LIQUOR BURNING PROCESS··ITS MODIFICATION AND INCREASE IN CAPACI'l'Y: Y. Shibue, A. Sakamoto, N. Arakawa, Y. Kawai &: ~ Showa Denko K.K., 8, Ebisucho, Kanagawa·ku, Yokohama City, Kanagawa !'ref., Japan A procea for removing all organic substances from the Bayer liquor, the Liquor Burning Process (hereinafter called LBP)·· reported to the 112th AIME meeting in 1982··eontinues today with ita atisfactory operation. In 1988, Shows Denko modified the process, replacing the wet-type venLuri scrubber with a dry·type elec:trostatic precipitator. Before the modifi~tion, it required a ,reat deal of labor to remove scales generated inside Lhe wet·type venturi scrubber and thi. uAed to affect the operating efficiency. The modification has resulted in elimination of the generation of acales, bringing about a subetantial increase in the operating efficiency. A. 8 result, the copacity of the I,BP equipment has improved to a level equivalent to 120 percent of the original deRian capacity. Showa l>enko has also developed a technology to effectively leparate sodium oxalate. Combination of this technology and modified LBP will enable the removing of lOdium oxalate in the motIt effective manner.
(4:30 p.m.) NEW HORIZONTAL BELT FILTERS FOR A WIDE RANGE OF APPLICATIONS: David J~ Smith and Neil C. Beaton, Dorr-Oliver, Suite 840, 400 Perimeter Center, Terrace, N.E., Atlanta, Georgia 30346-2091, U.S.A. Horizontal vacuum belt filtration is well-established for slurry dewatering and cake washing duties. DorrOliver Europe has introduced two new horizontal vacuum belt filters which offer a number of advantages over traditional products used in this area, (a) a rubberbelt type with a novel belt support system and (b) a Rigid-Belt type. The rubber belt type incorporates a roller system for belt support, while the unique Rigid-Belt type eliminates the rubber carrier belt by substituting a series of interconnected metal trays containing filter grids. The benefits of the RigidBelt filter over continuous rubber-belt types, and over intermittent-vacuum types are reviewed. The commercial-scale application of both Dorr-Oliver filter types to a variety of washing duties such as pigments, alumina, and gypsum is reviewed, and aeveral applications involving dewatering of coarae alurries are alao discuseed.
CARBON TECHNOLOGY IT Sponsored by TMS Aluminum Committee (formerly Light Metals) Monday PM February 19, 1990
Room: Salon 2 Marriott
Session Chainnen: Euel R. Cutshall, Manufacturing Technology Laboratory, Reynolds Metals Company, Sheffield, Alabama 35660; Thomas R. Oaks, Northwest Aluminum Company, 3313 West Second Street, The Dalles, OR 97058 (2:00 p.m.) ANODE CAST IRON THICKNESS OPTIMIZATION, G. E. Goeres, M. D. Ohlswager, Aluminum Company of America, Wenatchee, Washington 99S01; R. W. Peterson, Aluminum Company of America, Alcoa Center, Pennsylvania 15069, USA Cast iron thickness of 24 commercial anode connections was varied by using stub holes 6.5 and 6.75 inches diameter with stubs 5.6, 5.8 and 6.0 inches diameter. Stub temperatures at time of casting were either 25'C or 150"C. Stub temperatures and stub· carbon resistances were measured during the first two day's pot operation. Stub-earbon resistance decreased and then leveled off with increasing stub temperature. This inflection point was an indication of tightness of the connection.
To avoid anode
breakage, this tightening should not occur prematurely. A stub hole 0.5 inch diameter larger than the stub was best for castings poured with 25'C stubs. (2:30 p.m.) ELECTRICAL BAKING OF IABORATORY..scALE CARBON ANODES UNDER HIGH PRESSURE, Feng Naixiang, Qiu Zhuxian, Northeast University of Tech., Shenyang, Uaoning Province, People's Republic of China; Uu Uanshuen, Zhang Shicheng, Fushun Aluminum Smelter, Fushun, Uaoning Province, People's Republic of China; Kai Grjotheim, Department of Chemistry, University of Oslo, P. o. Box 1033, Blindern, N.Q3IS Oslo 3, Norway; HaWr Kvande, Hydro Alliminium, P. o. Box SO, N· 1321, Stabekk, Norway. Laboratory.scale carbon anodes were baked electrically while the carbon·pitch matrix was maintained at pressures lip to 320 kg/cm2. Various mixtures of calcined coke, green petroleum cokes, natural graphite and coal tar pitch were made. The baked samples were analysed with respect to density, specific electrical resistance and crushing strength. Densities up to 1.71 g/cm3, specific resistances of abollt SO ohm mm2 m· 1 and crushing strengths above 400 kg/cm2 were measllred for samples made of SO mass % calcined coke, 10 mass % green petroleum coke and 10 mass % coal tar pitch, baked at 800"C under high pressure. 0:00 p.m.)
THE EFFECTS OF FEEDSTOCK AND COKE CALCINATION DEGREE ON THE PROPERTIES OF BENCH·SCALE ANODES, T. Mllftlloglu, R. Fernandez, Petroleum Coke Lab., STATOlL · Mongstad Div., P. O. Box 1103, N·S1S4 Mongstad, Norway. Bench·scale anodes were prepared from petroleum coke of different origin and varying degree of calcination. Aggregate size distribution, pitch type and content, paste mixing and fonning parameters and anode baking conditions were the same for all anodes. The aggregate contained 16% butts in all cases. However, the cleanliness of the butts was different and butts with four levels of sodium contamination were utilized in preparing
Annual Meeting-29 anodes from each coke type. The sensitivity of the anodes to increasing content of sodium was determined by measuring the reactivity in CO2 and air. CO2 reactivity of the anodes mcreased significantly with increasing sodium content. The catalytic effect of sodium was strongly affected by the coke calcination degree and to a lesser extent by the feedstock which was used for green coke production in this work. The air reactivity of the anodes, on the other hand, showed only a slight dependence on the sodium content and the trend was similar for the different types of coke. Coke calcination degree combined with the feedstock type, however, had a strong effect on the air reactivity of the resultant anodes. The anodes produced from low-calcined coke showed very poor physical and mechanical properties while the effect of feedstock on these properties was not significant.
0:30 p.m.) SODERBERG PASTE. EFFECT OF FINE FRACTION VARIATIONS, Per Stokka, Norsk Hydro Research Centre, N·39!O Porsgrunn, Norway; Ivar Skogland, Hydro Aluminium, Karmlly Plants, N-4265 H~vik, Norway. When producing SIlderberg paste the fine fraction is the largest and most difficult fraction to control. This work demonstrates the effect of varying fineness of the fine fraction on the vibrated bulk density of the whole aggregate and the amount of pitch needed to obtain a certain Ilowability.
(4:00 p.m.) APPLICATION OF ROBOT TECHNIQUE IN ANODE PLANT, H. A Meier, B. Zito, G. Coppo, ALUCENTRO, Divisione dell'A1usuisse Italia SpA, Stabilimento di Porto Marghera, Casella postale 569,1·30170 Mestre P.T., Italy. Anode cleaning is an unpleasant operation characterized by repetitiveness, noise, dust and heat. A first automated cleaning unit, using advanced robot technology, was installed in 1987 at ALUCENTRO, an ALUSUISSE plant in Porto Marghera (Venice, Italy). The System works successfully and allowed to eliminate the physical presence of the workers in a hygienically critical environment.
11ffi PYRElRON VARIABLE RATIO OXYGEN/AIR/FUEL BURNER SYSTEM FOR ALUMINUM MELTING: H. E. Phipps, H. E. Phipps Company, P. O. Box 1164, Roxboro, NC 27573 and G. M. Gitman, American Combustion, Inc., 2985 Gateway Drive, Norcross, GA 30071. The PYRElRON Burner System was developed to increase melt rate up to 50%, decrease fuel consumption, lower oxides of nitrogen generation on a per pound throughput basis, decrease combustion volume by up to 50%, decrease melt loss and increase furnace refractory life. This is accomplished by simultaneously increasing flame temperature, velocity and luminosity to enhance heat transfer efficiency. The versatility of the process extends from impingement conditions present.when charging prime metal into the interior of a circular, top charge melter to the sItuation where all heat is transferred through the interior hearth bath surface for an exterior charge well type furnace melting contaminated scrap.
(2:50p.m.) SETTLING OF INCLUSIONS IN HOLDING FURNACES MODELING AND EXPERIMENTAL RESULTS. C. Sztur, F. Balestreri, J L Meyer, B. Hannart, Pechiney CRV S.A., B.P. 27, 38340 Voreppe (France) Description of settling phenomena usually refers to falling particles in a liquid, following Stokes law. But the thermal convection always takes place in holding furnaces due to temperature heterogeneity, and the behaviour of the inclusions can be dramatically influenced by the liquid metal motion. A numerical model based on turbulent fluid flow calculations in an holding furnace and on trajectories calculations of a family of inclusions has been developped. Results are compared with experiments on a lab. scale and on an industrial scale furnace. An analysis of the governing parameters will be presented.
(3:15 p.m.)
(4:30 p.m.) CARBON PLANT RETROm TO MEET HIGHER PRODUCTION AND QUALI1Y REQUIREMENTS, Ana Mabel Vivas, Ram6n Betancourt, Venalum, Industria Venezolana de A1uminio, CA, Apdo. 51.963, Caracas 106, Venezuela. Venalum is expanding its smelting capacity to 456.000 M.T., in order to accomplish it, anode production must be increased. This paper highlights the green mill retrofit to achieve the required anode production. A description of the major modifications is given together with the benefits obtained by the retrofit project.
CAST SHOP TECHNOLOGY IT: Melting Technology Sponsored by the TMS Aluminum Committee Monday PM February 19, 1990
(2 :25 p.m.)
Room: Salon 4 & 5 Marriott
Session Chairman: S.A. Levy, Reynolds Metals Company, Metallurgy Laboratory, Fourth and Canal Streets, Richmond, VA 23261 (2:00 p.m.) OPTIMAL FUEL CONTROL OF A CASTING FURNACE. R.T. Bui, R. Ouellet and J. Perron, Universite du Quebec a Chicoutimi, Chicoutimi, Quebec, Canada, G7H 2Bl This furnace plays a central role in the fabrication of aluminum. In 1987 a 50-equation analytic model was built and validated. , It was shown to be a reliable tool for simulating and analyzing the dynamic behavior of the furnace. But i t is too cumbersome for control purpose. In this work, the model is reduced to 10 equations using least square approximation applied to data obtained from simulations made with the analytic model. Then an optimization is performed using the Pontryagin maximum principle, to bring a given mass of metal from an initial state to a target temperature in a prescribed time with minimum fuel. It is formulated as a two-point boundary-value problem and solved by the relaxation method. This leads to a substantial fuel savings. The conclusion to be retained is that this complex metallurgical process can not only be represented by an analytic model but also be optimally controlled using a simplified model. This should induce new thinking in the important field of fuel-optimal furnace control.
BREAK
(3:30 p.m.) INTERFACIAL TENSION IN MOLTEN ALUMINUM ALLOYS AND SALT SYSTEMS: Francis K. Ho and Yogesh Sahai, Materials Science &Engineering, The Ohio State University, 116 W. 19th Ave., Columbus, OH 43210. In the secondary industry, melting of hard-to-melt aluminum scrap such as Used Beverage Cans (UBC's) is normally done under a cover of molten salt flux usually based on sodium and potassium chlorides. During the melting, the salt prevents oxidation of the metal and may assist in the coalescence of molten aluminum alloys. The wetting characteristics and the interfacial tension between the metal and salt strongly influence the movement and separation of the dispersed metal in the salt matrix. In this paper, an apparatus for the measurement of interfacial tension in molten aluminum alloys and salt systems by drop detachment method is described. The interfacial tension results of aluminum alloys in clean salts of varying composition are presented. The potent i ali nfl uence of salt fl ux compos it i on on recycling of UBC's is discussed. (3:55 p.m.) FUNDAMENTAL INVESTIGATIOOS AOOUT Mm'AL LOSS DURING REMELTING OF EXTRUSlOO AND ROLLING FABRICATICN SCRAP. H. Rossel, Vereinigte Aluminiurn-Werke Nl., Georg-von-BoeselagerStraBe 25, D-5300 Bonn 1, West Gernany.
!-Etal loss is an inevitable part of aluminum scrap rrelting and represents one of the main cost factors in a smelting plant. The purpose of this paper is to shCM basic correlations between rrelt loss, alloying elerrents (M:J and Mn), scrap gearetry (specially the thickness of the scrap) and process parameters as teITq?erature during rerrelting of different types of scrap. Resuits from experi.rrents with scrap of the alloys AA 1050, AA 3004, AA 3103 and AA 5082 will be presented and discussed.
(4:20 p.m.) IMPROVING CAST-HOUSE QUALITY - TIIE REFRACfORY WAY: S. R. Sibley, Marketing Manager, Light Metals Division, Foseco International Ltd., 285 Long Acre, Nechells, Binningham B7 5JR, United Kingdom. No matter how efficient any in-line filtration system is, there is always a distance between the cleaning device and mold. Refractories used for transportation and control are critical for the optimum quality of billet, slab, etc., that are being produced. This paper discusses critical properties of refractories and how recent development work of end user requirement has been harnessed with technology in order that tailor made refractories can be produced for all areas of the cast-house including filtration media thus minimizing refractory media as a source of harmful inclusion to the billet, slab, etc. This paper will also briefly review the history of refractory demand in line with cast-house technology and give some vision as to how re.fractory technology and awareness can lead casting technology as opposed to followmg such technology. Impact on costs are summarized too.
30-Annual Meeting (3:20 p.m.)
CREEP TMS General Abstract Session Monday PM February 19, 1990
Room: Plaza B Inn at the Park
Session Chairman: J.C. Gibeling, Department of Mechanical Engineering, University of California, Davis, CA 95616 (2:00 p.m.) TRACE ELEMI!1IT EFFECTS ON HIGH TEKPERATUIIE IlITERGRABULAIt FRACTIJlIE. F. Cosandey, Dept. of Mechanics & Materials Science, Rutgers University, P.O. Box 909, Piscataway, NJ 08855-0909. The creep behavior of Ni-20 Cr alloys containing trace additions of B, ee, Zr and Kg has been investigated as a function of stress and te.perature. For all trace ele.ent content, mini.u. creep rates are reduced and increases in creep activation energy have been ~asured. For B, Ce and Zr, increases in ductility and fracture tiae are associated with change in fracture mode from intergranular to transgranular while embrittleaent is induced by Mg. The effect of various trace element content on grain boundary structure, .orpbology and segregation has been measured by TEM~ The creep results have been analyzed in ter.s of creep cavitation models. The relative .agnitude of various trace ele.ent content on grain boundary cohesion, stress concentration and stress relaxation processes will be discussed. This work is supported by the Materials Division of NSF. (2:20 p.m.) HIGH-TEMPERATURE CREEP DEFORMATION AND TEM STUDIES OF A W-4.0Re0.33HfC ALLOY: B. l. Chen, D. l. Jacobson, and K. S. Shin, Dept. of Chemical, Bio and Materials Engineering, Arizona State University, Tempe, AZ 85287 Tungsten-base alloys have great potential for future space power systems. For such applications the understanding of creep behavior of these alloys at ultra-high temperatures is very important. In the present study, the creep behavior of a W4.0Re-0.33HfC alloy was examined at various temperatures above The 1900 K in the applied stress range of 20 to 70 MPa. activation energy and the stress exponent for high-temperature creep were determined. The dislocation structure was examined with transmission electron microscopy and the structure was correlated with the creep data. It was found that subgrains formed during creep deformation and the subgrain size was inversely proportional to the applied stress. The effect of stress on the dislocation structure was also examined. In the light of these experimental observations, the creep mechanism of a W-4.0Re-0.33HfC alloy will be discussed. (2:40 p.m.) Creep in an A1-5t Ag Alloy: Ho-Kyung Kim and Farghalli A. Mohamed, Materials Section, Department of Mechanical Enqineerinq, University of California, Irvine, California 92717.
The creep characteristics of A1-5t wt Ag, including the stress exponent, the activation energy for creep, and the stress dependence of the subgrain size, were investigated over wide ranges of temperature and stress. Also, the stress exponent obtained form stress reduction tests was compared with that determined from uninterrupted tests. The experimental data obtained for the alloy are discussed with reference to the results of recent studies on the creep behavior of Al and its alloys. (3:00 p.m.) Creep Behavior and Substructure in an AI-Li Alloy: ~ TAe Park, Enrique Lavernia and Farghalli Mohamed, Materials Section, Department of Mechanical Engineering, University of California, Irvine, California 92717. The ~ffect of stress and temperature on the creep behavior of an Al-2.l wtt Li alloy was studied in the solid-solution range. In addition, the dislocation substructure that developed during creep was examined by means of transmission electron microscopy. The results show that under the experimental conditions used in this investigation, AI-2.l wtt Li behaves as a class II alloy (metal type).
Creep Behavior of Silicon Carbide particulate Reinforced 6061 Aluminum: Kyunq-Tae Park, Enrique .Lavernia and Farghalli A. Mohamed, Materials section, Department of Mechanical Engineering, University of California, Irvine California 92717. ' The effect of stress and temperature on the creep behavior of SiCp-6061A1 was investigated. The experimental data show that the steady-state creep rate of the composite depends strongly on the temperature and applied stress and that the stress exponent, when measured over a wide range of strain rate, is not constant but increases with decreasing stress. The experimental data obtained are examined in the light of recent results on the mechanical properties of discontinuous metal-matrix composites at elevated temperatures. (3:40 p.m.) CREEP RUPTURE OF 304 STAINLESS STEEL UNDER MULTIAXIAL STRESSES: H. K. Kim, F. A. Mohamed and J C. Earthman, Materials Se~tion, Department of Mechanical Engineering, University of California, Irvme, CA 92717 Creep rupture by intergranular cavity growth has been studied extensively, most commonly under uniaxial stress conditions. Although uniaxial stress experiments have led to a good understanding of the physical processes involved, they do not provide sufficient information to predict cavity growth and creep rupture under multiaxial stress conditions. To examine the effect of stress state on creep rupture behavior, specimens of 304 stainless steel were tested at elevated temperatures under one of three stress states: uniaxial tension, biaxial shear and triaxial tension. The effect of stress state on the development of cavitation damage was examined using SEM techniques. The relationship between the rupture time and the stresses which drive intergranular cavitation is discussed. (4:00 p.m.) DIFFUSION-ACCOMMODATED SLIDING OF IRREGULARLY SHAPED GRAIN BOUNDARIES.* J. H. Schneibel and P. M. Hazzledine, Metals and Ceramics Division, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831 The Raj and Ashby model for diffusion-accommodated grain boundary sliding is restricted to periodic boundary shapes with small amplitudes as compared to the wave length. As an extension to Raj and Ashby's work we present, for the case of grain boundary diffusion control, a method to evaluate the sliding rates and normal stress distributions at arbitrarily shaped boundaries in 2-dimensional bicrystals. Our results are found to be consistent with Raj and Ashby's analysis. Several examples for irregularly shaped boundaries are presented and discussed. It is found that the boundary conditions at the grain boundary end points can strongly influence the stresses at the grain boundaries. *Research sponsored by the Division of Materials Sciences, U.S. Department of Energy under contract DE-AC05·840R2l400 with Martin Marietta Energy Systems, Inc.
(4:20 p.m.)
CREEP BEHAVIOR AND MICROSTRUCTURE OF TiAl/Ti 3Al XDTM COMPOSITES: C R Feng-, H. H. Smith, D. J. Michel and C. R. Crowe, Materials Science and Technology Division, Naval Research Laboratory, Washington, DC 20375-5000. The creep behavior and microstructure of extruded Ti-46at%Al alloy and XD™ processed and cast/extruded Ti-47at%Al/7.5v% TiB2 composite were investigated. The creep tests were conducted at 10 ksi at 760°C following duplex heat-treatment at 1200°C/18h plus 900°C/8h. The results show that the composite exhibited an increased creep strain and creep rate when compared with the non-TiB 2 containing alloy. The microstructure of the materials both prior and subsequent to creep testing were examined by TEM methods. The implication of the observations will be discussed. (4:40 p.m.) CREEP OF COPPER UNDER CONSTANT STRUCTURE CONDITIONS: G.S. Nakayama and J.C. Gibeling, Division of Materials Science and Engineering, University of California, Davis, CA 95616. A series of stress reduction experiments conducted on pure copper was analyzed. Initial stresses of 68.9 HPa at 673 K and 30.3 HPa at 873 K were used, and stress reductions of 2.5% to 15% were made at 10% strain during steady state creep. The exponential dependence of constant structure strain rate on the reduced stress is consistent with the result previously obtained for pure aluminum. These data provide additional support for the exponential form of the kinetic law for obstacle-controlled dislocation gl ide. In addition, the average activation free enthalpy for dislocation glide determined from the constant structure data is approximately equal to typical measured values of the creep act i vat i on energy. Based on a compari son of the
results for aluminum and copper, it is suggested that the stacking fault energy may influence both structure evolution and dislocation glide past obstacles . (5:00 p.m.) A STUDY OF THE GROIITH OF GRAIN BOUNDARY CAVITIES IN TUNGSTEN/COP· PER COMPOSITES DURING HIGH· TEMPERATURE FATIGUE : Yong'Suk Kim, M.
J . Verrilli , and T. P. Gabb, NASA Lewis Research Center, Cleve · land , OH 44135 .
Previous research on the cyclic failure of ~he W·fiber reinforced Cu at high temperatures has shown that " the composite fails primarily due to the Cu matrix degradation, which is characterized by the growth and coalescence Qf grain boundary cavities. The current research has focused on evaluating this grain boundary cavity growth mechanism . The understanding of the mechanism is necessary for a proper lifetime prediction of the composite in high· temperature engineering structures. A pre ' creeping technique is used to create a regular and uniform initial cavity distribu· tion along grain boundaries of the Cu matrix. Cyclic failure data of these pre'cavitated specimens are obtained under different temperatures and stresses . The comparison of the data with several theoretical predictions, together with metallographic and fractographic analysis provides an understanding of the operative cavity growth mechanism . Special attention has been paid to the effect of grain boundary sliding and grain size of the Cu matrix .
DEFECTS AND DEFECT REDUCTION PROCESSING IN SEMICONDUCTOR HETEROSTRUCTURES TI: Epitaxial Stabilization: a-Sn and Sn1.KGeKAlloys Sponsored by the TMS Electronic and Photonic Device Materials and Electronic Materials Committees Monday PM February 19, 1990
Room: Orangewood II Quality Hotel
Annual Meeting-3! (2 :50 p.m . ) GROWTH AND CHARACTERIZATION OF n- Ge,Snl-l< HETEROEPITAXIAL ALLOYS ON (001) InSb, GaSb. and InSb/GaAs. E.A. Fitzgerald. P.E. Freeland. M.T. Asom. W. Lowe, R. MacHarrie, Jr., A.R. Konan, Y.H. Xie. F.A. Thiel. A.M. Sergen~ L. Cooper, G.A. Thomas. K.A. Jackson. B.E. Weir. G.P. Schwartz. GJ. Gualtieri. and L.C. Kimeriing. AT&T Ben Laboratories, Murray Hill. NJ 07974. We have grown MBE n-Ge, Snl_, (O-dimensional. a gradual shift 10 three-dimensional. twinning. and finally phase separation. Double and uiple crystal x-ray results show that totally strained (up to £..0.84%). n-Ge,Snl-l< is stabilized willI x<=0.08. For x
(3: 10 p.m.) The Growth and Characteristics of Metastable SnGe Alloy Films Prepared by Molecular Beam Epitaxy. L.W. Tu, ~ Zhang, X.J. Yi, G.K. Wong, and J.B. Ketterson, Northwestern University -- We report molecular beam epitaxy growth of SnGe alloy films. Using (001) CdTe substrates, we have grown homogeneous alloys of Snl _ ~Ge with x < 0.1. X-ray diffraction studies show tnat' these films grow pseudomorphically, resulting in a tetragonal dilation in the growth direction for the pure gray Sn films and a contraction for the SnGe alloy films. Hall effect measurements indicate that the alloy films in general have higher carrier concentrations and lower mobilities than the pure gray Sn films. The alloy films with higher Ge concentration were observed to change from n- to p-type conduction as the temperature was lowered. We have also studied the growth of alloy films using (001) Si and GaAs substrates. The results of these studies will also be reported.
(3:30p.m.)
BREAK
(3:45 p.m.)
Chainnan: L.C. Kimerling, AT&T Bell Laboratories, Murray Hill, NJ 07974
GROWfH AND PROPERTIES OF SNGE FILMS GROWN BY MOLECU· LAR BEAM EPITAXY. S. S. Iyer, Alex Harwit, 1. C. Tsang. V. P. Kesan , and 1. AngileUo , IBM Research Division, T. 1. Watson Research ("..enter. P. O. Box 218, Yorktown Heighls. NY 10598. U.SA
(2:00 p.m.)
x up 10 0.3. - The fUms were deposited at 140°C on 5OO0A thick gennanium buffer layers grown on silicon (I (0) substrates. The fIlms have been analyzed structurally
Invited
HIGH RESOLUTION X·RAY DIFFRACTION FROM EPITAXIAL FILMS . A. R. KOrlan, AT&T Bell Laboratories, Murray Hill, New Je~ey 07974. High Resolution X·ray Diffraction (HRXD) is a powerful and non·destructive probe for the structural characterization of epitaxial films. The kinematical part of the diffraction is routinely used to study the atomic structure and to quantitatively identify the strain parameters with a very high precision; typically to a few parts of ten·thousandth of the bond length. In the triple crystal geometry instrumental resolution for the rocking scans can be chosen as small as few arc· seconds. For high quality films, the dynamical diffraction contribution to the scattering also becomes significant which, through careful analysis yields the film thickness. In this talk the state of the art HRXD technique, instrumentation and its applications to epitaxial films will be reviewed in detail.
(2: 30 p.m.)
Invited
EPITAXY·INDUCED STRUCI1JRAL PHASE TRANSFORMATIONS. I Sverre Froyen, S.·H. Wei and Alex Zunger, Solar Energy Research Institute, Golden, CO 8040 I. * Using our calculated structural energies for CdTe, MgS, and NaC!, we describe how hydrostatic pressure versus volume data can be related to epitaxial structural energies. We shown that coherent epitaxial growth away from lattice-matched conditions can stabilize pseudomorpbic phases thai are normally inaccessible (by pressure). We find that the stability region of j3-Sn CdTe is increased and that MgS can be grown in the zincblende phase. This can be modeled by geometric constraints imposed by the substrate and by simple elastic models. Finally, we discuss how the nucleation of misfit dislocations will limit the thickness of these pseudomorpbic phases to a few monolayers; even when the pseudomorphic phase is lattice matched to the substrate. It is likely, however, that energy barriers will permit tbe growth of much thicker metastable films. *Supported in part by OER-BES, Materials Research Division. IS. Froyen, S.·H. Wei, and A. Zunger, Phys. Rev. B 38, 10124 (1988).
Using molecular beam epitaxy. we have succesfully grown thin films of Ge, _,Sn, with using Auger Electron Spectroscopy. X-Ray Diffraction , and Rutherford Backseattering. The electrical and optical propenies have been investigated by Hall effect. Raman, and infrared absorption spectro""opy. The fIrst layers of tin accumulate on the surface, however. as the growth proceeds tin begins to incorporate until a sleady state growth is achieved. Increased optical absorption has been observed at long wavelengths in higher tin content alloys. and the magnitudes suggest that it is from the direct band-gap of the semiconductor. On exposure to air a thin oxide layer is fonned . and Ihe fIlms are found to be unstable above 190°C. We are presently in· vestigating the photoconductive properties of SnGe. • P.R. Pukite, Alex Harwit, and S.S. Iyer. Appl. Phys. Lett . 54. 2142 (1989).
(4:05 p.m.) A TWO-DIMENSIONAL ELECTRON GAS (2DEG) AT TIlE INTERFACE OF A NONPOLARJPOLAR SYSTEM: a-SnJInSb(IOO) AND (111~ W. T. Yuc:n.· W. K. Liu, t s. N. Ho1mea.t
Labor-,..
R. A. Sttadlinl·t ond B. A.Joyce-. tBIackeu Imperial eon... of Science, Tedmolol)' ond Medicine, Prince COIIIOft Rood. London 2W7 2BZ. -lnterdUciplirury R _ C..- of S.bondo. The properti.. of IhiJ 2D£G .... pudy o&cted by 1M Ihiclrnesl of 1M a-Sn films ond 1M orimtation of 1M _ t t..... CNly danonslr&ling 1M interfacial charOCI directions. Finally steps are observed IS the epiJayer thickness approaches O.21J.m. 1lUs behaviour 5uggeslS that • mechanism involving bond breaking ond bonding of Sn wi!h In ond Sb llOms is takin, place at !he inIerface. This meclwUsm has been proposed for explaining lIIe absence of on inIerface dipole II 1M heterojunc:tion of 1M Iaaicemacched nonpolar/polar GelGaAs( IOO) sYSJem. Possible origiN of !hi. 2DEG in the a-SnJInSb .YSJem willihen be discussed in !he liel" of Ibe prosenl daIa.
,IS
.ud..,..
(4:25 p.m.)
Shubnikov-de Haas Effect in Thin Epitaxial Gray Tin Films. LoW. Tu, S.N. Song, Z. Zhao, G.K. Wong, and J.B. Ketterson, Northwestern University -- The transverse
32---Annual Meeting magnetoresistance and Hall effect have been studied for n-type gray tin epilayers grown on (001) CdTe substrates by molecular beam epitaxy. Shubnikov-de Haas oscillations are observed in samples having Hall 2 mobilities ~104 cm /Vs at low temperatures. Measurements were carried out using the fieldmodulation techniques in static fields up to 10 Tesla and for temperatures between 1.2 to 10 oK. Beat patterns are observed in the Shubnikov-de Haas spectra, which we ascribe to either inhomogeneous doping, arising from the diffusion of Cd and Te from the substrate, or quantization of electron motion in the direction parallel to the film normal. When the magnetic field is oriented at 45° and 90° away from the film normal, clear oscillations set in at a field much lower than the expected size-effect-cut-off field, implying that specular reflection at the film boundaries plays a significant role.
FRACTUREll TMS General Abstract Session Monday PM February 19, 1990
Room: Plaza C Inn at the Park
model which incorporates the crack velocity dependence of energy absorption by crack tip plasticity. Supported by Los Alamos National Laboratory and the Office of Naval Resp.arch. 0:00 p.m.)
MICROMECHANISMS OF STATIC CRACK INITIATION AND GROWTH IN AN a. TITANIUM ALUKINIDE ALLOY*: K. S. Chan, Southwest Research Institute, 6220 Culebra Rd., San Antonio, TX 78228-0510. The deformation and fracture behaviors of Ti-24Al-IINb with an equiaxed a. + ~ microstructure have been characterized by performing tensile and JIC tests at 25·C to 600·C. In addition, the micromechaniBms of crack initiation and growth have been studied by postmortem examination of the fractured specimens and by in-situ observation of the fracture events in a SEM equipped with a loading stage. The results indicate that the yield and the ultimate tensile strength are decreased, while the strain hardening exponent, elongation, true fracture strain, the J 1C value, and the tearing modulus are increased with increasing temperatures. Crack growth in Ti-24AI-IINb occurs by a process that involves nucleation and linkage of microcracks and crack bridging in the process zone. The microcracks are initiated either by decohesion of slip bands in a. or by tearing at or near the a. + ~ interfaces. *Supported by the MOSR through Contract No. F49620-89-C-0032. 0:20 p.m.)
Session Chainnan: K.S. Chan, Southwest Research Institute, P.O. Drawer 28510, 6220 Culebra Rd., San Antonio, TX 78284
CRYOGENIC MECHANICAL PROPERTIES OF VINTAGE III ALUMINUM-UTIllUM ALLOY 2090, * ll..!Jnl and J.W. Morris Jr., Center for Advanced Materials, Lawrence Berkeley Laboratory, and Department of Materials Science and Mineral Engineering, University of California, Berkeley, CA 94720.
(2:00 p.m.)
The mechanical properties of a new vintage 2090- T81 plate material were evaluated at 300, 200, 77, and 4K by means of tensile and toughness tests and compared with those of an earlier vintage. Previous work on the earlier vintage has shown a simultaneous increase in both strength and toughness with decreasing temperature. Previous work has also shown extensive delamination and inhomogeneity in the through-thickness properties. The new vintage was produced in hopes of reducing this delamination and inhomogeneity. Results on the strength and toughness of the new material will be reported.
MOLECULAR MECHANICS··A BLEND OF MOLECULAR DYNAMICS AND CONTINUUM MECHANICS. D J Quesnel and D. A. Ekeren. Mechanical Engineering Department. University of Rochester. Rochester. NY 14627 Even with our rapidly advancing progress in computational resources. we will not be able to achieve molecular dynamic representations of life sized objects. Avogadro's Number is just too large. Continuum mechanics and its numerical counterpart. finite element analysis. have been shown to represent the macroscopic elastic behaviors well while molecular dynamic representation of small portions of material appear to have realistic representations at the molecular and hence mechanistic level within their capability. In this work. we explore a method of joining these computational tools for the purpose of examining the fracture behavior of materials. A method is developed to transform a crack tip region represented by continuum mechanics into a region represented by molecular dynamics at the crack tip and a region represented by continuum mechanics in the far field. all the while maintaining a coherent boundary between the two computational domains. Some preliminary results will be presented and discussed. (2:20 p.m.)
MICROSTRUCTURAL CONTRIBUTIONS TO THE ONSET OF CRACK INSTABILITY: J. E. Hack. Department of Mechanical Engineering. Yale University. New Haven. CT 06520-2157. A recently developed model for crack instability in quasi-brittle materials which incorporates the crack velocity as a controlling parameter will be extended to include precursor events which or-cur at the microstructural scale. The creation of propagating microcracks in the highly stressed region ahead of a pre-crack and the role of pre-existing dislocations will be treated. It will be shown that a criterion for crack instability based upon the microstructure of a material can be deduced by balancing the microcrack and resulting macrocrack velocities. Distinctions between the fracture behavior in low dislocation density materials. such as as-grown single crystals. and thermo-mechanically processed structural materials also will be discussed. Where possible. predictions of the model will be cOlpared with available experimental data Supported by LANL through the ECUT Program of the USDOE (2:40 p.m.) CRACK VELOCITY DEPENDENCE OF FRACTURE TOUGHNESS IN REFRACTORY METALS AND ALLOYS: D. Rittel. J. G. Kosinski. Jr. and J. E. Hack. Department of Mechanical Engineering. Yale University. New Haven. CT 06520-2157.
Results of experiments aimed at correlating fracture toughness levels with initial crack velocity and the level of crack tip plastic deformation will be reported. Estimates of crack tip plastic deformation will be obtained with Electron Backscattering Diffraction. Materials used in the study include pure polycrystalline Wand a W based heavy alloy. In addition. initial experiments on single crystal NiAI will be discussed. Where possible the data will be compared to predictions of a
* Supported by the DOE Research, Office of Basic Energy Science, U.S. Department of Energy. D. Chu is supported under a National Science Foundation Graduate Fellowship. 0:40 p.m.)
~~~~~:~I~~S ALOJMI~~~~~i~~M ~~tgis~Ri. T~~~n~~e~;;'~ ~:a
and R. O. Ritchie, Center for Advanced Materials, Lawrence Berkeley Laboratory. University of California, Berkeley. CA 94720.
Prior studies on the cryogenic properties of AI-Li alloy 2090- T8 have shown remarkable improvements in strength. elongation and toughness with decrease in temperature from ambient to 4 K. In this presentation. fracture-toughness (K Ie> and resistance-curve (R-curve) behavior in 12.7 mm plates and 1.6 mm sheets of the peak-aged 2090- T8 alloy are examined as a function of temperature. with emphasis on the mechanistic role of crack-divider delamination toughening. In the absence of changes in the microscopic fracture mechanism, thick plates showed an increase in toughness with decrease in temperature due to a greater tendency for through-thickness splitting at low temperatures. Conversely. thin sheets display reduced toughness at lower temperatures. associated with changes in the fracture mode (plane-stress/planestrain) over this temperature range. Models for these toughening mechanisms are presented. ·Sponsored by the Department of Energy, Office of Basic Energy Sciences. (4:00 p.m.)
STUDY OF LIQUID OXYGEN (LOX) COMPATIBILITY BEHAVIOR AND RELATED MICROSTRUCTURAL EFFECTS OF AI-Li ALLOYS, *R.M. DIWAN, Mechanical Engineering Department, Southern University, Baton Rouge, LA 70813 The aluminum-lithium Al-Li based alloys provide significant advantage~ over the current conventional alloys for advanced aerospace applications, and in particular as cryogenic mate-
rials. The LOX compatibility evaluations are carried out of the Al-Li 2090 and Weldalite 049 alloys employing liquid oxygen impact test conditions under ambient pressures and upto 1000 psi. The behavior of these alloys is analyzed and related to microstructural effects using optical metallography, scanning electron microscopy, and surface studies using Secondary Ion Mass Spectrometry SIMS, Electron Spectroscopy in Chemical Analysis ESCA and Auger Electron Spectroscopy. Differences in the behavior of these alloys will be discussed and related to their microstructural aspects. This research has been supported by the ASEE/NASA sponsored program NASA NGT-Ol-008-02l.
Annual Meeting-33. (4:20 p.m.) DUCTILITY AND FRACTURE IN B·DQPED NiAl Alloys: * c....I...LilI. and E. P. George, Metals and Ceramics Division, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831-6115. Stoichiometric NiAl showed mainly brittle grain-boundary fracture, with limited tensile ductility (-2%). Our Auger studies reveal that grain boundaries in NiAl are extremely clean and free from any detectable impurities. The boundaries are thus considered to be intrinsically weak. In an attempt to improve the grain boundary properties a series of NiAl alloys was prepared with up to 300 wt ppm B. Boron is found to be very effective in suppressing intergranular fracture in NiAl. However, boron is a potent solid solution strengthener, and it reduces the tensile ductility of NiAl at levels greater than 100 ppm. Careful control of boron level is thus needed in order to achieve both strength and ductility of NiAl. *Research sponsored by the Office of Energy Utilization Research, Energy Conversion and Utilization Technologies (ECUT) Program, U.S. Departroent of Energy, under contract DE-AC05-840R21400 with Martin Marietta Energy Systems, Inc.
(4:40 p.m.) CHEMISTRY OF GRAIN BOUNDARIES IN MICROALLOYED NiAl:* E P George and C. T. Liu, Metals and Ceramics Division, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831-6093. Stoichiometric NiAl was microalloyed with boron, carbon and beryllium in an attempt to overcome its intrinsic intergranular brittleness. It is found that boron is very effective in suppressing grain boundary fracture, whereas carbon and beryllium do not affect the predominantly intergranular fracture mode of NiAl. In order to understand these macroscopic observations, we are using Auger electron spectroscopy to study the microchemistry of grain boundaries in both doped and undoped NiAI. Boron is found to segregate to and strengthen grain boundaries in NiAl, thereby suppressing intergranular fracture. Carbon, on the other hand, does not segregate to the grain boundaries; as a result, it is unable to suppress intergranular fracture. We will also discuss the segregation behavior of the constituent elements (Ni and AI) in these alloys, and relate the overall grain boundary chemistry to fracture behavior. *Research sponsored by the Office of Energy Utilization Research, Energy Conversion and Utilization Technologies (ECUT) Division, U.S. Department of Energy, under contract DE-AC05-840R21400 with Martin Marietta Energy Systems, Inc.
mGH PERFORMANCE COPPER·BASE ALLOYS II Co-sponsored by the TMS Structural Materials and Powder Metallurgy Committees Monday PM February 19, 1990
(2 :00 p.m.)
RELATIONSHIP OF STRENGTH AND CONDUCTIVITY IN DEFORMATION PROCESSED COPPER-REFRACTORY METAL ALLOYS: J.D. Verhoeven, H. Downing, W.A. Sp;tzig, E.D. Gibson, L.S. Chumbley, I.E. Ames
Laboratory*,
Iowa State University,
(3:00 p.m.)
QUENCH HARDENED COPPER-BERYLLIUM ALLOYS FOR WEAR RESISTANCE: A. Guha and W. R. Cribb, Brush Wellman Inc., Alloy Division, 17876 St. Clair Avenue, Cleveland, OH 44110 A series of binary copper-beryllium alloys were investigated close to the eutectoid composition containing 6.10 wt.% beryllium. These alloys exhibit unusual hardening during rapid quenching from high temperature above the eutectoid temperature of 620 C. Room temperature evaluations show that hardness as high as Rockwell C58 can be obtained in the as-quenched state. A characteristic lamellar structure is observed as a result of isothermal transformation below the eutectoid temperature. Long term exposure at elevated temperatures up to 600 C shows unusual thermal stability of the lamellar structure. Tensile behavior at room temperature shows that high strength with limited ductility is obtained from these alloys. Preliminary evaluations indicate that outstanding wear resistance with reasonable conductivity is obtained from these alloys. Extreme hardening in the as-quenched state is explained by the transformations that occur during quenching. (3:30 p.m.) DEVELOPMENT OF COPPER ALLOYS BY NON-EQUILIBRIUM AND HYBRID PROCESSES: RAPID SOLIDIFICATION, SOLID STATE MICROBLENDING, AND PHASE REDISTRIBUTION PROCESSING:.s. ~, Battelle, Columbus, OH 43201 and A. Patel, Lanxide Corp., Newark, DE 19711
There is now general recognition that significant morphological, compositional, and crystallographic modifications can be effected in many alloy systems by non-conventional processing methods. This has given rise to vigorous alloy development efforts to meet the challenging requirements that modem materials must address. Alloy systems which earlier were difficult or even impossible to produce with acceptable microstructure and properties and in useful forms can now be regarded as potential candidates for commercial applications. Copperbased alloys, because of their ductility and electrical and thermal conductivity and other desirable properties, are among those receiving increased attention. In this paper a general overview will be presented of some of the recently developed processes which are being employed to engineer new copper alloys. Examples of rapid solidification, mechanical alloying, solid state microblending, and phase redistribution processing as applied to Cu-Cr, Cu-W, Cu-Ta, and CuPb-Sn alloys will be discussed. (4:00 p.m.) TUNGSTEN FIBER REINFORCED OFHC COPPER. R. L. Ammon and R. W. Buckman, Jr., Westinghouse Advanced Energy Systems, Pittsburgh, PA 15236
Room: ParkA Inn at the Park
Session Chairmen: I.E. Anderson, 122 Metals Development Bldg., Ames Laboratory, Iowa State University, Ames, IA 50011; R.E. Lewis, Lockheed Missiles & Space Co., Inc., R&D Div., Dept. 93-10, Bldg. 204, 3251 Hanover Street, Palo Alto, CA 94304
Anderson,
matrix microcomposites are strengthened by discontinuous, finely spaced second phase filaments or a fine dispersion of second phase particles. The degree of embrijtiement of the microcomposite by hydrogen was found not to be determined solely by fhe degree of embrijtiement of the reinforcing phase. Embrittlement was measured in terms of room temperature tensile properties following elevated· temperature, elevated·pressure hydrogen exposure as well as tensile properties of specimens tested in the elevated-pressure environment.
Ames,
IA
50011, and L. G. Fritzmeier, Rocketdyne Division, Rockwell International, Canoga Park, CA 91303.
Copper-refractory metal alloys, Cu-R, have been prepared by deformation processing of cast ingot material with R being Nb
or Ta or Cr. Measurements have been made of both the tensile properties and the electrical conductivity of these materials
as a function of the amount of deformation. These materials have an unusually good combination of strength plus electrical conductivity as will be demonstrated by comparison with the
best commercial alloys presently available. .The variation of these properties with composition and the R component will be presented and discussed. Interesting differences of these properties are found for axisymmetric versus plain strain conditions and possible reasons for this behavior will be presented.
*Operated for the U. S. Department of Energy by Iowa State University under contract No. W-740S-Eng-82. (2:30 p.m.) HYDROGEN EFFECTS IN COPPER MATRIX MICROCOMPOSITES: D.G. Ulmer, P.D. Krotz, J.L. Yuen and R.P. Jewett, Rocketdyne Division of Rockwell International, 6633 Canoga Avenue, Canoga Park, CA, 91303.
The hydrogen compatibility of a new class of copper matrix microcomposite materials, being developed for high temperature applications, was investigated. These copper
Tungsten fiber reinforced composite OFHC copper plates were produced. Metal arc spraying was used to form a monotape which
was subsequently consolidated by HIPing. Four- and ten-ply, ten and forty volume percent of continuous 0.2 mm diameter tungsten fiber reinforced OFHC copper composite plates were
produced. Good chemical bonding occurred between the copper matrix and tungsten fiber as determined by transverse tensile
testing. The as-HIPed copper matrix (1 hr @ 1200 K) exhibited a fine equiaxed grain size (ASTM 8-9). Little change was
measured in the tensile strength of the tungsten wire after
winding, spraying, and heating for one hour at 1200 K.
How-
ever, the room temperature ductility was affected, which has been attributed to the prior processing schedule used to produce the tungsten fiber.
(4:30 p.m.) EFFECT OF THERMAL EXPOSURE ON GRAPHITE FIBER REINFORCED COPPER COMPOSITES: B.A. Ash, S.W. Fowser, J.L. Yuen and N.E. Paton, Rocketdyne Division of Rockwell International, 6633 Canoga Avenue, Canoga Park, CA 91303
Graphite fiber reinforced copper is an attractive composite for high temperature applications. Addition of the continuous fibers reduces the density of the OFHC copper matrix and increases the strength and elevated temperature thermal conductivity. The fiber/matrix interface is an important parameter that influences the properties of the Gr/Cu composite. However, since copper and graphite do not form a metallurgical bond, copper migration has been observed under isothermal and thermal cycling conditions. With time at temperature this leads to microvoid formation around the graphite fibers. Extended exposure results in microvoid coalescence and eventual cavitation. Rationale for this phenomenon is discussed along with effects on mechanical and thermophysical properties.
34-Annual Meeting (4:20 p.m.)
HOME-ROTHERY MEMORIAL SYMPOSIUM IT: High-Tc Superconductors and Related Metallic Oxides (B)
CHANGES IN THE CRYSTAL CHEMISTRY AND MORPHOLOGY OF BULK HIGH-T c SUPERCONDUCTORS BY PROCESS CONTROL, P. Haldar, Johnson Matthey Electronics, Spokane, WA 99216, and B. C. Giessen, Department of Chemistry and Barnett Institute, Northeastern University, Boston, MA 02115.
Sponsored by the TMS Alloy Phases Committee
The superconducting behavior and crystal chemistry of the perovskite-like cuprates are greatly influenced by preparation and heat treatment conditions (e.g., temperature, atmosphere). By regulating these variables, it is possible to control the crystal structure and microstructure of these materials. Thus various members of the rare-earth, Bi, Tl and Pb containing compound families can be synthesized. Further, melt processing is shown to be an efficient method of obtaining long, plate-like, textured grains extending over large areas. Magnetic alignment with/without rare-earth substitution provides another effective way of controlling the grain orientation. The effects of a combination of approaches such as (i) magnetic alignment and annealing or (ii) magnetic alignment and melt processing on the microstructure of oxide superconductors will also be discussed.
Monday PM February 19, 1990
Room: Orangewood I Quality Hotel
Session Chairmen: D.E. Polk, Materials Division, Office of Naval Research, Arlington, VA 22217; E.W. Collings, Battelle Memorial Institute, Columbus, OH 43201 (2:00 p.m.) DEFECTS AND INSTABILITIES IN HIGH·T c SUPERCONDUCTORS, e......R... llii&hl;, Department of Chemis try, Oregon Sta te Univers i ty , Corvallis, OR 97331. Our highest Tc superconductors are based on Tl, Pb,Bi/Ba, Sr, Ca/Cu/O compositions, and here the real structures and compositions deviate considerably from the ideal situation. These deviations are crucial because they affect the carrier concentration and thus the actual Tc' However, these deviations can also cause inhomogeneities which can degrade some important properties, but which may actually improve critical current.
The compositional deviations are
associated with point defects, and these defects may be required for the stabilization of this class of superconductor, 1. e., they appear to be entropy stabilized phases.
(2:30 p.m.) CATION DISORDER AND OXYGEN VACANCIES IN Tl·Ca-Ba-Cu-O SUPERCONDUCTORS, B Morosin, E.L. Venturini, C.P. Tigges, J.E. Schirber, J.F. Kwak, D.S. Ginley, and R.J. Baughman, Sandia National Laboratories, Albuquerque, NM 87185. The significant range of transition temperatures (Tc) observed in TI-Ca-Ba-Cu-O superconductors with the identical structural phase and nominal composition results in part from Tl/Ca disorder and oxygen vacancies. Such site disorder and anion vacancies lead to variations in carrier concentration
which directly affects Tc' Additional factors may be the role of syntactic intergrowths leading to polycrystal formation and the presence of strain effects. A major synthetic challenge is
composition and
phase
control,
particularly
for
single
crystals and thin films. These issues and their relationship to superconducting properties will be discussed. This work was supported by the U.S. Department of Energy under
contract
#
(3:00 p.m.)
DE-AC04-76DP00789. BREAK
(3:20 p.m.) PROCESSING-STRUCTURE· PROPERTY RELATIONSHIPS IN SUPERCONDUCTING OXIDES, B.T. Ahn, V.Y. Lee, and R Beyers, IBM Research Division, Almaden Research Center, San Jose, CA 95120; T.M. Cur
and R.A.
Huggins,
Department
of Materials
Science
and
Engineering, Stanford University, Stanford, CA 94305. The structures and properties of superconducting oxides that are observed at and below room temperature depend critically on how the materials are processed at higher temperatures.
Consequently, there is a need to understand the hightemperature chemistry of these oxides. Here we describe the use of a solid-state ionic technique to determine the thermodynamic stability of YIBa2Cu307_x and the quaternary phase relations near the compound at elevated temperatures by oxygen coulometric titration. Moreover, we use the same apparatus to investigate oxygen ordering and phase separation in Yl Ba2Cu307 -x at lower temperatures by preparing samples with precisely controlled oxygen contents.
(3:50 p.m.) THE PROSPECTS FOR HIGH Tc WIRES AND TAPES: HOW CLOSE ARE WE? R. D. Blaugher, Intermagnetics General Corporation, Guilderland, NY 12084. This talk will present the status of the current worldwide research to develop a technologically useful high Tc wire or tape. The major technical approaches will be reviewed, which are divided into either vapor deposition methods or powder
precursor techniques.
An attempt will be made to present a
realistic assessment of the current understanding and the prospects for obtaining a high current carrying high-Te conductor.
HYDROMETALLURGY AND AQUEOUS PROCESSING IT: Physical Chemistry of Precipitation and Effiuent Treatment Processes Sponsored by the TMSIEPD Aqueous Processing Committee Monday PM February 19, 1990
Room: Salon 1 Marriott
Session Chairmen: Fiona M. Doyle, University of California at Berkeley, Dept. of Materials Science & Mineral Engineering, Hearst Mining Building, Berkeley, CA 94720; Lorin Redden, U.S. Bureau of Mines, Salt Lake City Research Center, 729 Arapeen Drive, Salt Lake City, UT 84108 (2:00 p.m.) JAROSITE PRECIPITATION IN ZINC HYDROMETALLURGY; F. Elgersma, O.S.L. Bruinsma, P.J. Jansens and G.M. van Rosmalen. Delft University of Technology - Leeghwaterstr. 44 - 2628 CA DELFT. A high purity electrolysis solution is essential for hydrometallurgical zinc winning processes. For this reason ferric iron containing compounds are usually precipitated as iron and impurity collector during the calcine leaching. One of the most common processes for this purpose is the jarosite process. The present study is aimed at precipitating a low-impurity jarosite separately from other zinc electrolyte contaminating heavy metals. If the low impurity jarosite and the other contaminants (Pb, Ag, As .. ) were collected separatelv~~th~_larosite can be regarded as an iron containing basic material, avoiding the environmental problems of storage ponds or dumping of over 80% of the zinc processing solid waste. Results will be shown on jarosite precipitation kinetics and uptake of impurities based on continuous precipitation experiments. During the experiments pH, temperature and residence time were automatically controlled All experimental conditions correspond to zinc hydrometallurgy plant practice.
(2:20 p.m.) THE STABILITY AND SOLUBILITY OF FERRIC ARSENATE; AN UPDATE; Robert G. Robins, The University of New South Wales, Kensington, NSW 2033, AUSTRALIA. Precipitation of ferric arsenate is an effective method for removing arsenic (v) from hydrometallurgical process solutions.
The solubility and stability of this compound, FeAs0 .2H 0, has 4 2
received some attention in recent Hears and the reported solubilities of amorphous and crystalline material have varied considerably. This paper reviews previous studies and reports some recent work with a consistent model to explain the variable results. Free energy data for ferric arsenate and its associated dissolved complexes are presented.
(2:40 p.m.) THE DETOXIFICATION OF EFFLUENTS CONTAINING ARSENIC WITH IRai SULPHATE AND HYDROGEN PEROXIDE: Luiz Alberto Teixeiro and Alexandre Guerra ItJnteiro. Catholic University of Rio de Janeiro. Dept. Materials Science & Metallurgy, CP.38008 - Rio de Janeiro -22453 - RJ Brazil An experimental investigation on the kinetics of oxidation of
Annual Meeting-35 As(III) 'with H2D2 in synthetic effluent solutions followed by precipitation of Fe/As(V) compounds was carried out. The variables investigated were: As concentration (200 - 3000 ppm), oxidation pH (2.0 - 11.0), oxidation temperature (25 - 1000 C), H2D2 stoichiometric excess (50% - 100%), precipitation pH (5.0 7.0), precipitation temperature (25 - 80 0 C) and precipitation agent (Fe 2 + - Fe 3+). Fixed conditions were Fe/As molar ratio (6:1) and stirring speed (560 rpm). The precipitates obtained were amorphous and believed to be FeAs04.xFe(OH)3 type of compounds. The combined Fe/H2D2 detoxification process, under selected conditions, was able to reduce the As contents of effluents containing 1600 ppm As to less than 0.1 ppm, in less than 4 hours of total processing time. (Work done in cooperation with Peroxidos do Brasil Ltda., a company member of INTEROX.)
(3:00 p.m.) METAL RECOVERY AND WASTE PROCESSING USING FERRITES: James D. Navratil, Department of Mineral Processing & Extractive Metallurgy, The University of New South Wales, P.O. Box 1, Kensington NSW 2033 Australia.
Ferrites have been used to separate a wide range of substances such as dissolved metal species, particulate matter, and organic and biological materials. Ferrites have been used almost exclusively for metal waste treatment applications. The versatility of the ferrite waste treatment technique is evident in the wide variety of application methods. Furthermore, ferri tes can be used to remove selected impurities in feed solutions containing metal products requiring purification. A brief overview of ferrite properties and recent applications for metal recovery and waste processing will be presented. (3:20 p.m.) Break
(3:40 p.m.) KINETICS OF NUCLEATION AND GROWTH DURING HYDROLYTIC STRIPPING: Yasuhiro Konishi, Dept. Chemical Engineering, University of Osaka Prefecture, Sakai, Osaka 591, Japan, and Fiona M. Doyle, University of California at Berkeley, Dept. Materials Science & Mineral Engineering, Hearst Mining Building, Berkeley, CA 94720. Hydrolytic stripping is the process in which metal ions are simultaneously stripped from a loaded cation-exchange solvent extractant and precipitated as oxides or hydroxides. Hydrolytic stripping is usually effected by contacting the loaded organic phase with water at elevated temperatures. The possibility of using hydrolytic stripping to prepare ceramic powders with controlled size and morphology is now being evaluated for different systems, using carboxylic acid extractants. As part of this work, mathematical models were developed to describe the kinetics of nucleation and growth of both hematite and nickel ferrite particles. This paper describes the development of the models, along with estimation of key parameters appearing in the rate equations from experimental data.
(4:00 p.m.) SPECIES ACTIVITY DISTRIBUTION DIAGRAMS, Zn-, S-, AND Zn-S-H ,0 SYSTEMS M.F. El-Demerdash and S.M. EI-Raghy Department
of
Metallurgy,
Faculty
of
Engineering,
Cairo
University, Cairo, Egypt.
The activity distribution diagrams for different species in aqueous systems are developed. Computational method may be considered as an extension of the Activity-Term method introduced by the authors to construct E -pH diagrams. In the Activity-Term method, all activities 01- different species are evaluated before deciding the most stable one. In the present study, a factor is assumed to allow for the conservation of the element. At any point, the total concentration of all species should contain only the input m moles of the element (elements) considered. This approach is applied to Zn-, S-, and Zn-S-H,O and activity distribution diagrams are plotted as a function of pH, at different potential levels and at different temperatures.
INTERFACES IN METAL-CERAMICS COMPOSITES I: Thermodynamics and Kinetics A Sponsored by Process Fundamentals Committee and Composite Committee Monday PM F~bruary 19, 1990
Room: Plaza D Inn at the Park
Session Chairmen: R.Y. Lin, Dept. of Mat. Sci. & Eng., University of Cincinnati, Cincinnati, OH 45221; M.R. Krishnadev, Dept. of Mining and Met., Pavillion Pouliot, Laval University, Quebec, G1K 7P4, Canada (2:00 p.m.) INTRODUCTION BY SYMPOSIUM COORDINATOR (2:05 p.m.) Key Note Speech: "Ramification of Recent Micro-Mechanical innovation on Understanding in Composite Interfaces S. G. Fishman, Office of Naval Research, 8001 N. Quincy St., Arlington, VA 22217. (Abstract Not Available) ll
,
(2:50 p.m.) INTERFACIAL REACTIONS IN INTERMETALLIC ALLOY MATRIX COMPOSITES, S. M. Jeng and J.-M. Yang, Department of Materials Science and Engineering, University of California, Los Angeles, CA 90024-1595
Fiber-reinforced intennetallic alloy matrix c.omposites are under intensive development for various high temperature structural applications. They possess excellent specific strength and stiffness, as well as good oxidation resistance. However, the nature and strength of the fiber/matrix interface playa dominant role in detennining the perfonnance and reliability of the composites. Hence, it is necessary to understand the fundamental nature of the complex interfacial phenomena between various potential reinforcements and the intennetallic alloy matrices. In this paper, the chemical interactions between various continuous fibers and intennetallic alloy (Ni3A1, Ti3Al) matrix will be discussed. The microstructure and chemical compositions of the reaction layers have been characterized by microscopy and electrOn microprobe. The interfacial reaction kinetics and its dependence on the temperature, environment and the structural characteristics of the materials will be presented. The effectiveness of using diffusion barrier coating to control the interfacial reactions will also be discussed. (3:10 p.m.) INFLUENCE OF COATING OF SiC PARTICUIATES ON THE MICROSTRUCTURE AND MECHANICAL PROPERTIES OF ALUMINIUM ALLOY MATRIX COMPOSITES, C.G Krishnades Nair, Hindustan Aeronautics Ltd. Bangalore, India, M.R. Krishnadev, Visiting Professor, University of Cincinnati, Dept. Materials Science and Engineering, Cincinnati, OH 452210012.
Particulate SiC reinforced aluminium alloy Composites are gaining importance in the aerospace industry, however, there is the need to improve
the bond between the particulates and the matrix to
further upgrade their properties. To achieve this, the authors have used copper-coated SiC particulates for reinforcing aluminium alloys. Results indicate that the coating improves the bond between the particulate and the matrix, conferring overall improvement in the properties. Composition of the interface has been analyzed via EDAX and Scanning Auger and the results have been used to explain the observed improvement.
(3:30 p.m.) BREAK (3: 40 p.m.)
MICROSTRUCTURAL CHARACTERIZATION OF THE INTERFACIAL REGION SiC REINFORCED Al-Mg ALLOYS: L.G. Legoux*, L. Salvo**, Ribes**, G. L' Esperance* and M Suery**, **Genie Physique Mecanique des Materiaux (URA CNRS 793) ,Inst. Nat. Poly tech. Grenoble,St.Martin d'Heres Cedex,France.
IN H. et de
Surface oxidation of SiC particles playa major role during the fabrication by casting techniques of Al-matrix composites and can influence their mechanical properties.The aim of this work is to characterize the interfacial reactions occurring between oxidized SiC particles and Al·Mg alloys. The composites were elaborated using the compocasting technique with 10 volt SiC particles. The parameters were the extent of oxidation of the particles and the thermal history of the melt, including normal fabrication procedure and prolonged treatment above the liquidus temperature. Characterization of the interfaces was performed using SEM and TEM with microdiffraction, EELS and EDX. The effec t of interfacial reactions on mechanical properties was also studied particularly on age-hardening behaviour of the matrix which was found to be greatly affected by such reactions.The work clearly demonstrates that controlled interfacial reactions are needed for optimum mechanical properties. *Dept. de Genie Metallurgique, Ecole Poly tech . de Montreal,Montreal.Canada.
3S-Annual Meeting (4: 00 p.m.) RELATIONSHIP BETWEEN MICROSTRUCTURE AND PHYSICAL PROPERTIES OF A1203 AND SiC REINFORCED ALUMINUM ALLOYS. G Selvaduray** D.Quinn* , R. Hickman* , D. Richard* , and P.Das* San Jose State University , School of Engineering, Dept . of Materials Engineering, One Washington Square, San Jose , CA 95192·0086. Microstructural features of an A1203 reinforced 6061·T6 and SiC reinforced 356 · T6 aluminum alloys were studied, and correlated to the phYSical properties of these alloys. Commercially available alloys were obtained, and tensile, impac t and wear samples were prepared according to ASTM standards.
After testing, the samples were sectioned, polished
and examined by optical and electron microscopy. Microstructural features examined include particle size distribution and particle·matrix interaction . *HC Corporation, San Jose, CA **San Jose State University, San Jose , CA
(4: 20 p.m.) THE INTERFACIAL REACTION KINETICS OF THE AL/SIC COMPOSITE,~ Dept . of Materials Science and Engineering, University of Cincinnati, ML #12 , Cincinnati, OR 45221 . USA
~The
The interfacial reaction between aluminum matrix and silicon carbide reinforcement in Al/SiC composites has been studied . Examinations wi th a scanning electron microscope (SEM) and an energy dispersive x·ray analyzer (EDAX) revealed that SiC dissolved in molten aluminum during the processing of composites. The extent of the dissolution reaction increases with the increasing time and temperature of the casting operation. Thermodynamic calculations suggest that the S1 content in molten
Al in the vicinity of SiC surface can be as high as 15.5 wti at 800 o C. Kinetically, however, it is concluded that the equilibrium condition is not reached at the AI/Sic interfaces . The diffusion of Si in molten Al is not the rate determining step in the dissolution of SiC in molten aluminum. The chemical dissolution of solid SiC in molten Al appears to control the AI/SiC interaction. The rate of the SiC dissolution has been derived as a function of the processing temperature based on the experimental results of this study . (4:40 p.m.) The Influanc. of th .. Interface ' On 'Mechanical Behavior of Tl-6Al-4V/SCS~6 · Compo.ites
P.K. Wright, G.A. Smith, M.D. SansDRler, and R. Nimmer., GE-Aircraft Engines, Cincinnati, OR The mechanical properties of SCS-6 reinforced Ti·6Al-4V ,compoaite. are profoundly affected by the nature of the interfaetal bondini between the fiber and matrix, The principal (mechanical) features of this inurface are the weak carbon-rich aurfece coeting of the fiber, and the 4egr.e of thermally induced reaidual atre •• from fabrication, The combination of theee cwo affecte leade to different behavior. under dtfferent str •• a atatee and temp_returea, Theae ,ffacts will b. illustrated by obaervations on the tren.varae, shaar, and longitudinal deformation and fracture behevior of these compo.itea. *GE-Corporate Reeearch and Development Center, Schenec tady, NY
1990 LEAD·ZINC AND RELATED METALS SYMPOSIUM IV: Physical Chemistry of Pb·Zn·Sn Sponsored by the TMS-EPD Lead-Zinc-Tin Committee Monday PM February 19, 1990
Room: Salon A & B Marriott
Session Chainnen: T.J. O'Keefe, Materials Research Center, University of Missouri-Rolla, Rolla, MO 65401; A. Morris, Materials Research Center, University of Missouri-Rolla, Rolla, MO 65401 (2 :00 p.m.) A PRACTICAL MODEL OF THE IMPERIAL SMELTING ZINC · LEAD BLAST FURNACE , Herbert H. Kellogg , Henry Krumb School of Mines, Columbia University, New York, NY 10027 . A model of the ISF is described which reproduces with accu· racy the "normal" operation of three industrial furnaces. The model , written in Fortran, is based on detailed heat and material balances, and the concept that the ascending gases reach equilibrium with zinc oxide and with the water·gas
"Solution los s" of carbon in the lower shift reaction. shaft , and partial reoxidation of zinc vapor in the upper shaft are also essential features. For prediction of fur· nace behavior under altered operating cond i tions (bl ast rate , temperature, moi sture content , oxygen enrichment , or injection of oil or coal through the tuyere s ) the model modifies th e heat lo ss , temperature of slag and bullion, amount of reo xidation in the upper shaft and the condenser efficiency to reflect the altered operating parameters. Model predictions for such operating conditions are in rea· sonab1 e agreement with plant experience . (2:30 p.m.)
INVESTIGATION AND MODELING OF THE NON-FERROUS BLAST FURNACES or METALEUROP, B. Madelin and G. Sanchez, Minemet Recherche, I, Avenue Albert Einstein, 78190 Trappes - France, and A. Rist, Ecole Centrale de Paris, Grande Voie des Vignes, 92290 Chatenay Malabry, France.
Over the last ten years, Mineaet Recherche and Metaleurop have used proving to investigate the operation of its Iaperial Saelting furnace and lead blast furnaces. This allows modeling to predict the influence of operating parameters on furnace perforaance . In the Iaperial saelting furnace, probing shows the existence of both an isothermal reserve zone at 1015°C and a chemical equilibrium between ZnO and gas. A practical model for heat and material balances can predict the influence of operating parameters on the furnace. Lead blast furnace results show that thermal and chemical equilibria are not reached. Classical models for staged heat and mass balances are not sufficient. Other means are used to characterize the influence of operating paraaeters on the thermal instabilities of these furnaces .
~6~U~~R MODELUNG OF MASS AND ENERGY BALANCES
FOR THE OUTOKUMPU LEAD FLASH SMELTING PROCESS, E.M. Weenink and E.R. de Puy, Dept. of Raw Materials Processing, Delft Univ. of Technology, P. O. Box 5028, Delft, The Netherlands, and Will em P.C. Duyveste~, BHP-UTAH International, Inc., 1190 Bordeaux DrIve, Sunny. vale;cA 089. One of the new direct smelting technologies for the production of lead from sulfide concentrates that has reached the commercial stage is the Outokumpu lead flash/electric furnace smelting process. Using data from pilot plant test work, energy and mass balances for a commercial plant were Initially calculated by hand. To optimize process conditions, a large number of such balances had to be generated and a computer program using Lotus 123 was developed which provided an excellent opporturuty for computer simulation. Operating parameters can be calculated by iteration using input values for feed compOSItion, feed rate, smelting temperature, and lead content of the slag. The output is presented by means of various views of the appropriate process steps, or in 68 different graphs. (3 : 30 p .m. )
BREAK
(3:50 p.. m.l
TOP SUBMERGED LANCING FOR RECOVERY OF ZINC FROM ISF SLAG, M. Waladan, G.K. Williams Laboratory, University of Melbourne, G.R. Firkin, Sulphide Corporation Pty. Limited, and 1. Bultitude-Paull, B.w. Lightfoot, and I.M. Floyd, Ausmelt Pty. Limited, Australia. The top submerged lancing Sirosmelt system has been developed through laboratory scale testwork and pilot plant trials for recovery of zinc from1SF slag. The effects of the process vanables of gas injection rate, method of coal feeding, and temperature on zinc fuming rates were studied in the laboratory. It was found that the rate of zinc elimination was dependent on all of these variables. The rate of zinc removal from slag was independent of zinc content of slag down to approximately 1.5% Zn, below which the rate was proportional to zinc content in slag. Pilot plant trials were carried out to evaluate a proposed process and results followed predictions from the laboratory scale tests. A comparison of the proposed process with conven· tional slag fuming operations is made and it is shown that the process has a number of advantages which have a significant impact on the economics of zinc slag fuming.
(4:20 p.m.)
THE REDUCTION OF A CASSITERITE CONCENTRATE WITH CHARCOAL, R.L. and A.E. Radwan, Delft University of Technology, Department of Raw Materials Processing, P. O. Box 5028, Delft, The Netherlands.
~
An investigation of the reducibility of a cassiterite concentrate with charcoal was aade in the temperature range of 11231373 K. The course of the reaction was aonitored by a coabination of sample weight loss aeasurements, product gas analysis, and chemical analysis of the reaction product. The cassiterite could be best reduced by a combination of high temperature, reasonably high charcoal proportions, high charge porosity, and a small percentage of lime. Analysis of kinetic data obtained from isotheraal experiments using theory from the oxidation of carbon by carbon dioxide gave an activation energy value of 225 kJ/aole for the consumption of carbon by reaction with cassiterite.
Annual Meeting-37 (4:50 p.m.) FUTURE CHANGES IN THE PHYSICO-CREIIISTRY OF ZINC-ELECTROllINNING, A. von Ropenack, Ruhr-Zink GmbH, lIittener StraBe 1, D 4354 Datteln, lIest Germany. The hydrometallurgical winning of zinc is at a crossroads. New processes promise better economics, better maintenance of resources, and siapler operation. Pressure leaching promises to eliainate roasting and thus one complete process step. The hematite process solves a difficult and always costly iron problea. RDA opens a possible road to a totally new generation of electrolysis. This paper describes a modern hydrometallurgical zinc plant including pressure leaching, iron control, purification and electrolysis via a hydrogen diffusion anode.
Silver contained in the concentrates is distributed between the Cu-Pb-Ag residue and the copper cake produced in the purification process. The Cu-Pb-Ag residue, which contains gold and silver, is sent to the Oyachi flotation plant of the Kosaka Smelter: about 82% of the gold and about 85% of the silver are recovered in the copper flotation concentrate. In this paper, the distribution of gold and silver in the Iijima Refinery is discussed and operational results of the Oyachi flotation plant are presented. (3:30 p.m.)
BREAK
(3:50 p.m.) RECOVERY OF SILVER FROM RESIDUES OF THE ZINC JAROSITE PROCESS: C.S. Ek, Universite de Liege, Rue A. Stevart, 2, Liege B-4000,
"BeTgiuin.
1990 LEAD·ZINC AND RELATED METALS SYMPOSIUM V: Precious Metals Recovery in the Zinc Industry B Sponsored by the TMS-EPD Lead-Zinc-Tin Committee
As a continuation of previous work, further tests have been carried out to recover silver from the residues obtained in a zinc jarosite process. Selective flotation has been performed with different reagents, and cleaning steps have been introduced to upgrade the silver concentrate. Leaching of crude residues, as well as flotation tailings, has been carried out with cyanide and thiourea. Finally, the different beneficiation methods are compared, and the industrial aspects of their application are reviewed.
Monday PM February 19. 1990
(4:20 p.m.) THE RECOVERY OF SILVER AND SCARCE ELEMENTS AT ZHUZHOU SMELTERS: Zuoren Huang , Department of Technological Development, Zhuzhou Smelters, Zhuzhou, Hunan, China.
Room: Salon C & D Marriott
Session Chainnen: R. Menge, Preussag-Weser-Zink, Johannastrasse 2, D-2890 Nordenham, Gennany; R.W. Stanley, Noranda Technology Centre, 240 Hymus Boulevard, Pointe Claire, Quebec, Canada H9R IG5 (2 :00 p.m.) A OIEMICAL AND MINE:RAI.mICAL S'IUDY OF SILVER, l.EI\D AND CAl:MIlM IN KIIXl mEEK ZOC
D.ltrizac, T.T. Chen, CANMET, 555 Booth Street, ottawa, K1A OGl
C&iB.da
The mineral fontlS of silver, lead and cadmium in the zinc concentrates an::! the various roaster products made by the Kidd Creek Division of Falconbridge Inc. have been elucidated. In the concentrates, all three elements occur mainly in sulphide fOIlll, either in solid solution in the principal ore minerals or as discrete mineral grains. During roasting, FIg, Pb and Cd are preferentially concentrated in the hot cyclone an::! ESP dusts of the roaster train. Silver is converted to <0.5 JlIlI grains of silver metal and to Flg2S04 which dissolves in or inpregnates silicates. Lead is converted to ~bearing silicates or to PbS04 and PbO which are present either as tiny free grains in znO or as an intergrowth with lead silicate. Lead is also a minor constituent of the ferrite phase. caanium occurs Irostly as cadmium-bearing znO an::! ZnFe;P4' rut a significant amount also reports as Cd-bearing silicates or zinc sulphate. (2:30 p.m.) RECOVERY OF LEAD, SILVER AND GOLD FROM ZINC PROCESS RESIDUES BY A SULFIDIZATION-FLOTATION METHOD: J. Rastas, Helsinki University of Technology: J. Leppinen and V. Hintikka, Technical Research Centre of Finland: S. Fugleberg, Outokumpu Research Centre. The recovery of lead, silver and gold from existing and potential zinc process residues using a sulfidization-flotation method was investigated both in the labortory and at the pilot plant scale. The sulfidization of lead and silver is carried out at low pH by the addition of a stoichiometric amount of sulfide at a controlled rate. To increase the particle size of the sulfides, seeding with the lead concentrate obtained is employed. The flotation is carried out at pH 2-4 using low pHcollectors such as dithiophosphates. Special attention must be paid to the role of oxygen in the conditioning stage. The effect of different parameters on both the sulfidization and the flotation has been evaluated . Recoveries of 85-90, 90-95 and 80% were obtained for lead, silver and gold, respectively. Typically, concentrates contalnlng 50-60% lead, 3-5 kg/t silver and 10-15 glt gold were produced.
The flowsheet of the hydrometallurgical zinc plant which is in use at Zhuzhou Smelters is described. The principal features of the process are fluid bed roasting, continuous calcine leaching, solution purification, zinc electrolysis, electrolyte cooling, and finally, the melting and casting of zinc forms. The Zhuzhou process for the flotation of silver from the leach residue collected on the Moor filters is described. The technical details of the process are outlined and the operating data are analyzed. The flotation facilities are described in detail and concentrate grades and recoveries are presented. The Waelz process is used to concentrate In, Ge and Ga remaining in the leach residue after silver flotation. These elements subsequently are recovered by selective leaching in sulphuric acid, and details of this operation are presented.
MATERIALS DEGRADATION IN LOW EARTH ORBITll Sponsored by the Jt. TMS/ASM Corrosion & Environmental Effects Committee Monday PM February 19. 1990
Room: ParkC Inn at the Park
Session Chainnen: Bruce A. Banks, NASA Lewis Research Center, Cleveland, OH 44135; Jon B. Cross, Los Alamos National Laboratory, Los Alamos, NM 87545 (2:00 p.m.)
ATOMIC OXYGEN EFFECTS ON SPACECRAFT MATERIALS: R.C. Tennyson, W.O. Morison, University of Toronto Institute for Aerospace Studies, Toronto Canada, M3H 5T6 A facility has been developed based on a SURFATRON, microwave powered device that is capable of producing an atomic oxygen beam energy of 3e>t, with flux levels as high as 10 17 atoms/cm 2 -s. A variety of materials has been studied (polymer based composites, thin film dielectrics, carbon) to determine their mass loss rates and reaction efficiencies. Protective coatings have also been investigated . The validity of accelerated testing will be demonstrated, together with surface morphology changes. The presence of U.V. radiation will also be discussed.
(3:00 p.m.) RECOVERY AND SUBSEQUENT TREATMENT OF GOLD AND SILVER IN ZINC PROCESSING AT THE IIJIHA REFINERY: A. Onozaki, H. Nagata and N. Ishimori, Iijima Refinery, Akita Zinc Co ., Akita 011, Japan.
(2: 25 p.m.)
The Iijima Refinery was put into operation in 1972 and today has an annual production capacity of 156,000 tons of SHG electrolytic zinc. At the Iijima Refinery, the recoveries of gold and silver are 98.0% and 98.2%, respectively. Gold contained originally in the zinc concentrates reports in the Cu-Pb-Ag residue produced in the residue treatment process.
We describe a device for production of low energy (5-10 eV) neutral atomic beams for surface modification studies, which recreates the flux of atomic oxygen in low Earth orbit. The beam is produced by acceleration of plasma ions onto a negatively biased plate of high-Z metal; the ions are neutralized and reflected by the surface, retaining a large fraction of their in-
ATOMIC OXYGEN BEAM EXPERIMENTS FOR EROSION SIMULATION J. Cuthbertson, W. Langer, and R. Motley, Princeton University, P.O. Box 451, Princeton, NJ 08543
3S-Annual Meeting cident kinetic energy, forming a beam of atoms.
The device is
based on a magnetically confined (3-4 kG) co-axial plasma source and the atom energy can be varied by adjusting the bias voltage. The source provides neutral flux ~ 5 x 10l6(cm2 sand fluence ~ l02l(cm2 in 5 hours. The composition and energy of the beam is diagnosed using a mass spectrometer. Samples of carbon film, carbon-based paint, Kapton, mylar, teflon, and photoresist exposed to atomic 0 beams show erosion quite similar to those
observed in orbit on the Space Shuttle. (2:55
p.m.)
ETCHING POLYMERS IN THE OXYGEN AFTERGLOW: EFFECT OF REACTOR PARAMETERS ON ETCH RATE. N.R. Lerner. NASA Ames Research Center. Mail Stop 230-3. Moffett Field. CA 94035. The etch rate of polymers in the afterglow of an RF discharge in oxygen is shown to be extremely sensitive to reactor design. At constant power input, etch rates are especially sensitive
to the design and placement of the electrodes and to the
diameter of the reactor.
Those reactor configurations which
laboratory-simulated micrometeroids. After being dormant for years, the SERT reactivated in July of 1989 for data retrieval. Values Al/89 sensors were recorded after close to 20 years exposure in space. This data which provides valuable baseline information concerning in the micrometeroid environment and the lifetime of a potential solar concentrator surface will be presented.
MATERIALS ISSUES IN ELECTRONIC & OPTOELECTRONIC CONNECTORS II: Electrical Issues Sponsored by the TMS Non-Ferrous Metals Committee and the TMS Electronic Packaging and Interconnection Materials Committee
lead to significantly higher etch rates promote larger
concentrations of ground-state atoic oxygen for a given input power, and also result in observable light emission from
excited CO and atomic oxygen in the vicinity of the sample.
When such configurations are used, significant heating of the
sample is also observed.
The dependence of the etch rate of
polymers on atomic oxygen concentration and temperature is also reported. Theoretical and practical implications of these
dependencies is discussed.
p.m.) EXPERIMENTAL INVESTIGATIONS OF LOW-ENERGY (4-40 eV) COLLISIONS OF 0-(2p) IONS AND 0(3 p ) ATOMS WITH SURFACES A. chutjian,' O.J. Orient' and E. Murad+ 'Jet Propulsion Lab., Caltech. Pasadena CA 91109 +AFGL, Hanscom Air Force Base, MA 01731-5000
(3:20
Using a newly-developed. magnetically confined source low-energy. ground state oxygen negative ions and neutral atoms are generated. The energy range is variable, and atom and neutrals have been generated at energies varying from 2 eV to 40 eV and higher. We find that the interaction of these low-energy species with a solid magnesium floride target leads to optical emissions in the (at least) visible and infrared regions of the spectrum. We will discuss briefly details of the phodetachment source. We will then present spectra of the neutral and ion "glows" in the wavelength range 250-850 nm (for 0-) and 600-850 nm (for 0), and discuss the variability of the emissions for incident energies between 4 and 40 eV, and demonstrate the effect of playing gases over the target surface during exposure. p.m.) ATOMIC OXYGEN EFFECTS ON REFRACTORY MATERIALS: NASA Lewis Research Center, Cleveland, OH 44135.
(3: 45
Atomic oxygen in low Earth orbit (LEO) may have undesirable effects on exposed refractory materials, such as are proposed for nuclear reactors in orbit, high temperature radiators, solar dynamic collectors, etc. Time resolved measurement of the volatile efflux from such materials at high temperatures is being done in an ultrahigh vacuum atomic oxygen ion beam facility at Case Western Reserve University, under grant from the Lewis Research Center. Results of measurements of the efflux of volatile oxides of molybdenum. niobium - 1 zirconium, and carbon-carbon composites at temperatures as high as 1400 K will be presented, along with a discussion of the roles of adsorption, desorption, and diffusion in atomic· oxygen reactions on surfaces at high temperatures. The dependence of reaction rates for certain materials on the energy of the incident atomic oxygen beam will be emphasized. (4: 10
p.m.)
PROTECTIVE COATINGS FOR LOW EARTH ORBIT APPLICATIONS AND LABORATORY SIMULATION OF LOW EARTH ORBIT: Bawa Singh. David Sarnoff Research Center. Princeton. NJ 08543-5300.
Monday PM
Session Chainnan: Robert McRoczkowski, AMP Inc., 2100 Paxton St., Harrisburg, PA 17105 (2:00 p.m.)
REQUIREMENTS AND TRENDS IN ELECTRICAL CONTACT APPLICATIONS. R. D. Ma1ucci. Mo1ex. Inc., Lisle. IL The variety of applications that exist for electrical contacts is diverse and changing. Present requirements and future trends in various industries are discussed with regard to designing modern day connector systems. The trends of increasing packaging density. speed and power are analyzed in terms of their impact on design and material requirements. The general classification of electrical contacts is discussed according to their function. Characteristics such as activity. power level. Signal type and speed are considered along with specific design criteria by considering fundamental contact theory and the electrical and mechanical properties of contact material. The impact of various industry trends are discussed in terms of present and future material requirements. (2:30 p.m.)
THE EFFECT OF ZINC DIFFUSION ON INTERMETALLIC GROWTH AND SOLDERABILITY OF HOT DIPPED CU-ZN ALLOY. F. J. Puckert. Wie1and-Werke AG. U1m. Federal Republic of Germany The retarding action of the intermeta11ic phase on zinc diffusion and solderability of hot dipped and thermally treated samples of Cu-in alloy was investigated. The solderability was examined visually using a DIN test and a quantitative wetting balance method. Diffusion profiles were determined on hot dipped and aged samples by SEM/EDX and GDL-ana1ysis. During extended room or elevated temperature aging. the zinc concentration in the tin layer remained below the detection limit or at impurity levels corresponding to the solder coating bath. The intermeta11ic phase acts as a zinc diffusion barrier to maintain excellent solderability for critical electrical connections. Growth of the intermeta11ic phase to the surface causes dewetting due to oxidation and loss of solderability. (3:00 p.m.) FRETTING CORROSION OF TIN PLATED ELECTRICAL CONTACTS,
Tin plated
tin
sion
011 44130.
A resurgence of interest in placing large solar dynamic concentrators 1n a space for power generation, has brought up again a concern for maintaining the integrity of the optical properties of highly specular reflecting surfaces in the near earth environment. One of the environmental hazards of concern and needing evaluation is the micrometeroid environment. On board the SERT II spacecraft, launched in Feb. 1970, is a micrometeroid sensor (2000 A 01 on Stainless Steel) designed and ground-calibrated by exposure to
to
We
(1)
process,
affect
TilE EFFECT OF TilE NEAR EARTH MICROMETEROID ENVIRONMENT ON A HIGHLY REFLECTIVE MIRROR SURFACE AFTER 20 YEARS IN SPACE: M.J. Mirtich,
General
copper alloys
because
resistance.
NASA Lewis Research Center, Cleveland, Oft 44135: W.R. Kerslake,
Lee,
Motors
Research
Lab-
are extensively used
as
connector contact material in automotive electrical circuits. Such tin plated contacts are susceptible to fretting corrosion which is a result of cyclic relat1ve motion between the contact surfaces. Fretting corrosion is unde-
of
Sverdrup Tech. Inc .• Middleburg IIts.,
Anthony
oratories. GM Tech Center. Warren. MI 48090 and Michael S. Mamrick. GM Packard Electric Division. P.O. Box 431. Warren. OH 44486.
sirable
ABSTRACT NOT AVAILABLE.
Room: California TIIIIV Quality Hotel
February 19, 1990
it.
it
have
can
(2)
and
lead
studied
understand
investigate
(3)
to higher contact
fretting
and quantify
corrosion
the corro-
parameters
evaluate various
which
techniques
to prolong contact life in the presence of fretting. We will report on the experimental technique used and the results of our study. The dependence of fretting corrosion on contact force, wipe length! contact feometry, tempera~~c~e ~f i:b~'l~aen~cJiil be ed\~~~Scs:J. oad, and the pres-
(3:30 p.m.)
CONNECTOR STABILITY TESTING W. D. Emmons. IBM. 11400 Burnet Rd .• Austin. Texas. 78758. This presentation describes a method for testing small system connectors from a users viewpoint. This method has been deve10p-
Annual Meeting-39 ed in order to achieve a consistent test sequence, test environment, test data analysis, and test control methodology across the numerous connectors being evaluated. The connector Test Methodology and desired characteristics are discussed in this presentation. Also included is the description of the gaseous environmental test that is used for determining a contacts resistance stability in a gaseous corrosive environment. In addition, the development of the user using stability ,rates versus failure rates is presented. Also, the purpose of this presentation is to give an overview of the testing and data analysis methodology along with the test controls required for each evaluation. The test method presented evaluates the initial Time Dependent and Recovery contact resistance stablilities and allows the user to screen out marginal or bad connectors. (4,00 p.m.)
A IIIGH-DENSITY, HIGII SPEED, BOARD-TO-BOARD STRIPLINE CONNECTOR, Matthew Sucheski, Douglas W. Glover, AMP Inc., 2100 Paxton st., Harrisburg, PI\. 17105.
(ABSTRACT NOT AVAILABLE)
MATERIALS SCIENCE OF MAGNETIC MATERIALS II Sponsored by the TMS Magnetic Materials Committee (EMPMD) Monday PM February 19, 1990
Room: California I-ll Quality Hotel
Session Chainnan: Raja K Mishra, Physics Department, General Motors Research Laboratories, Warren, MI 48090-9055 (2 :00 p.m.) ATOMIC FORCE MICROSCOPY: H. K. Wickramasinghe, IBM T. J. Watson Research Center, P. O. Box 218, Yorktown Heights, NY 10598
(Abstract not available)
(2:45 p.m.) MATERIALS SCIENCE AND MAGNETIC THIN FILMS: M. Hong, AT&T Bell Laboratories, Murray Hill, New Jersey 07974 Magnetic thin films in the form of homogeneous alloys, compounds, or heterostructures such as superlattices, multi layers and very thin overlayers, have played important roles in industrial applications as well as fundamental studies of magnetism. For the present applications in the area of magnetic or magnetooptical recording, homogeneous films are now being employed. However, films of a heterostructural nature carrying novel magnetic properties may soon find practical applications, besides interests in their fundamental properties. The ability to produce artificial structures with atomic-scale dimensions by the modern thin film techniques is one of the most exciting recent developments in materials science. This talk is to review the recent development of magnetic
thin films with new magnetic structures and to address the materials science challenge and excitement in these magnetic films.
(3:30 p.m.) MATERIALS FOR MAGNETO-OPTIC DATA STORAGE: Chien-Jung Lin, IBM Research Division, Almaden Research Center, 60 Harry Road, San Jose, CA 95120 Magneto-optic (MO) data storage technology is currently the most highly developed high-density reversible optical data storage technology. A typical MO data storage medium consists of a substrate, MO active layer, and dielectric layers, each involving quite different materials. Recording of information is accomplished by thermomagnetically writing magnetic domains on the thin MO film with a micron-sized laser beam. The recorded information is read back with the same laser beam at a lower intensity based on the magneto-optic Kerr or Faraday effects. This talk will review how the properties of materials used in the magneto-optic data storage media influence the performance and reliability of the MO media, with emphasis placed on key magneto-optic material requirements.
(4: 15 p.m.) MAGNETIC RECORDING: MA TERIALSSCIENCE CHALLENGES: Pantelis Alexopoulos, IBM Almaden Research Center, 650 Henry Road, San Jose, CA 95120-6099
(Abstract not available)
MODELING OF MICROSTRUCTURE EVOLUTION DURING SOLIDIFICATION II: Cellular and Dendritic Growth Sponsored by the TMS Solidification Committee Monday PM February 19, 1990
Room: Plaza A Inn at the Park
Session Chainnen: R. Trivedi, Ames Laboratory, U.S. Department of Energy and Department of Materials Science and Engineering, Iowa State Univerisity, Ames, IA 50011; M.E. Glicksman, Department of Materials Engineering, Rensselaer Polytechnic Institute, Troy, NY 12181-3590 (2:00 p.m.) SIDEBRANCH FORMATION: J S Langer. Institute of Theoretical Physics, 6808 Ellison Hall. University of California. Santa Barbara, CA 93106 New theoretical ideas which have emerged recently on the way in which patterns are formed in dendritic crystal growth and in the closely analogous phenomenon of viscous fingering in ftuids will be presented. Specific emphasis will be placed on the dynamics of dendrite sidebranch formation. A theory of sidebranch formation, based on the amplification of noise at the dendrite tip, will be developed for non·local models of solidification. Current thoughts on the selection mechanism of primary spacing will also be presented.
(2: 30 p.m.) OSCILLATORY INSTABILITY OF DEEP CELLS IN DIRECTIONAL SOLIDIFICATION: ~,Laboratoire de Recherche en Combustion, Universite de Provence . SI. Jerome, 13397 Marseille Cedex 13, France. Dendritic structures are commonly observed to ongInate from finger-like cellular structures in directional solidification of alloys. the vicinity of cell-dendrite tranSltlon, experimental In observations show coherent sidebranching of deep cells. This phenomenon of coherent sidebranching will be discussed and we shall propose that it results from an oscillatory instability of deep cells in which cell widths and tip positions oscillate coherently in phase. A linear stability analysis of a periodic array of deep cells will be presented in the small Peelet number limit and the stability limits of oscillatory mode will be derived.
(3:00 p.m.) THE EFFECTS OF INTERFACE ANISOTROPY PROPERTIES ON THE DEVELOPMENT OF CELLULAR MICROSTRUCTURES: R. Trived. Ames Laboratory, U.S. Department of Energy and Department of Materials Science and Engineering, Iowa State University, Ames, IA 5001 1, USA. The anisotropies in interface energy and interface kinetics play an important role in the development of solidification _microstructures. The effects of these anisotropic
properties on the development of interface patterns during the directional solidification of alloys will be examined. Specifically, the linear and the weakly nonlinear analyses of planar interface instability, and the subsequent development of cellular structures, for anisotropic materials will be discussed. These theorectical models will then be compared with the results of directional solidification experiments in which the anisotropic effects lead to the formation of cells and dendrites which are tilted with respect to the heat flow direction. It is shown that dendrites form in the preferred crystallographic direction, whereas the angle of tilt for cells is governed by the relative effects of heat flow and the anisotropic property of the crystal. From the tilting of cells, the travelling wave part of the purturbed interface is characterized and compared with the theory to obtain the magnitude of anisotropic effects. It is also shown that the cellular spacings, as well as the amplitude of cells, alter significantly with the angle of tilt under identical conditions of growth rate, temperature gradient and composition. (3:30 p.m.)
BREAK
(3:40 p.m.) CELL STRUCTURE NEAR ABSOLUTE STABILITY, S. H. Davis, Department of Engineering Sciences and Applied Mathematics, Northwestern University, Evanston, IL 60208 A dilute binary material is directionally solidified. As the pulling rate is increased, the planar interface becomes cellular, dendritic, cellular and then regains planarity at the absolute
stability limit. Nonlinear analytic methods are used to describe the high-pulling-rate cellular structures in two- or three-
dimensions and the roles of the parameter involved. Emphasis is placed on the mechanisms of transition from one pattern to another. Anisotropic-interface-kinetic effects are discussed as well.
40-Annual Meeting (4:20 p.m.) THREE-DIMENSIONAL COMPUTATIONS OF NONPLANAR INTERFACE MORPHOLOGIES DURING DIRECTIONAL SOLIDIFICATION OF A BINARY ALLOY. G.B. McFadden, S.R. Coriell, and R.F. Boisvert, NIST, Gaithersburg, MD 20899 Three-dimensional steady-state solutions for nonplanar interface morphologies are computed numerically by using finite differences. A linear temperature field is assumed, corresponding to the case of equal thermal properties in the crystal and melt, with negligible latent heat release. The solute field in the melt and the unknown crystal-melt interface position are obtained self-consistently. For a model of an aluminum-chromium alloy with distribution coefficient greater than one, steady-state solutions corresponding to two-dimensional bands and three-dimensional hexagonal nodes are obtained, as well as solutions with rectangular interface planforms. The stability of the computed steady-state solutions is suggested by the behavior of the iterative scheme used in the calculation. Near the onset of instability, the calculations predict hexagonal nodes, which is consistent with weakly nonlinear theory. (4:40 p.m.) The Effects of Convection on Ostwald Ripening: N. Akaiwa and P W Voorhees, Northwestern University, Evanston, IL 60208. The coarsening kinetics of stationary spherical solid particles in a quiescent liquid are characterized the well-known temporal power laws of the Lifshitz-Slyozov theory of Ostwald ripening. However, in many solidification processes the particles are not stationary in the liquid but can move, for example, as a result of a density difference between the solid and fluid. We have examined the effects of particle motion on the temporal power laws describing the kinetics of the average particle radius, number of particles per unit volume and the average liquid concentration during Ostwald ripening in the limits of both large and small Peclet numbers. We find that only under certain conditions can scaled time-invariant solutions to the equations describing Ostwald ripening be found and that convection can alter the exponents of the temporal power laws from their classical values. Research supported by the Microgravity Sciences Division of NASA
NOVEL PROCESSING METHODS AND PRO· PERTIES OF DIAMONDS AND DIAMOND· LIKE THIN FILMS IT: Characterization and Properties Sponsored by the TMS-SMD and Electronic and Photonic Device Materials and Electronic Materials Committees and ASM and TMS-EMPMD Monday PM February 19, 1990
Room: Orangewood m Quality Hotel
Session Chairmen: J. Prater, Mats. Engr., Army Research Office, P.O. Box 12211, Research Triangle Park, NC 27709; M. Imam, Naval Research Laboratory, Phys. Met. Div., Code 6320, Washington, DC 20375-5000 (2:00 p.m.) (Invited) CHARACI'ERIZAnON OF GROwm PROCESSES OF DIAMOND TIIIN FILMS BY RAMAN SPECTROSCOPY, R J Nemanicb, Department of Physics and Department of Materials Science and Engineering, North Carolina State University, Raleigh, NC 27695-8202 Raman scattering measwements are used to characterize the atomic structwes of thin films produced from CVD deposition of methane diluted with hydrogen and oth,er growth methods. The films are, in general, composite structures which 'include amorphous and crystalline diamond and graphitic regions. Spectral features are identified and attributed to diamond, sp2 (graphitic-like) and disordered sp3 bonded regions. We describe how the domain sizes of the regions are manifested in the Raman spectra. Defect and doping properties are evidenced by changes in the Raman cross-section and the phot~-luminescence. The growth process is modelled in terms of the initial nucleation and the interactions with the substrate surface. ' • Supported in part by ONR under contract NOOOI4-86-K-0666, and the Research Triangle Institute.
(2:30 p.m.) (Invited) X-RAY AND ELECTRON OPTICAL CHARACTERIZATION OF CVD AND COMBUSTION DEPOSITED DIAMOND. by C. S. Pande, C. L. Void, R. G. Vardiman. H. A. Hoff and B. B. Ratb. Naval Researcb Laboratory. Washington. DC 20375-5000 Diamond deposits grown in acetylene flame and also CVD grown deposits have been examined by transmission electron microscopy. Electron channelling (TEM), electron channelling and by x-rays. has been used to measure quantitatively the presence of strain. and x-ray to obtain texture patterns. TEM and electron diffraction showed the presence of intersecting twins inQicating the possible SEM and TEM have also been used to presence of lonsdellite. examine the fracture behavior of diamond deposits. (Invited) (3:00 p.m.) STRUCTURE-GROWTH RELATIONSHIPS IN FILAMENT ASSISTED CVD DIAMOND FILMS:* R. E. Clausing, Metals and Ceramics Division, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831-6093. Scanning and transmission electron microscopy (SEM and TEM), Raman spectroscopy, and x-ray diffraction have been used to characterize the topographic and internal (crystal-defect) structure of diamond films grown by the hot filament assisted chemical vapor deposition method. Films were grown under conditions chosen to provide several distinctly different diamond microstructures. This paper relates internal and external growth features to each other, to properties including hardness, elastic modulus, friction and wear, and infrared transmission and to the growth conditions. The growth features and internal structure are briefly discussed in terms of growth mechanisms. *This research was jointly sponsored by the U.S. Army Strategic Defense Command under MIPR No. W31RPD-8-P4030 with the U.S. Department of Energy under contract DE-AC05-840R21400 with the Martin Marietta Energy Systems, Inc. (3:30 p.m.)
BREAK
(4:00 p.m.) (Invited) VAPOR PHASE DEPOSITION AND CHARACTERIZATION OF DIAMOND FILMS ON VARIOUS SUBSTRATES: B.E. Williams, Y.H. Lee, D.A. Asbury. H.S. Kong, P. Richard, B. Stoner and J.T. Glass, Dept. of Mats. Sci. & Engr., North Carolina State Univ., Raleigh, NC 27695.
Microwave plasma and hot filament chemical vapor deposition have been utilized to deposit diamond on various substrates Including 51. NI. Ta. Mo and W. Since the final goal of the research Is heteroepItaxy. substrates were chosen from atomic spacing and/or surface energy conslderaUons which promote 2-D growth. The quality of the diamond films has been characterized via Raman spectroscopy and SEM. Films grown on 51 and NI exhibited the best quality according to sp2/ sp3 peak raUos and the full width half maximum of the 1332 em-I Raman peak. Shifts In this peak were
correlated with coefficient of expansion mismatches.
Furthermore. X-ray
diffraction. Auger electron spectroscopy and X-ray photoelectron spectroscopy have been utilized to examine the Interface layer and the first stages of nucleaUon of the diamond on some of the substrates. For example. on 51 (Ill). AES and XPS of the carbon core and valence levels reveal rapid formaUon of a SIC layer prior to ti,e appearance of any observable diamond crystallites. The concentration of carbon due to diamond and SIC as determined by XPS parallels the observed diamond coverage on the surface by SEM. (Invited)
(4:30 p.m.) MECHANICAL PROPERTIES CHARACTERIZATION OF DIAMOND-LIKE AND DIAMOND FILMS:* M. E. O'Hern and C. J. McHargue, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831-6117. The mechanical properties of diamond-like carbon and diamond films have been studied using indentation techniques. Values for hardness and elastic modulus are given for both types of films relative to natural diamond and single crystal sapphire. Determination of hardness and modulus of elasticity in thin films (particularly those with high values such as diamond) requires more information than can be obtained from a conventional (e.g., Knoop or Vicker's) microhardness tester. The use of an ultralow load indentation instrument (a mechanical properties microprobe) to characterize these properties is addressed. The problems associated with characterization of these films by indentation are briefly reviewed and discussed. *Research sponsored by the U.S. Army Strategic Defense Command under MIPR #W31RPD-8-C4114 with the U.S. Department of Energy under contract DE-AC05-840R21400 with the Martin Marietta Energy Systems, Inc.
Annual Meeting-41
PROCESS FUNDAMENTALS AND MATERIALS PROCESSING I: Chemical Vapor Deposition Sponsored by the TMS Process Fundamentals and Synthesis and Analysis in Materials Processing Committees Monday PM February 19, 1990
concentration fields in the reactor, as well as the rate of The agreement between the theoretical deposition of Si 3N4 . predictions and experimental measurements is reasonably good. The relevance of this work to the design of CVD reactors will be discussed.
(4: 15 p.m.) THE MODELING OF MOCVD SYSTEMS - A COMPARISON OF PREDICTIONS WITH MEASUREMENTS: A. H. Dilawari, J. Szekely and R. Daly, Dept. of Materials Science and Engineering, MIT, Cambridge,
MA
02139.
A mathematical representation is developed to describe the
fluid flow, heat flow and mass transfer in a typical barrel
Room: Salon H Marriott
Session Chairmen: T. DebRoy, Department of Materials Science and Engineering, The Pennsylvania State University, University Park, PA 16802; N. El-Kaddah, Department of Metallurgical and Materials Engineering, The University of Alabama, Tuscaloosa, AL 35487-6974
type reactor, used for the preparation of gallium arsenide wafers. In the formulation allowance is made for axial, radial and azimuthal velocity components and for simultaneous heat and mass transfer. The theoretical predictions are found to be in excellent agreement with measurements.
REACTIVE METALS I: Ti Production and Use of Emulsions and Precursors Reactive Metals Committee
(2:00 p.m.) ROLE OF MODELING IN MATERIAL PROCESSING BY CHEMICAL VAPOR DEPOSITION (CVD): M. K. Alam, Associate Professor of Mechanical Engineering, Dept. of Mechanical Engineering, Stocker Center, Ohio University, Athens, Ohio 45701.
Monday PM February 19, 1990
Chemical Vapor Deposition (CVD) processes are currently attracting special attention in materials processing. High temperature materials, such as compounds of tungsten, boron, tantalum, and a variety of ceramics, can be produced at temperatures much lower than the melting point of the solid. This process produces material of high purity, in complex shapes, and in a variety of microstructures.
Session Chairmen: Renato G. Bautista, Mackay School of Mines, University of Nevada-Reno, Reno, NV 895570047; KNona C. Liddell, Department of Chemical Engineering, Washington State University, Pullman, WA 99164
CVD is a capital intensive process, involving use of high cost raw materials. Modeling simulations have the potential of reducing the time for CVD process design, and the need for experimentation. Modeling would, therefore, appear to be an attractive option for producing a product of acceptable quality. Progress has been made in modeling of CVD processes, specially for semiconductor processing. (2:30 p.m.) ROLE OF MACROSCOPIC TRANSPORT IN THE CHEMICAL VAPOR DEPOSITION OF DIAMOND: K. Tankala and T. DebRoy, Department of Materials Science and Engineering, The Pennsylvania State University, University Park, PA 16802 (Abstract not available) (3:00 p.m.) COMPUTER MODELLING OF CHEMICAL VAPOR DEPOSITION PROCESSES: Dewei Zhu and Yogesh Sahai, Material s Science and Engineering, The Ohio State University, 116 W. 19th Ave., Columbus, OH 43210. general purpose computer model for descri bi ng the transport phenomena and resulting rate of deposition has been described. Partial differential equations describing the conservation of mass, momentum, energy and chemical species are solved by a computer program employing finite difference method. The system cons i de red is depos it i on of s i 1 icon ina vert i ca 1 stagnat ion flow reactor by the reaction of silicon tetrachloride and hydrogen. The program allows for multiple chemical species and natural convection effects. Predicted silicon deposition rates along the substrate are in reasonable agreement with experimental values available in literature. The effect of gas inlet configuration on the uniformity of deposition has been studied. The model can be used as a tool for design optimization of such reactors. A
(3:30 p.m.)
BREAK
(3:45 p.m.) A COMBINED THEORETICAL AND EXPERIMENTAL INVESTIGATION OF THE CVD OF SILICON NITRIDE ON CEMENTED CARBIDE TOOLS: N. EL· KADDAH*, J.W. MCKELLIGET**, C.CHEN*, AND J. LYNCH**, * Dept. of Mat. and Met. Eng, The University of Alabama, Tuscaloosa Al 35487, ** Dept. of Mech. Eng., University of Lowell, MA 01854. Experimental results are presented for the depos i tion rate of Si3N4 on tungsten carbide cutting tools in a conventional horizontal flow CVD reactor. The precursor gases used were silane and ammonia. The effect of substrate position and gas temperature and pressure on the reaction kinetics are investigated. Based on a numerical solution of
the
Navier·Stokes
equation,
the
thermal
energy
balance
equation, and the mass transport equation a mathematical model was developed to predict the velocity, temperature, and
Room: Salon G Marriott
(2:00 p.m.) MATHEMATICAL MODELING OF THE SELECTIVE CHLORINATION OF IRON FROM ILMENITE ORE BY CO-Cl 2 MIXTURE IN A FLUIDIZED BED': K. I. Rhee and H. Y. Sohn, Department of Metallurgical Engineering, University of Utah, Salt Lake City, Utah 84112-1183. By modifying a two-phase fluidized-bed model, two mathematical models which describe the behavior of gas and solid movement and reactions in a fluidized bed as well as interpret the experimental results for the selective chlorination of iron were formUlated. One model is based on treating the bubble and the emulsion phase as separate continuum phases with mass exchange between them, resulting in differential governing equations. The other is based on the compartmentalization of the fluidized bed into a network of perfectly mixed reactors. Predictions from these models produced satisfactory agreement compared with experimental results. The effects of important variables affecting the bed performance -- superficial gas velocity, exchange rate between the phases, and reaction rate constant -were tested. 'Work supported by DOl Mineral Institutes program under Grant No. G1144149 and by NSF under Grant No. INT82-11631.
(2:30 p.m.) RECOVERY OF SYNTHETIC RUTILE AND IRON OXIDE FROM ILMENITE ORE BY SULFATION WITH AMMONIUM SULFATE: Chul Tae Lee, Department of Chemical Engineering, Dankook UniverSity, Seoul, Korea; and H. Y. Sohn, Department of Metallurgical Engineering, University OfUtah, Salt Lake City, Utah 84112-1183. The reaction between ilmenite and ammonium sulfate was investigated to test the feasibility of producing synthetiC rutile and high-purity iron oxide using ammonium sulfate as a sulfating agent. The optimum sulfation conditions were determined to be a temperature of 400·C, 3 hours reaction time, an ammonium sulfate-to-ilmenite weight ratio of 11, and an ilmenite particle size of -250 mesh. Water leaching of the reaction product under these conditions resulted in the recovery of 94.5% of Ti0 2 and 99.8% of iron. From the leachate, Ti0 2 of 94.2% purity as a mixture of rutile and anatase and ~-Fe203 of 98.3% purity were obtained. (3 :00 p.m.) UPDATE OF INDUSTRIAL PLANT PRODUCINS ELECTROLYTIC TITANIUM M.V. Ginatta, R. Berruti, G. Orsello; G.T.T. S.A.-Via Brofferio,3 10121 Torino ITALY
Electrowon titanium has reached industrial commercialization.The development stages which led to the construction of the last type of electrowinning plant are reviewed. The electrochemical concepts and the engineering criteria which have been chosen for plant construction and operations are outlined. The equipment of
42-Annual Meeting the electrolytic plant and the molten salt cells are operated with the ease of an aqueous tankhouse. The cost effectiveness of plant design and low cost materials used for its construction are illustrated. Continuous steady state production of titanium crystals from the extraction compartments is obtained by supplying tetrachloride to the dissolution compartments. The cells electrodes are suplied with direct current from standard rectif~ ers. The titanium metal produced is of excellent quality having a typical BHN 70 - 80.
'(3:30 p.m.)
BREAK
(3:45 p.m.) Zirconium Powder from Ammonium Zirconyl Oxalate Precursor Ching-Lun Huang
Luh-HuR R("In9
of Nuclear P.o. Box 3-7 ~nstitute
~nergy
Kesearch
Lung-Tan, Taiwan 32500 Republic of China Abstract Zirconium powder is an important commodity for the thermal battery, delayed explosive mixture, and flash bulb. Hydride-dehydride, magnesium and calcium reduction methods were tried. A novel method to prepare zirconium oxide from ammonium zirconyl
oxalate precursor was developed.
This precursor was a soluble
zirconium compound fabricated from zirconyl chloride, oxalic acid, and ammonium oxalate.
Following crystallization,
drying~
attriting, calcining and deagglomeration, ultra fine zirconium oxide with tetragonal crystal structure was obtained. Calcium reduction of zirconium oxide in a furnace with carefully-controlled atmosphere was a necessary step to manufacture thermal battery grade zirconium powder. Characterization of raw material , precursor, zirconium oxide, and metal powder were also reported
(4: 15 p.m.) PREPARATION OF LANTHANIDE OXALATE POWDERS USING CARBOXYLATE-BASED EMULSIONS: Jung H. Yoon, IBM Corporation, Advanced Semiconductor Technology Center, Route 52, Hopewell Junction, NY 12533, and Fiona M. Doyle, University of California, Dept. Materials Science & Mineral Engineering, Berkeley, CA 94720, USA. Many of the technological uses of lanthanide powders require closely controlled particle sizes and morphologies. Lanthanide oxalate powders were prepared by contacting loaded carboxylic acid solvent extractants directly with aqueous oxalate solutions, using ultrasonic stirring and surfactants to create emulsions. The position of equilibrium is such that lanthanide ions were effectively stripped from the organic phase, and precipitated from the aqueous phase. The effect of lanthanide type, stirring energy input, phase ratio, surfactant concentration, lanthanide concentration and lanthanide/oxalate ratio on the precipitate characteristics was studied. The results are discussed in terms of the effects of the operating conditions on the emulsion type (oil-in-water or water-in-oil), and on the kinetics of the stripping, nucleation and growth processes.
REDUCTION TECHNOLOGY IT: Cell Efficiency Sponsored by TMS Aluminum Committee Monday PM February 19, 1990
Room: Salon J & K Marriott
Session Chairman: Alton Tabereaux, Reynolds Metals Company, P.O. Box 1200, Sheffield, AL 35660 (2:00 p.m.) THE DYNAMICS AND PERFORMANCE OF REDUCTION CELL ELECTRO!..YTES: M P Taylor Xiaoling Uu, K.J. Fraser, Comalco Research Centre, 15 Edgars Road, Thomastown. Vic. 3074 The performance of smelting electrolytes has traditionally been difficult to correlate with composition. Average cell current efficiency varies widely with the degree of control and thermal stability. Similarly, the operating temperatures of different cell types are not easily reconciled against their quoted average electrolyte compositions, while laboratory measurements of liquids temperature show an even poorer correlation between the different investagators. The above problems clearly indicate that it is not possible to consider the phase equilibria and chemical activity of the electrolyte in isolation from the heat transport rates within it. this paper addresses the interaction between thermal state and phase equilibria in a smelting electrolyte and relates the results to obselVed operating parameters.
(2:30 p.m.) A PROCESS CONTROL SYSTEM FOR REDUCTION CELLS BASED ON DYNAMIC HEAT AND MATERIALS BALANCE: M P Taylor and G.I. Blatch, Comalco Research Centre, 15 Edgars Road, Thomastown. Vic. 3074 An aluminium reduction cell is ideally operated at steady-state. Thermal steady state can be defined as operation with constant heat losses from the cell and constant energy consumption within the cell. In reality, operations such as anode setting and feeding cause thermal disturbances to a cell. A control system has been developed at the Comalco Research Centre which integrates and systematizes responses to thermal disturbances. In essence, the energy requirements of a cell for metal production and individual operations (including anode setting and feeding) are calculated. The target resistance of the cell is then changed to ensure that energy available for dissipation from the cell (as heat) is maintained at a constant, target rate. Examples of the control system performance are included in the paper. (3:00 p.m.) CURRENT EFFICIENCY IN MODERN POINT FEEDING INDUSTRIAL POTLINES: B . Langon and J. Peyneau, Laboratoire de Recherche des Fabrications, Rue Sainte Claire Deville, B.P. 114, 73300 Saint-Jean-De-Maurienne, France. The mechanisms of current efficiency in modern point feeding pots are analysed both from the theoretical and practical points of view. Various factors influencing the current efficiency, such as bath circulation under magneto-hydrodynamic forces, metal diffusion, thermal balance and bath chemistry, etc ... are reviewed. The industrial results presented cover 9 potlines started since 1979 for a total production of 850 000 tpy. The average current efficiency varies between 94 and 95%. (3:30 p.m.) CONTROLLED POTENTIAL COUWMETRY AS A TOOL FOR TIlE DETERMINATION OF CURREN[ EFFICIENCY IN COMMERCIAL HALL CELLS: G P Tarcy and D. R. DeCapite, Aluminum Company of America, Alcoa Technical Center, Alcoa Center, PA 15069 Shan term current efficiency measurements in the Hall cell are performed by either anode gas analysis or by measuring metal production with a correction for cell inventory fluctuations. The cell inventory corrections are done by a metal dilution procedure in which a known quantity of another metal (i.e., Ag) is dissolved into the Al cathode and then chemically analyzed for its concentration in the metal. Metal dilution analysis is preferred over anode gas analysis when accurate and precise current efficiency results are needed but the technique is dependent upon the size of the sample analyzed and the precision of the analysis method. None of the past analytical techniques is satisfactory because of either inadequate precision or legal considerations resulting from the use of radioactive materials. A method which uses a silver addition and controlled potential coulometry to perform the chemical analysis has been developed to circumvent shortcomings in the other methods. The best precision to date on Hall cell samples is ±o.5% at a 95% confidence interval in the 50-100 ppm Ag range. This represents an ability to determine current efficiencies on a weekly basis in commercial Hall cells and pilot Hall cells to ±o.5% and ±o.25% respectively.
(4:00 p.m.) APPUCATION OF THE RADIOISOTOPE DILUTION TECHNIQUE IN EXAMINING FACTORS AFFECTING CURRENT EFFICIENCY IN AlUMINIUM REDUCTION CELLS: K J Fraser. W. Uru and W. Wiblin, Comalco Research Centre, 15 Edgars Road, Thomastown. Vic. 3074 The technique of radioisotope dilution has been used to determine inventories of molten metals and other liquids such as slags in a number of metallurgical industries. This technique applied to the measurement of molten metal inventory in aluminium reduction cells allows accurate determination of current efficiency in such cells over single tapping cycles. This allows performance comparisons to be made between cells under different operating conditions. This is particularly useful in the evaluation of plant trials over a relatively short period of time as well as being an important tool in the study of pot dynamics and the effects of process control variations on current efficiency. This paper reports the application of the radioisotope dilulion technique for the determination of cell current efficiencies in an aluminium smelter on a routine basis.
(4:30 p.m.) INDUSTRIAL CONSEQUENCES OF ANODIC PHENOMENA IN HALL HEROULT CELLS, Nolan E. Richards, Manufacturing Technotogy Laboratory, Reynolds Metals Company, P. O. Box 1200, Sheffield, Alabama 35660-1200, USA
The electrolytic reduction of alumina dissolved in a fluoride molten salt, probably the most extensively practiced industrial high temperature electrochemical process, is characterized by some phenomena at the anode difficult to isolate yet intellectually stimulating. There is a surprisingly high anodic overpotential, the anode is consumable and the byproduct, which modulates the real current density. also, by serendipity, promotes mass transfer in a process that would otherwise be uneconomical. In this review, laboratory studies of the etectrode kinetics, concepts and data describing the ionic species, double layer formation and expectations for ion transfer are correlated with data obtained on these and selected parameters in actual industrial cells. Interest focuses on how the potential impediment to successful replication at the intensity and scale required of an economical cell, actually work to make the process all of operational, manageable,
Annual Meeting-43 predictable. The problems of high overpotential and gas bubbles can even be turned into advantages through artificial intelligence.
(5:00 p.m.) THE CATALYTIC ACTION OF SOME ADDITIVES ON CARBON ANODE REACTION OF ALUMINIUM ELECTROLYSIS CELL: Y.G. Chun, Q. Zhi xian, Zang jinseng and L. Jing jiang, Northeast University of Technology, Shenyang, China During aluminium electroiysis, ionic oxygen is discharged on the carbon anode surface and reacts with the carbon to yieid CO2. The overall electrolytic process may be represented by: AI~3(diss) + 312C(s) • 3I2C02(g) +2AI(1)' The reaction of oxygen with carbon is slower, that causes considerably higher anodic over·vo~age (400-600mV) and will lead to waste much electric energy. We carried out experiments in laboratory in order to study the catalytic action of some additives with the purpose of decreasing the anodic over-voltage. Some additives (K2C03, Na2C03,Li2C03 and RE2(C03)i3) in some amount were added to the carbon peste samples, respectively, then the carbon samples were baked up to 9500C. Electrolysis experiments were carried out in a small electrolysis cell with a graphite crucible as cathode, and a carbon sample prepared as described above as anode, a graphite electrode in spectral purity as reference, recording I-V plot by YEW model 3067 MEMORY RECORDER. It was found that all the additives mentioned above had catalytic action on the carbon anode reaction, decreasing the anodic over-voltage more or less.
SUPERPLASTICITY IN AEROSPACE II, 2nd International Symposium Sponsored by the TMS Shaping and Fonning Committee Monday PM February 19, 1990
Room: Garden Rooms 1-2-3 Inn at the Park
Session Chainnan: A.K. Mukherjee, Division of Materials Science, Department of Mechanical Engineering, University of California, Davis, CA 95616
experiments involving SiCp , one or more jets of strengthening phases were injected into the atomized spray at a prescribed spatial location. This approach minimizes extensive interfacial reactions by injection of the reinforcing particulates at a spatial location where the atomized matrix spray contains a limited amount of volume fraction liquid. (3:45 p.m.)
THE EFFECT OF ADIABATIC HEATING AND CAVITATION ON THE TENSILE FLOW STABILITY OF HIGH RATE SUPERPlASTICITY: Thomas R. Bieler and Amiya K. Mukherjee, Department of Mechanical Engineering, Division of Materials Science, University of California, Davis, 95616. The tensile behavior of IN90211, a mechanically alloyed aluminum, was characterized at strain rates between 0.0001 and 340/sec at 425-485·C. At strain rates above O.I/sec, superplastic elongations were obtained. Cavitation was precluded by several factors, including the grain shape, orientation, and the high lattice diffusivity (0.76-0.83 Tm)' However, a dip in elongation to below 200% in the middle of the superplastic regime was observed. An analysis of heat conduction showed that a "thermal flaw size" of 25-50'C occurred along the tensile axis, causing a deformation instabil ity, and consequently poor elongations. Elongation predi ct ions based upon Hart I s criterion for flow stabil ity provided good agreement with observed elongation of the "thermally flawed specimens. Work supported by AFOSR-86-0091. (4: 15 p.m.)
THE STRENGTH AND FRACTURE OF SOLID STATE DIFFUSION BONDED JOINTS IN AI-Li ALLOY (8090) SHEET: P.G. Partridge, D.V. Dunford, Materials & Structures Dept, Royal Aerospace Establishment, Farnborough, Hants UK. There is increasing interest in joining AI-Li alloys by diffusion bonding (DB). DB joints have been produced in 8090 alloy above the superplastic forming temperature of ~515°C. The shear and peel strengths and the fracture behaviour of these joints are described and compared with the shear and fracture of the base alloy. The effect of test piece design, surface oxidation, bonding procedure and testing technique on the measured joint
properties are summarised.
(2:00 p.m.)
SUPERPLASTIC BEHAVIOR IN CERAMICS, CERAMIC COMPOSITES, METAL MATRIX COMPOSITES,AND INTERMETALLICS; J Wadsworth T. G. Nieh, Lockheed Research and Development, Lockheed Missiles and Space Co., 3251 Hanover St., Palo Alto, CA 94025, and meg D. Sherby, Stanford University, Stanford, CA 94305. Superplastic behavior has been observed in relatively recent times in ceramics such as yttria-stabilized zirconia polycrystals (Y - TZP), ceramic composites such as Y - TZP containing alumina particles, metal matrix composites such as ceramic whisker and particulate reinforced aluminum alloys, and intermetallics such as nickel silicides and certain aluminides. In this paper an overview will be presented of these recent observations. The microstructures and corresponding strain rates for superplastic flow will be shown and comparisons with conventional superplastic metals will be made. (2:45 p.m.) SUPERPLASTICITY IN Al-SiC MMC'S: John Pilling and Pauline M.
Virro, Department of Metallurgical Engineering, Michigan Technological University, Houghton, MI
49931
Metal matrix composites based on aluminium alloys 2014 and 7475 with 15 vol.% SiC produced using the co-spray process, were thermomechanically processed into sheet material. A combination of solution heat treatment, ageing, warm and cold rolling resulted in grain sizes of ~ 13 ~ and 7 ~m respectively following recrystallisation. Strain rate sensitivities of 0.4 and greater were recorded for both materials under conditions of strain rate and temperature at which the un-reinforced materials are normally superplastic. Elongations to failure of 300 to 450% were attained with only moderate superimposed pressures. Analysis of the failed test pieces showed that elongation to failure were limited, not by cavitation, but by strain enhanced grain growth. (3: 15 p.m.)
Superplastic Behavior of Spray Atomized and Deposited AI-4Li1Mg-O.5Ge-O.2Zr and AI-4Li-l Mg-O.5Ge-O.2ZrISi~p Materials: J. Swatl, FA Mohamed and E.J. Lavernia Materials Section Department of Mechanical Engineering, University of California: Irvine, CA 92717. In the present study the microstructures and superplastic behavior of AI-4Li-l Mg-O.5Ge-O.2Zr and AI-4Li-l Mg-O.5Ge-O.2ZrISiCp materials produced by spray atomization and deposition are studied. The spray atomization and deposition studies involved disintegration of a stream of molten metal into a fine dispersion of droplets using high energy inert gases. The resultant distribution of droplets was then directed towards a substrate, where it impacted and built up into a deposit of predetermined thickness. For
The results are discussed in terms
of the deformation and fracture of modes of this alloy. The possibility of combining superplastic forming with DB of AI-Li alloys is considered. (4:45 p.m.)
MECHANISMS OF DIFFUSION BONDING IN TITANIUM ALLOYS*: Chimata Gandhi, Rockwell International Science Center, Thousand Oaks, CA 91360 Titanium alloys and titanium aluminides are readily diffusionbondable at high temperatures because of their high solubility for oxygen and the consequent removal of the tenacious oxide layer in an inert atmosphere. Analytical models are being developed to understand the micromechanisms of diffusion bonding and to predict bonding times in these alloys. The diffusion bonding occurs in two stages. In the first stage, large asperities collapse during initial loading by plastiC deformation or by power law creep deformation, In the second stage, pores shrink and sinter by a number of mechanisms. The rate of shrinkage of the pores depends on their size and shape, on the magnitudes of surface and boundary diffusion coefficients and, to a lesser extent, on power law creep deformation. The predicted diffusion bonding times are in good agreement with the experimental data for titanium alloys. *Research supported by Rockwell International IR&D funds.
44--Annual Meeting
ADVANCED METAL & CERAMIC MATRIX COMPOSITES ll: Processing Metal Matrix Composites
TUESDAY AM
Sponsored by the TMS Powder Metallurgy Committee, TMS Physical Metallurgy Committee, and ASM-TMS Composites Committee Tuesday AM February 20, 1990
Room: ParkD Inn at the Park
Session Chainnen: Allan H. Clauer, Battelle Columbus Division, 505 King Avenue, Columbus, OH 43201; P. Ramakrishnan, Department of Metallurgical Engineering, Indian Institute of Technology, Powai, Bombay 400076, India
Special Joint Symposium sponsored by The Minerals, Metals & Materials Society (TMS) and The American Ceramic Society (ACerS~ Advanced Aerospace Materials From User and Supplier Perspectives New metals, ceramics and composites are rapidly evolving as candidates for applications within commercial and military aerospace market. Panelists representing major aerospace manufacturers will discuss their requirements for these new high performance materials with industry suppliers. Major application areas to be examined include missile, airframe, engine and auxiliary equipment needs. These demands will be analyzed with respect to state-of-the-art supplier capabilities and their projections for technology advancement. Overall industry trends will be examined, cost and technological limitations will be reviewed and an outline of evolving opportunities for advanced materials within the aerospace marketplace will be constructed. Symposium participation will be opened for audience questions and discussion after brief presentations by each of the panelists.
(8:30 a.m.) APPLICATIONS OF RST TO THE PROCESSING OF METAL MATRIX COMPOSITES: F. H. Froes, WRDC/MLLS, Wright-Patterson AFB, OH 45433; J. V. Wood, University of Nottingham, Nottingham, England; Y-W. Kim, G. Das. and S. Krishnamurthy, MetcutMaterials Research Group, P.O. Box 33511, WPAFB, OH 45433 The RST approach can be used to fabricate metal matrix composites (MMC) in three basic ways: via a direct powder route, by production of a RS ribbon which is subsequently used to fabricate the composite, and by a co-spray process in which the reinforcement and liquid matrix are sprayed together onto a substrate. These three methods will be discussed drawing examples from MMC's produced from aluminum. titanium and magnesium and intermeta1lics based on these elements; reinforcements considered will include particulate, whiskers and continuous fibers.
ADVANCED AEROSPACE MATERIALS
FROM USER AND SUPPLIER PERSPECTIVES
Joint Organizing CommittH
User Panelists
Suppll.r Pan.llsts
Dr. Clifford P. Ballard, Program Chairman Allied-Signal Inc. P. O. Box 1021 R Morristown, NJ 07962·1021
Robert Harrison Manager Technology Transfer Genaral Electric Aircraft Enginas H85 1 Neuman Way Cincinnati, OH 45215
Wa~er Cebulak
Mr. James D. We~erlen Inc. 7460 Girard Avenue, Su~e 8 La Jolla, CA 92037 We~erlen,
Dr. R. Nathan Katz SlCMT·DA U. S. Army Materials Technology Laboratory Watertown, MA02172
"Who wUI be Int.mteel" • • • • • •• Aerosplce Engl",e,.. Engina Airframe Miss~e
• Aerosplce Suppliers Materials Cof11lonants Slbsystems • Mat,rills Scientists Government Plannars
James Munier Chief Engineer Allied-Signal Aerospace Company Garren Auxiliary Power Division 2739 Easl Washington Street P.O. Box 5227 Phoenix, AZ. 85010 Robert Petrusha Senior Stall Engineer lockheed Missile & Space Company Inc. 0181·30 81167 1111 lockheed Way Sunyvale, CA 94089-3504 Dr. William Quist Senior Principle Enginaer Boeing Commercial Airplane Co. Mail Stop 73-43 P.O. Box 3707 Seattle, WA 98124 Dr. K K. Sankaran Section Chief McDonnall Douglas Aircraft Company Mail Code 0341200 P. O. Box 516 St. louis, MO 63116
Manager Government Technology Aluminum Cof11lany of America Alcoa laboratories Alcoa Technical Center Alcoa Center, PA 15069
Dr. Richard C. Dickenson Manager, CalborH:arbon Technologies lTV Millise & Electronics Gro~ Mail Stop SK-03 P. O. Box 650003 Dallas, TX 75265-0003 William Frieman V. P. Qual~ Assurance & Technology Howmet Corporadon 475 Steamboat Road W. Donaid Johnson Deputy Difector, Composites Division Dupont de Nemours & Company, Inc. P. O. Box 80702 Wilmington, DE 19880-0702 Dr. William Rhodes Senior Stall Scientist GTE Laboratories Inc. 40 Slyvan Road W~ham, MA 02254
Special Joint Symposium sponsored by TMS and The American Ceramics Society (ACers)
Annual Meeting-45 (9:00 a.m.) STEREOLOGY a= SOME LIQUID PHASE SINTERED METAL-METAL COMPOSITES: .G.S. upadhhaya, Department of Metallurgical Engineering, Indian Institute of Tec no!ogy, Kanpur - 208016, INDIA The present investigation describes the densification behavior and its basis on quantitative metallography of transition metal (Fe, Co, Ni) - 20 mass% Aluminium bronze systems. The green compacts pressed at a pressure of 350 MPa were sintered at 1150·C in dry hydrogen (dew point - 35·C) up to 60 min. at an interval of 15 min. Iron based composites showed better densification than the remaining ones. It was noticed that grain coarsening is less in the case of better wettable Fe-AI bronze system. A general fall in the contiguity of the matrix spheroids with increase in the sintering period during the initial stage suggests the contribution of 1st stage of liquid phase sintering. (9: 18 a.m.)
MATRIX/REINFORCEMENT DEVELOPMENT FOR ALUMINUM-BASED COMPOSITES. T. Sritharan, K. Xia, J. Heathcock and J. Mihelich, Commonwealth Aluminum, 1200 Meidinger Tower, Louisville Galleria, Louisville, KY 40202. One of the major factors hampering the commercial development of liquid metallurgy particle reinforced composites is the matrixparticle interface. The challenge is to control the reaction between the matrix and reinforcement particles to achieve limited wetting so that good distribution and interfacial bonding are obtained. The composite designer can vary the reinforcement type and/or the matrix alloy type to achieve the optimum degree of matrixparticle interaction. This paper describes Comalco's work on the development of Al MMC' s with novel matrix and reinforcement materials to achieve this end. Particular attention is paid to the
interfacial
reactions
that occur
between various matrix
interaction
on
alloys and reinforcement materials. the matrix reinforcement composites is discussed.
In addition, the effect of the
properties
of
(9:36 a.m.) THERMAL ANALYSIS OF THE SOLIDIFICATION PROCESSING OF METAL MATRIX COMPOSITES. D. Bhanja, R. B. Bhagat and M. F. Amateau, Engineering Science and Mechanics Department, 227 Hammond Building, The Pennsylvania State University, University Park, PA 16802. We have attempted to model the infiltration of molten metal into the fiber preforms during the fabrication of metal matrix composites by high pressure infiltration casting (HiPIC). This leads to the prediction of a threshold pressure for complete infiltration. Next, we incorporate the effect of pressure on solidification temperature using Clausius-Claypeyron relationship. Finally, higher values of heat transfer coefficient for conduction through die wall are used to reflect the reduced contact resistance. A two-dimensional non-linear finite element analysis provides a complete thermal history during the solidification processing of composites. The thermal analysis results are of inunense importance to the optimization of the processing parameters for fabricating composites with improved mechanical properties.
(9:54 a.m.) PROCESSING AND PROPERTIES OF LAMINATED METAL MATRIX COMPOSITES. Mark A. Kelley and Maurice F. Amateau, Applied Research Laboratory, The Pennsylvania State University, Post Office Box 30, State College, PA 16804. Laminated nickel matrix metal composites containing silicon carbide whisker reinforcements were fabricated by tape casting techniques. Orthotropic lamina with fiber fractions up to 0.30 have been successfully produced in which the reinforcements are aligned in the casting direction and are uniformly dispersed in the matrix. Elastic and strength properties of these laminates can be successfully predicted using the classical laminate plate theory. Laminates have been designed in which favorable residual stresses can be induced by thermoelastic tailoring. Mechanical properties of various simple and complex laminate designs have been measured and compared with predicted properties.
(10: 12 a.m.)
WHISKER REINFORCED COMPOSITES PREPARED FROM WET BALL MILLED ALUMINIUM POWDERS: J.Shimizu,J.Kusui,A.Tanaka and O.Iwao,Toyo Aluminium K.K.,R&D I,aboratories, 4-chome,Aioi-cho,Yao-shi,Osaka,Japan. Whisker reinforced aluminium alloy composites were com30lidated by vacuum hot press followed by hot extrusion. To get an uniform distribution of whinkerc it i8 indispensable to use powders having a small particle diameter. Wet ballmilled fine aluminium powders were found to be quite effective for the uniform distribution of whisker and hence mechanical properties of extruded composites. The effect of wet
ballmilling and consolidation conditions on the tenSile strength of composites has been investigated. The result showed that the tensile strength of AA 6061 reinfoced by 20 volume % of SiC whiskers reached to 630MPa with 2% of elongation. This high strength is concluded to attribute to uniform distribution of whiskers and fillely dispersed aluminium oxides. (10:30 a.m.) MEASUREMENTS OF PHYSICAL PROPERTIES OF HOT PRESSED GRAPHITE/ COPPER AND GRAPHITE/ALUMINUM COMPOSITES. Irwin G. Greenfield, Department of Mechanical Engineering, University of Delaware, Newark, DE 19716. The coefficient of thermal expansion (CTE), electrical resistivity and thermal conductivity were measured in a series of high purity copper and aluminum matrix composites with volume fractions ranging from about 0.1 to 0.6 of graphite chopped fibers. The aspect ratio of the high-modulus fibers were distributed between 30 to 100. Experimental results demonstrate the influence of the powder-metallurgical processing parameters, such as temperature and pressure, on physical properties of the hot pressed billets. A CTE of 6.6 microns/mK was found for the 0.5 volume fraction graphite/aluminum composite; whereas, for a copper matrix material the CTE at 0.5 volume fraction was 5.5 microns/mK. In addition to the volume fraction of graphite fibers, the thermal conductivity and electrical resistivity depend on voids remaining after processing and the character of the interface.
(10:46 a.m.)
BRAZE AND DIFFUSION BONDING OF ALUHINUH-IIATRU COIIPOSITES: L.!L...R.G. Thompson, Univ. of AlabllllUl, Birminghall, AL 35294, G.R. Edwards, Colorado School of Hines, Golden, CO 80401
~,
Aluminum-lI&trix cOllposites were studied in an attempt to develop a technique for bonding at relatively low tellperatures (400C) in air. The general technique used for joining tvo aluminum-lI&trix parts vas to isothermally solidify a zinc-aluminum eutectic. Several lIethods for applying zinc to aluminum vere studied including zinc foils, zincating, and sputter-coated silver interfaces between the aluminum and the zinc. Bonding vas accollplished using a range of tellperatures (380-450C) and tilles (15-60 lIinutes) at a pressure of 1000 psi. A 1I0dification of the above technique utilized an aluminum fluxing agent to assist the wetting of the eutectic liquid on the aluminum II&trix. Fluxes active over the temperature range of 400-600C vere studied. The bonding process vas phenomenologic'lly a three stage process of brazing, isothermal solidification, and homogenization. Results vill be reported in terms of bond strength, phase relationships, and rates of solidification and hOllogenization. (11 :06 a.m.)
Structures and Properties oi High Temperature AI-Ti/SiC Metal Matrix Composites processed by Spray Atomization and Deposition: E.J. Lavernia and FA Mohamed, Materials Section, Department of Mechanical Engineering, University of California, Irvine, CA 92717; William E. Frazier, Advanced Metallic and Ceramic Branch, Naval Air Development Center, Warminster PA 18974. This work reports on a study of the microstructures and mechanical behavior of spray deposited AI-Ti/SiC metal matrix composites for elevated temperature applications. The microstructural characteristics of the spray depOSited material is investigated as a function of processing conditions. The mechanical behavior studies include both room and high temperature properties. The microstructural studies are accomplished uSing the scanning, transmission and optical microscopes. (11 :24 a.m.) Effect of N/c Ratio on Mechanical Properties in Ti(C,N) Base Cermet Haiime Yasui-, Hiroshi Tanaka- and Toshiro Kobayashi" - N'l'K Technical Ceramics Division NGK Spark Plug Co., Ltd. 2808 Iwasaki, Komaki, Aichi 485, Japan •• School of Prodution Systems Engneering Toyohashi Un1 versi ty of Technology 1-1 H1barigaoka, Tenpaku-cho, Toyohashi, Aichi 440, Japan The present experiment has been carried out to investigate the effect of N/c ratio on the mechanical properties in Tl(C,N) base cermet. The mechanical properties to be mainly examined are as follows: (a) the static fracture toughness and the comparison of its meaSurement-methods. (b) the dynamic fracture toughness by Instrumented Charpy impact testins system and the effect of loading rate on its value. (c) the cyclic fatigue properties by Rimorph type actuator and the effect of frequency response on its life. Moreover, the meaSurement of thermal shock resistance and transverse rupture strength, etc. and the observation of material microstructure were carried out.
46-Annual Meeting (11 :42 a.m.)
FABRICATION OF HIGH TEMPERATURE INTERMETALLIC MATRIX COMPOSITES: S. Nourbakhsh F.L. Liang and H. Margolin, Department of Metallurgy and Materials Science, Polytechnic University, 333 Jay Street, Brooklyn, NY 11201. The need for strong lightweight materials for elevated temperature use has led to the exploration of aluminide intermetallics. Properties of such intermetallics have been improved by the incorporation of ceramic fibers (particulates). Enhancement of mechanical properties can only be achieved by proper tailoring of the fiber (particulate)/matrix interface. The key to the development of successful intermetallic matrix composites relies extensively on the fabrication route. The aim of this paper is to provide an up-to-date review of the current fabrication techniques. Special attention will be paid to the pressure casting technique which is currently being pursued at Polytechnic University. This work was supported by the SOlO/1ST under ONR Contract Number NOOOI4-86-K-0552.
ALUMINA AND BAUXITE II: Bayer Process
Sponsored by TMS Aluminum Committee TuesdayAM February 20, 1990
Room: Salon 3 Marriott
Session Chairman: W.C. Sleppy, Alcoa Laboratories, Alcoa Center, PA 15069
(8:30 a.m.) NABALCO CONTINUES DEVELOPMENT OF QUANTITY WITH QUALITY: G. P •
Brown, Nabalco pty Limited, Gave, Northern Territory, Australia. Nabalco commenced production of smelter grade sandy Alumina in 1979 using a high productivity process. To meet increasing market demands during the past 10 years, the Gave Plant has continued to produce a top quality Alumina and increase production rates by 30%. This has been achieved with minimal capital expenditure, by progressively increasing the precipit~tion liquor productivity and improving the plant operational availability.
(9:00 a.m.) PREDESILICATION OF PIJIGUAOS BAUXITE: M. Faneitte, R. Gil, O. Hernandez, Interalumina, Apartado 289, Puerto Ordaz, Venezuela. Pijiguaos bauxite is a trihydrate bauxite characterized by low reactive silica content of 1-2% and high 5-8% quartz. Bauxites of such low silica content are known to cause operational and product impurity control problems in the alumina plants. The current studies were undertaken to quantify the likely problems and develop if needed, optimun and suitabl solutions to INTERALUMINA considering the expansion to 2 x 10 5 TPA. The studies confirm that in the straight forward trihydrate bauxite digestion as currently practiced in the alumina plant, the concentration of silica in the digested liquor will rise to unacceptable levels; in the range of 0.8-1.40 g SiO /100 g Na O. The most effective and economic solution for the pI~nt was f06nd to be the introduction of a predesilication step prior to digestion. Most of the desilication product was formed during the predesilication. (9:30 a.m.) A NEW METHOD OF SODIUM OXALATE DESTRUCTION: Kwat The, Robin Brassinga, George D. Fulford, Alcan International Llmlted, P.O. Box 1250, Jonquiere, Quebec, Canada G7S 4KB, Lee' Beaudette, Carleton University, Ottawa, Ontario, Canada K1S 5B6, W. Doug Gould, Ronald G.L. McCready, CANMET - Energy, Mines and Resources Canada, Ottawa, Ontario, Canada KiA OGl. Sodi um oxal ate, a by-product from the Bayer process, is a health hazard and must therefore be disposed of in an environmentally safe manner. Recently Alcan, Carleton Uni versity and CANMET have successfully developed a 1aboratory process for the decomposition of the compound by microbial metabol ism. Thi s paper describes the experiment carried out and discuss the results obtained.
(10:00 a.m.) A SIMPLE, REAGENT-TECHNIQUE TO IMPROVE HYDRATE FILTRATION: M.E. Keeney, Industrial Minerals Research Laboratory, Curtin University of Technology, School of Applied Chemistry, Bentley, WA 6102, Australia and R.L. Harvey, Nabalco Pty Ltd, Nhulunbuy, NT 0880, Australia. The alumina industry makes extensive use of chemical reagents in hydrate filtration to reduce filter cake moistures. Most of the commercial dewatering aids for hydrate are simple reagent blends, usually employing an inexpensive tall oil/tallow fatty acid in combination with a nonionic surfactant. Recent work by Nabalco/IMRL to improve hydrate filtration has shown that fatty acids alone can be utilized very effectively as hydrate drainage aids, provided the condensate in the spray/ trough wash system is sufficiently caustic to fully convert the fatty acid to a water-soluble soap. A simple drainage aid technique based upon on-site, fatty acid causticisation has been fully implemented at the Nabalco refinery (Gove, Australia) with significant plant savings and operational improvements. The new technique is currently under evaluation by a number of alumina refineries worldwide. This paper briefly describes the physiochemical processes involved in the use of drainage aids in hydrate filtration, the benefits and down-stream effects associated with the use of fatty acids as a hydrate drainage aid and possible methods of implementation of the fatty acid-based process, both in deliquoring and dewatering applications. (10:30 a.m.) INCINERATION OF SODIUM OXALATE AS OXALATE/FUEL OIL SLURRY IN A ROTARY CALCINER: S. Kumar, Alcan Jamaica Company, Kirkvine P.O., Jamaica, West Indies. In the Bayer process, presence of organic carbon in the bauxite is the source of dissolved sodium oxalate impurity in the aluminate liquor. Sodium oxalate, when dissolved in liquor, is known to have a deleterious effect on hydrate precipitation rate and hence the liquor productivity. However, when the oxalate concentration in the liquor exceeds the supersaturation level, it coprecipitates with hydrate resulting in poor agglomeration and hence a fine hydrate that is difficult to classify. Hydrate produced under these conditions is also known to be extra fragile as confirmed by excessive breakdown during the calcination step. In order to avoid co-precipitation of sodium oxalate with hydrate, an Oxalate Removal System using a dense bed crystallizer on a slightly concentrated evaporator product liquor, is being operated at Ewarton, Aljam. The sodium oxalate, so produced, is converted into calcium oxalate by reacting with lime slurry. Pilot plant tests have been carried out to incinerate oxalate as calcium oxalate/fuel oil slurry in a rotary kiln. Based on the pilot plant studies, it has been demonstrated that disposal of oxalate as a fuel oil slurry in a rotary kiln is a viable process. A plant scale system to dispose of sodium oxalate via incineration route is being implemented in the alumina plants of Alcan in Jamaica. (11 :00 a.m.) MgO AND CHLORITES AS ADDITIVES FOR AUTOCLAVING DIASPORIC BAUXITE: Cao Rong-jiang and C. Liang, General Research Institute for Nonferrous Metal, Beijing 100088, People's Republic of China. In literatures and Bayer plant practice, MgO from the bauxite ore or the dolomite carried in from lime stone are precipited as Mg(OH) in red mud. But in certain conditions MgO added intentional1y to autoclave the diasporic bauxite beneficially those are: (1) lengthening the induction period in dissolving AI 0 from diaspore bauxite. (2) alterating CaTi03 scale to be a2 3 Magnesium Ferro (Ferric) aluminum silicate i.g. (Mg, Fe)R [(AI Si ) 0 ]. (OH) (n<=6), preCipitated in red mud ad alleviatfng4t~e d¥fficult~ of the scale p~oblem. (3) Alleviating the so call "high temperature silicate scale" cleaning problem such as the scale from illite whic8 precipited on the tube at the temperature range, say 200-250 C. It is known a new process with MqO cont~ineo additives using tube digistion for chinese dIaspore baUXIte IS already at plant design stage.
Annual Meeting--47
CARBON TECHNOLOGY ill Sponsored by TMS Aluminum Committee (formerly Light Metals) TuesdayAM February 20, 1990
Room: Salon 2
Marriott
Session Chairmen: Euel R. Cutshall, Manufacturing Technology Laboratory, Reynolds Metals Company, Sheffield, AL 35660; Wayne R. Hale, Kaiser Aluminum and Chemical Corp., East 2107 Hawthorne Road, Mead, WA 99021 (8:30 a.m.) PROPERTIES OF CATHODE CARBON BWCKS AFTER 10 YEARS OF OPERATION, J. Mittag, Sigri GmbH, Werk Griesheim, Stroofstrasse 27, 0-6230 Frankton I Main 83, West Germany; E. Bernhauser, Schweizerische Aluminium, Feldeggstrasse 4, CH - 8034, Ziirich, West Germany; H. Friedli, Schweizerische Aluminium AG, Werk Steg, CH 3945 Steg, Wallis, West Germany. The aim of this paper is to give a brief survey of the properties of cathode carbons, both amorphous and semi-graphite, which were more than 10 years in pot operation and taken out of service but not as a result of cathode failure. Typical characteristics of these old, well behaved cathodes are: Uniform grade of infiltration Absence of aluminium carbide and cryolite No venical (only some horizontal) cracks Wide range of electrical conductivity over the entire height of the blocks. From these findings, conclusions can be drawn about the causes for regular and early deterioration of cathode carbons. (9:00 a.m_) CEMENT FOR COLLECTOR BAR - CATHODE BWCK JOINT, L. Castonguay, S. K. Nadkarni, Alcan International Limited, Arvida Research and Development Centre, P. O. Box 1250, Jonqui~re, Qu~bec, Canada G7S 4K8. In the installation of the cathode assembly of the aluminium reduction cells, the joint between the steel bar and the cathode block is made either by mix rodding or by cast iron rodding. Mix rodded joints are easy to install as no specialized equipment is necessary. However, mix rodding is associated with problems of exposure to noise and fumes. These problems were overcome by replacing the conventional hot mix by a cement which can be filled in a cold condition in the space between the block and the bar without any need for tamping. (9:30 a.m.) ULTRASOUND FOR QUALfIY CONTROL OF CARBON CATHODE MATERIALS, T. Foosnaes, S. Jarek, T. Oksefjell, H. A !?lye, Institute of Inorganic Chemistry, Norwegian Institute of Technology, N-7034 Trondheim, Norway. Ultrasound has been used for detection of inner cracks in baked cathode blocks. Transverse and longitudinal cracks were predicted based on the change in sound velocity by different placement of the detector. The actual cracks were later found by cutting while no cracks were found in blocks where the difference in sound velocity was below a threshold value. It was also found a linear relationship between Young's modulus and the square of the velocity times density. Ultrasound was also used for measurement of ramming paste compaction.
00:00 a.m.) COMPUTER SIMULATION OF A VERTICAL RING FURNACE, T. BourgeoiS, R. T. Bui, A Charette, Universite du Quebec a Chicoutimi, 555 Universite Blvd., Chicoutimi, Quebec, Canada G7H 2Bl; B. A Sadler, A O. Tomsett, Comalco Research Center, 15 Edgar's Road, Thomastown, Victoria, Australia 3074. A transient two-dimensional mathematical model of a vertical flue ring furnace is presented. Emphasis is laid on the main heat balance elements. Heat sources (fuel, volatiles, packing coke) and heat sinks (gas, brick, coke, anode and losses through cover, sidewall and foundation) are all considered together with air infiltration and gas mass flow. Heat balance, temperature distributions, gas composition, infiltration levels, fuel consumption are determined for a given set of operation and design parameters. Model characteristics and capabilities are discussed. Results obtained for a standard furnace fire cycle are compared with available on site measurements.
(10:30 a.m.) METHODOLOGY FOR THE EVALUATION OF A BAKING ANODE FURNACE, Luis Bonomie. Wilfredo Salas, Jost A Rojas P., Alcasa, Aluminio del Caroni, SA, Zona Industrial Matanzas 8016, Ciudad Guayana, Venezuela. The baked furnace is the last process for manufacturing anodic block whicb are used in aluminum reduction pots in the type of prebaked anodes. They must gather various types of pbysical and cbemical characteristics whicn- determines the quality required for optimum function in the pots. BaIting process plays a very important roll in the development of these cbaracteri$tics therefore, it makes it necessary to develop a methodology whicb ensures to estimate the incidence of the variables that inIluence the process of anodic block qualities. The purpose of this work is to develop a methodology wbich measures the variables mentioned above. The methodology suggest two steps: L Evaluating quality of anodes and variables that interfere furnace functions; pbysical and chemical analysis of anodes before and after baking; temperature distn'bution in tbe anodes; temperature distribution at Duid walls; suction pressure; efficiency of burners; b. Relationship between Oato-Oiagbostic of evaluated furnace.
CAST SHOP TECHNOLOGY m: Metal Treatment - Fluxing Sponsored by the TMS Aluminum Committee Tuesday AM February 20, 1990
Room: Salon 4
&;
5
Marriott
Session Chairman: H. Yu, Alcoa Laboratories, Molten Metal Processing Division, Alcoa Center, PA 15069
(8:30 a.m.) A MATHEMATICAL MODEL OF THE POROUS TROUGH REACfOR IN HYDROGEN REMOVAL FROM AN ALUMINUM MELT: J G Stevens. H. Yu, C. E. Eckert, Alcoa Laboratories, Alcoa Center, PA 15069.
In multi-alloy ingot casting units, frequent alloy changes are often necessary. An economical means of metal treatment in such casting units is the porous trough reactor which does not require flushing with a large amount of metal. Costs of alloy changes are therefore reduced. A mathematical model has been created for performance prediction and optimization of the porous trough reactor. In the mathematical model, plug flow of the metal stream was assumed and axial diffusion was considered. Mass transfer coefficients were detennined using plant reactor data. Model predictions were tested with 6061 alloy experiments and found to be reasonable.
(8:55 a_m.) MATHEMATICAL AND PHYSICAL MODELLING OF MELT TREATMENT PROCESSES: J. F. Grandfield, D. W. irwin, S. Brumale and C. J. Simensen*, Comalco Research Centre, 15 Edgars Road, Thomastown, Victoria, 3074, Australia. (*Now at Centre for Industrial Research, Oslo, Norway). Both physical and mathematical modelling have proved useful in controlling aluminium melt refining processes. This paper details some examples. Water modelling of various impeller designs was carried out in a full size tank. Deoxygenation rates were measured at various rotor speeds and gas flow rates; the characteristics of the bubble clouds produced were compared. The ratio of rotor speed to gas flow was found to be a key parameter. The mathematical approaches which have been used to describe sodium, hydrogen, calcium, etc., and particle removal are reviewed and compared with experimental data. While sodium and hydrogen removal seem to be readily predictable, particle flotation modelling requires more work. Filter modelling is reviewed. Means of calculating filter efficiencies are given. A good correspondence exists between predicted and measured data in water analogue and aluminium systems.
(9:20 a.m_) GAS FLUXING OPERATION IN ALUMINIUM MELT REFINING LABORATORY AND PLANT INVESTIGATIONS: M. Nilmani and P. K. Thay, Department of Chemical Engineering, The University of Melbourne, Parkville, Victoria 3052, Australia and C. J. Simensen* and D. W. Irwin, Comalco Research Centre, IS Edgars Road, Thomastown, Victoria, 3074, Australia. (*Now at Centre for Industrial Research, Oslo, Norway). This paper discusses the gas injection process in aluminium pretreannent furnaces. A general theory of gas fluxing for the removal of sodium is outlined. Results obtained from both physical modelling experiments and plant tests using a novel injection device, called Helical Plug Lance, are presented. Other injection devices such as plain lance and diffuser are also compared. Under the same experimental conditions, Helical Plug lance has been found 50% more efficient than the plain lance.
48-Annual Meeting
With more than 100 units inslalled worldwide in the eighties. the A1pur technology has been fully proven.
Results of laboratory studies and experiences with aluminum in production operations show comparisons for various alloys within the sheet, plate, and extrusion industries and for various furnaces and in-line fluxing operations. Data including alkali metal removal, hydrogen removal, and solid particulate removal will be reviewed as compared to other degassers. Air samples have been taken in the stack and working area to show the non-toxic nature of AGA Mix 14. Finally, some aspects of gas dispersion into melts will be discussed and their effect on how AGA Mix 14 actually works in aluminum melts.
The present and future applications of the process are discussed : In-line treatment during casting, new developments with the A1pur rotary mixer (foundry casting alloys, continuous casting, furnace treaanen~ ladle treatment).The recent improvements of the technology are reviewed (new immersion healers, automatic operation etc...) as well as the future developments of bigger units for higher flowrales.
DEFECTS AND DEFECT REDUCTION PRO· CESSING IN SEMICONDUCTOR BETEROS· TRUCTURES ill: Defect Reduction in Misfitted Heterostructures
(9:45 a.m.) ALPlJR TECHNOLOGY· PRESENT AND FU1URE J. Bildslein - I. Venire - Pechiney Aluminium Engineering
BP 24 - 38340 Voreppe - France
(10: 10 a . m.)
BREAK
(10 : 20 a.m.) PRETREATMENT IN POlLINES CRUCIBLES: TIlE MIXAL PROCESS: ~,Aluminium Pechiney, b.p. 114 - 73300 Saint-Jean-de-Maurienne, France and~, Pechiney Recherches, b.p. 27 - 38340 Voreppe, France. An industrial treatment of aluminum in potlines crucibles (MIXAL) has been commissioned in Pechiney Saint-Jean-de-Maurienne plant in 1988. The objective of the process is to reduce the alkali content. The principle is based on the injection of an Ar-CI2 mixture through a rotary mixer. A complete description of the installation is given including automatization and effluent treatment station. The main parameters of the process are presented as well as plant results concerning alkali removal efficiency, elements of running cost, ergonomics ... (10:45 a.m.) THE ALCOA TELEGAS JITll'INSTRUMENT: D A Anderson, Equipment Development Division and D. A. Granger, Molten Metal Processing Division, Alcoa Laboratories, Alcoa Center, PA 15069, R. R. Avery, Alcoa Massena Operations, Massena, NY 13662. Thc Telegas instrument represented a major breakthrough in hydrogen measurement when it was introduced 30 years ago. It provided a rapid, quantitative measure of hydrogen content at reasonable cost. It is still the only quantitative method' which can be used for feedback control of the fluxing unit. However, the Telegas instrument is bulky and the readings are easy to misinterpret. Alcoa has developed an improved instrument, Telegas JITII', which addresses these problems. This paper discusses ingot plant experience with the new instrument. It is compact and lightweight while using the same basic measurement principle as the original Telegas instrument. The instrument package contains a microprocessor based control system. operator interface via a microterminal and a printer for hard copy output. Operation is under control of the microprocessor and is automatic; the operator only tells it when to start and stop. The software is menu driven, fricndly and frees the operator of the error prone task of determining the hydrogen content from the Telegas reading. a separate temperature measurement and a chart.
Sponsored by the TMS Electronic and Photonic Device Materials and Electronic Materials Committee Tuesday AM February 20, 1990
Room: Orangewood IT Quality Hotel
Session Chairman: E.A. Fitzgerald, AT&T Bell Laboratories, Murray Hill, NJ 07974 (9 : 00 a.m.)
Invited
GROWTH, DEFECfS AND CONTROL IN HETEROSTRUCfURE MATERIALS: John M. Parsey, Jr. and C. L. Reynolds. AT&T BeU Laboratories. Reading. PA 19612
Epitaxial growth of compound semicOnductor materials suitable for heterostructure field-efkct transistors is a demanding procedure. To realized reproducible and uniform characteristics in individual devices and integrated circuits, control over all aspects of the crystal growth process is required; e.g.. growth rates. material compositions, and impurity concentrations. Detailed characterization of the molecular beam epitaxy (MBE) growth of GaAs-AlGaAs materials reveals control within a few monolayers per unique layer in the material structure. Defect density is an issue in materials quality and device yield; "oval" defects. and other defect structures. can be fatal . However. the densities may be influenced at various stages of the preparation and growth. Electronic charge density and location in the density structure is also crucial and strongly impacted by details of the growth process. Uniformity of the layers and impurity species is very important to realizing consistent device characteristics with high yields. In 3" diameter wafers carrier concentrations can be maintained uniform at roughly 2-3%, the Al fraction is controlled to ±I % across the wafer and ±2% wafer-to-wafer. and thicknesses. similarly. The relationship between this level of control and the concomitant effi:cts on devices wiu be discussed. (9 : 30 a.m.) Invited GENERATION OF MISFIT DISLOCATIONS IN HETEROSTRUCTURES
R.IJ.l1ru1!
ALCAN EXPERIENCE WITH THE USE OF SULPHUR HEXAFLUORIDE FOR AlUMINIUM FLUXING: P. Waite, D. Bernard : Alcan International Limited Arvida Research and Develop~ent Centre, P.O. Box 1250, Jonquiere, (Quebec), Canada, C7S 4K8 . Recently, it has been suggested that chlorine-free gas mixtures composed of sulphur hexafluoride (SF~) and an inert carrier gas, can be employed for both furnace and in-I ine fluxing. Metal treat.ent performance and emission data obtained from laboratory and plant trials involving sulphur hexafluoride/nitrogen mixtures (furnace)t and sulphur hexafluoride/argon (in-line), is reviewed. n addition results will be presented from plant trials which examined the coinjection of fluoride containing gases such as sulphur hexafluoride with chlorine and argon as a means of reducing the oxidation problems usoc'iated with chlorine fluxing. Finally, it has been found that the severe oxidation which co. .only occurs at the liquid metal surface of high Mg alloys can be effectively inhibited by use of a cover gas coaposed of sulphur hexafluoride and nitrogen. Relevant laboratory and plant data is summarized .
PRODUCTION AND LABORATORY RESULTS USING AGA MIX 14, Tim Rack, AGA Gas, Inc., P.O. Box 94737, Cleveland, Ohio 44101-4737 and Ron Corns, AGA Specialty Gas, 6421 Monclova, Maumee, Ohio 43537. AGA Mix 14 successfully replaced Chlorine and Halocarbon-12 as well as other alternatives to effectively degas and flux molten aluminum with no toxic or corrosive side effects.
a...
~
Department of Materials Science University of Virginia Charlottesville. Virginia 22903-2442 The use of elastic mi-s fit strains to generate misfit disloca-
tions and thereby control both the lattice parameter and the
dislocation density is receiving increasing attention in the
heteroepitaxial growth of compound semiconductors. Models based on equilibrium calculations have met with limited success when quantitatively applied to the experimental data from nonmetallic systems. Discrepancies between the experimentally determinated valves of critical thickness and residual strains and the smaller valves of these parameters calculated from models, are felt to be due to barriers to the acquisition of misfit dislocations. In this paper problems with both the experiments and the models will be discussed in terms of resolving the discrepancies by the incorporation of temperature and time dependencies in the data as well as in the models.
(10 : 00 a.m.) A CONTINUUM MECHANICS DESCRIPTION OF STRAIN INTERACTIONS IN gPITAXIAL LAYERS: Brian J. Bartholomeusz, Research Laboratories, Eastman Kodak Company, Rochester, N.Y. 14650 and Krishna Rajan, Materials Engineering Department, Rensselaer Polytechnic Institute, Troy, N.Y. 12180. Lattice disregistry that exists in expita xial overgrowths is often accomodated by interfacial dislocation arrays. Frank and Van der Merwe developed the first satisfactory treatment for such interfaces assuming simple harmonic interactions across them. In this presentation the calculation of the interaction energy in a hi-layered misfitting film using the continuum meth-
Annual Meeting-49 ods of Eshelby are presented. This method has the advantage over the Frank - Van der Merwe approach in that it can detail the distinct effect of each interface and permits estimates to be made of their individual contribution to the total-energy. The results of these calculations are examined in the context of observed dislocation configurations in InGaAs/GaAs strained layer superlattices. Of particular interest is the role of strained layer superlattices to inhibit dislocation propogation.
(10:20 a.m.) REDUCTION OF DEFECf DENSITY IN HETEROEPITAXIAL ae. Si I-x GROWN ON PATTERNED Si SUBSTRATES. E.A. Fitzgerald, P.E. Freeland, Y.·H. Xie, J. Michel, AT&T Bell Laboratories, Murray Hill, NJ 07974. We have grown MBE Ge.Sil_x semiconductor alloys on mesas fabricated on (001) Si substrates. Single layer films with x
BREAK
(10:55 a.m.) CRITICAL SIZE OF FINITE HEMISPHERICAL EPILAYERS: K. Jagannadham and J. Narayan, Dept. of Materials Science and Engineering, North Carolina State University, Raleigh, NC 27695-7916 Previous analysis [1] of critical size of hemispherical epilayers of GaAs on Si using surface dislocation analysis are improved in order to represent the distortions more accurately. The shear dislocation loops at the interface are replaced by the prismatic array to describe the lattice misfit between the epilayer (GaAs) and the substrate (Si). The surface dislocation analysis is used to satisfy the boundary conditions at the free surface of the epilayer. Discrete dislocation analysis using a finite number of loops is used to determine the critical size. 1. Critical thickness during three-dimensional epitaxial growth-a self-consistent approach, Mat. Sci. Engng., 1989, All3, 65. (11: 15 a.m.)
THE ISOLATION AND NUCLEATION OF MISFIT DISLOCATIONS IN STRAINED EPITAXIAL LAYERS GROWN ON PATTERNED, IONDAMAGED GaAs: G. P. Watson, M. O. Thompson, D. G. Ast, Materials Science Dept. Cornell University, Ithaca, NY 14853, A. Fischer-Colbrie and J. Miller, HP Laboratories, Hewlett Packard Co., Palo Alto, CA 94304. The misfit dislocation density of strained epitaxial III-V layers has previously been shown to be significantly reduced by isolating sections (by patterned etching) of a GaAs substrate before epitaxial growth. A disadvantage of this technique is that the wafer surface is no longer planar, which can complicate subsequent device fabrication. As an alternative method, we have investigated growth of 35ooAof Ino.osGao.9sAs by molecular beam epitaxy on substrates which were patterned and selectively damaged by Xe ion implantation (300 KeV, 10 15 cm- 2 ). Although the misfit dislocations appeared to be stopped by the ion-implanted regions, the damaged portions also acted as copious nucleation sources and the misfits that formed occurred in virtually only one direction. Results of cathodoluminescence, ion channeling, x-ray diffraction, and other techniques are discussed. (11 :35 a.m.) USE OF ION IRRADIATION DURING GROWTH OF III-V SEMICONDUCTORS ON Si(100) TO REDUCE DEFECT DENSITIES: C.-H.Choi, R. Kaspi, L. Hultman, and S.A. Barnett, Department of Materials Science and Engineering and the Materials Research Center, Northwestern University, Evanston, IL 60208. High flux, 25 eV Ar ion irradiation during the first - 10 monolayers (ML) of epitaxial InAs growth on Si(100) decreased the defect density as evidenced by a decrease in the X-ray diffraction rocking curve full-width at half maximum by a factor of - 2. Reflection high-energy electron diffraction studies showed that 25 eV bombardment also increased the critical thickness for the formation of three-dimensional islands from 2.5 to 10 ML for substrate temperatures in the range from 250 to 400 'C. Similar results are observed for InGsAs and other InAsbased alloys on Si (100). The data suggest that the decreased defect density was due to the ion-induced change in nucleation mechanism. Transmission electron microscopy studies of defect formation mechanisms and scanning tunneling microscope images of nucleation morphology will be presented.
(11 :55 a.m.) REPLICATION AND ANNIHILATION OF DISLOCATIONS DURING EPITAXIAL GROWTH: E.A. Beam III, S. Mahajan, Department of Metallurgical Engineering and Materials Science, Carnegie Mellon University,
Pittsburgh, PA 15213, W.A. Bonner, Bellcore, Red Bank, NJ 07701 The replication behavior of dislocations whose Burgers vector is
parallel to the (001) InP surface has been investigated during
homoepitaxy. Results indicate that these dislocations are replicated into the epitaxial layer without exhibiting enhanced dissociation at the epitaxial layer-substrate interface. In addition, the replicated dislocations tend to align themselves parallel to the growth direction. Computer simulations have been carried out to evaluate the atomic arrangement at the emergence points of edge dislocations whose Burgers vector is
parallel to the (001) surface.
Results show that the emergence
points are associated with protrusions and depressions. These features are responsible for the observed replication behavior.
The annihilation of dislocations during epitaxial growth will be illustrated using several examples. The governing processes are statistical in nature and are not going to playa major role in
reducing dislocation densities in epitaxial layers over that in
a low dislocation density substrate.
EXPERT SYSTEM APPLICATIONS IN MATER· IALS PROCESSING & MANUFACTURING I Sponsored by the TMS Shaping & Forming Commit· tee, the TMS Synthesis & Analysis in Materials Processing Committee and the ASM-MSD Computer Simulation Activity Tuesday AM February 20, 1990
Room: ParkB Inn at the Park
Session Chainnan: M.Y. Demeri, Research Staff, Ford Motor Company, P.O. Box 2053, Room S-1067, Dearborn, MI 48121 (8:30 a.m.) AN INTEGRATED ENVIRONMENT FOR CASTING DESIGN: R. Natarajan, C N Chu and R.L. Kashyap, School of Engineering, Purdue University, W. Lafayette, IN 47907. Casting process is heavily experience-oriented and casting design is an iterative task between casting designers and foundry experts. A knowledge-based expert system called EXCAST was developed in order to facilitate the design and manufacturing of casting components. The EXCAST system uses a fixedfeature design approach, where a generic group of features is used to facilitate the design process. Tolerancing, rounding, shrinkage, simple junction, and wall thickness are incorporated in the rule-base. These rules are suitable for the evaluation of local casting features. However, casting components generally have complicated geometric shapes, and design rules cannot always address the complicated solidification behavior arising from inter-feature interactions. An efficient geometry based sinoulation method is used to model the solidification process, and predict global casting soundness. A characterization of the solidification mechanism allows the prediction of centerline macroscopic shrinkage as well as distributed microporosities. The integration of the procedural and rule based approaches makes this system efficient for predicting castability of parts. (9:00 a.m.) Prediction of Longitudinal Surface Cracking In Continuously cast Slabs, Using a Frame Based Facility J. Suni, Carnegie Mellon UniversHy, Pittsburgh, PA 15213 and H. Henein, UniversHyof Alberta, Edmonton, Alberta, T6G 2G6 The system developed is a frame based variable handling facility. Variable types which are provided by the system include simple variables (input), table variables and function variables. Table variables, which involve a number of rules for determining a result value, allow for direct application to dispositioning in continuous casting. Function variables are determined by Lisp function expressions, which can include calls to external Fortran routines. Any level of nesting is allowed, so that a table variable might check the value of a function variable, which might check the value of another table variable, etc. Hence, this type of implementation provides the flexibility to deal wHh qualny issues in a detailed manner, but retains the simple framework of standard dispositioning guidelines. Both simple guideline approaches and more detailed approaches to predicting cracking have been developed. Using simple guidelines involves checking easily observed processing parameters (eg. carbon content, su~ur content, casting speed, etc.) against critical levels which have been established by the practice. More detailed prediction of 10ngHudinai surface cracking involves consideration of important aspects of the problem, such as heat transfer and shell formation stabilijy, steel ductility, etc. Results of both approaches will be presented.
50-Annual Meeting (9:30 a.m.)
Part design for Net·Shape Manufacturing, R. A. Miller, NSF Engineering Research Center for Net Shape Manufacturing, The Ohio State University, 1971 Neil Ave., Columbus, Ohio 43210 1be ERC for Net Shape Manufacturing has an extensive research program in many aspects of the engineering of manufacturing processes involving
dies and molds. Research activity involving knowledge-based-systems focuses primarily on part, process and die design. Ongoing wode includes forming sequence design for cold forging, design assessment tools for injection molded parts, design/redesign systems for die casting, and feature extraction and part representation for net shape parts including complex surfaces. In this paper, our work on part representation, necessary to support knowledge-based systems development and to integrate CAD/CAM systems with knowledge-based systems, will be described and discussed. 'This wode is basic to much of or design related activity and will be illustrated with examples from die casting design.
((((((10:00 a.m.)
USING AN EXPERT SYSTEM TO INTERPRET FINITE ELEMENT ANALYSIS PROCESS HISTORIES: K J Meltsner. GE-CRD, K-1, MB207, P.O. Box 8, Schenectady, NY 12301. Finite element analysis (FEA) is an essential part of many manufacturing design systems. However, FEA programs produce large data files which must be interpreted by the design engineer. As part of a jet engine design workstation, a metallurgical expert system has been provided to interpret FEA simulations of Ti-6AI-4V thermomechanical joining. The expert system includes a history scanner to reduce the FEA data files and a metallurgical knowledgebase to predict the part's microstructures and properties. The program also includes a library of micrographs and an explanation module which are available to assist the user in understanding its predictions. The current system is limited to the analysis of a particular part, but its methods are applicable to other materials and manufacturing methods. (10:30 a.m.)
AN EXPERT DECISION SUPPORT SYSTEM FOR DEFECT PREDICTION IN FORMING PROCESSES: Namita Goyal, Graduate Research Assistant, University of Pittsburgh, Pittsburgh, PA 15261. H. A. Kuhn, Technical Vice PreSident, Metalworking TeChnology Inc., Johnstown, PA 15904. An overriding concern in today's manufacturing industry is to manufacture quality parts ecOnomically without compromising their mechanical and physical properties. In precision forging processes manufacturing defect-free parts is of prime importance. This requires systematic determihation of localized potential fracture regions which need to be explored through further analysis. The approach taken for automated application of fracture criteria is to build an integrated design framework based on a set of advanced software tools which perform reasoning with hybrid data (both symbolic and numeric). An Artificial-Intelligence based decision-support system is implemented using KEE, an expert system shell, on the SUN workstation. The Finite Element Method is used selectively for simulation and validation of the potential fracture regions identified by the decision-support. The design and control of the system is influenced by the selection of part and preform geometry, material, and other processing conditions. 01:00 a.m.)
AN M. D. S.
EXPERT SYSTEM TO RECOMMEND ROLL ADJUSTMENTS IN A RAIL MILL: M. Vyas, Bethlehem Steel Corp., Bethlehem, PA 18016, J. Renn, Bethlehem Steel Corp., Bethlehem, PA 18016, Harpster, Bethlehem Steel Corp., Steelton, PA 17113
At Bethlehem Steel's Steelton Plant, floormen and rollers are using an expert system that recommends roll adjustments for the 28" rail mill. The objective is to produce rails within strict dimensional tolerances. Based on the rail sample dimensions and other process data, appropriate roll adjustments are made to maintain consistent rail dimensions. A PC-based expert system was developed and installed in the mill to recommend the "best" of several possible complex roll adjustments using real time data input by mill operators. The experience and knowledge of the mill rollers and theoretical rolling knowledge were used to establish the roll adjustment criteria for the expert system. It also includes utilities for rail dimension trend charting, data reporting and statistical quality control charts.
(tt :30 a.m.) BILLET SEQUENCING OF OPEN DIE FORGING OF STEEL 718. H. Henein Dept. of Mining Metallurgical & Petroleum Eng., Univ. of Alberta, Edmonton, AB T6G 2G6 and I. Hulthage The Robotics Institute, Carnegie Mellon, Pittsburgh, PA 15213 There is a growing demand for improved control in the production of 718 billets in order to forge fine grain products. Heuristic and algorithmic modeling appear to be vital for the achievement of this goal. Thus greater knowledge is required of the forging operation and its
relationship to the deformation and microstructure of 718 as well as advanced techniques to utilize such knowledge. This paper describes a prototype system that advises the forging Pressman on the sequencing of billets. The system utilizes a schema based hierarchical knowledge representation of information associated with forging. Sequences are found through the utilization of both heuristic and algorithmic models of the forging process and associated variables. The results are here compared with typical production forging sequences.
EXTRACTIVE METALLURGY TMS General Abstract Session Tuesday AM February 20, 1990
Room: Salon G
Marriott
Sesson Chainnan: Dr. Charles O. Bounds, Horsehead Resource Development Company, 300 Frankfurt Road, Monaca, PA 15061 (8:30
a.m.)
THERMOCHEMICAL NATURE OF MINOR ELEMENTS IN COPPER MATTES: s. Akagi, Saganoseki Smelter & Refinery, Nippon Mining Co., LTD, Saganoseki, Ohita 879-22, Japan; D.C. Lynch and W.G. Davenport, Dept. of Materials Science & Engineering, University of Arizona, Tucson, Arizona 85721. Equilibrium distribution measurements of As, Sb, and Bi between molten copper and white metal, and between molten copper and
matte were conducted at 1473 K using a static system. results of this investigation have been activity of those elements in matte and develop, in conjunction with other data partial empirical equation representing
The
used to evaluate the white metal, and to in the literature, a the activity coefficient
of As and Sb in copper mattes of widely varying compositions. The results of the investigation indicate that control of minor
element content in matte is strongly affected by the sulfur
deficiency of the matte.
In addition, the analysis has yielded
several ternary interaction coefficients for arsenic in molten
copper.
(8 : 50
a.m.)
Simulation of Teniente's Convert Through GPSS/H, R. Bustomante, G. Giacamon ond M. Borra , Foculty of Engineering ond Geological Sciences, Universidad del Norte, P. O. Box 1280 , Antofagasta , Chi-
le .
The use of a simulotion language , such os GPSS/H, allows us to appreciate the following advantages in regard to old simulation models done in FORTRAN, 0 generol purpose language, - The programs are shorter and more efficient. - It provides a natural frame. - The changes made to the simulation sustems are easier to implement. - The statistics are generated automaticaly . - Etc . The principal objetive of this study is to give for a process, a set of best decisions that one can take on time, clearly, preci sely and reliable, such as the process that takes ploce in the Te niente's convert, which is in continuos development.
(9:10 a.m.)
RATE PHENOMENA IN THE FLASH PROCESSING OF ELECTRIC ARC FURNACE DUST: ~ HY and N.J. Theme 1is , Henry Krumb School of Mines, Columbia University, New York, NY 10027. The flash processing of (EAF) dusts is a promising method for recovering zinc and producing non-toxic slags for direct disposal. In addition to experimental tests in a flash reactor (0.13m I.D.*1.5m high, l400 0 C), this study has encompassed characterization of EAF dusts, construction of a matrix of the energy/material balances and thermodynamic constraints, and development of a mathematical model of the flash reactor. In the experimental work, the controllable factors were a) zinc content of the EAF dust, b) specific smelting rate, c) temperature level of the reacting gas-solids stream (inlet rate of bulk oxygen), d) level of CO/C02 ratio in the bulk gas (inlet mix of CO, C02, 02 and N2 rates) and e) residence time of the particles in the reaction zone (total gas flowrate). This paper describes the reaction system and the preliminary results obtained.
Annual Meeting-51 (9:30 a.m.) REDUCTION OF ZINC FROM SLAGS USING A CARBON-SATURATED IRON MELT. W.J. Rankin, Dept. Chemical Engineering, University ofMelbourn~ Parkville, 3152 Australia, and S. Wright, CSIRO, Division of Mineral and Process Engineering. P.O. Box 312, Clayton 3168, Australia. In a laboratory investigation it was found that zinc was reduced from slags by reaction of ZnO and FeD to form Zn and Fe203. The Fe203 is reduced back to FeO by reaction with Fe when an iron-carbon melt is present. FeO is reduced by carbon in the iron melt and whether there is a net consumption or production of iron depends on the relative rates of the two competing reactions. In a system with no forced convection the rate of reduction of ZnO appears to be limited by the mass transfer of ferric ions in the slag. Conceptual processes for reducing zinc using an iron bath have been examined using thermodynamic simulation and laboratory experiments. Well-mixed, co-current and conter-current systems have been considered and their advantages and disadvantages compared. (9:50 a.m.) COLD PLASMA PROCESSING OF METALLURGICALLY SIGNIFICANT ORES: D.E. Bullard, R.A. Ortega, D.C. Lynch, and W.G. Davenport, Arizona Materials Laboratories, University of Arizona, 4715 E. Fort Lowell Road, Tucson, Arizona 85712. Preliminary results will be presented on a new experimental program aimed at examining the role cold plasmas may play in the processing of ores. Refractory oxide materials are being reacted in a microwave induced plasma in order that the effect of radicals, ions and energetic electrons on chlorination processing and hydrogen reduction can be determined. These initial results will focus on analysis of the gas-solid reactions and on the physico-chemical nature of the plasma. (10:10 a.m.) REMOVAL OF IMPURITIES FROM MOLTEN COPPER BY INJECTION REFINING. T. Stapurewicz and N.J. Themelis, Henry Krumb School of Mines, Columbia University, New York, N.Y. 10027. The removal of antimony from molten copper by sodium carbonate injection was investigated in 10-20 kg experimental melts. Analysis of the experimental results, by means of mathematical models·, showed that the overall process of impurity removal is controlled by diffusion of the reacting species through the metal phase. The reaction between metal and injected flux can be divided into two subprocesses: i) "transitory contact" reaction to the injected flux particles as they rise through the melt and ii) "permanent contact" reaction across the interface between the metal bath and the supernatant slag layer. The rate of the transitory contact reaction was fou~d to depend principally on the rate and depth of powder injection. The rate of the permanent contact reaction is enhanced by gas injection through the melt.
on these conslderat Ions, the heat transfer coefficient bet~een bath and .Ide ledge In aluminium cells could be estimated (11:10 a.m.) ~
CAST-lWSE QJALITY
-
'mE REFRAcroRY WAY:
..s...B.
.5il:lley, Marketing Manager, Light Metals Division, Foseoo Internatiooal Ltd, 285 Ia1g lIcre, Nechells, Binningham, United Kirg1an No matter hew efficient any In Line filtraticn system is
there is always a distance between the cleartin:l" devioe and Refractories used for transportaticn and ClCIltrol are critical for the optimlrn quality of billet, slab etc. that is being produced. 'Ihis paper discusses critical properties of refractories and hew recent develqmant \
nold.
(11:30 a.m.)
PROTOTYPE LOCAL CONTROL SYSTEM FOR THE V-350 VENALUM CELL. L. Maldonado, J. Ramones and ~ Sarda. Industria Venezolana de Aluminio C.A., Venalum. APDO. 312, Puerto Ordaz, Estado Bolivar, Venezuela. This paper describes a prototype microcomputer based local control system, developed to operate the V-350 VENALUM cell. As implemented, this prototype enables data acquisition, processing and control actions required for the optimum contra I of the mass and ener9Y balance of the cell. Apart from performing trad1tional functions of pot voltage and alumina feeding, the outstanding features of this prototype are, among others, the monitoring of voltage signals related to individual anode current flow so that anode current distribution and stability is established for each anode. It also has a failure safety scheme which allows a non-stop normal rate of feeding in case of failures of the microcomputer unit. The prototype is composed of three main modules: central processing unit, data acquisition module and an intelligent control panel.
ruGH PERFORMANCE COPPER-BASE ALLOYSm
(10:30 a.m.)
Co-sponsored by the TMS Structural Materials and Powder Metallurgy Committees
BASIC DESIGN OF THE VENALUM CASTING PILOT UNIT FOR ALUMINIUM ALLOYS. Pedro ~ Orsetti R. Industria Venezolana de Aluminio C.A., Venalum. APDO. 312, Puerto Ordaz, Estado Bolivar, Venezuela.
Tuesday AM February 20, 1990
The present paper gives some highlights of the V-UPC project be~ng developed at Venalum, whose main objective 1S the design and installation of a casting pilot unit for aluminium alloys. This pilot unit will let us study and optimize casting and operation parameters for several aluminium alloys, having emphasis on the semi-continuous vertical casting process. A brief description of the main components is given together with the resources required to carry out the V-UPC project. The benefits which will be obtained by Venalum from the development of this project are also discussed. (10:50 a.m.) HEAT TRANSFER BETWEEN MOLTEN AND SOLID CRYOLITE BATH: Y.R. Gan and J .. Thonstad, Laboratories of Ind. Electrochemistry~ Norweg1an lnst. of Technology, N-7034, Trondheim, Norway.
The heat transfer coefficient between molten 8nd solid cryolite vas measured by recording the temperature course Inside a cold sphere ~hlch vas Immersed Into the melt. The data obtained Indicate that the measured heat transfer coefficients are higher than vhat can be calculated from dlmenslonle •• equations for convect Ive heat transfer or from measurements In lo~-temperature phy!Ic~l models. Thlg means that the heat trlln~fer between molten and sol'd LZ'lth 'S not It purely convective he~t transfer process. Heat transfer by radiation should he taken Into account In this case. The Influence of the freezing/melting process on the heat transfer coefficient In the case of free convection should also be considered. Based
Room: ParkA
Inn at the Park
Session Chairmen: R.E. Lewis, Lockheed Missiles & Space Co., Inc., Dept. 93-10, Bldg. 204, 3251 Hanover Street, Palo Alto, CA 94304; W.E. Spitzig, 211A Metals Development Bldg., Ames Laboratory, Iowa State University,Ames,~ 50011 (8:30 a.m.) THE POTENTIAL OF RAPID SOLIDIFICATION IN ODS COPPER ALLOY DEVELOPMENT: M.S. Nagorka, G.E. Lucas and C.G. Levi, University of California, Santa Barbara, CA, 93106, and S.D. Ridder, National Institute of Standards and Technology, Gaithersburg, MD, 20899. ODS Copper alloys offer an excellent combination of conductivity and strength retention above ~TM/2, but current formulations do not achieve the strength levels required by some advanced aerospace components, e.g. rocket combustors. Theoretical considerations suggest that the creep strength could be modified by suitable control of the elastic constants and interfacial characteristics between the dispersoid and the matrix, which in turn control the dislocation-particle interactions. Y203 and various RE oxides offer interesting alternatives to the usual A1 20 3 dispersoids. However, they cannot be processed by conventional internal oxidation since the alloying elements are practically insoluble in copper and form intermetallic compounds. RSP is under investigation as a way to circumvent the processing problems. Microstructure evolution during solidification and subsequent internal oxidation of Cu-Y alloys will be discussed and a preliminary assessment of their mechanical behavior will be presented.
52-Annual Meeting (9:00 a.m.)
(11 :30 a.m.)
OXIDATION BEHAVIOR OF Cu·Nb MICROCOMPOSITES: K K Chiang. K. J. Kallenborn, J. L. Yuen, and N. E. Paton, Rocketdyne Division of Rockwell International, 6633 Canoga Avenue, Canoga Park, CA. 91303 The oxidation behavior of Cu·1Sv/o Nb microcomposites was studied in air over the temperature range 600 o ·900°C. The patterns of scaling and subsurface oxidation were examined by optical metallography, SEM, XRD and microprobe analysis. In this two phase microcomposite in which the second phase takes the form of fine filaments spaced on the order of 1 jlm apart, two types of oxidation morphology were observed. For the transverse direction samples, an external scale consists of CU20 and CuO was formed with a niobium oxide inner layer. In the longitudinal direction, preferential oxidation along the niobium filament was found underneath the external copper oxide scales. Thermogravimetric analyses show the kinetics to be parabolic. The activation energy for oxidation parallel to the Nb filament direction was measured to be 22.9 kcal/mole. The oxidation of Cu·Ta and Cu-Cr microcompos~es were also investigated and compared to that of Cu-Nb. The effect ot alloy composition, microstructure of the two phase mixture, and temperature on the oxidation behavior is also considered. (9:30 a.m.)
JOINING OF AN ALUMINA-STRENGTHENED COPPER ALLOY": E Michael Hosking. Sandia National Laboratories, Physical Metallurgy Div.1831, Albuquerque, NM 87185 Alumina-strengthened copper (ASCu) alloys are receiving wide interest in the engineering community because of their stable microstructure and retention of strength following elevated temperature cycles. The purpose of this study was to examine the feasibility of brazing or inertia welding pyrolytic graphite and AISI 316L stainless steel to ASCu. All-Cu active filIer metals containing Ti were used when brazing. The fine wamed structure of ASCu neccesitates a surface barrier to prevent Ag diffusmg into the ASCu grain boundaries. Ag diffusion lowers the ASCu strength by approximately 30-35%. Cu plating was found to be the most effective barrier. Compliant interlayers were used between the graphite and ASCu braze joints to reduce brazing stresses and tile spallation. The graphite-ASCu braze joints survived high heat flux exposures of 1 to 2 kW cm2 . ASCu-316L jomts were made by inertia welding. Excellent metallurgical welds were produced which withstood a 10.34 MPa hydrostatic pressure test. The welds were subsequently tested to failure and failed in the ASCu piece with strengths ranging from 420 to 480 MPa.
* This work performed at Sandia National Laboratories supported by the U. S. Department of Energy under contract number DE-AC04-76DPOO789.
TN
STRUCTURE AND PROPERTIES OF XD COPPER: R M Aikin Jr, Martin Marietta Laboratories, 1450 S. Rolling Rd., Baltimore, MD 21227 High-purity copper XDTN composites reinforced with up to 20 vol% of a submicron in-situ dispersion of TiB2, ZrB2, or zrC have been fabricated. The tensile properties of these materials have been evaluated from room temperature to 1400°F, and as a function of time at elevated temperature. The alloys show very good elevated temperature strength, including excellent strength retention following long thermal exposures and extremely good microstructural stability. These mechanical properties will be discussed with respect to the microstructure and microstructural stability of these alloys. (10:00 a.m.)
ELEVA1ED 1EMPERATURE MECHANICAL PROPERTIES OF GLIDCOP AL-15: F, A, Crossley, Aerojet Techsystems, 9935/2019, Sacramento, CA 95852-1699 Glidcop AL-15 is essentially pure copper dispersion strengthened by fine particles of alumina in the amount of 0.15 wt. percent An anneal of 9500C for 1/4 h was applied to 22-mm diameter bar to simulate a braze cycle. This material was evaluated in tensile test properties and in strain-controlled, low-cycle-fatigue over the temperature range from 25 to 7320 C. Creep was evaluated at 538OC. Tensile reduction in area was found to decrease with increasing temperature. Low-cycle-fatigue performance was found to decrease in parallel with reduction in area. This behavior appeared to be due to void formation around the noncoherent alumina particles, and their growth rate which increased with increasing temperature. (10:30 a.m.) LO'lG TERM STABIUTY OF TWO DISPERSIO'l STRENGTHENED COPPER ALLOYS AT ELEVATED TEMPERATURES', J J Stephens and C.R. Hills, Sandia National Laboratories, Albuquerque, New Mexico 8718S. We have examined the long term microstructural stability of two dispersion strengthened copper alloys at elevated temperatures: an alumina dispersion strengthened alloy with 0.7 vol% dispersoid (Glidcop AI-1S) and a recently developed alloy reinforced with S vol% titanium diboride (MXT-S). Glidcop AI-15 has been found to possess excellent long term stability (>2000 hrs.) at 850°C without recrystallization, but long term anneals at higher temperatures can lead to secondary recrystallization and abnormal grain growth. This appears to be coincident with a change in dispersoid type from y-AJ203 to 9AI20:3·2B20:3. By comparison, softening in MXT-5 appears to proceed by means of normal grain growth processes. TransmiSSion electron microscopy resuhs on the chemical stabil~y of the TiB2 dispersoids in MXT-5 will also be discussed. 'This work performed at Sandia National Laboratories supported by the U.S. Dept. of Energy under contract number DE-AC04-76DP00789. (11 :00 a.m.)
GLIDCOtro DISPERSION STRENGTHENED COPPER: fROPERTIES AS A FUNCTION OF PROCESSING HISTORY AND SERVICE TEMPERATURE: P. K. Sarnal and J. D. Troxell, SCM Metal Products Inc., Cleveland, Ohio.
GlidCOP~ Dispersion Strengthened Coppers, made via powder metallurgy, comprise of a pure copper matrix and a fine dispersion of sub-microscopic particles of aluminum oxide. These materials possess superior mechanical strength, high electrical and thermal conductivities and an excellent resistance to softening ~~en exposed to elevated temperatures. A wide variety of GlidCop forms and section sizes have been successfully produced and have found applications in electronics, aerospace, and a variety of other high performance applications. The thermo-mechanical processing history has significant effects on the properties of the final product. The mechanical, ~Jbermal and electrical properties of various grades of GlidCop over a wide range of temperatures will be presented and the effects of processing history on these properties will be discussed.
mGH TEMPERTURE MATERIALS I: Fracture Mechanics of High Temperature Materials Sponsored by the TMS High Temperature Alloys Committee Tuesday AM February 20, 1990
Room: Plaza B Inn at the Park
Session Chairmen: Richard G. Menzies, GE Aircraft Engines, Mail Drop M85, Cincinnati, OR 45215; Michael V. Nathal, NASA, Lewis Research Center, MS 49-3, 21000 Brookpark Road, Cleveland, OR 44135 ( 8:30 a.m.)
FRACTURE MECHANICS OF SUPERALLOYS FOR TURBINE ENGINES, AN OVERVIEW, B.A. Cowles, Pratt & Whitney, P.O. Box 109600, West Palm Beach, FL 33410-9600 Advances in fracture mechanics understanding for superalloys have made possible the application of damage tolerance design approaches for gas turbine engines and liquid-fuel turbopumps. Integration of fracture mechanics with non-destructive evaluation has resulted in successful implementation of both the Air Force Engine Structural Integrity Program (ENSIP), and the engine component Retirement-far-Cause Maintenance program for the FIOO engine, with many benefits. Selected aspects regarding fracture mechanics are presented. Advanced engine applications provide new development challenges for super alloys and fracture mechanics. Key technical challenges remain, in applications of fracture mechanics, in alloy and process development to improve fatigue and fracture properties, and for assessing fracture behavior in hostile environments for advanced rocket and hypersonic propulsion systems. Some areas for future research are suggested. ( 9:00 a.m.) FATIGUE CRACI PllFASlTItw OF Ii,ll IS A FUlCTItw OF C!B.'IJSlTItw AID !IiII.'I!RUIIII
Graham Webb. Stephen D. Antolovich, School of Materials Engineering, Georgia Institute of Technology, Mechanical Properties Research Laboratory, Atlanta, GA., 30332·0245, Randy Bowman, NASA·Levis Research Center, Cleveland, OH Four B-modified Ni3Al alloys were studied. as a function of c:c::ap::lSition and temperature. AI and Cr content were varied to change the stoichicmetry and. microstructure. At 25°C, it was found that the lowest Al bearing alloy exhihited the slowest FCP rate. This rate was cc:mparable to the lowest rate observed for 1IIJde1 superalloys. At 600°C, it was found that the alloy having the least Al +Cr bad the best FCP properties. The results were interpreted in terms of fundamental deformation mechanisms, microstructure, and envi..rotxreot. In addition, crack closure was measured at 25°C and was found to have little effect on FCP. Two alloys were bina.ry Hi and II and two were ternary Ni, II and Cr.
Authors are grateful to the AFOSR 87-0162, Dr. A. Rosenstein for support of this project.
(9:25 a.m.) PREDICTION OF CRACK GROWTH BEHAVIOR IN ADVANCED AIRCRAFT ENGINE MATERIALS: R.H. Vanstone, K.S. Kim, and M.D. Sensmeier, GE Aircraft EngInes, Mall Drop A333, Cincinnati, OH 45215 (Abstract Not Available)
Annual Meeting-53 ( 9:50 a.m.)
FLOW, FATIGUE AND FRACTURE OF SINGLE CRYSTAL NICKEL BASE SUPERALLOYS: W S Walston, M. Dollart , I.M. Bernstein!, and A.W. Thompson, Carnegie Mellon University, Pittsburgh, PA 15213, t Illinois Institute of Technology, Chicago, IL 60616. The last decade has seen an enormous effort in research concerning the behavior of single crystal superalloys, and today many of the unique aspects of these alloys are well understood. An overview will be presented which covers the current understanding of the flow, fatigue and fracture of single crystal nickel base superalloys. Emphasis will be placed on two of the most studied of this class of alloys, CMSX-2 and PWA 1480, while contrast and similarities between these alloys and others will be given. The flow characteristics will be covered as a function of temperature, and the development of the deformation structure up to fracture will be discussed. The results of low cycle fatigue tests at different strain amplitudes will be presented along with the corresponding dislocation structures. The well-known crystallographic fracture of these alloys will be discussed in-depth with special emphasis on the relationship between the microstructure and fracture. Despite the many similarities in the behavior of the various single crystal superalloys, there are enough differences and variables to puzzle researchers into the next decade. This work was sponsored by NASA Lewis Research Center under the direction of Dr. R.L. Dreschfield (10:15 a.m.)
INFLUENCE OF MICROSTRUCTURE ON CRACK GROWTH BEHAVIOUR OF PM ASTROLOY AT HIGH TEMPERATURE J.-C. LAUTRIDOU, C. DUCROCQ and J.-Y. GUEDOU SNECMA, Materlals and Processes Laboratory- B.P. 81 91003 EVRY CEDEX (FRANCE) Crack propagation tests on short and long cracks specimens have been performed in air at 600 and 650°C on'PM Astroloy, on which various thermal mechanical treatments have been applied (HIP + conventional forging, extruding + isothermal forginq) in order to produce different microstructures: fine grain, necklace. dUDlex.Several fatigue waveform have been used, inclOd'irrg dwe11 time at maximum load. It appears that the microstructure of the alloy has a limited influence on crack propagation at 600°C. On the opposite, a strong effect of grain size is evidenced at 650°C ~·10reover, the waveform may alter the crack propagation rate in a ratio about 5 or more, depending upon dwell time. Main axis for an optimization of the microstructure in view of fati oue crack orooaoilti on resi st~ncp. ilrethen di scussed and the applicability in industrial way of such procedures is considered. (10: 40 a.m.)
THE EFFECT OF ENVIRONMENT, MICROSTRUCTURE, AND HOLD TIME ON THE FATIGUE CRACK GROWTH BEHAVIOR OF RENE'95 AT 649°C: ~,DD Krueger, and RH VanStone, GE Aircraft Engines, Cincinnati, OH 45215; N Jayaraman, Univ. of Cincinnati, Dept. of MatI. Sci. and Eng., Cincinnati, OH 45221. The fatigue crack growth behavior of PM Rene'95 was investigated 10 determine the influence of age treatment, environment, and hold time at 649°C. Material was solutioned above the y solvus and subjected to either a standard 760°C/8hr age (SA) as a baseline condition or a 982°C/8hr age (OA) 10 alter the y characteristics. The OA treatment coarsened the y precipitates and decreased both creep resistance and tensile strength compared to the SA treatment. Crack growth tests were conducted at 10 Hz and with a 300 second hold time in both air and vacuum. Crack growth rates were always lower in vacuum than in air for both age treatments and frequencies. In comparison to the SA material, the OA material exhibited accelerated hold time crack growth in air but decelerated hold time crack growth in vacuum. LCF and creep tests were conducted to study fatigue and creep damage via TEM. Creep deformation consisted primarily of dissociated dislocations while fatigue deformation consisted primarily of dislocation tangles concentrated in 'bands.
Hold time fatigue
deformation consisted of both. In general, the dissociated dislocation deformation mode was associated with intergranular failure. Deformation characteristics were correlated 10 the FCG behavior observed. (11: 05 a.m.)
TIME DEPENDENT MECHANISMS OF FATIGUE CRACK PROPAGATION IN IDGH-STRENGTH SUPERALLOYS:K-M. Chang, GE Corporate Research and Development Center, P.O. Box 8, Schenectady, NY
12301
Tinle-dependence of fatigue crack propagation in high-strength superalloys, has been investigated in air and in vacuum by employing alloy 718 as a model alloy. The accelerated crack growth rate induced by a hold tinle at the maxinlum load of each fatigue cycle can be attributed to two mechanisms: the "internal" mechanism of creep cracking and the "external" mechanism of environmental embrittlement. The former contributes to the tinle-dependent behavior of cracking in vacuum, and the latter dominates the crack growth in air. With the aid of a modern crack monitoring technique, the distinction between two mechanisms can be clearly attained. The difference between the two tinle-dependent mechanisms has been observed in the crack growth characteristics within one loading cycle and in the existence of a damaged zone in front of the crack tip, as well as the required conditions to start tinle dependent behavior and the fracture mode of tinle dependent cracking. (11: 30 a.m.)
MICROSTRUCTURAL INFLUENCE ON NEAR THRESHOLD FATIGUE CRACK GROWTH BEHAVIOR OF A NICKEL BASED SUPERALLOY: ~ K putatunda, Andrzej Spis, and J. Raghavan. Wayne State University, Detroit, MI 48202 Fatigue crack growth behavior of a nickel based superalloy (Incoloy 825) in four different heat treated conditions has been studied
in air in the threshold region. The Investigation also examined the influence of grain size and material strength on fatigue threshold and near threshold crack growth rate. The Influence of load ratio (R) on fatigue threshold was also examined. The heat treatment studies revealed a peak hardness condition at 434 hours of aging at 750 0 C. The near threshold crack growth rate in this condition was lower than In any other condition. This resulted in an increase in fatigue threshold with simultaneous increase in material strength. The fatigue threshold was found to increase with increase in grain size of the material in all load ratios. This increase In fatigue threshold can be attributed to the increased roughness induced crack closure in these materials.
mGH TEMPERATURE SUPERCONDUCTING COMPOUNDS I: Plenary Session Sponsored by the TMS Supervconducting Materials Committee Tuesday AM February 20, 1990
Room: Orangewood I Quality Hotel
Session Chairman: Amit DasGupta, Department of Energy, Chicago Operations Office, 9800 South Cass Ave., Argonne, IL 60439; Sung H. Whang, Polytechnic University, 333 Jay Street, Brooklyn, NY 11201 (8:30 a.m.)
( 8:35 a.m.)
OPENING ADDRESS
KEYNOTE LECTURE
CRITICAL CURRENT DENSITY OF 111<;11 TEMPERATURE SUPERCONDUCTORS, A. P. M~loz~2!1 IBM Research, Yorktown IIeights, NY IOS9S-02IS USA Progress in processing high temperature superconductors for high critical current densities is reviewed. Values of order 5x10' II/em' at 77 K have been reported in YilaCuO films with comparable values in other materials systems. Ilowever the concept of critical current density has heen questioned hecause pinning barriers to nux motion arc often small in these materials, leading to ohmic behavior at very low voltages. Fortunately, there is some experimental evidence that co-operative effect.s between nux lines can create a "vortex glass" state which restores a true critical current below some field-dependent "glass" transition temperature. Progress in increasing pinning barriers and optimizing the vortex glass state by processing will be summarized.
( 9:20 a.m.) PLENARY SESSION FABRICATION, CHARACTERIZATION ANO APPLICATION OF HIGH CURRENT DENSITY, HIGH Tc SUPERCONDUCTING THIN FILMS: R.B. Laibowitz. P.O. Box 218. Yorktown Heights. NY 10598 High Te superconducting thin fi lms have been fabricated using several techniques (electron beams. sputterin9 and laser ablation). While the highest current dnsity films (V-based) are generally expitxial and grown on single crystal SrTi 0 substrates, compari sons with other substrate materials wi 11 3 be shown. In addition to studying critical currents and fields. we have also measured Raman and microwave properties which will be discussed. Lithographic patterning of these films has produced reproducible structures with linewidths as small as 2 1»11. Patterns containing dc SQUIDs and multilayer junction structure have been developed using these techniques. These lithography studies have been carried out on both the V-based and TI-based materials. Recent work on much finer linewidths will also be discussed.
( 9:50 a.m.) MICROSTRUCTURE DEPENDENCE OF CRITICAL CURRENT DENSITY IN Ag SHEATHED HIGH T OXIDE TAPES: Kozo Osamura, Department of Metallurgy, Kyoto UniveFsity, Sakyo-ku, Kyoto 606, Japan The effect of heat treatment and cold working on the critical current density (J ) and the resultant microstructure change have been investigated For the Ag sheathed Y-Ba-Cu-O and Bi(Pb)-Sr-Cao tapes. The J of Y-tapes increased with increasing degree of cold-working ana roughly depended on the absolute tape thickness. PreSSing ~as more effective for improving J. The specimen with 3300 A/cm at 77K consisted of textured gratn structure with low porosity. In the Bi(Pb)-tapes, a most effective technique was the repeat of pressing and heat treatment, and the selection of 2 the chemical composition. J increase abruptly with decreasing tape thick2ess below a critiga1 thickness. The specimen with 10800 A/cm at 77K consisted of two major superconducting phases and three minor ones. The pole figure indicated significantly the (001) preferred oriented sturucture parallel to the tape surface. The microstructural factors for optimizing J have been campared for both oxide systems. C (10:20 a.m.)
BREAK
54-Annual Meeting (10,35 a.m.)
(8:50 a.m.)
Chemical and Physical Properties of Bi-Sr-Ca-Cu-O Superconductors: M. Schwarz, J. Bock, 8. Hettich, M. Peuckert, 1toechst 1\G, P.O.
TANJ(I!OOSE MECilANIZATION:
Box 80 03 20, 6230 Frankfurt 80, F.R.G.
2846 Bankhead Highway, Austell, GA 30001
Pure Bi-Sr-Ca-Cu-O superconductors have been prepared with single, double
~nd
triple Cu02-1ayers and corresponding Tc-values of 75K,
85K and 110K. While the single and triple layer compounds may be prepared by reacting solid mixtures of the metal oxides a process starting with a stoichiometric oxide melt was found to be advantageous for the double layer compound. Additionally to X-ray diffraction, HR-TEM investigations and microstructure analysis we have characterised all of the three compounds by XPS-spectroscopy which surprisingly exhibits a significant change of the electronic state for all of the four containing metals 8i, Sr, Ca and eu by going from superconducting to not superconducting samples.
(11:05 a.m.)
STRUCTURE/PROPERTY RELATIONSHIPS IN HIGH T COPPER OXIDES CONTAINING THALLIll1 AND BISMUTH: M.A.cSubramanian Central Research and Development Department, E.I. du Pont de Nemours and Company, Experimental station, Wilmington, DE 19880-0262 A brief survey of composition and structures of hightemperature copper oxide superconductors containing thallium, bismuth and lead will be presented. An interesting common structural feature is that the part of the structure between nonconsecutive Cu0 sheets is ill-defined. Possible correlations of T2 with the structure, bonding and charge carrier concentration will be discussed. Nonstoichiometry and the ~~ist~~ce of extended defects in connection with the Cu /Cu ratio (hole concentration) in these cuprates will also be discussed
AUTC»1ATIC GRAB SYSTEMS
Tam Marttila and Ronald L. Pariani - Wenmec Systems, Inc. During the last few years, when a new tankhouse or a major modernization has been planned, automatic or semiautanatic crane and grab systems have at least been considered. It seems that in the future tankhouse managers for new, modernized and even old tankhouses will be looking for a proven and reliable system for automatically moving and
positioning electrode loads in the tankhouse and in storage areas.
Our paper will descr ibe the Wenmec
solution for automatic electrode handling.
The advantages
of automatic handling and very accurate electrode
positioning will be discussed, and the way Wenmec has solved the problems in mechanical construction and in automatic, accurate and reliable positioning will be presented.
(9: 10 a.m.) INCREASING MASS TRANSFER IN AQUEOUS SOLUTIONS BY USE OF MICROPOROUS MEMBRANES F. Tailoka and D. J. Fray, Department of Materials Science and Metallurgy, University of Cambridge, Pembroke Street,
Cambridge CB2 3QZ
In order to increase mass transfer in aqueous electrowinning cells, gas sparging has frequently been used and, usually, the
bubbles are created from gas spargers consisting of holes drilled into a tube, held beneatb tbe electrode. This work reports tbe use of microporous membranes to create a very fine dispersion of bubbles which can greatly enhance the mass transfer of electroactive species to the electrode. Using this system, it has been
shown tbat it is possible to electrowin copper as a coherent planar deposit, at high current density, from copper chloride solutions.
(9:30 a.m.)
(11:35 a.m.) RECENT WORK ON HIGH T SUPERCONDUCTING OXIDES AT THE DEPT. OF INORGANIC CHEMISTRY MOSCOW STATE UNIVERSITY, Y.D. Tretyakov,
or
Moscow State University, Leninskie-Gory, Moscow V-234, USSR.
(Abstract Not Available)
ELECTROLYTIC REMOVAL OF MERCURY FROM Au-Hg-CN ELECTROLYTES, by Jae-Ho Lee and G. P. Martins, Department of Metallurgical and
Materials
Engineering,
Colorado
Colorado, 80401.
School
of
Mines,
Golden,
Mercury, present in the leach solution of some ClP circuits, poses a health hazard during smelting of steel wool cathodes
on which the gold (and mercury) are electrodeposited from the
HYDROMETALLURGYANDAQUEOUS PROCESSING m: Electrolytic and Separations Processes Sponsored by the TMSlEPD Aqueous Processing Committee
strip solution. The research conducted has examined the selective removal of mercury from synthetic Au-Hg-Na-CN "strip" electrolyte using carbon, glassy carbon and copper electrode-substrates. On the basis of cyclic voltammetry conducted with gold, mercury, and mixed gold/mercury cyanide electrolytes, a mercury-amalgamated copper-electrode offers a viable electrochemical technique for selective removal of mercury. This inherent feature has been incorporated in the design of a packed-bed (high surface area) copper/mercury amalgam electrode electrolytic-cell for effecting this purification scheme. J
(9,50 a.m.) Break
TuesdayAM February 20, 1990
Room: Salon 1 Marriott
Session Chairmen: James E. Hoffman" Jan H. Riemers & Associates, 11531 Meadowlake Drive, Houston, TX 77242; Rolf Wesely, Davy McKee Corporation, 2303 Camino Ramon, P.O. Box 5500, San Ramon, CA 94583
(10: 10 a.m.) MIXED SOLVENT EXTRACTION SYSTEMS FOR F-ELEMENT SEPARATION AND PURIFICATION: James D. Navratil Department of Mineral Processing
and
Extractive
Metallurgy,
The
University
South Wales, P.O. Box 1, Kensington NSW 2033 Australia.
of
New
Neutral bifunctional organophosphorus solvent extractants (such
as
carbamolymethylphosphonates
and
carbamoylmethylphosphine
oxides) strongly extract lanthanides and actinides and do not extract more than 40 other elements from mineral acids. Because
of their excellent properties they have been found useful for
(8:30 a.m.)
OPPORTUNITIES IN ELECfROLYTIC PROCESSING, Dr. c.L. Kusik, Arthur D. Little, Inc., 20 Acorn Park, Cambridge, MA, 02140, Joseph Goodwill, Center for Metals Production, 4400 Fifth Avenue, Pittsburgh, PA, 15231 In an effort funded by the Center for Metals Production (EPRI), work was undertaken to (1) evaluate technologies with respect to the recovery of selected metals from both primary and secondary resources as well as waste resources, and (2) identify opportunities that might permit consideration for joint development by industry and EPRI. The technologies of interest are classified as "electrolytic technologies" including electrorefining and electrowinning. In the course of this effort, more than 40 metal commodities were screened and reviewed. Seven metal commodities selected for more indepth analysis were copper, iron/steel, lead, magnesium, sodium, titanium, and zinc. Sectors considered in this analysis included extraction from ore ("primary industry") and recovery from scrap metal ("secondary industry") as well as waste products. For these seven metals, market, industry, and current production overviews were developed followed by the identification and evaluation of over 60 potential electrolytic processes in nine research categories.
f-element recovery and purification processes. A brief overview of the lanthanide and actinide extraction chemistry of these reagents will be given. A discussion of recent results with mixing these reagents with monofunctional extractants and using the mixtures for rare earth recovery and separation will also be
presented.
(10:30 a.m.) EXTRACTION OF ZINC FROM CHLORIDE SOLUTION WITH DBBP: H K Lin, G.C. Fang and P.D. Rao Mineral Industry Research Laboratory, University of Alaska Fairbanks, Fairbanks, Alaska 99775-1180 Extraction of zinc from chloride solution with dibutyl butyl phosphonate (DBBP) in ESCAID 110 has been investigated. Distribution coefficient of zinc was found to be independent of equilibrium pH. This independence indicates that the mechanism of the extraction may be solvation. Slope analysis for the system of dilute zinc chloride solution reveals that two moles of DBBP react with one mole of zinc. A series of tests were conducted to bring DBBP to its full loading capacity and the analYSis of the results confirms this stoichiometric factor of 2. Extraction isotherms were established by either changing AIO
Annual Meeting-55 ratio or by changing concentration of zinc in tne aqueous feed and both curves are found to be identical. (10:50 a.m.) RARE EARTH RECOVERY USING PSEUDO LIQUID MEMBRANES: Harrying Cheng and James D. Navratil, Processing and Extractive Metallurgy,
Department of Mineral The University of New
South Wales, P.O. Box I, Kensington 2033 NSW, Australia.
Al though liquid membranes have several advantages over conventional liquid-liquid extraction, problems such as membrane instabili ty remain. In order to solve these problems, we have combined
an electrostatic
field with
the
principle
of
liquid
membranes and developed a novel reactor system for rare earth recovery. Our cylindrical reactor, filled with extractant and diluent, is divided by means of a perforated cylinder into an outer annular extraction cell and an inner stripping cell. Below the extraction and stripping cells are two settlers for the aqueous solutions. The design of the reactor permits the organic solution to flow freely through the special perforated cylinder and the added feed and strip solutions, outside and inside the perforation, respectively, to be completely separtated. We have been investigating aerosol and droplet dispersion by physical means as well as by electric field and ultrasonic methods. A summary of progress will be given.
(11: 10 a.m.) RARE EARTH SEPARATIONS USING AN IMPROVED ANNULAR CHROMATOGRAPH: Anthony Murphy, James D. Navratil and Delai Sun, Department of Mineral Processing and Extractive Metallurgy, The University of
New South Wales, P.O. Box I, Kensington 2033 NSW, Australia.
Advances in hydrometallurgy require the introduction of new separation and purification processes, but they can be complex, costly and energy intensive. Chromatographic processes have low energy requirements and possess very high separation efficiencies. However, they have not yet made the great industrial impact that might be expected from their benefits. The large-scale and difficult separation of rare earths would greatly benefit from an improved continuous chromatographic operational device. Truly continuous, multicomponent chromatographic separations may be achieved by means of the continuous annular chromatograph. Work has been initiated at The University of New South Wales involving the improvement and development of such equipment for the separation and purification of rare earths.
INTERFACES IN METAL-CERAMIC COMPOSITES IT: Theoretical Modeling of Interface Properties Sponsored by Process Fundamentals Committee and Composite Committee Tuesday AM February 20, 1990
Room: Plaza D Inn at the Park
Session Chainnen: RJ_ Arsenault, University of Maryland, College Park, MD 20742-1572; P. Jena, Virginia Commonwealth University, Richmond, VA 23284 (8:30
a.m.)
ELECTRONIC STRUCTURE AND ENERGETICS OF INTERFACES IN CERAMIC COMPOSITES, W.R.L. Lambrecht and B. Segall, Department of Physics, Case Western Reserve University, Cleveland, OU 44106. When two radically different crystal structures come into contact at an interface, a "frustration" of bond hybridisation may result, even when their lattice constants are matching. This is shown to be the case for interfaces between TiC, which is octahedrally coordinated, and SiC which is terahedrally coordinated. Although the local densities of states show evidence for both TiC and SiC bonds across the interfaces, the net adhesion is found to be very weak for abrupt interfaces between these materials. The presence of interface layers of a different chemical composition for which there is some experimental evidence and/or significant atomic relaxations may change the adhesion energy and requires further study. (9: 00
a.m.)
Electronic and structural properties of metal matrix composites.
P. Jena, Physics Dept., Virginia Commonwealth Univ., Richmond, VA 23284-2000 and R. J. Arsenault. Engineering Materials Group, University of Maryland, College Park, MD 20742.
The present theoretical understanding of the nature of bonding at the SiC/AI interface would be reviewed. Our discussion would concentrate on the electronic structure at the interface probed through charge transfers between atoms and the electron distri-
bution. The structural composition of the interface obtained from energy minimization calculations would also be presented. The results will be analyzed with respect to available experimental data. (9: 30
a.m.)
FINITE ELEMENT MODELING OF FRICTIONALLY RESTRAINED COMPOSITE INTERFACES, Roberto Ballarini and Shamim Ahmed, Department of Civil Engineering, Case Western Reserve University, Cleveland, OU 44106.
This paper presents the results of several finite element simulations of fiber push-out and pull-out tests. The analyses rely on a special finite element whose response is governed by a Mo hr-Coulomb yield criterion. Various interfacial characteristics, such as the distribution of stresses along the fiber-matrix interface, the extent of slip, load transfer from fiber to matrix, and the amount of fiber extraction or depression are studied for various specimen configurations as well as for several values of interfacial bonding and friction coefficients. The numerical technique is first checked by comparing the results of pull-out test simulations with those obtained using a singular integral equation approach. The model is then used to simulate push-through tests on a AVCO SCS-6 fiber-reinforced reaction bonded Si 3 N. (RBSN) composite. (10:00 a.m.)
BREAK
(10: 15 a.m.)
COMPUTER SIMULAnON OF ELASTIC PROPERTIES OF LAYERED MAlERIALS*. D. Wolf, Argonne National Laboratory, Argonne, IL 60439 The anomalous elastic behavior of layered materials is investigated by atomistic simulations. The systems considered include superJattices of grain boundaries in fcc metals and in silicon as well as strained-layer superlattices fonned between different fcc metals. The observed elastic behavior, including the supermodulus effect, is discussed in terms of the structural disorder due 10 the interfaces and the dimensional changes in these superJattices as function of the modulation wavelength. *This work was jointly supported by the U.S. Department of Energy, BES Materials Sciences, under Contract W-31-109-ENG-38 and by the Office of Naval Research, under Contract No. NOOOI4-88-F-OOI9. (10: 45 a.m.)
INTERMETALLIC PUASE FORMATION AND BREAKDOWN OF Mo DIFFUSION BARRIERS IN Ni-Mo-Cu DIfFUSION TRIADS, Yinsheng Shueh l , John P. Hirth2 and Robert A. Rapp", Dissociation and recession of the compound layers in the Ni-Mo-Cudiffusion triads initiated when the Mo layer was nearly consumed. The product phases of the dissociation reaction are consistent with those predicted from the ternary phase diagram. Numerical methods based on a finite difference technique, and an analytical solution based on diffusion-controlled growth and quasi-steady-state approximation were used to analyze the results. Both solutions correlated well with experimental results. '-TN-;w Materials R&D, China Steel Corp., Kaohsiung 81233, Taiwan. 2 Mechanical and Materials Engineering Dept., Washington State University, Pullman, WA 99164 3 Materials Science and Engineering Department, The Ohio State University, Columbus, OU 43210. (11:15 a.m.) ANALYSIS OF THE HIGH TEMPERATURE INELASTIC BEHAVIOR DURING PROCESSING OF FIBER REINFORCED METAL MATRIX COMPOSITES: Simon ~ ~ and D.C. Stouffer, Department of Aerospace Engineering, University of Cincinnati, Cincinnati, OH 45220
The residual stresses introduced during cool down in manufacturing processes have a significant influence on the interfacial strength of the metal matrix composites. This effect is even more complicated because inelastic flow can occur in the matrix material at elevated temperature. A unit cell model has been developed based upon the modification of the Navier equations of elasticity to include the inelasticity of the matrix material. This boundary value problem is then solved by a finite difference technique with an efficient matrix direct solver. A unified state variable constitutive equation is used to model the inelasticity of the material. Numerial results are obtained for the cooling process of a metal matrix composite from the annealing temperature. (11: 45 a.m.)
INTERFACE CHARACTERIZATION OF PLASMA-PROCESSED Ti-AI KETAL MATRIX COMPOSITES: W. N. Gungor and W. G. Burke, Westinghouse Science l Technology Center, 1310 Beulah Road, Pittsburgh, PA 15235 Recent efforts to produce intermetallic matrix composites have focused on the titanium aluminide (TiAI) system with various ceramic reinforcements. In order to optimize the interface structures of these composites, plasma processing approach was used. In the process, by melting one or more of the component
56-Annual Meeting materials in a plasma and controlling deposition onto a cooled substrate, a range of structures can be produced and the method is highly suited to examining experimental composite systems. In this work, premixed Ti3AI powders and various ceramic particulates were deposited on a water-cooled metal substrate. Metallography and AEM characterization were performed. The ceramic/matrix interface and matrix structures were correlated with the processing parameters.
1990 LEAD·ZINC AND RELATED METALS SYMPOSIUM VI: Lead Recycling
batteries. l\ctual plant q>eratin:J data indicates that the desulturized paste contains less than 0.6\ sulfUr and. less than O.st~. The ex system utilizes mechanical screenin;J and. an up-flaw hydrodynamic separator to acxx:uplish the cx:upanent separatiCl1. SOdiUlll hytIroxicle or sodiUlll cart>onate is used for desulfurizatian. '!be sodiUlll sulfate brine is evaporated and. crystallized to pro:Iuce clean anhyCrc>Js sodiUlll sulfate suitable for use by detergent manufacturers. Plant data also indicates that an increase in furnace productivity of aba.It 25 to 30 percent 18 achieved usinq feed fnn the ex system. 'lbe desulfuriztd past:a 18 very suitable for alecttolytic processin:J as an altemative to furnace processin:J. (10:50 a.m.) RSR'S FULL SCALE PLANT TO ELECTROWIN LEAD FROM BATTERY SCRAP:
R. D. Prengaman and H.
B.
McDonald,
Mockingbird Lane, Dallas, TX
Sponsored by the TMS-EPD Lead-Zinc-Tin Committee TuesdayAM February 20, 1990
Room: Salon A & B Marriott
Session Chairman: E.K. Hudson, Lake Engineering Inc., 6000 Forest Drive, Suite 350, Atlanta, GA 30328; D.J. Eby, RSR Corporation,UU West Mockingbird Lane, Dallas, TX 75247 (8:30 a.m.)
SECONDARY LEAD BLAST FURNACE SMELTING AT EAST PENN MANUFACURING CO., INC., Kenneth N. Pike, Deka Road, Lyon Station, PA 19536 The need to comply with environmental regulations and increase production has resulted in modificiations to the
East Penn blast furnace operation.
Changes made include
furnace design modifications, closed loop water cooling with heat recovery, automated weighing and charging, automated
tuyere punching, ammonia scrubbing of tail gases and improved metal and slag handling to minimize environmental impact.
These changes have resulted in a marked increase in yearly 'production and compliance with many of the air, water and
solid waste regulations.
(9:00 a.m.) ~EAD RECYC~ING UTI~IZING SHORT ROTARY FURNACES, H. Forrest and J.D. Wilson, Cookson Industrial lIaterials ~imi ted, Crescent House, Newcastle upon Tyne, NE99 IGE.
smelting at Cookson Industrial lIaterials was formerly carried out using a combination of Blast and Reverberatory Furnaces. In the late 70's these furnaces were replaced by two 20 tonne capacity Rotary Furnaces. This furnace type has proved to be more flexible in the range of feedstock that can be processed, including lead drosses, battery plate and even whole batteries. High rates of production have been achieved whilst attaining a good level of environmental control. The majority of the sulphur in the feedstock can be fixed by use of appropriate fluxes. ~ead
(9:30 a.m.) ~EAD RECYCLING AT THE HARTZ SlIEnER OF IIETALEUROP, G. Schenker, Hartz-lIetall GmbH, Rammelsberger, Strasse 2, 3380 Goslar, Federal Republic of Germany.
The secondary lead smelting plant of lIetaleurop SA in Oker has been engaged in smelting battery scrap since the early 1960's using long rotary kilns. In 1983 modern battery breaking equipment was installed. The equipment separates oxides fro. metallics, recovers an upgraded polypropylene for sale and separates PVC, ebonite, and separators. A modern effluent treatment plant neutralizes acid and cleans waste water. Oxides and sludges are smelted in the rotary kiln. lIetallics are smelted in campaigns. ~ow antimony bullion from oxides is refined to soft lead or Pb-Ca while metallic bullion is refined to grid alloys. The plant also processes ashes, flue dusts, and drosses. The plant plays an important part of the recycling strategy for lead in the Federal Republic of Germany. (10: 00 a.m.)
BREAK
(10:20 a.m.)
AD'lANaS IN I.EAI>oACID BM'l'ERY RECYc:t.n«;: ~'S 1ol1ltIIM!D ex I!REAI(ER SYSTEX: R.M. Fsvnolds. P,E. and. E.K. 1IIxlsal, Lake En;J~in:J, Inc., 6000 Lake Forrest Drive, SUite 350, Atlanta, Georgia 30328.
En;Jitec lDpianti S.p.A. of Milan, Italy has successfUlly designed, canstruc:ted and. has q>erated sta~f-the-art battery breaker systems at two EUrcpean locations. A third facility is partially c:x:mpJ.eted in Toronto, ontario. '!be ex system crushes IoIbole, undminad batteries, separates the vari<>JS cx:ap:>nents and. desulturizes the battery paste. '!be system also produces sodiUlll sulfate fnn the sulfUric acid and. lead sulfate cxntained in the
RSR
Corporation,
1111
w.
75247.
RSR has developed a process to produce high purity electrowon cathode lead from battery scrap. The process includes crushing the batteries I separating the lead bearing material from the case, HR and separators, and separating the paste fran the metallic lead fraction. The paste is treated with Na2C03 to convert the PbS0 4 to PbC0 3 • The paste is further treated to decompose the Pb02 to PbO. The lead from the treated paste is leached into H2SiF6 and electrowon from solution using Pb0 2 coated graphite anodes. RSR is in the final design stages of a plant to produce 36,000 tons per year of electrowon lead from battery paste in a new plant adjacent to the Indianapolis refinery. The paper describes the patented process in detail and compares present lead recycling plants with the new electrolytic plant. The new process is environmentally superior to the present pyranetallurgical process.
(11 :20 a.m.) THE DOE RUN COMPANY I S BUICK RESOURCE RECOVERY FACILITY, BOSS, MISSOURI: A. W. Worcester and J. A. Moenster, The Doe Run Company, Suite 400, 11885 Lackland Road, st. Louis, MO 63146-2436. The plant is designed to produce 60,000+ tons of refined lead per year from a nominal feed consisting of 110,000 tons of SLI and industrial batteries, and 10,000 tons of lead contained in drosses and scrap. The plant will also produce as by-products polypropylene chips and detergent grade sodium sulfate, using the Engitec Impianti CX Technoloqy. The battery separation circuit will consist of a battery breaking hammermill and a vibrating screen with water sprays to separate the battery paste from the grid metal and other battery components. The battery components, consisting of ebonite, and separator materials, will be separated from the grid metal and classified with the use of hydrodynamic water separators. The battery paste will be treated with sodiUm carbonate to produce a lead carbonate paste for smelting in a reverberatory furnace. The slag from the reverberatory furnace will be campaigned through a traditional primary blast furnace to produce hard lead. The sodium sulfate liquor from the lead carbonate filtration step will be crystallized to produce anhydrous sodium sulfate. The metallic grid metal will be treated through a small rotary melter of the Engitec design to produce a hard antimonial lead. The Plant is scheduled to be commissioned 12 months after receipt of all of the Federal and State permits. The plant will be integrated into the Buick Primary Lead Smelter, located in the center of Missouri's New Lead Belt, at Boss, Missouri. This will allow The Doe Run company to fully utilize all of its existing assets.
1990 LEAD-ZINC AND RELATED METALS SYMPOSIUM Vll: Zinc Pressure Leaching
Sponsored by the TMS-EPD Lead-Zinc-Tin Committee Tuesday AM February 20, 1990
Room: Salon C & D Marriott
Session Chairmen: C.A Sutherland, Cominco Ltd., Trail, British Columbia, Canada VIR 4L8; B.N. Doyle, Sherritt Gordon Mines Ltd., Fort Saskatchewan, Alberta, T8L 2P2,Canada (8:30 a.m.)
RECENT EXPERIENCE WITH ZINC PRESSURE LEACHING AT COIIINCO, D.W. Ashman and W.A. Jankola, Cominco Ltd., Trail, British Columbia, Canada, VIR 4L8. The zinc pressure leaching plant at Cominco's Trail Operations is now in its ninth year of production. The plant produces a zinc sulphate and jarosite slurry which is further processed in the calcine leaching circuit. Elemental sulphur, a byproduct of the process, is sold in the North American market. During the past eight years, improvements have been made to remove production bottlenecks and to improve leaching extraction. The plant currently accounts for approximately 20 percent of Trail's zinc production and will be increased further. Possible future expansion of the zinc pressure leaching plant would account for approximately 48 percent of Cominco's zinc production.
Annual Meeting-57 (9:00 a.m.) ZINC SULPHIDE PRESSURE LEACHING AT KIDD CREEK, A.C. Mollison and G.W. Moore, Falconbridge Limited, Kidd Creek Division, Bag 2002, Timmins, Ontario P4N 7Kl, Canada.
In November of 1983 a fire inside the autoclave caused the curtailment of operation of the Kidd Creek pressure leaching facility. Subsequent repairs and recommissioning of the autoclave in 1984 allowed Kidd Creek to increase its production of zinc as originally planned. This paper describes the operation of this pro~ess from 1984 to 1988 and details the repairs carried out in 1988 following a major mechanical failure in the vessel. (9:30 a.m.) THE ZINC PRESSURE LEACHING PROCESS, M.J. Collins, B.N. Doyle, E. Ozberk, and .I.M. Masters, Sherritt Gordon Limited, Fort Saskatchewan, Alberta T8L 2P2.
Direct pressure leaching of zinc sulphide concentrates was started commercially in October 1980. Continuing research and development studies at Sherritt and feasibility studies for a number of clients have defined various pressure leach configurations for integrations with existing roast-leach operations and for grass roots plants. Existing and potential applications of zinc pressure leaching are discussed. (10:00 a.m.)
BREAK
(10:20 a.m.)
THE KINETICS OF THE SHERRITT GORDON ZINC PRESSURE LEACH PROCESS D.B. Dreisinaer, E. Peters, M. Talaba, G. Owusu, R. Swiniarski, The University of British Columbia, 309-6350 Stores Road, Vancouver, B.C., V6T lW5, and K.B. DeGraaf, Bacon, Donaldson and Associates Ltd., 2036 Columbia Street, Vancouver, B.C. The zinc pressure leach process developed by Sherritt Gordon, Ltd. permits the direct leaching of sulphide concentrates according to the overall reaction, ZnS + H.SO. + 1/20. - ZnSO. + H.O + S· The kinetics of the leach are affected by three rate controlling steps; (1) oxygen mass transfer from the gas to the leach solution, (2) homogeneous oxidation of ferrious ion to ferric ion by dissolved oxygen and (3) ferric ion leaching of zinc sulphide in the presence of a sulphur dispersing detergent (lignin solphonic acid). Fundamental studies have been conducted to identify the importance of the factors that influence the kinetics of each individual rate controlling step. Ferrous oxidation by molecular oxygen and zinc sulphide leaching by ferric ion have been studied.
(10:50 a.m.)
INTEGRATION OF DIRECT LEACHING OF ZINC-BASED SULPHIDIC CONCENTRATES AND THE VIEILLE-KONTAGNE GOETHITE PROCESS, T. De Nys and F. Terwinghe, Vieille-Kontagne, Zinkstraat 1, B-2490 BalenWezel, Belgium. A direct leaching process of sulphidic zinc-based concentrates in aqueous sulphuric acid solutions has been developed by the V.K. R&D Department. By a very accurate control of the oxidation potential of the leach solution, this process enables a separate recovery of the zinc, copper, silver, lead and sulphur of the concentrate and makes it extremely attractive for treating more complex ores. The iron, solubilized during leaching, is eliminated out of the zinc solution by means of the V.M. goethite process. Alternative flow sheets for a conventional zinc electrolysis plant are presented.
(11 :20 a.m.)
DIRECT LEACHING OF ZINC CONCENTRATES AT ATKOSPHERIC PRESSURE, R.W. Adams, P. Mangano, E.G. Roche and S.J. Carpenter, Electrolytic Zinc Company of Australasia Ltd, Division of Pasminco Ketals, GPO Box 377D, Hobart, Tasmania Australia 7001.
A new process has been developed for leaching zinc sulphide concentrates at atmospheric pressure using air or oxygen plus nitrogen dioxide oxygen carriers. By controlled addition of concentrate a leach solution with low residual nitrate concentration can be produced. Following separation by filtration of unleached solids and elemental sulphur and conventional oxidation/precipitation of dissolved iron, the leach liquor can be treated with metallic zinc to remove the remaining nitrate and other metallic impurities such as copper, cadmium and cobalt to provide a solution suitable for electrolytic recovery of zinc. Pilot scale testing has given zinc recoveries in the high nineties, a high grade jarosite-free Pb/Ag residue and saleable elemental sulphur. Zinc concentrates containing substantial levels of lead and copper can be treated. The new process is cost competitive with pressure leaching.
MATERIALS DEGRADATION IN LOW EARTH ORBITm Sponsored by the Jt. TMS/ASM Corrosion & Environmental Effects Committee Tuesday AM February 20, 1990
Room: Parke Inn at the Park
Session Chainnen: H. Gary Pippin, Mechanical Systems Technology, Boeing Aerospace and Electronics, Seattle, WA 98124; V. Srinivasan, Universal Energy Systems, Inc., Dayton, OH 45432 (8:30 a.m.) ELLIPSO~lETRIC ANALYSIS OF t-1ATERIALS DEGRADATION IN SPACE: John A. Woollam, Paul G. Snyder, and Bhola N. De, Department of - Electrical Engineering, University of Nebraska, Lincoln, NE 68588-0511
Variable angle spectroscopic ellipsometry (VASE) is a rapid, sensitive, non-invasive optical technique performed at ambient pressures. Thus a vacuum is not needed, and materials properties changes can be monitored with a resolution of an atomic monolayer. In addition, surface roughness and material microstructure can be determined quantitatively. This talk will review the fundamentals of VASE, and give specific examples of applications to Indium Tin Oxide, Silicon Dioxide, Aluminum Oxide, and Tin Oxide coatings exposed to an atomic oxygen envi ronment. (9:00 a.m.)
VACUUM ULTRAVIOLET RADIATION AND THERMAL CYCLING EFFECTS ON ATOMIC OYXGEN PROTECTIVE PHOTOVOLTAIC ARRAY BLANKET MATERIALS: J. A. Bra~y, B. A. Banks, and S. K. Rutledge, NASA Lewis Research center, C eveland, OH 44135; T. J. Stueber and C. R. Stidham, Sverdrup Technology, Inc., Middleburg Hts., OH 44130; J. E. Merrow, Ohio University, Athens, OH 45701; R. Lebed, Michigan State University, E. Lansing, MI 48825. Ultraviolet radiation and thermal cycling can cause degradation of materials in the low earth orbital (LEO) environment. To assure long-term material durability, it is necessary to expose samples of candidate materials to simulated environments prior to their selection for space flight use. Two facilities have been constructed at NASA Lewis Research Center to evaluate the . durabil i ty of candi date space power system materi a1sunder conditions of steady-state vacuum UV radiation or combined vacuum UV exposure and thermal cycl ing. Characteristics of these facilities and the results of testing atomic oxygen protective photovoltaic array blanket materials will be presented. (9:30 a.m.) STUDIES OF THE INTERACTION OF 8 km/s OXYGEN ATOMS VITH SELECTED MATERIALS. George E. Caledonia and Robert H. Krech, Physical Sciences Inc., Research Park, P.O. Box 3100, Andover, MA 01810. The technique of laser-induced gas breakdown is used to generate a high flux pulsed beam of 8 km/s oxygen atoms. The beam contains approximately 10 18 oxygen atoms per pulse, can be operated at a PRF as high as 10 Hz, and be expanded to encompass an area >1000 cm 2 . The beam is used to simulate the interactions between the ambient atmosphere and spacecraft materials to be encountered in low earth orbit applications. The device and its operating characteristics will be described and example results
on material erosion, surface modification, and surface glow will
be presented.
(10:00 a.m.) CHARACTERIZATION AND CALIBRATION OF THE EIOM-III FLIGHT MASS SPECTROMETER IN A HIGH VELOCITY OXYGEN ATOM BEAM: S.L. Koontz, H. Ehlers, NASA Johnson Space Center, ES531, Houston, TX 77058; D. Hunton, The Geophysics Lab., Hanscom AFB, MA 01713; J. Cross, Los Alamos National Laboratories, CLS2, Los Alamos, NM 87545; E. Lan, McDonnell Douglas Space Systems Co., Huntington Beach, CA 92647. ABSTRACT NOT AVAILABLE (10: 30 a.m.) REACTION EFFICIENCY OF 5 eV OXYGEN IONS ON CARBON, Gary hl. Sjolander This paper will compare laboratory results of oxygen reaction chemsitry on carbon samples for oxygen energy of 5 eV. A brief description of our laboratory apparatus, status and direction,
58-Annual Meeting will be given, and a comparison with oxygen plasma asher device chemistry will also be given. The paper will include a discussion that compares oxygen ion chemistry with low earth orbit neutral chemistry. (11:00 a.m.) THE REACTION EFFICIENCY OF THERMAL ENERGY OXYGEN ATOMS WITH POLYMERIC MATERIALS: S.L. Koontz, NASA Johnson Space Center, Houston, TX 77058; Paul Nordine, 443 West 66 Terrace, Kansas City, MO 64113.
(10:30 a.m.)
RECENT ENVIRONMENTAL DEMANDS ON AUTOMOTIVE CONNECTORS. Y. Matsuda,Sumitomo Wiring Systems, Ltd., Yokkaichi, Japan Development of automotive electronics has increased the number
ABSTRACT NOT AVAILABLE. (11:30 a.m.)
the convection of corrosive gases to the gas/metal boundary layer; their diffusion through the concentration boundary layer; and the reaction rate of the corrodents with the metal surface.
&need for reliable connectors. The increasing number of con-
Discussion and Concluding Remarks: V. Srinivasan.
Bruce A. Banks and
MATERIALS ISSUES IN ELECTRONIC & OPrOELECTRONIC CONNECTORS m: Environmental Issues Sponsored by the TMS Non-Ferrous Metals Committee and the TMS Electronic Packaging and Interconnection Materials Committee Tuesday AM February 20, 1990
Room: California llIIIV Quality Hotel
Session Chainnan: D. Jeannotte, IBM Corporation, Hopewell Junction, NY 12533 (8:30 a.m.) AN OVERVIEW OF ENVIRONMENTAL ISSUES RELATED TO ELECTRONICS CORROSION: W. Abbot, Battelle Laboratories, Columbus, OH.
(Abstract Not Available) (9:00 a.m.)
ELECTROPLATED Ni/X (X. p. As. Sb. and Gel FILMS AS PRECIOUS METAL SUBSTITUTES IN CONTACTS. M. Robbins. J. T. Ple.es. and T. D. Schlabach. AT&T Bell Laboratories. Murray Hill. NJ 07974 Electroplated fil.s of the type Ni/X (where X : P. As. Sb. and
nectors is accompanied by shortage of space in engine compartment & instrument panel and increasing circuit pitch for PCBtype circuits. The need for low insertion force per terminal to maintain modest housing assembly force conflicts with contact force required for Signal or power &minimizing fretting when coupled to lOG accelerating vibration . Miniaturization imposes connector fabrication problems. Engine temperatures to 1600 C, increasing use of road salts, sensors, and high pressure car washing, create an increasingly hostile environment. These trends require connector substrate materials with better combined formability/spring properties/conductivity; resistance to stress relaxation. corrosion and stress corrosion; and compatibility with coatings. Coatings must withstand temperature and corrosive environment; not add to insertion force while limiting and with~tandiryg fretting; maintajn low contact resistance; & be compatlble wlth substrate materlals. (11:00 a.m.)
THE EFFECTS OF CRITICAL CORROSION TESTING VARIABLES ON CONNECTOR METALLURGIES, R. R. Gore &J. L. Chao. IBM Corporation, Research Triangle Park, North Carolina 27709 Studies using IBM's gaseous corrosion environmental test, called G1T, have identified some of the critical variables in accelerated tests used to predict the reliability of electrical connectors in business office environments. Systematic studies were performed by varying corrosive gas concentrations. air velocities. air exchange rates, and sampling conditions in a manner not previously reported. These data show that concentrations of gases such as chlorine and sulfur dioxide may be more critical than previously thought. The velocity and exchange rate of the gases also appear very important in generating uniform exposures to corrosive gases and should be used to provide adequate protection from loading and shadowing within the chambers. Data will be reported for copper. nickel. porous gold, and palladium-nickel coupons and selected connectors. Copper wire and thin film corrosion indicating devices will be described.
Ge) have been investigated as protective layers in electrical
contacts for precious metal replacement. The layers exhibiting the best contact resistance (Rc) stability are those where the concentration of X is in the Ni solid solution region. Stability with respect Rc .as tested in the laboratory ky heating sa~les (OFHC Cu strips coated with Hi/X) at 75 C and 951 RH for periods of 3-14 days and .easuring Rc under 5Oy.. Sulples where Rc <50~n .ere considered to be stable. NflX "fil.s coated with "'0.05II1II of Au .ere found to be stable when aged in atmospheres containing corrosive agents (e.g. HZS. CIZ). (9:30 a.m.) THE EFFECT OF TEMPERATURE AND HUMIDITY ON THE MACRO AND MICRO WEAR OF TIN-COATED COPPER ALLOYS AND THEIR IMPACT ON CONTACT RESISTANCE, ~ ~ and J. c. Fister, Olin Metals Research Laboratories,
06511.
91
Shelton
Ave.,
New
Haven,
CT
Tin coatings have been used successfully on copper alloys in many connector applications. One shortfall of these coatings preventing their
wider
utilization
has
been
in
their
wear
pro-
perties, manifested in a limited number of ,nsertion/withdrawal cycles and a tendency toward fretting. Experiments have been per-
formed, characterizing the wear of tin-coated copper alloys for macroscopic wipe distances (Simulating insertion/withdrawal) and micro-
scopic
wipe
~~
distances
(fretting
mE:::.t ur~oeafnli c\~~\d \,t 1 f: t ~
controlled
resistance
environments.
measurements
In
were
t
conditions)
in
i ~ ~i eadndfrco~ n::: ~
these
experiments,
performed
in
each
csse. The implication of using tin coatings in elevated temperature connector applications will be discussed. (10:00 a.m.)
Mass-Transport Limitations in Atmospheric Corrosion, P. J. Peterson &L. Volpe, IBM Corporation, San Jose, CA 1222 The corrosion rate of metals in environments containing gaseous corrodents in the sub-ppm concentration range can be strongly influenced by mass-transport limitations of the corrosive gas delivery to the surface of the metal. The mass transport constraint must be considered when studying atmospheric corrosion in the field as well as in the laboratory. Accelerated atmospheric corrosion tests that disregard these limitations can lead to erroneous conclusions. In this talk. theoretical mOdels and experimental data will show the interaction between:
METALI.IZATION FOR ELECTRONICS APPLICATIONS I: Thin Film Phenomena A . Basic Processes, Microstructure Development Sponsored by TMS Electronic and Photonic Device Materials and Electronic Materials Committees Tuesday AM February 20, 1990
Room: California I·n Quality Hotel
Session Chainnan: J.E. Sanchez, Jr., CAM Bldg. 66, Lawrence Berkeley Labs., Berkeley, CA 94720 (8:30 a.m.)
A THERMODYNAMIC ANALYSIS OF DIFFUSIVE STRESS-VOIDING, F. G. Yost, Sandia National Laboratories, Albuquerqup., NM 87185.
In recent years the silicon integrated circuit industry has experienced a new failure mechanism . The aluminum conductor lines in such circuits are prone to void formation and growth to such an extent as to cause circuit failure even before the circuit 1s exercised. These conductor lines are geometrically constrained by an overlying passivation layer of glass and are subjected to very large thermal stress. The voiding phenomenon is modeled as a stress-driven diffusive process in which each grain boundary acts as a reservoir for mass . The thermodynamics of stressed solids is used to derive an expression for the net chemical potential acting on an atom diffusing in an arbitrary grain boundary. The gradient of this potential is then used to calculate the flux of mass in the arbitrary and the average grain boundary. A field equation which preserves mass and predicts relaxation of thermal stress is derived and solved. The results show that void growth is sensitive to conductor line geometry, grain size and stress magnitude.
Annual Meeting-59 (9 :05 a.m.) CHARACfERlZATION OF ELEMENTAL DISTRIBUTIONS IN ALUMINUM THIN FILM METALLIZATIONS BY ANALYTICAL ELECTRON MICROSCOPY*, A D Romig. Jr., D. R. Frear and C. R. Hills, Sandia National Laboratories, Albuquerque, NM 87185
The texture is found to vary dramatically depending on the bias potential. High index {531} planes with a nearly epitaxial relationship, epitaxy , and fiber textures are found in copper/bare silicon. Deposition on oxidized silicon produces fiber textures in both aluminum and copper where texture sharpness and volume fraction of various components vary.
The distribution of alloying elements, especially Si and Cu, in aluminum thin film metallizations has a significant influence on the properties of the films, including resistance to failure by electromigration and stress voiding. Of particular interest is the distribution of alloying elements in the vicinity of precipitates and at grain boundaries. Analytical electron microscopy (AEM) is an ideal technique for characterizing thin film metallizations because it allows for simultaneous morphological, crystallographic and chemical characterization of the films. This paper will review the AEM procedures required to characterize thin films, including the composition profile deconvolution techniques required to obtain elemental distributions on a nm scale. Specific examples will be cited including recent evaluations of the formation of Si nodules in as-deposited Al - I wt. %Si and the formation of CuAl2 precipitates in as-deposited and isothermally aged Al- 2 wt.%Cu. The analytical observations will be rationalized with classical thermodynamic and kinetic models of precipitate growth ·Supported by the U. S. Department of Energy under contract number DE-AC04-76DPOO789.
( 11 :00 a. m.) CAPILLARY INSTABILITIES IN THIN FILMS ; ~ Bnd C.V. Thompson . Depa rtment of Materials Science a nd Engineering. M.I . T ., 77 Ma s s achusetts Av e ., Cambridge , MA 02138. r"" ' Very thin films, less than' lOOnm . thick, are used in a variety of applications, including as catalysts and for thin film reactions to form patterned silicides in electronic devices. Because of their h i gh surface to v o lume ratio , these v e r y thin films are subj e ct to capillary instabili ty and will agglomerate well below their melting temperatures . In orde r to de velop a gene ral understa nding of agglomeration i n very thin films , we have studied initially continuous and patterned films of gold on fused silica substrates . Two in situ techn i ques were us e d to monitor agglomeration : 1) heating and video recording in a transmission electron microscop e , and 2) measure ment of the intensity of laser light transmitted through a sample heated i n a furnace. Electron mic r oscopy allowe d inve stigation of t he role of the micros t ruct ure o f the Au f i lm Bnd analysis of light transmission during heat i ng allowed determ i nation of t empera ture-dependent and film-thickness-dependent a gglome ra t i on rates. The se results will be de scribed along with mode l s for the agglome ration process .
(9:40 a.m. ) 2D COMPUTER MODELLING OF POLYCRYSTALLINE FILM GROWTIl. A.J. Darnmers, S Radelaar Center for Submicron Technology, Delft University of Technology, PO Box 5046, 2600 GA Delft, the Netherlands Structural features and texture of vapour-deposited polycrystalline films may depend on a variety of parameters. These include substrate temperature, direction and magnitude of deposition flux, density and orientation of initial crystal nuclei, anisotropy of crystal growth rates and substrate topology (e.g. contact holes). Understanding the dependence of film properties such as grain size, preferred orientation of crystallites and surface roughness on these parameters is of utmost importance for controlled growth of thin films as applied in microelectronic devices. We initiated a computer simulation project, aimed at getting insight into the development of thin film structure in the temperature regime where surface diffusion allows formation of crystallites, but where (bulk) recrystallization and grain growth are absent. Essentially the (2D) model we study consists of growing and mutually impinging polygons. We successfully implemented the particularly cumbersome bookkeeping associated with this type of geometrical problem, where developments in the field of computer graphics proved to be valuable. First results, obtained recently, look quite promising. These will be presented, along with some technical aspects of the computational model. (10:00 a.m.) EVOLUTION OF MICROSTRUCTURE DURING ION BEAM ASISTED FILM GROWTH. C.M. Yang, H.A Atwater, Califomia Institute of Technology, Pasadena, CA 91125
Low energy ion bombardment during thin film deposition has been shown to significantly modify the microstructure of deposited polycrystalline thin films. However, little of a fundamental nature is known about the interaction of a low energy ion beam with a discontinuous film at the early stages of growth, and the resulting effect on microstructure. We have modeled the effects of low energy ion bombardment on post·nucleation growth and coarsening of islands in discontinuous films, using a procedure similar to the classical coarsening analysis of Lifshitz, Slyozov and Wagner (LSW). Solution of a continuity equation, subject to the constraint of mass conservation, coupled with interface-limited rate laws for island growth and coarsening allow various types of ion-surface interactions and island size distributions to be modeled. It has been speculated that several different phenomena may occur as a result of low energy ion bombardment during film growth, including enhanced ada tom diffusion, sputtering, dissociation of small islands and generation of new nucleation sites. Our analysis suggest that each of these phenomena leads to a unique kinetic path for coarsening and very different film microstructures Hence it may be possible, through a study of island coarsening kinetics to assess the relative importance of adatom diffusion, spunering, island dissociation, etc. in determining film microstructure
resulting from ion beam-assisted deposition. (10:20 a.m.)
A MODEL FOR THE EFFECTS OF ENERGETIC PARTICLE
BOMBARDMENT ON THE CR THIN FILM PROPERTIES: J.H. Hsjeh, R.F. Hochman, School of Matertials Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0245 Concurrent energetic particles (ions, atoms, etc.) bombardment on the growing film has been known to change the film properties. However, the bombardment effects can vary from case to case, depending on the deposition conditions. To explain the effects of the bombardment, a simple model is proposed. The model shows the change of the film properties may be dominated by particle momentum transfer or particle energy transfer or both. In the study, ion-plated Cr thin films are used as the example. The film properties examined are residual stress, microstrain, grain size and morphology. (1 0 : 40 a.m. ) PROCESSING EFFECTS ON TEXTURE IN THIN FILM METALLIZATION: DB Knorr, T.-M. Lu, Center for Integrated Electronics, Renassler Polytechnic Institute, Try, NY 12180-3590
Preferred crystallographic orientation is an ubiquitous consequence of the deposition of thin films . This study uses the partially ionized beam deposition technique where 0 to 5% of evaporated atoms are ionized by electron bombardment, then are accelerated by biassing the substrate. Nominally room temperature deposition of copper on both bare silicon and oxidized silicon and of aluminum on oxidized silicon are reported. Effective in-situ cleaning and increased ada tom mobility produce novel textures in the polycrystalline, equiaxed ftlms. The relationship between deposition conditions (ion content and bias potential) and texture are studied by x-ray diffraction pole figure analysis.
(11 :20 a.m.) COMPUTER SIMULATION OF TRANSPORT IN THIN FILMS : M.P. Anderson and S. Ling , Exxon Research and Engineering Company, Route 22 East, Annandale, NJ 06601 It is well established that the transport phenomena in materials are strongly affected by polycrystalline microstructure (e.g., grain size and morphology) and the relative conductance/diffusivity of the grains and grain boundaries. A two-dimensional computer model, in which these parameters are explicitly included, has been developed to describe the steady-state transport (both charge and mau) in thin films. In this approach, a thin film i s first created using the Monte Carlo microstructural dynami cs procedure, in which grains are described as collections of small area elements. The local resistivity/diffusivity is then assigned to each of the area elements, and the effective resistance/diffusion coefficient and steady state fluxes calculated employing an iterative averaging technique . Application of the methodology to realistic microstructures has been carried out, and the results will be discussed.
(11 :40 a.m.) r~TERIALS SCIENCE ISSUES IN PLANARIZATION TECHNOLOGY--AN OVERVIEW: Jeff Trogo10 and Krishna Rajan, Materials Engineering Department, Rensse1ear PolytechniC Institute, Troy, NY 12180-3590.
Chem-mechanica1 polishing planarization technologies are in use to prepare flat silicon wafers as well as a variety of components requiring flat surfaces in the optical industry . The application of such p1anarization technologies to flatten the topography of multilevel interconnections and insulators on processed silicon wafers is now being given serious consideration. However, the use of such polishing procedures demands a fundamental understanding of the behavior of thin films exposed to chemical attack and surface deformation . In this paper we review some of the materials science issues that must be considered in order to fully develop this type of p1anarization technology for multilevel metallization.
ThlS work is supported by the Semiconductor Research Corporation .
60-Annual Meeting
MODELING OF MICROSTRUCTURE EVOLUTION DURING SOLIDIFICATION ill: Micro! Macro Modeling Sponsored by the TMS Solidification Committee Tue.dayAM
February 20, 1990
Room: Plaza A Inn at the Park
Session Chairmen: D.M. Stefanescu, Solidification Laboratory, Department of Metallurgical and Materials Engineering, University of Alabama, Tuscaloosa, AL 35487-0202; C.G. Levi, Materials Department, University of California at Santa Barbara, Santa Barbara, CA 93106 (8:30 a.m.) PREDICTION OF COLUMNAR AND EOUIAXED MICROSTRUCTURES: ~ and Ph. Thltvoz, Ecole Poly technique F~derale de Lausanne, Laboratoire de M~tallurgie Physique, 34 ch. de Bellerive, CH·l007 Lausanne, Switzerland. During solidification of most technical alloys, two length scales can be identified: the first one is associated with heat and mass transport and is of the order of the size of the whole process. The second scale which is that of the microstructure is primarily dictated by solute diffusion and capillarity effects. Accordingly, modeling of columnar and equiaxed microstructure formation during solidification processes can be achieved by solving the continuity equations of heat. mass and momentum at the macroscopic scale. coupled with existing theories 01 growth kinetics and heterogeneous nucleation. Recent developments in this field lor both dendritic and eutectic alloys will be presented with emphasis placed on casting and laser remelting 01 metallic surfaces. Effects of grain orientation on microstructure selection will also be discussed. Although numerical methods employed in such modeling will not be detailed. micro·macroscopic coupling schemes will be briefly examined. Calculated cooling curves and microstructures will be compared with experimental results.
(10:45 a.m.) MODELLING OF MACROSEGREGATION IN DENDRITICALLY SOLIDIFIED ALLOYS: Itsuo Ohnaka, Dept. of Materials Science and Processing, Faculty of Engineering, Osaka University, Yamadaoka 21, Suitashi, Osaka 565, TAP AN
A numerical simulation method applicable to various kinds of macrosegregation and some examples are presented. Microsegregation, laminar and D' Arcy flow due to thermal and solutal convection, solidification contraction, and solid phase movement by bulging in the case of continuous casting are considered. Discrete equations for energy, mass and momemtum conservation have been derived directly from the physical phenomena by the direct finite difference method. Examples include inverse-, normal- and negative-segregation in ingots and center-line segregation in a continuously cast steel, demonstrating a good agreement with real macrosegregation. (11 : 15 a.m.)
(9:00 a.m.) COUPLED DENDRITIC/EUTECTIC EOUIAXED SOLIDIFICATION: PM Stefanescy and D. Bandyopadhvav. Dept. 01 Metallurgical/Materials Engineering. University of Alabama, Tuscaloosa. AL 35487·0202. Solidification Kinetics . Heat Transfer (SK· HT) modeling, also known as micro·macro modeling has made significant progress over the last few years. One dimensional SK models are now available for modeling of nucleation and growth of eutectic and dendritic equiaxed grains which allow fairly good calculation of fraction of solid generated during solidification. They are coupled with 3·0 heat transfer codes. Although these models can predict microstructural features as well as the path of the isotherms through the casting. they lack the accuracy required by industry because of inherent simplifying assumptions. This paper will discuss the issues which must be solved in order to transfer SK·HT modeling from the computer of the researcher to that of the shop engineer. They include: nucleation models also involving secondary nucleation induced by fluid flow, 3·0 models lor equiaxed dendritic growth. models for coupling dendritic and eutectic solidilication. and generation 01 a data base for SK modeling. Validation on a number of commercial alloys with regular and divorced eutectics will be demonstrated.
(9:30 a.m.) MODELUNG SOUDIFICATION OF CAST IRON : G R J Lesoy!!. Laboratoire de Genie Metallurgique. Ecole des Mines. Parc de Saurupt. 54042 Nancy Cedex. France. The aim of this paper is to review models presently available for simulating the formation of as·cast microstructures and the kinetics of solidification of cast irons. Special attention is paid to spheroidal graphite cast irons (S.G. irons). The know· how related to modelling of elementary phenomena like nucleation and growth of phases from iron melts is made out firstly. Then. examples of modelling either columnar of equiaxed solidification are presented. Concerning directional solidification of S.G. irons, a model is presented which accounts for the effect of the solidification rate on the onset of the eutectic reaction, on the temperature interval for Its completion and on the density of spheroids. Concerning equiaxed solidification of grey and S.G. irons. few models are successful in predicting the recatescence and the density of eutectic grains or graphite spheroids in different locations of simple shaped castings. However. in the case of S.G. iron. the granulometry of the spheriods is still badly simulated. Finally. developments in modelling the formation of microporosity during solidification of cast iron are examined. (10: 00 a.m.)
(10: 15 a.m.) MICRO- AND MACROSTRUCTURE EVOLUTION IN THE PRESENCE OF FLUID FLOW: David D. Goettsch, Long-Sun Chao and Jonathan A. Dantzig, University of Illinois, Dept. of Mechanical & Industrial Engineering, 1206 W. Green Street, Urbana, IL 61801. Grain size and other micostructural features such as second phase morphology and distribution are the primary factors in determining the mechanical properties in cast metals. For a given material, careful control of the temperature history during solidification can produce desirable microstructures, but the results can be strongly influenced by the presence of fluid flow in the melt during solidification. This presentation will cover recent developments in computer simulations to predict the microstructure and properties of cast irons in real casting processes. These processes necessarily include fluid flow from both filling transients and from natural convection, heat transfer and mass transport. Several examples will be given to demonstrate the important phenomena and the roles of the various process parameters in influencing microstructure evolution.
BREAK
CONSERVATION EQUATIONS FOR MODELLING t1ACROSEGREGATION IN DENDRITIC ALLOYS by D.R. Poirier and S. Ganesan. Department of Materials Science and Engineering. The University of Arizona, Tucson, AZ 85721
There have been many works on modelling macro segregation in dendritic alloys, for which the mushy zone is considered to be a porous medium. However, the form of the momentum equation has not been consistently represented. Therefo re, the momentum equation i s derived based on the method of "vol ume averaging". The solute conservation equation is written so that it accounts for the e ffect of solute diffusion in the local solid during
so lidification. Also, the energy equation is presented in a manner that properly accounts for enthalpy in terms of temperature and composition.
Finally, the results of "numerical
experiments" for permeability in high volume fractions of liquid, in which laboratory expe riments fail, are given so that permeability can be estimated throughout the entire mushy zone.
Annual Meeting-61
NOVEL PROCESSING METHODS AND PROPERTffiSOFD~ONDSANDD~ONDLnrn
THIN Fll..MS llI: Advanced Processing
Sponsored by the TMS-SMD and Electronic and Photonic Device Materials and Electronic Materials Committees and ASM and TMS-EMPMD Tuesday AM February 20, 1990
Room: Orangewood m Quality Hotel
Session Chairmen: R. Kossowsky, Pennsylvania State Univ., Engr. Mats. Dept., P.O. Box 30, State College, PA 16804; J.C. Hurt, US Army Research Office, Met. & Mats. Sci. Div., P.O. Box 12211, Research Triangle Park, NC 27709 (8:30 a.m.)
(Invited)
Microwave Plasma Assisted Oeposltlon Techniques for Diamond and Diamond·Like Coatings. Richard S POGt arid Lawrence P. Bourget, Applied Science and TechnoloQY, Inc. (ASTeX), 35 Cabot Road, Woburn, MA 01801 Microwave plasma assisted deposition has proven to produce excellent diamond and dlamond·like coatings (DLC). The diamond films are polycrystalline, grain size 10l1m to O.ll1m, with hardness and thermal conductivity about half of that of natural diamond. Dlamond·llke coatings, In contrast, are amorp~ous, hard carbon films with widely varying properties. In diamond deposition the plasma serves as a radical generator. while In DLC the plasma Ion bombardment plays a decisive role. We will present techniques for both dePOSition processes and clarify the differences In plasma conditions. We will examine applications of diamond coatings to windows and cutting tools, and of dlamond·like coatings to optical and magnetic disk applications. The scaling of these deposition techniques requires different approaches, a8 the pressure range and plasma conditions are very distinct. (Invited)
(9:00 a.m.) LOW-PRESSURE DIAMOND SYNTHESIS AND APPLICATIONS: K. Upadhya, GE Medical Systems, Milwaukee, WI 53219. The unique combination of optical, physical and electronic properties of diamond have been the main driving force for the development of new and Ie .. expensive methods for diamond synthesis. Potential appl1catlons for diamond Include reI lable Infrared Window coat Ings, high temperature semiconductors, high energy microwave ampl Hiers, powerful pulsed lasers, wear coating, and as hut sink materllls. Diamond ftlm Is hard, stiff, chemlcilly Inert and has • vory low coefficient of thermal expansion. Currently, there Ire severl' methods, such as chemical vapor deposition (CVD), Plasma assisted CVD, Hot ftlament,lASER enhanced CVD, and Ion bum enhanced deposition (lEBO) techniques which have been employed for diamond/or dhmond l1ke-carbon flll1 fabrication. The potenthl appltcatlon of diamond In space technology, consumer 'ppl hnces, and electronic materials wll1 require greater depOSition rate,methods for doping diamond during crystall lzatlon and better fllm RlOrphology. This hlk will provide, broad background on the crystallization process of diamond from the gas phase. Physical and chemicil Justlflcltion of diamond growth at low pressures will be presented from the diverse viewpoints of thermodynlmlcs, chemical kinetics, surflce phenomena, Plasma chemistry and nucleatIon and growth ",echlntsm(s). Also, the difficulties encountered during nucllltion on different substrates, adhesion of the diamond fllM to the substrate, fl1m morphology, and lattiCI defects of diamond crysUls wll1 be presented. The future erl of diamond blsed electronics and optiCS, tho hopes of achieving this Ind the dlfflcultl.s to be over come .. 111 be discussed.
(9:30 a.m.)
Comparative study of hot filament CVD of diamond on silicon partially illuminated by a 193nm ArF laser beam A.Rengan, G.Matera, M.Longo, A.R.Srivatsa, J.Krishnasawamy, and J.Narayan, Materials Science and Engineering, North Carolina State University, Raleigh, NC-27695. Hot filament(-2000 0 C) chemical vapor deposition of diamond on a (100) silicon substrate at -7500C partially illuminated by a 193 run ArF laser beam will be reported. Additionally the substrate is electrically biased with respect to the hot filament. The UV light from the laser in addition to removing the graphitic and other non-diamond phases is expected to enhance the biasing effect. Some preliminary results on this novel technique will be discussed using TEM, Raman, and SIMS techniques on substrates illuminated by the laser as compared with a substrate which did not receive UV illumination during deposition. #We acknowledge research support by the NSF under grant No.DMR8703621.
(9:50 a.m.) Morphological features of hot filament CVD Diamond
G.Matera, M.Longo, A.Srivatsa, A.Rengan, J.Krishnaswamy, and J.Narayan, Department of Materials Science and Engineering, and R.Nemanich and E.Buehler, Department of Physics, North Carolina State University, Raleigh, NC-27695 Crystalline diamond deposition on silicon(IOO) substrate at -7500C by hot filament(-2000°C) decomposition of methane and hydrogen(1 SCCM CH4 : 100 SCCM H2' 20 to 60 Torr total pressure) has shown considerable diversity in morphology and carbon-carbon bonding over the deposited area. These are examined by SEM, TEM, and Raman techniques and the diversities in the deposited film are primarily related to the inhomogeneities in the deposition process. Preliminary results of the effect of an excimer laser beam irradiation of the substrate during deposition will be described. . We acknowledge research support by the NSF under grant No.DMR-8703621. (10:10 a.m.)
BREAK
(10 :40 a.m.) (Invited) TRIBOLOGICAL STUDIES ON DlAMONDLIKE FILMS DEPOSITED WITH AN ARC DISCHARGE METHOD: J-P Hjrvonen, Los Alamos National Laboratory, Materials Science and Technology Division, P.O. Box 1663, Ms-k765, Los Alamos NM 87545; R. Lappalainen, University of Helsinki, Department of Physics, Siltavuorenpenger 20 M, 00170 Helsinki, Finland; J. Koskinen, Cornell University, Department of Materials Science and Engineering, Bard Hall, Ithaca NY 14853; A. Anttila, University of Helsinki, Department of Physics, Siltavuorenpenger 20 M, 00170 Helsinki, Finland, and M. Trkula, Los Alamos National Laboratory, Materials Science and Technology Division, P.O. Box 1663, Ms-E549, Los Alamos NM 87545.
Diamondlike carbon films were deposited on high chromium tool steel and silicon using an arc discharge method. Characterization of the films was carried out with Raman spectroscopy. In dry sliding wear and friction tests with a hardened steel pin as a counterpart friction coefficient decreased from the initial value 0.20 to 0.13 after a sliding distance of I 130 m. Wear coefficient of the order of 10. 17 m 3/mN was obtained. Moreover, a transfer layer on the pin was formed and is believed to have the dominating effect on the sliding behavior. Degradation of good tribological properties was found to be caused by a partial wear-through and a subsequent redeposition of the transfer film on the wear track during prolonged sliding. (11: 10 a.m.) PRELIMINARY INVESTIGATION ON THE DEPOSITION AND CHARACTERI@ATION OF DIAM~ND/DIAMOND kIKE CARBON FILMS ~. Purohit, J. Mazumdar , A.S. Rao, D.P. Arora Argonne National Laboratory. Argonne, IL 60439
~niversity
of Illinois, Urbana - Champaign, IL 61801
'David Taylor Research Center, Annapolis, MD 21402-5067 #permanent Address: University of Cincinnati, Cincinnati, OH Diamond and/or diamond like carbon films have been deposited ~t temperatures in the range 600 - 1IOaoe on silicon and niobium substrates using dilute mixtures of methane and hydrogen gases, using few different plasma deposition
techniques. The deposited film microstructure and composition was characterized using scanning electron optical microscopy and Raman spectroscopy respectively. The results suggest that both crystal structure and the physical characteristics of the deposited films depend upon the precursor gas chemistry, process temperature and the plasma - ion density. These results and some thermophysical and electrical resistance properties of these films will be discussed in detail.
(11 :30 a.m.) Hydrogen free DLC film by laser ablation and plasma hybrid technique: J.Krishnaswamy, A.Rengan, and I.Narayan, Materials Science and Engg, North Carolina State University, Raleigh, NC-27695-7916 and C.McHargue and M. OHern, Metals and Ceramics Div, ORNL, Oak Ridge, TN-37831- 6118, Y.Cong, R.Collins and K.Vedam, MRL, ,University Park, PA-16802 Hydrogen control is a basic requirement for several applications as
it controls the mechanical, electrical, and optical properties of DLC films. Recently a novel technique of depositing hydrogen free DLC films by a laser ablation and plasma hybrid technique(1 )was described and the deposited films were shown to possess mechanical and optical properties superior to those films deposited by the conventional technique. Herein a method of controlling hydrogen incorporation into the films is described. The films are characterized for hardness, and by FT-IR, Raman, and SIMS techniques. 1. J. Krishnaswamy, A. Rengan, J. Narayan, K. Vedam and C. J.McHargue, Appl. Phys. Lett.,~, 2455 (1989).
62-Annual Meeting
PROCESS FUNDAMENTALS II: Thermodynamics and Reaction Kinetics in Metallurgical Processes A Sponsored by the TMS PrQcess Fundamentals Committee Tuesday AM February 20, 1990
Room: Salon H Marriot
Session Chairmen: J.F. Elliot, Department of Materials Science and Engineering, MIT, Cambridge, MA 02139; Y.K. Rao, Department of Materials Science and Engineering, University of Washington, Seattle, WA 98195
(9 : 45 a_m.) REDUCTION OF HEMATITE WITH HYDROGEN: NUCLEATION AND GROWTH: M. Moinpour. Research Assistant Professor. Dept. of Electrical Engineering, University of Washington, Seattle, WA 98195; Y.K. Rao, Professor. Dept . of Materials Sc ience and Engineering, University of Washington, Seattle, WA 98195. Analysis has been made of the nucleation and growth phenomena attending the reduction of hematite with hydrogen and hydrogen helium gas mixtures. The experimental data correlated well with the Avrami-Erof eyev equations. Whereas the Avrami-Erofeyev model envisions the reduction mechanism in terms of formation of iron nuclei and their subsequent growth in a reasonable fashion, it does not s eem very precise f or describing the early stages of the reduction. Another model , called "Pore Cra ter" model , has been developed to explain the reduction of hematite in terms of formation of "pores" and subsequent growth with interpenetration and overlap . The model takes into considerati on the structural changes dur i ng the transforma tion of hematite to magnetite as well as the solid state diffusion of Fe 2+ (Fe X) and 0 2 - (OX) ionic spec i es. (10 ,10 a .m .)
(8 : 30 a.m.) COHPREHENSIVE EQUILIBRIUH FOR HYDROHETALLURGICAL SYSTEHS ; HETAL-AHHONIA-WATER SYSTEM: Shaolin Zhong and Diji Zheng , Dept. of Hetallurgy, Central South University of Tech ., Changsha, Hunana , P. R. of China 410083 Based on mass·balance and charge-balance equations, a new calculation method has been proposed , which gives equilibrium relations between solid phase and associated aqueous phase in terms of total concentrations rather than the activities of individual dissolved species . This technique is described with Hetal(Cu, Hg, Ni)-Ammonia-Water systems as examples . When an initial aeuqous solution system is specified, along with the addi'tion of ammonia progressively the corresponding equilibrium diagrams: potential-PH diagram for CopperAmmonia-Water system , solubility curve for Hagnesium-AmmoniaWater system, and voltage-concentration ratio curve for Nickle-Ammonia-Water system can be calculated. Hore quantitative information on real hydrometallurgical processes can be obtained from these diagrams when compared with the conventional diagram. I
(8 : 55 a.m.)
THERMODYNAMIC STUDIES ON HF-HN03 -H60 SOLUTIONS B Chen and R_ G_ Reddy Chemical and Metallurgical Engineering Department, Mackay School of Mines, University of Nevada-Reno, Reno, NV 89557_
BREAK
(10:25 a_m .)
KINETICS OF CARBOTHERMIC REDUCTION OF ALUMINA IN THE PRESENCE OF A SOLVENT METAL, 1700-1850·C - EXPERIMENTAL: R. A. Frank, Union Carbide Industrial Gases, In c ., Tarrytown, NY 10591, and J. F. Elliott, Mass. Institute of Technology, Cambridge, MA 02139 The rate of reduction of alumina by carbon in the temperature range of 1700 to 1850·C has been measured (a) by determining the rate at wh i ch CO gas was evolved by the reduction reaction, an d (b) by thermogravimetric analysis in which the change of weight of a crucible containin9 alumina, carbon, and liquid tin was measured. · The total pressure in the reacting systems was held constant in the range of 0.5 to 0.2 atm. abs. The effects of type and particle size of alumina, particle size of carbon, temperature , total pre s s ure, and several minor expe rimental variables were studied. The overall r a te process follows a simple pseudo-firft order kinetics model for which the rate constant, k ( sec- ) has an apparent activation energy of 328 ± 28 kcal, and it is proportional to the pressure (atm.) to the -2.30 ± 0.36 power. ( 10 : 50 a .m .) KINETICS OF CARBOTHERMIC REDUCTION OF ALUMINA IN THE PRESENCE OF A SOLVENT METAL, 1700-1850· C - THEORETICAL : R. A. Frank, Union Carbide Industrial Gases, Inc., Tarrytown, NY 10591, and J. F. Elliott, Mass. Institute of Te chnology, Cambridge, MA 02139
Partial preaaures of HN03 , HF and H60 over HN03 -HFternary solutions were determined by a H60 transpiration technique at 298 and 313K _ The samples were analyzed using liquid ion chromatograph. The concentration of HN03 and HF varied from 1 to 8 molalities_ Experimental results show that the partial preasure of HN03 and HF increases with the increasea in concentration of HN03 and HF, and increases with increase in temperature . The partial pressure of H60 decreases slightly with increase in concentration of HF, and alao decreases significantly with increase in concentration of HN03
A detailed study has been made of the experimental results from a study of the kinetics of reduction of alumina by carbon in the presence of liquid tin as the s olvent for the aluminum metal. The results hav e been tested again s t a number of kinetic models of mass and heat transport, and chemical reaction rates, including the intrinsic-transport control model. The rate data cannot be fitted in the classical way by any single reaction rate model, nor a model that includes a chemical rate process and a transport process in series. Instead, it appears that a rate controlling mechanism includes the combined effects of a chemi c al rea c tion and heat transport, possibly in a parallelseries path. Application of the various possible models to interpretation of the data is discussed.
(9:20 a_mol
(11: 15 a_m_)
PREDICTING ACTIVITIES OF WATER IN CONCENTRATED IONIC SOLUTIONS: Dr. c.L. Kusik, Arthur D. Little, Inc., 20 Acorn Park, Cambridge, MA, 02140.
EFFECT OF Ir AND Hf ADDITIONS ON THE SINTERING BEHAVIOR OF TUNGSTEN: L.A . D'Cruz , D.L. Jacobson and J.F. Horris , Department of CfiemlcaI , BlO and Material s Engineering , Ari zona State Unive rsity , Tempe, AZ 85287-6006 , USA .
Calculations of equilibrium concentrations of strong electrolytes involved in reactions in aqueous solutions such as in leaching, metal pickling, stack gas scrubbing, causticising, and other hydrometallurgical operations generally have been aided by a number of developments in predicting ionic activity coefficients in aqueous solutions leading to imprOVed methods for predicting solubility limits of salts in multicomponent solutions at varous temperatures by use of a thermodynamic solubility product. For a hydrated salt or crystal species, the thennodynamic solubility product K, can be represented by an equation of the following standard form : log K, = v, log 'Yl2m, + U, log 'Yllm, + n log a.,. Although methods have been presented to predict 'YI2' estimates of the activity of water, a." have relied on the Gibbs-Duhem relationships using graphical or integral techniques which often have proved to be time consuming. This paper presents a rapid way to estimate activities of water over multicomponent solutions. Comparison is made with the results of prior work.
W-based high-temperature a lloys are promising candidates f or application s in high-power thermionic e n e rgy convert er s. Eutecti c Ac t i vated Sinterin g is being studied in compacts prepared by a dding up to 2wt% of a eutectic blend of Ir and Hf powders (Hf-17a/ o Ir) t o pur e tungsten powder . Sintering e xperiments conc entrat e on t e mpe ratur e s between llOO° C and 2000·C. Expe rimental de n si fi c ation ra te s will be c omp a r e d with an existing ma t hematical mod e l based on mult i pl e diffusion me chanisms. Opt i cal and SEM observations will als o be presented.
Annual Meeting-63
REDUCTION TECHNOLOGY m: Bath Chemistry Sponsored by TMS Aluminum Committee Tuesday AM February 20, 1990
Room: Salon J & K Marriott
Session Chairman: Mark Taylor, Comalco Aluminium Ltd., Comalco Research Centre, 15 Edgars Rd., Thomastown, Victoria 3074 Australia (8:30 a.m.) BATH CHEMISTRY CONTROL SYSTEM: D. J. Salt, Anglesey Aluminum Metal Limited, Penrhos Works, P. O. Box 4, Holyhead Gwynedd LL 65 2 UJ. A system of bath chemistry control that takes account of alumina composition, cell age, type of break and feed mechanism, type of fume treatment has been devised. Aluminum fluoride usage versus cell age curves have been developed for half break and point feed cells using both pure ore and a mixture of pure ore and scrubber reagent. These curves are adjusted to take account of the soda and lime content of the alumina. Bath temperature is taken into account when calculating corrective additions of aluminium fluoride. These are increased when the pot is cool and decreased when the pot is warm.
(9:00 a.m.) ACTION OF CALCIUM IN THE REACTION WITH DILUTE FLUORIDE SOLUTIONS:
Z. Yang and R. W. Smith, Department of Chemical and Metallurgical
Engineering, University of Nevada-Reno, Reno, NV 89557.
Fluorides can be discharged into the atmosphere from high temperature industrial processes such as fused salt electrolysis production of aluminum. Such fluorides may end up in surface and ground waters. This paper investigates the interaction of dilute HF and NaF solutions with calcium oxide, calcium hydroxide and a common carbonate rock mineral - calcite. When dilute HF and NaF solutions w~e reacted with CaD and Ca(OH)2 and each studied as a function of time, in dilute HF solutions within 5 minutes virtually all fluoride was removed, while NaF solutions required almost 1 hour to remove equivalent amounts of fluoride. When dilute HF solutions were reacted with calcite at different temperatures and the reaction studied as a function of time, the reaction was found to obey a second order rate law. When dilute NaF solutions were interacted with calcite the reaction was much slower and appeared to be independent of temperature and to obey an apparent first order rate law.
(9:30 a_m_) PHASE DIAGRAM DATA IN THE SYSTEM Na3AIFS-LizAIFS-A1203' PART I: LIQUIDUS TEMPERATURES FOR THE PRIMARY CRYOLITE CRYSTALLIZATION: A. Rostum, A.Solheim and A. Sterten, SINTEF, the foundation for Industrial and Scientific Research at the Norwegian Institute of Technology, N-7034 Trondheim, Norway. Liquidus temperatures for primary crystallization of Na3AIFS in the system Na3AIFs-LizAIFs-A 1203 was determined by means of thermal analysis. The effect of CaF2 on the primary crystallization temperature was established. The results are presented in the form of quasibinary liquidus diagrams. An empirical equation for the calculation of liquidus temperature is presented, which is valid from 1011 0 C to approximately 800 0 C.
(10:30 a_m.) STUDIES ON THE DISSOCIATION REACTION OF MOLTEN CRYOliTE BY COMPUTER SIMULATION: Qui Zhuxian and Xie Gang, Northeast University of Technology, Shenyang, China. The dissociation reaction of A 1F6- ions was studied by Monte Carlo technique. The most probable reaction is the formation of A 1F4- and Fions In the total 126 F- ions studied there are 103 ions in the combined form (81.75%), 5 F- ions are in the bridge form (3.96%), 18 F- ions are in free state (14.29%). Free F- ions from Al F6 3- are 10.33%, which equals to 2a16. Hence we get the degree of dissociation (l to be 31 %, which is in good agreement with cryoscopic method by Grjotheim. (11:00 a.m.)
REDUCTION OF SILICON IN AN ALUMINUM ELECTROLYSIS CELL: Rudolph Keller, EMEC Consultants, Export, PA 15632 USA; B.J. Welch, University of Auckland, New Zealand; A.T. Tabereaux, Reynolds Metals Co., Sheffield, AL 35660 USA_ Silicon dioxide dissolved in the electrolyte of an aluminum electrolysis cell is reduced by direct electrOlysis and/or reduction by aluminum metal. Silicon-aluminum alloys can be directly prepared in a Hall-Heroult cell by adding, e.g., quartz, sand or quarzite to the bath. When this was done in an industrial cell, significant losses in current efficiency were experienced. Laboratory investigations indicate that such losses may be due to only part of the silicon directly alloying with aluminum upon reduction. Another part forms incoherent solid silicon that gets dispersed in the electrolyte and may then be reoxidized.
(11 :30 a.m.) EXPERIMENTAL RESULTS OF UF EMPLOYMENT TO INDUSTRIAL POTS Liubo Vujasinoyic and Stipe Gulin, Sour Industrija Aluminija, "Boris Kidric", Ro Centar Za Aluminij, 59000 Sibenik, Cesta Samoupravljaca BB, Yugoslavia. In the period of May 88- March 89 we carried out in our plant testing of the usage efficiency of Li2C03 at 12 electrolytic potswith concentration ofLiFin the electrolytic bath of 1.5% and 2.5%. Our paper would comprise: -Influence of lif upon technological parameters for the electrolytic process control and movement of Li in the electrolytic metal. -Comparative results of LiF usage efficiency in the plant of electrolysis. -Behaviour of Ii through the casting process and its influence upon the castings quality.
SOLIDIFICATION BEHAVIOR OF OPTICAL AND ELECTRONIC MATERIALS I Sponsored by the TMS Solidification Committee Room: Conference Theatre Quality Hotel
Tuesday AM February 20, 1990
Session Chairmen: M.E. Glicksman, Materials Engineering Department, R.P.I., Troy, NY 12180; J. Parsey, AT&T Bell Laboratory, 2525 North 12th St., Reading, PA 19612 (8:15 a_m.) OPENING REMARKS (8:30 a.m.) ON THE SOLIDIFICATION OF OPTO-ELECTRONIC MATERIAL: K. A. Jackson, AT&T Bell Laboratory, Hill-Side Road, 600 Mountain Ave., Murray Hill, NJ 07479. (Abstract Not Available) (9: 10 a.m.)
(10:00 a_m.) PHASE DIAGRAM DATA IN THE SYSTEM Na3AIFs-Li3AIFS-AIF3-A1203' PART ii: ALUMINA SOLUBILITY: E. Skybakmoen, A. Solheim and A. Sterten, SINTEF, the foundation for Industrial and Scientific Research at the Norwegian Institute of Technology, N-7034 Trondheim, Norway. The solubility of alumina in the system Na3AIFs-Li:Jft.IFs-AIF3 was determined by measuring weight loss of a rotating sinter-corundum disc. The results were fitted to an empirical equation which covers the composition range 0 - 50 mol% LixAiF3+x where 0 < x < 3 and temperatures ranging from 10300 C and down to the temperature of the Na3AIFslA1203 univariant line. The effect of CaF2 and MgF2 on the solubility was established. Based on these data and available liquidus temperatures for primary crystallization of Na3AIFs (part 1) the liquidus diagrams were constructed for the ternary systems Na3AIFs-Li3AIFs-A 1203 and Na3AIFs-AIF3-A 1203
BULK GROWTH OF II-VI SOLID SOLUTION SEMICONDUCTING ALLOYSAN OVERVIEW, R. N. Andrews, Department of Materials Engineering, University of Alabama at Birmingham, Birmingham, AL 35294 Considerable attention has been focused on attaining a basic understanding of the mechanisms involved in the bulk growth of single crystals of II-VI semiconducting alloys. However, there are several problems associated with the directional
solidification
and solidus coefficient
difficult.
of these alloys.
curves
makes
Also,
The widely separated
liquidus
which results in a large segregation bulk growth of homogeneous materials
in
the
II-VI systems,
there
is
a
large
variation in the thermophysical properties both with temperature and
composition.
This presentation will address the advantages
of bulk growth of II-VI compounds, the problems associated with the growth, and the results of work underway by the author and co-workers
on
the growth and characterization
of
Hg
1
Cd Te,
Hg1 _ ZnxTe, Hg1_xCdxSe, and HgJ_xZn Se. The prospects t~r this impo~tant area of crystal growtfi will also be discussed.
64-Annual Meeting (9 : 50 a .m.)
(9:05 a.m. )
SOLIDIFICATION BEHAVIOR OF A NONLINEAR OPTICAL MATERIAL, Tl-As-Se K. C. Yoo, N. B. Singh and R. H. Hopkins, Westinghouse Science and Technology Center, 1310 Beulah Road, Pittsburgh , PA 15235. A nonlinear optical material of the sulfosalt compound, T13AsSe3 (TAS), has received considerable attention as an efficient harmonic frequency converter of the output wavelengths from carbon dioxide lasers. The solidification behavior of TAS at different growth conditions (growth rate, temperature gradient, impurity level and composition) has been studied to understand the way that high quality TAS crystals can be grown. A key issue is optical homogeniety required for efficient harmonic generation. The control of impurity and lattice imperfections during solidification has been evaluated to improve crystal quality. In this presentation, we will discuss our observation of the solidification behavior of TAS controlled by growth parameters.
PHASE DIAGRAM STUDIES I N Ba l . • K. BiO): D G Hi n k s , B . Da brows k i , J.D . Jo rgen sen, D . R . Ri cha rds , Y. Zheng, Sh i yo u Pe i a n d A. W. Mi t c he ll, Material s S cien c e Di v isi on , Argonne Nati o nal La b o rat o r y , Argo nne II 60439
(10 : 30 a.m.) ENCAPSULATED GROWTH OF GaAS : G. J. Abbasch i an , Department of Mater ials Science and Engineering, University of Florida, Gainesville, FL 32611. (Abstract Not Available)
Ba l _.K. BiO) single-pha se sampl es can b e synthesized from x= O t o 0.5. At each c omposi ti on a t wo -s t ep firi n g p r ocedure i s necessary to p r e p a r e the material fr om the st a r ting ox ide powders. K is inco r porated into the la tti ce by firing at 725 C in a comp os ition- dependent P02 a tmosphere followed by a l o w-t e mperature anneal in 02 t o fill the oxygen va c an c i e s . At x= O, the material i s mo n oc li n ic with both fro zen- out rigid rotations and symmetri c breathing mo de s of the Bi 0 6 o ctahedra. As x inc r eases , the breathing mode i s r api dl y removed l e adin g to a n o rth o r hombic st ru c tur e an d finall y at x=O.375 t h e r ig id r otat i o n distort i o n dis a ppear s and t h e mat e rial becomes c ubic a nd s upe r co nduc ting. Wor k s u ppor te d b y th e U.S. DOE under con t r act No. W- 3 1- 109ENG- 38 an d t h e NSF-Of f i ce o f Scien ce a n d Te chno l ogy Cente rs u n de r Co n tra c t No. STC- 880985 4 .
(9 :40 a.m.)
(11 : 10 a.m.) SOUDIFICATION OF TERNARY CdTe COMPOUNDS; P J Larson Jr, Senior Staff SOientist, Corporate Research Center/A02·26, Grumman Corporation, Bethpage, NY 117t 4·3580 CdTe and ternary modifk:atlon thereof are Important members of the II·VI
family of compound semiconductors. They are used as: nuclear, gamma ray, and Infra·red detectors; laser optics; and non·llnear optical materials. The primary needs within these application communities are for Increased structural perfection and yields within the bulk crystals and uniform specific propertle8 wtthin substrates cut from the bulk crystals . TheBe crystals are typically grown using modified Bridgman crystal growth techniques. This talk will present the state·ol·the·art In bulk ternary CdTe growth and the means under Investigation for Improving the bulk and wafer properties. Modelling and empirical results with respect to active atmosphere control, conveotion control, and thermoplastic stress minimization will be
disoussecl.
The T*-phase rare earth cuprates at R2Cu04 stoichiometry are intermediate in structure between the La2Cu04 and Nd2Cu04 structure types. The phase forms when R is an approximately equal mixture of two rare elements . Sr on the R site is found to significantly stabilize this phase . We present data on the phase stability limits, structural refinement , physical properties and attendant superconductivity within the T* phase field. The maximum superconducting Tc in the T*-phase is equal to that found in Sr-doped La20u04 , in spite of signi ficant structural differences . *
Work supported by the U. S . Department of Energy, OBES.
( 1 0 : 15
THERMODYNANUCSANDPBASE RELATIONS OF EXOTIC MATERIALS I: Superconducting Materials Sponsored by the ASM-MSD Thermodynamic and Alloy Phases Committee TuesdayAM February 20, 1990
PHASE STABILITY AND SUPEROlNIJUCTIVITY IN T*-PHASE R20u04 RARE EARTH aJPRATES: * Z. Fisk, S.-W. Cheong , J. D. Thompson, M. Hundley, R. B. Schwarz. and G. Kwei, Los Alamos National Laboratory, Los Alamos. NM 87545 .
Room: Plaza C Inn at the Park
Session Chainnan: A.S. Nash, Electrical and Biomedical Engineering Department, Marquette University, Milwaukee, WI 53233 (8:30 a.m.) GRAIN GROWIll AND INTERGRANULAR PHASES IN CYCLICLY ANNEALED YBa20u307.x : * A. P. Clarke, R. B. Schwarz. and J. D. Thompson. Los Alamos National Laboratory. Los Alamos, NM 87545. We have studied the effects of repeated oxygen anneals between 7SOoC and 9SQPC on intergranular material composition and grain morphology for bulk YBa20u307- x. On heating the material to 95QPc for the first time, YBa20u307- x reacts with CUO present at grain boundaries and forms Y2BaOuO,; and melt. Cyclic annealing causes a smooth change in the intergranular material composition . After eight cycles the 950°C reaction has disappeared and has been replaced by the eutectic melting of YBa20u307 - x . OuO, and BaCuO:!. The cyclic annealing promotes a monotomic increase in average grain size. an overall narrowing in the grain size distribution , and a significant increase in the volume fraction of superconducting phase. * Work supported by the U. S. Department of Energy. Office of Basic Energy Sciences.
a.m . )
BREAK
(10 :35 a.m . ) 1HERMODYNAMICS OF ONE, 1WO AND TIiREE-DIMENSIONAL OXYGEN ORDERING IN YBa2Cu30x. M. Asta*; G. Ceder, C. Carter, M.E. Mann*, M . Kraitchman, E. Salomons and D. de Fontaine. Dept. of Materials Science and Mineral Engineering, V.c. Berkeley and Lawrence Berkeley Lab, Berkeley, CA 94720; *Dept. of Physics, V .C. Berkeley.p It is shown that the (three-dimensional) tetragonal to orthorhombic transition in the 1·2·3 superconducting compound can be mapped onto a two dimensional Ising model with anisotropic second neighbor effective pair interactions. The CVM has been used to calculate a phase diagram based on interaction parameters provided by first principles LMTO calculations performed by Ph. Sterne at Lawrence Livermore National Lab. At high temperature, agreement with experimental phase transition data from Oak Ridge National Lab is excellent. At low temperatures, two-dimensional ordering gives way to quasi one-dimensional states of order as described by the linear chain ISing model for which the thermodynamics are known exactly. Monte Carlo simulation produces structures and diffraction patterns in strildng agreement with those found experimentally by the Antwerp TEM group. [Supported by grant from Dept. of Energy, under contract DE-AC03-76SF00098].
(11 : 10 a. m. ) STRUCTURAL INSTABILITIES IN SUPEROlNDUCfING aJPRATES:
T. E. Mi t chell, Los Alamos National Laboratory, CMS, MS K765, Los Alamos , NM 87545
Inve stigations of the crystal chemistry of the family of superconducting cuprates has yielded a rich harvest of crystallog raphic structures. Further investigation has revealed t hat many of the structures undergo small distortions which are manifested in a variety of ways. For example , the tetragonel-to-orthorhomic transformations observed in the Y-Ba-Ou-Q and La-Sr-Ou-Q systems are due respectively to oxygen va cancy ordering and rota tion of the Cu06 octahedra. The incommensurate periodicity observed in the Bi-Ga-Sr-Ou-Q system is due to displacements of the Bi ions . The commensura te periodicities observed in Gd2Cu04 and other lanthanide cuprates are also probably due to displacements of the rare-earth ions. On the other hand, commensurate structures in reduced Y-Ba-Ou-Q are due to oxygen vacancy ordering. These various structural instabilities are discussed i n terms of structural nuances in the perovskite family.
Annual Meeting-65
TMS ELECTRONIC MATERIALS LUNCHEON/SPEECH
processes. The importance of the interrelationship between transport phenomena and chemical reactions in metallurgical systems, which are almost always heterogeneous in nature, will be emphasized. The paper then takes a brief look at the prospects for process engineering in extractive metallurgy.
Guest Speaker: Harry C. Gatos, Massachusetts Institute of Technology, Department of Materials Science and Engineering, Room 13-4122, Cambridge, MA 02139 Tuesday, 12:00 Noon February 20, 1990
Room: Tiffany Terrace Inn at the Park
Topic: "The Revolution and Evolution of Electronic Materials: The First Forty Years" (12,00 Noon)
TUESDAY PM
TMS ELECTRONIC MATERIALS LUNCHEON SPEECH
Forty years ago the discovery of the transistor signaled a materials revolution of an unprecedented magnitude. The materials community was totally unprepared for the new chemical
and crystalline perfection demands made by semiconductors and their new technologies. Metallurgists took on many of the new challenges. Such developments as purification by zone refining, growth of large-single crystals, and defect control were essential for sustaining of the revolution. A few years later, the developments made in semiconductors were being transplanted to metals and other classes of materials. These striking advances in materials science and engineering, spearheaded by electronic materials will be discussed in this luncheon presentation.
TMS EXTRACTION PROCESSING DIVISION
ADVANCED METAL & CERAMIC MATRIX COMPOSITES m: Fracture and Fatigue Sponsored by the TMS Powder Metallurgy Committee, TMS Physical Metallurgy Committee, and ASM-TMS Composites Committee Tuesday PM February 20, 1990
Room: ParkD Inn at the Park
Session Chainnen: Prabhat Kumar, Cabot Corporation, Boyertown Plant, County Line Road, Boyertown, PA 19512; Henry Rack, Clemson University, Department of Mechanical Engineering, Clemson, SC 29634-0921
METALLURGYLUNCHEON~PEECH
Guest Speaker: Dr. Thomas S. Mackey, P.E., Key Metals & Minerals Corporation Tuesday, 12:00 Noon February 20, 1990
Room: Grand Ballroom E Marriott Hotel
Topic: "Extractive Metallurgy of Non-Ferrous Metals with Special Emphasis on Lead-Zinc-Tin, Rare Earths, and Copper" (12,00 Noon)
EXTRACTIVE METALLURGY LUNCHEON SPEECH
The world-wide status of lead-zinc-tin, copper and rare earths including the latest extractive metallurgy and future technical and commercial developments within these industries will be covered in some detail. The effect of environmental laws on these industries in'the USA, Japan and Europe will be covered along with current and future supply/demand projections.
TMS EXTRACTIVE METALLURGY LECTURE Speaker Awardee: H.Y. Sohn, University of Utah Tuesday, 1:45 PM February 20, 1990
Room: Salon 4 & 5 Marriott Hotel
Topic: "The Coming-of-Age of Process Engineering in Extractive Metallurgy" (1,45 p."'.)
EXTRACTIVE METAI,LURGY LECTURE
The increasing importance of the application of process engineering principles and techniques in extractive metallurgy is reviewed. The significant developments made in recent years in quantitively describing metallurgical processes are discussed. Some general approaches and building blocks of process engineering are presented, together with examples related to a number of unit operations and unit processes encountered in extractive metallurgy. A considerable amount of discussion will be devoted to the advances made in the engineering analysis of rate processes in metallurgical
(2:00 p.m.) FRACTURE OF CERAMIC AND METAL MATRIX COMPOSITES: A. G. Evans, University of California - Santa Barbara, Materials Department, Santa Barbara, CA 93106.
(Abstract Not Available) (2:30 p.m.) EVALUATIONS OF TOUGHNESS AND MECHANICAL PROPERTIES IN AI/SiCp COMPOSITE:T.Kobayashi, H.Iwanari, Toyohashi University of 'Technology, Department of Production Systems Engineering, Tempaku-cho, Toyohashi, Aichi 440, Japan, K.Yamamoto, TYK Corporation Limited, Ohata-cho, Tajimi, Gifu 507, Japan. Dynamic fracture toughness and several mechanical properties
of composite of 6061 aluminum alloy reinforced with SiC particle were evaluated. It was prepared by powder metallugy (P/M) method. Dynamic fracture toughness parameters were evaluated by newly developed CAl (computer aided instrumented impact testing) system. That is, Jd value and crack propagation toughness (i.e.
tearing modulus Tmat) were
evaluated with a precracked Charpy type specimen. As a result, it was found that strength and hardness were increased with increased addition of SiC particle in AI/SiCp composite, while
elongation, J d value and Tmat were significantly decreased. Failure or decohesion at the particles was obvious from the observation of fracture surface. This was discussed from a view of decrease of the toughness in such composite.
(2:45 p.m.) MODELING TOUGHENING EFFECTS IN WHISKER-REINFORCED COMPOSITES Richard G. Hoagland, Washington State University, Pullman, WA 99164-2920; Charles H. Henager, Jr., Battelle Pacific Northwest Laboratory, Richland, WA 99352. Most explanations of toughening due to whisker reinforcements consider effects associated with whiskers bridging the crack or contributions from crack deflection. In this paper we present a numerical simulation model that considers the crack tip screening effects of a set of discrete, arbitrarily positioned whiskers. Screening (and antiscreening) derive from the stresses created in the matrix due to matrix/whisker elastic incompatibility. The method used is similar to that employed earlier in modeling toughening effects by microcracks and by dilatant centers and permits an examination of a range of microstructural interactions and their effects on strength and toughness. Results are presented for single whiskers, for several different random arrays of whiskers, and for arrays which show advantages of whisker texturing.
66---Annual Meeting (3:00 p.m.) Effect of Dislocation and Recovery on Si3N4 Whisker Reinforced Aluminum P/M Cornposite,Mamoru Mabuchi,Tsunemichi Imai,Yoshihisa Tozawa & Yoshio Murase, Government Industrial Research Institute, Nagoya,l Hirate-cho,Kita-ku,Nagoya 462,Japan After very fine aluminum alloy powder(2124 aluminurn,average size 7~m)
was blended uniformly with Si3N4 whisker in organic solvent,
Si3N4 whisker reinforced composite was fabricated by hot pressing before hot extrusion. Effect of temperature and strain rate on mechanical properties such as strength and ductility etc. was investigated. Tensile strength of the composite increased more remarkably than that of aluminum only with raising strain rate at elevated temperature. And in this case a fracture strain decreased monotously with increasing strain rate. The strain rate had an effect on dimple size of fracture surface of the composite. This tendency was found to connect with the increase of dislocation density and the recovery developted during tensile testing at elevated temperature
(3: 15 p.m.)
Microstructure and Mechanical Behavior of AVSiC Metal Matrix Composites processed by Spray Atomization and Deposition: E.J. Lavernia and FA Mohamed, Materials Section, Department of Mechanical Engineering, University of California, Irvine, CA 92717. In the present study the microstructures and mechanical behavior of several spray deposited AVSiC materials are investigated as a function of processing conditions and alloy chemistry. The microstructural characteristics of the spray deposited materials is investigated as a function of cross·sectional thickness for a single experiment, as well as a function of processing variables for distinct experiments. The mechanical behavior studies include both room and high temperature properties. The microstructural studies are accomplished using the scanning, transmission and optical microscopes. (3:30 p.m.) EFFECTS OF HEAT TREATMENT ON THE MICROSTRUCTURE AND TENSILE PROPERTIES OF A F/M SiC PARTICULATE/7091 Al MMC. S. Dionne and M. R. Krishnadev, CANMET, 555 Booth Street, Ottawa, Canada KIA OGl. Most investigations conducted to date on SiC-reinforced metal matrix composites have utilized heat treatment procedures very similar to those which are used for the unreinforced matrix
alloys.
However, evidence in the literature indicates that the
presence of the reinforcement phase can modify the aging response
of some matrix alloys significantly. A commercially fabricated SiC particulate reinforced 7091 aluminum composite was therefore subjected to controlled heat treatments to evaluate the aging response of this composite system.
Microstructures of the as-fabricated and heat-treated materials were characterized using optical microscopy t SEM and TEM to determine the effects of heat treatment on the interface chemistry and on the size and distribution of the matrix precipitates. Hardness and tensile properties of the as-fabricated and heat treated composites were evaluated.
(3:45 p.m.)
ARE THERE FATIGUE EFFECTS ON CERAMICS AND CERAMIC MATRIX COMPOSTES UNDER CYCLIC LOADING? ; I...Euili.:, A. P. Majidi" and T. W. Chou", ·Dept. of Mech. Eng., Doshisha University, Kyoto, 602 JAPAN, "Center for Composite Materials and Dept. of Mech. Eng., Univ. of Delaware, Newark, DE 19716 USA This review focuses on the existence of cyclic effects on fatigue strength of ceramics and ceramic matrix composites (CMCs) through the survey of pub· lished papers. The definition of the term, "cyclic effects on fatigue life", is not always clear. If it also includes the observed SCG (Slow Crack Growth) accel· eration under cyclic loading as well as cyclic fatigue effect depending on the number of loading cycles, the survey of published data on cyclic fatigue of glasses, some ceramics and CMCs might give the conclusion that there are some cyclic effects on fatigue of these materials except glasses and SiC. In many cases, their fatigue lives are time·dependent, not dependent on the number of cycles. Glass matrix composites do not seem to show the pure cyclic effect based on the number of cycles. The observed cyclic effects for some ceo ramics could be explained by the virtual SCG acceleration mechanism. Hence, we propose a simple fatigue acceleration mechanism based on the pure SCG under cyclic loading. Friction between fibers and matrices could be a source of cyclic effects on fatigue in CMCs having continuous fibers. (4:00 p.m.) CASTING PARTICULATE AND FIBROUS METAL-MATRIX COMPOSITES BY VACUUM INFILTRATION OF A LIQUID METAL UNDER AN INERT GAS PRESSURE: Jingyu Yang and D.D.L. Chung, State University of New York, ·Buffalo, NY 14260, USA Reported here is a method of making metal-matrix composites.
This method combines the essentials of three liquid-phase fabrication methods: (i) vacuum infiltration, (ii) infiltration under an inert .gas pressure, and (iii) squeeze casting. In this method, the particulate or fibrous preform is placed in a
chamber and the matrix alloy is placed above the preform. The matrix alloy is heated to the liquidus temperature together with the chamber and the preform under vacuum. Then an inert gas like argon is compressed on to the top surface of the matrix alloy melt, forcing the melt to infiltrate the preform. The pressure is 1,000 - 2,500 psi. Because the temperature of the melt is just at the liquidus, it is much lower than that in squeeze casting. Moreover, the pressure is an order of magnitude lower than that in squeeze casting. The low temperature lessens the interfacial reaction between the matrix and the filler, while the low pressure essentially eliminates preform compression.
(4: 15 p.m.) SiC FIBER REINFORCED Al-Si ALLOY COMPOSITES BY DIFFUSION BONDING OF PREFORMS. P.V.Hegde,V.Gopinathan and P.Ramakrishnan, Dept.of Metallurgical Eng., Indian Institute of Technology,Bombay 400076, India. Al-12.4 wt.% Si alloy matrix composites reinforced with SiC fibers were fabricated by hot pressing of infiltrated composite preforms. The effects of pretreating the fibers with an aqueous solution of K2ZrFsflux on the fiber-matrix wettability as well as on the microstructure, strength and fracture characteristics of the composites were examined. The composites exhibited fairly uniform distribution of fibers with little indication of any voids either within the matrix or at the fiber-matrix interface. A fiber volume fraction of 0.15 resulted in composite strength of about 220 MPa as compared to a matrix strength of only 110 MPa. SEM examination of the fracture surfaces showed limited fiber pull-out indicating a strong interfacial bonding. (4:30 p.m.) SINTERED 6061 ALUMINIUM ALLOY - PARTICLE COMPOSITES: PROCESSING AND MECHANICAL BEHAVIOUR, A.K. Jha*. S.V. Prasad*, G.S. Upadhyaya**. *RRL (CSIR), BhopaI-462 026 **IIT Kanpur - 208 016, INDIA The present paper describes the sintering characteristics and mechanical behaviour of Aluminium alloy-particle composites. Aluminium alloys of 6061 grade, was selected as the base alloy whereas solid lubricants (graphite & talc) and hard dispersoids (alumina and tiC) were selected as the dispersoids. Composites containing up to 14 vol. % of each dispersoid were prepared by the conventional P/M route involving the steps: mixing compaction Three sintering atmosphers viz. argon, nitrogen and sintering. & vacuum were used. Sintered composites were subjected to repressing - resintering and heat treatment also. Densification behaviour of the composites was monitored by evaluating the dimensional changes, porosities and densification parameter. Hardness (Brinel!) and tensile properties of the composites were measured as a function of volume fraction of dispersoid. Metallographic samples were examined using optical and scanning electron microscopes. Fractography was also carried out using SEM. (4:45 p.m.) FATIGUE-CRACK PROPAGATION IN METAL-MATRIX COMPOSITES:' R. O. Ritcbie and J.-K. Sbanll, Department of Materials Science and Mineral Engineering, University of California, Berkeley, CA 94720, USA Microstructural mecbanisms affecting fatigue-crack propagation in metalmatrix composites are reviewed, witb empbasis on bigh-strength aluminum alloys reinforced with SiC. Fatigue-crack growtb bebavior in general is found to be influenced by several mechanisms, unique to specific growth-rate regimes. In particulate-reinforced composites at near-tbreshold levels for example, tbe crack path tends to avoid the particles; tbe resulting deflection leads to enlarged fracture-surface asperities wbich promote crack closure. As crack-tip plastic zone sizes here are comparable with particle sizes, the reinforcement particles can also act as "crack traps" along the crack front. Conversely, at higher stress intensities, i.e. above _10- 9 m/cycle, tbe limited fracture of the reinforcement phase ahead of tbe crack tip can result in uncracked ligaments along tbe crack length, which act as crack bridges. Models for such crack-tip sbielding mechanisms are presented, and shown to be consistent with results in a wide range of composites. ·Supported by the U.S. Air Force Office of Scientific Research.
Annual Meeting-67
ALUMINA AND BAUXITE m: Simulation, Modeling, Control Sponsored by TMS Aluminum Committee Tuesday PM February 20, 1990
Room: Salon 3 Marriott Hotel
Session Chairmen: Kwat The, Alcan International Limited, Jonquiere, Quebec, Canada G7S 4K8 (2:30 p.m.)
SIMULATION OF BAYER DIGESTION PROCESS WITH SOLIDS & ELECTROLYTES: Yoshio Kumagae, Pree. Tech. Lab., Sumitomo Chern. Co.Ltd., Niiharna-Shi, Ehime-Ken, 792 Japan. Process simulation plays an important role in process development and retrofitting. As the success of process simulation depends heavily on the modeling method of real process, special attentions should be paid for choosing reasonable models. The feature of Bayer process in terms of modeling is the strong non-ideality of phase equilibria. To realize the good description of the phenomena, it is of great importance to treat solids and electrolytes rigorously. In this work, Chen's model was applied for the representation of boiling point rise and alumina solubility. Parameter analysis of the phase equilibria was made using experimental and plant data for simulation run. Process simulation using ASPEN was done for the digestion and evaporation process. Evaluation of the results and some applications of this model are dis,cussed.
(3:00 p.m.) MATERIAL BALANCE ANALYSIS USING SIGNAL FLOW CHART AND ITS APPLICATION IN ALUMINA PROCESSING: Hong Niu, Shen Guo and Heng-Long Li, Shenyang Alum. & Magnesium Engr. & Research lnst., 50 Block 3, Heping St., Shenyang, Liaoning, P.R.C. Material balance analysis is the fundamental work for the design, operation and control of chemical processes. We have developed a special program for alumina processes instead of using a readymade software. The signal flow chart method can conveniently describe the process studied. Based on the linearity of the problem, transition matrices are used to express the unit operations. Circle elimination is an important step to simplify the calculation. The non-linearity existed in alumina processes has been fully counted. The signal flow chart method is quite flexible. It can solve the material balance calculation for a variety of alumina processes, such as sintering process, Bayer process and combined process all of which are in popular use in China, and meet the requirement to adjust a certain process in detail. It is also a useful tool in analysising and optimizing process parameters.
(3:30 p.m.) APPLICATION OF THE PHILOSOPHY OF EXPERT SYSTEM ON RED MUD THICKENERS OVERFLOW QUALITY MONITORING: C.E. Suarez, E. Sulpizi, F. Peleato and A. Saraullo, Interalumina, Apartado 289, Puerto Ordaz, Venezuela. A practical application of the philosophy employed by EXPERT SYSTEMS is presented, which provides an effective tool to aid process control supervisors in detecting and correcting process upsets. The system, TURFREE, serves as a diagnositc guide in case of process disturbances that are closely related to quality, in terms of suspended sol ids content, of the overflow from a red mud clarifier/thickener operation, leading to rapid and logical selection of appropriate corrective actions. TURFREE is structured such that it incorporates associated knowledge gained through experience of a team of process engineers and research and development personnel at the INTERALUMINA plant. Implementation of the TURFREE system has proven successful in reducing reaction time required for the operating group to initiate appropriate control actions to rectify upsets in this unit operation and to minimize their effects on other parts of the alumina refining process. The system is also an adequate resource for training, and its modular structure permits easy updating of the knowledge base. The positive results obtained through the use of TURFREE and its supporting philo$ophy have led to consideration of other process applications using the same approach. (4:00 p.m.) DESIGN OF A BAYER IMPURITY REMOVAL PROCESS WITH ASPEN: G.D. Horne and S.C. Sharma, Alcoa of Australia, P.O. Box 161, Kwinana, Wa 6167, Australia. This paper describes the application of the ASPEN process simulation package to the in-house design of an impurity removal process in a Bayer alumina refinery. The application is introduced with a description of the process and where it fits into
the existing refinery. The company's choice of the simulation package is also explained. The development of the process model is then described, as well as problems in verification. Sensitivities were studied for various equipment configurations taking advantage of the foreground flowsheet flexibility built into the model. Case studies were also run to evaluate various operating options and to set crucial operating parameters. Major gains from the use of simulation are identified. Finally, the transition of the design model into the operating phase is discussed. (4:30 p.m.) PHYSICAL CHEMISTRY CONSIDERATIONS IN ALUMINUM HYDROXIDE PRECIPITATION: Milind V. Chaubal, Alumina Division Technology, Reynolds Metals Company, P.O. Box 9911, Corpus Christi, TX 78404. Importance of physical chemistry principles in the preCipitation of aluminum hydroxide in a Bayer alumina plant is discussed. Specifically, the roles of solution chemistry and kinetics of precipitation reaction in developing the criteria for precipitation circuit operation are delineated. Application of these criteria to develop strategies to maximize precipitation yield in both batch and continuous precipitation circuits is also addressed. (5:00 p.m.) RESIDENCE-TIME DISTRIBUTIONS IN BAYER PROCESS VESSELS. DEVELOPMENT OF A SUITABLE LIQUOR TRACER: S.C. Grocott and L. McGuiness, Research and Development Dept., Alcoa of Australia Ltd, P.O. Box 161, Kwinana, 6167, WA, Australia The flow behaviour of solutions in continuous processes exerts an enormous impact upon the efficiency of plant operation. Liquor flow patterns or residence-time distributions (RTD's) have typically been measured by using a tracer. In the Bayer process this has been severely restricted because of the aggressive chemical environment and the very high concentration of interfering substances. Alcoa of Australia Ltd has developed a family of tracer substances for use in the Bayer process. These substances (halogenated organic compounds) are stable under almost all conditions in the Bayer process. They are very similar to organic impurities already present in Bayer liquor. They do not adsorb onto solids and do not impact upon the process. They can be analyzed and detected at very low (less than 0.1 milligram per litre) concentrations. A full RTD study can be performed on a one megalitre precipitation vessel with only one kilogram of tracer material. These compounds are easily the best substances that have been developed for investigating the hydrodynamics of Bayer and other process vessels. As a demonstration of the exceptional suitability of these substances for RTD studies, examples of their use on a mud thickener, an oxalate crystalliser and a digester train are given. A description is also given of the most important areas for tracer studies in the Bayer process.
CARBON TECHNOLOGY IV Sponsored by TMS Aluminum Committee (formerly Light Metals) Tuesday PM February 20, 1990
Room: Salon 2 Marriott
Session Chairmen: Euel R. Cutshall, Manufacturing Technology Laboratory, Reynolds Metals Company, Sheffield, AL 35660; Julian V. Copenhaver, National Southwire Aluminum Company, P.O. Box 500, Hawesville, KY 42348 (2:30 p.m.)
OPTIMIZATION OF COKE CALCINING KILN OPERATION USING MATIlEMATICAL MODELLING, J. Perron. T. H. Nguyen, Ecole Polytechnique, Dtpanement de gtnie m&:anique, P. O. Box 6JJT9, Succ. A, Montrtal, Quebec, Canada, H3C 3A7; Use Castonguay, Alcan International Umited, Arvida Research & Development Centre, P. O. Box 1250, Jonquiere, Qutbec, Canada, G7S 41<8. Petroleum coke used for the production of anodes is usually calcined in rotary kilns. The calCination process involves a number of different parameten which render the optimization of the calciner operation through plant trials very complicated and costly. A mathematical model was developed to simulate the operation of the coke calcining kilns. The model was used to optimize the operation of the kiln in terms of improved calcined coke quality.
68-Annual Meeting (3:00 p.m.)
COKE BULK DENSITY MEASUREMENT USING A NUCLEAR GAGE, R. w. Peterson, T. R. Hornack, Aluminum Company of America, Alcoa Center, Pennsylvania
15069, USA Coke bulk density is an important measure of coke quality. Measurement is conventionally done manually by vibrating a small sample in a graduated cylinder. The device described here utilized a 15·liter sample of coke in an aluminum cylinder. The cylinder was rotated on a screw so that the entire contents of the cylinder could be scanned by a nuclear gage. Agreement with densities calculated from weight and volume measurements was good. (3:30 p.m.)
CAPIllARY AND ClASSICAL DIIATOMETRY OF GRANUIAR AND COMPOSITE CARBON, T. Foosna:s, Hydro Aluminium AS., R&D Carbon, N·5875 Ardalstangen, Norway; K. Grjotheim, K. R. Kvam, University of Oslo, N.Q315 Oslo 3, Norway; T.
Naterstad, Sales of Technology, N·1321, Oslo, Norway. Two methods are described for measuring the coefficient of thermal expansion (CTE) of petrol coke. The capillary method developed in this work makes use of push·rod
dilatometers to measure the thermal expansion of petrol coke grains in stacks of 70 mm height Push-rod dilatometers were also used in the 'classical' method to measure the CTE of core samples from pilot~scale prebaked anodes. The variation of the CTE of petrol coke grains with temperature differs from that of anode samples. The CTE of petrol coke grains in the size range 5 to 12 mm was found to have the same average value for the entire size range, 3.0 x 10""'OC1. For the Iaboratory·made samples, the average CTE for the different size fractions varied between 3.4 and 3.8 x 1O"""C-1. Anode samples showed increasing thermal expansion with increasing bulk density. The consistently higher expansion of the composite compared to the aggregate samples indicates that local expansion into inter·particle voids is partly restricted by the binder coke. Average CTE values for all samples gave the following values: Petrol coke 3.0 x 10"",·CI, pilot scale anodes 3.6 x 1O"""CI, industrial prebaked anodes 4.4 x 1O"""C1. (4:00 p.m.)
PETROGRAPHIC EVALUATION OF CALCINED COKE BY IMAGE ANALYSIS TECIINIQUE, Jacques Boutin, Lise Castonguay, Alcan International Limited, Arvida
Research & Development Centre, P. O. Box 1250, Jonqui~re, Qu~bec, Canada G7S 4K8. Calcined coke properties are of a prime importance to ensure the quality and the perfomtance of anodes used by the aluminium industries. It is known that the coke properties are affected by its microstructure. The actual optical microscopic method for the determination of coke microstructure is tedious and unstable statistically. A method was developed to perform petrographic evaluation of calcined coke by a fully automatic intage analysis technique. More than one hundred calcined coke specimens were sampled from different sources and were investigated by both optical microscopic evaluation and image analysis technique. Results were correlated with other coke properties.
CAST SHOP TECHNOLOGY IV: Metal Treatment - Filtration and Recycling Al-Li Sponsored by the TMS Aluminum Committee Tuesday PM February 20, 1990
Room: Salon 4 & 5 Marriott
Session Chainnan: D. Apelian, College of Engineering, Drexel University, Philadelphia, PA 19104
(2:50 p.m.) A PROPERTY CHARACfERIZATON OF COMMERCIAL FILTER MEDIA AND TIlE RELATIONSHIP TO PERFORMANCE: D Apelian, R. Mutharasan, College of Engineering, Drexel University, Philadelphia, PA 19104; C. E. Eckert, Alcoa Laboratories, Alcoa Center, PA 15069; C. Schade, College of Engineering, Drexel University, Philadelphia, PA 19104. The prudent selection of filter media is germane to the effective, reliable and economical filtration of aluminum alloys. In this investigation, a property characterization of selected commercial reticulated foam and rigid granular filter media was performed that includes macroporosimetry, bulk density, pore topography and structural modulus. Since the optimum selection of filter media is highly situational, a property/performance discussion is provided for several filtration scenarios. (3: 10 p.m.)
A PRACTICAL GUIDE ON HOW TO OPTIMIZE CERAMIC FOAM FILTER PERFORMANCE: J. E. Dore, SELEE Corporation, 700 Shepherd Street, Hendersonville, NC. Ceramic foam filters of SELEE(R) structure represent stateof-the-art technology for removing inclusions from molton aluminum alloys. They are effective and at the same time inexpensive and easy to use. This paper presents a series of practical suggestions on how to obtain optimum performance with ceramic foam filters and hence improved product quality. The importance of all phases of filter use are discussed including filter bowl design and construction, selecting proper filter area and pore size, the filter gasket and its role, preheating, priming, steady state operation and melt sampling. Examples are given on how improper practices can adversely affect melt cleanlin.ess and product quality.
(3:30 p.m.) UPDATE ON THE FILTRATION OF ALUMINUM ALLOYS WITH FINE PORE CERAMIC FOAM: D. B. Rogers, P. A. Bosomworth, F. B. Tadayon; SELEE Corporation, 700 Shepherd Street, Hendersonville, NC. The increased demands for metallurgically clean aluminum have placed more and more stringent requirements on liquid metal processing. Filtration using ceramic foam filters of the SELEE(R) structure provides positive elimination of nonmetallic inclusions. The use of ceramic foam filters is discussed and product improvement for several sheet products is reviewed. The continually increasing quality demands on metal cleanliness requires finer pore filters than have been classically used. The use of No. 50 and No. 60 pore filters in the production of clean metal is described.
(3:50 p.m.)
BREAK
(4:00 p.m.)
MEASUREMENT OF FIL1'RA'l.'ION EFFICIENCY OF AN ALCOA 528 FILTER: F.Frisvold, SINTEF, N-7034 TRONDHEIM, Norway, T.A.Engh, Metallurgisk inst., NTH, N-7034 TRONDHEIM, Norway, E.Bathen, Microscan, Tonstadgrenda 23, N-7075 TILLER, Norway. Samples were taken before and after an Alcoa 528 filter 'I'he samples were analysed for carbide and oxide inclusions and oxide films. Also hydrogen was measured. The number size distribution of the films was determined metallographically and carbide distribution was obtained by automatic image analysis. The distribution as a function of depth was also found. The distributions were decreasing exponential functions of size. This exponential size distribution can be given a theoretical justification. The results indicate that removal of oxide particles is better than for films. Also the presence of chlorides (C1 -gas) improves 2 removal of carbides.
(4:20 p.m.) (2:30 p.m.)
CAKE BEHAVIOR IN TIlE FILTRATION OF ALUMINUM ALLOYS:
R Mutharasan, College of Engineering, Drexel University, Philadelphia, PA 19104; C. E. Eckert, Alcoa Laboratories, Alcoa Center, PA 15069; D. Apelian, C. Schade, College of Engineering, Drexel University, Philadelphia, PA 19104.
It was previously shown that both incompressible and compressible cakes can exist in the cake mode filtration of aluminum alloys. Although a strong dependence of alloy type and upstream metal processing history on cake type was exhibited, the specific effects of inclusion morphology/size distribution were not mechanistically understood. Cake stability was also ambiguous.
In this investigation, aluminum melts were carefully contaminated with various inclusions of known morphology and size. Isobaric filtration experiments were then conducted to generate temporal flow rate data, that subsequently yielded cake characterization, stability and overall flow resistance information. Inclusion morphology, concentration and size distribution, in combination with filter media variables, were shown to strongly influence these parameters. The enhanced capture of small «1511) ideomorphic inclusions brought about by film morphology oxides was also experimentally demonstrated and discussed.
RECYCLING OF ALUMINUM-LITIIIUM SCRAP: T J Robare. R. S. James, G. M. Kallmeyer, R. H. Keenan, Alcoa Laboratories, Alcoa Center, PA 15069. Aluminum-lithium alloys are intportant strategic materials in future aerospace and aircraft development. These alloys offer weight reduction due to lower density and provide intproved mechanical properties. The usage of Al-Li alloys is projected to grow significantly in the next five years as airframe designe~s search for increased payloads, longer range and/or faster, more maneuverable aIrcraft. The introduction of Al-Li alloys into the common recycling loop for aircraft alloys can create several problems for the recyclers. Strict segregation of these alloys is viewed as cost-prohibitive by most major airframe manufacturers. Even small contamination of the normal scrap with Al-Li alloys may cause problems in the areas of (1) metallurgical property alterations in the secondary aluminum alloys, (2) refractory life in common reverbatory melting furnaces, (3) higher melt losses to skim, (4) safety and (5) environmental implications. This paper will address each of these problem areas and will discuss recycling methodologies to minimize the impact of Al-Li alloy contamination due to the lack of segregation. These methodologies include dilution with subsequent chlorine fluxing, removing lithium in a vacuum, transport cells to recover lithium and salting out the lithium using solid salts.
Annual Meeting-69 (4:40p.m.) LITIIIUM RECOVERY FROM ALUMINUM-LITHIUM SCRAP: R. A. Christini, Fabricating Technology Division and M L Weaver. Chemical Systems Division, Alcoa Laboratories, Alcoa Center, PA 15069 An experimental program was undertaken to evaluate the electrochemical separation of lithium from AI-Li alloys. A lithium transport cell, similar to a Hoopes cell, was developed. The sensitivity of the process to various operating parameters was detennined. Preliminary experiments were conducted in two different types of bench cells. The results of this work led to the construction and successful operation of a pilot scale system for lithium recovery. An optimum electrolyte composition was detennined that maximized lithium recovery and coalescence. Highest purities and recoveries of lithium were obtained with anode current densities between I and 6 amp/in.2 The largest contribution to total cell operating voltages was due to the electrical resistance of the electrolyte. Prefluxing of the electrolyte with HCI gas removed moisture and decreased the rate of sludge formation. Lithium recoveries of greater than 95% of the lithium transported at purities greater than 99% lithium were obtained through careful process control.
EXPERT SYSTEM APPLICATIONS IN MATER· IALS PROCESSING & MANUFACTURING II Sponsored by the TMS Shaping & Forming Committee, the TMS Synthesis & Analysis in Materials Processing Committee and the ASM-MSD Computer Simulation Activity Tuesday PM February 20, 1990
Room: ParkB Inn at the Park
Session Chainnen: N. Karam, Spire Corporation, Patriots Park, Bedford, MA 01730; P. MacNeille, Research Staff, Ford Motor Co., Dearborn, MI 48121 (2:00 p.m.)
DESIGNING AN EXPERT CONTROL SYSTEM FOR THE PRODUCTION OF SILICON CARBIDE WHISKERS: W. J. Parkinson, P. D. Shalek, E. J. Peterson, Los Alamos National Lahoratory, Los Alamos, NM 87545, G. F. Luger, Department of Computer Science, University of New Mexico Albuquerque, NM 87131 Silicon carbide whiskers, a very strong material produced primarily for strengthening ceramics and metals, are being considered for various commercial uses. Whisker production is a semi-batch process that is difficult to model mathematically, so rules accumulated from experience are used to set up and run the process. Several sets of process conditions are possible. When the correct set is chosen, the proper whisker for the desired use is produced. We designed a two-phase, PC-based expert system to help inexperienced users. In phase one, an expert consultant provides users with information that enables them to set up the run. This information is incorporated into the rule base, which makes up the second phase, the control system. Because this project is laboratory scale, no automatic controls were added to our system. Instead, we use a human controller: the operator. The operator asks the expert system if the process is behaving correctly. The expert system makes the decision and suggests any corrections the operator should make. With this system design, the automatic controls can be added later. (2:30 p.m.) ALLOY MELTING EXPERT SYSTEM: R.G. Trimberger, GE Co. CR&D, P.O. Box 8, K1-MB-139, River Road, Schenectady, NY 12301, B.R. Hathaway, GE Consulting Services, CR&D, P.O. Box 8, K1-4C9A, River Road, Schenectady, NY 12301 General Electric's Corporate Research and Development Center melts over 1200 experimental alloys a year. The Alloy Melting Expert System (AMES) is a knowledge-based expert system that is used to select the furnace, crucible, mold(s), and raw . materials and to generate specific melting instructions for each alloy. Mold, furnace, and crucible selection are each based on a number of factors such as reactivity, thermal characteristics, physical chemistry, and experience with similar type alloys. After evaluating all of the factors an ordered list is generated with the candidates molds, furnaces, and crucibles ranked according to suitability. Raw materials are selected primarily on the basis of purity, cost, volatility, and reactivity. Melting instructions are generated based on deoxidation requirements, alloy composition, type and form of the raw materials and vapor pressure of the raw materials. AMES consists of nine modules and five databases. 0:00 p.m.)
DEVELOPING AN EXPERT SYSTEM FOR INDUSTRY: E.D. Jamieson, Lindberg Algonquin Rd., Rol11ng Meadows, IL As computer based control and
THE HEAT TREATING Heat Treating, 3601 60008. manaqement systems
become more integrated into heat treating operations, the development of expert systems becomes more feasible. While expert systems reside in all manufacturing operations with eXI?erienced personnel, the ability to access that 1nformation remains limited. L1ndberg Heat Treating has begun to develop an expert system based on experienced personnel and process control signals. The system wil~ be designed to be interactive with shop personnel, engineers and managers. The program will address the development of expert systems for real time process control based on computer modeling and process signals. It will also address process troubleshooting based on decision-trees developed for SPC programs. Historical process parameter data will be imputed to the aecision-tree from the computer based control unit to aid in decision making. Equipment maintenance and troubleshooting are also part of the program. In addition, a training program using tutorial and simUlation techniques will be developed to aid shop personnel in utilizing the expert system. 0:30 p.m.)
ANALYTICAL FURNACE AND CLOSED-LOOP FEEDBACK CONTROLLED SINTERING J. T. Strauss, and R. M. German, Materials Engineering Department. Rensselaer Polytechnic Institute, Troy, NY 12180-3590. A high temperature batch furnace has been constructed for real-time analysis of sample weight (TGA),'.temperature profile (DTA), and effluent gas composition using various gas analytical methods. (FTIR, oxygen and hygrometer sensors). A separate precision dialatometer is used to provide dimensional change date. These analytical capabilities are controlled by a microprocessor. The long-range goal is for the computer to fully govern the thermal/atmosphere cycle in a closed-loop feedback mode that will allow optimization of the sintering pathway (time/temperature/atmosphere) for any component. Early results on iron powder samples indicate a complex series of interactions between the powder and atmosphere during heating. The presentation will include an update on the system. its design, software development, and early experiments with injection molded compacts. This research is supported by the New York State Center for Advanced Technology at Rensselaer. (4:00 p.m.)
WELDEXCELL: A BLACKBOARD SYSTEM FOR OFF-LINE INTELLIGENT PLANNING AND CONTROL OF ROBOTIC WELDING .Im.ld..E.. Jones Ph D; Senior Scientist; American Welding Institute; 10628 Dutchtown Road; Knoxville, TN 37932; USA. Dawn R. White, Ph.D.; Director of Program Development; Advanced Technology Development Division; MTS Systems Corporation; 14000 Technology Drive; Minneapolis, MN 55344 The joining of metals into fabricated components and structures is a difficult task. The most common method of joining metals is welding, but the welding process is complex and requires several important steps to be performed in a carefully integrated manner. Although the process may seem SImple to an experienced welding engineer, when analyzed tn sufficient detail, the engineerinlllplanning processes are extremely complex. It is clear that the type of distributed' problemsolving in multiple knowledge domains involved in~this multi-disciphnary engineenng problem cannot be addressed using a single knowledge source. The problem solving technique being applied is a blackboard expert integration environment. The main blackboard IS functionally divided into five sub-blackboards: Joint Designer; Structural Integrity Analyzer, Material Analyzer; Procedure Specifier; Path Planner. Each of the sub-blackboards consists of several expert systems and databases. This Welding Job Planner (WJP) is then integrated with an advanced robotic Welding Job Controller (WJC) to form the two pans of this WELDing EXpert manufacturing CELL (WELDEXCELL). The project is sponsored by the Navy Manufacturing Technology Program. (4:30 p.m.)
A Gallium Arsenide Meta-Controller Paul Nielsen, Vivek Badami, and Jim Comly General Electric Corporate Research and Development P.O. Box 8, Schenectady, NY 12345 We describe a supervisory controller that incorporates both a procedural and rule-based language capable of responding to asynchronous real-time sensor interrupts. This controller uses an English-like syntax that allows expressing process actions in both a sequential and exception handling mode. The notion of time is explicitly incorporated into the syntax of the language. This system, referred to as a Meta-Controller (MC), has been used to implement a supervisory control system for the growth of Gallium Arsenide crystals using the Liquid Encapsulated Czochralski process. The MC provides for consistent, repeatable growth of boules of Gallium Arsenide. It automates such process operator functions as monitoring and data logging. Through the use of actuators, it changes control loop set points and control modes and downloads new gain schedules to a digital controller based on changing conditions or timed events. It handles faults during the process run through recovery or system shutdown. This paper focuses on both the architecture for supervisory control of materials processing and an implementation on a physicaJ puller using artificial intelligence techniques. It also describes the special demands of real time performance.
70---Annual Meeting
mGH TEMPERATURE MATERIALS IT: Modeling of High Temperature Materials Sponsored by the TMS High Temperature Alloys Committee Room: Plaza B Inn at the Park
Tuesday PM February 20, 1990
Session Chairmen: Richard G. Menzies, GE Aircraft Engines, Mail Drop M85, Cincinnati, OR 45215; Michael V. Nathal, NASA Lewis Research Center, MS 49-3, 21000 Brookpark Road., Cleveland, OR 44135 (2 :00 p.m.)
The pa r ti t ion coeff ic ients of tantalum, titaniu~, and alumin ~m I,ere observed t o be 1es s than uni ty ; wh i 1e th at for tungsten 3n~ cocc l t were greats, than un ity . TGis is qual i tati vel y si ~i l ar to their ni ckel base binari es. The partiti on coefficients ~er! temperature i ndependent in th e ran;e from 1534 to 1608 K. Micr~segre g ation ~rofiles compare well wtth predictions from
anal yt i cal models develo ped for binary al l oys.
(3:30 p.m.) FLUlD FLQ/ VlSUALlZlTIOI AJID IIlDELDIG OF IIIERT GAS A'l'CIIlZlTIOli
S. D. Ridder, P.I. Espino aDd F.S. Biancaniello The Supersonic ioert Gas !letal Ataoizatioo (SiGIIA) tecboique produces ultra-fine (
gas atomizatioo. llatheolatical .:xIeliDg, pitot tobes aDd schlieren pbotograpby have beeo used to study the supersonic flow of tbe Si GIIA gas jets, whI.le high ~ (10,000 fps) cineaatograpby aDd 2S ns pulse duratioo bolograpby were used to study the extr....ly chaotic gas-liquid flow in the atCDiziDg spray pllDe. Results iDclude a graphical IIKldeliDg tool usef ul for designiDg aDd optimiziDg gas jet atomizers (or spray foraincJ systems), as well as dramatic still and III)tion pictures of droplet formatioo duriDg biQb t ellpE!rature Eta! atcaizatioc .
THE ROLE OF IIlDELIIIG Of DIVES'I!!EIT CASTIIIG L.D. Gr_
pce Airfoils, !DC. , 23555 Euclid Avenue, ",5 32, Cl eveland, OH 44.117
Ficite Eleaoent Tbenol IIodeling is beiDg used OIl • productioo DasU for tbe design of gatiDg systeaos aDd tbe selectioo of castiDg par_ters. Predicti v• .,&ls are used for llicroshrink., MCI'OShriak, bot tears aod grain size. Casting process da.sigo caD be devel~ for a . - part whI.le tbe t ooling u beiDg built. ODe of tbe advan~ of !!Oleling u tbe speed of • siIIllatioo i teratiOil. A key aspect of 0Ddeling u tbe ability to select a proceas that penli.ts .all YariaUOIlS in the proc:eas without produciDg defective castiDga. DuriDg tbe last few yeatS, tbe ability to hoild aDd run !!Olels bas increased so tbet tbe tiDe aDd cost bas been significantly reduced.
11111(2 :30 p.m.) VlSUALUATIOli '!OOLS FOR
IIl!lELOO
OF SIlIGLE CRYSTAL
C1STII!iS
A.F. GiaEi and D. E. Edwards, United Tec:bnologies Research Center Silver Lane, East Hartford, cr 06108
"nle rate of advaDoe in ~ter hardware aod software capability over the past five years bas been truly ~ve. ~lex _ t ri es aDd prtlCe$S88 """ DOlI be !!Oleled io reasoaable tiEs. '!'be probl_ re.ins, bowver I in tl:1mai.tting the volllliDous infoIWllltiOO to the eod user i.o a fom whicb is infOI1llltive aod efficient. 'l'be type of infomatioo generated is ~ture in three diaenai.aDal space 118 it varies with tiae aDd also certaio reduced quantities such as tbenIU gradient, growth rate aDd cooling rat.. These latter quanti ties can be IIIIIlipulated to be C
crystal superalloy castiDga.
(2 :50 p.m . )
SOLIDIFICATIOIf Jll)l)tLIlfG OF SDIGLE CRYSTAL IIIVES"Il!!lI'I CASTIlIGS
1.0. Yu, M.J. Beffel, M. Robinson, D. Goettsch, B. 'l'baIas, D. Pinella, and i.G. Carlson
SiDgle-crystal superalloys provide tbe hi\lbest ..tal t-""ture COI'abllity for gas tubine airfoils . "rbese parts are -.ofactured by tbe lmidirectiaool. vi tbdraval 9Olidificatioo ...tbod. !be final castiDg iDtegrity depeD:ls upoo their structures as well as i.nbereDt defects, which are detemLoed duriDg solidificatioo. Efforts were undertaken in the UJft'E0i p~ and at PCC AirfoUs to correlate the solidification CODditioos aod tbe llicrostructores aDd defects of siDgle-crystal castiDga. The approach caobines ficit....leaoent tbenIU analysis aDd experilleotal results. Clusters of instruEnted cast cylinders with various ~ters, core sizes, core .terials and sbrood sizes were fatEd at specific furnace t""""""tures aDd vit.bdraOlal rates. A ficite-elOllOllt !!Olel, devel~ to analyze t.belIIIIIl CODditioos duriDg castiDg, geDOrlIted data 00 oool.ing rates , local solidi.ficatiOD tiEs, tbema.l gradieot (G), and solidificatioo veloci ty (R). Results shoo that the ficite-el....,t tbenIal analysu can accurately -=del tbe casting process with reuooable ~er nm tilles. Iklreover, correlations of casting microstructures and defects with calculated solidification CODditioos indicates that ficite-elOllOllt !!OleliDg is a powerful tool for predictiDg llicrostructures aDd defects in single-crystal investment castiDga.
(3: 10 p. m.)
MICROSEGREGATION IN DIRECTIONALLY SOLIDIFIED PWA-1480 SINGLE CRYSTAL NICKEL BASE SUPERALLOY: S. N. Tewari, M. Vijayakumar, Cleveland State UniverSity, Cleveland, OH 44115, J. E. Lee, P. A. Curreri, NASA Marshall Space Fl ight Center, Huntsville, AL 35812. Microsegregation behavior in a multicomponent nickel base superalloy, PWA-1480, has been investigated. Solutal profiles in single crystal specimens, which were partially directionally solidified and quenched, have been examined on several trans ·.~:- " ~ dendrites. Isoconcentration contours across the primary dendrites were used to identify the most appropriate method to obtain the relationship betlteen fraction solid and the correspondi ng solute concentration at the liquid-solid interface.
(3 :50 p.m.) THE DESIGN OF COMPONENTS TO REDUCE THE RISK OF QUENCH CRACKING, R. A. Wall i s, Cameron Forge Company, Houston, Texas 77251. Designers of components for aircraft engines are continually seeking improved properties from alloys. One important variable that determines the strength of many superalloys is the cooling rate from its solutioning temperature. The desired cooling rate is often obtained by quenching the component. Heat transfer during the quenching process is highly transient and can produce extremely large thermal gradients in the component being quenched. If the temperature gradients are too high, the resulting stresses may cause the part to crack . Process models may be use~ to predict the temperature and stresses in parts during the quenching operation. These models, while in their infancy, can be used to design a component shape, and a process, which reduces the risk of quench cracking. Several practical examples will be presented. (4: 10 p.m.) RESIDUAL STRESSES IN FCRGllGS AND THEIR DISTORTIOOS DURIN:; SUBSE-
ELE}lENl' APPROACll: U. Chandra, GE Aircraft Engines, 1 Neurrann Way MD H-85, Cincinnati, Ohio 45215.
QUEm' MAOUNIN3 - A FINITE
Forgings of expensive superalloys are often found to contain large residual stresses inspite of aging. Unacceptable distortions may oc= when such forgings are being machined to their net shapes. Understanding the reasons for the formation of residual stresses and machining distortions is basic to developing any successful techniques of avoiding them. This paper presents the following; (a) the basic concepts (applicable to any manufacturing process) of residual stress generation and machining distortions, (b) specialization of these concepts to the forging process, and (c) finite elen-ent analyses of sane sirrple geanetries. The finite elen-ent ~k addresses all three parts of the problem, i.e. formation of residual stresses, their relief due to aging and, finally, distortions during machining. It utilizes proper non-linear, transient thermal as well as stress/displacerrent analyses. The focus of the paper is on basics, and mathematical details are kept to a minimum.
(4:30 p.m.) A MATHEMATICAL MODEL OF THE WIDE GAP BRAZE REPAIR PROCESS: Ernesto Gutierrez-Miravete, The Hartford Graduate Center, 275 wind sor St., Ha rtford, CT 06120, and No n" Pietruska, UTC/ Pratt &Whitney, 400 Main St . , East Hartford, CT 06108. Small cracks in turbine vanes are routinely repaired by the wide gap brazing process. The brazing paste is placed on the cleaned cracked part and the entire component is then heated under a protective atmosphere . At temperature, the paste melts and the boron contained in it slowly diffuses into the side walls of the crack. Bya process of isothermal solidification a repaired component with excellent properties can be obtained. A mathematical model of the isothermal solidification process during wide gap brazing of cobalt base superalloys has been developed . The model provides information on the boron concentration profile as well as on the rate of solidification and repair of the crack. It can be used to assist the design of optimal heat treatment cycles for crack repair .
Annual Meeting-71
HIGH TEMPERATURE SUPERCONDUCTING COMPOUNDS II: Bulk A:. Synthesis/Oxygen Dissolution Sponsored by the TMS Superconducting Materials Committee Tuesday PM February 20, 1990
Room: Orangewood I Quality Hotel
Session Chairmen: Dave Sharma, EPRI, 3412 Hillview Ave., Palo Alto, CA 94303; Paul Berdahl, Lawrence Berkeley Laboratory, Berkeley, CA 94720 (2:00 p.m.) WEAK LINKS IN HIGH T, SUPERCONDUCTORS, Dayjd Larba1estier, Susan Babcock, Xue-Yu Cal, Manfred Daewnling, Damian Hampshire, and Jeff Seunt;jens, Applied Superconductivity Center, University of Wisconsin, 1500 Johnson Drive, Madison, WI 53706. Raising the transport critical current density (J,,) of HTSC remains one of the most desirable goals of HTSC research. In our group we have
tried to develop an integrated view of the weak link problem by combining electromagnetic, microstructural and fabrication studies. In this paper we will summarize recent work concerning bulk and single crystal studies of 123 compounds, melt textured 123 and BSCCO wires. In particular we will discuss the apparent reduction in weak link density observed in recently made wires and melt-textured samples.
(2:25 p .m.) CONTACTS FOR Y-, Bi-, AND TI-BASED HIGH-Tc SUPERCONDUCTORS, J. W. Ekin, National Institute of Standards and Technology, 325 Broadway, Mail Code 724.05, Boulder, Colorado 80303
(3:30 p.m.)
BREAK
(3:50 p.m.) THE EFFECT OF ANNEALING ENVIRONMENT ON THE SUPERCONDUCTIVE PROPERTIES OF PLASMA SPRAYED YBaZCu30x COATINGS: H. Wang, B. Gudmundsson, R. A. Neiser, H. Herman, SUNY-Stony Brook, NY 11794-2275, M. Suenaga, D.O. Welch, Brookhaven National Laboratory, Upton, NY 11973. Plasma spraying is a deposition method with the potential of producing large quantities of high temperature superconductive coatings. The as-sprayed YBa2Cu30, coatings are semiconducting tetragonal X-phase (a=3.87Ac-11.61A). Heat-treatment of this phase at 930°C/24h results in a poorly superconducting orthorhombic phase. However, this study shows that the superconductive properties can be improved by heat-treating the as-sprayed coatings in argon above 850°C prior to the 930°C oxygen anneal. The argon pre-annealing causes the X-phase to be transformed to Y2BaCu05' YBaZCu306, CuO, and BaO. These phases will transform into a good orthorhombic superconducting phase during the heat-treatment at 930'C. This research was sponsored by the Department of Energy contract no . DE-AC02-76CH00016.
(4 :05 p.m.)
°
PREPARATION AND CHARACTERIZATION OF POLYCRYSTALLIC YBa Cu *: B K Hj11jams, J. Z 4 D.M. Kroeger, R.A. Padgeet and J . ~. Scarbro ugh , 'bak Ridge National Laboratory, Oak Ridge, Tennessee 37831-6116
Srynestad,~
Dense polycrystallic samples of the compound YSa cu 0 were prepared by a three z 8 step procedure. In the first step. a mixture of Y2 3' Ba0 and CuD was converted 0 2 to YSa Cu by heating to 960 C under an oxygen pressure of 55 atm. The material did nol sfn~er under these conditions. In the second step. densification was pro~ 0 duced by hot isostatic pressing at 875 C. This degraded the superconducting properties severely, and a third high tenperature anneal in pressurized oxygen was required to restore optiFnlJ'l1 properties . Samples were characterized by measurenen· ts of the critical temperature. critical curuent density. and normal state electrical resistivity. Auger spectroscopy was used to define the composition of grain boundary regions .
e
°
*Research sponsored by the Office of Energy Storage and Distribution, Conservation and Rer.-1wable Energy, U.S. Department of Energy, under contract
DE-AC05-S40RZI400 with Martin Marlett. Energy Systel1s, Inc. -Chemistry Division, oak Ridge National Laboratory, Oak Ridge, Tennessee.
6 8 A reduc tion of 10 to 10 in contact surface resistivities have been achieved using noble metal contacts to Y-, Bi-, and Tlbased high-T superconductors. Auger microprobe depth profiling of the contagt interface and the V-I characteristics distinguish
(4:20 p.m.)
the favorable interfacial chemistry for silver and gold contacts
Solbd state reaction of appropriate a"",unts of YZO j , BaCO j , and CuD for Zh at SOD C In flowing oxyaen with a total pressure of 2Iiin Hg, followed by cooling and annealing at 750 C for lOh in oxygen at ambient pressure, has produced orthorhombic YBa Cu 0 ("124") as the ... in phase with YBalU30 ( ' IZ3") as a 2 4S impurity phase. The phase purity improved considerably and ph:se-pure lZ3 was obtained upon annealing the as-calcined powder in oxygen at BOOOe for 24h under ambient pressure. Magnetization neasurements perfonned using a low-field d.c SQUID magnetometer showed superconductlng transition temperature of BOK. No diamagnetic behavior observed at temperature above SDK, indicating that lZ3 was not present as an impurity in the samples synthesized by this method.
compared witn indium and other metals with a higher oxygen affinity. Methods of forming noble metal contacts in both bulk and thin-film hi gh-T
c
superconduct ors are discussed.
(2 : 50 p.m.) THERMALLY ACTIVATED FLUX CREEP AND CRITICAL CURRENT DENSITIES IN HIGH TEMPERATURE SUPERCONDUCTORS, Teruo Matsushita, Department
University.
of
ElectroniC,
Faculty
of
Engineering,
Kyushi
6-10-1 Hakozaki, Higashi-ku, Fukuoka 812, Japan.
Thermally activated flux creep in high temperature superconductors seriously affects their critical current charact e ristics. The effect of flux creep becomes more
significant when the superconductors are used at higher fields
and/or higher temperatures. In principle, this effect can be reduced by increasing the flux pinning strength in the superconductors. Our largest interests are (1) the possibilit y of finite c ritical current densi t y even at high fields and 77K and (ii) the possibility of persist e nt current that can hold in the superconductors for sufficiently long time. These points are discussed from the aspect of theoretical expectation of attainable pinning strength. It is shown that the first bar is easy to clear.
SYNTHESIS OF SUPERCONDUCTING PHASE-PURE YSalu 0 : U. Salachandran, M.E. Sisnek, 4S G.W. Tomlins, B.W. Veal, and R.B. Poeppel, Argonne National Laboratory, Argonne, I L 60439
Work supported by the U.S. Dept. of Energy, Offeces of Energy Storage and Distribution, Conservation and Renewable Energy (UB. RBP); Basic Energy
Sciences, Materials Science (GWT, SWV). under contract W-31-109-Eng-3S; National Science Foundation, Office of Science and Technology Center under contract 10MB-
8809854 (HEB); and partial support from the 01visfon of Educational Prograllls,
Argonne National Laboratory (MES, GWT).
(4 :35 p.m.) THE NATURE OF OXYGEN IN OXIDE SUPERCONDUCTORS: N. A. Gokcen, Bureau of Mines, U.S. Department of the Interior, Albany, OR 97321, C.-H. Hsu, E.I. duPont de Nemours and Co., Fibers Department, Wilmington, DE 19 880 -0302.
(3 : 15 p.m.) SYNTHESIS AND PROPERTIES OF POLYCRYSTALLINE T1ZBaZCaCu20x and TlzBazCazCu30x: Gerald J. Vogt and David S. PhilIps, Materials Science & Technology Division, Ceramics Science & Technology Group, Los Alamos National Laboratory, Ma il Stop G77l, Los Alamos, NM 87545.
The di ssoluti on process for gaseous oxygen in YBa2Cu306+2r is 0 .5
Nearly single phase samples of TllBalCaCulOx and TllBa2Ca2Cu30x were prepared by a solid state reaction of T1203 with BaCuOl, CalCu03 and CuO. Starting compositions of Tll,lBaCaCuZOx and T12,ZBaZCa2Cu30x were fired in quasi-tight reaction vessels for durations of lZ to 48 hours in flowing oxygen. Samples fired at a temperature below 870'C produced the 22lZ phase, while samples above 870'C yielded the 2223 phase. Pressed samples treated in a single firing step produced lowdensity « 50% theoretical), porous material. Samples prepared by two firing steps with an intermediate regrinding yielded higher density material (>70% theoretical)_ The synthesis procedure with the quasi-tight reaction vessel will be described, showing minimal thallium loss during firin g for 12 to 48 hours. The
type c rystal structure is essential and (b) at least one element capable of forming a peroxide or superoxide is necessary to
characterization results by x-ray diffraction, magnetic suscept-
ibility , electron microscopy, and reSistivity will be presented.
02(g) - [0], and r is the fraction of sites occupied by oxygen on the crystal basal planes (r<0.6). The corresponding statistical thermodynamic equation is RTlnpo. 5 - -88,300 + 56.55' T -(75,ZOO - Z18.3 T)r + RTln[r/(l -r)]. where R - 8.3144 J/mol·K. T is the temperature in K. and P is the oxygen pressure in atm . Analyses of existing data suggest that (a) a layeredprovide the -0-0- chains.
The presence of copper is effective
but it must be tentati ve ly regarded as nonessential if new superconductors with high cri tical t e mpe ratures were to be
discovered .
A possible route for this purpose might rest on an
analysis of the d·orbitals of the transition elements.
(4:50 p.m.) ELECTROOIEMICAL INSERTION
OF OXYGEN INTO HIGH Tc SUPERCCNDUC'I'ORS
J _ MacManus, D. J. Fray and J. E_ Evetts. Dept. of Materials
Science and Metallurgy, University of Cambridge, Cambridge. By interfacing YIBa2Cu307-x with solid state ionic conductors, it has been shown that both qualitatively, by continuous x-ray diffractometry and resistive measurements, and quantitatively,
72-Annual Meeting by X-ray step scanning and Guinier X-ray analysis, that oxygen can be titrated both into and out of YjBa2Cu307-x' Use of a mixture of CU20/CU as a source of oxygen has enabled the oxygen activity or partial pressure to be measured within the YjBa2CU307_x and has allowed oxygen transfer to take place totally in the solid state. Diffusion coeffiCients, as a function of stoichiometry and oxygen content, have been determined by monitoring the change in pctential with time.
HIGH TEMPERATURE SUPERCONDUCTING COMPOUNDS m: Thin Film A Sponsored by the TMS Superconducting Materials Committee Tuesday PM February 20, 1990
Room: Orangewood IT Quality Hotel
Session Chainnen: K.E. Gray, Materials Science Division, Argonne National Laboratory, Argonne, IL 60439; H.S. Kwok, State University of New York at Buffalo, Bonner Hall, Buffalo, NY 14260 (2:00 p.m.) QUALITY THIN FILMS: A MUST FOR THE STUDY OF BASIC MECHANISMS AND FOR ADVANCING SUPERCONDUCTING ELECTRONICS: T. VENKATESAN, BELLCORE, RED BANK, NJ 07701 The major challenges In measuring Intrinsic properties in the high temperature superconductors (HTSC) as well as for the progress of HT Superconducting electronics are predominantly materials problems. Hence, a reliable source of single crystalline films would not only be essential for advancing the technology of superconducting electronics but also for increasing our understandlnll of the fundamental mechanisms of superconductivity. How far are the films from singlecrystallinity? In this talk I will summarize the progress to date. (2 : 25 p.m.) ON THE NATIJRE OF IN-SITU SUPERCONDUcnNG FILM FORMATION. H.S. Kwok, Institute on Superconductivity, State University of New York at Buffalo, Bonner Hall, Buffalo, NY 14260 Several experiments were performed to examine the nature of in-situ film formation for the high temperature superconductors. In one experiment, the resistivity of a Y-Ba-Cu-O film was monitored during laser deposition in real time. Both deposition parameters leading to in-situ films and those requiring high temperature (850°C) post- annealing were examined. It was found that in all cases, the resistivity was dependent on the oxygen partial pressure during deposition, and that filling the chamber with oxygen immediately after deposition was crucial to the formation of the supercondueting phase. Oxygen out-diffusion during deposition at high temperature (600°C) was found to be the major obstacle to true in-situ film formation. In a separate experiment, the plasma emission from the laser generated plume was examined. Both emission and absorption spectroscopic data were obtained relative to the formation of atomic and molecular beams in the plume. Their relevance to thin film form ation will be discussed. (2: 50 p.m.) IN-SITU LASER PROCESSING OF EPITAXIAL AND TEXTURED HIGH-Tc SUPERCONDUCTING THIN FILMS. J. Narayan, Dept. of Materials Science and Engineering, North Carolina State University, Raleigh, NC 27695 Superconducting RBa2Cu307 and RBa2CII40106 (where R=Y or Ho) films have been formed by "in-situ" laser processing in the temperature range 500 to 650·C. The thickness of the films, obtained by baised laser deposition, ranged from 400 to 5000A with Tc (critical superconducting transition temperature associated with zero resistance) varying from 87 to 9OK. The films grown on (100) SrTiO:l, (100) LaAI03 and (100) YSZ substrates were epitaxial with following relations: <110>123 II <110> SrTiO:!, [001]12311[001] YSz, [100]12311[110] YSZ. The above epitaxial growth is referred as lattice-matched and planar-matched on (100)SrTi03 and YSZsubstrates, respectively. The film growth on (loo)MgO and (100) Si was textured, where only C-axis was aligned: <001> 123 11<001> MgO ' or Si. The textured films on (100) MgO and Si exhibited Tc of 85K, and 78K, respectively. The critical current density (Ie) up to 6.0 x 106 Amp cm- 2 (zero magnetic field at 77K) was obtained for best films on (100) SrTi03 and (100)LaAI03 with minimum channeling yield (Xmin) of 3.0%. Detailed X-ray diffraction, transmission electron microscopy, and electron channeling studies were used to investigate the nature of epitaxial or textured growth and defects such as dislocations. twins, domains and grain boundaries.
(3:15 p.m.)
BREAK
(3:35 p.m.) FABRICATION AND CHARACTERIZATION OF YBCO TIDN FILMS ON STAINLESS-STEEL SUBSTRATES, R.E. RUllO, P. Berdahl, R.P. Reade, aDd J.M. MeMillt.n, Mail S~p 90-2024, Lawrence Berkeley Labora~'1, Berkelo)" CA gt720, and B.L. Ol&en, Le.wrence Linrmore Nt.tional Lt.bort.~r)', 1.-350, Livermore, CA 94550. Pulsed I..,er deposition haa been demoDitrated .., an excellent technique for fa.brict.ting thin lihno or the IIOW high temport.tu •• con.mle .uperconduc~l'II. Mucb ot tbe previous wo.k bill been devoted ~ deposition on sin,le crysta.l sub$trate. suitable Cor the electronics indll8tr),. For electric power tran.miNion or energy .~rar;e applicatiolUl, other more practical lubatn.tee t.re required. We ht.ve ra.bricated _deposited thin-film YBCO luperconductora on .tainle...teel aub.tratea by uoin, metal buffer la),er" Our deposition cbamber provides the ct.pt.bility or rotating rour bulk materiall .. puately ill to tbe path or an excimer I_r beam. This approach allow. inlitu I...... deposition 01' severa.l m..teriLis without exposing the chamber (and 8lms) to the atmosphere between I..yel'll. W. depoeited the metal buft'•• lay... on the .ta.inleM steel before depositing the luperconducting film •. Th. d.poeition parameters for metal buffer layel'll hr.ve been optimized &lid ..... eigni6ca.ntly different than those employed for tbe superconducting ma~rial. Electricr.1 chiIT.cteri.ation including R vs T and critical current density sug,est tbr.t the films bave good superconducting tranlitiolll but prllOntly bYe reduced current capacity compared ~ those on lingle crYltallubitu,tea. (3 : 50 p. m.) STRUCTURE, CHEMISTRY AND METALLURGY OF HIGH TEMPERATURE SUPERCONDUCTING THIN FILMS; Ivan K Schuller University of California San-Diego, La Jolla, Ca. 92093, A. Fartash, Argonne National Laboratory, Argonne, Ill. 60439 and Y. Bruynseraede, Katholieke Universiteit-Leuven, Leuven, Belgium. We have performed extensive studies regarding the preparation and structural, chemical and metallurgical characterization of high temperature superconducting thin films prepared by Molecular Beam Epitaxy techniques. The structural and chemical uniformity and the surface roughness are strongly affected by preparation conditions. It is pointed out that many of the measurement techniques require calibrations using a·priori characterized standards. The chemical interaction between the substrate and the overlayer is the main factor which determines the structure of the film, so that under identical growth conditions different types of films can be grown. Work supponed by the Office of Naval Research grant NOOO14-88-0480 (at UCSD) the US Department of Energy, BES-Materials Sciences under contract No. W -31-1 09-ENG38 (at ANL) and the Belgian Interuniversity Institute for Nuclear Science (at KUL). International travel was provided by NATO. (4:15 p.m.) LPE GROWN Bi-Sr·Ca·Cu·O SUPERCONDUCTOR FILMS: C S Yang and A. S. Vue, Department of Materials Science and Engineering, University of California, Los Angeles, CA 90024 Four samples, made of varying amounts of Bi203, SrC03. CuO, and CaC03 powders, were melted at 890°C and held at this temperature 'or one hour in an alumina crucible. Sapphire substrates were immersed into the melt and the temperature was then lowered to 879°C in two hours for the LPE growth . The LPE·grown films were slowly and carefully withdrawn from the crucible and cooled inside the furnace to room temperature at a cooling rate of 120°C per hour. X-ray diffraction patterns of all the BCSCO films show strong (OOt) reflection peaks, indicating that the c-axis of the film is perpendicular to the substrate surface. The data of the resistivity versus temperature the LPE-grown BCSCO films indicate that the resistivity decreases from room temperature to about 100oK. Below this temperature, it deviates from linearity with a downward curve to 98°K and remains constant until it reaches 89°K. At this temperature the resistivity deviates again from linearity with another downward curve to a zero·resistance value of 70oK. However, in another case, the double· transition phenomenon was disappeared and Tc (R=O) was 8ooK.
0'
(4 : 40 p.m.) EFFECT OF OXYGEN ORDERING ON BULK AND THIN FILM CERAMIC SUPERCONDUCTORS. Y. Bruynseraede, B. Wuyts, J. Vanacken, Laboratorium voor Vaste Stof-Fysika en Magnetisme, Katholieke Universiteit Leuven, 3030 Leu· ven , Belgium; LK. Schuller, Physics Department - B019, University of California . San Diego, La Jolla, CA 92093, USA . It is now well established that tbe physical properties of high Tc oxide compounds are very sensitive to the oxygen ordering and stoichiometry. Changes in the oxygen content can induce structural transformations, metal-insulator transitions, variations in the transition temperature and may influence the critical current and field properties. A review will be given of the effect of oxygen deficiency a.nd vacancy ordering on the normal, superconducting and magnetic properties
of various bulk and thin film ceramic superconductors. Results obtained from gas evolution, superconductivity, x-ray diffraction and TEM experiments will be discussed. This work is supported by the Belgian F.K.F.O., LU .A.P. and G.O.A. Programs (at K.U .L.) and the U.S. Office of Naval Research contract N" 00014-88K·0480. (at U.C .S.D.). International travel was provided by NATO grant R.G. 0005/89.
Annual Meeting-73
HYDROMETALLURGYANDAQUEOUS PROCESSING IV: Biotechnological Treatment of Minerals and Effluents
(4:00 p. m.) EFFECT OF BACTERIA ON THE ELECTROCHEMICAL BEHAVIOR OF PYRITE: Fiona M. Doyle, Abbas H. Mirza, Wumao Ye, and Mariola Mianowska, University of California at Berkeley, Dept. Materials Science and Mineral Engineering, Hearst Mining Building, Berkeley, CA
94720.
Sponsored by the TMSlEPD Aqueous Processing Committee Tuesday PM February 20, 1990
Room: Salon 1
Marriott
Session Chainnen: Batric Pesic, University of Idaho, Department of Mining and Metallurgical Engineering, Moscow, ID 83843; Rong-yu Wan, Newmont Metallurgical Services, 417 Wakara Way, Suite 210, Salt Lake City, UT 84108 (2:30 p.m .) MIXING EFFECTS I N REACTORS FOR THE BACTERIAL LEACHING OF ORES Dr. Graham Andrews , Biotechnol ogy Unit, Idaho National Engr. Laboratory, Idaho Falls, ID. P. O. Box 1625 83415 Biological proc ess es are inherently slower than chemic al processes, so in the bioprocess ing of large volumes of or e efficient bioreactor design is essential to mi nimize the size a nd cost of t he reactors. Essential features are a high pulp density , a microbial concentration proporti onal to t he available solids surface are a, and nut rient and inhibitor concentrati ons kept in the optimum range for microbial metabolism. Mixing the bioreac t or has three ma i n effects; incoming ore i s automaticall y inoculated wi th mi crobes grown i n t he reactor ; some sol id part icles go strai ght fr om the
i nl et to the outl et rec ei vi ng vi rtual ly no t reatment; inhibitors , including
er senat es , pr oduced by t he process are
spread throughout the reac to r l owering the metabolic rate. An analysis shows that the overall effect is usually very negative, and a plug flow type of reactor with cell recycle can produce a much higher th roughput . Convent ional reactors, inc luding heaps and stirred tanks i n ser i es, are evaluated based on this analysis .
(2:55 p.m.) THE CATALYTIC ROLE OF THIOBACILLUS FERROOXIDANS IN THE LEACHING OF PURE NATURAL CHALCOPYRITE: Tomas Vargas, Jacques Wiertz, Blanca Escobar, Ricardo Badilla-Ohlbaum, Depto de lngenier[a Qu[mica, Facultad de Ciencias F[sicas y Matematicas, Universidad de Chile, Gasilla 2777, Santiago, Chile. The catalytic role of Thiobacillus ferrooxidans on the dissolution of chalcopyrite depends on the bacterial oxidation of dissolved ferrous iron (indirect action) and the direct attack by bacteria attached to the sulfide (direct action), but the relative importance of both mechanisms is not yet well determined. Further insight into this problem was gained from leaching experiments using a pure strain of T. ferrooxidans (ATCC/19859) where, in addition to the usual measurements of bacteria in solution, the evolution of the attached bacterial population was evaluated separately. Moreover, the use of highly pure, natural chalcopyrite prevented the interference of galvanic interactions, permitting accurate chalcopyrite kinetic determinations and yield calculations to be obtained. The catalytiC act ion of T. ferrooxidans is discussed in relation to the fundamental leaching mechanism of chalcopyr ite and the observed growth patterns of free and attached bacteria. (3:20 p.m.)
Many sulfide leaching reactions involve electrochemical mechanisms, which have been elucidated using electrochemical studies. There is still, however, appreciable uncertai nty in the.interpretation of the electrochemical behavior exhibited in the presence of bacteria such as Thiobacillus ferrooxidans. Under aerobic conditions, oxygen reduction occurs within the bacteria, using electrons transported biochemically across the cell wall. Thus the bacterial cell walls are part of the electrochemical circuit . A corresponding flow of current would only be observed in an external circuit, however, if the bacterially mediated oxidation process occurs at the mineral surface. Oxidative leaching could occur without current flow if electron transfer occurs in solution. This paper discusses our own observations of the behavior of pyrite, along with results from the literature, in terms of probable reaction mechanisms. (4:25 p.m.) CHEMICAL AND BIOLOGICAL CYANIDE DESTRUCTION AND SELENIUM REMOVAL FROM PRECIOUS METALS TAILINGS POND WATER : R. H. Lien, B. E. ninsdale, K. R. Gardner, and P. B. Altringer , Bureau of Mines, Salt Lake City Research Center, 729 Arapeen Drive, Salt Lake City, UT 84108 . The Bureau of Mines, U.S. Department of the Interior, is investigating chemical and biological decontamination of complex waste waters such as tailings pond water containing 280 ppm CN and 5 ppm Se, as well as significant concentrations of arsenic, copper, iron, silver, and zinc . The most effective chemical procedure involves cyanide oxidation using hydrogen peroxide or sodium hypochlorite followed by selenium reduction using ferrous hydroxide. The effluent contains <1 ppm CN and 150 ppb Se; concentrations of other major contaminants meet EPA drinking water standards . Preliminary research indicates that biological cyanide oxidation is possible even in highly alkaline solutions (pH 10.5) containing high cyanide concentrations; indigenous bacteria destroyed 95 pct of the cyanide. Exploratory research shows that other bacteria removed at least 70 pct of the selenium from oxidized, cyanide-free water .
INTERFACES IN METAL· CERAMICS COM· POSITES III: Thermodynamics and Kinetics B Sponsored by Process Fundamentals Committee and Composite Committee Tuesday PM February 20, 1990
Room: Plaza D
Inn at the Park
Session Chainnen: G.P. Martins, Dept. of Met. Eng., Colorado School of Mines, Golden, CO 80401; P.X. Lee, Dept. of Mat. Sci. & Eng., Shanghai Jiao Tong Univ., Shanghai, 200030, PRC
(2: 00 p.m.)
EIECIRXlIEMIS'I OF T. ferroaxidans INI'ERl\CrIOOS Wl'lH P¥RITE: * B. Pesic am Inbeum Kiln, Department of Metallurgy, am D. J. Oliver, Department of Bacteriology am Bic:x:ilemistry, University of Idaho, M:lscxM, Idaho 83843
MATRIX ALLOYING ELEMENT EFFECTS ON CIAL INTERFACES, Li Pingxing Institute of Composite Materials, Shanghai Jiao Tong University, Liu Guobin , Institute of Composite Matl' s , Shanghai Jiao Tong Univecsity, Shanghai, China .
'!be interactions of pyrite with T. ferroaxidans Io'ere studied by usin] cyclic voltametIy tedm.i.ques (c.v). '!be stOOy was perfonned on crystalline mineral pyrite. '!be cyclic voltametIy was perfonned in different type of studies, rut the JOOSt characteristic one was the c.v durin] the fermentation. very inportant di.sooveries Io'ere made. Fbr exanple, the pyrite
The influence of matrix alloying on the compatibility and interfac ia l bonding of CIAI composites has been investigated. It appears that wettability between fiber and aluminum could be improved in some extent by a small amount of alloying of titanium and copp er. High tensile strength of CIAI composites can be obtained with an adequate interface bonding by adding suitable
surface invariably ~ its prcperties after 200 day of fermentaion, as evi.deooed by the cyclic voltarrograms. 'Ihls elcect:rochemical fiIxtin:J perfectly correlated with the results fran the scannin;J electron microsc:qJy stOOy. '!be finiin3s reported in this paper are very inportant, unique, am infonnative. (*) 'Ihe work supported by the OOE
(3:45 p.m.) BREAK
am USIIVINEL
alloying elements such as titanium or copper to aluminum ma trix.
Alloying additions change the interfacial behavior by influenc ing the formation of interfacial phases, resulting in the variation of compos ite wire strength .
The tensile strengths of composite
wire are rapidly decreased after thermal exposure above 550 0 C due
to the i nt e nsive interfacial reaction , in the mean time corresponding changes of fracture appearance occurred. According to the characterization of the interface between fiber and
aluminum, the effects of alloying elements Ti and Cu are discussed from the thermodynamic and fracture mechanism po i nt of
v iew.
74-Annual Meeting p.m.) INTERFACIAL REACTIONS IN METAL MATRIX COMPOSITE CONTAINING INTERMETALLIC RIBBONS,R A Varin, H. Hetelnick , Z. Wronski, Department of Hechanical Engineering, University of Waterloo, Waterloo , Ontario, Canada N2L 3Gl. (2:20
The focus of the present work is on the investigation of the interfac ial zones formed during fabrication and simulated
elevated
temperature
service of a model
intermetallic·metal
composite system. In this composite the melt· spun ribbons of Ni3 (Al, Hf/Zr, B) intermeta11ic compound were cast into 1100 Al
matrix. the interfacial zones composed of several intermetallic layers were observed and their composition was investigated by
quantitative EDS. be AINi ~ A13Ni2
The general sequence of layers is observed to A13Ni. Temperature stability of these layers
drop method in vacuum from 700 to 1000C. The effects of
temperature and alloying elements such as Mg,Cu,Zn, Si, etc. on
the wettability
investigated.
of Alumina .
Aluminium systems
were
The wetting phenomena and interfacial reactions
were observed and analyzed with metallographic microscope , SEM, microprobe,
X-ray diffraction,
etc . There
is no interfacial
reactions detected between pure Aluminium drop and a-Alumina
plaque below 900 o C;
above
this temperature Q-Alumina reacted
with Aluminium and formed the gaseous A120 which eroded the surface of placque and resulted in improving the wettability of the system. The effects of Hg and Cu on improving wettability will also be discussed. * No . 52 Research Institute, Baotou 014034 P.R. ,China
at elevated temperature service was simulated by annealings at
(4:30 p.m.) "INTERFACIAL STABILITY OF SiC FIBER REINFORCED NICKEL AND COBALT
in
COMPOSITES,": Zhang Zhiming, Beijing Research Institute of Materials and Technology, P.O. Box 9211, Beijing, China; and Wei Ketai, Department of materials Science and Fuels, National University of Defense Technology, Changsha, Hunan,
~
various temperatures and times. Considerable growth of A13Ni2 interfacial layer was observed upon annealing in the range of 400·500 0 C for several hours. The obtained results are discussed terms
of
interdiffusion of nickel
and aluminum
in
the
interfacial zone. The importance of these findings for the potential application of these composites is also discussed.
China.
p.m.) EFFECTS OF THERMAL EXPOSURE ON INTERFACIAL CHARACTERISTICS OF GRAPHITE/ALUMINUM COMPOSITE, ~,* S. Dionne , * G. Carpenter,* D. Zimcik,** *Dept. of Energy, Hines, and Resources 568 Booth St., Ottawa, Ontario KIA, OCl, Canada, **Dept. of Communications, 3701 Carling Ave., Ottawa, Ontario K2H 8S2, Canada.
(4: 50 p.m.) THE INTERFACE REACTION OF C/Al AND Gr/Al COMPOSITES, Sui Quanwu Guo Shuqi, Guo Wenxu and Tang Fenjun, Institute of Metal Research, Academia Sinica, Wenhua Road 2·0, Shenyang, China.
(Abstract Not Available)
(2 :40
It has been known that the degredation induced on graphite/aluminum (Gr/Al) composites, when exposed to elevated temperatures
I
is
due
to
the
excessive
interfacial
reaction
between the g~aphite and aluminum. One common remedy to this problem, is the application of TiB2 surface coating on graphite fiber to inhibit such reaction . In this study, effects of thermal exposures on Gr/Al composite, containing graphite fibers coated with TiB2 were evaluated. This includes examination of
interfacial characteristics with both scanning and transmission electron microscopy, and evaluation of mechanical properties . In
addition,
fractographic examination were performed to determine
CVD, PVD and a new method without any coating are employed to manufacture precursors of C/Al and Gr/Al. The composites are fabr icated by different technology from lay-up C/Al or Gr/Al precursors alternating with LD2 foils. Scanning electron microscopy (SEM) , transmission electron microscopy (TEM) , X-ray energy spectroscopy (XES) and secondary ion mass spectroscopy (SIMS) are applied on the fracture surface of the composite and· the cross sections. Based on the detected compounds at the interfaces via these four techniques, it is concluded that the structure and the reaction products at the interface observed by
different technologies are obviously distinctive.
the operative mechanisms to failure .
(3:00 p.m.)
T!lEmK)DYNAMIC ANALYSIS OF CHEMICAL INrERACTIONS BEnmlN REINFORCEMENI MATERIALS AND MATRICES IN AUlMINIDE-BASED HIGH TE}lPERAnJRE INTERMETALLIC MATRIX COMPOSITES: A. K. Misra, Sverdrup Technology, Inc., Lewis Research Center Group, 16530 Ccmnerce Court, Middleburgh Heights, OH 44130. Chemical reaction between the fiber and the matrix in a composite system is of major concern in selecting a reinforcement material for a given matrix. In this paper, chemical interactions in high temperature intermetallic-based composite systems are examined from thermodynamic considerations. Different reaction modes between the reinforcement material and the matrix material are described and criteria for chemical compatibility between these two are defined. Based on thermodynamic analysis, several chemically compatible reinforcement materials are identified for intermetallic matrices based on Fe-Al, Ni-Al, and Nb-Al systems.
(3:20 p.m.) Break (3:30 p.m.) PRODUCTION OF MG - SIC COMPOSITES BY INJECTION, Nina Loge Hansen Thorvald Abel Engh, Otto Lohne, The Foundation for Scientific and Industrial Research, 7034 Trondheim, Norway. A method for producing particle reinforced magnesiUII composites is described . Particles are injected by means of a N2 carrier gas through a lance down into a Hg·alloy . The ceramic particles are distributed evenly throughout the melt; however, there are a certain nUllber of clusters and agglomerates . The composites are suitable for extrusion and give a fairly good quality of the profiles. Tensile tests showed that the particle additions caused a marked increase in stiffness and yield· strength properties compared to the referenced material. The improvements in tensile strength were only marginal. Particle additions result in a 50-70t decrease in ductility. To produce a satisfactory material the dispersion of the particles in the gas phase must be improved and/or the clusters in the melt must be broken up by electromagnetic stirring. (3: 50 p.m.)
"INFILTRATION MECHANISMS
lIND
KINETICS OF LIQUID ALUMINUM
J.D. Seiz, G. R. Edwards, and G. P. Martins, Department of Metallurgical and Materials Engineering, Colorado School of Mines, Golden, CO 80401. INFILTRATION OF -SiC COMPACTS,":
(Abstract Not Available)
(4:10 p.m.) AN INVESTIGATION ON WETTING BEHAVIOR AND INTERFACIAL REACTIONS OF ALUMINIUM - 0: ALUMINA SYSTI!!!, Wu Jinbo Zhong Lijun, Qin Fangliang, Qi Peixiang*, Dept. of Haterials Science and Engineering, Zhehiang University, Hangzhou 310027, P.R. China. The contact angles of molten Aluminium and some Aluminium alloys on polycrystalline a-Alumina were measured by sessile
1990 LEAD-ZINC AND RELATED METALS SYMPOSIUM VITI: Zinc Practices Sponsored by the TMS-EPD Lead-Zinc-Tin Committee Tuesday PM February 20, 1990
Room: Salon A & B Marriott
Session Chainnen: J.V. Derby, Zinc Corporation of America, 300 Frankfort Rd., Monaca, PA 15061; G. Reis, Vielle-Montagne, Zinkstraa 1, B-2490 Balen-Wesel, Belgium (2:30 p.m.) USE OF NATURAL GAS AND EVAPORATIVE COOLING FOR SLAG FUMING AT THE CHIMKENT LEAD PLANT, USSR, K.S. Izbathanov, M.A. Naimanbaev, N.B. Tabynbaev, and B.V. Kolosov, Chimkent Lead Saelter, USSR. Tbe Chiatent Lead Plant was constructed in 1934 and since that tiae has undergone many revisions. In 1975 the slag fuaing operation was changed from pulverized coal to natural gas fuel. In 1982 the slag fuaing furnace water jacket system was converted to evaporative cooling with high pressure steam production. The impact of these changes and the method of their application are described.
(3:00 p.m.) EQUIPMENT AND PROCESSES FOR ROASTING ZINC CONCENTRATES AT COHINCO'S ZINC PLANT, TRAIL, B.C., D.H. Magoon, K.A. Metcalfe, A.R. Babcock, and V.A. Van Beek, Co.inco Ltd ., Trail, British Columbia, Canada V1R 4L8. The zinc smelter of Coainco Ltd. at Trail, B.C. is the largest in tbe western world. A total ot 550,000 tonnes of concentrate are processed at tbe smelter annually. Approximately 80' of tbe zinc concentrate is roasted and the remainder is treated by aqueous oxidation pressure leaching. The zinc concentrate roasting and sulphur dioxide gas processing plants have undergone many changes over the years_ This paper describes the processes and equipment currently used for concentrate roasting, S02 gas cooling and cleaning, mercury removal, sulphuric acid production, liquid S02 production, and tail gas S02 scrubbing. The recent installation of an advanced process control systea tor the operation ot these integrated processes is also described.
Annual Meeting-75 (3:30 p.m.)
THE CONTROL OF LEAD IN ELECTROLYTIC ZINC AT BARTLESVILLE, J.K. Boxall and S.E. James, Zinc Corporation of America, P.O.Js.O:K 579, Bartlesville, OK 74005. The electrolytic zinc plant at the Bartlesville Division of ZCA originally used lead anodes containing only 0.25% silver because only HG zinc and alloy was to be produced. In 1984, a small addition of strontium carbonates was made to control the surface morphology of the cathodes which reduced short circuits and increased current efficiency. Later a drastic reduction in colloid addition further reduced short circuits. Increased strontium additions allowed the production of some SHG zinc. Increasing the silver content of the anodes reduced the lead in cathodes only slightly. Kore important, the anode scale that was not removed during anode cleaning, entered the electrolyte and contaminated the cathodes with lead. A piping system has been installed to purge the contaminated electrolyte directly back to the leaching section. This has allowed the plant to produce SHG zinc consistently. (4:00 p.m.)
BREAK
(4: 15 p.m.)
THE CELLHOUSE BASED ON NEW CONCEPT ATKITSUBISHI AKITA ZINC REFINERY, K. Kaneko, S. Watanabe, Y. Kiyano, T. Ishikawa, A. Katai, T. Araki, Kitsubishi Ketal Corporation, Akita Zinc Refinery, 3-1-18 Barajima, Akita 010, Japan. In accordance with the new concept described in the paper presented at the TKS annual meeting in 1989, a new zinc cellhouse was constructed and successfully operated. The new concept was proposed in order to reduce energy requirements and to mechanize the electrode handling in the cellhouse. Kajor modifications involved the use of an electrode unit consisting of assembled anodes and cathodes with frame bodies between each cathode and anode. The space between a cathode and an anode could be halved by this unit structure. In this paper actual operation data are summarized and discussed in comparison with the calculated figures.
tecnnical, electrical, and process designs of electric furnaces peculiar to the reduction of non-ferrous metals. Electric and various thermic furnace processes are compared. (3:00 p.m.)
THE OUTOKUKPU METHOD FOR LEAD RECOVERY IN ELECTRIC FURNACE, J.S. Sulanto, Outokumpu Oy, Engineering, P.O. Box 86, 02201 Espoo, Finland. A series of tests have been carried out in a 1000 KVA electric furnace adjacent to a pilot Outokumpu Flash Smelting Furnace for smelting lead concentrates. The furnace has been fed continuously with molten Flash Furnace Slag. The feed rate has varied from 0.5 to 3.1 KTPH (0.55 - 3.4 STPH) depending on the feed rate and the quality of concentrate to the Flash Smelting Furnace. The inlet slag contained 15 to 40% of lead and the cleaned slag 1.5 to 3.0%. The lead recovery from solid raw material was tested in the same continuous process by feeding lead calcine into the furnace. Feed rate was 0.7 to 0.95 KTPH (0.8 - 1.05 STPH) and the calcine contained 33 to 34% of Pb. Several reduction agents and procedures were tested. The process proved to be suitable for continuous recovery of lead from solid oxide raw aaterials. (3:30 p.m.) ELECTRIC ARC SMELTING AT RSR CORPORATION: Don Eby, RSR Corpora-
tion, 1111 W. Mockingbird Lane, Dallas, TX 75247.
The application of electric arc furnace smelting of lead, antimony, and tin bearing slag fran a reverberatory furnace will be discussed. The selection for the 4.0 MVA electric arc furnace in Indianapolis, Indiana is presented. The design criteria for the transfonner, electrodes, refractories and associated equipment is presented. The operation of the furnace including charging, bullion tapping, slag tapping, ventilation, slag canposition, metal recoveries, and toxic leach results of thrCM away slag are discussed. FlCM charts and the equipment specifications are given.
(4:45 p.m.)
THE KONACA ELECTROTHERMIC SMELTER - THE OLD BECOMES THE NEW, Roger L. Williams, Zinc Corporation of America, 300 Frankfort Road, Monaca, Pennsylvania 15061. The Zinc Corporation of America Monaca smelter continues to use the St. Joe-developed electrothermic process for producing zinc metal and zinc oxide. As we enter the 1990's, the electrothermic process flexibility permits treatment of larger quantities and lower-grade zinc-bearing recycle materials. A new sinter plant having one large sinter machine will increase sinter feed production required to meet our capacity target of 150,000 tons per year.
1990 LEAD·ZINC AND RELATED METALS SYMPOSIUM IX: Electric Furnace Smelting of Lead, Zinc and Tin Sponsored by the TMS-EPD Lead-Zinc-Tin Committee Tuesday PM February 20, 1990
Room: Salon C & D Marriott
Session Chairmen: R.D. Prengaman, RSR Corporation, 1111 West Mockingbird Lane, Dallas, TX 75247; R.H. Maes, Metallurgie Hoboken Overpelt, A. Greinerstraat 14, B-2710, Hoboken, Belgium (2:30 p.m.)
LEAD SMELTING IN A SUBMERGED ARC FURNACE, H. Cepin, H. Koenig, G. Rath, and T. Vlajcic, Mannesmann Demag Huttentechnik, Wolfgang-Reuter-Plate, Postfach 101507, D-4100 Duisburg 1, Federal Republic of Germany. Production of non-ferrous metals is still principally performed in shaft and flame fired furnaces. Electric furnaces increase metal recovery, reduce environmental burdens and decrease energy consumption compared to conventional processes. Because non-ferrous metals have low melting and boiling points and aggressive slags design of the furnace, energy input, and slag conductivity and composition are very important. Secondary materials are easily handled in electric furnaces. Since additional amounts of lead, zinc and tin will become available from secondary sources in the future, electric furnaces are expected to replace conventional smelting furnaces. The paper describes
(4:00 p.m.)
BREAK
(4: 15 p.m.)
ELECTRIC SKELTING OF TIN CONCENTRATES AT CAPPER PASS LTD., A.K. ~ and I.A. Linton, Capper Pass Limited, North Ferriby, North Humberside, England. The application of submerged electric arc furnace smelting techniques to tin concentrates at the Capper Pass tin smelter in England is discussed. Some experimental data from earlier small scale 50 KVA electric smelting testwork on various low and mediua grade tin concentrates is presented. Design considerations for the construction of a 3 MYA electric furnace to treat tin concentrates from the East Kemptville mine in Canada are considered, together with necessary precautions peculiar to the construction of a high grade smelter in close proximity to an existing low grade complex non-ferrous metal smelting operation. Slag composition mOdifications to optimize operational smelting conditions for various concentrates treated are discussed, together with refractory selection. The effect of concentrate impurity levels on the permissible degree of recycle to produce a satisfactory metal product is highlighted. (4:45 p.m.)
THE ELKEK MULTI-PURPOSE FURNACE FOR THE RECOVERY OF ZINC AND LEAD FROK WASTE KATERIALS, T. Pedersen, J .A. Aune and R.H. Cundall, Elkem Technology a/s, P.O. Box 4376 Torshov, N 0402 Oslo 4, Norway. Elkem processes waste aaterials, dusts, sludges and other residues in a Multi-Purpose Furnace. The Kulti-Purpose Furnace is an electric slag resistance reduction furnace equipped with a hot aedia, Modular Freeze Lining and proprietary, air tight electrode seals. The oil cooled freeze lining allows higher heat flux than conventional furnaces. Reduction activities are free from the reoxidation by infiltrated air via gas tight electrode seals. Small quantities of highly concentrated off gas are produced. High thermal efficiency is achieved by generating heat within the slag itself. The Kulti-Purpose Furnace operates at high melting rates and intense power density. "lIkem's EAF Dust Processing Technology" combines proven technologies for the recovery of contained metal values from K061 waste at the same tiae prodUces a disposable slag product.
76-Annual Meeting
LOW TEMPERATURE Si AND Si-BASED EPITAXIAL STRUCTURES: ELECTRICAL PROPERTIES AND DEFECTS I: Low Temperature CVD Si Epitaxy Sponsored by the TMS Electronic and Photonic Device Materials and Electronic Materials Committees Tuesday PM February 20, 1990
Room: Orangewood III Quality Hotel
Session Chainnan: M.L. Green, AT&T Bell Laboratories, Murray Hill, NJ 07974
(3:10 p.m.) QUALITY OF SILICON EPITAXIAL LAYERS GROWN FROM SILANE/HYDROGEN MIXTURES AT 650-850·C: A J Pidduck, D J Robbins, I M Young, J L Glasper, A S R Martin, C Pickering and Y Leong, Royal Signals & Radar Establishment, Malvern, Worcs, WR14 3PS, UK. Silicon epitaxial layers grown at reduced temperature and pressure (O.l-lmbar)in our ultrahigh vacuum reactor have been characterised using many techniques, including optical and electron microscopy, defect etching, mass spedaDrnetry, and insitu light scattering. We summarise the range of impurities which have been encountered, and the different ways in which
they
affect epitaxial growth and layer quality.
(3: 30 p.m.)
(2:00 p.m.)
(Invited)
LOW TEMPERATURE, DEFECT-FREE SILICON EPITAXY USING A LOW-KINETIC ENERGY PARTICLE PROCESS: T. Shibata and T. Ohmi, Dept. of Elect., Tohoku University, Aza-Aoba, Aramaki, Aoba-ku, Sendai 980 Japan. Low kinetic-energy particles (Ar ions) having energies comparable
to typical inter-atomic binding energies have been utilized to control the kinetics of thin film growth as well as to perform very effective predeposition substrate-surface cleaning. By combining this technology with Ultra Clean Technology for processing environment control, we have succeeded in growing devicegrade epitaxial silicon layers with perfect crystallinity at temperatures as low as 300°C. In this low temperature epitaxial growth process, in situ impurity doping and electrical activation are possible with a doping concentration identical to that of the silicon block used as a sputtering-target. pn junction diodes formed by directly depositing p-epilayers on p type substrates exhibited very low reverse current density of 1.88xlO- 9 A/cm 2 at a reverse bias voltage of SV. Further reduction in epitaxial growth temperature has been achieved by increasing the ion dose while keeping the ion bombardment energy at the optimum value. (2: 30 p.m.)
REMOTE PLASMA-ENHANCED CVD: REACTION KINETICS AS A FUNCTION OF GROWTH PARAMETERS B Anthony, T. Hsu, L. Breaux, S. Banerjee, and A. Tasch, Microelectronics Research Center, Department of Electrical and Computer Engineering, The University of Texas, Austin, TX 78712. In this paper we will discuss the reaction kinetics of Remote Plasma-enhanced Chemical Vapor Deposition (RPCVD). In RPCVD, a remote noble gas glow discharge is used to generate energetic species to drive the CVD reaction at low temperatures. The excitation of the silane is limited to interaction with the plasma-excited argon gas over a distance of 5-10 cm between the gas dispersal ring and the substrate.
The dependence of growth rate on temperature, r·fpower, and the silane flow rate has been studied over a range of 200-400·C, 4-8W r·f power, and 10-20 sccm of silane at constant pressure. The deposition rate has been found to increase sharply with increased r-f power which implies that energetic electrons are more important than argon metastables since metastable concentration is maximized at low powers. A plot of depositon rate vs. temperature yields a very small activation energy (0.25 eV) which is much smaJJer than that of low· temperature thennal CVD. (2: 50 p. m. )
(Invi ted)
REDUCED PRESSURE AND TEMPERATURE EPITAXIAL SILICON CVD: KINETICS AND LAYER CHARACIERIZATIONS 1.L.Regolini and D.Bensahel, CNET-CNS 38243 Meylan France and I.Mercier CNRSILEPES 38040 Grenoble France In the rapidly advancing VLSI technology the scaling of devices for higher density and better performance imposes new requirements on semicond. materials and process. The capability of depositing thin Si epi-layers with sharp transition width and high crystallographic quality is essential for high speed technology. Using a Rapid Thermal Processing system we have studied the kinetics aspects of the more usefull reacting gases like: silane, dichlorosilane and disilane. We have obtained relavant results specific to the working temperatures (650 to 1100°C) and pressures (around 2 Torr). A comparative study will be presented between these systems concerning growth rate, selectivity and defects production. The extention of these results to the deposit of titanium disilicide, without substrate consumption will also be discussed.
Hydrogen
is
shown to influence epitaxial growth mechanism and rate below 750°C, and also to be non-inert during pre-epitaxial bake at 850-900°C to remove substrate surface oxide_ Residual carbon, oxygen and particulates, when present on the substrate surface after prebake, are found to cause epilayer dislocations and stacking faults. We describe the optimisation of wafer precleaning chemistry and handling procedures needed to control these defects. The or1g1ns and control of unintentional dopants and metallic contaminants will also be discussed.
BREAK
(3:45 p.m.)
LOW TEMPERATURE EPITAXIAL GROWTH IN RADIANTLY HEATED BARREL EPI REACTORS: John o. Borland, Applied Materials, Inc., 3050 Bowers Ave., Santa Clara, CA 95054. Recent developments in low temperature epitaxial growth in radiantly heated epi systems have resulted in the growth of single crystal silicon epilayers in the 700°C to 850°C temperature range at pressures down to 10 Torr. A comparison between silane and chloro-silane based epilayers will be presented at these low temperatures and the effects of oxygen and carbon at the epi/substrate interface. ULSI quality SiH2C12 based epitaxial structures have been grown at 850°C with very low defect levels, high minority carrier lifetimes and excellent junction leakage characteristics. Also, electrically active insitu doped p+ (boronb epilayers >lxl0 20 /cm 3 , n+ (arsenic) epilayers 2xlO l /cm 3 and n+ (phosphorus) epilayers 9xlO l9 /cm 3 have been selectively grown at 850°C/lO Torr. (4:05 p.m.) SILICON EPITAXIAL FILM-PROCESS AND MATERIAL CHARACTERIZATION: s. B. Kulkarni, IBM E. Fishkill, Hopewell Junction, NY 12533 Silicon epitaxial films were grown by chemical vapor deposition using a commercially available low pressure AMC 7800 reactor. NPN devices with isolation were fabricated with further processing and tested after first metal in order to evaluate the device performance and quality of the epitaxial films. Films were characterized using optical, physical and electrical characteri-
zation techniques. Electrically active defect density was measured as a function of prebake temperature, depositionpressure and temperature, and the chemistry of the process. Dichlorosilane was found to give superior quality films as compared to silicon tetrachloride especially at low temperature and pressure conditions. Understanding of this phenomenon appears to be related to the growth kinetics (surface vs mass transport). Identification of defects and their behaviour with respect to the pre and post epitaxial processing is described. (4:25 p.m.)
(Invited)
LOW TEMPERATURE SI EPITAXY USING Si,H.: F. Hieno , Y. Furumura, and K. Wada Semiconductors Group. FWITSU LTD. Kamikodanaka, Kawasaki 211, JAPAN In recent ULSI technology, a low-temperature process for silicon epitaxial growth is necessary to minimize autodoping and solid-state diffusion effects. Also, reduction of parasitic capacitances has been playing important role to obtain a high speed performance of the devices. We have developed low temperature and low pressure Si epitaxy using Si,H 6 and SPEG (Selective Polyand Epitaxial-Si Growth) technique which make possible to form active region on Si -substrate and contact region on SiO, simultaneously. We evaluated a crystalline quality of obtained epi layers. Also, we evaluated defects at poly-Si/epi boundary and at oxide/substrate boundary of SPEG layer. We fabricated BJT and EBT(Epitaxial Base Transistor).
Annual Meeting-77
MATERIALS ISSUES IN ELECTRONIC & OPTOELECTRONIC CONNECTORS IV: Optical and Opto-Electronic Connectors & Splices Sponsored by the TMS Non-Ferrous Metals Committee and the TMS Electomic Packaging and Interconnection Materials Committee Room: California llIIIV Quality Hotel
Tuesday PM February 20, 1990
Session Chairmen: S. Wen, IBM Research Div., T.J. Watson Research Center, Yorktown Heights, NY 10598; J.C. Williams, IBM Systems Integration Division, Manassas, VA 22110 (2:00 p.m.) OPTICAL FIBER CONNECTOR AND SPLICE APPLICATIONS AND TECHNIQUES, ~. C. Williams, IBM Systems ~.rt2~r{0~tion Div1s10n, 9500 Godwin Dr., Manassas, Optical fiber is an increasingly frequent choice as the "best" transmission medium for high performance communication applications1ranging from
transoceanic
to
interch1p.
The
light-guiding
core of widely used sing.1e-mode optical fiber is less than ten microns ~n diameter and fiber optic interfaces must maintain submicron preci810n alignments throughout their service lifetimes,_ In this fap~r J the i.mportant application requ1rements 0 f1ber OEt1C connect10ns are reviewed and examples of connector and splice techniques which can maintain high precision in adverse environments are described.
(2:30 p.m.) IMPACT OF MATERIALS ISSUES ON FIBER OPTIC CONNECTORS AND· SPLICES· COST AND RELIABILITY, !L. ~, Bellcore, 44~ South St., Morristown, NJ 07960. Fiber optic splices and connectors are composite structures incorporating a variety of materials. They are required to maintain the alignment of optical fibers to submicron precision, in adverse environments, for long periods of time,
at
moderate
cost.
Many
of
the
challenges
encountered are materials issues. A critical material is the matching fluid used in splices. The temperature dependence of refractive index has been observed to result in unacceptable reflectance at temperature extremes. Occlusion
of the fiber cores by solid particles, air droplets, or absorbed water can lead to large
and errat1c splice loss or total splice failure. The real-world severity of this problem has not yet been established; however, we propose several solutions. Other examples of materials issues are the sub-Tg creep of thermoplastic materials which may impact their use in injection molded connectors and the possibility of premature reaction of UV curable adhesives dur1ng storage. The increased viscosity of the adnesive can lead to long-term failure. (3: 00 p.m.)
PROPER TECHNIQUES OF TERMI~lATION AND END PREPARATION OF OPTICAL FIBERS: John Schnecker, Buehler Ltd., Lake Bluff, IL. ABSTRACT NOT AVAILABLE. (3: 30 p.m.)
DELAYED FAILURE OF OPTICAL FIBER CONNECTOR ASSEMBLIES, Walter R. Wagner, IBM, Research Triangle Park, NC 27709 Although rigid quality control techniques and in-line proof testing are applied to the manufacture of optical fiber in order to assure a high-strength, long-life fiber, the portion of the fiber incorporated in the assembly of the connector is not the same as that taken from the draw tower. The fiber must be stripped of the protective coatings and buffer material that have been applied during manufacture in order to be installed in a connector. This exposes the fiber to ambient conditions and possible contact damage during the buffer/coating stripping and subsequent epoxy procedure. This paper reviews the fundamental aspects of flaw growth in glass when moisture is present, and shows how mechanical stripping of the fiber will induce flaws that drastically reduce the strength of the fiber. This reduction in strength is ultimately reflected in the early-life failure of connectors. Several examples of optical connector failures are presented which demonstrate how assembly techniques may influence these early-life failures. The current limits in making accurate life-time predictions for optical connectors are reviewed, and future areas of research are recommended.
(4:00 p.m.)
PACKAGING MATERIALS FOR OPTOELECTRONIC CONNECTORS, S. Wen, IBM, T.J. Watson Research Center, Yorktown Heights, NY 10598 The materials chosen for optoelectronic connectors not only need to satisfy electrical requirements but also optical requirements. This paper will review the materials and process to package lasers, chips, microlenses and fibers. The thermal expansion coefficient matching, heat conductivity and mechanical stability are crucial parameters. Choice of solders are important for optical alignment tolerance and stability of the package after assembly. Residual stress relaxation plays crucial role for single mode fiber packages. Active alignment during soldering or laser welding will be discussed. To facilitate multi channel optics, waveguide and micro lens array has been developed. Leverage of incorporating them into packages will be described. Combination of high speed optics and electronics drives needs for further materials development.
METALLIZATION FOR ELECTRONICS APPLICATIONS II: Thin Film. Phenomena B . Interfacial Diffusion and Reactions Sponsored by TMS Electronic and Photonic Device Materials and Electronic Materials Committees Tuesday PM February 20, 1990
Room: California I·ll Quality Hotel
Session Chairman: A.H. King, Department of Materials Science and Engineering, SUNY, Stony Brook, NY 117942275 (2:00 p.m_) ELECTRICALLY CONDUCTING THIN-FILM DIFFUSION BARRIERS, M-A. Nicolet, California Institute of Technology, Pasadena,-CX-
9iT25'
Thin-film diffusion barriers have become an integral part of
the metal-semiconductor contact for all advanced semiconductor
devices. The function of the barrier is to impart metastability to a basically unstable system. The dilemmas created by attempting to make thin electrically conducting diffusion barriers will be discussed. Various ways of resolving the difficulties will be described, including the ideas of a sacrificial barrier. a stuffed barrier, and amorphous metallic barriers. Examples of successes and failures will be given. General considerations on contact
design will conclude the talk, with particular concern for GaAs and other compound semiconductors.
(2:30 p.m.) STRUCTURE AND STABILITY OF METAL-SEMICONDUCTOR INTERFACES, Robert Sinclair, Department of Materials Science and Engineering, Stanford University, Stanford, CA 94305. Most metals are thennodynamically unstable with both elemental and compound semiconductors. As a consequence reactions can take place at metalsemiconductor interfaces which influence contact behavior and properties. We have used cross-section electron microscopy to study the extent and microstructure of such reactions, and thennodynamic analysis to understand the reaction products. At metal-silicon interfaces interdiffusion and amorphous phase fonnation, upon deposition alone, is quite common. The thennodynamic and kinetic considerations for interpreting growth of the amorphous alloy will be elaborated. In multicomponent systems, interface stability must be discussed in tenns of higher-order phase diagrams, and their application to contact metallurgy, diffusion barriers and various metallization schemes will be described.
(3:00 p.m.j Surface and Interface Energies in a Three-Dimensional Silicon-Silicide Structure: K. N.
Th,
IBM Research Division, Thomas J. Watson Research Center, Yorktown lIeights,
NY 10598 In a hetero-epitaxial growth, while it is crucial to have a close lattice match at the interface between the overlayer and the substrate, the surface energy of the overlayer must be less
than or very close to that of the substrate, otherwise the overlayer is unstable and tends to ball-up. In extending the hetero-epitaxial growth of a single layer of B on A to a superlattice structure of AIBIAIB, where A and B are of different matters, we encounter a sequential growth which is against the principle of minimizing the surface energy, i.e., if A. wets R, B will not wet A. Hence, it is intrinsically difficult to build a 3-dimensional
hetero-epitaxial structure consisting of silicon and silicide. Clearly, the fundemantal issue here is to understand the surface and interface energies involved. This talk presents an analysis of the relative surface energy of 8i and CoSi1 on the basis of a direct comparison
of the kinetics of crystallization of amorphous Si and amorphous CoSi, thin films. It is
a general analysis and can be extended to the '<:urface and interface energies of other superlattices
78-Annual Meeting (3:30 p.m.) GRAIN BOUNDARY DIFFUSION AND THE GROWTH OF TITANIUM SILICIDE LAYERS ON SILICON: Y.L. Corcoran and A.H. King, Department of Materials Science and Engineering, State University of New York, Stony Brook, NY 11794-2275. Most previous studies of the kinetics of silicide formation have relied upon experimental techniques that provide no microstructural information, and typically yield very few data points. We have performed a study of the thickening of titanium silicide films during solid state reaction, using transmission electron microscopy. It is found that the microstructure of the thin films of titanium silicide exerts a profound influence upon the kinetics. In particular, the grain size in the reacted film increases during the growth of the film, resulting in the loss of diffusive paths (i.e. grain boundaries) for the transport of silicon to the reaction front. This results in kinetics that are markedly different from the typically postulated .;t. growth law. The kinetics are analysed in detail and shown to fit a suitably modified theory with great accuracy. Acknowledgement: This work is supported by the National Science Foundation under grant number DMR-8901994.
(4:00 p.m.) TITANIUM SILICIDE-SILICON INTERFACE DEGRADATION DURING HEAT TREATMENT: N. de Lanerolle, B. Kim, L. Moser, Y. Zheng, J.Berg, D. Sterner. Standard Microsystems Corporation, 3S Marcus Boulevard,Hauppauge New York 11788, USA. Thin titanium silicide films were grown on different silicon substrates by rapid thermal annealing in a NZ ambient. The silicide film was then heat treated at various temperatures and environments in a furnace. The effect of such heat treatment on the morphology of titanium silicide and the titanium silicide-silicon interface was studied. The interface was observed by stripping the silicide film back to the substrate silicon, thereby exposing the morphology of the titanium silicide interface with silicon. This also enabled the correlation of changes of the surface of the titanium silicide film with that of the interface. (4:20 p.m.) THE EFFECf OF HYDROGENATED a·SILICON IN THE REACfION KINETICS OF PALLADIUM AND NICKEL SILICIDES*: Haydn Chen and N R Manning. Dept. of Materials Science and Engineering.University of Illinois at Urbana·Champaign. Urbana. IL 61801. The rate kinetics of the Pd·a·Si:H and Ni·a·Si:H reactionss has been quantitatively studied by monitoring the in.....si.UL growth of the silicide film and/or the depletion of the metal overlayer by employing x-ray diffraction and a position sensitive detector (PSD). The specimens studied included samples of Ni on a-Si:H on c-Si; Ni on a -Si on silica; and nominally 100nm of Pd on three different thicknesses of a -Si:H. each on both c-Si and silica. The 11- S i: H thicknesses correspond to sufficient Si to yield excess Pd. stoichiometric
Pd2Si. and excess Si on complete reaction of metal to
silicide. Results from samples will be presented. as well as comparisons to published reaction data for c-Si and nonhydrogenated Il-Si. Possible mechanisms and implications will be discussed *This work was supported by the US Dept. of Energy. Contract DEFG 05-85ER75231. (4:40 p.m.) OXIDATION OF TITANIUM SILICIDE: Tom Sandwick and Krishna Rajan, ~laterials Engineering Department, Rensselaer PolytechniC Institute, Troy, NY 12180-3590. Titanium disilicide is known to have two crystallographic forms: the C49 and the C54 structures. Microstructurally the metastable C49 structure is known to be heavily faulted. In this study we report the effect of oxygen on the phase stability of TiSi2. It has been found, for example, that oxidation of the C49 phase can result in a transformation to the C54 phase. The results of detailed TEt~ studies have shown that the C54 phase is not completely fault free but contains some twins. Some of the twin boundaries show facetting. The crystallography of the facets are consistent with a "chemical twinning" model that has been proposed by us earlier This work is supported in part by the Program at Rensselaer.
IBI~ I~icroelectronics
PHASE STABILITY & DIFFUSION PHENOMENA TMS General Abstract Session Tuesday PM February 20, 1990
Room: ParkA Inn at the Park
Session Chainnan: Dr. Brent Fultz, 138-78 Caltech, Pasadena, CA 91125 (2:00 p.m.)
MICROSTRUCTURAL STABILITY OF Ti 3AI-Cr/Nb ALLOYS PRODUCED BY SPUTTERING*: Raghavan Ayer, STEM, Inc., 305 Bic Drive, Milford, CT 06460 and Rong Wang, Failure Analysis Associates, 8411 154th Ave, N.E, Redmond, WA 98052 There is considerable interest in the development of fine grain alloy systems based on Ti3Al by sputtering. The talk describes an electron microscopy study conducted to investigate the chemistry and structure of the crystalline and amorphous phases which form during the synthesis of Ti3Al based alloys. CBED and xray microanalysis methods were used to characterize the phases both in the as-sputtered condition and subsequent to annealing in the range of 873-1073K. The details of the phase transformations in Ti 3AI-cr and Ti 3AI-Nb systems will be discussed. * Research was supported by NASA Langley Research Center, Contract ~ NASl-18693 (2: 17 p.m.) THERMAL STABILITY OF ORTHORHOMBIC PHASE IN TI-25Al1ONb-3V-l Mo ALLOV: M Soojb*. C. H Ward*" and I. Weiss*; *Materials Science and Engineering, Wright State Uniyersity., Dayton, OH 45435; **Wright Research and Deyelopment Center, WPAFB, OH 43433. Microstructural characterization of cast and heat treated ond forged and heot treated Ti-25AI-l0Nb-3V-1Mo (at. :1:) alloy was corried out using Optical Microscopy. Transmission Electron Microscopy, Selected Areo-, Mlcro-. and Conyergent Beam Electron Diffraction, Energy Dispersiye X-roy Spectroscopy (EDX) and X-ray Diffraction. Microstructure of the alloy in the cast or forged and then ~-solotion treotment and quenching consist of ond B2 phases.The ordered orthorhombic phase (O-phase) was obseryed during aging of the quenched material. Results of thermal stability of the O-phose following thermal exposure at different temperatures and times will be presented and discussed.
a,
This work was supported by WRDC, WPAFB, Ohio ond Edison Moteriols Technology Center, Dayton, Ohio. (2:34 p.m.) COMPATIBILITY OF POTENTIAL COMPOSITE REINFORCEMENTS WITH A TUNGSTEN-METAL MATRIX:* E. K. Ohriner and E. P. George. Metals and Ceramics Division. Oak Ridge National Laboratory. P.O. Box 2008, Oak Ridge. TN 3783l. The compatibility at elevated temperature of a variety of oxides, borides. carbides, and nitrides with a tungsten matrix were initially evaluated by thermodynamic calculation. The interactions of tungsten with oxides AI2~. Y203. Ce203, and Hf02, and with NbC, HfN. and HfB2 were studied experimentally. Blended powder mixtures were cold pressed. sintered. and exposed in vacuum at temperatures to 2273 K. In addition. full density materials were produced by hot consolidation at 1373 K at 1400- MPa pressure. The experimental and calculated compatibilities are compared. *Operated by Martin Marietta Energy Systems, Inc .. under contract DE-AC05-840R21400 with the U.S. Department of Energy. (2:51 p.m.) A MOSSBAUER EFFECI' STUDY OF HYDROGEN IN ORDERED Ni3Fe J. Okamoto and B. Fultz, 138-78. California Institute of Technology. Pasadena, CA 91125. Mossbauer spectroscopy was used to study the distribution of hydrogen and its interaction with iron in hydrogenated ordered Ni3Fe. Hydrogen was introduced into the sample cathodically in a three electrode cell. The spectra' were taken while the sample was kept hydrogenated at constant electrode potentials varying from 0 to 1.5 volts. The observed structures in the spectra were correlated to occupancies by hydrogen of the [Ni61 and [Ni4Fe21 octahedral interstitial sites. These hydrogen site occupancies are discussed in terms of site and interaction energies. This work was supported by the U.S. Dept. of Energy under contract # DE-PG0386ER45270.
Annual Meeting-79 (3:08 p.m.)
(4:33p.m.)
MlCROSlRUcruRAL CHARACfERIZATION OF LASER MELlED AL-U L S Chumbley, P. Molian·, J.E. Shield, Ames Laboratory, ·Mechanical Engineering Department, Iowa State University, Ames, IA 50011
PERCOLATION PHENOMENA IN DIFFUSION COUPLES H. Ouyang and B. Fultz 138-78 Caltech, Pasadena, CA 91125
Transmission Electron Microscopy has been used to examine the microstructure
In a previous Monte-Carlo computer simulation of diffusion in binary alloys, we studied a transition where diffusion is controlled by either the "slow" or the "fast" atomic species, depending on whether the concentration of the fast species is above or below its percolation threshold. In the present experimental study, we prepared thin film diffusion couples of pure Fe against Mo-V alloys with V concentrations above and below the percolation threshold. X-ray diffractometry and Mossbauer spectrometry were used to study the amount of interdiffusion in these thin film couples, and to identify phenomena related to the percolation behavior.
of 2090 AI-Li which has been laser melted.
In the as-received state the
microstructure consists mainly of solid-solution Al containing S' precipitates, as revealed by selected area diffraction patterns. Upon laser surface melting, the microstructure consists of large grains in which the S' reflections are much weaker, or absent altogether. The matrix contains a large number of dislocations, the actual density being ~ 1010 cm2 . A C03r5e interdendritic phase, un-identified at the time of this writing, is present and what appear to be small amorphous regions are found within the grains. The results of studies to identify the tnterdendritic phase, as well as some preliminary mechanical property data, will
be presented. (3:25 p.m.)
BREAK
(3 :42p.m.) A SIMPLE METHOD FOR IDENTIFYING THE LAUE ZONE NUMBER IN HCP SYSTEMS : A G Jacks o o, SRL, Dayt o n, OH, 45440 and M. Saqib, Wright State Univer s ity, Dayt o n, OH 45435. Identification of CBED rings from hcp systems is compli c at e d b y the presence of integral multiples o f H when e x perimental and theoretical values are compared. Combining extin c tion conditions on plane indices and the zone equati o n pro duces a simpl e equati o n useful for predicting the Laue zone number and r e lating it t o the experimentally observed multiple . Data from CBED rings fr o m TiB 2 and o ther s y stems are presented and explained using this method . A related problem stemming from use o f Miller-Bra v ais indices is discussed. ·This work was supported by WRDC c o ntract F33 615-B 6-C5013 and the Edi s on Material s Technology Cent e r.
This work was supported by the U. S. Dept. of Energy under contract # DE-FG0386ER45270. (4:50 p.m.) MICROSTRUCTURAL EVOLUTION AND THE THERMAL FATIGUE PERFORMANCE OF MODIFIED 9Cr-lMo STEEL: W. B Jones," Sandia National Laboratories, Albuquerque, NM and D. H. Polonis , University of Washington, Seattle, WA Thermal and mechanical cycling experiments have been conducted on Modified 9Cr-1 Mo Steel to Characterize microstructural instabilities and their effect on high temperature fatigue performance. Tempering for one hour at 760·C produces a relatively stable microstructure for subsequent exposure at temperatures up to 650·C, primarily due to the replacement of Cr2C by more stable V.C s along the martensite lath boundaries. Thermal and mechanical cycling conditions increase the tendency for lath break-up in the range 550·C650 · C, resulting in the formation of a fine equiaxed structure containing a reduced dislocation density and coarsened (Fe-Cr-MohsC a carbides. These changes are accompanied by a decrease in hardness and a fatigue life similar to that of the standard 9-1 alloy. The microstructural evolution is more rapid when compression hold periods are included at the peak temperature of each cycle. The observations suggest the growth of existing cracks is favored by compressive holding a peak temperature, whereas crack initiation predominates with tensile holding at the minimum temperature of each cycle. 'This work performed at Sandia National Laboratories supported by the U. S. Department of Energy under contract number DE-AC04-76DP00789.
( 3: 59 p.m.)
POWDER PRODUCTION & THERMAL MECHANICAL PROCESSING
MULTIPHASE DIFFUSION IN THE Cu-Ni-Zn SYSTEM. C. L. Liu and M. A. Dayananda. School of Materials Engineering, W. Lafayette, IN 47907.
TMS General Abstract Session
Multiphase diffusion was investigated with diffusion couples assembled with a. (fcc) and ~ (bcc) single phase alloys as well as (a. + ~) two-phase alloys in CuNi-Zn ternary system at 775°C. The main objectives of this smdy were to examine the development of diffusion structures and diffusion paths for the multiphase couples involving (a. +~) two-phase alloys. Both nonplanar interfaces and zero-flux planes (ZFPs) were observed for the multiphase couples assembled with a. or ~ phase alloys bonded to (a. +~) two-phase alloys. For selected (a. + ~)/(a. +~) multiphase couples, single phase layers of a. and ~ phases developed between the terminal alloys. A diffusion couple whose terminal compositions were selected in the a. region close to the aI(a. + ~) phase boundary on Cu-Ni-Zn ternary isotherm exhibited complex diffusional structures with (a. +~) two-phase layers. The diffusion structures for all the diffusion couples are described on the basis of experimental diffusion paths. Also, the development of ZFPs can play an important role in the control of diffusion structures. (4 : 16 p.m.) DIFFUSION CONTROLLED REACTIONS BETVEEN LIQUID TELLURIUM AND SOLID 316L STAINLESS STEEL*: J. E. Smugeresky, Sandia National Laboratories, Livermore, CA 94551-0969; A. D. Romig, Sandia National Laboratories, P.O . Box 5800, Albuquerque, NH 87185 The reaction kinetics between liquid Te and solid 316L stainless steel, pure Fe, Ni and Cr have been studied. An understanding of the binary reactions is required to model the reaction kinetics between the Te and stainless steel explicitly. The thermally activated rate equations were determined experimentally by dipping wires of the solid material in molten baths of Te (500900·C). The dissolution reactions shov a parabolic time dependence vhich is indicative of a volume diffusion controlled reaction. The reaction rate for thel~~ainless steel is given by the Arrhenius equation as: 6(cm'sec ) = -21.4 exp(-16,950/RT). The reaction rates for the pure components are similar. During the dissolution process, a series of intermetallic compounds form at the wire/Te interface. The compositions of these phases can be rationalized from the appropriate binary phase diagrams. The relationship between the nature of the intermetallic compounds and the observed reaction kinetics will be discussed. *Supported by the U.S. DOE under contract DE- AC04-76DPOO789.
Tuesday PM February 20, 1990
Room: Parke Inn at the Park
Session Chairman: B.H. Rabin, Idaho National Engineering Laboratory, P.O. Box 1625, Idaho Falls, ID 834152218 ( 2:00 p.m.) COMBUSTION SYNTHESIS OF IRON ALUM IN IDES AND PARTICULATE REINFORCED IRON ALUMINIDE COMPOSITES:* B. H. Rabin and R. N. Wright, Idaho National Engineering Laboratory, P.O. Box 1625, Idaho Falls, ID 83415-2218 Iron a1uminides containing 15 to 30 wt% aluminum have been fabricated by the exothermic reaction of elemental powder mixtures during heating . Densification and microstructural development were studied as a function of powder and proceSSing variables and the results are discussed in terms of transient liquid phase sintering mechanisms. Full density alloys and ceramic particulate reinforced composites were produced. through application of external pressure in a hot press. The results of microstructure characterization and initial mechanical property assessments will be presented .
*
Research supported by the U.S. Interior Department's Bureau of Mines under Contract No. JOI34035 through the U.S. Department of Energy Contract No. DE-AC07-76IDOI570. (2:20 p.m.) EFFECT OF THERMOMECHANICAL PROCESSING ON THE PROPERTIES, MICROSTRUCTURE AND ANISOTROPY IN Ti-l0V-2Fe-3A1, S. V. Thamboo and P. R. 8howa1, Cameron Forge Company, P. O. Box 1212, Houston, Texas 77251-1212 Different combinations of strength and K can be obtained in Ti-l0V-2Fe-3A1 through variations in the~omechanica1 process-
SO-Annual Meeting ing (TMP). This study was undertaken to examine the effects of the var ious steps of the TMP on final properties. The conventional 2-step forging (B forging plus al B forging) process was evaluated with respect to deformation induced anisotropy, cooling rate and aging conditions. In addition, alternative approaches of B anneal and/or alB solution treatment following I-step B forging were examined with a view to minimizing anisotropy while retaining acceptable properties. The effect of these TMP variations on the mechanical properties, microstructure and fracture morphology will be discussed, and some reference will be made of the application with I-step forging process for parts of complex geometry.
bonding interface characteristics. The high alloy powder needs to have no interparticle pore
(2 :40 p.m.)
inert gas atomisation technique. The powders were cold compacted and extruded in both rod and plate form in order to study the mechanical properties along the longitudinal and transverse directions. Detailed electron microscopy s tudy has been performed to study the evolution of the microstructure from the as-atomised form to
AN ANALYSIS OF DEFORMATION, TEMPERATURE AND MICROSTRUCTURE FOR HOT EXTRUDED TITANIUM ALLOY :K,Kimura M.Ishii, H. Yoshimura, Nippon Steel Corp., Hikari Yamaguchi Japan, 743. In hot extrusion, microstructure a nd mechanical properties of materials are
subjected to change considerably du e to adiabatic local heat generation. In the present work strain and temperature distributions and microstructual changes in the hot extrusion of Ti-6AI-4 V were studied by using visioplasticity method,
thermal caluculation and optical microscopy. Billets of 62<;1 mm diameter were heated at 950°C in the a + Ii region, I!OO°C in the Ii region and extruded with the extrusion ratio of 6 and 12. The result of visioplasticity calcu lation shows that in the deformation zone st ra in is relat ively high at the surface of the billet and is gradually decreased with the depth. Estimated strains of a bar extruded with the extrusion ratio of 12 at 950°C are 3.5 at the surface and 2.5 at the center respectively. But estimated temperature at the surface is lower than that at the center. Microstructures of the bar were bi-modal structure at the surface and acicular transformed structure at the center. A bar extruded with the extrusion ratio of 6 at 950 °C had all bi-modal microstructure. The patterns of strain and temperature distributio ns of bars extruded at lI00°C were similar to those at 950°C. In the case of extrusion at lIOO°C materials showed good hot-workabi lity and had all acicular structure. By applying visioplasticity method to the practical hot extrusion we are able to supply nonsymmetric cross-section extrusions with good accuracy.
0:00 p.m.) ATOMIZED FINE POWDERS--THEIR EFFECT ON DUCTILITY: Mo~er, 4 Gr~~n Briar Lan~, Cinnaminson, NJ 08077, W~st Cr~~k Road, St. Marys, PA 15557.
H. B. Ryan,
K~nn~th
Jam~s
(3:20 p.m. )
ANALYSIS OF PIPING DEFECTS IN FORGINGS AND EXTRUSIONS: ~ Y.H. Moon, Department of Metallurgical and Material s Engineering, Colorado School of Mines, Golden, CO , 80401, W.A . Gordon, The Torrington Co., 59 Field St., Torrington, CT, 06790.
~,
An upper bound analysis has been applied to the process of forward extrusion of cylindrical shapes through conical converging dies . The material is assumed to be a perfectly plastic metal which obeys the Mises yield criteria and the tool-workpiece friction is assumed to obey the constant friction factor formulation.
The
initiation of the piping defect has been determined . The size and shape of this type of surface defect has also been determined as function of the various
process parameters
(4:00 p.m.) CHARACTERISTICS OF RAPIDLY SOLIDIFIED ALLOYS BASED ON THE AI-Cr-Zr and AI-Cr-Fe SYSTEMS: E.K. Ioannidis and T. Sheppard,
Imperial College o f Science, Technology and Medicine, Dept. of Mats., Prince Consort Road , London SW7 2BP, Engl and. Rapidly solidified alloys contain additions of Crt Zr and Fe have
been inves tigated. All alloys have been produced by an up-draught
fully consolidated material. Decomposition of the rapidly solidi-
fied micros tructure occurs during consolidation of the powder material. It has been shown that the higher Cr content the more hetergeneous microstructure. Zr additions a re more preferab l e resulting in higher strength. Fe promotes heterogeneity originating from fine eutectic phases formed in the intercellular network of the rapidly solidified powder particles. Howe ver, higher short rod fracture t oughness have been noted in expense of proof and tensile strength. No significant drop of strength was observed in the transverse direction whilst significant drops in ductility were r ecorded especia lly for an Cr-rich a lloy (Al-6.4 3Cr1.67Zr). The fracture has occurred along primary powder particle boundaries .
(4:20 p.m.) Al-Li-Mg-CU-Zr ALLOYS PREPARED BY RAPID SOLIDIFICATION POWDER METALLURGY ROUTE: T. Sheppard and E.K. Ioannidis and X.Yin, Imperial ColI. of Sci., Tech . & Med., Dept. o f Mats., Prince Consort Road, London SW7 2BP, England.
Current P/M parts ar~ limit~d in ductility. Magn8-T~ch has r~ c~ntly d~v~lop~d alloys with t~nsil~ str~ngths of 110,000 psi and ~10ng8tion great~r than l~. Th~ pap~r will describ~ th~ir proc~ssing and prop~rties. A growing magnetics mark~t requires the improved ductility of phosphorus irons to sust.in s~v~re t~n sil~ forc~s r~quired in cold d~formation of thes~ parts. Powder p8rticl~ siz~ has be~n shown to dictst~ characteri8tics or poro.ity in th~ finished part, influencing th~ part's ability to sustain tensile forces during cold deformation. The results of this work will b~ describ~d. Extended automotive warranties r~quire .that ferritic stainl~Bs st~~ls b~ considered for improved corrosion resistance. The ductility of P/M stainless st~el will be compared with that of th~ phosphorus irons. The paper will conclude with s discussion of furth~r improvements in ductility that may result from using fin~r powders.
a
and better workability is capable of producing utilizing high
pressure fine-grain gas atomization technique.
(i. e.
reduction
ratio, die angle, friction and remaining billet length .) An analogous surface defect in axisymmetric forgings, often called "die pull away", can be characterized with the same analysis. The results of this analysis provides die design and processing criteria for manufacturers interested in avoiding these types of surface defects.
(3:40 p.m.) DEVELOPMENT OF HIGH-ALLOY CLAD PIPE PRODUCTION THROUGH P/M METHOD & SUITABLE POWDER PREPARATION: T. Fukuda, M. Nakanishi, T. Kubat Y. Ohashi, 1-3 Nishinagasu Hondori, Amagasaki , Hyogo,
Rapid solidification technology has been extended to the development of high s trength, high modulus low density AI -Li-Zr al loys containing alloying elements Cu and Mg. Six a lloy compositions have been produced by us ing an inert gas atomi sation technique . Two consolidation practices were performed: 1. cold-compaction/ hot-extrusion and 2. canning/degassing/extrusion. The mechanical properties of the as-extruded a nd heat treated material have been assessed in terms of hardness , tensile tests and short fracture tests. The ageing response has been studied a t 130°C, 170°C and 190°C. Detailed electron microscopy study has also been carried out to relate the mechanical properties to the developed microstructure. It was shown that the degassing process does not significantly increase the mechanical properties of the consolidated product, since cold compaction and hot extrusion were performed immediately after atornisation a nd the cold compacts were induc tion preheated to the desired extrusion temperature . CU additions are more beneficial for better ductility and short rod fracture toughness.
PROCESS FUNDAMENTALS AND MATERIALS PROCESSING n: Welding and Solidification Sponsored by the TMS Process Fundamentals and Synthesis and Analysis in Mateials Processing Committees Tuesday PM February 20, 1990
Room: Salon H Marriott
Session Chairmen: D. Stefanescu, Department of Metallurgical and Materials Engineering, The University of Alabama, Tuscaloosa, AL 35482; T. Piwonka, Metal Casting Technology, The University of Alabama, Tuscaloosa, AL 35482
660, Japan
Among high-alloy clad pipes in great dema nd for excellent corrosion resistant line-pipes, Incone l 625 clad pipe which shows most poor worka bilit y has been developed by using newly
developed powder metallurgy method. Typical defects occurring at extrusion of Ni-base high-
alloy clad pipes has been analysed. revealed to cause from the difference in flow stress between low alloy steel and high
alloy and cause from poor workability for mobility. Employing powder capsule using better workable high-alloy powder and also refined extru sion technique, no-defect clad pipe production has been fulfilled and re vealed excellent
(2:00 p.m.) ALLOY SOLIDIFICAnON: MODELING TECHNIQUES J.A.Sekhar, Associate Professor. University of Cincinnati,Cincinnati, Ohio 45221'()()12 Solidification processing/casting is increaSingly being adopted as the preferred manufacturing technique for the fabrication of net shaped complex geometry components. Modeling of solidification processes is important both for design as well as for understanding the physics of morphological evolution of the solidification microstructure. A fundamental aspect of such modeling involves the precise understanding of heat transfer through the two phase solid/liquid region. A scheme to do this will be presented where the nonlinearity of the fraction liquid with change
Annual Meeting-81 in temperature can be incorporated into the numerical calculations. The importance of properly accounting for the non-linear mushy wne will be discussed. Typical processes like squeeze casting and electromagnetiC stirring will be discussed to illustrate how precise modeling of the solid/liquid region can aid in the calculation of accurate heat transfer coefficients for these processes.
(2:30 p.m.) FLUID FLOW AND SOLIDIFICATION IN MELT - SOLID SLURRIES: O. J. Ilegbusi and J. Szekely, DeFt. of Materials Science and Engineering, MIT, Cambridge, MA 02139. A mathematical representation is developed to describe the electromagnetic force field, the fluid flow field and the solidification of melt - solid slurries. In this treatment
we use a non 0- newtonian, shear thinning model for the
constitutive relationship and make allowance for partial solidification in the bulk. The overall system behavior is related to the stirring intensity and also the system geometry.
(3:00 p.m.) MODELS FOR MHO MOL1EN MErAL ATOMIZER AND THE RAPID SOLIDIFICA nON OF METAL DROPLET IN FLIGHT:· T.e. Tszeng. D. O. Hobson, and V. K. Sikka, Metals and Ceramics Division, Oak Ridge National Laboratory, P.O. Box 2008; Bldg. 4508, Oak Ridge, TN 37831-6083. An analysis based on mathematical modeling has led to better understanding and improvement of a MHO molten metal atomizer for powder production and spray casting. Conformal mapping technique was utilized to solve the electric field which in turn determines the body force in the metal flow . On the other hand, similarity solution has been sought for the metal flow in the diverging nozzle with exponentially increasing radial body force. The solutions of electric field and the flow pattern have been used as a guideline for the MHO nozzle design. A separate numerical model for the solidification of a droplet in flight was also developed. Newtonian cooling was assumed for the heat transfer of the molten metal droplet. The model can predict the effects of droplet size, velocity, environmental gas type and pressure. It has been determined that these factors share about the same importance. "Research sponsored by the Office of Industrial Programs, Division of Improved Energy Productivity, U.S. Department of Energy, under contract DEAC05·840R21400 with Martin Marietta Energy Systems, Inc. (3:30 p.m.)
(3:45 p.m.) INTELLIGENT PROCESSING FOR PLASMA ARC MELTING D.G. Backman, D.Y. Wei, Y. Pang GE Aircraft Engines Modeling tools are being developed for plasma src assisted melting, atomization and casting to predict process behavior based upon process design and control parameters for achieving clean atomized and ingot products. The modeling approaches include development of finite control volume based models for detailed analysis of cold hearth melting and construction of a process simulator for interactive simulation of process status . The simulator consists of a group of reduced-order models which embody process physics and valid assumptions to reduce mathematical complexities The detailed models are used to validate the reduced-order models, to provide precalculated results for tabular inclusion in the simulator, and to define and interpret physical experiments. The reduced-order models are linked together in a single computer program which provides a straightforward user interface and fast response to enable predictions of melt flow pattern, melt temperature, skull thickness and
impurity particle trajectory and residen ce time. The process simulator is further validated via critical experiments and prepares a strong foundation for the development of an intelligent
controJ. system.
Sponsored by the TMS Reactive Metals Committee Tuesday PM February 20, 1990
Room: Salon G Marriott
Session Chairmen: Clifford B. Wilson, Dow Chemical Co., Freeport, TX 77541; Donald R. Sadoway, Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139 (2 :30 p.m.) THE ELECTROWINNING, ELECTROREFINING, AND ELECTROPLATING OF REFRACTORY METALS IN FUSED SALT MEDIA: Donald R. Sadoway, Department of Materials Science and Engineering, Massachusetts
Institute of Technology, Cambridge, Massachusetts
02139.
Fused salt electrolysis is used extensively for primary extraction of me tals (electrowinning). for refining of metals (e lectrorefining ) , and for the formation of coatings
In most cases . fused salt electrolysis
competes with nonelectrochemical processes to make the same products. Environmental concerns as well as health and safety issues in the workplace have the potential to change the economics of these proce sses in favor of electrolytic
technologies.
The theory and practice of fused salt
electrolysis will be reviewed in the context of processing refractory metals. Recent advances reported in the literature
will be cited.
The potential of fused salt electrolysis to
generate technologically important materials containing refractory elements will be assessed.
(3:00 p.m.) ELECTROWINNING OF CA, CE AND NO INTO A LIQUID MAGNESIUM CATHODE TO PRODUCE A STABLE ALLOY PRODUCT. Matthew R. Earlam, Dwaine W. Schoppe, K. G. Claus, all of Inorganic Research & Development, Dow Chemical Co., Freeport, TX 77541. Using a liquid magnesium cathode pool calcium, cerium and .neodymium can be electrowon from a molten salt bath. The typical current density was 2 amp/cm2 at 715·C. Alloy composition is limited t o molten magnesium rich side of the phase diagram. In the work with calcium, a molten salt composition of 1:1 KaCl to CaC12 was found to be relatively non-hygroscop ic, allowing the experiments to be performed in an open cell. Current yields in the electrowinning of calcium were 80%; the maximum obtain-
able composition was 42 Wt.% calcium. The current yields in the lathanides were lower. The alloys produced were brittle due to intermetallic formation, but have good handling proper-
(4:15 p.m.) A COMPREHENSIVE REPRESENTATION OF PLASMA - ARC WELDPOOL INTERACTIONS IN TIG WELDING: R. Chue, R. Westhoff and Cambridge, MA
REACTIVE METALS II: Electrolytic Processes and Mg Production
(electroplating).
BREAK
J. Szekely, Dept. of Materials Science and Engineering,
have been experimentally found out. An assessment of the metallurgical quality of weld metal and the dependence of mechanical properties on pool overlaps has been made. Also, the solution of Rosenthal etc. for a moving point source of welding, which was used as a guide line in formulating the empirical model for the calculations of weld pool dimensions, is compared with the solution of empirical model for submerged arc weld pool dimensions. *Present Address : Mr. N.S . Raghavendran, Dept. of Mathematics, I.I.T .• Bombay 400076, INDIA.
ties in air.
~IT,
02139.
A formulation is developed and computed result s are presented describing the two-way interaction of plasma arcs and the weldpool behavior in tungsten inert gas (TIG) welding operations. In order to do this, the equations describing the plasma arc and those representing the transient growth and collapse of the weldpool are so l ved together. It is shown that the coupling between these two processes can markedly modify the overall behavior of the system.
(4:45 p.m.) MODELLING AND APPRAISAL OF TANDEM WIRE SUBMERGED ARC WELDING FOR THE WEL~ABILITY OF 25 mm THICK LOW CARBON STEEL PLATE IN TWO PASSES: N. S. RAGHAVENDRAN, School of En99 .& App . Sc.,UCLA, LA,CA-900 24 ASBTRACT : Based on the we ld poo I length and width ca lcu lat ions due to an empirical model for submerged arc weld metal, an elegant quantitative definition of tandem submerged arc welding is established. On the basis of this theoretical consideration optimum welding parameters for joining 25 mm thick low carbon stee I p late in two passes by tandem wire submerged arc we ld ing
(3:30 p.m.)
BREAK
(3:45 p.m . ) THE EFFECT OF SALT IMPREGNATION ON GRAPHITE WEAR RATES IN A MAGNESIUM CELL USING HYDROUS FEED . Don J. Chandler, InorganiC Research and Development, Dow Chemical Co., Freeport, TX 77541 The consumption of graphite is a major cost in the operation of an electrolytic magnesium cell when hydrous magnesium chloride is used as the feed. Investigations into methods to reduce graphite usage have been hampered by the fact that laboratory wear rates were always considerably higher than production cells. It was found in these studies that making the graphite cathodic for a short period of time at low amperage caused electrolyte to impregnate the graphite, and resulted in the graphite achieving close to theoretical density. When laboratory cells were run using salt impregnated anodes, the graphite wear rate decreased by 75%, resulting in rates comparable with production cells. Further investigation showed that production anodes are naturally impregnated in situ. A mechanism tor the cathodic impregnation has been postulated.
82-Annual Meeting (4 : 15 p.m.)
PRODUCTION OF MAGNESIUM BY VACUUM CARBOTHERMIC REDUCTION OF CALCINED DOLOMITE: Rene Winand, Van Gijsel, and L. Segers, Department of Metallurgy·Electrochemistry, Universite Libre de Bruxelles, B 1050 Brussels, BELGIUM. It is well known that MgO can be reduced by carbon at atmos· pheric pressure at temperatures in the range of 1950· to 2000·C . This project deals with the selective reduction of MgO · CaO by carbon under moderate vacuum (1 mm Hg) and at temperatures around l400·C . After a short thermodynamic introduction the experimental setup allowing a continuous production of 20 kg magnesium per day is described, including condensation on a cool surface. The best condensate obtained was 80% in magnesium metal, so that purification by sublimation was necessary to recover a molten ingot .
(3 : 30 p.m.) IRON /\NO SILICON IMPURITIES IN THE PROCESS OF IILUMINIUM ELECTROLYSIS: ~illi~~, J. \lorvath, Hungalu Engineering' Development Centre 11-1389, Budapest P.O.B. 128, Hungary.
There are iron and silicon compounds in the electrolyte of aluminium reducti o n cells in concentrations more than million times higher than that calculated using equilibrium constants of possible reaction. Amounts of iron carried away by anode gases from the electrolyte exceeds many times amount of silicon. This phenome non can be explained, based on the measurements carried out on HUNG1\LU's smelting pots and on 50-J\ laboratory cell. The concentr a tion anomaly can be caused by stabile iron and silicon c o mpo unds floating as undissolved particles in the bath
and/or di s solved In it.
lIigh iron c o ntent in dusts c a ptured by gases is generated by reactions between carbon and iron compounds . The same explanation can be applie d to enrichment of carbon foams with iron .
Environmental Matters
REDUCTION TECHNOLOGY IV:
SOLIDIFICATION BEHAVIOR OF OPTICAL AND ELECTRONIC MATERIALS IT
Sponsored by TMS Aluminum Committee
Sponsored by the TMS Solidification Committee
Room: Salon J &; K Marriott
Tuesday PM February 20, 1990
Session Chainnan: Ray Peterson, Reynolds Metals Company, P.O. Box 1200, Sheffield, AL 35660
A SURVEY ON THE PAH PROBLEM IN ALUMINIUM INDUSTRY: M. Skogland, Hydro Aluminium a.s., Karmoy Plan Is, N·4265 Haavik, Norway.
"'':.:.:.r.1'N
Polycylic Aromatic Hydrocarbons (PAH) in carbon d in aluminium smelters have strongly focused in the last ... ~ -.., ~~ ~ ~ Jives an overview of PAH in raw material •. ' ~. IIssion, effluent and solid waste from alumi,,;·· c-c.1"trt~.. ~,tl obtained by measurements in anode paste ",. ~ ~ rnaces, soderberg smelters and prebake smelter -0 ~ P .. ",e been analysed by a reference methode used in the Norwegl J; ... ," lIum industry, and now also recommended as a reference methode for the European aluminium industry. The results of the measurements are evaluated from a theoretical aspect in an attempt to give a PAH-mass-balance for the total process.
:Ol.\.
..
dS)
,:"
(2 : 30 p.m.) ANGLESEY ALUMINIUM - ENVIRONMENTAL CONTROL AND COMt.t.JNrTY RELATIONS: ~ Anglesey Aluminium Metal limited, Penrhos Works, P .O. Box 4, Holyhead Gwynedd LL &5 2 W . The paper describes the current environmental control and monitoring programme of the 120,000 TelA aluminium reduction plant located in a region with no other heavy industry and with a major interest in agriculture and tourism.
An outline is given of past environmental performance, factors which have influenced emission levels from the plant and details are presented of outplant monitoring programmes to measure the impact of the operation of the plant on the surrounding country side. The methods used to communicate details of the plant's activities and environmental performance to the general public and the farming community are discussed. Informa1ion will also be presented, in outline, of process improvemenl developments which have been introduced and the influence these have had an environmental control. Future plans to modify the process further to improve environmental performance will be outlined. (3:00 p.m.) MECHANICAL SORTING OF ALUMINUM METAL FROM SPENT POTLINING: G.!. Mathieu, Chercheur scientifique Centre canadien de la technologie des mineraux et de I'energie Laboratoire des science minerales ,552 rue Booth Onawa (Ontario) K1A 001 and J.G. Tellier, Directeur, project Recyclage de materiaux Societe d' electrolyse et de chimie Alcan Llee, C.P. 1500, Jonquiere (Quebec) G7S 4L2 and J.G. Tellier. A novel technology based on body current was developed for recovering aluminum pieces found in ALCAN spent potlining. The separation is achieved by aclion of a repulsive force produced by permanent magnets of opposite poles mounted on a rotary drum. The new apparatus can be easily inserted in the end pulley of a conveyor system at minimal operaling and capital cost. The principle of operation, Ihe design and the other features of the seperator (capacity, efficiency, etc .... ) are discussed in detail. An analysis of the important parameters of the separator is given along with typical experiments and their results. In the specific case under study, i.e. aluminum separation from the other components of spent potlining, high grade concentrates were produced wilh recovery in excess of 95%.
Tuesday PM February 20, 1990
Room: Conference Theatre Quality Hotel
Session Chainnen: N.B. Singh, Westinghouse Science & Technology Center, Pittsburgh, PA 15235; Q.T. Fang, Molten Metal Proc. Div. Alcoa Tech. Center, Alcoa, PA 15069 (2 :00 p.m.) GROHTH AND CHARACTERIZATION OF LARGE-DIMENSION LASER CRYSTALS BY THE BRIDGMAN PROCESS
L.J . Atherton, J.J. DeYoreo, D.H. Roberts, J.F. Cooper Lawrence Livermore National taboratory P.O . Box 5508, L-490 Livermore, California 94550 We are deve10pl ng the capability of producl ng 1arge-dimensi on (1 x 10 x 20 cm") laser crystals for a variety of applications, based on the Bridgman process using induction heating. The Laser Program is currently developing a laserpumped-laser approach for a multimegajoule driver. The concept is based on flashlamp excitation of LiCaA1F 6 :Cr, (LiCAF) which then pumps a neodymium-doped storage material . Our experience in growing and characterizing LiCAF will be presented, along with modeling results of induction heating and the temperature profile in the furnace and crystal . (2:40 p.m.) SOLIDIFICATION BEHAVIOR OF ELECTRONIC ~TERIALS: N. B. Singh, M. E. Glicksman, R. Mazelsky, and R. H. Hopkins: Crystal Growth and Superconductor Research Department, Westinghouse R&D, Pittsburgh, PA 15235. A representative rather than exhaustive review has been given on the solidif i cation behavior of organic nonlinear optical
crystals. Many of the developments reported here represent quick experimental responses to the rapid theoretical advances made in the field of nonlinear optics. Here, recent work by authors is reviewed on the purific ation of source material,
solidification behavior and crystal growth. Var i ous interactions between morphology , mic rosegregation, defects, growth velocity and temperature gradient are discussed. Specific examples of substituted ani lines are presented. A stronger interaction is suggested between optical physicists and crystal growers to understand the problems of inhomogeneity, stress, damage threshold and doping in the organic crystals.
(3 : 20 p. m.) VGF GROWTH OF LARGE DIAl·IETER, HIGHLY PERFECT III-V COMPOUND SEMICONDUCTOR CRYSTALS: E. 11 . Monberg, AT&T Bell Laboratories, Murray Hill, NJ 07974.
The Vertical Grad.ient Freeze (VGF) technique is used to grow high quality, lar ge diameter single crystals of GaP, InP and GaAs. The crystals are grown under low axial and radial temperature gradients. The low levels of thermoelastic stress generated using this technique result in a 100-fold reduction in dislocation levels compared to LEC grown crystals. Transmission cathoduluminescence and x-ray topography reveal a marked reduction of striations due to inhomogeneities in dopant incorporation. This is attributed to reduced temperature fluctuations and a stabilizing a xial temperature
Annual Meeting-83 gradient against buoyancy-driven convection. A 24 zone furnace has been developed to grow InP under a number of temperature profiles. The effects of various thermal configurations upon the liquid-solid interface shape is dramatically shown using infra-red imaging techniques.
(4:00 p.m.) RECENT ADVANCEMENT~ IN SINGLE CRYSTAL GROwrn IN OPTICALLY TRANSPARENT OXIDE AND FLUORIDE SYSTEMS. B.H.T. Chai, CREOL, Univ. of Central Florida, 12424 Research Parkway, Orlando, FL. 32826 The desire to obtain and ultilize highly efficient coherent light sources covering a broad spectrum of wavelengths from short UV « 200 nm) to near IR (~ 3000 nm) for a variety of applications has become one of the major driving forces for the rapid progress in the research and development of new materials in oxide and fluoride systems. In almost all applications, high optical quality single crystals are needed. Some crystals doped with active ions are used as gain media, while others are used as linear and nonlinear optical devices. I will review several of the recently developed new crystal systems. The primary consideration is the crystal structure. The crystal should have the desired symmetry and contain the specific ion or ion group. The physico-chemical stability of the compound limits the choice of growth method. While congruent melting is desirable, phase equilibrium study can help the selection of proper flux for growth. The rapid improvement of PC power and speed enhances the crystal growth control stability and reproducibility. I will discuss the feasibility of using multiple sensor inputs and control loops for single crystal growth. (4:40 p.m.) GROWTH OF LiNb0 : Hill, NJ 07974.3
C. D. Brandle, AT&T BEll Laboratories, Murray (Abstract Not Available)
SUPERPLASTICITY IN AEROSPACE ill, 2nd
International Symposium
(3:15 p.m.)
COMPUTATIONAL MODEL FOR SUPERPLASTIC PANS OF COMPLEX GEOMETRY WITH FRICTION: N. Chandra, K. Chandy, S. Ramalingam, Florida State University, Tallahassee, FL 32306 Computational process models are developed in this paper for the formation of plane-strain boxes of complex cross-sectional details. Many of the practical superplastic components manufactured in industry have sloping side walls with die bottoms either corrugated or at angles to the sides, so designated from strength considerations. A cost-effective membrane model which will be useful in the parametric study die geometry, die wall friction, initial thickness. and material property is presented in this work. Numerical results of the model are compared with continuum finite element solutions and also with experimental data. (3:45 p.m.)
NUMERICAL MODELLING OF SUPERPLASTIC FORMING OF SHEET METAL: Hong Sheng Yang and A.K. Mukherjee, University of Cal ifornia at Davis, Department of Mechanical Engineering, Division of Materials Science and Engineering, Davis, CA 95616. Superplastic forming of a circular sheet diaphragm into a circular cavity has been numerically modelled. Whereas most of the previous analyses have assumed that the surface during bulging remained spherical and/or strain-rate sensitivity (m) was independent of the strain and strain rate in the present analysiS, allowance is made for deviation from a spherical shape as experimental observation shows that the free-formed bulge is more accurately described as an ellipsoid. Results on the predicted shape of bulges suggest that lower m value of the bulging material gives rise to a more conical shape, which is consi stent with experiment observations. Furthermore, it is found that, for moderately superplastic material with a relatively low m value, the predicted thickness strain distribution is significantly less uniform if an ellipsoidal shape of bulge is allowed than that if spherical shape is assumed in the modelling. (4: 15 p.m.)
Sponsored by the TMS Shaping and Fonning Committee
EFFECT OF HYDROSTATIC PRESSURE ON SUPERPLASTIC DEFORMATION OF AL-LI ALLOY: Ho-Sung Lee and A. K. Mukherjee, Department of Mechanical Engineering, University of California, Davis, CA 95616
Tuesday PM February 20, 1990
The effect of gas pressure in minimizing cavitation and its effect on instability to failure have been studied in AlLi alloys. Good correlation was obtained between the experimentally measured and theoretically predicted values of thinning parameters and bulging heights during the superplastic biaxial deformation. Emphasis is given in comparing the instabilities encountered in uniaxial and biaxial tests. It is observed that the limit strain during biaxial forming is due to flow stress concentration in the presence of intergranular cavities. Application of hydrostatiC pressure substantially minimized the incidence of cavitation.
Room: Garden Rooms 1-2-3 Inn at the Park
Session Chainnan: H.C. Heikkenen, General Dynamics, P.O. Box 748, Fort Worth, TX. 76101 (2:00 p.m.)
MODELING OF SUPERPLASTIC FLOW IN HETEROGENEOUS MATERIALS, AK Ghosh, Department of Materials Science and Engineering, University of Michigan, Ann Arbor, MI 48109·2136. Many of the superplastic materials of engineering importance are two-phase deformable alloys or composites containing nondeformable phases dispersed in a deformable alloy matrix. The deformation of such materials is analyzed by simple modeling techniques, and the resulting constitutive behavior and flow localization processes are analyzed. The importance of the basic flow parameters of the constituent phases on the overall material response is described. Comparison with experimental results are presented for selected material systems. (2: 45 p.m.) MECHANICAL CHARACTERIZATION OF SUPERPLASTIC MATERIALS USING MODIFIED CONE TEST: N. Chandra, D. Kannan, Florida State University, Tallahassee, FL 32306, R.E. Goforth, Texas A & M University, College Station, TX, 77840, L. Phillips, LTV Aerospace Corporation
The establishment of material parameters under multiaxial loading condi tions is critical in the determination of the forming behavior of superplastic materials. A constant angle cone test was originally designed to achieve constant true t during the test. Analysis shows that this condition is realized only approximately, since the variation of strain in the unsupported region has not been taken into account in the original design. In this paper, a comprehensive analysis is used to model the deformation into a cone with varying angles which produces a constant value of strainrate. Also a new method to determine variable optimum t is suggested which is useful for materials that undergo dynamic recrystallization. These procedures are applied in the mechanical characterization of superplastic Al and Al-Li alloys.
This work was supported by AFOSR grant No. AFOSR-86-0091 p.m.) THE APPLICATION OF FRACTAL IN THE FRACTURE OF SUPERPLASTIC DEFORMATION: Jiang Xinggang, Chui Jianzong and Ma Longxang, Department of Metal Forming, Northeast University of Technology, Shen Yang, P R C
(4: 45
In this paper, Fractal is used to describe the fracture behavior of superplastic deformation of high strength aluminum alloy. After measuring the fractal dimensions of the fracture surfaces, it is found that under the same deformation condition, when the initial grain size increases, the fractal dimension of fracture surface decreases and the total elongation decreases. So we may estimate the superplasticity from measuring the fractal dimensions of fracture of the specimens. The higher the fractal dimension of the fracture surface, the better the superplasticity of the alloy.
84-Annual Meeting
SURFACE INTERACTIONS AND CORROSION OF REFRACTORY METALS AND ALLOYS IN LIQUID METAUVAPOR METAL SYSTEMS Sponsored by: TMS Refractory Metals Committee and TMS Corrosion and Environmental Effects Committee Tuesday PM February 20, 1990
Room: Plaza A Inn at the Park
Session Chainnen: L.B. Lundberg, Los Alamos National Lab., P.O. Box 1663, MS-J576, Los Alamos, NM 87545; M.G. Adamson, Lawrence Livennore National Lab., P.O. Box 808 L-350, Livennore, CA 94550 (2:00 p.m.)
MODELING THE BULK AND SURFACE PROPERTIES OF LIQUID METAL SOLUTIONS: Wayne J. Howell, IBM Corp., Hopewell Junction, NY and Charles A Eckert, Georgia Institute of Technology, Atlanta, GA. Liquid metal solutions have unique properties that are proving useful in improving existing materials technology and in developing new technologies. Important applications of liquid metals include use as solvents for reduction of metal ores, separations and purifications, liquid-phase sintering of refractory metals, rapid solidification of metastable alloy phases to create metallic glasses, soldering and welding, and production of composite materials. Process development for these and other applications requires an ability to model the thermodynamic properties of liquid alloy systems. We have developed an approach to modeling the bulk and surface phase thermodynamic properties (e.g. Gibbs free energy, enthalpy, entropy, excess volume, structure factor, and surface tension) of liquid alloy systems which exhibit evidence of intermetallic compound formation. The inclusion of both chemical forces (which represent liquid-phase compound formation) and physical forces have resulted in models which are able to represent the properties of a wide range of systems. The number of model parameters is minimized by exploiting newly . developed relationships between the thermodynamic properties.
(2:30p.m . ) Embrittlement by liquid U in Group VB and VIB Metals During Tensile loading at 1473 K: J.S. Huang and G.F. Gallegos, lawrence livermore National laboratory, livermore, CA 94550. An investigation was conducted into the effects on tensile behavior when Ta, W, V, Nb were exposed to liquid U. Although Ta and Wwere embrittled with negligible reduction of crosssection, V and Nb remained ductile with the samples completely necking down to sharp tips. Embrittlemp.nt in Ta and Wwas associated with intergranular penetration of U, and the fracture surfaces showed evidence of grain boundary dissolution at low strain rate. At higher strain rate, however, this evidence was less obvious. Current results indicate that intergranular penetration of liquid U and tensile embrittlement in Groups VB and VIB metals were favored by systems that have low mutual solubility between the solid and liquid elements. It was found that the thermodynamic criterion for intergranular penetration of liquid into a solid (grain boundary energy <2X liquid-solid interfacial energy) only applies to these systems. Work performed under the auspices of the U.S. Dept . of Energy by lawrence livermore National lab. under contract #W-7405-Eng-4B. (3:00 p.m.) INTERACTIONS BETWEEN NIOBIUM AND LEAD AT 1575 TO 1874 K: Lynn B Lundber~, Los Alamos National Laboratory, P.O. Box 1663, Mail Stop J576, Los Alamos, NM 87544. Niobium and Nb-5.4 at.% Hf-2.0 at.% Ti (C103) capsules were filled with lead and heated isothermally at temperatures from 1575 to 1874 K for 4 to 8 hours to study the effects of liquid and vaporous lead on these materials. Gross reactions were not observed in either pure niobium or CI03. However, both niobium and C\03 were found to dissolve in liquid lead to levels as high as 0.6 at. %, and lead was found to dissolve in both solid materials to levels as high as 3 at. %. A tentative Nb-Pb phase diagram is proposed for the experimental temperature range. Only terminal solutions were observed - solid at the niobium-rich end of the phase diagram and liquid at the lead-rich end. Diffusion of lead into both solid Nb and CI03 was observed by electron microprobe analysis, and consistent diffusion coefficients were calculated from the determined concentration/position data and from a consideration of the tentative Nb-Pb terminal solid solution phase boundary. There were only minor variations in the diffusion and phase equilibria behavior between pure niobium and C103.
( 3:30 p.m.)
BREAK
(4 : 00 p.m.)
THE INFLUENCE OF THE CONCENTRATION OF NONMETALS IN LITHIUM ON THE REACTIONS WITH THE ALLOY V-ISi-3Ti: Hans U Bor~stedt and 1. Konys, Kernforschungszentrum Karlsruhe GmbH., Instit.ut fUr Material- und Festkorperforschung, P.O. Box 3640, 0-7500 Karlsruhe I, Fed. Rep. of Germany. Reactions between the vanadium-base alloy V-lSi-3Ti and lithium were studied at 550°C in a forced convection loop made of stabilized austenitic stainless steel. Pre-heating the lithium in contact with titanium sponge reduced the nitrogen and carbon levels below 100 wppm. In spite of the low concentrations of non-metallic elements, the vanadium alloy picked-up these contaminants producing a hardened surface layer. Nitrogen concentrations in the range 10 to 70 wppm significantly influenced the dissolution rate of the solid alloy. Carbon uptake did not reach these levels. Surfaces of the alloy remained unprotected at the lowest nitrogen levels in the lithium. At higher contents, protecting carbonitrides were detected at the surface of the alloy reducing the dissolution rate during the corrosion test. Dissolution of vanadium and the minor alloying elements became evident when nitrogen concentration was increased further. Material loss from V-ISi-3Ti was only one tenth of that seen from either austenitic or ferritic stainless steels at the same temperature and flow velocity. (4:30p.m.)
COMPATIBILITY STUDY OF BN INSULATOR COATING WITH LIQUID LITHIUM AND STRUCTURAL MATERIALS;ll.....L ~ A. Hull, and D. L. Smith, Argonne National Laboratory, Materials and Components Technology Div., 9700 So. Cass Ave., Argonne, IL 60439. Liquid metals such as lithium and .t~e lithium-I~ad eu.tectic are attractive materials for both the coolant and tnttum breeder 1R fUSIOn reactor blankets. However, one of the major concerns for liquid metal blankets is the magnetohydrodynamic (MHO) forces and their subsequent influence on thermal hydraulics and corrosion. One possible approach to counteract these adverse MHO forces is to place an insulator coating at the liquid metaVstructural material interface and thus eliminate MHO-generated currents from passing through the structural walls. Based on a previous thermodynamic analysis, boron nitride and a few other metal oxide materials were rated as the most promising coating candidates in such environments. In this study, a ferritic substrate alloy, HT-9, was coated with a thin layer of BN and immersed in flowing liquid lithium at 400°C. The results of studies of chemical compatibility with liquid lithium. electrical conductivity, and the physical stability will be presented.
THERMODYNAMICS AND PHASE RELATIONS OF EXOTIC MATERIALS II:
Intermetallics
Sponsored by the ASM-MSD Thermodynamic and Alloy Phases Committee Tuesday PM February 20, 1990
Room: Plaza C Inn at the Park
Session Chainnan: R.B. Schwarz, Center for Materials Science, Los Alamos National Laboratory, Los Alamos, NM 87545 (2:00 p.m.) FORMATION OF METASTABLE L12 PHASES IN AL3Zr INTERHETALLICS:* P S Desch and R. S. Schwarz, Center for Materials Science. Los AlaJIos National Laboratory, Los Alamos, NM 87545; Philip Nash. Illinois Institute of Technology, Chicago, IL 60616 . Alloys of A15XZr2' where X is AI, Li, or Cu, are stable in the tetragonal D0 23 structure. We prepared powders of these alloys having the metastable cubic L12 phase by mechanically alloying mixtures of the elemental powders in an inert atmosphere. Differential scanning calorimetry was used to determine the thermal stabilities of the Ll2 phases. We have correlated the L12 - D0 23 transformation temperatures and transformation enthalpies with the thermodynamic properties of the element X added to Al 3Zr. * Work supported by the U. S. Department of Ener,gy, Office of Basic Energy Sciences.
(2 :30 p.m.) EFFECT OF NITROGEN ON a STABLIZIATION/NUCLEATION IN Ti-AL PHASE DIAGRAM. Rlmal Kad, and Ben F. Oliver, The University of Tennessee, Knoxville, TN 37996-2200. Stable a formation from Ti-50at%AI melt has been reported 1. While a can be nucleated under phase equilibria conditions, there are additional a stabilizing/nucleating events made possible by the presence of nitrogen in the melt. Ti-54at%Al alloys were arc cast (moderate growth rate) or
Annual Meeting-85 directionally solidified (very slow growth rate) with a varying N content. As cast solidification structures show increasing volume fraction of a with increasing N content. Presence of N or nitrides forming from the melt preferentially nucleates a and not y-TiAI. The homogenized alloys are single phase y with a distribution of Ti2A1N nitride particles. Resulting microstructures are reconciled with the solidification rate and the N content. Results indicate that the two parameters may be interdependent Optical, SEM, AEM data in support of the solidification path arguement will be reported. 1. C. McCullough et al., Scripta Met. vol. 22, p. 1131 (1988).
(3:00p.m.) A STUDY OF a/y PERITECTIC OSCILLATIONS USING CONTAINERLESS LEVITATION MELTING FOR HIGH TEMPERATURE Ti-AI PHASE DIAGRAM. Ben F. Oliver and Bimal Kad, The University of Tennessee, Knoxville, TN 37996-2200. Oscillating peritectic solidification structures have been reported for low melting point alloys. Similar studies for high temperature peritectic phase fields are limited because of the lack of encapsulating media and the reactive nature of the melt. A container less levitated zone melter was used to study peritectic solidification in Ti-AI phase diagram. A a+Liq=y phase field has been suggested [1] for the Ti-AI phase diagram. However, the most recent phase diagram [2] does not report this phase field. Controlled solidification experiments with an essentially flat interface reveal direct solidification structure evidence for such an oscillating peritectic reaction. Process fundamentals, solidification parameters and resulting microstructures will be reported. 1. C. McCullough et al., Scripta Met. vol. 22, p. 1131 (1988). 2. J. L. Murray, "Phase Diagrams of Binary Titanium Alloys." ASM Int., 1987.
I
WEDNESDAY AM ADVANCED METAL & CERAMIC MATRIX COMPOSITES IV: Interface and High Temperature Performance Sponsored by the TMS Powder Metallurgy Committee, TMS Physical Metallurgy Committee, and ASM-TMS Composites Committee Wednesday AM February 21, 1990
Room: ParkD
Inn at the Park
Session Chairmen: Ann M. Ritter, General Electric Corp. R&D, Schenectady, NY 12301; Rizhang Zhu, Beijing University of Science and Technology, Department of Surface Science and Corrosion Engineering, 30, Xue Yuan Lu Street, Beijing, China
(3:30 p.m.) PHASE STABILITY OF ION IRRADIATED NI2AL3:
Michael Nastasi.
Los Alamos National Laboratory. Los Alamos. NM 87545. P. R. Okamoto. Argonne National Laboratory. Argonne. IL 60439. R. S. Averback. University of Illinois at Urbana. Urbana-champaign. IL 61801. and J. W. Mayer. Cornell University. Ithaca. NY 14853. The low temperature phase stability of thin film Ni2Ala under 400 keV Kr+ion irradiation was examined in the Argonne National Laboratory high voltage electron microscope (HVEM). At 20 K the disordering transformation from Ni2Ala to CsCI to bec was observed during an exposure to 7.5 x 1013 Kr/cm2 . A combination of amorphous and bec phases was observed at 1.5 x 1014 Kr/cm2 with complete amorphization occurring after 3 x 1014 Kr/cm2 . This transformations sequence indicates that the free energy of the amorphous state lies above the disordered state and that additional forms of radiation-induced damage beyond chemical disordering are required for the amorphization of Ni2A13.
(4:00 p.m.) PHASE EQUILIBRIA IN THE Ni-Sb SYSTEM: G.H. Cha and P. Nash, Department of Metallurgical and Materials Engineerin~nois Institution of Technology, Chicago, IL 60616 A portion of the Ni-Al-Sb phase diagram containing from 15 to 45 atomic % Al and 5 to 15 atomic % Sb has been studied by x-ray diffraction, optical and electron microscopy, energy dispersive spectrometry and microhardness. Partial isothermal sections have been established from samples annealed at 1000 0 C and 1200 0 C for 7 days. Regions of primary solidification have been established from examination of the as-cast vacuum induction melted ingots. Preliminary data on the mechanical properties will also be presented and correlated with the microstructures.
TMS AWARDS DINNERIRECEPTION
(8:30 a.m.) INTERFACIAL REACTIONS IN SiC FIBER-REINFORCED TITANIUM ALLOY AND TITANIUM ALUMINIDE COMPOSITES: G. Das, Metcut-Materials Research Group, P.O. Box 33511, Wright-Patterson AFB, OH 45433-0511 SiC fiber reinforced metal matrix composites based on Cl-Ti, 8-Ti, Ti3Al ("2)' and Ti-Al (y) matrices were prepared by HIP'ing matrix powders (RSP) with SiC fibers. As-HIP'd specimens
were
exposed
to
elevated
temperatures
various
Microstructural characterization of reaction
zones in these composites was done by a combination of optical microscopy, SEM, EMPA, TEM, and EDS techniques. Hardness variations
in the matrix adjacent
to
the
reaction zone were
measured by microhardness indentation technique. Results will be discussed in terms of stability of reaction zones in these composites, load transfer capabilities of the matrix to the fiber and designing matrices suitable for composites.
(8:55 a.m.) HIGH-TEMPERATURE INTERMETALLIC MATRIX COMPOSITES FORMED BY HIGH-ENERGY HIGH-RATE P/M PROCESSING: B.-H. Lee. C. Persad. Z. Eliezer, and H.L. Marcus. Center for Materials Science and Engineering, The University of Texas, Austin, TX 78712. Intermetallic matrix composites containing discontinuous ceramic or intermetallic particulate reinforcements have been formed by high-energy high-rate powder consolidation.
BCC-stabilized Ti3A1 powder matrices have been consolidated with second phases including SiC, TiB 2 , and TiAl powders. Processing was accomplished by discharge of a single high electrical energy pulse through a powder blend under pressure. Control of energy input produced solid-state and liquid-phase-assisted consolidation. The high-temperature stability of the matrix/reinforcement interfaces was compared by annealing treatments of up to 100 hours at 1073K - 1273K. X-Ray diffraction, analytical electron microscopy, and microhardness profiling were employed in characterizing the interfacial phases. This work was supported by DARPA under ARO Contract
Tuesday, February 20, 1990 6:00 PM Reception 7:00 PM Dinner Marriott Hotel Room: Gand Ballroom E & F
for
times to determine thermal stability of reaction zones and their growth behavior.
DAAL 0387·K·0073,
(9: 12 a.m.) AmThffiru~SnmYOFmGHTIThW~~CRffiW
DEFORMATION IN METAL MA1RIX COMPOSITES, T L Dragone and W.D. Nix, Department of Materials Science and Engineering, Stanford University, Stanford, CA 94305 The creep defonnation behavior of metal matrix composite systems has been srudied using finite element techniques with the objective of understanding the underlying mechanisms which strengthen fiber reinforced materials at high temperatures. Axisymmetric and plane strain unit cells are used to model the material which consists of stiff, elastic fibers in an elastic, creeping matrix which follows a simple power law constirutive behavior. Results indicate that large triaxial stresses are induced in the matrix due to the constraint imposed by the more rigid fibers, and that these stresses have a strong effect on reducing the creep rate for the composite. The effect of reinforcement phase geometry on the overall defonnation rate is investigated with particular
8S-Annual Meeting
emphasis on fiber aspect ratio, unit cell geometry, and various forms of particle c:Iustering. Theoretical predictions from this modelling are compared to expenmental results of creep deformation in metal matrix composite systems. (9:29 a.m.) MICROSTRUCTURAL ASPECTS OF CREEP IN SiC WHISKER-REINFORCED A1 20 3 : J.R. Porter, Rockwell International Science Center, Thousand Oaks, CA 91360; K. Xia, Comalco Research Centre, Thomastown, Victoria 3074, Australia; T.G. Langdon, University of Southern California, Los Angeles, CA 90089-1453. The incorporation of SiC whisker reinforcement dramatically reduces creep deformation rates in A1203_ However, the amount of creep rate reduction depends on the processing method used to fabricate the composite. Creep rates at l500°C can be reduced by up to two orders of magnitude in conventionally processed SiC whisker-reinforced Al 20 3 and by up to four orders of magnitude in SiC whisker-reinforced A1203 processed by using a colloidal method. A mechanical testing program showed that there is a transition to tertiary creep and failure in the conventionally processed composites whereas the creep rate tends to decrease continuously with time in the dispersion processed composites. This paper describes the mechanical properties of the composites and the microstructural characteristics after deformation.
(9 :46 a.m.) MECHANISMS FOR THE EARLY ONSET OF TERTIARY CREEP IN A SILICON CARBIDE REINFORCED ALUMINUM COMPOSITE, E. P. Barth, J. T. Morton M. W. Kopp, and J. K. Tien, Strategic Materials R&D Laboratory, Materials Science and Engineering Center, University of Texas,
78712.
It has been previously demonstrated that above a threshold
stress P/M silicon carbide reinforced aluminum composites do not exhibit conventional creep behavior, but rather, immediately enter the tertiary creep regime following primary deformation. Several mechanisms for this phenomenon present themselves for consideration. These include local plastic failure of the matrix near the whiskers, void growth by power law creep of the matrix, and void nucleation and growth by stress assisted vacancy coalescence. The various candidate mechanisms will be
discussed in light of the observed deformation behavior and SID! and TEM microstructural analysis. The implications of this behavior on design and applications will also be discussed.
(10:03 a.m.) Development of Mi-Aluminide Based Ca.posites for Elevated Temperature Applications; G. E. Fuchs*, The Aerospace Corp., Los Angeles, CA, 90009. *Current Mailing Address: General Electric Co., Rm. Al-1Z3, P.O. Box 1072, Schenectady, NY lZ30l. mechanical properties and chemical compatibility of Ni3Al-based composites have been examined. The
are
based
on
IC-ZZI
(Ni-16AI-8Cr-IZr-0.IB)
with
AlZ03, TiC and TiB2 particulate reinforcement. Powder metallurgy techniques were used to fabricate the intermetallic matrix composites (IMC ' s). The resulting microstructures are discussed with respect to the processing parameters. In addition, the reinforcement/matrix interfacial reaction has been characterized
at 1373K for up to 1000 hours.
(10:54 a.m.)
In the present work, aluminum alloy 6061 was reinforced with particulate silicon-carbide (SiC). An evaluation of the feasibility of near-net shape HIP processing was achieved by the examination of the structure and properties of the resulting composite materials.
Metallurgical features and also optimal densification have been shown to produce considerable improvements in the properties when
compared with extruded materials.
This relationship between
structure and properties has been clarified using scanning and transmission electron microscopy of prior particle boundaries, metal-ceramic interfaces, and the bulk structure of the matrix.
*BillitonResearch is the central laboratory of the Metals Sector of the Royal Dutch/Shell Group of Companies.
(11: 11 a.m.)
Scientific Research.
composites
Effect of tantalum, as alloying element, on the kinetics and mechanism of interaction of tungsten fiber in W-Ni and W-Fe composites, fabricated using P/M processing, have been studied at 1l00-1300°C for 1-50 hrs. In order to establ ish thermodynamic equilibrium between the W-fiber and Ta-alloyed Ni- and Fe-base matrices, Ni-W-Ta and Fe-W-Ta tarnary phase diagrams, investigated by the author, have been utilized. Thi s revea 1ed tha.t f.ormati on of i ntermeta 11 i c compound Ni .. W at the flber-matrlx lnterface can be avoided and equilibrium between the Ni-base matrix and W-fiber can be achieved by alloying these matrices with 2 at. % Ta and raising the temperature of the sample rapidly. Tantalum has been found effective in decreasing the diffusion of Ni and Fe atoms into the W-fiber and supressing its recrystallization. The effectiveness of tantalum has been explained on the basis of Fermi energy level and electron density Nd (E.).
INTERFACIAL REACTIONS IN HIGH TEMPERATURE METALLIC, INTERMETALLIC, AND CERAMIC MATRIX COMPOSITES: A STATUS REVIEW, J.K. Tien and M.W. Kopp, Strategic Materials R&D Laboratory, The University of Texas at Austin, Austin, TX 78712.
This work has been supported by the Air Force Office of
The several
INTERACTION OF TUNGSTEN FIBER WITH TANTALUM ALLOYED Ni - AND Fe-BASE MATRICES DURING HEAT TREATMENT: N.Asrar, Research Centre for Iron & Steel, Steel Authority of India Ltd., Ranchi-834002, India.
& Development
THE STRUCTURE AND PROPERTIES OF PARTICULATE-SiC/Al ALLOY 6061 METAL-MATRIX COMPOSITE MANUFACTURED VIA HIP. M. A. Sargent, C. Rensen, W. H. M. Alsen, Billiton Research BV, Arnhem*, Post Office Box 40, 6800 AA Arnhem, The Netherlands.
l
Austin, TX
(10:37 a.m.)
The TiBZ reacted extensively
with the matrix. No particulate/matrix reaction was observed in either the TiC or the AlZ03 composites. Tensile tests were conducted on the AlZ03 and the TiC reinforced composites in the temperature range 298-lZ73K. Low tensile strengths and
Efforts to extend the operating temperatures of today's gas turbine engines beyond that permitted by single crystal superalloys have led to an emphasis on metalliC, intermetallic, and ceramic matrix composites (MMC's, IMC's, and CMC's). Although there are several structural benefits that are derived from utilizing a composite architecture, additional problems associated with multicomponent materials present themselves. These potential barriers that must be addressed include, but may not be limited to, fiber/matrix bonding, coefficient of thermal expansion mismatch, environmental resistance of both fiber and matrix materials, and thermally activated processes affecting both fiber and matrix components. This paper reviews recent work in the latter area of interfacial reactions in composite systems that are intended for high temperature applications. Examples have been chosen to illustrate the behavior of metal, intermetaillc, and ceramic reinforcements in each of the three classes of matrix materials.
reinforcement/matrix bonding. The TiC-IMC exhibited tensile properties superior to those of conventional Ni-base superalloy~.
ALUMINA & BAUXITE IV: Alumina and Chemicals
(10:20 a.m.)
Sponsored by TMS Aluminum Committee
CONTROL OF INTERFACE ROUGHNESS IN A1 2 0,-Cr2 0, COMPOSITES REINFORCED WITH A1 2 0, FIBERS AND Cr INCLUSIONS. Carol A. Handwerker, David C. Cranmer, and Uday Deshmukh, National Institute of Standards and Technology, Gaithersburg, MD 20899
Wednesday AM February 21, 1990
ductilities
were
observed
in
the
A1Z03/IC-Z21
due
to
poor
Composites with controlled fiber roughness and volume fraction of reinforcing metal have been fabricated by hot pressing
solid
solution A1 2 0,-Cr 2 0, particulate matrices with Al 2 0,
fibers at various oxygen partial pressures.
Room: Salon 3
Marriott
Session Chainnan: M.V. Chaubal, Reynolds Metals Co., Corpus Christi, TX 78440
By exploiting
features of the equilibrium AI-Cr-O phase diagram and the surface facetting phase diagram, the fiber interface debonding strength can be tailored. Processing conditions, microstructures, and mechanical properties of the composites
will be presented. Research sponsored in part by the Office of Naval Research.
(8:30 a.m.) EFFECT OF TEMPERATURE, IMPURITIES, RETENTION TIME AND SEEDING ON THE RATE OF CRYSTAL GROWTH, NUCLEATION & QUALITY OF ALUMINA HYDRATE DURING PRECIPITATION: B.K. Satapathy and P. Vidyasagar, R&D Laboratory, Nalco, Orissa, India. Grain size control for hydrate/alumina is quite vital for its value in international market. Studies have b~en made on the
Annual Meeting-87 effect of temperature, impurities (with reference to oxallic carbon), retention time and seeding on the grain size, nucleation and surface area of the hydrate during precipitation. The rate of crystal growth of 0.5 to 0.7 micron/day (for our plant) has been established at constant temperature. Nucleation of 1-2 x e S at 3.55 micron has been found to produce coarse hydrate. Oxallic carbon and seeding with fine hydrate increases fines in precipitation.
(9:00 a.m.)
The KgP2 AID AlF3 IIPLUEICE 01 Q Alz03 FORMATIOI AS WELL AS FORK AID SIZE OF GRAIIS II CALCIIATID. PROCESS. A.DerdAcka Grzymek, Z.Kanlk, A.Stak, J. Iwanclw.Academy of Klnlng and Ketallurgy, Krak6w, Poland.
The influenle of "gF. lnd AIF, lineralizer! on I AhO, forlation II well u on fori and size of grlins in the lallination proltss WlS investigated. The industrill Al(OHlo frol 6romwile plant nllr Opole WlS lallintd in the Inpmtures 1173, 1523, 1573, 1623 K in 1 hour with addition of 0.1 and 0.2 I of "gF, and AIF" The aiuliniul hydroKide frol 6ro§zowilt, oblaintd by sinter-disintegntion J,6rzyui's lethod, lIinly ollurs in fori of hydmgili\e, The Slipies me uilined by Ihe IRO, SE" and under optiul licr05!ope, Ihe specyfic gravity WlS delerlined 100, The IRO analysis reveal Ihal pure I AhO, phue WlS obhined in 1523 K \eopenture, bolh for lhe nlples with and without linenlizer. In the noplt without lintnliur cllcined in Ihe Itlperalure 1473 K Ihe I AhO, phase is absenl, lhe lransienl fom of AhO. - I, e, I oClured only, On the other hand the sllpln lith addition of lintralizer conhin a grut lIount of I AhO., The lid ... specific gravity (om 4g/n') liS obtained for two nlpln lith 0,2 I addition of "gF. and m., Invntigated Sllpln differ tach other in shape and size of gnins. In the higher Itlpenturll of calcination anc! with the gruler addition of lineralim, the lost regular grain habit lIS observed. The relationship between the grain silt of I AhO. and Ihe ulcination lelpenture and the Iype and quantity of lineralim,
(9:30 a.m.) DEVELOPMENT AND PRODUCTION OF NON METALLURGICAL ALUMINAS IN HUNGARY: M. Toth Benjaminn~ J. Csige J. M~tytsi, Aluterv-KI (Hungalu) Hungalu Engineering and Development Centre, H-1389, P.O.B. 128, Budapest, Hungary. In addition to metallurgical uses the Hungarian alumina production the ambition to produce aluminium-oxid for other purposes has recently been prassed. This paper describes: -the motive factors of this activity -the activity till now -and (mainly) the results of actual research and development works (10:00 a.m.) THE PRODUCTION OF ALUMINA CATALYST SUBSTRATE FROM ALUMINO-SILICATE PRECURSORS: ~ and K. Perry, Council for Mineral Technology (MINTEK), 200 Hand Strijom
Road, Randburg, 2123 South Africa, South Africa.
The manufacture of sintered fine-grained alumina catalyst substrate from aluminosilicates other than bauxite, namely andalusites and kaolin, was investigated. for andalusi te the reaction that was considered was AI,:l.SIOG" + C -i AlA 0,..1 + SIO (g) + CO (g) which is thermodynamically favorable at temperatures above 1500 C. Pellets comprising of a stoichiometric mixture of alumino-silicate and reductant were heated in a bed of carbon granules for various periods of time at temperatures ranging between 1500 and 1800' C and above. Using optical microscopy, xray diffraction and scanning electron microscopy it was found that porous aluQlina having a purity in excess of 90 per cent can be produced; but only at the highest temperature or at lower temperatures under conditions of extended heating times. High temperatures also favor crystal growth which is required for the sintering of the alumina particles - so imparting mechanical strength to the pellets. Measurements indicate that the alumina pellets have high porosity (approx. 50%), good thermal shock resistance, adequate mechanical strength (2.5kN) as well as a large surface area - characteristics which indicate that they would be ideal materials for catalyst substrate. 0
lower temperatures favor the formation of the intermediate namely sintered mulli teo It was found that, at these temperatures, the production of composite pellets comprising of clusters of alumina particles embedded in a matrix of mullite and produced from alumino-silicate precursors resulted in catalyst substrate pellets also having "cceptable physical properties.
(10:30 a.m.) THE PRODUCTION OF COARSE-GRAINED MOSAIC ALUMINIUM TRIHYDROXIDE FROM BALL-MILLED SEED: N. Brown, VAW, Vereinigte AluminiumWerke AG-Berlin, Postfach 24-68, D-5300 Bonn 1, Germany. Coarse-grained aluminium trihydroxide of mosaic particle structure is produced in a three-stage precipitation process starting from ball-milled seed. The three stages of precipitation are designated seed deactivation, agglomeration and crystal growth. Since the starting seed for the process is created mechanically, the fines creation step is taken out of the precipitators thus eliminating the need for product stream classification. The 'once-through' nature of the process leads to the production of strong, coarse particles of aluminium trihydroxide at high liquor productivity, while reducing the overall dimensions of the precipitation circuit with savings in capital investment and energy consumption.
(11 :00 a.m.) THE DEVELOPMENT OF SIMPLE, REAGENT-TECHNIQUES TO IMPROVE ALUMINA PRODUCTION: M.E. Keeney, Curtin University of Tech., School of Applied Chemistry, Bentley WA, 6102, Australia. If properly implemented, chemical reagents can significantly improve process operations in alumina refining. To optimise reagent performance, however, it is important to understand both the chemical and engineering fundamentals of the particular process under modification. This paper describes some of the areas of alumina refining under investigation by the IMRL, in collaboration with the alumina industry, in which simple, reagent techniques are being employed to dramatically alter process operations. These areas include drainage aid and antifoam development, hydrate and oxalate crystal modification, and DSP scale control. The underlying mechanisms being altered by the reagents to control the processes are described.
CARBON TECHNOLOGY V Sponsored by TMS Aluminum Committee (formerly Light Metals) Wednesday AM February 21, 1990
Room: Salon 2 Marriott
Session Chainnen: Euel R. Cutshall, Manufacturing Technology Laboratory, Reynolds Metals Company, Sheffield, AL 35660; David M. Leslie, Alumax, 800 Rene Levesque Blvd. West, Suite 2950 Montreal, Quebec, Canada H3B lZl (8: 30 a.m.)
PROPERTIES OF BENCH SCALE ANODES PRODUCED USING BINDER PITCHES VARYING IN PRIMARY AND SECONDARY QI CONTENTS, David Belitskus, Walter W. Hill, Aluminum Company of America, Alcoa Laboratories, Alcoa Center, PA, 15069, USA Although it has been known for a long time that electrode binder pitches contain materials that are insoluble in a strong solvent such as quinoline, it has been more recently established that quinoline insoluble (QI) material can be of fundamentally different types, 'Primary QI' is a carbon black·like material originating from the gas phase during coal coking, from which tar (subsequently distilled to produce pitCh) is a by· product. 'Secondary QI' is mesophase formed from the liquid phase during thermal treatment of tar or pitch. The work described in this paper was carried out to help quantify effects of secondary QI on aluminum production cell anode properties. Excess consumption during electrolysis of bench scale anodes produced using experimental pitches varying in primary and secondary QI contents decreased with increasing total QI content, irrespective of type, up to about 9 wt.% total QI. The secondary QI increased binder requirement and resulted in a slight increase in electrical resistivity. The effects can be attributed to wettability and coke yield differences. (9: 00 a.m.)
PERFORMANCE OF BINDER PITCHES WITH DECREASED QI-CONTENT IN ANODE MAKING: FORMATION, NATURE, PROPERTIES AND SUBSTITUTION OF QUINOUNE INSOLUBLES, Dr. Arnold Alscher, Dr. Rolf WildfOrster, Verkaufsgesellschaft fur Teererzeugnisse (VIT mbH), Varziner Str. 49, 0-4100 Duisburg 12, West Germany. Five different kinds of quinoline insolubles in binder pitches were investigated concerning their influence on the performance in anode making .- primary QI, secondary QI, coaVcoke breeze, carbon black, and ash. Experimental and commercial pitches were fully characterized using conventional anode binder specification tests and some additional tests including wetting behaviour, ageing stability, mesophase content and inorganic material. Bench·scale test anodes were produced using a prebaked·type anode formulation with petroleum coke. Their chemical, electrical and mechanical propenies were measured. The results indicate that ,binder pitches with 6-9% QI can provide optimum wetting propenies and thus can be used to produce baked carbons of high performance. (9:30 a.m.)
QI IN COAL TAR PITCHES, PART 1 - INVESTIGATION OF COKE OVEN PARAMETERS ACTING ON TAR CHARACfERlSTlCS, P. Lagassie, J. L. Saint Romain, Norsolor, H.G.D. Dept., BP 6 - 62880 Vendin Ie Vieil, France; R. Bertau, B. Souffrey, Centre de Pyrolyse de Marienau, Usine de Marienau, 57612 Forbach Cede.. France. The Objective of this paper is to understand through the different studi::s published in literature, the parameters acting on the coke oven tar propenies. A special attention is
8S-Annual Meeting devoted to quinoline insolubles which are considered today as a very important characteristic of binder pitches used by the aluminium industry. Quinoline insolubles present in tars, and pitches derived from them, are classified as spheroidal primary QI, carried over particles, and mesophase. The influence of the coke oven design and operating conditions are discussed. QI levels as well as tar molecular structure are related to the different reactions occurring during fortnation, cracking and collection of tar vapours. (10:00 a.m.)
QI IN COAL TAR PITCHES, PART 2 • PROGRAM DEVELOPED BY HGD TO UNDERSTAND, TO FOLLOW AND TO ANTICIPATE THE COKE OVEN TARS EVOLUTION, J. L Saint Romain, P. Lagassie, Norsolor, H.G.D. Department, BP 6 62880 Verdin Ie Vieil, France. The coke oven design and operating conditions are changing, which results in variations of the tars produced. In order to understand and to follow the evolution of the tars processed in its plants, HGD has developed a special program. Analytical techniques such as NMR, liquid chromatography and scanning electronic microscopy have been used to investigate the structure of different tars. Correlations between the characteristics of these tars and the properties of the pitches derived from them have been established. Statistical quality control has been realized in order to detec.t any change in tar and pitch quality and to anticipate further evolutions. (10: 30 a.m.)
SULPHUR LEVELS IN PETROLEUM COKE: A CRISIS FOR ALUMINUM SMELTING?, Franz Vogt, Rod Zabreznik, Calciner Industries, Inc., P. O. Box 1306, Chalmette, LA 70044; James Waller, The Pace Consultants, Inc., P. O. Box 53473, Houston, TX 77052-3473. SUlphur levels in petroleum coke used in aluminum smelting have increased about 12% since 1986. This increase is due to developments in the world petroleum refining industry. The outlook for this industry supports a continuation of this trend. The increased demand for petroleum coke by aluminum smelters around the world will strain available lower sulfur (less than 2.5%) green coke sources available for calcining. Concurrently, environmental requirements for the smelters will tighten the allowable sulphur emissions. This industry must either obtain lower sulphur raw materials or install costly S02 scrubbing systems. This presentation surveys the trends in coke sulphur content and outlines the alternatives for coping with these trends. These alternatives include possible solutions in the refining, calcining and smelting industries.
(11 :00 a.m.)
LABORATORY EVALUATION OF A LOW QUINOLINE INSOLUBLES COAL-TAR PITCH AS ANODE BINDER, P. J. Rhedey, A1can International Limited, Kingston Research and Development Centre, P. O. Box 8400, Kingston, Ontario, K7L 5L9, Canada.
A coal-tar pitch containing 5 WI. % primary QI was evaluated in prebaked-type model anodes. Comparison of the results with those obtained on a coal-tar pitch containing 12 WI. % 01 shows a reduction of 0.12 WI. % in binder requirement for each 1% decrease in QI, a loss of 0.75 WI.% in binder coke yield per 1% decrease in QI, an overall loss of 5 to 10% in the compressive strength of the baked carbon, but no adverse effect on other baked electrode properties such as electric resistivity, air bum, and electrolytic anode consumption. Based on the results it is concluded that low-QI pitch, alone or in a blend with medium-QI pitch, can be a suitable binder for prebaked anodes.
CAST SHOP TECHNOLOGY V: Grain Refining and Modification
establishing the method as a useful tool for the cast house. Specifically, the problem of lower metal temperatures has been addressed by changes in the sampling method. Also, two distinctly different types of curves appear, and the liquidus temperature of the alloy seems to control which type of curve results.
(8:55 a.m.) ASSESSMENT OF ALUMINIUM GRAIN REFINER PERFORMANCE USING TIffi ALCOA TEST: S. M. Ahmady and D G. McCartney, Dept. of Mat. Sci. and Eng., University of Liverpool, P. O. Box 147, Liverpool L69 3BX, U.K. and S. R. Thistlethwaite, Anglo Blackwells Ltd., Ditton Road, Widnes WAS ONT, U.K. It is common practice to add inoculants to aluminium alloys prior to casting to provide an equiaxed grain structure. There is, however, a need for (a) improved grain refiner and (b) optimization of addition levels for particular alloys and casting processes. The use of the Alcoa Grain Refining Test to assess the perfonnance of AI-6Ti, AI-5Ti-\B and AI-5Ti-0.2B master alloys in aluminium alloys 5657, 6063 and 7050 will be discussed. The Alcoa test has been characterized in tenns of heat and fluid flow, and it will be shown that, as a grain refining test, it simulates well the environment encountered in the DC casting process. Results will be presented which demonstrate that the minimum grain refiner addition required to just suppress columnar growth is directly related to base alloy constitution. The relative perfonnance of the three grain refining master alloys will be presented.
(9:20 a.m.) CHARACTERISTICS OF A NEW GENERATION OF GRAIN REFINER: Rodney F. Carver, Gary W. Boone, Frank P. Koch, KB Alloys Corporate Technology, 3293 McDonald Road, Robards, KY 42452. The need for superior grain refiners continues to grow. More emphasis is being placed on effectiveness, consistency and cleanliness than at any time in the past. Producers of wrought alloy products for critical applications continue to upgrade their products through quality improvements and gauge reduction. In response to the need for a cleaner, more consistent and effective grain refiner; KB Alloys has developed a new boron-free grain refiner alloy. This paper uses the Aluminum Association Test to characterize the grain refining effectiveness and consistency. Advantages are shown for the use of this product in both cast and wrought alloy products.
(9:45 a.m.) THE EFFECTIVENESS OF A GRAIN REFINER REINFORCED BY A BUILT-IN ENERGY CONTENT: M. Vader and J. Noordegraaf, Kawecki-Billiton Metaalindustrie, Kloosterlaan 2, 9936 TE Delfzijl, The Netherlands. Nucleating particles should contain enough built-in energy to form, above the T liquidus, a stable germ on which solidification can start. The cooling rate and alloy composition can largely influence this phenomena. Melt agitation via
electromagnetic or ultrasonic energy input can influence a
homogeneous start in the outer skin of the cast. However, in order to continue the original fine grain, the presence of active particles in the melt is essential. The dotation of these particles to the melt in an optimally effective form and the exclusion of any harmful side effects, is an advantage that the new generation of grain refiners offers. The theory behind the nucleation defines the particles needed. Optimal grain refining is achieved via Statistical Process Control (SPC) and special production methods. (10:10 a.m.)
BREAK
(10:20 a.m.)
Sponsored by the TMS Aluminum Committee Wednesday AM February 21, 1990
Room: Salon 4 & 5 Marriott
Session Chainnan: DA Granger, Molten Metal Processing Division, Alcoa Laboratories, Alcoa Center, PA 15069 (8:30 a.m.) MONITORING GRAIN REFINER PERFORMANCE USING THERMAL ANALYSIS: S. A. Levy, Reynolds Metals Company, Metallurgy Laboratory, Fourth and Canal Streets, Richmond, VA 23261 Thermal analysis is being used in the foundry for routine quality control relative to silicon modification and grain refinement. Such techniques have been much slower to be adopted in the wrought aluminum casting field. The current paper addresses the problems which have been studied and the progress that has been made toward
THE APPLICATION OF QUALITY ASSURANCE IN THE PRODUCTION OF ALUMINIUM ALLOYS USING SPC. Dr. J Pearson London & Scandinavian Metallurgical Company Ltd., 45 Wimbledon Hill Road, London SW19 7LZ, England.
Quality Assurance has been applied to the production of aluminilJID master alloys. A quality management system covers all commercial and production stages from the purchase of raw materials to delivery of the alloy to the customer. Statistical process control procedures have been installed at key stages of the production cycle. The implications and advantages of Quality Assurance are described with reference to the production of the grain refiner titanium boron aluminium.
(10:45 a.m.) A TOTAL QUALITY APPROACH FOR SELECTING A CHOICE SUPPLIER: Gilberto Acosta, Jr., Aluminum Company of America, P. O. Box 472, Rockdale, TX 76567.
In the last several years, the manufacturing world has undergone a total quality reawakening. Quality is no longer measured by the ability to simply meet the specifications, but by continued reduction of variation about a target. The
approach outlined herein demonstrates how this quality philosophy is used in the selection of a choice supplier from a pool of qualified ones. A five-step approach was developed to ensure a fair selection process. The steps include: I) establishing quantitative and qualitative criteria; 2) creating a value system for selected criteria; 3) collecting representative data; 4) scoring results; and 5) providing feedback. By following this five-step process, eight weighted criteria are used to determine an overall choice supplier for a grain refiner product. The five-step technique has proven to be an effective, fact-based decision tool which has the added benefit of strengthening supplier relations by establishing specific goals and targets. (11: 10 a.m.) EFFECT OF IMPURITIES ON STRONTIUM MODIFICATION OF AL-SI CASTING ALLOYS: Bernard M. Closset, Timminco Metals, 130 Adelaide Street West, Toronto, Ontario, Canada and Salim Khan, McGill University, Department of Mining and Metallurgical Engineering, 3450 University Street, Montreal Quebec, Canada. The effect of strontium and antimony on the microstructure.of AI-Si hypoeutectic casting alloys has been analyzed. The strOntium modification of an AI·Si-My alloy in the presence of antimony has been studied. A356.0 melts with various amounts of antimony have been treated with increasing strontium levels. The microstructure and the mechanical properties of melts contaminated with antimony have been determined at various strOntium levels. Antimony has to be kept at a level in the order of 0.02% for the strOntium to be effective. Finally, a correlation between porosity and hydrogen content has been established for different strOntium and antimony levels. (11 :35 a.m . ) STRONTIUM INFLUENCE AS A MODIFYING ELEMENT ON TIlE MICROSTRUCTURAL CHARACfERISTICS AND MECHANICAL PROPERTIES OF TIlE AL-SI (A-356) ALLOY: Marco Pemia and Luis Avarullo, Venalum, Industria Venewlana de Aluminio, c.A., Apdo. 312 Puerto Ordaz, Edo Bolivar, Venezuela. Venalum is diversifying the alloy production including recently the A-356 alloy to supply the automotive industry. This paper describes the progress obtained in our plant in the production of the A-356 alloy modified with Sr. Sr is added as master alloy AI-Si-Sr in different proportions and using different casting methods to detennine the effects of the Sr addition in the relation among cooling rate, velocity, alloy microstructure and mechanical properties. E~ONMENTALEFFECTS
TMS General Abstract Session Wednesday AM February 21, 1990
Room: Plaza A Inn at the Park
Session Chairman: Neville Moody, Sandia National Labs., P.O. Box 969, Livermore, CA 94550 (8:30 a.m.) THE INFLUENCE OF STRENGTH ON THE EMBRITTLEMENT OF ALUMINUM BY MERCURY by Y. Liu and R. G. Hoagland, Department of Mechanical & i1aterialsTngTileering, I,ashington State University, Pullman, WA 99164-2920. Like the active plastic lone, the plastically deformed wake behind a growing crack can be a major source of crack tip shielding. Additional shielding contributions can occur from ligaments left behind as the crack extends . For environmentally assisted crack growth of aluminum by mercury in which the crack grows slowly at velocities up to O. I mis, both effects make the near-tip field K(K . ) different from the applied K(K ) in a way that depends o~l~he flow strength of the materialag~en at very low K/~y levels. This work presents some results of LME crack growth studies of the AIIHg system showing the role played by the strength of the solid. Shielding effects by the plastic wake is suggested as an explanation of the effects. This work supported by the Office of Basic Energy Science, U.S. Department of Energy under Grant DE-FG06-87ER45287. (8 :48 a.m.) CREEP-CORROSION BEHAVIOR OF IlELDS IN ALLOYS 800H, 316, AND SEVERAL EXPERIMENTAL FEHRITIC STAINLESS STEELS
Ala1n Lefort, Centre d' Etudes Nuclea1res de Grenoble, 38041 Grenoble, France; Joelle Bressan, CERSTA, Gaz de France, 93211 La Plaine St. Denis, France; Ugo Franzoni, Centro Sperlmentale Metallurgico, Rome EUR, Italy. The behav10r of mater1als at h1gh temperatures in agressive env1ronments, such as those found 1n the coal gasification process, depends on the mater1al's ab1l1ty to form and retain a protectlve ox1de layer on the surface. In welded components,
Annual Meeting-89
locall1zed corros1on of the welded Jo1nt ls an add1t1onal problem. In th1s study, the creep behav10r of several commerc1al and exper1mental alloys, both w1th and without welds, was studied under carburizing and ox1d1z1ng gases wh1ch contained sulphur. No catastrophic rupture was observed for the base alloys, but the welded alloys presented several problems. (9 : 06
a.m.)
EFFECT OF V AND Hf ON THE CORROSION BEHAVIOR OF AN ALUMINA FORMER: y, Srinivasan and Si-Cheng Kung, Universal Energy Systems, Inc., Dayton, OH 45432. The objective of this investigation is to understand the sulfidation behavior of an alumina former in an environment that is typical of coal conversion and gasification condition. Coupons of Fe-25 wt.' Cr-6 wt.' Al with and without the addition of V and Hf were exposed to Hz/HzS/HzO/Ar gas mixture at 700'C for times up to 120 hrs. The pOz and pSz were 10- zoand 10- 1 atm. Scales were characterized by SEM/EOAX and sputter AES. The results were discussed. The addition did not have any significant effect on the sulfidation behavior at 700'C. (Research sponsored by the Martin Marietta Energy Systems, Inc. who operates ORNL for DOE; Program Manager: Mr. R. R. Judkins.) (9:24 a.m.)
ALLOY CORROSION AND EROSION IN COAL PREPARATION PLANTS, M.A. Pagano, S.B. Lalvani, and J.H. Swisher, College of Engineering & Technology, Southern Illinois University, Carbondale, IL 62901. Piping and other hardware in coal preparation plants are very susceptible to degradation by dissolved salts and suspended coal in the process water. It is suspected that, because coal mined underground in Appalachia and the midwest is now increaSing in chlorine content, there could be an adverse impact on materials performance. In-plant tests were conducted to determine the importance of chlorine concentration in process water on corrosion and erosion rates and to explore methods of corrosion control. The results were surprising in that the chlorine concentration did not have a dominant effect on corrosion and erosion rates. Cupronickel alloys not considered previously for this application performed well, but metal wastage of the cupronickels was greater than for 304 stainless steel. Success was achieved in cathodically protecting two wear-resistant alloys from corrosion in coal slurry piping (Hadfield steel and Ni-Hard cast iron). ( 9:42 a.m.)
AN EVALUATION OF CATHODIC PROTECTION OF MATERIALS IN HIGH CHLORIDE ENVIRONMENT, S. B. Lalvanj, V. V. Patel, and J. H. Swisher, Mechanical Engineering and Energy Processes, College of Engineering and Technology, Southern Illinois University, Carbondale, IL 62901. Degradation of alloys of 304SS, Monel, Cu-Ni (90:10), Hadfield steel, and carbon steel C-IOI8 in chloride containing aqueous solutions was studied using a potentiodynamic polarization technique. The solutions were obtained from two coal preparation plants. Another solution used was simulated seawater. The influence of aeration on the corrosion rate of the alloys was determined. The experimentally generated data was used to predict the extent of cathodic protection needed to passivate the alloys. As compared to simulated seawater, it was found that the cathodic protection of alloys is more difficult in the chloride-containing solutions obtained from a coal preparation plant . 00:00 a.m.)
THERMAL HEALING OF DEFECTS IN PROTECTIVE OXIDE FILMS ON CHROMIUM-CONTAINING STEELS. J,H. SWisher, W.O. Cho, and W.W. Qiu, Dept. of Mechanical Engineering and Energy Processes, Southern Illinois UniverSity, Carbondale, IL 62901. As a sequel to research completed recently on unalloyed iron, the thermal healing of mechanical defects in protective oxide films on chromium-containing steels was investigated. Most of the experiments were conducted on a series of Cr-Mo steels, which oxidize to form a non-adherent outer layer of FeO and an adherent inner layer of (Fe,Cr)304 spinel. When defects in the spinel layer were healed in a static vacuum, the resulting oxide was depleted in Cr. Similar experiments conducted on an Fe2SCr-6Al alloy resulted in a depletion of Al in healed flaws. When a Cr plating was added to an Fe·8.3Cr-I.OMo steel, two benefits were obtained. Metal wastage during oxidation was reduced, and the depletion of Cr in healed flaws was less significant. Thus Cr plated material may be an effective substitute for higher Cr alloys in some applications.
90-Annual Meeting (10:18 a.m. )
OXIDATION CHARACTERISTICS OF Ti-25Al-lONb-3V-lMo. T. A. Wallace, R. K. Clark, NASA Langley Research Center, Mail Stop 194, Hampton, VA 23665-5225; S. N. Sankaran, K. E. Wiedemann, Analytical Services and Materials , Inc., 107 Research Drive, Hampton, VA 23666. Static oxidation kinetics of the super alpha-two titanium·aluminide alloy Ti25AI-lONb-3V·IMo (at. %) have been investigated in air over the tempera· ture range of 650·1000°C using thermogravimetric analysis. The oxidation kinetics were complex at all exposure temperatures and displayed up to three distinct oxidation rates. Oxidation produ cts were determined using X-ray diffraction techniques, electron microprobe analysis, and energy dispersive X-Ray analysis. Oxide scale morphology was examined by scanning electron microscopy of the surfaces and cross sections of oxidized specimens. The oxide was multilayered , consisting primarily of TiO, doped with Nb, a top layer of AhO" and a thin bottom layer of TiN. (10:36 a.m.)
HYDROGEN TRAPPING AT COHERENT PRECIPITATES IN AN FCC ALLOY: APPLICATION OF EMBEDDED ATOM THEORY: N. R. Moody and S. Foiles, Sandia National Laboratories, Livennore, CA, 94551-0969. Previous work on cathodically charged , nickel-rich IN903 shows that deuterium concentration-versus-depth profiles are strongly influenced by trap effects. With the assumption that trapping is localized to the trap feature surface, the calculated trap site densities compare well with trapping at the "( precipitates. However, disagreement exists as to the whether trapping is a short or long range phenomenon in this and similar materials. As a result, energy minimization calculations have been conducted on L9 tilt boundaries in Ni using the Embedded Atom Method to simulate the extent of trapping from an "( interface. These calculations indicate that trapping is localized to a distance of one unit cell from the interface. The calculations also indicate that the trap binding energy is very high (--0.5 eV) on the interface plane but decreases rapidly with distance, exhibiting an average energy similar to published values for nickel-rich alloys. This work supported by U.S. DOE Contract DE-AC04-76DPOO789. (10:54 a.m.)
EFFECTS OF COMPOSITION AND MICROSTRUCTURE ON TRANS GRANULAR STRESS CORROSION CRACKING IN AUSTENITIC STAINLESS ;TEELS:* P 5 Maiya, Materials and Components Technology Division. Argonne National Laboratory. Argonne. Illinois 60439. Transgranular stress corrosion cracking (TGSCe) of Type 347 stainless steel in 289°C water containing 0.1 ppm sulfate. as observed in slow-strain-rate testS. is sensitive to different heat treatments . In particular. Nb in solid solution exacerbates TGSCC. optimal aging of the alloy imparts resistance to TGSee. and overaging of the alloy makes susceptible to TGSee. These observations can be shown to be generally consistent with film-rupture models for sec. For example. Nt in solid solution decreases stacking fault energy. promotes planar slip. and facilitates film rupture and hence TGSCC. However. i n the alloy containing NbC preCipitates. TGSCC susceptibility depends on whether the alloy is aged or overaged. In optimal aging heat treatment. the flow stress requi~ed to break the passive film increases (with a resultant increase in resistance to TGSCe) . while in overaged alloy, increased susceptibility to TGSCC can be associated with a decrease in flow stress. 'Work supported by the Office of Nuclear Regulatory Research U.S. Nuclear Re2ulatory Commi ssion. Washington. DC 20555.
(n:12 a.m.) SIr... Acceler.ted Grain Boundary Oxidation of Inconal 718: Barbara A. Young, linda B. Thompson. Rocketdyne, Canoga Park, CA 91303 lnconel 718 Is an age-hardenable, high strength material with good fatigue resistance over a wide range of temperatures. Its properties make It attractive for gas turbine engine components up to 1300F where smooth and notch rupture behavior may be a property consideration. This study discusses a program Initiated at Rocketdyne to determine whether the predominant lailure mechanism in Inconel 718 between 1200F and 1500F Is due to a non-environmentally affected minimum ductility range or to stress accelerated grain boundary oxidation (SAGBO) commonly reported in Incoloy 903.Slow strain rate tensile tests between 800F and 1500F In addition to combination .",ooth notch rupture tests at temperatures between 1200F and 1500F were performed in both air and argon. It was determined that tensile elongations on Inconel 718 exhibit a minimum ductility In the presence of oxygen between 1400F and 1450F. Smooth and notch rupture lives are also sensitive to oxygen, but In · the temperature range of 1250F to 1500F. Notch sensitivity within this range was found to be . dependent upon variations In applied stress. Scanning electron microscopy of air tested samples in the temperature range. of Interest demonstrated intergran;,lar failures associated w~h SAGBO. This phenomenon, which Is not typically observed In Inconal 718, is postulated to be heat treat dependent and 8880clated with an altered ·solution anneal, currently In use at Rocketdyne, developed for Improved hydrogen rasistance.
EXPERT SYSTEM APPLICATIONS IN MATERIALS PROCESSING & MANUFACTURING nI Sponsored by the TMS Shaping & Forming Committee, the TMS Synthesis & Analysis in Materials Processing Committee and the ASM-MSD Computer Simulation Activity Wednesday AM February 21, 1990
Room: ParkB Inn at the Park
Session Chairmen: H. Henein, Department of Mining, Metallurgical and Petroleum Engineering, 606 ChemicalMineral Eng. Bldg., University of Alberta, Edmonton, Alberta T6G 2G6; W. Pardee, Rockwell International Science Center, 1049 Camino d06 Ri08, Thousand Oaks, CA 91360 (8:30 a.m.)
EXPERT SYSTEM APPLICATIONS IN FORKING PROCESSES: M. Y . Demer!, Research Staff, Ford Motor Company , Dearborn , MI 48121. Today's highly competitive market mandates on the manufacturers to make major changes in their traditional product development cycle from concept through production. There is a strong need to reduce cost, shorten the product development cycle and improve the quality of the product. To achieve this goal, the traditional product development process, which is based on the engineer's reliance on his experience to build and test new products, needs to be improved . The time and expense increases as the number of design trials increase. For minor modifications, the traditional process works fine but for new designs and products it becomes highly inefficient. The engineer needs tools to evaluate alternative designs and manufacturing processes before prototype building and testing. The paper addresses the potential integration of Expert Systems with Process Simulation and the benefits that such an integration would extend to the materials forming process. (9:00 a.m.
MANUFACTURING CONTROL THROUGH THE DEVELOPMENT OF A SYSTEM FOR REAL TIME MATERIAL PROPERTIES MONITORING : e.O. Ruud, I.e. Conway, K. Kozaczek, C-J . Yu, The Pennsylvania State University, University Park, PA 16802. The goal of this investigation is to develop and demonstrate the feasibility of a system for in-process interrogation of copper alloy strip. The system is to continuously provide a characterization signature to be used to determine the instantaneous properties and forming characteristics of the metal. The need for in-process characterization is based upon the broad scatter in properties and fonning characteristics which exists in commercial wrought metal products, and the loss of productivity and product quality caused by that scatter. The investigative approach uses a unique combination of three concerted nonintrusive techniques, e.g., eddy current, ultrasound, and x-ray diffraction, all three of which individually have seen limited application in the real-time frames necessary. This paper will illustrate the correlations developed between nondestructive characterization parameters and the mechanical and fonning properties of the copper base alloys. (9:30 a.m.)
KlCAPP: AN EXPERT SYSTEM FOR PROCESS PLANNING : M. S Gouda, Detroit Edison , 1159 Falcon , Troy, MI 48098 and K. S. Taraman, Dean of Engineering, Lawrence Technological University, 21000 W Ten Mile Road. Southfield, MI 48075. Expert systems are being used to create Computer Aided Process Plans (CAPP) in manufacturing. A microcomputer system called MICAPP was developed for making rotational workpieces in a medium size shop . The paper addresses the development of the MICAPP system and its validation . A comprehensive survey was also conducted to identify expert systell based CAPP's and CAPP' s developed by generative , semi-generative and variant approaches . The advantages and shortcomings of the various CAPP systems are reviewed. The paper identifies 128 CAPP systems which are listed according to the type of input used to describe the design specification in the system . The survey includes computer hardware, data input and output, processing logic, part type, selection criteria, vendors as well as the countries utilizing such systems.
Annual Meeting-91 (10:00 a.m.) PC-BASED EXPERT COMPUTER AIDED PROCESS PLANNING SYSTEM FOR PRISMATIC PARTS: N.U. Qureshi, University of Detroit, Dept. of Mechanical Engineering, Detroit, Michigan 48221. Y.S. Chadda, University of Detroit, Dept. of Mechanical Engineering, Detroit, Michigan 48221. Computer aided process planning is a key link in the integration of CAD and CAM, which is so vital for the computer integrated factory of the future. A low-cost PC-based system MICROCAPP has been developed. It is a feature-oriented process planning system for prismatic parts of average complexity. The system developed is capable of processing sixteen different features commonly encountered in manufacturing of prismatic parts. MICROCAPP automatically selects machining operations, cutting tools, machining parameters, machine tools, and machining time estimation for each feature. The system makes use of the minimum cost and maximum production rate criteria for the optimization of process parameters. The results are very promising and can be used in small batch, discrete parts metal working industries. ClO:30 a.m.) AN EXPERT SYSTEMS APPROACH TO RULE-BASED DYNAMIC PROCESS PLANT SCHEDULING: W. Marcus Sztrirnbely, Peter J. Weymouth, Advanced Dynamic Systems Inc., 8130 Sheppard Ave., East, Suite 200, Scarborough, Ontario, MlB 3W3. The secure operation of a complex industrial process such as a Steelmaking Meltshop or Nickel Smelter depends almost entirely on a harmonic working relationship between the human operator and the real time process control environment. The introduction of sophisticated real time monitoring systems and faster than real time decision making modules requires a new approach to dynamic process plant scheduling and production optimization under varying operating constraints. Based on the proven Expert Systems Methodologies of Knowledge Engineering and Rule-Based Logistics, the ADSI Scheduling Advisor has been successfully implemented on the shop floor of several major base metals producing facilities. An overview of several successfull projects will be presented in an attempt to demonstrate how the new ~nd emerging science of Artificial Intelligence can be successfully integrated with the time tested 'rule of thumb' operating practice.
mGH TEMPERATURE DEFORMATION,
FRACTURE AND APPLICATION OF INTERMETALLIC COMPOUNDS I: Intermetallics with Lis-Structure
Jointly sponsored by ASM Flow & Fracture Committee and TMS High Temperature Alloys Committee Wednesday AM February 21, 1990
Room: Plaza B
Inn at the Park
Session Chairmen: J.H. Schneibel, Metals and Ceramics Division, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831-6116; A.K. Vasudevan, Office of Naval Research, Code 1216, Arlington, VA 22217
(8:30 a.m.)
Invited)
SUPERLATTICE LOMER DISLOCAnONS IN Ll2 ALLOYS: * P M. Hazzledjne, Y. Q. Sun, and M. A. Crimp, Department of Metallurgy, Parks Road, Oxford OXI 3PH, United Kingdom At temperatures above the peak in the yield stress ofLl2 alloys (1'01](010) slip becomes prevalent for all crystal orientations except those close to [001]. Weak· beam electron microscope observations show that edge dislocations (Lomers) are less mobile than screw dislocations (Kear-Wilsdorfs) and that they are dissociated into several partial dislocations. Lattice resolution observations conftrm that the dissociation involves climb at high temperatures. The super Lomers and double Lomers are important elements in the understanding of both the yield and creep properties of Ll2 alloys at high temperatures and the yield stress anomaly at lower temperatures. *Work supported by SERC and the National Physical Laboratory, United Kingdom.
(11:00 a.m.)
PRODUCI1VITY IMPROVEMENT TROUGH MACHINE DIAGNOSTIC EXPERT SYSTEMS H. Nivi, Manufacturing Development Center, Ford Motor Co., 24500 Glendale, Detroit, Ml 48239 Determining the root cause of machine failure plays a critical role in machine uptime and directly impacts productivity. Rapid and accurate machine diagnosis is key to determining the root cause of failure. Ford Motor Company has developed an expert system to diagnose induction heating equipment manufactured by Tocco Inc. It combines the expertise of experienced Ford personnel with the knowledge of Tocco's best diagnosticians. Plant personnel at Ford have used this user·friendly system on a portable computer. It asks a series of questions and provides possible answers, with comprehensive graphics to assist the user in performing the pertinent test. The Tocco Diagnostic Assistant has provided valuable insight into the development of machine diagnostic expert systems. Teaming the plants and machine suppliers is essential for the successful development and implementation of expert systems. The Tocco Diagnostic Assistant is a completed software package and is now commercially available. (11:30 a.m.) DIAGNOSTIC AND ADVISOR TOOLS FOR MANUFACTURING PROCESSES AND MACHINE TOOLS AT CUMMINS ENGINE COMPANY, INC. by Robert T. Easterling, Director - elM Development, Cummins Er'gine Company,
Inc. Box 3005, Columbus, IN
47202-3005
This presentation will focus on how Cummins Engine Company, Inc. is applying Expert Systems technology in manufacturing to maintain its position as the worlds largest independent producer of diesel engines. The benefits of improved productivity and capital equipment utilization from the application of expert
systems will be highlighted.
Expert systems that advice
machine operators, maintenance personnel and engineers on how to better manage critical manufacturing processes and diagnose machine failures will be covered. This will include an EDM process advisor, engine diagnostics, and transfer line diagnostics. Notable developments in diagnosing machines controlled by programmable logic controllers will be covered in more detail, particularly as it applies to transfer lines.
(Invited)
(9:00 a.m.) THE INFLUENCE OF ORDERING ON DISLOCATION PROCESSES IN DISPERSEDPHASE ALLOYS M. S. Duesbery and K. Sad an and a , Naval Research Laboratory, Washington D.C. 20375 The interaction of isolated and coupled dislocations with a re9ular array of precipitates is examined within the Brown self-stress theory using a novel computer model based on molecular dynamics. Details of the model and its applicability to relevant hardening mechani sms wi 11 be di scussed in general terms. Several spec ifi c cases will be considered in detail. These will include hardening due to differential stacking fault energies and due to superlattice dislocation formation, for non-trivial combinations of order and disorder in the matrix and the dispersed phase. The results will be compared with the qualitative predictions available in the 1iterature. (9:30 a.m.) THE DETERMINATION OF COMPOSITIONAL ORDERING AT GRAIN BOUNDARIES IN Ni Al USING HIGH RESOLUTION TRANSMISSION ELECT~ON MICROSCOPY-; M J Mills, Sandia National Laboratories, Livermore, CA 94551-0969 The mechanical properties of polycrystalline and multi· phase intermetallics are known to be strongly influenced by the properties of the interfaces present. A classic example of this effect is the dramatic increase in the ductility of polycrystalline Ni3AI upon the addition of small amounts of boron. It has been proposed that the formation of a compositionally disordered region near the boundaries promotes transmission of slip from one grain to the next, thereby reducing the propensity for intergranular fracture. High resolution transmission electron microscopy (HRTEM) is currently being used to probe the structure and ordering at grain boundaries in directionally solidifted Ni3AI (supplied by C.T. Liu at Oak Ridge National Laboratory). Image simulations indicate that the determination of an order/disorder transition using HRTEM is extremely dependent upon experimental parameters such as boundary orientation, specimen thickness and microscope defocus. However, a limited number of observations obtained under appropriate imaging conditions demonstrate no clear evidence for a disordered boundary region in boron·doped material. The implications of these results with respect to models of fracture in Ni3AI will be discussed.
*
Research sponsored by tM Department of ElU!rgy. Office of Basic ElU!rgy Sciences IUlder contract DE·AC04·76DPOO789.
92--Annual Meeting (9 :50 a.m.) ON THE REVERSIBILITY OF CROSS-SLIP IN NI3AL Walter ~ Milligan, Dept of Metallurgical Engineering, Michigan Technological University, Houghton, MI 49931, and Stephen D. Antolovich, School of Materials Engineering, Georgia Institute of Technology, Atlanta, GA 3033Z-0Z45. Ni3AI
and
other
LiZ
intermetallics
exhibit
a
tension
compression asymmetry in the flow stress which is cycl ically stable during f ully reversed fatigue. This stable asymmetry requires that the rate controlling cross-slip mechanism be reversible under a change in loading direction. This reversibility requirement can be used as a critical test of the available cros s-slip models. It will be demonstrated that the cross-slip mechanism proposed by Pope , Paidar and Vitek is consistent with this requirement, while the classically assumed total cube cross-slip me chan ism is not. Supported by NASA-Lewis under NAG3-S03 (10:10 a.m.)
BREAK
(10:20 a.m.) HIGH TEMPERATIJRE DEFORMATION OF DS NICKEL·ALUMINIDE INTERMETALLlCS: M. Nazmy and M. Staubli, Department of Material Technology, Asea Brown Boveri and Company, CH·540I Baden, Switzerland In this investigation Ni3Al base intermetallics, alloyed with Ta, Nb, and Si, have been prepared by vacuum induction melting and then were directionally solidified in the form of bars. In addition to the major alloying elements (i.e., Ta, Nb, and Si) these intermetallics were microalloyed with Band Hf for enhancing the ductility of these materials. The effect of temperarure on the tensile flow stress of these different intermetallics was investigated. The high temperature deformation behavior of these materials was studied by creep and fatigue testing at a wide range of temperatures, and the results were compared with the available creep data for other intermetallics and commercially known nickel·base superalloys. (10:40 a.m.) INFLUENCE OF STRAIN RATE AND TEMPERATURE ON THE MECHANICAL AND MICROSTRUCTIJRAL RESPONSE OF Ni3AL:* Q.L QmyjIl and P.S. Follansbee, Los Alamos National Laboratory, Los Alamos, New Mexico 87545 While the influence of strain rate on the structure / property response of a variety of metals and alloys has been extensively studied, the effect of strain rate on ordered alloys remains largely unknown. Extrapolation of known trends in conventional alloys to ordered systems is complicated by the unusual temperature dependence of the yield strength of some intermetallics such as Ni3Al. In this paper, the effect of strain rate and temperarure on the substructure evolution and mechanical response of Ni3Al will be presented. The strain rate response of Ni3Al have been studied at strain rates from 10.3 S·1 (quasi·static) to 5000 s . 1, utilizing a Split Hopkinson Pressure Bar. The Hopkinson Bar tests were conducted at temperatures from 25 to 970OC. Experimental results have revealed that Ni3Al displays a small rate sensitivity at room temperature and a strong temperarure dependent yield strength at high strain rate, similar to that observed at quasi·static rates. The substructure evolution has been characterized utilizing TEM. The defect generation and rate sensirivity of Ni3Al will be discussed as a function of strain rate and temperature and contrasted to that observed in nickel. *Work performed under the auspices of the U.S. Dept. of Energy (11 :00 a.m.) THE E~~ECT O~ MINOR ALLOYING SUBST ITUTIONS ON THE HIGH-TEMPERATURE DE~ORMATION OF AI2?Fe3TiS: K. S. Kuma r , R.A. Herring, Martin Ma rietta Laboratorles Batimore, MD 21227 and J.D. Wh i ttenberger, NASA Le~is Research Center, Clevel and, OH 44135. The compressive behav ior of quaternary L12 intermetallics, of the type A122~e3TiS with 2 atomic percent Nb or V, was examined as a function of temperature and s train rate . Although the niobium-containing quaternary and the ternary A122Fe3TiS behave similarly at warm t emperatures, the quaternary exhiblts Significant strength retention at higher temperatures. Serrated yielding ~as observed over a certain ~arm temperature range, coincident wi th the region exhibiting a positive t emperature dependence of strength. The serrations are believed to originate i n the cross-$lip behavior intrinsic to these L12 materials. Transmission microscopy was used to characterize the dislocations in the quaternary in termetallics following deformati on at different temperatures. (11 :20 a.m.) Elevated Temperature Strength and Deformation Mechanisms of High-Temperature LiZ Intermetallics: Charles H Henager. Jr., J. L. Brimhall, and S. M. Bruemmer, Pacific Northwest Laboratorya, Richland, WA 99352. Elevated temperature microhardness testing is being used in conjunction with bend testing and TEM to study the flow properties of high·temperarure Ll2 intermetaIlics produced by sputter deposition. IntermetaIlic deposits of Ir:3Nb and Ir3Hf have been tested and compared to data from Ni3Al. Preliminary microhardness data indicates that the high·temperature Ll2 intermetallics show
a yield stress anomaly with a peak strength at 800 C and maintain their room temperature strength to 1200 C. Bend tests are being performed at selected temperatures to obtain dislocation slip characteristics and fracture behavior. aoperated for the U.S. Department of Energy by Battelle Memorial Institute under Contract DE·Aal6-76RLO 1830.
mGH TEMPERATURE SUPERCONDUCTING COMPOUNDS IV: Bulk B: Melt Texturing/Substitution Sponsored by the TMS Superconducting Materials Committee Wednesday AM February 21, 1990
Room: Orangewood I Quality Hotel
Session Chairmen: D.K. Finnemore, Department of Physics, Iowa State University, Ames, IA 50011; C.S. Pande, Naval Research Laboratory, Washington, DC 20375 (8:30 a.m.) CRITICAL CURRENTS IN HIGH TC SUPERCONDUCTORS. S. Jin, J. E. Graebner and T. H. Tiefel, AT&T Bell Laboratories, Murray Hill, NJ, 07974. Copper-ox ide-based bulk high temperature superconductors have two major problems in their critical current behavior, i.e., i) Josephson weak-links at grain boundaries and ii) the lack of sufficient flux pinning within the grains. While the weak-link problem can be avoided by extreme alignment of grains (such as Melt-Textured-Growth), it would nevertheless be desirable if the weak-links could be eliminated without grain alignment. The insufficient flux pinning restricts the high-field JC(H) in bulk, weak-link-free Y-6a-Cu-0 to about ~ 10 4 A/cm 2 at 77K. which is somewhat lower than the desired values for many major superconductor applications . In this paper, the progress made in both problem areas will be reviewed. Various proceSSing methods will be compared, and the creation of desirable flux pinning sites will be discussed . (8:55 a.m.) TRANSPORT AND INDUCTIVE CURRENT MEASUREMENTS IN ORIENTED GRAINED YBa2Cu30x SUPERCONDUCfORS : K. Salama, V. Selvamanickam, Texas Center for Superconductivity, University of Houston , Houston, Texas 77004. H. Kupfer, and C. Keller, Kemforschungszentrum Karlsruhe, Institut fUr Technische Physik und Universitat Karlsruhe, 7500 Karlsruhe, FRG. A liquid phase processing method for the fabrication of bulk YBa2Cu30x superconductors with large current carrying capacity has been developed. Slow cooling through the peritectic transformation has been shown to control the microstructure of these superconductors. Direct current measurements show that the transport current density of these oriented grained samples is orders of magnitude higher than those of sintered bulk Y·Ba·Cu·O superconductors. The current carrying capacity of this material was also investigated by inductive flux profile and a.c. susceptibility measurements. The induced current was directed perpendicular to the a·b plane oriented grains where the transport current was measured. In spite of the unfavorable geometry of the investigated shielding current, no features of granularity were observed. Field and temperature dependences of the Ic are also measured and compared with those in granular material. (9:20 a.m.) EFFECf OF RARE EARTH SUBSTITIJTION ON QUENCH AND MELT GROWTH (QMG) PROCESS: Kiyoshi Sawano, Mitsuru Morita and Shoichi Matsuda, R&D Lab· I, Nippon Steel Corporation, 1618 Ida, Nakahara·ku , Kawasaki 211 Japan QMG process is capable to produce bulk superconductors with high critical current density (Jc). The process employs partial melting and subsequent slow cooling to grow YBa2Cu30x (123) grains by the peritectic reaction between Y2BaCu05 (211) and liquid. The resultant 123 grains are highly oriented and involve unreacted 211. Since yttrium is not necessarily the best for this process, other rare earth elements were used for full and partial substitution for yttrium. Since 211 of neodymium and lanthanum do not exist, the final 123 grains did not contain 211. Ytterbium exhibited the stable liquid phase formation during melting or partial melring. This contributed to obtain finely and homogeneously dispersed 211 and to improve microstructural development to form 123. Superconducting properties seemed to be similar to those of the yttrium system. Effect of the microstructure such as defects and grain size on the Jc is significant rather than the type of rare earth elements. Therefore, rare earth substitution in QMG process contributes to easier optimization of the process for the ideal microstructure and higher Ic.
Annual Meeting-93 (9:45 a.m.)
°
DIRECTIONAL SOLIDIFICATION OF YBa Cu + BY ZONE MELTING: 2 P.J. McGinn, W. Chen, University of No~r~ Dame, Department of Materials Science and Engineering, Notre Dame, IN 46556. Zone melting has been used to achieve textured structures in YBa Cu 0 +x. Uniform textured regions as long a 3 cm. have been 2 3 6 observed. Hot zone temperture, cooling gradient and cooling rate control the size and direction of the textured domains, which consist of many parallel plate-like grains. Typical grain dimensions are on the order of 5 mm. x 0.1 mm. x 3 mm. XRD, SEM and optical microscopy results confirm the high degree of texture. Magnetization studies indicate a much larger magnetic hystersis than in unaligned or silver doped samples.
(10:00 a.m.)
MELT TREATMENT OF YBa2Cu307 lAg BY USING A FLAME: *Keshun Shi and D.D.L. Chung, Composite-~aterials Research Laboratory. Furnas Hall, State University of New York, Buffalo, NY 14260. A flame from a mixture of oxygen and natural gas was used to melt in air composites of Ag particles in YBa2Cu307_o' Following the melt treatment, superconductivity was restored by heating in oxygen at 900°C. With a reducing flam~ and 52 wt.% Ag, Tc was above 90 K and J c was almost 300 Alcm at 40 K. In contrast, 2 with a reducing flame and no silver, J c was only 30 A/cm at 40 K. With an oxidizing flame and 52 wt.% Ag, both Tc and J c were lower than for the corresponding case of a reducing flame. The room temperature electrical resistivity for 52 wt.% Ag was lower for the composite treated by a reducing flame than for the untreated composite. * Supported by the New York State Institute on Superconductivity
(10: 15 a.m.) MOLTEN SALT PROCESSING OF YBa2Cu307_x, D. B. Knorr and C. H. Raeder, Materials Engineering Department, Rensselaer Polytechnic Institute, Troy, New York 12180-3590 Grain oriented YBa2Cu307-x superconductors are necessary for high critical current density levels in bulk conductor applications. High aspect ratio platelet powder particles are desirable starting materials to be oriented by subsequent powder processing. Direct synthesis of the superconducting phase and coarsening of pre-existing 1-2-3 phase powders are two approaches. The major problem is stability of the 1-2-3 phase upon exposure to molten salts. A series of molten salt systems are screened at 900°C for 2 hrs. to identify those that retain 1-2-3 phase. Systems include halides, sulfates, and nitrates. The relative stability of 1-2-3 phase is dichloride bearing salts and alkali halide salts are compared. Decomposition phases are categorized according to salt system. Similar results are detailed for nitrate salts, which decompose, and for sulfates whose behavior is analogous to the chloride systems. Analysis of particle size and shape indicate prospects for particle coarsening. Results of experiments to synthesize 1-2-3 phase from constituent powders are reported.
(10:30 a.m.) FUNDAMENTAL STUDIES OF PROCESSING STRATEGIES FOR HIGH-Tc SUPERCONDUCTORS. M. J. Fluss, Lawrence Livermore National Laboratory, Livermore. CA 94550' Experiments and theory are helping to solve the variety of processing questions surrounding the high-Tc superconductors. In our laboratories, first principles calculations (P.Sterne and L. Willis) of the electronic structure predict phase diagrams of the YBCO material. Cluster model calculations of the molecular orbital properties (P.Turchi) intrinsic to these materials have been studied in detail both by experiment (R.H.Howell. A.Wachs. R.West and J.Kaiser) and theory, and have emphasized the ionic character of the materials in general. This type of view. ionic sublattice with some MO bonding, has led us to explore the novel idea of substituting halogen species for oxygen (R.Glass. H. Radousky and M.J. Fluss). The fundamental and technological implications of these studies will be discussed with regards to the production of wires, tapes and cables. as well as the integration of high-Tc devices into existing silicon technologies. *Work performed under the auspices of the U.S. Dept. of Energy by the Lawrence Livermore Nat'l Lab under contract #W-7405-ENG-48. (10:55 a.m.) PREPARATION OF HIGH DENSITY SUPERCONDUCTING CERAMICS IN THE Bi-Sr-Ca-Cu-O SYSTEM. H. Sato. W. Zhu. M. M. Miller and T. Ishiguro. School of r~ateriaTSrri9Tneering. Purdue University. West Lafayette. IN 47907. A high density superconducting ceramics close to 100% theoretical density in the Bi-Sr-Ca-Cu-O system has been prepared from amorphous precursors obtained from melts. and growth characteristics of superconducting phases have been investigated. A series of homologous superconducting phases of the form Bi2Sr2Can_1CunOx (n=l, 2m and 3) has been known in this system and all of these structures can be obtained at a single composition by different heat treatments. The n=2 and n=3 compounds grow as platelets and both the phases coexist microsyn-
tactically. The high Tc phase (Tc both in the n=2 and n=3 phases.
~
II0K) seems to be found
(11: 10 a.m.) Effect of Ag on Ag doped YBa2Cu307_o superconducting ceramics Shin-Guang Shyu, Yimei Zhu and H. Suenaga, Materials Science Division, Brookhaven National Laboratory, Upton, NY 11973 A composite of YBa2Cu307_o-Ag ceramics is prepared by sintering a mixture of Ag20 and YBa2Cu307_o powder. Doping of Ag in YBaZCu307_o may enhance or reduce transport critical current density depending on processing conditions and the amount of Ag introduced. Low temperature sintering of Ag20 doped YBa2Cu307_o enhances J c ' Higher sintering temperature may decrease Jc as amount of Ag increased. Also, enhancement of grain growth was observed in Ag doped sample. The effect also depends on the processing conditions and amount of Ag20 used. Microstructure, AC susceptibility, resistance measurement, microanalysis, transport critical current densities and their relationships will be discussed.
(11 :25 a.m.) PHASE STABILITY AND MICROSTRUCTURE OF YBa2Cu307-x DOPED WITH Nb205: F.C. Rizzo Assuncao, M.F. Silva Lopes, A.H.P. Carvalho and C.A.A. Nunes, Department of Materials Science and Metallurg~ PUC/RJ, Rio de Janeiro. RJ, 22453, Brazil. The substitution of Cu by Nb in the compound YBa2Cu307-x has been investigated. Samples were prepared from carbonates and oxides using the solid state diffusion technique. The amount of Nb introduced varied from 1 to 8 at%. The resulting samples were analized by x-ray diffraction, and the microstructure was observed by SEM and optical metallography. Two phases were ider tified showing the imiscibility of Cu and Nb in the oxides formed. A thermogravimetric and thermodifferential analysis of samples prepared from the same starting materials showed the tendency of Nb to react with Ba and Y, forming niobates at lower temperatures, reducing therefore, the amount of Nb available for incorporation in the 1-2-3 compound at the temperature of formation of this phase. The paper will present the superconducting properties of the two-phase material obtained and will discuss the attempts of introducing Nb using a copper niobate in the starting mixture.
mGH TEMPERATURE SUPERCONDUCTING COMPOUNDS V: Thin Film B: Microstructure Sponsored by the TMS Superconducting Materials Committee Wednesday AM February 21, 1990
Room: Orangewood II Quality Hotel
Session Chairmen: J. Narayan, Department of Materials Science and Engineering, North Carolina State University, Raleigh, NC 27695; S.M. Wolf, Energy Conversion and Utilization Technologies Program, US Department of Energy, CE-121, 1000 Independence Ave., SW Washington, DC 20585 (8:30 a.m.) INTERFACIAL REACTION AND CATION DIFFUSION STUDlES RY RADlOTRACFRS IN YRa,Cup" EPITAXIAL FILMS ON (100) SrTiO, SURSTRATE: D. Gupta, R. R. l.aibowitz,.I. t\ Lacey and S. Shinde, IRM Research Division, Thomas.l. Watson Research Center, P.O. Rox 21R, Yorktown Heights, New York I059R USA Knowledge of atomic movements in the supcrconducting oxides themselves and the interfacial reactions with the substrate is important to optimize their processing and subsequent use. We will first descrihe our interfacial reaction studies hetween the SrTiO, substrate and the YRa,Cu,o, , epitaxial films during the gro\\1h process. For this purpose radioactive R~Sr isotope was incorporat{'n onto the the hare surface of the suhstrate and the YBai~u)076 epitaxial films were grown under various conditions. Significant migration of the R~Sr and Sr was observed from the suhstrate to the epitaxial films during the growth process. These effects can be related to the improvements in the supcrconducting properties of the films. Second, we will discuss our cation diffusion studies using the filNi radiotracer in the epitaxial film" :lnd hulk specimens of the YDa1 Cu J 0 1 ~ oxide in the 550-750 "C range and in an 02 pressure of I har. The 6JNi difTusivitics show an Arrhenius hehavior in the orthorhombic phase with n. - 1.1 cm 2 /scc and Q = 2.7 eV. The behavior in the orthorhomhic to tetragonal phase transfonnation region is heing investigated. The measured diffusion par:::lmeters \\-'ill he compared and contrasted with those fClr oxygen in this oxide reported in the lih'rature as well as in several transition metal oxides.
94---Annual Meeting (8:55 a.m.) SPUTIERED SUPERCONDUCTING FILMS OF Bi2SI2CaCU20x MADE BY ONE-STEP, LOW-TEMPERATURE, IN-SITU GROWTII--K E Grav. R. T. Kampwirth, P. H. Andersen and D. B. McDonald, Materials Science Division,
Argonne National Laboratory, Argonne, Illinois, 60439, USA
Low-temperature, in-situ processing of thin films of the high temperature superconductor has provided the smoothest films by discouraging grain growth, the highest current densities and allows for the potential compatibility with a variety of substrates which would melt or severely react at the customary annealing temperatures. Little work has been reported on films of the HTS cuprates based on Bi and n which exhibit higher Te, fewer insulating impurity phases and offer better compatibility with various substrates than YBCO. This paper reports the formation of superconducting films of BSCCO by sputtering using a one-step, in-situ process with the substrate temperature at only -65()oC. SEM analysis indicates that smoother films have not yet been obtained and that some segregation occurred. We will discuss these results and report the current status of this program. This work supported by the U.S. Department of Energy, Division of Basic Energy Sciences-Materials Sciences and Office of Energy Storage and Distribution, Conservation and Renewable Energy, under contract #W-31-109-ENG-38 and the National Science Foundation--Office of Science and Technology Centers under contract #STC-8809854.
(9:20 a.m.) EFFECT OF SUBSTRATE 'r-E'MPEMTURE AND RF BIASING ON TIlECOMPU,lTlON OF SINGLE-TARGET SPUTTERED Bi-BASED THIN FILMS, N. G. Dhere. A. R. Mason, J. P. Gor,ll and R. G. Dhere, Solar 'C'nergy Research Institute, 1617 Cole Blvd, Golden, CO 80401. Variation of composition with variation of substrate temperature and RF biasing in single target sputtered Bi-Pb-Sb-Sr-Ca-Cu-O an4 Bi-Sr-Ca-Cu-O thin films has been studied because of its influence on formation of high T superconducting phases. It was found that concentrations of Sr and to a lesser extent of Sb were the least affected by substrate heating and RF biasing. RF biasing enhanced preferential resputtering of Bi and Pb. At o substrate temperatures of above 47S C, Ph concentration was drastically 'reduced while above 600°C the relative concentrations of Cu, Ca and Sb were enhanced excessively. Methods for obtaining appropriate compositions have been discussed.
(9:35 a.m.l Preparation and Analysis of HTSC-Thin Films J. Geerk, X.x. Xi, Q. li, H.C. li, R.-L: Wang, G. Linker, O. Meyer, F. Ratzel, , R. Smithey, Kernforschungszentrum Karlsruhe, INFP, P.O.B. 3640, D-7500 Karlsruhe, FRG Thin films of YBaCuO have been deposited onto AI203, MgO, SrTi03, and Zr02 substrates by inverted cylindrical magnetron sputtering. The main advantage of this preparation technique is the high reproducibility allowing detailed and systematic studies of the film properties as a function of deposition parameters. The films grow highly textured on all substrates. For the 1-2-3 material zero resistance is obtained near 90 K. High critical currents between 4x10 s and 5.5x10 6 A/cm 2 were determined for films of the 1-2-3 material on the substrates MgO, Zr02 and SrTi03. A correlation between the critical current density and the growth quality expressed in terms of the mosaic spread has been found. On films on SrTi03 tunnel junctions could be prepared which showed a gap-like feature in their current-voltage characteristic. (10:00
a.m.)
EIECTRCN MICROSCOPIC STUDIES OF OXYGEN EFFECTS IN HIGH Tc SUPERCCN)u::TORS S. Arrelinckx, University of Antwerp,
Groenenborgerlaan 171, B-2020 Antwerp, Belgium.
RlX:A,
The inportance of oxygen content and of oxygen disorde~order phenarena in detennining the superconducting properties of high T oxycle superconductors is well established. Manifestations of these phenarena in the 1-2-3 carpounds are the occurence of transformation twins and of various vacancy ordered superstructures such as the nao (n;2,3,4,5) and the 2·.J2a x2V2a o o structures. The structural aspects of such phenarena can be studied in detail by electron diffraction and electron microscopy using various observation IOOdes. The ()(--+T transformation and the associated microstructural changes are observed "in-situ" and recorded on video-tape. The formation conditions of the various superstructures are derived fran heating experirrents in the microscope. The results are interpreted in terms of IOOdels. The relationship between the oc=ence of oxygen defects and structural IOOclulations in the biSlTUlth and in the thallium carpounds will be discussed.
(10:25 a.m.) MICROSTRUCTURAL EFFECTS OF HOT EXTRUSION OF YBa2Cu307-x K. Rajan, R. Garcia, W.Z. Misiolek and R.N. Wright, Materials Engineering Department, Rensselaer Polytechnic Institute, Troy, NY 12180=3590. In the present work, microstructural studies have been undertaken on YBaCu307_x hot extruded at 850°C. A superconducting powder was ball milled for two hours and then pressed into silver and copper tubes. These tubes were located in stainless steel jackets thus forming composite billets. The
outside diameter of the billets was the same, but the inner tube diameters differed. Powder packing pressure was varied. The extrusion process variables included die angle and nominal strain rate. The observation of different defect structures in hot extruded YBa2Cu307-x is reported. Of particular Significance is the observation of periodic defects at grain boundaries. The possible implications of these observations for the development of texture are discussed. Work supported in part by the New York State Energy Research and Development Authority: Contract No. l123-EEED-AEP-89
(10:40 a.m.) MICROSTRUCTURE AND CHARACTERIZATION OF Sb IN BISSCO: N G Wimer, Geo-Centers, Inc., Fort Washington, MD 20744, H.A. Hoff, and G.N. Candela, Naval Research Laboratory, Washington, DC 20375-5000. Recently there have been reports of a 132 K superconductor made by the addition of Sb203 to Pb-Bissco. We have found the inclusion of Sb when preparing PbBissco does not enhance Te above that found in undoped samples. By examining the microstructure using optical and electron microscopy, those grains which exhibit the grain morphology of Bissco are located. Using energy dispersive x-ray spectroscopy on these grains, we have found that Sb does occur in small quantities.
(10:55 a.m.) OPTICAL AND ELECTRON MICROANALYSIS OF CUP RATE SUPERCONDUCTORS: H.A Hoff, M.S. Osofsky, L.E. Richards, and C.S. Pande, Naval Research Laboratory, Washington, DC 20375-5000, and W.L. Lechter, Sachs/Freeman Associates, Inc., Landover, MD 20875.
Individual anisotropic grains in heterogeneous and opaque cuprate-based materials, when viewed in reflected polarized light, often have characteristic colors (colors of polarization). Of the cuprate superconductors, regardless of carrier type, examined so far, only one characteristic color has been observed. We have studied the presence of color and found a strong correlation with the existence of superconductivity. The gradation in color from insulator to superconductor and the changes in composition corresponding to the gradation found by quantitative energy dispersive x-ray spectroscopy will be discussed.
(11: 10 a.m.) MICROSTRUCTURE AND OPTICAL MICROSCOPY OF PR CONTAINING CUPRATES: K T Mulville, Naval Research Laboratory DOD SEAP Participant, Washington, DC 20375-5000, L.E. Richards, Naval Research Laboratory, Washington, DC 20375-5000, N.G. Wimer, Geo-Centers, Inc., Fort Washington, MD 20744, V.M. LeTourneau and H.A. Hoff, Naval Research Laboratory, Washington, DC 20375-5000.
PrBa2CU307-x, which is orthorhombic and twinned, is not a superconductor. Er, Pr-cuprates, without an alkali-earth, and Pr 1-2-3 have been prepared and examined with electron and optical microscopy. We have observed colors of polarization in the Er, Prcuprates and found a difference in color between the phases of Pr 1-2-3 similar to the difference between tetragonal ErBa2Cu306+x and orthorhombic ErBa2Cu307-x' The implications of these colors and color differences with regard to detecting cuprate superconducting grains at room temperature will be discussed.
(11 :25 a.m.) QUANTITATIVE METALLOGRAPHY OF Bi-Pb-Sr-Ca-Cu-O SUPERCONDUCTORS*. J.L. Albarran, B. Campillo, J. Juarez-Islas, G. G6ngora and L. Martinez. Instituto de Fisica-Cuernavaca. Apdo. Postal 139-B, 62191 Cuernavaca, Morelos, Mexico. We studied the superconductors Bi2Sr2CalCu,OX, BilPbo.,S~lCalCul.5 Ox and Bil.7Pbo.,Sr2Ca2Cu,O employing X ray, SEM and EDAX microanalysis and reSistivity measurements. We followed the preparation techniques of Maeda et. al and Zhu et al. Several phases were identified in each material including the superconducting L{T c=85k} and H{Tc=110k}. Quantitative metallographic studies revealed that the superconducting phases cover only from 20 to 40 percent of the materials. This is a clear effect of the Pb addition on the amount of the superconducting phases. *Work supported by CONACyT-MEXICO and Programa Universitario de Superconductividad-UNAM.
Annual Meeting-95
INTERFACES IN METAL-CERAMICS COMPOSITES IV: Interface Reactions Sponsored by Process Fundamentals Committee and Composites Committee Wednesday AM February 21, 1990
Room: Plaza D
Inn at the Park
Session Chainnen: S. Fishman, Office of Naval Research, 8001 N. Quincy St., Arlington, VA 22217; W. Barker, DARPA, Washington, DC 20375 (8:30 a.m.)
STUDIES OF BERYLLIDE PRECIPITATION IN Cu/Be/Ni ALLOYS: M. Aindow, P. Pirouz, A.H. Heuer and A. Guha *, Department of Materials Science and Engineering, Case Western Reserve University, Cleveland, OH 44106. Whilst binary copper beryllium alloys with low beryllium contents exhibit high conductivity, they do not undergo substantial age hardening. The solution is the addition of nickel or cobalt in order to stabilise the beryllide precipitate phases. The resulting alloy has excellent conductivity and enhanced mechanical properties following heat treatment. A TEM study will be presented of precipitation in a commercial copper· based ternary alloy containing 0.35 wt % Be and 1.60 wt % Ni. The atomic structure of the interface has been examined using HREM and weak·beam imaging to reveal the presence of growth ledges and misfit dislocations.
Ti-47Al/TiB2 intermetallic composites has been studied by transmission electron microscopy (TEM) and high resolution electron microscopy (HREM). The matrix consists of two variants of twin related ordered L10 O/O-TiAl and ordered DO"~"'2-Ti3Al phases with the topotaxial relationship. The lattice mismatch between the", 2 and '1 phases is accommodated by misfit dislocations. The "'2/-Y grain morphology formed by recrystallization during forging has an interphase interface with regularly spaced perfect misfit dislocations while the misfit dislocations at the interface of "'2 lathes grown inside '1 grains are dissociated with partials bounding stacking faults that run from the interface into the '1 grains. (9: 50 a.m.)
EXAFS AND GLANCING ANGLE X-RAY REFLECI1VITY STUDY OF METAL-MATRlX COMPOSITE INTERFACES: B. Maruyama, Naval Research Laboratory Code 63714555 Overlook Ave. SW, Wash. D.C., 20375-5000, E. V. Barrera, Brookhaven National Laboratory Bldg 480, Upton, NY 11973, R. K. Everett and W. Henshaw, Naval Research Laboratory, and S. M. Heald, Brookhaven National Laboratory. Extended X-ray absorption fme structure, EXAFS, and glancing angle X-ray reflectivity have been used to study the structure of model metal-matrix composite interfaces. EXAFS and glancing angle reflectivity allow the nondestructive determination of the structure of a buried interface, > lOooA helow a surface. EXAFS data yield the identity, coordination number and spacing between the target atom and its nearest, and next -nearest neighbor atoms. Glancing angle X-ray reflectivity yields information on the depth of the interface and the extent of the reaction. In particular, vanadium and gadolinium coatings used as diffusion barriers in the aluminum-graphite system have been evaluated in terms of their stability and resistance to reaction upon heat treatment. Rutherford back-scattering (RBS) data are used in conjunction with the EXAFS data to give a novel picture of the chemical and structural state of composite interface. ( 10: 10 a. m.)
BREAK
(10: 20 a.m.)
REACTION KINETICS OF TiJAl SCS-6 METAL
* Brush Wellman Inc., Alloy Div., 17876 St. Clair Ave., Cleveland, OH 44110. (8:50 a.m.)
THE DISLOCATION STRUCTURE OF INTERFACES IN THE NiO·pt SYSTEM F.·S. Shieu and ~, Department of Materials Science and Engineering Cornell University, Ithaca, New York 14853 The dislocation structure of (001) NiO-pt interfaces was studied using electron microscopy and electron diffraction techniques. Three categories of orientation relationships were identified: exact epitaxy with (OOl)Pt I I (OOl)NiO, [llO]ptll [llO]NiO; small rotations away from exact epitaxy about the common [001] direction, and high index planes of pt parallel to (001) of NiO. Theoretical calculations of the expected dislocation structures of interfaces were made using a Bollmann-type analysis. The experimental observations and theoretical predictions were shown to be in generally good agreement, with some differences with respect to the detailed structure. The reasons for the coexistence of the various categories of orientation relationships will be discussed. The relationship between the dislocation structure and the mechanical properties of interfaces will be examined. (9:10 a.m.)
THE STRUCTURE AND PROPERTIES OF INTERFACE REACTION ZONES IN TITANIUM ALUMINIDE-NIOBIUM COMPOSITES: G. R. Odette, H. E. Deve, A. Hasegawa* and G. E. Lucas, Materials Department, University of California, Santa Barbara, CA 93106
The intermetallic matrix, interface reaction product zones, and ductile reinforcement regions of y-TiAl-NbfTiNb composites were investigated. The composites were fabricated by either powder consolidation and forging or diffusion bonding. Analytical electron microscopy was used to determine the character and spatial extent of the interface phases along with their corresponding microstructures and compositions. The debonding and fracture processes were observed during mechanical testing using both optical and scanning electron microscopy (SEM) techniques, and the interface toughness was estimated. The deformation substructures in the various phases were also characterized using transmission electron microscopy, and fractographic features were evaluated by SEM. The effective toughness of the composites was analyzed in terms of the constrained deformation and fracture behavior of the ductile phase and the debonding characteristics of the interface region. The implications of these results to developing high toughness intermetallic composites will be outlined. • Pennanent address National Institute for Metals, Tskuba, Japan (9:30 a.m.)
INTERNAL INTERFACES IN FORGED XD PROCESSED Ti-47Al(TiB2 ALLOY, O. Popoola, C. Cordier, P. Pirouz and A.H. Heuer, Dept. of Materials Science and Engineering, Case Western Reserve University, Cleveland, OH 44106 TiAl alloys are potential candidates Cor aerospace and aeronautic applications. The structure of the boundaries and interphase interfaces in Corged
MATRIX
C(M'OSITE:
D.R. Schuyler and M.M. Sohi, Garrett Engine Div., Allied/Signal Aerospace, 111 S. 34th Street, M/P 93-394 503-4V, Phoenix, AZ 85034, R. Mahapatra, Naval Air Development Center, Code 6063 Warminster, PA 18974 Ti3Al metal matrix composites (MMCs) are being considered for appl i cati ons in advanced gas turbi ne engines. These materi a 1 s offer high temperature capability combined with excellent specific strength and fatigue resistance. During exposure to processing and servi ce temperatures, MMC' s develop fi ber/matri x reactions that form second phases at the interface. The chemi stry and morphology of these phases determine fiber/matrix bond strength and fracture mechani sms and are key to devel opi ng the des ired properties. Reaction products and kinetics were determined for .several material combinations of Ti3Al matrix alloys, and cerami c rei nforcements such as sil icon carbi de (SCS-6), sapphi re and zirconia. Ti3Al/SCS-6 combinations show less reaction zone formati on than the other rei nforcements. Adequate con sol i dation is obtained at temperatures as low as 1600°F while rapid reaction zone growth occurs at temperatures above 1800°F. (10:40 a.m.)
THE EFFECT OF MATRIX ALLOYS ON THE PROPERTIES OF GRAPHITE/ALUIIINUK COMPOSITES Guoding Zhang Cheng He, Renj ie Wu, Institute of Composite Materials, Shanghai Jiao Tong University, Shanghai 200030, China. The effects of two alwninum matrix alloys, pure Al and LY12 (AI-Cu-Mg alloy), on the mechanical properties of graphite fiber reinforced aluminwn composite wires were investigated. Samples of composites have been subjected to a variety of thermal treatment. The mechanical properties, fracture surface, composition and distribution of phases. interfacial state of the composites were determined with tension and shearing testing machines, scanning electron microscope (SEM) , X-ray energy spectroscope (XES) and transmission electron
microscope (TEM). The mechanism of the effects of LY12 matrix alloy was studied. The results show that alloying elements in the matrix have a significant influence on the composition and distribution of the phases in the composites, interfacial state and consequently on the final mechanical properties of the
graphite/aluminum composites.
LY12 matrix alloy, with high Cu
content, deteriorates greatly the properties of the composite materials. (11:00 a.m.)
NEAR-INTERFACE DEFORMATION IN METAL-MATRIX COMPOSITES: S. Suresh, Division of Engineering, Brown University, Providence, Rhode Island 02912 . This presentation will provide a summary oC our recent experimental and numerical results on deformation in whisker- and particulate-reinforced metal-matrix composites. The model systems examined include cast Al-3.5 weight percent Cu alloys with ditTerent volume Cractions oC SiC particulates and powder metallurgy 2124 aluminum alloys with SiC whiskers. The etTects of matrix microstructure and reinCorcement geometry and distribution on the evolution of matrix plasticity and triaxiality will be addressed and experimental observations oC Cailure modes will be described in the lights oC predicted behavior. Particular
96-Annual Meeting attention will be devoted to identifYing the microstructural and processing conditions which promote void nucleation within the composite matrix as well as at the particle-matrix interface.
1990 LEAD-ZINC AND RELATED METALS SYMPOSIUM X: Application of Lead, Zinc and Tin
Sponsored by the TMS-EPD Lead-Zinc-Tin Committee Wednesday AM February 21, 1990
Room: Salon A & B Marriott
Session Chainnan: N. Ghatas, Noranda Research Center, 240 Hymus Blbd., Point Claire, Quebec H9R 1G5 Canada (8:30 a.m.)
LEAD BATTERIES, POWER AND PROTECTION FOR THE 90'S, M.G. Mayer, Lead Development Association, 42 Weymouth Street, London WIN 3LQ, England. Since its invention in 1859 various designs of lead batteries have evolved to meet commercial needs for traction applications and standby power. In recent years major deveLopments have taken place in lead alloys and the chemistry and mechanical design of industrial lead batteries. These are reviewed in relation to their commercial application together with the new type of maintenance free gas recombination lead-acid batteries which are opening so many new markets for lead. (9:00 a.m.)
INDUSTRIAL APPLICATIONS OF LARGE LEAD-ACID BATTERIES FOR EMERGENCY AND PEAK POWER DEMAND, W.C. Spindler, batteries Included, 15174 Kellen Court, Riverside, California 92506. Five industrial installations of large lead-acid battery energy storage systeas in operation in the United States, West Germany and the Republic of South Africa will be described. One example is for a deep mine shaft in which air ventilation and personnel evacuation require the highest reliability for emergency power. The battery system is designed for additional costeffective benefits of peak load lopping and reactive power contrc1. Commercially available lead-acid batteries of the following types are in use: traction, truck, submarine, high power, and gel. Electrode grids of both the calcium and antimony alloy types are u~ed. Applications include deep and shallow cycling as well as float and intermittent use. The advantages of sealed lead-acid batteries in UPS (uninterruptible power source) systems are will established througho"~ the world, and such system suppliers could readily develop markets for industrial emergency and peak power lead-acid storage battery installations. (9:30 a.m.) LEIID-ffiEE SOIDER 'ro MEET NEW SAFE IRINKIN:> WATER REXiULATIOOS by~.S • .Barux>s, Produ::t Manager (author) and wayne Rupert, Teclmical Services Manager (p:esenter), Engineered Materials
Division, Engelhard Cbrporation, 603 West Street, Mansfield, M'issachusetts 02048.
Sectioo 1417 of the Safe Drinking water Act Aroondnents enacted by the u.s. Calgress prchlbits the use of lead 00Iltaining solders for potable water systans in new CCflStruCtion or when making repairs. The legislature defines "lead-free" solder as cootaining less than 0.2% lead. In the developnent of a new solder it was recognized in additioo to the, lead free requirement, the alloy shruld perfonn as well as the popular 50/50 tin lead solder, both in awlication and properties. Another objective was to have only nootaxic elements. '!his led to the developnent of a 4.0% cx:pper, 0.5% silver and 95.5% tin solder. Review of the merits of other alloy systems and the developnent of the physical properties aloog with test data are presented for the lead free tin rich alloy. (10:00 a.m.)
BREAK
(10:20 a.m.)
PROCESSING AND PROPERTIES OF CAST AND EXTRUDED ZINC-ALUMINUM ALLOY MATRIX - SIC COMPOSITES, M.M. Kim and F.E. Goodwin, International Lead Zinc Research Organization, Inc., 2525 Meridian Parkway, Research Triangle Park, NC 27709. An-27% Al was pressure infiltrated into a column of 20 micron diameter abrasive grade SiC particles to give billets with
particle loading of 50 volume %. Some billets were heated to a temperature where they became semi-solid and diluted with liquid Zn-27% Al alloy to give billets with particle loadings of 12 and 16 volume %. Some of these billets were then extruded into wire with an area reduction of approximately 98%. Tensile properties of the as-cast billets were always inferior to those of the matrix alloys; however, the extruded composite samples showed at 37% improvement in ultimate tensile strength compared to their extruded conventional counterparts. The fracture surfaces of these samples are described. (10:50 a.m.)
EFFECT OF HEAT TREATMENT ON SURFACE MORPHOLOGY AND CRYSTALLOGRAPHY OF ZN AND ZN-FE ELECTROGALVANIZED COATINGS, M.M. Kim and F.E. Goodwin, ILZRO, Inc., 2525 Meridian Parkway, Research Triangle Park, NC 27709. As-electrogalvanized coatings of pure Zn, Zn-I1% Fe, and Zn18% Fe coatings were characterized in terms of initial crystal structure and surface morphology. Samples were then subjected to heat treatments simulating the paint bake cycle of automobile manufacturers. The morphology of pure zinc electrogalvanized coatings varied with heat treatment from platelets to smeared platelets. Zn-11' Fe coatings change from tetrahedral nodules to a mixture of round and tetrahedral nodules, while the Zn-18% Fe coating always have round nodules. A critical transformation time can be defined at any selected temperature, below which the crystal structure does not change. This temperature is related to iron content. (11 :20 a.m.)
ZINC CASTING ON THE HAZELETT TWIN-BELT CASTER, D.R. von Gal, Hazelett Strip-Casting Corporation, P.O. Box 600, Colchester, VT 05446. The majority of zinc strip produced in the world today is cast on the Hazelett twin-belt casting machine. The casting process employs two parallel, moving, water-cooled belts as the major mold surfaces. Together with two strings of side dam blocks, the design offers a four-sided moving mold which produces a continuous strip or bar. The caster was originally developed for aluminum and copper alloy casting. The design was expanded into zinc casting. Within 10 years the industry converted from cast and rolled ingots to twin-belt continuous casting with inline rolling. The paper will cover the history and current technology of this process for zinc strip production.
1990 LEAD-ZINC AND RELATED METALS SYMPOSIUM XI: Processing of Zinc-Lead Bearing Steelmaking Dust
Sponsored by the TMS-EPD Lead-Zinc-Tin Committee Wednesday AM February 21, 1990
Room: Salon C & D Marriott
Session Chainnen: Dr. C.O. Bounds, Horsehead Resource Development Co., 300 Frankfort Road, Monaca, PA 15061; D.A. Paulson, US Bureau of Mines, Rolla Research Center, Rolla, MO 65401 (8:30 a.m.) THE PROCESSING OF STI!ELWORKS WASTE
Rolf Kola, Berzelius UnweIt-Service GMBH Richard-Seifert-Strasse 20, 4100 Duisburg 28 Since "",re than 15 years the Waelz process at "Berzelius" in Duisburg has proved suitable for reprocessing the zinc and lead containing dust and sludge which inevitable occur in making steel in EAF-plants. The produced Waelz-oxide is aggl.-rated in a hot-briquetting plant and charged to an Imperial Smelting Shaft Furnace to recover the valuable elements in form of metallic zinc and lead. In view of the actual importance of the dumping problems of the EAF wastes a description of the "Berzelius" treatment route is given but
also the situation in Western Europe is illustrated. Dust from stainless steel production is successfully treated in the Scan Dust Plasma Process in order to recycle Chromium and Nickel, while the resulting Zinc-Dust is treated via the Waelz-ISF-Route. For treatment of low Zinc-containing dust, sludge and fines from integrated steel~orks, pilot scale tests have been carried out successfully in the Circulating Fluid Bed in order to recover lletalized iron and zinc/lead fume.
30,000 tpy Demonstration Plant is planned.
A
Annual Meeting-97 (9:00 a.m.)
PRODUCTION OF ZINC OXIDE FOR ZINC SMELTING PROCESS FROK EAF DUST AT SHISAKA WORKS N. Tsuneyama, K. Takewaki and K. Yasukawa S1Dlit.-> Ketal Kining Co., Ltd., Nisibara, Nihama, Khime, Japan
been carried out resulting in the capacity of 60,000 tly of EAF
down
the
operating
cost,
fraction is treated in a pressure leaching process.
some
improvements
have
dust as compared with 32,000 t/yr in 1977. The current capacity is equivalent to 20% of EAF dust generated in Japan. The number of workers has been decreased by the introduction of an automatic control system, and the productivity per worker is now twice as much as 10 years ago.
This paper will
describe the progress of the operation at the Shisaka Works
after the previous paper reported in "Zinc • 85 11 and s1lDIIarize the status of EAF dust recycle in Japan.
Hydro-
chloric waste pickle liquors have been used as a leaching liquor. By creating optiaal leaching conditions more than 98% of the zinc is brought into an acid solution, whereas nearly
all the iron remains in the residue.
Sumitomo Ketal Kining Company has been operating Waelz Kiln at the Shisaka Works for 12 years to recover zinc and lead from electric arc furnace (EAF) dust. To increase the capacity and cut
trated in 25% of the dust (the fine particle fraction) by means of slurry-hydrocyclonage. The resulting fine particle
The zinc can be separat-
ed selectively from the solution. The residue can be reused in the blast furnace (via the sintering plant). In the
article
information
is
given about
the
research work
and
-results of the pressure leaching process as well as the possibilities of the process for solving caaparable probleas.
LOW TEMPERATURE Si AND Si-BASED EPITAXIAL STRUCTURES: ELECTRICAL PROPERTIES AND DEFECTS IT: Si-Ge Epitaxy by CVD
(9:30 a.m.)
RECYCLING LEAD AND CADMIUM, AS WELL AS ZINC, FROM EAF DUST S. E. James, ZCA, Bartlesville, OK and C. o. Bounds, !!RD, Monaca, PA 15061 USA production of EAF dust, contains more than 10,000 tpy of lead, 250 tpy of cadmium and 100,000 tpy zinc. Horsehead
Sponsored by the TMS Electronic and Photonic Device Materials and Electronic Materials Committees
Resource Development Company, Inc. (HRD), in conjunction with
its sister company, Zinc Corporation of America (ZCA), has taken a unique approach to achieving total recycle of the toxic heavy metals found in the EAF dust.
Impure zinc oxide,
Wednesday AM February 21, 1990
Room: Orangewood III Quality Hotel
produced by a Waelz kiln and containing the lead and cadmium, is calcined in a rotary kiln at Palmerton, PA, to produce a
refined zinc-rich intermediate for ZeAls zinc plant at Monaca, PA. During the calcining operation, lead and cadmium are
concentrated in a fume product and shipped to ZCA IS electrolytic plant at Bartlesville, OK where a new hydrometallurgical circuit has been constructed and the existing cadmium plant expanded. These facilities permit total recovery of cadmium to refined metal and the lead to a silver-rich intermediate, which is sold to lead smelters. In this paper the Waelzing, calcining and new leaching facilities
will be described. (10:00
a.m.)
BREAK
(10:20 a.m.) TIlE PROCESSING OF ELECTRIC ARC FllRllACE BAGHOUSE DUSTS IN TIlE TETRONICS PLASMA FllRllACE
P. Cowx, C. Chapman, C. P. Heanley, J. Pargeter*
~cs Research and Develo...,nt Co. Ltd.
5b Lechlade Road, Faringdon, Oxon, SN7 9AJ, U.K. International Kill Service Inc.*, 1818 Market Street, Philadelphia, Pennsylvania 19103, USA
The carbotherm.ic reduction of zinc oxide rich Electric Arc
Furnace baghouse dust in the Tetronics transferred arc plasaa furnace
to produce
a
non-toxic
slag
and
a
-.etallic
zinc
product is described. In particular, the operation of the zinc splash condenser coupled to the plasma furnace is exaained and the effect of process variable on zinc recovery is discussed. (10:50 a.m.)
EAF DUST PROCESSING IN THE GAS-FIRED FLAME REACTOR PROCESS C. O. Bounds and J. F. Pusateri, Horsehead Resource Development Co., Inc., 300 Frankfort Road, Monaca, PA 15061-2295 Managing electric arc furnace (EAF) steelmaking dust as an
RCRA-EPA listed hazardous waste places economic pressure on the steelmaking industry. Horsehead Resource Development
Company, Inc. (HRD) serves the industry by recovering metals from the dust in rotary kiln plants and has demonstrated the commercial viability of its patented coal-fired FLAME REACTOR
Process. However, because of location and/or regulation, many steelmakers could be more economically served via on-site
processing of its dust. Therefore, in 1987, through a contract with the Gas Research Institute, !!RD embarked on a program to develop and field test a gas-fired reactor. The development program has successfully proven the environmental, technical and economic performance of the concept.
Consis-
tently high zinc, lead and cadmium recovery and production of marketable slag were achieved. The FLAME REACTOR field test results and the economic performance of on-site or smaller regional plants will be highlighted. a non-hazardous,
Session Chairman: J.L. Regolini, CNET-CNS, 38243 Meylan, France (8:30 a.m.)
(Invited)
LOW TEMPERATURE EPITAXY OF SI:GE:B ALLOYS BY ULTRA·HIGH·VAClJUM/CIiEMICAL VAPOR DEPOSITION, B. S. Meyerson, F. K. LeGoues, P. .I. Wang, J. M. Stork, G. L. Patton, and D. L. Harame, IBM Research Division, T. J. Watson Res. Cntr., P. o. Box 218, Yorktown Heights, N.Y. \059R. We report a broad study of materials properties and devices fabricated em· ploying an isothermal method of epitaxy, UIIV /CVD, operating at a constant 550C, which will be shown to produce highly non.equilibrium yet crystallographically perfect heteroepitaxial films in the Si:Ge:B material set. Substitutionally bound B at concentrations of 5x10 21 cm- 3 is readily incorpo. rated into Si and/or Si:Ge alloy layers, imparting remarkable properties. Si:Ge beteroepitaxy by this method has also been employed to produce high quality !Ieterojunction Bipolar Transistors, as well as enhanced mobility Si:Ge struc· tures. The quality of the Si/Si:Ge interface has been probed via analysis of the electronic function of these devices, and is found to be excellent. IIole mobilites of approx. 4000cm2 (V - see)-' at sheet charge densitcs of 8xlO"cm- 2 are measured by both !Iall and Schubikov-deHaas methods at 14K. Aspects of the processing and electronic behavior ofthesc material sets will be the focus of this talk. (9:00 a.m.) LOW-TEMPERATURE EPITAXY OF Si, Ge AND Sil_XGeX USING S1 2H6/GeH4 /H 2 LOW-PRESSURE CHEMICAL VAPOR DEPOSITION: K. Fujinaga and T. Karasawa, NTT LSI Laboratories, 3-1, Morinosato Wakamiya, Atsugi-shl, Kanagawa 243-01, Japan. Epitaxial growth temperature reduction is demonstrated by a Si 2H6 and GeH 4 low-pressure chemical vapor deposition (LPCVD) technique and a purified H2 -gas flowing process on a surface at a growth temperature, prior to Si 2H6 and GeH 4 gas introduction into a CVD reactor. Epitaxial Si grown on 5i and Ge/Si substrates is obtained at temperatures as low as 520°C and 650°C, respectively, using 5i 2H6/H 2 . Dislocation density of !~e homoepitaxial 5i grown at 5 0oC is on the order of 10 7em , while, by contrast, over 10 cm- 2 dislocations that mainly originate from the Ge and Si lattice misfit (4%) are observed for the heteroepitaxial Si. The heteroepitaxial Ge grown on Si at a low temp rature of 410°C using GeH 4/H 2 includes dislocations of 2xl0 8 cmand no SF. The subsequent fieat-treatm nt ~t 700°C in nitrogen reduces dislocation density to 2xlO 7cm-. Sil_XGeX heteroepitaxy using 5i 2H6/GeH 4 /H 2 is confirmed in the temperature range of 550-700 oC. The dislocation densit of a 550°C-grown layer thicker than a critical thickness is 4xlO cm- 2 .
S
2
7
PRESSIlRK LEACHING OF ZINC-BEARING BLAST FllRllACE DUST Ing. R. Geutskens, Hoogovens Ijmuiden, P.O. Box 10.000, 1970 CA Ijuiden - The Netherlands
(9:20 a.m.) HEAVILY P-DOPED (>10 21 cm- 3 ) 3i AND SiGe FILMS GROWN BY PHOTO-CVD AT 250 oC: A. Yamada , Y.Jia, M.Konagai and K.Takahashi, Department of Electrical and Electronic Engineering, TokiO Institute of Technologi, 2-12-1 O-okaiama, Meguro-ku, TokiO 152, Japan.
Zinc-bearing dust from blast furnace gas-scrubbing operation is a well known environmental health problem in the iron and steel industry. Hoogovens Ijuiden (The Netherlands) developed processes to solve this problem. 90% of the zinc is concen-
Previouslj, we have developed a photo-CVD process for the lowtemperature Si epitaxial technologj using SiH4+SiH2F2+H2 reactants. In this studj, this novel process has been successfullj applied to a low-temperature heavilj-doping teChnique. PH3 and
(11 :20 a.m.)
9S-Annual Meeting
GeH4 were used as a doping gas and a Ge source gas, re spectivelt. Ttpical epitaxial temperature was 250oC. The fair lt good a5reement between the electron concentration and the incorporated ~hosphorus conce ntration was obtained even at this high carrier concentration. The maximum electron concentration of 1.3x10 21 cm- 3 was achieved at the Ge contents of 0.1. The r es istivitt of the sam~le monotonouslt decreased with decreasing the sample temperature, which implies that the carriers were not fr ozen out in l ow tem~erature regions (4K) . Furthermor e , diodes with heavilt doped n-ttpe Si epitaxial laters showed the good rectification. These heavilt doped Si and /or SiGe films were ap~ licable to the base later in Si hetero-bipo lar transistors. (9:40 a.m.) (9:55 a.m.)
BREAK
(Invited)
STRUCTURES AND PROPERTIES OF lRP-GROWN Si-Ge HETEROSTRUCTURES M.P. Scott, Hewlett Packard Circuit Technology R&D, 3500 Deer Creek Rd., Palo Alto, CA. 94303-0867 Significant advances in the growth of pevice quality Si-Ge heterostructures have been realized with the use of chemical vapor deposition techniques. The LRP, or Limited Reaction Processing, approach uses rapid changes in wafer temperature to initiate and terminate the growth of thin layers. The material' properties characteristic of such a growth technique will be reviewed in the context of heterojunction diode and bipolar device performance. The influence of dislocations associated with strain relaxation, interface quality, doping profiles and impurity content will be discussed. Materials requirements for high performance integrated structures will be reviewed relative to alternative growth techniques. (10: 25 a.m.) STRUCTURAL CHARACTERISTICS OF Si AND Si-Ge EPITAXIAL LAYERS GROWN BY RAPID TRERMAL CHEMICAL VAPOR DEPOSITION (RTCVD). M.L. Green, D. Brasen, B. Weir, L.C. Feldman, R. Hull, E. Stevi~ J::Benton, T. Harris, G. Higashi, and J. Bean, AT&T Bell Laboratories, Murray Hill, NJ 07974; *AT&T Bell Laboratories,Allento~ PA 18103. Thin Sil_xGe (x- O to 0.5) films have been grown by RTCVD, a x combines conventional chemical vapor deposition technique that and rapid thermal heating. The structural perfection and purity of these films have been studied by TEM, etch pitting. deep level transient spectroscopy (DLTS), SIMS, photoluminescence (PL) and ion-scatterin~. Intrinsic Si films grown at T~800°C show few defects ( <10 cm- 2) , less than 1012cm- 2 electrically active defects (DLTS). and exhibit bound exciton transitions (PL). These findings are consistent with long carrier lifetimes in the Si, and in fact, high quality diodes and transistors have been produced from these layers. Si-Ge layers produced by RTCVD at 650·C are relaxed to a greater extent than is predicted from the Matthews-Blakeslee equilibrium model. Nonetheless, high gain (' ~ 300) heterojunction bipolar transistors have alsobeen~odced • ......-(10: 45 a.m.)
GexSil_xlSi Heteroepitaxial Growth by the Conventional Chemical Vapor Deposition Technique H.M. Liaw Motorola Inc., 5005 E. McDowell Road, Phoenix, AZ 85008 Present investigation of GexSi l-xlSi heterostructure is focusing on its potential applications for gettering of defects/impurities from the overlay epitaxial layer, and for high speed bipolar transistors. The epitaxial growth was carried out at a temperature between 850°C and 1000 °C at 55 torr. GeH4 and SiH2Ci2 were used for Ge and Si-source gas respectively. Distribution of the Ge/Si ratio between the gas and solid phase is affected to a greater extent by gas composition, and less extent by growth temperature. GexSil-x films to be used for gettering require the formation of misfit dislocations. Density of the misfit dislocations is a functiona of Ge concentration and layer thickness. Gettering of shallow etch pits resulting from metallic impurities in an epitaxial layer was demonstrated by the misfit dislocations. GexSil-x films to be used as base for heterobipolar transistors require a low or no misfit dislocations at the epitaxial hetem-interface. Defect density was found to be inversely proportional to the film thickness. The film as thin as 700 A was obtained with abrupt profiles of Ge and B concentration. (11:05 a.m.) DEFECT-FREE REMOTE HYDROGEN PLASMA CLEAN OF Sll..ICON (100) FOR HOMOEPITAXY L Breaux, B. Anthony, T. Hsu, S. Banerjee, and A. Tasch, University of Texas, Austin, TX Plasma cleaning and plasma-enhanced epitaxy are very important tools in the area of low temperature processing «600°C) of silicon. For low temperature plasmaenhanced epitaxy, a critical requirement is the ability to provide silicon substrates free of surface contamination for subsequent film growth without the use of higher temperatures or introducing surface damage. Hydrogen has been shown to interact with the silicon (100) surface to provide etching of impurities for cleaning and to create ordered surface structures by terminating silicon dangling bonds. In this study, atomic hydrogen produced by a remote plasma is used to remove surface carbon, oxygen, and nitrogen and provide a defect-free surface. The hydrogen-terminated surface provides a passivated surface that effectively resists readsorption of contaminants. RHEED patterns from the surface after hydrogen plasma exposure at different temperatures (20°C-325°C) indicate (Ixl), (2x I), and (3x 1) surface reconstructions: These structures are correlated with
other published results of hydrogen on si/tcon (I (0) and indicate that the surface is ordered with different hydrogen-terminated structures for each surface pattern. Auger data indicates the removal of carbon, oxygen , and nitrogen and, along with residual gas analysis, suggests an etching process of Si and Si02 by hydrogen to remove the surface contamination. (11:25 a.m.) Sll..ICON TEMPERATURE MEASUREMENT BY INFRARED TRANSMISSION FOR LOW-TEMPERATURE EPITAXIAL GROWTH. J C Sturm P.V. Schwartz, and P .M. Garone, Dept. of Electrical Engineering, Princeton University, Princeton, N.J. 08544. While defect formation and stochiometry in heteroepitaxy on silicon can depend critically on substrate temperature, existing methods are not well-suited for the absolute measurement of actual wafer temperature during low-temperature growth, especially if no susceptor is used as in rapid therm al processing. We report the use of infrared transmission to accurately measure silicon wafer temperature in a rapid thermal processing environment during the growth of both silicon and silicongermanium layers. The technique relies on the temperature dependence of bandga p, and does not require the knowledge of emissivity, etc. Depending on the wavelength range selected, one can use the method from under 400 to over 800 C. In practice, the method has an absolute accuracy on the order of a few degrees.
METALLIZATION FOR ELECTRONICS APPLICATIONS m: Metallization in Silicon Based Technologies - Electromigration, Microstructure, Processing and Properties Sponsored by TMS Electronic and Photonic Device Materials and Electronic Materials Committees Wednesday AM February 21, 1990
Room: California I-II Quality Hotel
Session Chainnen: J.R. Lloyd, Digital Equipment Corporation, Hudson, MA 01749; J.E. Sanchez, Jr., CAM Bldg. 66, Lawrence Berkeley Labs., Berkeley, CA 94720 (8:30 a.m.) ELECTROMIGRATION, WHERE ARE WE AND WHERE ARE WE GOING? J.R. Lloyd, Digital Equipment Corp., Hudson, MA 01749 To those in semiconductor reliability, Electromigration needs no definition. Much effort has been expended in trying to control and understand it. Initially, control wasn't too difficult , lowering the current density or alloying Al with CU seemed to work, but understanding it was different. These traditional methods are either no longer adequate, nor appropriate, creating a need to expand understanding for increasingly more difficult problems. In this paper, I will explore questions whose answers have eluded us and propose some that may not yet have been asked. The simple lognormal distribution, so sacred, yet impossible, the relationship of structure to lifetime, activation energy, immortal contacts, and interfacial electromigration will be explored. There may be more questions than answers, but it is hoped that light may be shed on an important and perplexing problem. (8:50 a.m.) TIlE USE OF A lWO FREQUENCY AC BRIDGE TO MEASURE lIf NOISE
IN POLYCRYSTALLINE COPPER Fll..MS: K P Rodbell R.H. Koch, mM T.1. Watson Research Center, Yorktown Heights, NY 10598 We have measured the Iff noise in polycrystalline copper films over a temperature range of 300K to 600K using a two frequency ac bridge. Two lockin amplifiers were used to demodulate the voltage across the bridge to obtain two estimates ofR(t) for the film, each measured at a different modulation frequency. The averaged cross-correlation spectrum then measures the resistance fluctuations of the sample only, with no contribution from the sample's Johnson noise and/or preamplifier noise and hence no subtraction is necessary. Using this technique we have measured the noise magnitude and slope versus temperature and frequency for copper films . Using the Dutta-Hom equation to include a temperature dependent number of defects, we have measured the mean energy of migration and the mean energy of formation of the defects creating the lIf noise. The temperature dependence of the noise magnitude showed a peak at 490K, resulting in a mean energy of migration of 1.2 ev. Values of 0.86 ev for Ef were also measured. Both values are ± 0.10 ev. In calculating the mean energies a value of 1O- t3 S for "to was used.
Annual Meeting-99 (9: 10 a.m.)
MICROSTRUCTURAL EFFECTS ON THE 1/ f" NOISE OF THIN ALUMINUM BASED FILMS. J. G. Cottle & N. S. Klonaris, Dept. of Elec. Eng., Univ. of South Florida, Tampa, FL 33620. The 1/ f" noise was measured on a variety of AI-based thin metal films with widely varying microstructure using an AC bridge technique. The magnitude of the current noise in response to a small, non-destructive AC signal was found to vary several orders of magnitude between 0.01 and 10 Hz and was correlated to the microstructural differences of the films. These differences have a strong affect on film lifetimes as measured in an accelerated electromigration test. Variation in microstructure was accomplished by different deposition temperatures and annealing parameters, and verified using TEM micrographs. Following fabrication, the current noise magnitude was measured and found to be extremely sensitive to variations in film microstructure. The use of this technique to discriminate differences in film microstructure is discussed along with the correlations between excess noise and the quality of the thin film as a metallic interconnection. (9:30 a.m.) EFFECT OF MICROSTRUCTURE ON THE STATISTICS OF ELECTROMIGRATION-INDUCED INTERCONNECT FAILURE: C.V
Thompson
and J. Cho, Dept. of Materials Science and Engineering,
M.I.T., 77 Massachusetts Ave., Cambridge, MA
02138.
Using Al-2% Cu-0.3%Cr alloy films, we have obtained
interconnects with widely varying microstructures, including
multilayered fine-grained lines, large-grained lines, lines with bimodally distributed grain sizes and lines with bamboo
microstructures.
We have also fully characterized the
distribution of failure times for large populations of lines
having each of these microstructures. 'Using these results we have developed a failure unit model which accurately
describes the grain size and line geometry dependence of the median time to failure and the ·lognormal standard deviation in the time to failure. ~se results and their implications for developing and evaluating metallization technologies will be reviewed. (10:00 a.m.) THE EFFECT OF PLANARIZING DEPOSITION CONDITIONS ON THE MICROSTRUCTURE AND ELECTROMIGRATION FAILURES IN AL-2%CUI%SI THIN FILMS. L T McKnelIy Jr, J.E. Sanchez, Jr., R. Mercer, J.W. Morris, Jr., Dept. of Materials Science, University ofCalifomia, Berkeley and Center for Advanced Materials, Lawrence Berkeley Laboratory, Berkeley, CA 94720 Microstructural characterization and electromigration lifetime testing have been completed on 5000A thick AI-2%Cu-1 %Si thin films sputter deposited on thermally oxidized [100] silicon substrates. The matrix of film conductor conditions includes variations in deposition temperature, deposition bias, and annealing treatment, asdeposited condition, rapid thermal annealed (RTA), and furnace annealed. The planarizing (450°C, -100 volt bias) deposition condition produced a larger ultimate average grain size than for unheated-zero bias depositions for all unannealed and annealed conditions. Silicon and copper rich precipitation is shown to be a function of deposition conditions and annealing treatment. The electromigration median time to failure (MTF) for these film conductors is also shown to be a strong function of deposition conditions and annealing processing, with the planarized-RTA conductors exhibiting the longest MTF at the comparison test conditions. Correlations between microstructural features, transmission electron microscopy of failed lines, and MTF are discussed in terms of existing models for electromigration damage and failure. (10:20 a.m.) Morphology and Mechanisms of Electromigration Induced Damage and Failure in Al Alloy Thin Film Conductors: J E. Sanchez, Jr., L.T. McKnelly, Jr., J.W. Morris, Jr., Dept. of Materials Science, University of California, Berkeley and Center for Advanced Materials, Lawrence Berkeley Laboratory, Berkeley, CA 94720 Microstructural characterization, electromigration testing, and failure analysis via scanning and transmission electron microscopy were performed on AI, Al 2% Cu, and Al 2% Cu 1% Si thin film conductors in order to determine the relationships between conductor microstructure and electromigration damage and failure mechanisms. Microstructural characterization included the determination of grain size, grain size distribution, and precipitate reactions and evolution for various film conditions of deposition and annealing. Coarsening of copper rich phase particles is shown to be independent of electromigration induced mass flux. Electromigration testing was performed using the parallel line array technique allowing the simultaneous testing of 25-50 lines under constant current/voltage control conditions. Testing results show that conductor lifetimes are strong functions of annealing. Failure analysis on tested lines via SEM and TEM indicates that there are several mechanisms for void growth and conductor failure which are dependent on conductor composition, and that damage and failures OCCur in regions not depleted of copper. Models for electromigration damage and failure are assessed in terms of these results.
e
(10:40 a.m.) ELECTRICAL PROPERTIES OF AI-Ti/W-PtSi-Si SYSTEM COMPARED WITH THE AI-Mo-PtSi-Si AND Al-Ti/W-Si SYSTEMS. T.J. Zhang, S. Y. Li, Department of Physics, Lanzhou University, Lanzhou, Gansu, People's Republic of China The electrical properties of AI-TilW-PtSi-Si system, especially the ohmic contact and Ihe anti-electromigralion properties, have been sludied. The
surface components, interface profile and the variation of the surface morphology with the annealing conditions of this system have been observed by using AES, ESCA, and SEM methods. The activation energy of the electromigration process is presented. In addition, the application effects of the AI-TilW-PtSi-Si system for SIT devices and Ie are given, which show excellent effects. The action mechanism concerned with these effects are discussed. Finally, the experimental results of AI-TilW-PtSi-Si system are compared with those of the AI-Mo-PtSi-Si and AI-Ti/w-Si systems. (11 :00 a.m.) MICROSTRUCTURE AND BARRIER PROPERTIES OF REACTIVE SPUTTERED TiW NITRIDE. Ivo J.M.M. Raaijmakers, Philips Research Laboratories Sunnyvale, P.O:Box-3409,-Sunnyva:l;;'·CA 94088; and Tarshish Setalvad, Signetics, P.O. Box 10272, Albuquerque, NM 87184. The application of TiW as a barrier metal in the AI-alloy metallization of Si-based bipolar and MOS integrated circuits is well known. It is recognized that the b,arrier performance depends on the presence of impurities like N,O or C. TiW alloys with different N concentrations were sputter deposited from a TiW target in Ar-N 2 mixtures containing upto 50% N2 • The microstructure of these layers was analysed. The barrier performance was evaluated in Si-TiSi,-TiW(N)-AICu and Si-Ti-TiW(N)-AICu metal schemes by electrical and chemical techniques. A part of the TiWoN ternary phase diagram could be composed from the gathered data. The bcc Ti W alloy is found to be converted to an fcc ternary nitride by incorporation of sufficient nitrogen. Although the barrier performance of the bcc phase could be improved somewhat by "stuffing" the grain boundaries with N or an exposure to atmosphere prior to AICu deposition, the fcc nitride always showed better barrier performance than the bcc phase. (11 :20 a.m.) MECHANICAL PROPERTIES AND MICROSTRUCTURAL CHARACTERIZATION OF AI-0.5%Cu THIN FILMS: Ramnath Venkatraman', J.C.Bravman', P.W.Davies", P.A.Flinn", D.B.Fraser", W.O. Nix' 'Dept. of Materials Science and Engg., Stanford Univ.,CA 94305 "Intel Corporation, 3065 Bowers Av., Santa Clara CA 95052 Aluminum-Copper alloys are used extensively in VLSI metalization due to their improved electromigration resistance and reduced tendency towards hillock formation. We have characterized the mechanical behavior of sputter deposited films AI-0.5%Cu alloy composition. These offer greater ease of dry etching as compared to metalizations with higher Cu contents. The variation of stress with temperature in films of thicknesses 0.5 and 1.0 micron, deposited on oxidized silicon substrates was determined by measuring wafer curvature optically. Texture measurements by X-ray diffraction show most grains to be within 5° of {111} orientation with a slight increase in this texture on annealing. Evolution of grain structure and deformation with thermal treatment has been studied by in-situ TEM heating experiments. Stress evolution during thermal treatment is seen to be influenced by excess copper precipitation during cooling and also by film thickness and grain size parameters which are correlated with dislocation glide as a stressrelieving mechanism.
(11 :40 a.m.) THE PROCESSES OCCURING IN AI LAYER ION IMPLANTED WITH 0, N, B, Sn AND HEAVY METALS UPON Si-AI-Si02-AI STRUCTURE ANNEALING. V V Tokarev, Institute of Physics of Solids and Semiconductors, the BSSR Academy of Sciences, P. Browka 17, Minsk, 220726 USSR IC integration on Si brought about the necessity for the development of n-Ievel metallization layers. This required the revision of some process aspects concerning the
fOITllation of conventional Al-based metallization and the search for new materials to
replace AI. Definite progress is made in both directions of solving the problem of nlevel metallization. However it is still impossible to replace AI completely and the problem of hillocks still has to be solved in AI metallization. This paper presents the investigation of the process kinetics occuring in Si-AI-Si02-AI structures with a specified surface topography upon annealing at AI-Si alloying temperature. The samples were studied by electrophysical measurements, Auger electron spectroscopy, cross-section and transmission electron microscopy. The first AI layer in the structure was ion implanted with 0, C, N, Sn and heavy metals. The dopants are shown to suppress the generation of large grains. To say nothing about the obvious effect thermal expansion coefficient matching of the layers constituting that structure, there can be given 2 explanations to the observed effect. Firstly, grain growth is suppressed by dopant segregation at grain boundaries and, secondly, this is caused by metal silicide precipitate formation.
100-Annual Meeting (9:42 a.m.) ISOTHERMAL ANNEALING OF COLD-WORKED ZIRCALOY-4 AND Zr-1wt%Nb TUBING: Derek O. Northwood and John Robinson, Engineering Materials Group, Mechanical Engineering Dept., University of Windsor, Windsor, Ontario, Canada, N9B 3P4; and Zheng Jie, General Research Institute for Non-Ferrous Metals, Beijing, P.R. China.
PHASE TRANSFORMATIONS TMS General Abstract Session Wednesday AM February 21, 1990
Room: ParkA Inn at the Park
Session Chairman: J.K. Lee, Department of Metallurgical Engineering, Michigan Technological University, Houghton, MI 49931 (8:30 a.m.) THE EFFECT OF TERNARY ADDITIONS OF Ge AND Mg ON THE HETEROGENEOUS NUCLEATION OF Pb AND Cd by Al: D.L. Zhang & B. Cantor, Dept. of Met. & Sci. of Mats., Univ. of Oxford, Parks Road, Oxford OX13PH, UK. Hypermonotectic alloys of Al-5wttPb with ternary additions of Ge and A!-8wttCd with ternary additions of Mg have been manufactured by melt spinning. The resultinq microstructures have been examined by transmission electron microscopy. The melt spun AI-Pb-Ge alloys consist of 10-l00nm Ph particles embedded in an Al (Ge) matrix with a cub,e-cube orientation relationship and a cuboctohedral particle shape bounded by (lll) and {IOO} facets. The melt spun AI-Cd-Mg alloys consist of 10-100nm Cd (Mg) particles embedded in an Al (Ng) matrix, with an orientation relationship of (111 )AI! / {OOOI 'Cd' <110>Al/ /<1120>Cd and a distorted cuboctohedral particle shape bounded by (Ill) and (100) facets. Increasing the Ge concentration decreases the lattice disregistry between the Pb particles and the Al (Ge) matrix, while increasing Mg concentration increases the lattice disregistry between the Cd (Mg) particles and the Al (Mg) matrix. The solidification behaviour of the Pb and Cd(Mg) particles in the ternary alloys has been investigated by heating and cooling in a differential scanning calorimeter. The undercooling required for heterogeneous nucleation of Pb particle solidification by the Al (Ge) matrix increases with increasing Ge concentration. The undercooling required for heterogeneous nucleation of the Cd (Mg) particle solidification by the Al (Mg) matrix decreases with increasing Mg concentration. Chemical factors are dominant over structural factors in determining the catalytical efficiency of Al as a nucleant for solidificaticn of Pb and Cd.
(10:00 a.m.) ELECTRON EMISSION AND RECRYSTALLIZATION BEHAVIOR OF W-Re-Th0 2 ALLOYS: B.H. Tsao, D. Tang, B.L. Chen, and D.L. Jacobson, Department of Chemical, Bio, and Materials Engineering, Arizona State University, Tempe, AZ 85287-6006
Intensifying space-power demands require ultra-alloys with maximal melting point, recrystallization temperature, creep resistance for nuclear reactor and attendant high temperature thermionic energy converters. Such ultra-alloys derived from W incorporate Re to reduce creep, recrystallization and embrittlement of W. However, addition of ThO and HfC is desirable to improve the high temperature mechani~al properties of W-Re alloys. In this aspect, the effective work function and recrystallization behavior of W-Re-Th0 2 alloys were studied by using a thermionic emission microscope. The influence of ThO? and temperature compared to that of HfC on the effective work function and recrystallization characteristics is reported. Insitu observations of recovery, nucleation and grain growth in a thermionics emission microscope are also presented. 00:18 a.m.) THE sTRES::I-llmOCED
MAR'tENSl1'E 'tRANSFORMATION AND SHAPII
IIBMORY EFFECT IN Fe-29.5Mn-6Si ALLOY lIT Xu, JiBlong, Jiang (}go; !1!ng. end Ji,Zhiqiang. Shanghai lron&Steel Res.lnst.(SISRI), 'Hu-Song, ShenghBi 200940,China. ABSTRCT, Tha Strass-induced r~ £ martensite transformation and shape aeao_y effect in Fa '29.5 Mn-6Si alloy have been investigeted by meens Of dilatation X-ray diffractometer end electron tranals.toi"mtcroscopy. The original microstructure of speCimen is t phese vi th F .C.C. str ucture proved by X diffraction spectra. The l' ~ (martensite transformation is took piace by external stress. The X-diffraction spectra end TEM shov the phase structure and micrograph), raspectively. By subsiquent varming-up specimen above the As temperature, the specimen exhibits shape memory effect vhich i~ associated vi th ~ ... I" rauersible transformation. The nearly co-' mplete shape memory effect is obteined vith strsin 2.5~. The effects of strain and deformed temperatura on the shape reaavery efficiency have been discused.
(8:48 a.m.)
THE CONTINUUM THEORY OF NUCLEATION IN MULTICOMPONENT SYSTEMS. J.J. Hoyt, Dept. of Mechanical and Materials Engineering, Washington State University, Pullman, WA
99164-2920.
The continuum theory of nucleation, as formulated by Cahn and Hilliard, is extended to the case of multi component systems. The extremal in a general multi component free energy is determined which yields a system of nonl inear differential equations describing the concentration profiles across a flat interface and of a critical nucleus. Computations of the work of formation of a critical nucleus indicate that even small additions of a third ideal element to a binary solution can have a significant effect on the nucleation reaction.
(10:36 a.m.)
(9:06 a.m.)
A STUDY OF THE AGING RESPONSE OF REDUCED QUENCH RATE SENSITIVE URANIUM-TITANIUM-NIOBIUM ALLOYS: G. Mackiewicz Ludtka, Oak Ridge Y-12 P1ant*, P.O. Box 2009, Oak Ridge, TN 37831. The highstrength U-O.STi alloy exhibits significant quench rate sensitivity. This work is directed at developing a high strength, ductile alloy with reduced quench rate sensitivity. Several ternary alloys have been evaluated utilizing Jominy end quench bars. These data, tensile test results, and microstructural observations indicate that the U-O.STi-O.INb alloy is a very promising candidate alloy. This ternary alloy retains up to 10% martensite to a depth of 2.0 in., whereas, in the U-O.STi alloy only 5 to 10% martensite is retained up to a 1.0 in. depth. These results as well as the aging response of these alloys will be discussed. *Operated by Martin Marietta Energy Systems, Inc., for the U.S. Department of Energy under contract DE-AC05-840R21400. (9:24 a.m.) ALLOTRIOMORPH /I,ORPHOJ.OGY KINETICS DUE TO SURFACE DIFFUSION: L. Trice, J. K. Lee, and S. A. Hackney, Dept. of Met. Eng., Michigan Technological University, Houghton, MI
The processes of recovery and recrystallization at 600°C in cold-worked (tube reduced) Zircaloy-4 and Zr-1wt%Nb alloy nuclear fuel sheathing have been monitored using optical and electron metallography and thermoelectric power measurements. The temperature of 600°C was chosen so as to preclude the occurrence of a phase change in either alloy and to limit the microstructural changes to those associated with the removal of cold work. Although recovery produces a small increase in thermoelectric power for both alloys, the major increase in thermoelectric power takes place on recrystallization.
J.
Morphological development of grain boundary allotriornorphs due to capillarity-induced surface diffusion has been simulated. Interestingly, it was observed that a transition exists between the development of an equilibrium morphology and a necking instability due to grain boundary grooving for precipitates having a constant volume. The transition is dependent on the dihedral angle at the grain boundary and the initial dimensions of the allotriomcrph. Similarly, shape development during growth of an equilibrium allotriomorph was also modeled by assuming the flux of material to the growing precipitate phase was transported entirely along the grain boundary. Grain boundary grooving was aga.in found t_o be an important factor. This work was supported by the Department of Energy (DE-FG03-87ER453l5).
PATH PROBABILITY METHOD STUDIES OF B2 AND D03 ORDERING L. Anthony and B. Fultz, Department of Materials Science 138-78, California Institute of Technology, Pasadena, California 91125. The Path Probability Method of Sato and Kikuchi [Acta Metall. 24 (1976) 797] is used to study the kinetics of ordering in alloys exhibiting B2 and D03 ordering. Tetrahedral clusters with each of their vertices on one of the four fcc sublattices are considered in the pair approximation with first- and secondnearest-neighbor interactions. The time evolution of the point and pair variables is obtained together with Kinetic Paths [Acta Metall. 37 (1989) 823] of the order parameters. The path of the alloy through combinations of the different order parameters can be controlled on the basis of temperature and interaction strengths. Comparisons with results obtained experimentally and by our Master Equation Method [1. Mater. Res. 4 (Sep./Oct. 1989)] will also be presented. This work was supported by the U. S. Department of Energy under contract No. DE-
FG03-86ER45270.
(10: 54 a.m.)
Short Range Ordering in Fe3Al
z. O. Gao and B. Fultz,
Pasadena, CA 91125
MS 138-78, California Institute o/Technology,
Mossbauer spectrometry is used to study the evolution of short range order in Fe3Al. Rapidly quenched materials were subjected to thermal annealings at various temperatures well below the critical temperature for D03 ordering. The hyperfine magnetic field distribution reveals a wide variety of local Al environments around 57Fe atoms. By following the ratio of the intensities of those various Al environments during different thermal processings, it was possible to determine that the alloy passes along different kinetic paths on its way of short range ordering. These experimental results will be compared with the results of computer simulation. This work was supported by the U. S. Dept. of Energy under contract DE-FG03-86ER45270
Annual Meeting-lOl (11:12 a.m.)
ORDERING IN COMPLEX STRUCTURES: T F Lindsey and J.W.Morris, Jr., Center for Advance Materials, Lawrence Berkeley Laboratory and Department of Materials Science and Mineral Engineering, University of California, Berkeley, CA 94720. Many complex structures such as intermetallics and minerals are ordered structures. In our current work, ordering wave analysis is being used to predict the ordering transformation paths that lead to the ground state configuration. Our approach provides a simple visual and conceptual means of understanding complex structures. We have applied our method specifically to isomorphic fcc and bcc based structures. The general procedure is also applicable to other types of ordering such as hcp type isomorphism. "This research is funded by the Director, Office of Energy Research, . Office of Basic Energy Sciences, Material Sciences Division of the U.S. Department of Energy under Contract No. DE·AC03-76SF00098.
PROCESS FUNDAMENTALS ill: Thermodynamics and Reaction Kinetics in Metallurgical ProcessesB Sponsored by the TMS Process Fundamentals Committee Wednesday AM February 21, 1990
Room: Salon H Marriott
Session Chairmen: n.R. Gaskell, School of Materials Engineering, Purdue University, West Lafayette, IN 47909; H.Y. Sohn, Department of Metallurgical Engineer· ing, University of Utah, Salt Lake City, UT 84112 (8:30 a.m.) MODELING OF SULFIDE CAPACITIES IN TERNARY SLAGS B Chen and R.G.Reddy, Chemical and Metallurgical Engineering Department, Mackay School of Mines, University of Nevada-Reno, Reno, NV 89557; and M. Blander, Argonne National Laboratory Chemical Technology Division/Materials Science and Technology Program, Argonne, Illinois 60439-4837.
A model has been developed to predict the sulfide capacities in ternary silicate melts based on the Reddy-Blander binary silicate system model and Flood's slag theory. Calculations of Cs for FeO-CaO-Si02' MgO-CaO-Si02 and MnO-CaO-Si02 slags using this model at 1773, 1848 and 1923K are shown to be in good agreement with the available experimental data. (9:00 a.m.)
.
GAS-SOLID REACTION RATE ENHANCEMENT BY PRESSURE CYCLING: M. B. Aboukheshem and H. Y. Sohn, Department of Metallurgical Engineering, Universlty of Utah, Salt Lake City, Utah 84112-1183. An experimental study and mathematical modeling of the effects of external pressure cycling on gas-solid reactions have been conducted. The reduction of nickel oxide by hydrogen was used as an example. The effects of the frequency and amplitude of
oscillation were studied at various reaction conditions. The intrinsic kinetic parameters for the reaction were first measured, and the reduction reaction was carried out under constant
pressure and external pressure cycling. A mathematical model was developed from first principles to establish the extent of the overall reaction rate enhancement and to analyze the experimental observations. The calculated values from the mathematical model are in excellent agreement with the experimental results. Pressure cycling greatly increases the overall rate of a gas-solid reaction. The effects are most pronounced when the overall rate is strongly influenced by pore diffusion. Depending on reaction conditions and the frequency and amplitude of pressure cycling, an order-of-magnitude increase in the overall rate may easily be attained. 'work supported by tne Iron and Steel Complex of Libya. (9:30 a.m.)
REDUCTION OF IRON OXIDE IN AN IRON-OXYSULFIDE BATH BY INJECTION OF CARBON: G. G. Krishna Murthy and J. F. Elliott, Department of Materials Science and Engineering, Massachusetts Istitute of Technology, Cambridge, MA 02139 The rate of reduction of iron from Fe-O-S oxysulfide melts by injection of carbon particles into the melt has been studied by measuring the rate of evolution of CO(g) and CO 2 (g). The temperature was 1200°C, and five different sizes of carbon particles were employed. The effects of particle size, type of
carbon, rate of particle injection, and melt composition were examined. The overall rates obtained are similar to those reported in the literature for reduction of iron-silicate melts for similar experimental conditions. (10:00 a.m.)
BREAK
(10: 15 a.m.)
INTERPRETATION OF OUTPUT OF RADIATION SENSORS IN MEASURING RAPIDLY CHANGING TEMPERATURES OF INDIVIDUAL, SMALL PARTICLES: P. Bolsaitis, Energy Lab., and J. F. Elliott, Dept. of Materials Science and Eng., Mass. Institute of Tech., Cambridge, MA 02139 Opt i cal pyrometry is commonly used to measure the temperatures of particles in various high temperature processes. The radiation emitted from a particle from which temperature is calcu1ated depends on the actual temperature, area, emi ss i vi ty, and surface properties of the particle, but methods of signal detection and processing also impose limitations on the detectability of rapid temperature changes. Thus measurement of temperatures under conditions where particle properties are ill defined and where rapid changes of temperature occur is complex. The time response of one· and two-color pyrometry systems is evaluated and tested experimentally with various types of single particles being heated and cooled rapidly. The effects of vaporization, phase trans it ions, temperature and wave 1ength dependence of emissivities, and instrument response characteristics on apparent temperatures are discussed and methods for calculating true temperatures are presented. (10:30 a.m.)
MATHEMATICAL MODELING OF MINOR-ELEMENT BEHAVIOR IN A FLASHFURNACE SHAFT:' K. W. Seo, Department of Chemical Engineering, and H. Q. Tang and H. Y. Sohn, Department of Metallurgical Engineering, University of Utan, Salt Lake City, Utah 84112-1183. A comprehensive mathematical model for the flash-smelting process that combines turbulent transport phenomena of a particleladen gas jet, chemical reactions, and thermal radiation was extended to enable the prediction of minor-element behavior inside the flash-furnace shaft during copper flash smelting. This model has elucidated the relative importance of elimination of the four most detested minor elements, As, Sb, Bi, and Pb, to the gas phase by volatilization. The basic premise of the volatilization model is that, at the surface of the molten particle, the partial pressures of the minor-element species are those at equilibrium. Transport of minor-element species to the gas is described by external mass transfer and turbulent transport phenomena. Experiments were carried out in a laboratory flash furnace. Reasonable agreement was obtained between the measured and predicted behavior. The effects of target matte grade, oxygen enrichment, and other operating conditions on minor-element
behavior have been examined. 'work supported by 001 Mlneral Institutes program under Grant No. Gl125129 and by NSF under Grant No. CPE-8204280.
(11 :00 a.m.) RATES OF MELTING AND REDUCTION OF IRON ORE, PURE IRON OXIDE, AND METALLIZED ORE PELLETS BY Fe-C MELTS, 1450-1650°C: G. G. Krishna Murthy and J. F. Elliott, Dept. of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139
The rates of melting and reduction of commercial iron ore pellets, and pure hematite, pure magnetite and pure wustite pellets by Fe-C melts have been studied. The composition and vol umetric rate of evolution of product gases have been mea· sured, and the assocsiated changes in composition of the melt have been determined. The effects of temperature and bath composition on the reduction rates have been measured. Melting of meta 11 i zed pellets by Fe-C melts has also been studi ed in comparable experiments.
102-Annual Meeting
REDUCTION TECHNOLOGY V: New Anodel Cathode Materials Sponsored by TMS Aluminum Committee Wednesday AM February 21, 1990
Room: Salon J & K
Marriott
Session Chainnan: Larry Boxall, Comalco Aluminium Ltd., 38/55 Collins St., Melbourne Victoria 3000 Australia (8:30 a.m.) CURRENT EFFICIENCY OF A BIPOLAR lABORATORY ALUMINIUM ELECTROLYSIS CELL: F. NAIXIANG, Q. ZHUXIAN, Department of Non·Ferrous Metallurgy, Northeast University of Technology, Shenyang, Liaoning Province, Peoples Republic of China; K. GRJOTHEIM, Department of Chemistry, University of Oslo, P.O. Box 1033 Blindern, N·0315 Oslo 3, Norway; and H. KVANDE, Hydro Aluminium a.s, P.O. Box 80,1321 stabekk, Norway. A Bipolar laboratory aluminium electrolysis cell was studied experimentally with respect to current efficiency and back ~. The ~ part of the cell, between the bipolar electrode and t~e cathode, had the higher current effICiency and the lower back~. Both current effICiency and back wnLincreased with increasing cell current. The leakage of current from the bipolar electrode directly to the walls of the cell was reduced by increasing the cell current and increasing the height of the bipolar electrode. (9:00 a.m.) RFSULTS OF 100 HOUR ELECfROLYSIS TEST OF A CERMET ANODE: OPERATIONAL RFSULTS AND INDUSTRY PERSPECfIVE, R. D. Peterson, N. E.
Richards, A. T. Tabereaux, Manufacturing Technology Laboratory, Reynolds Metals Company, P.O. Box 1200, Sheffield, Alabama, USA; O. H. Koski, S. C. Marschman, L G. Morgan, D. M. Strachan, Pacific Northwest Laboratory, Richland, Washington, USA. The operating performance of a 15 em diameter prototype cermet anode was evaluated in a laboratory reduction cell at Reynolds' Manufacturing Technology Laboratory, Sheffield, Alabama. The anode was preheated in a muffle furnace and then successfully hot transferred to the electrolysis cell without thermal fracturing. No major operational '" difficulties were encountered in the electrolysis test which lasted for 113.5 hours. A""rage ",current to the anode during the test was 120.9 amps or 0.69 amp cm·2 bas~ on '" conduction through the bottom surface only. A graphite anode of comparable size was used to start the test and to provide a baseline for comparison purposes. The cell '" operating parameters including individual voltage components, anode current density, '" alumina content, current scans, and anode bubble characteristics are discussed.
"'(9:30 a.m.) RESULTS FROM A 100 HOUR ELECTROLYSIS TEST OF A CERMET ANODE: MATERIALS ASPECTS. PM Strachan. O.H. Koski, S.C. Marschman, L.G. Morgan and A.E. Westerman, Pacific Northwest Laboratory, P.O. Box 999, Richland, Washington 99352. A.D. Peterson, N.E. Richards and A.T. Tabereaux, Manufacturing Technology Laboratory, Reynolds Metals Company, P.O. Box 1200, Sheffield. Alabama 35660 In the first paper, the operational results were discussed. In this paper, the resulls from exlensive materials examination will be discussed. After testing, the anode was cut in half and samples taken from the anode that were above, at, and below the electrolyte. The microstructure and phases were found to be different in each of these areas. For instance, the copper oxidized to CuO above the electrolyte while CU20 appeared to be the favored oxidation product in 'he anode in the electrolyte.
(10:00 a.m.) A MATERIALS SYSTEMS APPROACH TO SELECTION AND TESTING OF NONCONSUMABLE ANODES FOR THE HALL CELL: Donald R Sadoway, Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139. Based upon a global analysis of the Hall cell as a dynamic
materials system a new set of selection criteria has been enunciated in order to identify materials for use as nonconsumable anodes and ledge-free sidewalls. The result has been the discovery not only of materials previously ignored for these applications but also of new embodiments of anode design.
Candidate anodes have been tested in cells at MIT at the scale of 1 A and in the research and pilot facilities of an American aluminum producer at the scales of a 10 A and 100 A. The results of this investigation are also of value in paving the
way for the design of radical new electrochemical processes for the extraction of other reactive metals. Research sponsor: Faradaics Incorporated, Boston, Massachusetts.
(10:30 a.m.) SOME IMPORTANT PERFORMANCES OF rIB:! COATED CATHODE CARBON BLOCK: X.A. Liao and YJW.iu, Dept. of Met. Central-South University of Technology, Changsha, Hunan 410012 PR China. Electrical conductivity, resistance to corrosion and wellability towards molten aluminum of TiB2 coated cathode carbon blocks have been measured. The results showed that the coating had beller conductivity than common cathode block. Modified Rapoport Test showed that the expansion of carbon block resulted from sodium penetration, for coated block is 40% less than uncoated block. The wetting angle between molten aluminum and graphite or carbon block are nearly 1800 in the cryolite melt-aluminum-electrode system, but less than 700 between molten aluminum and TiB2 coated block. No effect to the wetting angle were observed while the electrode was polarized.
(11:00 a.m.) PILOT REDUCTION CELL OPERATIONS USING TiBz-G CATHODES: I...B.AbmJ.: D.V. Stewart and A.T. Tabereaux, Manufacturing Technology Laboratory, Reynolds Metals Company Sheffield. AI 35660 L.A. Joo and KW. Tucker Great Lakes Research, P.O. Box 1031, Elizabethton, Tn 37643. A four week operation of Reynolds 3-12 kA pilot reduction cell using titanium diboride-graphite cathode elements manufactured by Great Lakes Research demonstrated the commercial viability of the material. Utilizing two different cathode shapes under separate anodes, the materials served as the primary cathode, effectively reducing anode-cathode distance and cell voltage. Testing under various operating conditions, including close anode to TiB2-G distances demonstrated excellent cell stability with use of the cathode material. The TiB2-G material exhibited excellent thermal shock properties withstanding startup and removal from the cell during operation. Dissolution of the TiB2-G material in the aluminum metal showed acceptable life, however SiC holders of the cathode materials exposed to cryolite were unacceptable. Recommendations are made as to design considerations for commercial cell implementation.
(11:30 a.m.) USE Ot' I:lAUXITE BASED LOW CEMENT CASTABLE IN ALUMINIUM SMELTING APPLICATIONS: L C Edwards and A. Shaw, Comalco Research Centre, 15 Edgars Road, Thomastown 3074. Comalco Research and Technology have developed a low cost ('.astable refractory offering outstanding performance in reduction ~.:;..N ~r layer and metal contact applications. 'Ultracast' is a low cemp·· ~ .~_- .,ich utilises a unique refractory aggregate material known p' :;.:, ~ .... flne bauxite. Ultrafine bauxite occurs naturally with ~~~;deposit located on the Cape York Peninsula in Northe~~rt'\. ~~ Australia. Sintered aggregate prepared from the ultrafinp.a .~ .. ~ lOW chemical reactivity due to its microstructural unif/'" ~~).." .. .,mely low porosity. The unique properties of the aggre"" ~ -() ~~"tutes over 90% of the castable mix are the key to Ultracast's hl'~ ~ance. Plant and laboratory trial in reduction cell barrier layers, "ot metal crucibles and furnace hearths have returned excellent results. Cryolite corrosion resistance and aluminiumnon-welting behavior were found to be superior to wide range of commercial refractory brick and castable products. With the added advantage of joint free monolithic construction, large scale plant trials are now underway with the Ultracast refractory.
REFRACTORY GOLD, SILVER & PRECIOUS METALS PROCESSING I: Precious Metal Plant Design Sponsored by: TMS Precious Metal Committee Wednesday AM February 21, 1990
Room: Salon 1
Marriott
Session Chainnen: R.J. Wesely, Davy McKee Corp., 2430 Camino Ramon, San Ramon, CA 94583; T. McNulty, P.O. Box 2740, Evergreen, CO 80439 (8:30 a.m.) ECONOMICS OF REFRACTORY GOLD ORE PROCESS.
John E. Litz and
R. Wayne Carter, Hazen Research, Inc., Golden, Colorado. Refractory gold ore pretreatment processes continue to receive much attention. This paper compares the relative economics of treating ores by oxygen pressure oxidation, roasting, and bacterial oxidation with or without preconcentration by froth flotation. Capital and operating cost estimates are compared for each method of pretreatment. These estimated costs, estimated recoveries, and sulfide grades are then used to compare the process economics. The opportunities for alternative pretreatment processes also are discussed.
Annual Meeting-lOa
(9: 15 a.m.) DESIGN AND START-UP OF THE COVE MILL, M. Jennings, Echo Bay Management Corp., 5270 Neil Road, Suite 101, Reno, NV 89502, . O. Walls, Echo Bay Minerals Co., #1 McCoy Road, Battle Mountain, NV 89820, R. Wesely, Davy McKee Corp, 2430 Camino Ramon, Suite 100, San Ramon, CA 94583 In 1987, Echo Bay discovered a major silver/gold ore body at their McCoy property in Nevada, and quickly made a decision to build a 5000 tpd mill which was expanded during the design to 750.0 tpd. Since the ore reserves consisted of cyanide leachabl. oxide as well as sulfide, the plant flow sheet involved an oxide leach, a gravity circuit, and lead sulfide and a pyrite auIfide recovery by flotation. The paper describes the process development, equipment selection, plant design anq the start-up of the oxide treatment. (10 : 00 a.m.)
I£OOnR:;I.OC a:HiIIlRATI!N; 'IN 'llE JElT!N IF A I!Il){]{I!)ID(N RANl': A ~ SltDi RR A mRICllRf ~ CINlNJlWE. Dr.
~
A. Spn:er
an. Julia R.
Ilrlm, 'l'r:nwrld ~
an. Mirernls
(Aiitriiba) Pty. lid., 13-15 MOOn St, K&mle 61ffi \e;tem Iustnllia.
'nE rretallurgicallJlIlllEterS re:essary for fu! cEs~ of a ClJIIIEICial scale bio-acictitim plmt ~ 1Em ret.emtirB:I for a refulctorv ~te arant::rnte ~ a I!'IXleratelv &eurqtri.lic la:terial rulbJre. 'nE lJIIlIlEterS exmideml W'!re bio-acictitim rate, resian:e tine, rulbJre gIOith rate, solID! dimity, leo.cl of bio-acictitim an. CJ<)'!\EI\ qltS
(11:15 a.m.) PROCESS DESIGN CONSIDERATION OF THE GOLD PLANTS IN CHINA, Tian Jianjun, Changchun Gold Design Ins titute , 12 South Lake Road , Change hun , Jilin, China
During the last decade, the go ld industry of China has achieved rapid development with the great attention paid by the government and the application of new exploration, mining and
metallurgical technology. Meanwhile, the metallurgical process design has been facing the new challenge. Not only have the process design had to attach great important to the feasibility of technology an d the reasonable ne ss of economics, but also t o the environmental protection, capita l cost control and mine management. Moreover, the development of refractory ore treatment proces s has become the important work for the
research and engineering people of China.
The paper will
discuss the process design consideration in China from the aspects technology, economics. environment protection and mi ne management.
solidifi ed and quenched , have been examined on severa l t ransverse dendr ites . Isoconcentrati on contours across the pr imar y de ndrites were used to identify t he most appr opriate method to obtain th e relationship between fraction so lid and the corresponding solute concentrati on at the liquid- solid interface. The partiti on coefficient s of tantalum, titanium, and aluminum were ob served to be less th an unity; while that for tungsten and cobalt were greater th an unity. Thi s i s Qualitati vely similar to their nickel base binari es. The partition coefficients were temperature independent in the range from 15B4 to 1608 K. Microsegregation profile s compare well with predictions fr om analytical models developed for binary alloy s. (8:50 a.m.) PRIMARY AND SECONDARY DENDRITE SPAC INGS IN DIRECTIONALLY SOLIDIFIED SUPERALLOY, PWA-1480: M. Vijayakumar, S. N. Tewari , Cleveland State University, Cleveland, OH 44115, J. E. Lee , P. A. Curreri, NASA Marshal l Space Flight Center, Huntsvi lle, AL 35812.
This presentation is part of a detailed investigation of solidification behavior of multicomponent commercia l superalloy, PWA1480. Directionally solidified single crystal speci mens have been used to investigate the dependence of primary dendrite spac ing s and secondary arm coarsening kinetics on process var iables. The side branch coarsening kinetics show a reasonable agreement with the predictions from a simple analytical model proposed by Kirkwood for binary alloys. Growth rate and thermal gradient dependence of primary dendrite spacings compare well with predi ctions from the model due to Hunt . (9: 10
a.m.)
MORPHOLOGICAL ANALYSIS OF DIRECTIONALLY SOLIDIFIED HYPERMONOTECTIC AI-In ALLOYS, J. B. Andrews, University of Alabama at Birmingham, Birmingham, AL 35294, A. C. Sandlin, National Institute of Standards and Technology, Gaithersburg, MD 20899 There
exhibit
has
been considerable interest recently in alloys
which
liquid immiscibil i ty . It is thought that if the immiscible liquid can b e incorpo rated into the sample in an aligned fashion during solidification, then these alloys may
possess unique and useful properties. In t his study, several hypermonotectic AI-In alloys we re directiona lly sol idified under various growth conditions in an attempt to understand the solidification mechanisms and to determine the feasibility of producing aligned, compo site~like structures. Incorporation of the immiscible liquid and production of aligned structures appears feasible . However , interesting results were obtained that were not predicted from a stri c tly theoretical ana lysis. Ex isting models for these structures will be presented a nd discut=:t=:ed.
(9:30 a.m.) MORPHOLOGICAL ANALYSIS OF IMMISCIBLE Cu·Pb·AL ALLOYS DIRECTION· ALLY SOLIDIFIED UNDER ALTERNATING GRAVITY LEVEL CONDITIONS, A. C. Sandlin. National Insti t ute of Standards and Technology, Gaithersburg , MD 20899 , J . B. Andrews, University of Alabama at Birmingham, Birmingham , AL 35294 Hypermonotec tic alloys exhibit a region of two liquid immiscibility which can lead to settling and massiv e phase separation
during
solidification
under
l-g
conditions.
Sedimenta tion
should be eliminated and the microstructure altered by solidifying under l ow-g conditions. In this research, immiscible eu-pb-AI alloys were dire ctional l y solidified on NASA's KC - 135 zero-g aircraft which provides a pe rio dic
SOLIDIFICATION PHENOMENA I
variation from low·g (O.Olg) to high·g (1 . 7g) by repeatedly flying parabolic trajectori es. Low mis c ibility gap height samples (low Al content) exhibited a marked decrease in mean particle siz e during low·g solidification. This was attributed
TMS General Abstract Session
to the lack of convection augmented coalescence of hypermonotectic phase in advance of the interface .
Wednesday AM February 21, 1990
Room: Salon G
Marriott
Session Chainnan: Dr. Hani Henein, Department of Mining, Metallurgy & Petroleum, University of Alberta, 606 Chemical-Mineral Engineel"ing Building, Edmonton Alberta T6G 2G6 Canada (8:30 a.m.) MICROSEGREGATION IN DIRECTIONALLY SOLIDI FIED PWA-1480 SINGLE CRYSTAL NICKEL BASE SUPERALLOY: S. N. Tewari , M. Vijayakumar , Cleveland State Uni versity, Cleveland , OH 44115, J . E. Lee, P. A. Curreri, NASA Marshal l Space Flight Center, Hunts ville, AL 35812. Microsegregation behavior in a multicomponent ni ckel base superalloy , PWA-1480, has been investigated. Solutal profiles in single crystal spec imens, which were partially directionally
dome·height
alloys
(high
Al
content)
formed
the High
large
hypermonotectic droplets du r i ng high-g solidification .
(9:50 a.m.) SOLUTE DISTRIBUTION DURING STEADY STATE CELLULAR GROWTH: F.S.Basak and S.K.Fabi, Dept. of Metallurgioal Engineering, Indian Institute of Teohnology, Kharagpur 721302, INDIA. A theoretioal model is developed to prediot the solute distribution in the viofnity of the steady state cellular growth front during unidirectional solidifioation of alloy melts using Gelerkin's Method. The influence of the overlapping diffusion fields of the neighbouring cells and the mass conservation at the interfaoe have been taken into aooount. The model indioates the existenoe of finite oell length at different growth conditione and the results are oompared with partially direotionally solidified and quenched experiments on the Sn-Fb alloys. Solute oompoeition profiles in the quenohed liquid ahead of the eolid-liquid boundary are in agreement .i~A the model. The effect of localized surfaoe energy on the smooth stable oell shapes hae also been examined.
I04-Annual Meeting (10: 10 a.m.) SOLUTION OF COUPLED THERMAL-DIFFUSION EQUATIONS WITH APPLICATIONS TO EUTECTOID GROWTH. Rahul Dey Basu Mechanical and Aerospace Engg. Dept .• NCSU. Raleigh.NC A coupled system o~ di~~erential equations describing steady state eutectic solidi~ication is modelled by including heat trans~er and mass di~fusion. Radial and axial ef~ects are included. A nonlinear equation · is obtained and lineariaed with the aid of the steady state solution for heat flow. An integral transform technique is used to obtain a solution ~or generalised boundary conditions. The effects of external and also internal sources and sinks is capable of inclusion. A comparison with the thermogradient coupling used in other coupled mass and heat ~rans~er formalisms is made.
SUPERPLASTICITY IN AEROSPACE IV, 2nd International Symposium Sponsored by the TMS Shaping and Forming Committee Wednesday AM February 21, 1990
Room: Garden Rooms 1·2·3 Inn at the Park
Session Chairman: N. Chandra, Department of Mechanical Engineering, Florida State University, Tallahassee, FL 32306 (8:30 a.m.) SUPERPLASTIC DEFORMATION CHARACTERISTICS DURING DYNAMIC RECR,(STALLIZATION IN AI-2.4Li-I.2Cu-0.6Mg-0.12Zr ALLOY: Hamilton B. Ren, Department of Mechanical and Materials Engineering, Washington State University; and H. C. Heikkenen, General Dynamics Fort Worth.
c.....a
Previously cold- or warm-rolled AI-Li alloys have been shown to undergo dynamic recrystallization during the initial 40-80% strain, and exhibit corresponding dynamic changes in the hardening and rate sensitivity of the flow SlCess during this microstructural evolution. These changes are found to effect the diffuse necking process during the superplastic deformation, and understanding of these characteristics can lead to enhancement in the forming process, Details of the deformation such as increased forming rate. characteristics during the dynamic recrystallization were evaluated in this study, including the development of the back stress, strain hardening, and strain rate sensitivity. These flow properties were employed in explaining the diffuse necking characteristics during the initial stages of superplastic deformation. and implications of the
dynamic changes on characterization methods is discussed. (9: 00 a.m.)
SUPERPLASTIC FLOW CHARACTERISTICS AND MICROSTRUCTURAL ANALYSIS OF ALCOA 2090-0E16 ALLOY: R.E Goforth and M. Srinivasan, Texas A & M University, N. Chandra, Florida State University Superplastic flow characteristics and microstructural evolution
in a commercial Aluminum-Lithium alloy (ALCOA 2090-0E16) were investigated over strain-rates from approx. 10-' to 10-2 s-l, temperatures from 490'C to S27'C, and back pressure of atm. and 400 psi. The material was thermomechanicially processed by Alcoa for dynamic (continuous) recrystallization. The Constitutive relationship which generally characterizes fine grained materials was
investigated as
well
as
the
instability parameter.
The
microstructural evolution that occurs during the deformation process, e.g., static and dynamic grain growth and cavitation was evaluated using light microscopy and a density gradient column. TEM was utilized to study such details as the substructure and precipitates. An rnstran uniaxial tensile test machine modified for computer~control in order to maintain constant strain-rates and acquire data . A retort was incorporated the system in order to test at hydrostatic pressures up to psi.
was true into 1000
(9:30 a_m.) THE INSTABILITY OF SUPERPLASTIC DEFORMATION IN P/M AND 11M 7475 ALUMINUM ALLOYS: K. Higashi, H. Imamura'" • T. Ito * and S. Tanimura, Dept. of Mech. Eng,. * Dept. of Met. Eng •• University of Osaka Prefecture, Mozu-umemachi, Sakai, Osaka. 591 Japan.
P/M 747S-0.7Zr produced by powder-metallurgy shown superplastic elongation of more than 500% in the range of strain rates from 10- 2 to 10 0 s -I and at temperatures from 723 to 798 K. In particular. the maximum elongation of 1200% was obtained at 788 K and a strain rate of 7*10- 2 5- 1 which is two orders of magnitude faster than that for 11M 7475 by ingot-metallurgy.
This excellent superplastic performance at higher strain rates is due to fine grain size of about 2 ~ m develo ped by d ynamicrecrystallization during the early stage of s uperplastic flow. In this work. the stability of deformation in p/M 7475 is considered by instability parameter, I(s)-(I-y-m)/m • The both values of m (=lnallnE) and y{= 00/0£'0) at any strain respectively are measured under the constant true strain rate test. For P/M 7475 the deformation proceeds unstably (O < 1 < 3) from small strain to failure, means tha t the formation and/or the growth of necking proceeds gradually during superplastic flow.
(10:00 a.m.) PROCESSING, MICROSTRUCTURAL REFINEMENT AND SUPERPLASTICITY OF NAVALITE: E -W Lee, Naval Air Development Center, Warminster, PA 18974 and T. R. McNelley, Naval Postgraduate School, Monterey, CA 93943 Significant structural weight savings may be attained by fabrication of components utilizing superplastic forming of light weight AI-Mg-Li alloys. Thermomechanical processing and superplasticity of NAVALITE (Al-2.7Mg-2.0Li-0.SCu) has been examined. The essential feature of the processing is warm rolling at ~300°C with controlled reheating intervals between consecutive rolling passes. For material given an initial averaging treatment prior to the warm rolling , superplastic elongations were attained at temperatures from 400'C to SOO'C, with the maximum elongation being 670% at 4S0'C and 8 X 10- 4 s-l strain rate. In the absence of the overaging the maximum elongation decreased to 430% at 400'C and 8 X 10. 4 s-l strain rate. Microstructural analysis indicates that the processing results in a refined microstructure which evolves by continuous recrystallization . NAVALITE is heat treatable to a yield strength of 400 MPa at room temperature .
(10:30 a.m.) THE ROLES OF Mn AND Zr ADDITIONS IN SUPERPLASTIC AI-Mg ALLOYS: T R.McNelley, Materials Group, Department of Mechanical Engineering, Naval Postgraduate School, Monterey, CA 93943 and S.J. Hales, AS&M, NASA-Langley Research Center, Hampton, VA 23665. Refined microstructures may be produced in an Al-10Mg-0 . lZr (wt. pct.) alloy by static continuous recrystallization during isothermal rolling at 300·C. Recovery produces boundaries of progressively increasing misorientation provided the evolving structure is stabilized by interme tall ic p-phase (Al.Mg,) precipitates, 0.5-1.0 ~m in size and located at triple junctions. However, superplastic ductility is limited to the range 27S'C to 32S'C, i . e . temperatures below the p phase solvus . The substitution of 0 . 5 wt.pct . Mn for the Zr in the alloy results in superplastic ductility over a broad temperature range from 27S"C to SOO·C . This is attributed to the formation of rod-like precipitates of the AI.Mn phase, O.OS~m in diameter and 0.2~m in length. This phase remains stable above the Mg solvus and prevents coarsening of the continuously recrystallized grains. (11:00 a.m.)
SUPERPLASTIC HERA VIOR AND MICROSTRUCfURE OF 7475 Al DEFORMED IN AN EXTERNAL ELECTRIC FIELD: W. D. Cw. X. P. Lu, A. F. Sprecher and H Conrad. Materials Science and Engineering Department. Raleigh. North Carolina 2769S-79f.J7. Effect of an external electric field on superplastic deformation and resultant microstructure of 7475 aluminum alloy has been studied as a function of test temperature and strain rate. The extcma1ly applied electric field can n:ducc the flow stress. as shown by an electric field cycling test, the work hardening of 7475 AI. These effects become more significant with iocreasing test temperature and decreasing strain rate. The strain rate hardening exponent was also increased by the externally applied electric field. The changes in microstructure resulting from the applied electric field. particularly the dispersoid-free zones (OFZ) ncar grain boundaries, were studied by optical microscopy, EDS and TEM. A wider DFZ was found in the 7475 alloy deformed in an electric field. The structure and composition of the DFZ in specimens with electric flCld were sigoifJCandy different from thosc without. According to thesc results, the possible mechanisms responsible for the electric field effect arc discussed.
Annual Meeting-105
THERMAL ANALYSIS OF METALLURGICAL SYSTEMS I: Principles and Techniques Sponsored by the TMS Chemistry and Physics of Materials Committee Wednesday AM February 21, 1990
Room: California llI·IV Quality Hotel
Session Chairmen: R.D. Shull, Metallurgy Division, lMSE, National Institute of Standards and Technology, Bldg. 223, Rm. B150, Gaitherburg, MD 20899; A. Joshi, Lockheed Palo Alto Research Laboratory, Bldg. 204, Orgn. 93-10,3251 Hanover St., Palo Alto, CA 94304-1191 (8 :20 a . m. ) INTRODUCTORY REMARKS
(8 : 30 a.m. ) Invit.ed APPLICATION OF THERMAL ANALYSIS AND COMBINED TECHNIQUES TO POLYMERIC MATERIALS, R. B Prime, IBM General Products Division, San Jose, CA 95193 . This lecture will focus on a general description of thermal analysis techniques , utilizing polymeric materials a s typical applications . Differential s canning calorimetry (DSC) , thermogravimetric analysis (TGA) , thermomechanical analysis (TMA) , and dynamic mechanical analysis (DMA) techniques will be described as well as the combined TGA/MS and TGA/IR techniques. Kinetic, compositional and mechanical property de termination will be emphasized. Kinetic topics will include cure of thermosetting polymers and decomposition of polymers by DSC and TGA techniques . Measurement of thermal expansion and glass transit i on by THA , and the use of DMA to measure modulus and damping characteristics will be discussed. Compositional analysis by TGA will be described , followed by characterization of weight loss processes by TGA/MS and TGA/IR.
(9 : 00 a.m.) (Invited) APPLICATION OF TH ERMAL ANALYS IS TO STUDIES OF THE SYNTHESI S OF CERAMICS AND GLASS , P. K. Gallagher , AT&T Bell Laboratories , Murray Hill, NJ 079 74 . Thermoanalyt i cal methods can be immensely us e ful t o the scientist or engineer engaged in the preparat i on of new materials or the innovation of new synthetic proc e sses. Five major areas of a pplication a re: 1 ) establish ing the phase equilibria a s sociated materia l s system including any metastable as pec ts; 2) under s tanding the f o rma tion and decompos ition of a ny precu rsor used i n the synthes is ; 3) eva lua t i ng and c omparing the reactivity o f t he individual r eact a nt s , e.g., Fe203 or A1 203; 4) following proc ess steps, e.g., binder burnout, sintering, etc.; and fina lly 5) characterization of the product t o as s ure succes s o f the synthesi s . Illustrative examples from ea ch stage are described. Invited) (9 : 30 a.m.) THERMOMAGNETOMETRY AS A MEANS FOR FOLLOWING CORROSION AND OTHER LIQUID· SOLID CHEMICAL REACTIONS , A REVIEW, R G Cha rles, Consultant , Westinghouse Sci ence and Technology Center, 1310 Beulah Rd., Pittsburgh, Pa. 15235. Thermomagnetometry is a vers a tile and convenient, but much n e glected , means for following corrosion and other chemical reactions in bo th aqueous and nonaqueous liquids. The method i s based on the cha nge in magnetic properties of the re ac tion mixture , t a ke n a s a whole , a s meta l (or other reac tant) is converted to a reaction product. Reactions (requi ring only a few mg of reactants) are readily ca rried out, over a wide temperature range, in small sealed glass or metal capsule s. With nonmagnetic sealed metal capsules, it is also practical and simple to carry out reactions under pressure (~aqueous systems above 100°C ). Fas t quantitative measurements may be performed in situ on the magnetic balance while at temperature or on a ba tch of capsules after heating and cooling. A number of examples of reaction thermomagnetometry are desc ribed which employ liquid reactants as different as water and liquid alkali metals. Both batchwise and programed on· balance heating examples will be included.
(10 :00 a.m.) (Invited) IN-SITU THERMAL MEASUREMENT OF COARSENING KINETICS IN SOLIDLIQUID MIXTURES : S . P. Marsh, Naval Research Laboratory, Washington, D.C., 20375-5000, and M. E. Glicksman, Materials Engineering Department, Rensselaer Polytechnic Institute, Troy, New York, 12180-3590.
A technique has be en developed by the authors to monitor the ki netics of phas e coarsening in pure material s t hrough thermo-
capillary effects. A platinum RTD/AC bridge system, with a resolution of ±100 ~K, is used to measure the decay of the mean undercoa ling associated with the decreasing interfacial curvature in so l id-liquid mixture s of succinonitrile, ethylene carbonate , and a-P 4 , all held adiabatically to preserve the initial phas e fractions. Results confirm the time- 1 / 3 kinetics predicted by coarsening theory, whereas the volume fraction dependence of the coarsening rate suggests that the thermal signal tracks the specific surface area in the microstructure , Sv. Application of this t echnique to metals will be discussed. This work is supported by the National Science Foundation, under grant DMR86-11302.
( 10:30 a.m. ) Invited HIGH- TEMPERATURE STUDIES OF LEVITATED SOLID AND LIQUID SAMPLES: John L. Margrave, Robert H. Hauge and Na tha n Norem , Dept . of Chemistry , Rice Univers i t y , a nd Houston Area Research Center, P. O. Box 1892, Houston , Texas 77251, and Shankar Krishna n, Intersonics, Inc., 3453 Commercial Ave., Northbrook, IL 60062. Levitation techniques supplemented with induction, radiative or laser heating, allow one to achieve hi gh-t emperatures withou t contain e r s a nd, thus , to minimize con t amina tion . High-speed photogr a phy of levitated liquid samples (Cu and Ni ) s hows drople-t os cill a tions which ca n be related to s ur f ac e tensi ons and viscositi e s. Ellipsometr ic measurements on Cu, Ag, Au, Ni, Pd , Pt, 2r, AI , Si , Ta, W, we and Tae have yield ed real time optical properti es , £<",1) and true temperatures of levitated samples . By combinin g l e vitation with traditional drop calorimetry , we have det e rmi ned enthalpy increments, hea t ca pa cities and heats o f fu s i on f o r man y tran si tion me tals and allo ys .
(Invited) ( 11 :00 a. m. ) LOW-TEMPERATURE SPECIFIC HEAT OF METALS Norman E. Phillips, Lawrence Berkeley Labo rat ory, Berkeley, CA 94720 Measurement s of the low-temperature specif ic he a ts of metals can give fundame ntal information related to the electronic , vibrat i ona l, magnetic, and superconduc ting properties. Some results of this approach t o the study of metals will be outlined and illustrated by examples, including examples drawn from work on the new, high-Ie' oxide superconductors.
(11:30 a.m . )
Invite:!
HEA TCAPACmES OF SOLID POLYMERS.·Bernhard Wunderlich, Department of Chemistry, University of Tennessee, Knoxville, TN 37996-1600 and Division of Chemistry, Oak Ridge National Laboratory, Oak Ridge, TN 37831-6197. The thermal properties of solid, linear macromolecules are accessible through heat capacity measurements from about 10K to the glass transition or the melting transition. By measuring and collecting data on over 100 polymers, a data bank was established and used as a base for detailed correlation with an approximate frequency spectrum for the polymers. Besides assessment of the entropy at zero kelvin of disordered polymers, this heat capacity knowledge has helped in the elucidation of partial phase transitions and conformationally disordered crystal phases. A link has also been established to measurements of mobility through solid state nuclear magnetic resonance . Most recently heat capacity measurements have been linked to full dynamic simula· tion of crystal segments with 1900 chain atoms. Questions of disorder and anharmonicity can thus be analyzed. The work is summarized as the Advanced lliermal Analysis System ATHAS.
THERMAL PROCESSING OF ALUMINUM ALLOYS TMS General Abstract Session Wednesday AM February 21, 1990
Room: ParkC Inn at the Park
Session Chairman: P.S. Gilman, Allied Signal, Inc., P.O. Box 1021 R, Morristown, NJ 09762-1021 (8:30 a.m.)
EFFECf OF PARTICULATE TYPE ON MICROSTRUCIURE OF STIR CAST ALUMINUM METAL MATRIX COMPOSITES·: 1. P. Lucas. N. Y. C. Yang, and J. J. Stephens, Sandia National Laboratories, Livermore. CA 94551-0969
Particulate reinforced aluminum metal matrix composites (MMC) show significant improvement in both mechanical and physical properties over
lOG-Annual Meeting unreinforced materials. However, to optimize properties of MMC systems, physical metallurgical processes between the matrix and particulate reinforcement must be well understood. In particular, reactions at the particulate-matrix interface are of major importance. In this investigation, it was observed that the interfacial precipitation near the particulate-matrix interface was dependent on the type of reinforcement. For an A3S6/SiCp MMC processed by stir casting, there was virtually no interfacial precipitation (i.e., no reaction wne). In contrast for an A3S6/B4Cp MMC, a significant interfacial reaction wne (- I jJlIl thick) with a host of precipitate phases was observed. When significant amounts of alloying elements are partitioned to the particulate-matrix interface in MMCs, a depletion of these elements is necessary in the bulk matrix material. Identification of interfacial and bulk matrix phases was obtained for these composites by various analytical techniques including TEM, SEM and optical microscopy. *This work supported by the U. S. Department of Energy, DOE, under contract NDE-AaJ4-76DP00789. (8:50 a.m.) Effect of reinforcement addition on the early stages of precipitation in alumina particulate reinforced 6061 Al MMCs I. Dutta and J. L. Hafley, Department of Mechanical Engineering, Naval Postgraduate School, Monterey, CA 93940. Electrical resistivity measurements were conducted during isothermal aging treatments on an alumina particulate reinforced 6061 AI MMC and a monolithic 6061 Al control material. In addition, differential scanning calorimetry was used to investigate the growth kinetics and thermal stability of the metastable phases in the MMC and the control sample. From the DSC experiments, the absorbed heats due to the dissolution of the metastable phases were determined as functions of aging time and temperature. The activation energies of formation/dissolution of these phases were also determined. These results were correlated with the results of the resistivity data to yield a comparison of the aging kinetics of the MMC and the monolithic alloy. * This work was sponsored by the Naval Postgraduate School Foundation (9:10 a.m.) Effects of HAZ Thermal Excursions on Microstructure and Properties of 2090 Al Alloy: A J Sunwoo and J.W. Morris, Jr., Center for Advanced Materials, Lawrence Berkeley Laboratory, I Cyclotron Road MS 66-200 Berkeley, CA 94720 In the weldment, there are three distinct microstructures: fusion zone, heat affected zone (HAZ), and unaffected base metal. The HAZ undergoes a microstructural changes due to a temperature gradient established by heat conduction from weld. The HAZ of 2090 Al alloy is of interest because 2090 is a precipitate-strengthened material, and thereby any thermal excursions will affect the precipitation behavior and properties. The purpose of this research is to ",,-characterize and to determine the effects of thermal excursions on the =microstructural and property changes of 2090 AI alloy. The HAZ thermal ",,-excursions were simulated using a GLEEBLE on both 2090-T3 and -T81, and the tension tests were conducted at 293 K and at 77 K. The results indicate that =for peak temperatures above 30S'C the base metal temper becomes irrelevant, and above SOO'C the properties are similar to those of solution heat treated =material. The microstructure and properties correlation will be presented. * This research is funded by the Director, Office of Energy Research, Office of =Basic Energy Science, Material Sciences Division of the U.S. Department of ::Energy under Contract No. DE-AC03-76SFOOO98. (9:30 a.m.) EFFECT OF DIFFERENT HOMOGENIZATION TREATMENTS ON THE HOT FORMABILITY OF ALUMINUM ALLOY AA2024, B. Verlinden°, P. Wouterso. H'I' McQueen * E. AernoudtO, L. Delaeyo and S. Cauwenberg+, °M~tal . Mat. Eng., University of Leuven, 3030 Leuven-Heverlee, BelgIUm, *Mech. Eng., Concordia University, Montreal, H3G IM8, Canada,+SIDAL, A. Stocletlaan 87, 2570 Duffel, Belgium Hot torsion tests have been carried out on samples of an AA2024 alloy homogenized at 460 or SOOoC for 8 or 24h and water quenched. The tests have been carried out with strain rates between 0.1 and IO/sec at temperatures between 340 and 460°C. The peak stress, the softening during torsion and the activation energy are higher but the ductility lower for the samples homogenized at SOO°C than for those treated at 460°C. The increase in peak stress and softening and the decrease in ductility with rising homogenization temperature is more pronounced at low than at high test temperature and at higher than lower strain rate. An increase of the homogenization time from 8 to 24h has almost no effect on the values of peak stress and softening but improves the ductility due to large particle solution. The effects are discussed in terms of solid solution hardening and dynamic precipitation. (9:50 a.m.) STATIC RECRYSTALLIZATION OF AI-Mg-Mn ALLOYS AFfER HOT TORSION OR ROLLING B. Crawford, J. Belling, H.J. McQueen* and A.S. Malin, Comalco Research Centre, Thomastown, Victoria, 3074, Australia, *Mech. Eng., Concordia University, Montreal H3G IM8, Canada Several Al alloys, notably 3004, S005 and 5 18 t , have been hot deformed between 2S0 and SOO°C (10 and 0.1 s- ) in both torsion and rolling. As a result of rapid cooling, the elongated grain structure was retained except after certain rolling reductIOns. Specimens were annealed at temperatures between 4S0°C and 280°C related to certain intervals in the rolling schedule. The torsion specimens were examined on both tangential and longitudinal of static sections which exhibited a regular progress
recrystallization with time. In 3004-, the growth of the new grains is limited by the rows of MnAI, particles so that they are rectangular in shape. The commercially rolled specimens showed different behaviours as a result of fixed combinations of rising reduction and diminishing temperature and of non-uniformity due to roll chilling. For example, in 3004 recrystallization of the surface layers (to 20% depth) occurred only after completion of hot rolling to 96% reduction (£=3.1). In addition, the effects of Mg and Mn additions on SRX behaviour are examined. (10: 10 a.m.) EFFECT OF HOMOGENIZATION AND PRECIPITATION TREATMENTS ON THE HOT FORMABILITY OF ALUMINUM ALLOY AA2024, P. Wouterso, B. Verlindeno; H.J. McQueen{, *E. Aernoude, L. Delaeyo, and S. Cauwenberg~,OMetall. Mat. tmg., University of Leuven, 3030 Leuven-Heverlee,-Belgium, *Mech. Eng., Concordia University, Montreal, H3G 1MB Canada,+SIDAL, A. Stocletlaan 87, 2570 Duffel, Belgium Hot torsion tests have been carried out at 310 and 370°C (1 s-I) on an AA2024 alloy homogenized 8h between 420 and 520°C and either just water quenched or air cooled and aged at 290 or 350°C. The flow curves exhibited a large peak stress (Jp and work softening with low strain to fracture, cf, in the quenched condition. The peak and work softening were reduced and the ductility improved as either the aging temperature, or the deformation temperature, T was increased. The homogenizing T had a big effect on (Jp and little on cf in deformation at 310°C, whereas at 370°C there was little effect on (Jp but a big one on cf which was much higher. The above effects arise from the relative amount of solute and dynamic precipitation after various treatments at different test temperatures. (10:30 a.m.) EFFECT OF COOLING RATE ON THE FORMATION OF 8' (Al3Li) IN ALUMINUMLITIllUM BASED AJ.l.OYS ~,
V. Radmilovic and G. Thomas
Center for Advanced Materials and the National Center for Electron Microscopy, Lawrence Berkeley Laboratory, 1 Cyclotron Road, Berleeley, CA 94720. The ordered nature of 8' (Al3Li) is known to encourage planar slip and hence shear failure in· aluminum-lithium alloys. The effect of cooling rate on the fonnation mechanisms of 8' and the degree of planar slip was investigated. Rapid quenching and air cooling from solution heat treannent temperatures were evaluated in AI-Li-Zr and AI-Li-Cu-Mg-Zr alloys. Air cooling was found to increase the total elongation to failure in tension tests due to a bimodal distribution of 8' size in a high Zr content alloy. When rapidly quenched, 8' fonns by spinodal decomposition on the ordered lattice; when slowly cooled, 8' forms by heterogeneous nucleation. The results of quenching to a 6S'C bath are also discussed. This work was sponsored by the Director, Office of Energy Research, Office of Basic Energy Sciences, Chemical Sciences Division, of the U.S. Department of Energy under Contract No. DE-AC03-76SF00098.
THERMODYNAMICS AND PHASE RELA· TIONS OF EXOTIC MATERIALS III: Theory Sponsored by the ASM-MSD Thermodynamic and Alloy Phases Committee Wednesday AM February 21, 1990
Room: Plaza C Inn at the Park
Session Chairman: P.G. Nash, Metallurgical and Materials Engineering Department, lllinois Institute of Technology,Chicago,IL 60616 (8:30 a.m.) MONTE CARLO TECHNIQUE FOR CALCULATiON OF INTERMET ALLIC COMPOUND ALLOY PHASE DIAGRAMS": Samuel M Allen Center for Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139 The thermodynamic behavior of inter metallic compound alloys is strongly influenced by critical pheI),omena associated with orderdisorder transitions. Of the various' models available for calculation of phase diagrams of IMC alloys, the Monte Carlo method gives the most realistic behavior near order-disorder transitions. Attributes and limitations of the technique will be described. An application of the technique to the calculation of equilibria in the ~-P' NiAI-Ni2AITi phases in the Ni- AI- Ti system will be discussed in detail. "This research was supported by the National Science Foundation, Grant DMR-87192 17.
Annual Meeting-l07 (9:00 a.m.) ELECTRONIC STRUCIURE AND PHASE SfABILITY OF AI5-BASED SUBSTITUTIONAL ALLOYS: * P. E. A. Turchi. Lawrence Livermore National Laboratory, L-268 , P. O. Box 808, Livermore, CA 94550 and A. Finel, ONERA, BP72, 92322 Chatillon, FRANCE After a brief overview on the methodology which has been developed to study alloy phase stability in general, configurational order in substitutional Al5-based alloys is studied by means of a 3D-generalized Ising model, as deduced from electronic structure calculations. From a ground state analysis of this model. including effective interactions up to
the 6 th neighbor pair, we obtain two new ordered states at the compositions AsB3 and A7B, besides the well know A3B state. Low temperature expansions and statistical mechanics based on the Cluster Variation Method are then applied to determine the stability of these predicted configurations as a function of temperature and concentration. Resulting prototype Al5-hased order disorder phase diagrams are discussed in relation with the experimental si tuation.
*
Work supported by U.S. Dept. of Energy, Contr. #7405-ENG-48.
(9:30 a.m.)
CALCULATION OF TERNARY AND QUATERNARY PHASE DIAGRAMS CONTAINING SEMICONDUCTOR/METALLIC COMPOUND PSEUDOBINARY EUTECTICS FOR LIQUID ENCAPSULANT CZOCHRALSKI GROWTH: * L. Kaufman, ManLabs, Inc., 21 Erie Street, Cambridge, MA 02139 and B. Ditchek, GTE Labs, Inc. 40 Sylvan Road, Waltham, MA 02254. CALPHAD analysis of a number of GaAs-Me and GaAs-MeB2 systems were performed to identify attractive eutectic systems which could be solidified to obtain regular rod-like structures for synthesis of new composite electronic materials for high power switching. The calculations which are based on explicit definition of the thermochemical properties of mUlticomponent liquids and compound phases provide a means for prediction of eutectic temperatures and compositions. Moreover, component activities, vapor pressure and compatibility with encapsulants can also be predicted.
* SPONSORED BY SDIO/IST and Managed by Office of Naval Research under Contract NOOO 14-86-C-0596 (10:00 a.m.) BREAK
(10:20 a.m.) ELECTRONIC STRUCIURE, ALLOY PHASE SfABILITY , AND PHASE DIAGRAMS: A. Gonis, M. Sluiter, P. E. A. Turchi, Chemistry and Materials Science, L-280, Lawerence Livermore National Laboratory, Livermore, CA 94550; G. M. Stocks and D. M. Nicholson, Oak Ridge National Laboratory, Metals and Ceramics Division - 4500S, Oak Ridge, TN 37831. We review the relevance of electronic structure calculations
to the determination of alloy phase stability and alloy phase diagrams. The Connolly-Williams method, the generalized perturbation method, the embedded cluster method, and the method of concentration waves are presented and their main features are discussed and compared.
TMS INSTITUTE OF METALS LECTURE Speaker Awardee: Hubert I. Aaronson, Carnegie Mellon University, Dept. of Metallurgical Engineering and Materials Science, Pittsburgh, PA 15213 Wednesday, 11:45 AM Room: Plaza D February 21, 1990 Inn at the Park
Topic: "Atomic Mechanisms of Diffusional Nucleation and Growth" (11:45 a.m.) TMS INSTITUTE OF METALS LECTURE The lecture will be based upon ideas developed from Dr. R.F. Mehl's demonstration, in a series of papers entitled "Studies upon the Widmanstatten Structure," published between 1931 and 1937, that the broad faces of precipitate plates are formed by particularly well matched conjugate habit planes in the matrix and precipitate phases. The point will first be made that purely diffusional nucleation processes can lead reproducibly to the lattice orientation relationships which make Widmanstatten morphologies possible. Confirmation of the essential correctness of nucleation theory by recent experiments on eu-Co alloys provides a firm base for this and all further considerations. Differences in the nucleation mechanics of martensitic and diffusional transformations will be shown to lead simply and directly to sharp differences in both the kinetics with which misfit dislocations are acquired and their orientations with respect to interphase boundaries. These differences will be seen to lead in turn to the drastically different shear (=martensitic) and diffusional mechanisms of growth. The unit atomic process in growth by shear is the tightly coordinated "gliding," through dislocation motion, of successive atoms across advancing partially coherent interphase boundaries. In diffusional growth, however, this process is the loosely coordinated, biased random walk, diffusional jumping of individual atoms across the risers of growth ledges, or kinks on these risers, through regions containing sufficient local atomic disorder to make such jumps energetically feasible. Both types of growth are shown to be capable of satisfying the invariant-plane-strain surface relief effect and the crystallographic requirements and consequences of the phenomenological theory of martensite crystallography. Only "derivative" requirements of this theory, including absence of changes in composition and in long-range order, availability of sufficient driving force and the presence of an appropriate interphase boundary structure are capable of distinguishing between these two different growth mechanisms and then only when all are simultaneously employed. The importance of growth ledges in diffusional transformations will be particularly emphasized through examples including the thickening of grain boundary allotriomorphs, the degeneration of the Widmanstatten ferrite morphology and the slowing down of ferrite growth through action of the solute drag-like effect, and both bainitic and pearlitic modes of eutectoid decompoSition. Support from AFOSR and NSF is gratefully acknowledged.
The results of calcu-
lations of effective cluster interactions in substitutionally disordered alloys and of phase diagrams for specific alloy systems, e.g. PdRh and AILi, are shown and work currently in progress is reviewed.
(10:50 a.m.) SYSTEMATICS, PHASE RELATIONSHIPS AND THERMODYNAMICS OF RARE EARTH ALLOYS: K A. Gschneidner Jr , Ames Laboratory·, Iowa State University, Ames, IA 50011. The regular and smooth variation of many of the physical properties of the trivalent rare earth metals allows one to make correlations concerning their alloying behavior and the thermodynamic properties of the compounds formed by these elements. The 91 possible binary phase diagrams of the trivalent intra· lanthanide and yttrium-lanthanide systems can be represented by a single generalized phase diagram. Anomalies found in the high pressure variation of the generalized phase diagram yields important information about the bonding in the rare earth metals. The systematic variation of the lattice parameters of the rare earth intermetallic compounds can be used to predict the relative variation of the free energy of formation of these compounds. ·Supported by USDOE, Office of Basic Energy Sciences, Division of Materials Sciences, under Contract No. W-7405-ENG·82.
TMS LIGHT METALS LUNCHEON SPEECH Guest Speaker: Jacques Bougie, Alcan Aluminium Limited, 1188 Sherbrooke Street West, Montreal, Quebec, Canada H3A 3G2 Wednesday, 12:00 Noon Room: Grand Ballroom F February 21, 1990 Anaheim Marriott Hotel
Topic: "Key Issues Facing the Aluminum Industry" (12:00 Noon) TMS LIGHT METALS LUNCHEON SPEECH The aluminum industry is challenged to remain competitive internationally in a world of no tariff barriers, no non-tariff barriers, no subsidies and no countervail, all within a set of rules (GATT). For aluminum to be the material of choice, it must be cost-effective against competing materials. This means keeping costs down and developing new uses, building on aluminum's strengths. Likely areas for development include composite materials, transportation (rail cars, automotive castings and automobile structures), and expansion of existing applications into new markets (e.g., cans in Japan and Europe, and consumer products in the Soviet Union and China) . Environmentally, the aluminum industry must cope with the impact of acid rain reduction measures on power costs and the impact of rising environmental standards on industry processes,
lOB-Annual Meeting from bauxite to anodizing. The industry must make a commitment to improve environmental performance at a sustainable pace. Finally, aluminum is lion the side of the angels with an impressive track record on beverage can recycling. Its intrinsic value is the economic engine that drives waste recovery systems. Aluminum is the only material with a "closed-loopl! recovery cycle (for cans). To continue the progress, the industry needs to improve recovery systems and recycling technologies, while designing products in forms that are intrinsically recyclable. Ultimately, the industry·will "lease" and not sell aluminum to its customers, wi1;.h a guaranteed take back at the end of the product life cycle. ll
(2:30 p.m.)
HIGH TEMPERATURE DEFORMATION BEHAVIORS OF SiC WHISKER REINFORCED Al-Cu-Mg MATRIX COMPOSITE . H. I. Lee and K. H. Oh, Functional Materials Laboratory, Korea Advanced Institute of Science and Technology, Post Office Box 131, Cheongryang, Seoul, Korea; and T. S. Kim and T. H. Kim, Department of Metallurgical Engineering, Seoul National University, Sinlim Dong, Kwanak Gu, Seoul 151-742, Korea. Deformation behaviors of SiC whisker reinforced Al-Cu-Mg alloy in the semi -solid region were studied using tensile testing machine, scanning electron microscopy, and transmission electron
microscopy.
WEDNESDAY PM
The materials processed by
powder metallurgy
techniques were hot extruded and rolled. Microstructures of the processed materials were investigated using scanning electron microscopy and transmission electron microscopy. Tensile tests at selected temperatures near the local melting point were conducted in the argon atmos~here at strain rates in the range
of 5.0
x
10- 3 to 5.0
x
10-l s - .
The fracture surfaces of tested
samples were analyzed using scanning electron microscopy. The materials exhibited superplastic behavior at high strain rate in the semi-solid region.
ADVANCED METAL & CERAMIC MATRIX
(2 :45 p.m.)
COMPOSITES V: Process Modeling and Deformation Processing
DYNAMIC CONSOLIDATION OF ALUMINUM-SILICON CARBIDE COMPOSITES:* B, H. Rabin, G. E. Korth, R. L. Will iamson, Idaho National Engineering Laboratory, EG&G Idaho, Inc., P.O. Box 1625, Idaho Falls, ID .83415-2218
Sponsored by the TMS Powder Metallurgy Committee, TMS Physical Metallurgy Committee, and ASM-TMS Composites Committee
Dynamic consolidation has potential as an economical method for producing P/M composites and has the ability to achieve full density using low bulk processing temperatures. In this study, full density silicon carbide particulate - 2l24 aluminum compOSites were fabricated by dynamiC consolidation of powder mixtures using explosives. The microstructure of the as-consolidated composite was dependent upon processing conditions; numerical simulations were performed which allowed the microstructural observations to be related to shock parameters. Mechanical properties of the composites were studied in the as-consolidated condition as well as after heat treatment. Microstructural characterization and mechanical property results will be presented .
Wednesday PM February 21, 1990
Room: ParkD
Inn at the Park
Session Chairmen: David Lloyd, Alcan International Limited, Jonquiere, Quebec, Canada; Steven G. Fishman, Office of Naval Research, Materials Division, 800 N. Quincy Street, Arlington, VA 22217
* Work supported by the U.S. Department of Energy under DOE Contract No. DE-AC07-76ID01570. (3 : 00 p.m.)
(2 : 00 p.m.) PROCESSING AND SUPERPLASTICITY OF MG·9L1 AND Mg-9L1-XB4C :.J. WollensJine. P. Metenier, G. Gonzalez-Doneel, O.A. Ruano, and 0.0. Shelby, Dept. of Materials Science & Engineering. Stanford University, Stanford, CA 94305. The superplastic behavior of a Mg-9wt"loLi two phase alloy, in the coarse grained as·cast condition. in fine-grained laminate form, and in fine-grained laminated composite form, where B4C particles (x= 5 to 20 wt%) have been added to the Mg-9Li laminate to form the composite, was investigated in the intermediate temperature range. The as-cast material exhibited a stress exponent, n. equal to five at all strain rates and temperatures tested. The fine-grained laminate and the fine-grained laminated composite exhibited n values close to two at all temperatures investigated. The deformation mechanism of the as·cast Mg-9Li material can be described as. dislocation creep controlled by lattice diffusion in the ~ phase. In the n equal to two region for the Mg·9U laminate. the deformation mechanism be interpreted as, grain boundary sliding accommodated by slip controlled by lattice diffusion in the ~ phase.
(2: 15 p.m.)
DEFORMATION PROCESSING OF ALUMINUM-ALUMINA METAL MATRIX COMPOSITES: Floyd R. Tuler, Worcester Polytechnic Institute, 100 Institute Rd., Worcester, MA 01609, Thomas F. Klimowicz, Dural Aluminum Composites Corp., 10326 Roselle St., San Diego, CA 92121 Recent developments in the casting of particle· reinforced aluminum alloy composites have brought the properties of these materials and their production quantities to levels where they can be considered for a wide variety of applications. Metal matrix composite materials are now available as DC cast ingots which can be shaped by conventional hot metalworking processes, such as extrusion, forging, and rolling . However, a fundamental understanding of the high temperature deformation characteristics of these materials is required to specify the parameters for fabrication processing. This paper describes the results of a study to determine the relationship among tensile properties, microstructure, temperature, and strain rate in aluminum oxide particle-reinforced 2000 series aluminum composites. Applications of the results to forging operations are also presented.
PARTICLE-REINFORCED COMPOSITE PROPERTY PREDICTION BY UNIT-CELL FINITE-ELEMENT MODELING. D. G. Thompson, Y. Fan, P. K . Liaw, R. E. Shannon and T . T. Butler, Westinghouse R&D Center, 1310 Beulah Road, Pittsburgh, P A 15235. The material properties of particle-reinforced metal matrix composites are dependent on both material and geometric properties of the reinforcement medium. The shapes and sizes as well as the spacing, distribution, and orientation are all important parameters in determining the macroscopic behavior of the composite. The use of unit cell finite element modeling in predicting the macroscopic behavior is investigated. Three-dimensional modeling is compared with two dimensional modeling and to experimental results. Insights into modeling approaches and their implications are discussed. Specific findings about the influence of geometrical parameters of reinforcements upon the material properties of siliconcarbide-reinforced metal matrix composites are presented. (3: 15 p.m.) UU'ORMATION PROCESSING AMD MODELLING OF METAL MATRIX COMPOSITES: S P Derham, F J Humphreys, W S Miller, BPRI, Chertsey Road, Sunbury, Middlesex, TWl6 7LN .
Extrusion is an established technology for the manufacture of near net shape aerospace components in aluminium base alloys.
The technology is eminently suitable for use with metal matrix composite for the same market. The objective of this paper is to examine the interaction between billet variables and processing parameters on the strength-toughness relationship of aluminium based metal matrix composites in extruded form. The billet
variables discussed will include alloy chemistry, reinforcement type, size and volume fraction . These will be discussed in terms of a single set of extrusion parameters . Conversely the effect of certain process variables (eg temperature compensated strain rate, extrusion size, shape and ratio), together with die
geometry effects will be examined for specific billet types. The paper will compare the results of physical modelling of the extrusion process with data obtained from mathematical
modelling. Particular attention will be paid to the influence of billet and process variables on the strengthening mechanisms and the fracture characteristics of these materials.
Annual Meeting-I09 (3:30 p.m.) MODEL FOR FIBER BREAKAGE DURING PIM CONSOLIDATION OF METAL·MATRIX COMPOSITES: T. C. Tszeng, E. K. Ohriner, and V. K. Sikka, Metals and Ceramics Division, Oak Ridge National Laboratory,* P.O. Box 2008, Bldg. 4508, Oak Ridge, TN 37831-6083. Fiber breakage was observed during the fabrication of unidirectional metal-matrix composites by the PIM processing. An examination of specimens revealed that the failure mechanism was the bending of fiber under the action of powder particles. This problem was divided into two subtopics in this investigation. The first dealt with the contact force between the particle and the fiber when the fiber is heavily deformed by the particle. The second is the failure of fibers due to the contact force. The present model consists of a deformable sphere and a rigid indenter. Plasticity analysis was carried out to deterntine the contact force between the sphere and indenter. The model was successfully applied to the cases of a spherical indenter in the preliminary calculations. Then, a setting of deformable sphere-rigid cylinder was investigated. The bending stresses were then calculated based on the simple beam theory and a safe criterion for preventing the fiber from breaking was found. The calculations were compared with experiments. *Operated by Martin Marietta Energy Systems, Inc., under contract DE-AC05-840R21400 with the U.S. Department of Energy.
(3 :45 p.m.) THEORETICAL MODELING OF THE CHEMICAL VAPOR INFILTRATION PROCESS FOR FORMING HIGHLY DENSIFIED CERAMIC/CERAMIC COMPOSITES: N. H. Tai and T. W. Chou, Center for Composite Materials and Department of Mechanical Engineering, University of Delaware, Newark, DE 19716 Two theoretical analyses of the chemical vapor infiltration process: the isothermal infiltration of a unidirectional fiber bundle and the infiltration of a 3-D woven fibrous preform under pressure and temperature gradients have been performed. Due to the non-uniform deposition a density gradient in the deposited matrix is often observed. A method is proposed in this paper to alleviate the deposition gradient; it considers the physical property variations of the reactants, movement of the densified front, deposition of the chemical reaction products as well as the pressure and preform orientation adjustments during processing. According to the results of these considerations, a more uniform growth of the matrix and a high density ceramic/ceramic composite can be achieved.
(4:00 p.m.) Elastoplastic Finite Element Analysis of the Effect of the 'Ihemal Processing Cycle on SiC/Al Crnposites: Alvin Levy and John M. Papazian, Grurman Corporate Research Center, A02,-26, Bethpage, NY 11714.
(Abstract Not Available)
(4:45 p.m.) A COMPUTER MODEL FOR ONE AND TWO DIMENSIONAL THERMAL PROFILES OF DROPLETS DURING SPRAY FORMING: P.S. Grant**, S. Rogers*, B. Cantor**, and L. Katgerman* A Lagrangian computational model has been developed to describe the velocity, flight time, heat transfer coefficient, temperature, cooling rate and fraction solid of alloy droplets produced by gas atomisation. Profiles are given as a function of axial distance travelled from the point of atomisation. The model was then used in conjunction with PHOENICS software, which solved the two dimensional momentum equations of gas flow subsequent to atomisation within a Eulerian grid, to give droplet trajectories and profiles under realistic flow conditions. Thermal profiles include droplet undercooling and subsequent recalescence; segregated solidification, and eutectic solidification. The model is utilised to investigate the effects of droplet size, gas velocity, melt superheat and solute concentration on thermal profiles. Individual droplet histories are integrated over a droplet size distribution probability to give the total thermal condition of the spray. Trends resulting from parameter variation are presented and the imRlications for spray-forming discussed. (5 :00 p.m.) An Experimental study of the In Situ Carbide Nucleation am GrcMt:h in Metal Matrix Crnposites: Purush Sahoo am Michael J. Koczak, Dept. of Materials Engr., Drexel University, Philadelphia, PA 19104.
(Abstract Not Available) (5: 15 p.m.) MODELING AND CHARACTERIZATION OF THE THERMOMECHANICAL BEHAVIOR OF GLASS MATRIX COMPOSITES: W.S. Kuo and T.W. Chou, Center for Composite Materials and Department of Mechanical Engineering, University of Delaware, Newark, DE 19716
'n analysis of the steady state and transient thermal effects, coupled with ."echanicalloading, on carbon reinforced borosilicate glass is presented. The thermal deformation and stresses in the fiber, matrix and interphase are studied by assuming that the carbon fiber is both elastically and thermally anisotropic. A model of multi-layer concentric cylinder is applied in this study. It has been found that the onset of microcracking and the thermal shock resistance, which are governed mainly by the stresses in the matrix. are sensitive to the anisotropy of the fiber. The effects of fiber orientation and volume content are also taken into account.
(4: 15 p.m.) INFLUENCE OF PROCESSING PARAMETERS ON THE MECHANICAL PROPERTIES OF P/M ALUMINIUM MATRIX COMPOSITES: Antero Jokinen, Technical Research Centre of Finland, Metallurgy Laboratory, Metallimiehenkuja 2, SF-02150 Espoo, FINLAND
ALUMINA & BAUXITE V: Alternate or
The processing variables affect considerably the mechanical properties of PIM . aluminium based composites. A proper combination of mixing, degassing and consolidfrtion is needed. Especially, the temperatures during consolidation are very important.
Sponsored by TMS Aluminum Committee
In the Technical Research Centre of Finland the hot extrusion of silicon carbide particulate reinforced aluminium PIM alloys corresponding to types AA2014, AA6061 and AA 7090 is studied. The influence of mixing, degassing and hot 'pressing variables on the properties of the extruded product have been investigated using microstructural analysis and tensile and wear testing. The results indicate that mechanical alloying leads to extremely homogeneous microstructures and uniform distribution of SiC-particulates compared to those obtained using conventional blending. The fabrication can be successfully carried out without encapsulation and vacuum degassing by cold isostatic pressing of the powder before heating it in a protective atmosphere. Hot pressing at a high temperature before extrusion improves the strength of composites. Obviously this is based on a stronger bonding between the matrix alloy and the particulates.
(4:30 p.m.) THIXOTROPIC EFFECTS IN SEMI -SOLID AI-67. Si COMPOSITE ALLOYS REINFORCED WITH SiC PARTICLES. M. Mada and F. Ajersch, Ecole Poly technique, Dep. de Genie Metallurgique, C. P. 6079, Succ. A., Montreal, Quebec, H3C 3A7 Canada. Rheocast semi-solid composite alloys exhibit thixotropic behavior when deformed at varying shear rates due to the changes in interparticle and inter-aggregate bonding. The degradation of the bonds reduces the viscosity to a dynamic equilibrium value for each specific shear rate where the destruction and reconstruction
of particle-agglomerate bonds reach a steady state. For AI-6% Si alloy reinforced with SiC particles, our experimental results indicate that for a constant concentration of added particles and constant composition of primary Al phase, the dynamic equilibrium state is obtained more rapidly with increasing shear rate. For a given shear rate, the kinetics of the degradation of agglomerates is nearly independent of the concentration of the added reinforcing particles, especially at very high shear rates. Also, the kinetics of the structural degradation of agglomerates is highly dependent on the concentration of the primary Al phase particles.
Modified Processes
Wednesday PM February 21, 1990
Room: Salon 3 Marriott
Session Chainnan: B.A. Hiscox, Alcan International Ltd., Montreal, Quebec, Canada H3A 302
(2 :00 p.m.) PHYSICOCHEMICAL BASIS OF SIITER-DISIITEGRATIOI COMPLEX METHOD OF ALUMIIIUM OXIDE AID CEMEIT PRODUCT 101.
J.Grzymek, A.Derdacka - Grzyaek. Acadeay of Mining and Metallurgy, Krakow, Poland.
In Ihe cOlplex \echnology of IIU1inl Ind porllind ulenl produclion by J.6rzYlek's Itlhod, Iht self-disinlegrllion Ilulopuiverililioni of sinltrs and Itlll tonhining cilciul orlhosiJicllt and pyro·ulciul IIUlinilt occurs. This disinlegnlion is I conuquenct of I -) 1 ulciul orthosiliult Irlnsition. Tile lIin difficully inhibiting Ihe Itlf-disinltgrllion prOCHS is Ihe slibilililion of Iht • polYlOrph of CbSiO., Tht Iheortliul lOllS rnoive sOle probltls conctrning lhis prottSs. Tile folloYing lapin lit of significanl ilporhm : Iht mhlnin of shbilizing ution of SOle ions influtncing Ihe I-ll CbSiO. Innsilion Ind tJilinllion'lIf Ihis shbilizing efftcl. 811ing on txptriltnhl rHul1l a fOllula hIS bltn proposed 10 Hlilile quanlillvely llIe shbiJizing efftct of sOle cllions .hich lit ablt 10 rthrd llIe disinltgnlion. The forlula cOlpriul such proptrlitl of cllions IS tltclronegatiYily, Yllency, ionic radius, Hln quanluallUlbtl, configurltion of Ylltncy tltelrons. The inttfplltalion of llIe propo5ld fonuh Iuds 10 llIe conclusion Ihat llIe IlllIiliulion of lilt I-phalt un bt 1V0ided .hen Ihe ulciua orlhosiliult is synlhHiztd in rtducing Itlosphtrts. Tht proms of reduction decrtlllS llIe ltabilizing Ifflct txtlltd by fortign calions. The rllt of disinltgntion 1111 btln found 10 bt deltllintd by Iht nit of nyc Itllion , Enhlpy Ind tnlropy changts hlYt bttn Ituurtd for CI2SiO. al huling and cooling.
nO-Annual Meeting (2:30 p.m.) DISCUSSION ON HEAT CONSUMPTION IN THE MANUFACTURE OF ALUMINA BY LIME-SODA SINTERING PROCESS: Ding-Ou Yan and Heng-Long Li, Shenyang Alum. & Magnesium Engr. & Research lnst., 50 Block 3, Heping Street, Shenyang, Liaoning, PRe. There exists abundant high silica bauxite in China, being some billion tones approved. This type of ore has been processed economically with high recovery, low soda consumption and high quality of product when an improved lime-soda sintering process developed by Chinese engineers since mid 50's is applied. The main demerit of this process is the extremely high consumption at 3 even 4 times as that of Bayer process. It is however a necessary compromise to the utilization of high silica bauxite, and also it can be reduced if some modification of the technology is introduced. The heat for the chemical reaction, the theoritical heat consumption and the lowest heat consumption which can be reached after effective improvements have been performed are calculated by authors. As a conclusion, the current operating cost for high silica bauxite by lime-soda sintering process can further be reduced.
CARBON TECHNOLOGY VI Sponsored by TMS Aluminum Committee (formerly Light Metals) Wednesday PM February 21, 1990
Room: Salon 2 Marriott
Session Chairmen: Euel R. Cutshall, Manufacturing Technology Laboratory, Reynolds Metals Company Sheffield, AL 35660; David L. Belitskus, Aluminum Company of America, Alcoa Center, PA 15069 (2:00 p.m.)
(3:00 p.m.) FACTORS EFFECTING ON THE PROPERTIES OF BAYERITE OBTAINED IN THE CARBONIZATION PROCESS OF ALKALINE ALUMINATE SOLUTIONS: M.VOjcik and M.Pyzalski, Interbranch Institute of Building and Refractory Materials, Academy of Mining and Metallurgy, 30-059 Cracow,Av.Mickiewicza 30,A-3,Poland The carbonization process is the one of the method for aluminium hydroxide production from alkaline solutions beside of the widely known Bayer's method. The hydrargillite is the main form of AI(OH)3 which is manufactured by Bayer'S method. The obtainment of the other form of AI
(3:30 p.m.)
ALUMINIUM OXIDES OF THE ELEVATED PURITY PRODUCED FROM THE ALKALINE METHOD:M.Pyzalski,M.Vojcik, Interbranch Institute of Building and Refractory Materials, Academy of Mining and Metallurgy, 30-059 Cracow,Av.Mickievicza 30,A-3,Poland The approachable professional references do not re,'eal the fact that the aluminium OXides containing less than 0.1% of Na.,o+K 2 0 can be produced by the direct manner from anyone of alkaline methods. Higher purity can be received only on the way of usage of an additional and various methods for purification of AI(OHh or AbO •., which increase the production costs. Ours experiments show that it is possible to obtain by the direct manner the aluminium oxides containing less than 0.05% of alkalies from sodium or potasium aluminates solutions. The XRD and SEM technics and chemical and granulometrlc analysis were app11cated for description of the physico-chemical properties of manufactured aluminium oxides, Also, the comparison between the properties of an american, german and examinated aluminas were carried out and are presented in this paper.
(4:00 p.m.)
THE PROBLEM OF THE FORMATION OF THE SOLID REACTION PRODUCTS IN THE ALUMINA LIXIVIATION PROCESS FROM CALCIUM ALUMINA-SILICATES: M.VOjcik,Academy of Mining and Metallurgy 30-059 Cracow,Av.Mickiewicza 30,A-3,Poland
It often happens in the chemical processes that originated the reaction products in the form of solid phase and do not removed from the reacting system have the delayed and harmful effect on the chief chemical reaction and thus they have an influence on the efficiency of the chemical process and on it economy. The problem of the formation of a calcium carbonate in the process of aluminium lixiviation from selfdisintegrated sinters containing calcium alumina-silicates and it effect on the efficiency of the lixiviation process was shown in this paper. Owing to the application of ultrasound in the lixiviation process,the removing of the solid reaction products from the surface of the particle of selfdisintegrated sinters and in the consequence the increase of the recovery of aluminium and an essential reduction of the lixiviation time were ascertained. The XRD,SEM,DTA technics and granulametric and chemical analysis were used for an examination of the selfdisintegrated sinters and post-lixiviation slimes.
(4:30 p.m.) TECHNICAL RESULTS AND FURTHER DEVELOPMENTS IN PRODUCING ALUMINIUMOXIDE AND ALUMINIUM COMPOUNDS BY HYDROCHLORIC ACID-LEACHING OF CLAY KAOLIN AND OTHER ALUMOSILICATE MINERALS: S. Ziegenbalg and G.X. Haake, GDR, Research Institute of Non-Ferrous Metals, GDR-9200 Freiverg (Sacks), Lessingstrasze 41, 9200 Freiberg, West Germany. (Abstract Not Available)
BINDER FOR TIlE IDEAL ANODE CARBON, Sam Jones, Industrial Carbon Consultant, P. o. Box 43698, Tucson, Arizona 85733; Ted Bart, Engineered Materials Sector, Allied - Signal, Inc., P. o. Box l053R, Morristown, New Jersey 07960, USA Experiments with many different kinds of binder·filler combinations for anode paste formulation bave sbown that the ideal binder, wbich can be made from either coal tar or petroleum, should produce a rather isotropic coke to best achieve the most desirable properties of anode carbon. The normal production process for coal·tar pitch results in sucb a binder, wbere the binder coke is disorganized by primary quinoline·insolubles formed in the vapor space above the coal in the coking ovens. However, for both coal·tar and petroleum pitches, the process can be improved by eitber changing the thermal treatment variables, or addition of selected artificial quinoline·insolubles. (2:30 p.m.)
IMPROVEMENTS TO ANODE BINDER PITCHES BY ADVANCES IN RAW MATERIAL'l HANDLING, MANUFACTURING, APPLICATION AND TRANSPORT,
N. R. Turner, Bitmac Umited, Dawes Lane, Scunthorpe, South Humberside ON 15 6UR England. Coal tar pitch technology is developing rapidly. Increased co·operation with cokeries on tar quality and increased knowledge on tar handling are giving big rewards. Changes in tar plant design have revolutionized the market with pitches distilled under high vacuum. This new generation of pitches has superb coke wetting/mixing performance, potentially better carbonization characteristics, lower volatiles emissions, all contributing to higher quality Soderberg and prebaked anodes. Enmples will be given and discussed. Anode makers are using these pitches with their new binding characteristics in development programmes on anodes and are reaping the rewards in the cell rooms with improved performance. The properties of vacuum distilled pitches change only very slowly in storage tanks. This makes them ideal for long-distance transport by sea in tankers. The new generation of pitches is allowing the replacement of solid pitch shipping, which is becoming increasingly unacceptable environmentally, with 'clean' liquid pitch cargoes. (3,00 p.m.)
DELIVERY OF PITCH IN SOLID PROCESSED FORM, J. L. Saint Romain, J. C. Decroix, Norsolor, H.O.D. Department, BP 6·62880 Vendin Ie Vieil, France; Oalib Abrahao Chaim, Jorge Henrique Franca dos Santos, ALBRAS, Estrada PA 483 Km 21, Barcarena . Pa., CEP 68447, Brasil. This study intends to understand the problems occurring during solid pitch processing and utilization. A fundamental approach, considering pitches as glassy materials, has been used to study their behaviour at temperatures below their softening points. Based on the results of these investigations proper conditions of granulation, storage and transportation can be proposed. Practical examples of industrial realizations are presented. (3: 30 p.m.)
HOW THE COAL CHEMICALS DMSION OF ARISTECH CHEMICAL CORPORATION IS IMPLEMENTING A SUCCESSFUL TOTAL QUALI1Y EFFORT, Donald M. Ferguson, Nathan H. Weinstein, Aristech Chemical Corporation, 300 North State Street, Clairton, Pennsylvania 15025, USA Top managers at Aristech Chemical Corporation decided in 1986 to develop an aggressive quality program using statistical process control. In early 1987 Aristech initiated a company·wide quality program. "Aristech Total Performance" or ATP was the name given to this statistically based total quality activity. The Coal Chemicals Division and the Tarben Plant have instituted many changes to incorporate ATP into our daily business life. All levels of employees have received training on statistical process control and are involved in quality improvement. Statistical Process Control is just part of a major cultural change taking place in Aristech, the Coal Chemicals Division and the Tarben Plant. This paper deals with how the Coal Chemicals Division has developed its SPC Program, removed obstacles to implementation and improved processes.
Annual Meeting-111 (4:00 p.m.)
POSSIBILITIES OF PRODUCING ANODE PASTE BINDER FROM INDIGENOUS BASIC MATERIAL, Istvan Varga, Zsuzsanna Palovits, ALUTERV·FKI, Hungalu Engineering and Development Centre, Fehervari ut 144, H·1389 Budapest P. O. B. 128,
Hungary.
94720.
From Hungarian coals derived coal tar pitch does not correspond as binder pitch of anode paste, without some further treatment. Our publication contains the experiments, that we made for testing the 'Iuality of coal tar pitch, the methods for increasing the content of {Jresin, and the detail properties of obtained binder pitch after heat-treatment under pressure.
CAST SHOP TECHNOLOGY VI: CastingAutomation and Control Sponsored by the TMS Aluminum Committee Wednesday PM February 21, 1990
Room: Salon 4 & 5 Marriott
(2:00 p.m.) THE COMPUTERIZED CASTHOUSE J. F. Jordan, Swiss Aluminium Ltd R&D CH-3965
Chi~is
In the last few years the global aluminum market has undergone considerable changes causing traditional producers to improve their performance. In this environment, electronics and information technology have found fruitful ground. Today where we have to produce just in time, where above all flexibility in using available production tools is essential, where quality is an absolute must, it is no longer possible to meet these requirements without the support of information technology. The integration of all the sectors of production in the casthouse in a ClM (computer integrated manufacturing) concept is no longer just a futuristic idea but is a working tool of today used to optimize casthouse operations. This paper describes the processing of an order in a factory. Each stage from the confirmation of the order until shipment of the product via the planning of the different phases of production is described and commented on. It will also give a general view of the concept applied to improve the competitiveness of the plants in today's market.
TIlE IMPORTANCE OF MICROBIOLOGICAL CONTROL IN CASTING: C R. Smyrniotis, C. A. Wiatr and 1. C. Flaugher, Na1co Chemical Company, One Na1co Center, Naperville, IL 60566-1024. Control of microbiological fouling and its related corrosion problems is a critical element in providing high quality/consistent casting cooling. In. this pa~er the authors discuss general microbiological concerns related to cast10g coohng water, development of advanced technology, microbial control ~nd som.e general . experiences of in-field use. A discussion of the role casnng lubncants play 10 microbiological activity is included.
DEFORMATION & MECHANICAL BEHAVIOR I TMS General Abstract Session Wednesday PM February 21, 1990
Room: Park C Inn at the Park
Session Chairman: W.E. Dowling, Jr., Research Staff, Ford Motor Company, Dearborn, MI 48121 (2:00 p.m.)
(2:25 p.rn.) AUTOMATION FOR THE CAST HOUSE. Goodrich~
A LONG STEP FORWARD.
S. Takeda, Kaiser Aluminum-Center for
Technol~,
P.O. Box 877, Pleasanton, CA 94566; D. W. Pattee, R. V. Twinillg, Kaiser Aluminum-Trentwood Works, P.O. Box 15108, Spokane, WA 99215 Kaiser developed a highly dependable station automation system
as a requirement for the Electromagnetic Casting process. This technology has since been applied to two conventional casting stations. The system has been in continuous operation since 1980 in the Trentwood Plant with complete acceptance by crews and management alike. Presently, five of the eight casting stations in their cast house use this control system. The system is unique in it's patented control strategy as well as the underlyin5 philosophy that the operator must be able to intervene safely at any point in the operation. Justification for conventional casting has been in better recovery as well as the elimination of a '·learning curve': when changing alloys and sizes.
NUMERICAL MODELING OF THE EFFECTS OF STACKING FAULT ENERGY ON DEFORMATION OF FCC METALS: A. M. Tabash and J. C. Gibeling, Division of Materials Science and Engineering, University of California, Davis, CA 95616. A composite model for inelastic deformation of FCC metals is extended to include effects due to variations in stacking fault energy. The model incorporates two structure parameters, represented as microstructural strengths, associated with the dislocation densities in cell walls and cell interiors. The hard cell walls are assumed to be composed of edge dislocations, while screw dislocations are envisioned to move within the cell i nteri ors. The i nfl uence of stacki ng fault energy is made explicit through expressions for the kinetics of dislocation glide, edge dislocation climb and screw dislocation cross-slip. This treatment leads to natural variations in the structure parameters and in the calculated deformation behavior at high and low temperatures. These predi ct ions are compared with experimental data for aluminum, copper and nickel alloys. (2:20 p.m.)
(2:50 p.m.) UNDERSTANDING TIIE STARTING PHASE: A KEY FACfOR TO TIIE SUCCESS OF EMC (ELECI'ROMAGNETIC CASTING): Y. Krahenbiihl, R. Von Kaenel, B. Carrupt, J -C. Weber, Swiss Aluminium Ltd. R&D CH-3965, Chippis. Today more than one million tons/year of rolling slabs and extrusion billets in soft and hard alloys are produced with the proven Alusl!isse EMC techn.ology: Since EMC has also been applied to cast hard alloys, the Importance of thIS cas~g technique has considerably increased. The most critical phase of any c~st1Og process is the starting phase, and that applies for EMC even more than 10 . conventional casting, since additional parameters have to be controlled. This paper provides a basic theoretical understanding of the starting phase of a cast and confirms results by measurement on plant size installations. It discusses the important casting parameters and how to control them. (3:15 p.m.)
Mathematical and physical models of electromagnetic casting of aluminum have been constructed, with emphasis on electromagnetic fields, meniscus sh~pes and electromagnetically driven flow. The mathemat~cal models calculate all three as functions of parameters under control of the designer/operator, such as frequency, current and geometry. The physical m,?del was constructed principally to test the mathemat~cal model. It comprises a pool of Wood's alloy surrounded by inductors of various geometries (some with screen), powered by a 3 kHz power supply at various currents. Magnetic and electric fields, meniscus shape and alloy velocities are measured by various probes positioned by microcomputer. Results of the past year show reasonable fit between predictions and measurements. (3:55 p.m.)
Session Chairman: R. Figueres, Casthouse Technology, Pechiney, Voreppe, 38340 France
D. G.
(3:30 p.m.) FURTHER RESULTS FROM PHYSICAL AND MATHEMATICAL MODELS OF ELECTROMAGNETIC CASTERS: D. P. Cook, B. Q. Li and J. W. Evans, Dept. of Materials Science and Mineral Engineering, University of california, Berkeley CA
BREAK
EFFECT OF GRAIN SIZE ON YIELD STRENGTH AND FRACTURE MODE OF B2 NiAI AND FeAl. P. Nagpal, F. Liu and I. Baker., Thayer School of Engineering, Dartmouth College, Hanover, NH 03755. Rods of five FeAl and NiAI alloys whose compositions span the B2 phase fields (34-51 at. % Al in FeAI and 42-55 at. % in NiAI) were produced by multiple hot extrusions of ingots. Compression tests are being performed to determine the yield strength as a function of grain size and the result plotted in a Hall-Petch type relation. Preliminary results show that 10 NIAl, the lattice resistance increases and the Hall-Petch slope decreases with decreasing aluminum content. By comparison, forFeAl the lattice resistance increases with increasing aluminum content whlie the effect on the HallPetch slope is as yet unclear. The effect of grain size and composition on the fracture mode will also be discussed. The authors acknowledge the support of the U. S. Department of Energy, Office of Basic Energy Sciences Grant No. DE-FG02-87ER4531I.
112-Annual Meeting (2:40
p.m.)
HIGH TEMPERATURE GRAIN BOUNDARY STRESS RELAXATION.
F. Cosandey,
Dept. of Mechanics & Materials Sci., Rutgers University, Piscataway, NJ 08855-0909; B. Cao, R. Schaller and W. Benoit, Institut de Genie Atomique, EPFL CH-I015 Lausanne, Switzerland.
The high temperature anelastic response to a small cyclic stress
has been measured in 8i-20 Cr alloys containig trace additions of Ce. These measurements have been made as a function of temperature in an inverted torsional pendulum. Tvo anelastic relaxation phenomena have been observed at temperatures of 680°C and 815°C. The low temperature relaxation phenomenon, which was also observed for a single crystal, has been attributed to dislocation motioo; while grain boundary sliding is responsible for the high temperature one. Stress concentration and stress relaxation mechanisms associated with grain boundary sliding will be presented. The effect of Ce segregation on relaxation strength and related activation energy will be discussed. This work is supported by the Materials Division of NSF.
0:00 p.m.)
DEFORMATION STRUCTURES IN SUPER-0<2 TITANIUM ALUMINIDES:· L H Edelson, K. T. Venkateswara Rao, and R. O. Ritchie, Department of Materials Science and Mineral Engineering, University of California, Berkeley, CA 94720. Deformation structures and fracture and fatigue surfaces in super-"'2 titanium aluminides have been extensively. characterized using conventional TEM, SEM and optical microscopy. Dislocation loop artifacts from jet polishing have been identified which may result from shear wave instabilities in the cubic phase. TEM and slip line analysis suggest deformation proceeds by planar slip of t< Ill> Burgers vectors screw dislocations in the cubic B2 phase and that both basal and prismatic slip is active in the hexagonal "'2 phase. SEM analysis of fatigue and fracture surfaces has revealed cyclic fracturing and preferential secondary cracking in the discontinuous hexagonal phases depending on the stress-intensity range. Preliminary results of in situ TEM straining will be presented. ·Sponsored by the Air Force Office of Scientific Research.
(3:20 p.m.) EFFECT OF IRIDIUM ON HIGH-TEMPERATURE MECHANICAL PROPERTIES OF ARC-MELTED TUNGSTEN: A. Luo, D. L. Jacobson, and K. S. Shin, Dept. of Chemical, Bio and Materials Engineering, Arizona State University, Tempe, AZ 85287 Tungsten-base alloys have exceptional mechanical strength at ultra-high temperatures and are most suitable for applications such as future high-power space systems. In the present study, the effects of iridium on the room temperature hardness and the high-temperature mechanical properties of arc-~elted tungsten were examined. It was found that a solid solution softening occurred with a small addition of iridium in tungsten. The maximum softening at 298 K was observed at an iridium concentration of 0.4 wt.%. The results obtained from the tensile tests on dilute W-Ir alloys in the temperature range of 1600 to 2600 K showed that iridium had a moderate strengthening effect up to 2000 K. The effects of iridium on high-temperature deformation and fracture behavior of arc-melted tungsten were also examined with transmission and scanning electron microscopy. In the 1ight of these experimental results, the effect of iridium on high-temperature deformation mechanisms of tungsten will be discussed. (3:40 p.m.) ACTIVATION ENERGY FOR SU8GRAIN GROWTH IN Al-0.6Fe ALLOY DURING RECOVERY: Ming-Wei Tseng and S.K. Varma, Department of Metallurgical Engineering, The University of Texas at El Paso, El Paso, TX 79968 Wires of Al-0.6Fe alloy, with a true wire drawing strain of 1.7, have been annealed in a temperature range from 150 to 300°C to study the kinetics of subgrain growth during recovery. A linear relationship between the square of subgrain size and annealing time has been established in this temperature range according to Sandstrom's Model. However, the variation of the slopes of these lines with temperature do not follow the concept of a constant value for the activation energy involved for diffusion in this model. It will be shown that an assumption of a linear relationship between activation energy and temperature can lead to the minimization of the differences between the theoretically calculated and experimentally observed values of these slopes. (4:00
p.m.)
STRENGTHENING MECHANISMS IN THE AGED 2024 AND 7075 ALUMINUM ALLOYS: K. Janghorban and S. Ahmad Jenabali Jahromi, Mats. Sci. & Engr. Dept., School of Engr., Shiraz University, Shiraz, Iran.
Aluminum 2024 and 7075 alloys which are widely used in aerospace and marine applications were chosen to investigate their strengthening mechanisms. Using differential thermal analysis (DTA), metallography and tension test, the best solutionizing conditions were determined to 500 ± 5°C and 2 hours for 2024 and 480 ± 5°C
and one hour for 7075 alloy. Aging was performed in the range of 100 to 200°C for different times. It was concluded that the maximum strength in 2024 was developed at l80-190 o C after 10 hrs and in 7075 at 120°C for 28 hrs. In order to determine the strengthening mechanisms in these alloys qualitatively, the T8 treatment (cold work plus aging) was carried out. The results show an increase in the mechanical strength of 2024 whereas the 7075 alloy was not affected appreciably by cold working. It can be concluded that the strengthening mechanism in 2024 is due to the stress field around the precipitates, whereas in 7075 it is due to chemical strengthening.
(4:20
p.m.)
CONSTRUCTION OF DEFORMATION - MECHANISM MAPS FOR METALS AND SOLID SOLUTION ALLOYS, A NEW APPROACH: K. Janghorban & Shahrzad Esmaeili, Dept. Mats. Sci. & Engr. , School of Engr., Shiraz Univ., Shiraz, Islamic Republic of Iran.
A new approach has been introduced to construct deformationmechanism maps for metals and solid solution alloys. For this purpose two computer programs named DEFOR and GRAPH have been written in BASIC to be run on IBM compatible personal computers. Using these programs the maps for a number of pure metals and a solid solution alloy have been plotted and effects of grain size and alloying element have been studied.
(4:40 p.m.) COMPARISON OF TENSILE AND TORSIONAL PROPERTIES OF PSEUDOELASTIC
NiTI OVER THE TEMPERATURE RANGE· 196°C to 200°C: P.H. Adler, Raychem Corp., 800 Constitution Drive, Menlo Park, CA 94025. NI-49.2aloTI undergoes a B.-+8,.' (monoclinic) stress-aSSisted martensltic transformation with an accompanying overall shape change of ~0.13. Samples cold worked 30% and annealed at 375°C were found to exhibit pseudoelastic behavior i.e., recoverable strain, In both tension and torsion over the temperature range -20°C to 150°C, with a maximum measured recoverable strain of =0.08 occurring at T.50°C for both stress states. Spontaneous athermal transformation of the B structure occurs at M.=-20°C and the resulting low·temperature flow curves sAOw "rubber-like" behavior due to mobile Intervarlant, or twin, boundaries. AtT>150C, flow Is controlled by more traditional deformation processes In the austenite. These three flow regimes are best shown by plotting yield stress, 0" , vs. test temperature T. For either T<·20°C or T>150°C i.e., outside of the'transformation range, a conventional negative dO" !dT Is found, while for -20°C
ELECTRONIC & MAGNETIC MATERIALS TMS General Abstract Session Wednesday PM February 21, 1990
Room: Conference Theatre Quality Hotel
Session Chairman: John E, Sanchez, C.A.M., Building 66, Lawrence Livermore Laboratories, Berkeley, CA 94720 (2:00 p.m.) DEVELOPMENT OF ELECTRICAL STEEL PLATE WITH EXCELLENT MAGNETIC PROPERTIES Y. Tomita, T. Kumagai, R. Yamaba and Y. Tuda, Nagoya R&D Lab., Nippon Steel Corp., Tokai, Aichi, 476, Japan. Electrical steel plates with excellent magnetic properties are demanded for particle accelerator magnets, magnetic shield material for medical use and maglev vehicle magnets. In these uses, high magnetic flux density at low magnetizing force is required as magnetic property. To meet the requirement, (1) impurities must be decreased, (2) grain size must be coarsened, and (3) crystallite orientation must be given in easy magnetization. Impurities such as carbon, manganese and nitrogen must be as low as possible. To coarsen grain size, high temperature reheating of slabs, high temperature rolling and low reduction ratio for thick steel plates and low temperature reheating of slabs, low temperature rolling and high temperature annealing for thin steel plates are the best manufacturing procedures. To give crystallite orientation in easy magnetization, grain coarsening and texture introduction are necessary. The metallurgical factors in these manufacturing procedures to improve the magnetic properties were investigated from the standpoints of magnetic domains and domain walls.
Annual Meeting-113 (2:20 p.m.)
THE THERMIONIC EMISSION OF CHEMICALLY VAPOR DEPOSITED RHENIUM: Ralph N. Wall, David R. Bosch, Dean L. Jacobson, Arizona State University, Chemical, Bio, &Materials Engineering, Tempe, AZ
85287-6006
A renewed interest in developing thermionic energy converters for space applications has increased interest in studies of candidate emitter materials. An important parameter in evaluating these high temperature materials is the work function. The present effort examined the work function behavior of chemically vapor deposited rhenium. Duplex structures of rhenium on molybdenum; rhenium on tungsten; and rhenium on an alloy of tungsten and rhenium were considered. The work fU9ction was obtained from high temperature electron emission measurements with a guard-ringed vacuum emission vehicle designed to use a Schottky technique. Data were acquired over time and as a function af temperature. Residual gases were monitored by a quadruple mass spectrometer to determine the extent of their effect on the work function. (2:40 p.m.)
DEPENDENCE OF BORON IMPLANT CHANNELING IN SILICON ON TILT AND ROTATION ANGLES: C. Park, K. Klein, A. Tasch, Microelectronics Research Center, University of Texas at Austin, Austin, TX 78712, R. Simonton, Eaton Corp., Austin, TX 78758 and S. Novak, Charles Evans and Assoc., Redwood City, CA 94063 For the first time, a very detailed characterization of boron implant proftles has been performed for a large range of both tilt and rotation incident implant angles in order to understand the effects of incident angle on the channeling of boron ions. An extensive, high-sensitivity SIMS analysis has been performed for energies of 15, 35, and 80 keY, tilt angles of 0-10°, and rotation angles of 0-45°. For all energies, the amount of channeling varies more with tilt angle than with rotation angle, and the dependence on tilt angle is dominant at low tilt angles, with the effect of rotation evident only at larger tilt angles. This result differs substantially from previously reported theoretical considerations, which predict a substantially larger dependence of channeling on rotation angle, and it can be interpreted in terms of reduced axial and planar critical angles. With these results optimum angles for minimum channeling and/or minimal profile variation across an implanted silicon wafer can be identifted. 0:00 p.m.)
EFFECT OF TEMPERATURE AND OXYGEN PRESSURE ON TIIE CREEP BEllAVIOR OF YBa2Cu307-0 Kramer M J , Chumbley, L.S., and McCallum, R.W.; Metallurgy and Ceramics Division, Ames Laboratory, Ames, IA 50011 Recent work has shown that flux creep is a barrier to application of high temperature superconductors (HfSC) where high magnetic fields are present In conventional superconductors, cold working of the Nb3Sn wires introduces line defects which effectively pin the flux lines. Mechanical creep experiments have been carried out on polycrystalline Y 123 at various temperatures and oxygen pressures to elucidate the mechanisms of deformation at 0.8 < Trrm < 1.0 and to detemine the effects the introduced defects have on the superconducting properties. TEM work to date on deformed material shows the presence of dislocations in the (001) planes with line vectors of [100] and [110]. The twin 'pacing is narrower for the deformed material compared to the as sintered material suggesting that strain is also accomodated by deformation twinning. 0:20 p.m.)
IN-SITU DEPOSITION OF THIN FILMS OF HIGH TEMPERATURE SUPERCONDUCTORS AND RELATED MATERIALS ON SILICON SUBSTRATE BY MOCVD: R. Singh, S. Sinha, N.J. Hsu, A. Kumar, R.P.S. Thakur, P. Chou, University of Oklahoma, Norman, OK 73019 and J. Narayan, North Carolina State University, Raleigh, NC 27695-7916. Both thermal and rapid isothermal processing assisted MOCVD techniques have been used for the first time for the in-situ deposition of superconducting thin films of Y-Ba-Cu-O and buffer layers of BaF2 on silicon substrate. These films have onset temperature of 9OK, and zero resistance at 73K. X-ray diffraction shows that the films are mostly c-axis oriented. Details of the deposition technique, properties of the films and potential device applications will be presented. This work was supported by DARPA contract No. MDA-972-88-D0004. 0:40 p.m.)
LASER INDUCED MICROCRYSTALLIZATION OF AMORPHOUS Ge-SeTe: P. Sikka, Department of Science and Technology, Technology Bhawan, New Delhi-l1oo16, India Ge-Se-Te is a widely known infrared window material. The bulk s~mples. of Ge-Se-Te were grown by rapid quenching technique. The x-ray dIffraction
studies revealed that the as· grown samples were amorphous. In order to understand tha laser induced changes, the samples were irradiated with Qswitched Nd:Y AG laser pules of various energy densities (2-50 rnJ/cm2). The structural changes were studied by x-ray diffraction powder pattern and back reflection Laue pattern techniques. These studies revealed that the low energy pulses (~ IOmJ/cm2) introduced damage whereas high energy pulse ~ IOrnJ/cm2) irradiation resulted in micro crystallization of the samples. However it has been observed that the recrystallization is a surface effect. The depth of recrystallization was estimated by etching the samples and studying the x-ray diffraction patterns. The depth of recrystallization was found to be a function of energy density of the laser pulses. Infrared transmission measurements were made in the 2.5-50 ~m range using PoE infrared spectrophotometer. Electrical properties such as d.c. conductivity, magnetoresistance and thermopower were studied in the temperature range 77-300K. The changes in the electrical and optical properties were analyzed in the light of existing theories. (4:00 p.m.)
INFLUENCE OF DOPANTS ON THE ELECTRICAL TRANSPORT PROPERTIES OF Pbo.SSnO.2Te THIN FILMS; C. Jagadish, Physics Department, Queen's University, Kingston, Canada, A.L. Dawar, Defence Science Center, Delhi 110054, ~ Dept. Science and Technology, New Delhi 11 00 16, India Thin films of Pbo.sSno.2Te were grown by flash evaporating the stoichiometric polycrystalline powder and also the powder containing different concentrations (0.3 at.%, I at.%, 1.5 at.%, and 2 at.%) of Sb, Bi, T1 and In. X-ray diffraction and transmission electron microscope studies revealed that the films are preferentially oriented in the (200) direction of the cubic phase. Films doped with Bi and In were found to be n-type whereas undoped, Sb and T1 doped mms were p-type. D.C. conductivity and Hall coefficient studies were made on these films in the temperature range 77-400K. The results of p-type films were analyzed in the light of double valence band model and band parameters such as population ratio, effective mass ratio, valence band separation etc. have been calculated. Mobility-temperature data have been analyzed to identify the predominant scattering mechanisms. (4:20 p.m.)
TIIERMAL FATIGUE OF EXPERIMENTAL NEAR EUTECTIC TERNARY Pb-Sn-X SOLDER JOINTS, D Tribula. T.S.E. Summers, J.W. Morris, Jr., Center for Advanced Materials and the Department of Materials Science, University of California, Berkeley, CA 94720. New packaging technologies within the electronics industry constitute severe operating conditions for solder joints. Speciftcally, the thermal fatigue of surface mounted devices raises serious concerns about solder joint reliability in this application. Recent fundamental research identifies the inherent instability of the as-cast solder microstructure and its tendency toward inhomogeneous shear deformation as contributing factors that initiate the fatigue failure. In this work the thermal fatigue resistance of 58Sn-40Pb-2X solders is investigated and compared with the simple binary alloy. The alloys s~udied include ternary additions of Bi, Cd, and In. The effect of the addition on the as-cast microstructure is discussed. Microstructural observations reveal different fatigue mechanisms between the binary and ternary alloys. The beneficial effects of the third element additions appear as an inhibition of crack propagation under thermal fatigue conditions. This work was supported by the Director, Office of Energy Research, Office of Basic Energy Sciences, U.S. Department of Energy, under contract No. DEAC03-76SFOOO98.
(4:40 p.m.)
COMPOSmONAL DEPENDENCE OF TIlE FATIGUE BEllAVIOR OF TINLEAD SOLDERS,· T S E Summers D. Tribula and lW. Morris, Jr., Center for Advanced Materials, Lawrence Berkeley Laboratory and Department of Materials Science and Mineral Engineering, University of California, Berkeley, CA 94720. The isothermal fatigue behavior of five compositions within the binary Sn-Pb system (5Sn-95Pb, 20/80, 40/60, 50150 and 63/37) was studied as a function of strain, strain rate and temperature. The fatigue life data resulting from this study will be presented. In addition, the microstructural evolution during fatigue and the subsequent failure mechanisms will be discussed. Past work has shown that at 75°C with a 10% strain amplitude and a strain rate of lO-4s-l, recrystallization is involved in failure of all of the above compositions, and in the eutectic-rich alloys, the Pb-rich proeutectic has a beneficial effect on the failure mechanism. The effect of strain, strain rate and temperature on these fatigue mechansims will be correlated to their effect on the fatigue life. *This research is funded by the Director, Office of Energy Research, Office of Basic Energy Sciences, Material Sciences Division of the U.S. Department of Energy under Contract No. DE-AC03-76SFOOO98.
114-Annual Meeting
EXPERT SYSTEM APPUCATIONS IN MATERIALS PROCESSING & MANUFACTURING IV Sponsored by the TMS Shaping & Forming Committee, the TMS Synthesis & Analysis in Materials Processing Committee and the ASM-MSD Computer Simulation Activity Wednesday PM February 21, 1990
Room: ParkB Inn at the Park
Session Chainnen: R. Harrison, U.S. Army Materials Technology Lab., Watertown, MA 02172; L. Roth, Inference Corporation, 5300 W. Century Blvd., Los Angeles, CA 90045
incorporates both feedforward and feedback elements. Informal control objectives based on user and device fabrication requirements are used as input in the determination of feedforward trajectories for the control of formal objectives to be achieved during the growth process. A real time thermal imaging system is employed for non· intrusive characterization of the melt thermal field distribution and is,in parallel,used for state reconstruction of principal processing variables. The complex relationships between processing inputs and resulting electronic materials properties require the use of elements of artificial intelligence in the determination of feedforward trajectories to perform a trade-off analysis to balance competing requirements of device properties, yield and processing cost. (4:00 p.m.)
LOAD SCHEDULING EXPERT SYSTEM FOR HIGH QUALITY LARGE STEEL COILS: Ramesh Bhashyam and Lewis Roth, Inference Corporation, 5300 West Century Boulevard, Los Angeles, CA 90045 This paper describes a prototype load scheduling expert system designed to improve the quality of large steel coils. A prototype has been implemented using Inference's ART expert system tool. The line load schedule indicates the order in which available coils must be loaded onto the production line. Given qualitative descriptions of available coils, for example coil hardness, thickness, roughness and width among others, the system arrives at the optimum loading pattern by applying constraints aimed at improving the output quality of steel and improving production line speed. The loading pattern is then presented back to the supervisory operation control staff.
(2:00 p.m.)
(4:30 p.m.)
PREDICTIONS OF YIELD STRENGTH FROM COMPOSITION USING NEURAL NETWORKS: M. A. Przystupa, P.O. Box 39, Redondo Beach, CA 90277.
DEVELOPMENT OF A FRACTURE ANALYSIS EXPERT SYSTEM USING BASIC CONCEPTS OF KNOWLEDGE ENGINEERING: Ma Cheng and GIen A. Stone, South Dakota School of Mines and Technology, Rapid City South Dakota 57701
Following author's experience with the expert system based approach to the yield strength predictions (ALADIN Expert System for Alloy Design) a neural network approach to the same problem has been explored. The network used in this work has been of a back-propagation type with the hidden layers and it has been trained to predict yield strength from composition for various aluminum alloys. During the talk the final structure of the network will be discussed and compared with other possible architectures. Also the training process, the necessary set of data and predictive capability of the trained network will be described in details. (2:30 p.m.)
AI APPLICATIONS DURABILITY/LIFE PREDICTION OF COMPOSITE MATERIALS, Gary L. Ha~nauer and Suzanne G.W. Dunn, U.S. Army Materials Technology La oratory, Watertown, Massachusetts 02172 AI technologies are being employed to automate procedures for determining properties and evaluating the long-term durability of composite materials. Intelligent robotic work cells have been designed to compensate for the lack of technicians and yet provide needed flexibility for the automation of tests that are repetitive, operator-sensitive and/or hazardous. AI techniques facilitate real-time monitoring, control and integration of robotic work cells; planning and scheduling of tests; acquiring and handling data; and automating knowledge acquisition (i.e., assists experts in correlating data, interpreting test results, proposing and evaluating models to simulate and predict material behavi or, and preparl ng reports). AI app I i cat ions for .experimental design, robot integration, image analysis and data interpretation will be discussed. (2:55 p.m.)
PROSPECTS FOR INCORPORATION OF ARTIFICIAL INTELLIGENCE TECHNIQUES INTO MATERIALS RESEARCH: Ralph J Harrison, U.S. Army Materials Technology Laboratory, Watertown, MA 02172 We will review requirements, progress and prospects for the implementation of artificial intelligence techniques which can benefit the materials scientist in carrying out research. We do not of course propose replacing the researcher by an expert system; the BACON program for making scientific deductions would, if it had existed at the time, not have threatened Francis Bacon's place in the history of science. Instead we will describe AI tools which promise to become useful additions to the materials researcher's toolkit. These may include, for example, the utilization of pattern recognition or neural net techniques in addition to more conventional statistical analysis to discover correlations among various phenomena. More generally, knowledge based software for materials research (KBMR) can go beyond algorithmic evaluation of data in helping the researcher in problem solving. Some of this software may be applicable to generic scientific or engineering research, while other parts must be specific to the requirements for research in materials or even the particular research domain. 0:20 p.m.)
APPLICATION OF ARTIFICIAL INTELLIGENCE FOR ADVANCED PROCESSING OF ELECTRONIC MATERIALS: Michael J. Wargo, Massachusetts Institute of Technology, Cambridge, MA 02139 Electronic devices today fail to achieve theoretically predicted limits of performance due to an inability to control the chemical and crystalline perfection of the solid state matrices from which the devices are fabricated. Because performance and yield of electronic devices depend strongly on the close interaction between materials properties and device processing procedures, the 'perfect' material for one fabrication line produces unacceptable devices on another which utilizes a different processing scheme. A muItivariable control system is proposed for the growth of advanced electronic materials which
This paper describes what is expected of an expert system, introduces the basic concepts and skills of knowledge engineering, and discusses the role of knowledge engineering and domain experts in development of expert systems. In particular, the organization of the knowledge base, the representation of the inference algorithm and organization of the rules are presented. The paper emphasizes the role of domain experts in developing expert systems and focuses on the principles underlying design of expert systems. The method used to develop a user friendly interface is described. An example "Fracture Analysis Expert System" is introduced to illustrate these principles.
mGH TEMPERATURE DEFORMATION, FRACTURE AND APPLICATION OF INTERMETALLIC COMPOUNDS IT: Nickel and Titanium Aluminides Jointly sponsored by ASM Flow and Fracture Committee and TMS High Temperature Alloys Committee Wednesday PM February 21, 1990
Room: Plaza B Inn at the Park
Session Chainnen: J.D. Whittenberger, NASA Lewis Research Center, Cleveland, OH 44135; D.F. Hasson, Dept. ofMech. Engr., U.S. Naval Academy, Annapolis, MD 21402 (2:00 p.m.)
Invited)
High Temperature Deformation Behavior of B2 NiAI R Darolia, DF Lahrman and RD Field, GE Aircraft Engines, 1 Neumann Way, Cincinnati, Ohio 45215. Alloys based upon near stoichiometric NiAI offer potential applications in high thrust to weight turbine engines because of their low density and high melting points. Poor ductility at low temperatures has been for a long time the major obstacle to the development of this material, but, the high temperature yield, creep and rupture strengths also need to be improved. Fortunately, alloying approaches to this problem employing both solid solution and precipitation strengthening have been identified, and encouraging results have been reported in the literature. However, specific mechanisms for strengthening and the reasons for the widely varying potency of different alloying elements are not fully understood. In this talk, the high temperature properties and specific strengthening mechanisms to improve the high temperature properties of NiAI will be described. The yield, creep and rupture behavior will be described as a function of temperature and correlated with composition, microstructure and dislocation mechanisms.
Annual Meeting-U5 (2 :30 p.m.)
(Invited)
ON THE MECHANISM OF DUCTILITY IN NiAl: Daniel B. Miracle, WRDC/MLLM, Wright Patterson AFB, OH 45433-6533. The mechanism responsible for the ductility of polycrystalline NiAl has previously been proposed as the stress-assisted diffusional climb of <001> dislocations above RT. However, there are a number of inconsistencies with this model. A new mechanism based on the local activation of additional slip systems at internal stress concentrations 1s proposed to explain the ductile-to-brittle transition in NiAl. A model is developed to calculate the stresses at grain boundaries in NiAl, and to resolve these stresses onto potential slip systems. The results of these calculations show that <111> dislocations may be generated locally in polycrystalline NiAl at applied stresses well below those required for general activation of <111> dislocations in cube-oriented single crystals. The temperature dependence of this effect shows that the local activation of <111> dislocations is able to account for the ductile-to-brittle transition in polycrystalline NiAl.
(3:00 p.m.) STRUCTURES IN B2 ALLOYS: C R Feng and K. Sadananda, Materials Science and Technology Division, Naval Research laboratory, Washington, DC 20375-5000.
DISLOCATION
The dislocation structures in ~everal hot-extruded B2 alloys including FeAl, NiAl, NiAl-Zr and CoNiAl were investigated by TEM methods. For FeAl, edge dislocations of 'Burgers vectors were observed only in grains that were favorably oriented. In majority of grains, however, the dislocations were of <100> type. The <100> dislocations aligned themselves along cube directions because of elastic anisotropy of the material. The similarities and the differences in dislocation structures among various alloys were compared and the implications in relation to deformation process will be discussed. (3:20 p.m.) ELEVATED 1EMPERAlURE MECHANICAL BEHAVIOR OF Ni·20AI·30Fe. ~ P.R.Munroe and LBaker. Thayer School of Engineering, Dartmouth College, Hanover NH03755 An alloy Ni·20AI·30Fe was produced by casting and hot extrusion. The resulting microstructure consisted of a B2·structured primary phase in a fine L!2+B2 eutectic. At room temperature this alloy exhibited a yield strength of 760MPa and 22% tensile elongation. At higher temperatures the yield strength decreased dramatically from 610 to 310 to l50MPa at 700, 900 and I lOOK respectively. At the same time the tensile elongation increased until at I lOOK it was greater than 70%. Tensile testing of alloys which have the same compositions as the constiruent B2 and Ll2 phases in Ni·20AI·30Fe, revealed that the B2 phase retains its strength up to 700K, whereas the Ll 2·based alloy retains its strength up to 9OOK. These results will be discussed in terms of the microstructures and operating deformation mechanisms of Ni·20Al·30Fe at elevated temperarures. The effect of boron on the strucrure and properties of this alloy and its constiruent phases will also be presented. This work was supported by a grant from NASA· Lewis Research Center under contract no. NAG 3-775 (3,40 p.m.)
BREAK
tension and compression in the temperature range of 200-800 o C, around the observed peak flow stress temperature_ Both polycrystalline and directionally solidified large grained material were subjected to identical treatments. Yield strengths, tensile strengths and ductiltiy parameters will be presented in conjunction with observed dislocation configurations_ In addition fracture behavior of materials will be discussed as a function of temperature. (4:30 p.m.) THE DEFORMATION AND FRACTURE OF ALPHA·2 TITANIUM ALUMINIDE ALLOYS AT ELEVATED TEMPERATURE: BJ Marquardt and GK Scarr. GE Aircraft Engines, Cincinnati, Ohio 45215 Several alpha·2 titanium aluminide alloys have been tested for tensile and creep behavior at elevated temperatures. The alloys were melted to a 6 inch diameter by VAR, forged to 2.75 inches in diameter, and then extruded below their beta transus temperatures at a ratio of 22:1. Following extrusion, the materials were heat treated to produce varying alpha·2 and beta phase morphologies. Tensile tests were conducted over the temperature range of room temperature to 760°C (1400°F) and creep tests were conducted over the range of 540 to 760°C (1000 to 1400°F). Post·testing analysis has included SEM fractography and evaluation of creep deformation mechanisms with the use of TEM. The effects of composition, morphology, and temperature will be discussed as they relate to changes in deformation and fracture mechanisms. This study was conducted under the General Electric Company Independent Research and Development Program. (4:50 p.m.) THE STRUCTURE, MECHANICAL PROPERTIES AND DEFORM ATION BEHAVIOR OF TiAI·X ALLOYS. by M. A. imam, J. Feng, H. Jones and K. Sadananda, Naval Research Laboratory, Washington, DC 20375·5000 Ordered intermetallic materials offer a considerable advantage in mechanical properties such as high modulus, especially at high temperatures, high strength at elevated temperatures, high strain hardening rate and low diffusion with resulting low creep rates, but their use have been limited because of their poor ductility at room temperature. In the present investigation efforts are being made to understand and improve the room temperature ductility by alloying. Results of preliminary investigation on structure. mechanical properties and deformation behavior of TiAI with varying amounts of manganese additons will be discussed.
mOB TEMPERATURE SUPERCONDUCTINO COMPOUNDS VI: DeformationlMechanical Properties Sponsored by the TMS Superconducting Materials Committee
(3:50 p.m.)
Wednesday PM February 21, 1990
ELEVATED TEMPERATURE FRACTURE AND FATIGUE CRACK GROWTH RESISTANCE OF TITANIUM ALUMI· NIDE INTERMETALLICS: S.Suresh and P.E. Aswath, Division of Engineering, Brown University, Providence, Rhode Island 02912, USA.
Session Chainnan: R.B. Poeppel, Argonne National Laboratory, Argonne IL 60439
This work focusses on the effects of microstructure and temperature on the fracture characteristics of Ti·AI and Ti·AI-Nb intermetallics. Sys· tematic and controlled heat treatments are designed to produce duplex microstructures with different mixities (}:2 and {3 or (}:2 and I phases. The fracture toughness, R-curves and stable fatigue crack growth rates are determined over the temperature range of 20 DC to SOODC. The effects of microstructure on the overall crack growth resistance are investigated in the light of optical and TEM observations of crack-tip damage, scanning .auger spectroscopy of fracture surface of oxidation and scanning electron fractography. The fracture and fatigue characteristics of the intermetallics are compared and contrasted with those of the more conventional Ti·AI·V alloys and super (}:2 microstructures. Mechanisms of subcritical crack growth and their dependence on the microstructure are examined in detail.
(4: 10 p.m.)
DISLOCATION STRUC11nRKS AND DEFORMATION BEHAVIOR IN
TiAl+X ALLOYS_ Bimal Kad and Ben F_ Oliver, The University of Tennessee, Knoxville, TN 37996-2200. Ternary alloyed TiAl appears to be a promising route to enhance strength and ductility in this material. TiAl alloys with ternary additions of gallium and manganese alloys were deformed in
Room: Orangewood I Quality Hotel
(2:00 p.m.) MECHANICAL a:NSlDERATICNS IN '!HE FI
current density are considered.
116-Annual Meeting (2:30 p.m.)
SHOCK-INDUCED DEGRADATION OF SUPERCONDUCTIVITY AND ITS RECOVERY AFTER HEAT TREATMENT OF EXPLOSIVELY FABRICATED Y-Ba-Cu-O AND Bi-Pb-Sr-Ca-Cu-O: L.E. Murr, C.S. Niou, M. Pradhan-Advani, and L.H. Schoenlein,* Dept. of Metallurgical Engrg. &Materials Science, The University of Texas - EI Paso, EI Paso, Texas 79968 (*on leave from Battelle Pacific Northwest Laboratories, Richland, WA). it has been demonstrated that superconducting copper-oxide based powders can be consolidated and encapsulated within a supporting metal matrix by explosive fabrication, the residual superconductivity is observed to be compromised by shock-induced degradation which appears to be related to crystal defects created in the shock front. YBa2Cu307 exhibits no change in the onset Tc (at 90K) within the pressure range 4-8 GPa, but the Ilid-point Tc decreases with increas ing pressure. In Bi -Pb-SrCa-Cu-O, the Tc does not appear above 77K, even at 4 GPa, and the normal state resistance, like that in Y-Ba-Cu-O, exhibits a semiconducting behavior. Extracting the consolidated superconductors from fabricated monoliths shows that anneals as high as 930·C in Y-Ba-Cu-O and 860·C in Bi-Pb-Sr-Ca-Cu-O, respectively, are required to recover near ideal R-T signatures.
ments conducted on warm compacted billets, are found to substantiate the phenomenological model proposed earlier by the author~ Effects of different heat treatments on the microstructure and mechanical and electrical properties of the sintered billet and of the extruded as well as rolled wires (0.75mm diameter), have been studied. Based on these results, a microstructure controlled scheme for fabricating superconducting composite products of useful shapes, has been suggested.
~hile
:2:50 p.m.)
EFFECT OF ADDITIVES ON THE MORPHOLOGY AND SUPERCONDUCTIVITY JF YBa Cu 0 CERAMIC MATERIALS 2 3 6 ~.S. Rao, O.px Arora and L.F. Aprigliano David Taylor Research Center, Annapolis, ~ID 21402-5067
In order to improve the superplastic behavior of iBa2C~306+x ceramic materials, the grain size of the ,eramlC was reduced by the addition of both aluminum oxide .nd silver oxide in the concentration range 0 - 15 %(wt). rhe results suggest that the addition of these additives .ffect both the superconducting phase of the ceramic and the sintered grain size. The average grain size of pure .nd 15 %(wt) additive containing bulk ceramic was in the range of 20 - 60 and 1 - 5 microns respectively. It was .lso found that the additives used in this investigation stabilized the tetragonal phase of the 123 compound. As a result the Tc of the 123 material has dropped from 90 K to 50 K for samples containing the additives at high concentrations (~ 5 %(wt». The detailed analysis of these results will be presented at the meeting.
:3: 10 p.m.) ~EFORMATION PROCESSING OF YBa Cu 0 : K.C. Goretta, A. Brandstadter, J.P. Singh, 2 R.B. Poeppe1, A.J. Schultz, oong~u"Shi, and J.L. Routbort, Argonne National Laboratory, Argonne IL 60439
Plastic deformation can be used to fabricate usefg1 shapes of YBa Cu 0 2 3
(YBCO) YBCO defonns via diffusional flow above about 850 C. Flow rates are, ~owever, low and hence fanning processes must be slow. Results of high-temperature deformation, and hot rolling of YBCO and YBCO + Ag will be presented and discussed. Microstructures can be tailored by manipulation of starting powders and processing temperatures. The relationship between microstructures and properties will be examined.
Work supported by the U.S. Department of Energy, BES-Materials Sciences, and the Office of Energy Storage and Distribution, Conservation and Renewable Energy, under Contract W"31-109-ENG"38. A. Brandstadter was funded by a grant from the American Nuclear SOCiety.
(3:30 p.m.) TOUGHENEO YBa Cu O/ZrO Composites: K.C. Goretta, M. Kullberg, 0.0. Lacy, U. 2 Balachandran, R.~. xpoep~el. Donglu Shi, Ming Xu. and J.t. Routbort, Argonne National Laboratory, Argonne, IL 60439 Tetragonal ZrO particles have been added to YBa cu 0 (YBCO) to improve the 3 fracture toUgh~eSS of the superconductor. When ~ure ~ro2 is added, a chemical reaction with YBCO occurs and a semiconducting composite results. Sol-gel coatings have been appl ied to the ZrO. The deleterious reaction has been mitigated, but the improvements in to~ghneSS have been retained. Details of the coating composition and procedure will be presented. The electrical and mechanical properties of the resultant compOSites will be dicussed. Work supported by the U.S. Oepartment of Energy, Office of Energy Storage and Distribution, Conservation and Renewable Energy, and BES-Materials Science, under Contract W-31-109-ENG-38. 0.0. Lacy performed research under appointment to the Historically Black Colleges and Universities Nuclear Energy Training Program administered by Oak Ridge Associated Universities for the U.S. Oepartment of Energy.
(3:50 p.m.) STUDIES ON MICROSTRUCTURE-PROPERTY RELATIONSHIP OF YBa2Cu307-x+ Ag COMPOSITE WIRE: S. Samajdar and S.K. Samanta, Plasticity Laboratory, The University of Michigan, Ann Arbor, MI 4Sl09. Microcomposites of YBCO and silver formed through powder proces~ ing route have already shown attractive properties, particularly in regard to flexibility, formability and zero resistivity (at 77K) into usable shapes. The material has been characterized, in this effort, by evaluating the microstructure at the various stages of the fabrication route. Experiments on powder mixing procedure show interesting results. Recent results of experi-
mGH TEMPERATURE SUPERCONDUCTING COMPOUNDS Vll: Wireffape/Filaments Sponsored by the TMS Superconducting Materials Committee Wednesday PM February 21, 1990
Room: Orangewood IT Quality Hotel
Session Chairman: Sungho Jin, AT&T Bell Laboratories, Murray Hill, NJ 07974 (2:00 p.m.)
PROCESSING AND CHARACTERIZATION OF CERAMIC SUPERCONDUCTING POWDERS AND COMPOSITE MULTIFILAMENTARY WIRE M.R. Notis, B. Avitzur, M.S. Oh and a.F. Uu Lehigh University, Bethlehem, Pennsylvania 18015 Processing technology for fine multifilamentary superconducting wire requires unique control of the starting ceramic powder, the metal forming operation during fabrication of the co-drawn or co-extruded composite, and the post fabrication heat-treatments which can cause both microchemical modification and microstructural rearrangement of the fine ceramic filaments as well as interfacial reactions with the containment tubing. Each of these aspects of processing technology (starting powder, wire forming, and heattreatment) will be discussed in detail and results of characterization will be shown with the goal of producing multifilamentary wire with optimum properties. Processing parameters for both wire drawing and hydrostatic extrusion are compared. Results will be shown for a number of different superconducting ceramic compositions, including YBa2Cu 30 7 (1-2-3) and BiSr-Ca-Cu-O, and different containment metals, including Ag, Ni and Cu.
(2: 30 p.m.) FIBER FORMATION IN BISMUTH BASED SUPERCONDUCTOR, D.K. Finnemore, Department of Physics, Iowa State UniverSity, Ames, Iowa 50011. Long slender fibers of the high temperature superconducting compound, Bi2Sr2CaICu30S-o, can be prepared by shaping and freezing the material with a gas stream in a special nozzle. Fibers, typically 1 to lO~m in diameter and 1 cm long, are amorphous or nanocrystalline Bi(22l2). When they are amorphous, the fibers are very strong and can be processed into a microfilamentary composite. The goal is to make magnet conductors based on these materials.
(3:00 p.m.) PROCESSING OF SUPERCONDUCTING COMPOSITE MATERIALS DIRECTLY FROM THE MELTS: S.H. Whang, Z.X. Li and D.X. Pang, Department of Metallurgy & Materials Science, 333 Jay St., Brooklyn, NY 11201 Preparation of continuous superconducting oxide composites from the melts through two different techniques has been established. The two processings will be described in terms of processing methodology, oxygen annealing, microstructures and superconducting properties. Improvement in J is possible using the melttexture technique. Thecmicrostructures and the superconducting properties resulted from the texturing will be presented along with weak link effect in these materials.The stability of the interface between the oxides and the substrate due to the prolonged annealing will be discussed. The potential for this approach will be addressed in comparison with other comparable processing methods. (3,25 p.m.)
BREAK
Annual Meeting-1l7 (3:45 p.m.) FUNDAMENTAL PROPERTIES AND MICROSTRUCTURE OF HIGH-Tc SUPERCONDUCTING TAPE PREPARED BY DIRECTIONAL MELT-GROWTH TECHNIQUE T.Matsumoto, M.Okada, K.Aihara, and S.Matsuda, Hitachi Research Laboratory of Hitachi Ltd., Hitachi, Ibaraki, 319-12, Japan, M. Seido, Metal Research Laboratory of Hitachi Cable Ltd., Tsuchiura, Ibaraki, 300, Japan. Improvements in microstructure, designed to enhance the critical current density (Jc) of HTSC superconducting wire in an applied magnetic field, is a matter of some urgency. A directional melt-growth technique was successfully realized to improved the grain boundary weak-link behavior in HTSC superconducting tape by aligning the superconducting crystallites. Jc's of several tapes will be discussed in relation to each individual microstructure.
(4: 15 p.m.) POWER DEVICE DEVELOPMENT USING HIGH TEMPERATURE SUPERCONDUCTlNG POWOERS: .hJ'.,. Roth~
T.C. Bickel. Sandia National laboratories, Div. 6224. Albuquerque, NM
87185
The manufacture of electrical power devices using high temperature supereonducting materials first requires the development of wires with the necessary mechanical and electrical properties of ruggedness and current density. Prototype superconducting wi res have been made from bul k powders of YBa Cu 07 Prepared by a unique chemical co-precipitation process. Hot isost~ti~ p~essed samples of this material have shown a zero-field J
>
1000 A/em
at 76K.
This
uchem prep" method has been shown feasible for proauction of-collmef'chl-quantities of superconcucting powder. The preparation of wires using plasma sprayed 123 materials on Ag substrates will be described as well as conductors prepared from co-extruded billets of Ag/123 powder.
(4:35 p.m.) HTS MATERIALS REQUIREMENTS FOR POWER TRANSFORMERS: James A. Dirks, Tom A.
Wi 11 iams, Steven B. Merri ck, Paci fic Northwest laboratory, P. O. Box 999, Richland, Washington 99352
The key HTS material and fabrication requirements necessary for HTS power transformers to be economically competitive with conventional designs are explored. The critical temperature of the HTS material is assumed to be suffecient to utilize LN2 cooling. Material and fabrication requirements considered in this paper include: - Critical current density. Jc; - Superconducting twist pitch, L ; - Superconducting filament diame~er, 0; - Superconductor strand radi us. rS This study employs a family of computer models which. for a given set of HTS properties. can investigate a wide range of alternative component designs and select the design which is o.ltimal from a life-cycle cost standpoint.
INTERFACES IN METAL-CERAMICS COMPOSITES V: Thermodynamics and Kinetics C Sponsored by Process Fundamentals Committee and Composite Committee Wednesday PM February 21, 1990
Room: Plaza D Inn at the Park
Session Chainnen: T. Nicholas, WRDCIMLLN WrightPatterson AFB, Dayton, OR 45433; J.M. Yang, Dept. of Mat. Sci. & Eng., UCLA, Los Angeles, CA 90024 (2:00 p.m.)
THIN ALUMINA COATINGS ON BOROSILICATE GLASS: INTERFACIAL INTERACTIONS DURING HEAT TREATMENT, G.Se1vaduray, and L.Martinez, Dept. of Mat.Engrg, San Jose State University,San Jose, CA 95192. The possibility of using a thin alumina coating on borosilicate
glass coated ceramic fiber composite tile insulation (RCG/AETB
tiles) for space applications was investigated to determine the potentia1 use of alumina in improving the high temperature resistance of the glass coating. Three RCG/AETB tiles were spray coated with O. 55jjfficc-alumina powder suspended in ethanol. The average alumina coating thickness was 5. O.um. . Two tiles were
the underlying borosilicate glass after both radiant and convective heat exposure. Sintering of the alumina particles was confirmed. The sintered alumina particles were found to remain as a separate entity, without becoming incorporate into the borosilicate glass as particles. WDS revealed evidence of alumina dissolution into the borosilicate glass. (2: 20 p.m.)
INTERFACIAL CHARACTERIZATION OF SILICON NITRIDE REINFORCED CERAMIC-MATRIX COMPOSITES: A THERtl0DYNMIC APPROACH: t1. Singh and H. Wiedemeier, Dept. of Chemistry, Rensselaer Polytechnic Institute, Troy, NY 12180. Ceramic-matrix composites (CMCs) with higher operating temperature capabilities are required in a variety of structural components in high temperature jet engines and other aerospace applications. Silicon nitride attracted wide attention due to its potential as matrix and reinforcement for some of these applications. But the chemical stability of silicon nitride under different processing and operating conditions has been a major concern. In the present study, thermochemical analysis has been carried out to evaluate the stability of silicon nitride under different conditions and its interfacial reactions between some coating and matrix materials. The effect of certain environmental conditions i.e., different nitrogen and oxygen partial pressures will be discussed. (2:40 p.m.)
INTERFACIAL REACTIONS IN FIBER-REINFORCED GAMMA TITANIUM ALUMINIDE COMPOSITES: S. Krishnamurthy, Metcut-Materials Research Group, P.O. Box 33511, Wright-Patterson AFB, OH 45433 The quest for low density materials having high specific properties, thermal stability, and increased resistance to environmental effects has led to widespread interest in titanium aluminide composites. The present study concerns composites of gamma (TiAI) aluminide alloys reinforced with ceramic fibers such as silicon carbide and alumina. Small quantities of these composites were prepared by HIP'ing matrix alloy powders with fiber samples. As-produced and heat treated composite specimens were characterized by SEM, STEM, EDS, and other techniques to determine the extent of fiber/matrix reaction, the reaction products formed, and the changes in composition and microhardness across the interface. The results of this reaction zone analysis in the gamma aluminide MMC's will be discussed. (3:00 p.m.)
INTERFACIAL EVALUATION IN A Ti-25AI-17Nb/SCS-6 COMPOSITE: P.R. Smith and W.C Revelos, WRDC/MLLN Wright-Patterson AFB, OH 45433, and CG. Rhodes, Rockwell Science Center, Thousand Oaks, CA 91360 A Ti-25AI-17Nb/SCS-6 composite was fabricated using vacuum hot pressing techniques causing the formation of a multi-phase reaction zone at the fiber/matrix interface. This reaction zone was further grown by isothermal exposure at elevated temperatures. Microstructural characteristics of the reactions products in both the as-consolidated condition and in the thermally exposed conditions were determined by optical microscopy, scanning and transmission electron microscllpy, and wavelength dispersive spectroscopy. Reaction zone kinetics were determined in the range of 1200°F-1800°F and microhardness correlated with the various reaction product forms. The results will be compared to those previously determined for the SiC-reinforced Ti-24Al-llNb and Ti25Al-10Nb-3V-IMo. (3: 20 p.m.) Break (3: 30 p.m.)
MICROSTRUCTURAL CHARACTERIZATION OF REACTION ZONE IN SiC FIBER REINFORCED TI-ALLOY AND TITANIUM ALUMINIDEBASED MATRICES: Gopal ~ Metcut- Materials Research Group, P.O. Box 33511, WPAFB, OH 45433-0511 Microstructural characterization of reaction zones in SiC fiber reinforced titanium alloy and titanium aluminide matrices will be presented. Four matrices such as IMI 829 ( near a titanium alloy), Ti-l0V-2Fe-3Al (~ titanium alloy), Ti3Al-Nb and Ti-Al-Nb are considered. Changes in mechanical properties in the matrix adjacent to the reaction zone are monitored by the microhardness indentation technique and attempts will' be made to correlate such changes with chemical composition. Results are discussed in terms of load transfer capabilities of the matrix to the fiber through the reaction zone and designing matrices suitable for composites.
exposed to radiant and convective heating at 1426°C (2600 0 F) for
(3: 50 p.m) THERMODYNAMIC STABILITY OF TRANSITION-METAL SILICIDE/REFRACTORY CERAMIC MIXTURES.* P. J. Meschter. McDonnell Douglas Research Laboratories, P. O. Box 516, St. Louis, MO 63166.
scanning electron microscopy (SEM) and wavelength dispersive spec troscopy (WDS). The alumina layer was found to be banded to
Silicide-matrix/ceramic reinforcement composites (e.g., MoSiZ/SiC) are attractive for high-temperature applications prov1ded that matrix and reinforcement are thermodynamically com-
50 minutes. The third was kept as a reference. The alumina/borosilicate glass interface interactions were studied by
118-Annual Meeting patible upon prolonged exposure to temperatures up to 1600·C. In this study, stable Ti Si /ceramic or MoSi 2 /ceramic combinations 5 3 Ti-Si-B, TI-Si-C, Mo-Si-B, and Mowere predicted from calculated Si-C phase diagrams. The stabilities of candidate composites predicted by these diagrams, as well as Ti Si 3 /ZrB , 2 Ti Si 3 /A1 0 , MoSi 2 /TiB 2 , MoSi /ZrB 2 , MoSi 52 /TiC, and MoSi 2 /A1 0 2 3 2 2 3 5 tested were by annealing hot-pressed silicIde/ceramic powder mixtures for 100 h at 700-1600 o C.
The extent of reaction was
determined by SEM, EDX, scanning Auger microscopy, and x-ray diffraction. Promising thermodynamically and mechanically compatible combinations are Ti5Si3/TiB2 and MoSi with TiB , TiC, 2 and SiC, in general accord with thermodynamic 2predictions.
*
Research conducted under the McDonnell Douglas Corporation Independent Research and Development program.
(4:10 p.m.) STIFFNESS REDUCTION OF COMPOSITE LAIIINATES DUE TO INTERFACIAL DEBONDING, Yong-Oiu Jiang, Department of Engineering Mechanics,
Xian Jiaotong University, Xian, Shaanix, P.R. China.
The fracture mechanisms point to that a fiber break induces very
high shear stresses at the fiber-matrix interface of polymer matrix composites near the break.
If this stress exceeds the
shear strength of the interface, the fiber will debond from the matrix in the vicinity of the break with certain debonding length. The numerical results from a model considering a single fiber and the nearby cylindrical matrix field show that the debonding
has
a
major
effect
on
stiffness
reduction
of
Glass/Epoxy. The model considered is also good to estimate the stiffness reduction of metal or ceramic matrix composites due to
interfacial debonding.
p.m.) WETTING CHARACTERISTICS AND INTERFACIAL REACTIONS OF ALUMINIUMGRAPHITE SYSTEK, Chen Zhongyu Wu J inbo, Huang Xiangui, Department of Materials Science and Engineering, Zhej iang University, Hangzhou, 310027, P.R., China.
(4:30
Sessile drop experiments were carried out in order to measure
contact angles and to investigate wetting characteristics of commercial pure Aluminium and some binary, trinary Aluminium
alloys on polycrystallise Graphite system were studied, and an
was made an approach to the interface structure and phases of the system with microscope and X-ray diffraction.It was shown
that the interfacial phenomena of Aluminium-Graphite system were characterized of the reactive wettability. The change of the contact angles with the temperature was considered to have
two different features. The large obtuse angles were exhibited at the temperature below 950 0C although the wetting behavior was found to be improved with the temperature increasing from melting point to 950 0C.When temperature was higher than 950 0C, the interfacial reaction took place and a continuous interfacial phase was formed. The interfacial reaction caused
pure Aluminium or Aluminium Alloys to wet Graphite substrate, the contact angles of the system were 20-55 0 at 11000C. It was further indicated by X-ray diffraction that the continuous interfacial phase is A14C3.
1990 LEAD-ZINC AND RELATED METALS SYMPOSIUM XII: Lead, Zinc and Tin Minerals Processing Sponsored by the TMS-EPD Lead-Zinc-Tin Committee Wednesday PM February 21, 1990
Room: Salon C & D
Marriott
Session Chainnen: J.E. Hoffmann, Jan H. Reimers & Assoc., 11531 Meadowlake Drive, Houston, TX. 77077; W. Petruk, Canmet, 555 Booth St., Ottawa, Ontario KIA OGI Canada (2:00 p.m.)
KINERALOGICAL CHARACTERISTICS THAT AFFECT RECOVERIES OF SPHALERITE FROK VOLCANOGENIC BASE KETAL ORES, W. Petruk, Canmet, 555 Booth Street, Ottawa, Ontario, Canada K1A OG1. Sphalerite in base metal ores commonly occurs as layers and interstitial material. Layered sphalerite is liberated during primary grinding, but interstitial sphalerite requires regrinding to achieve liberation. Around 95% of liberated sphalerite is generally recovered in zinc concentrates from zinc circuit feeds. A series of tests was conducted on sphalerite in the ore of Brunswick Kining and Smelting during a continuous fine tuning operation of the concentrator. It was found that a grind of 80% minus 37 micrometres was required to liberate the interstitial sphalerite. The fine grind increased Zn recoveries from 75%. to 78-80%, but produced liberated slimed
sphalerite which was lost. Since regrinding to a fine grain size is needed to obtain the desired sphalerite liberation, a method of recovering the slimed sphalerite is required. (2:30 p.m.) MINERALOGICAL CHARACTERISTICS THAT AFFECT RECOVERIES OF GALENA FROM BASE KETAL ORES, W. Petruk, J.M. Wilson and T.T. Chem, Canmet, 555 Booth Street, Ottawa, Ontario, Canada.
Galena in base metal ores occurs as discrete grains, as an interstitial mineral among pyrite and sphalerite, and as inclusions in pyrite and sphalerite. The inclusions and interstitial fillings are very small (a few micrometres in diameter) hence much of the galena is not liberated during grinding and regrinding. On the other hand large discrete grains are liberated during rod mill grinding and slime during ball mill grinding and regrinding. The general view is that the large free galena grains should be removed from rod mill discharge before ball mill grinding, but generally this is not implemented. Because a large proportion of the galena is unliberated, galena concentrates grading only 30 to SO wt% Pb are commonly produced in order to obtain lead recoveries of 65-80%. The galena concentrates contain significant amounts of silver. (3:00 p.m.) MINERALOGICAL CHARACTERISTICS THAT AFFECT RECOVERIES or CASSITERITE FROM THE BASE KETAL ORE OF KIDD CREEK KINES IN ONTARIO, W. Petruk, Canmet, 555 Booth Street, Ottawa, Ontario, Canada.
The Kidd Creek base metal deposit near Timmins, Ontario contains around 0.15 wt% Sn, mainly as cassiterite. The cassiterite occurs largely as inclusions in sphalerite, with 85% being in grains smaller than 18.5 micrometers in diameter. Concentrates and tails from the company's Cu, Pb and Zn circuits were studied to determine the liberations and recoveries of cassiterite in the products in order to establish whether the cassiterite can be recovered from concentrator tailings. The ore had been ground to about 65% minus 44 micrometers in diameter (-325 mesh); 55% of the cassiterite was liberated. About 92% of the liberated cassiterite and 70% of the unliberated cassiterite were in the concentrator tails, and 20% of the unliberated cassiterite in the zinc concentrate. (3:30
p.m.)
BREAK
(3:50 p.m.) TIN, TANTALUM, AND NIOBIUM RECOVERY FROM KOUGAROK, SEWARD PENINSULA, ALASKA, J. L. Johnson and W. R. McDonald, U.S. Department of the Interior, Bureau of Mines, Salt Lake City Research Center, Salt Lake City, Utah; Appendix on Economic and Technical Evaluation, Frank A. Peters and Thomas A. Phillips,
U.S. Department of the Interior, Bureau of Hines, Washington D.C.
The U.S. Department of the Interior, Bureau of Mines, investigated recovery of concentrates containing tin, tantalum,
and niobium from the Kougarok deposit on the Seward Peninsula, AK. The resource is estimated at 25 million short tons at average grades of 0.75 pct Sn, 0.025 pet Ta, and 0.025 pet Nb.
Deslimed are was concentrated by a Humphrey's spiral and tabling
to give a concentrate containing 54 pet Sn, 0.3 pct Ta, and 0.3 pct Nb, with recoveries of 74 pct Sn and 44 pct Ta/Nb. Slimes
were treated separately using a Bartles-Mozley separator and
flotation, followed by fuming, to recover an additional 6 pct Sn and 2 pct Ta/Nb. A flowsheet with mill-design parameters
resulted from the metallurgical testing. was made based on the proposed process.
An economic analysis
(4:20 p.m.) A NEW PROCESS TO UPGRADE ZINC CONCENTRATES USING A PARTIAL DESULPHURIZATION TECHNIQUE, D.R. Spink and K.D. Nguyen, Turbotat, Inc. 550 Parks ide Drive, Suite A-14, Waterloo, Ontario, Canada N2L 5V4, and E. Marrella, University of Waterloo, Waterloo, Ontario, Canada N2L 3Gl.
A partial desulfurization roasting (PDR) technique has been developed for use on zinc concentrates whereby the sulfide sulfur associated with the iron content is selectively removed in the roaster. The resulting PDR product is then leached for the selective removal of the iron. The leach residue is an upgraded zinc sulfide concentrate of very low iron content. The up-graded zinc concentrate can be subjected to a conventional dead roast without formation of deleterious zinc ferrite so that hot acid leaching and jarosite (or other) iron precipitation steps are obviated. Examples are presented for the treatment of three very different types of zinc concentrates. (4:50 p.m.) DEWATERING OF FINE LEAD-ZINC CONCENTRATES BY HYPERBARIC FILTRATION, W. Stahl, Institut fur Kechanische Verfahrenstechnik und Kechanik der Universitat Karlsruhe, 7500 Karlsruhe, FRG and H. Perchthaler, Kaschinenfabrik Andritz, A-804s Graz-Andritz, Austria.
The basic principle of hyperbaric filtration and technological advantages are described. An economic evaluation of these ad-
Annual Meeting-119 vantages is described taking lead and zinc as an example. Test procedures and scale-up as well as a general layout of a hyperbaric filtration plant, air supply, slurry pumping and safety features are described. A disc filter for hyperbaric filtration can be designed to put the advantages to full use. These include: uniform flow in the filter tank to get uniform solids concentration; design of filter elements, piping and control bead to get optimum gas and liquid flow; controlled filter cake removal from fabric using a blast of air, controlled by a snapblow valve; and filter cake discharge by double-trap lock.
LOW TEMPERATURE Si AND Si-BASED EXPITAXIAL STRUCTURES: ELECTRICAL PROPERTIES AND DEFECTS ITI: Si-Based lIeteros~ctures
Sponsored by the TMS Electronic and Photonic Device Materials and Electronic Materials Committees Wednesday PM February 21, 1990
Room: Orangewood III Quality Hotel
the MgA1 0 layer during the M8E growth of 5i ....as essential for improvements 2 4 of the NHOSFET characteristics. And we conclude that the present Sil -A1 0 / 2 Si structure is very promissing for fabricating 501 devices. This work was performed under the management of the R&D Association for Future Electron Devices as a part of the R&D of Basic Technology for Future Industries sponsored by NEDO(New Energy and Industrial Technology Development Organization) .
r
(3:10
Selective epitaxiaJ growth (SEG) and epitaxiallateraJ overgrowth (ELO) of Si over oxide have received recent attention as technologies for CMOS, advanced bipolar, BICMOS, and DRAMS. We fabricated a stacked inverted P·MOS device in crystalline Si on top of the n·channel device with an oxidized poly· gate sandwiched in between; thereby eliminating all wells and with reduced circuit size. The criticaJ "as·grown" interface state densities between the ELO and poly·oxide were measured to be 1.5XlO+ 1l /(cm2·eV). Hydrogen post annealing conditions effect these surface states. A Dichlorosilane·HC1·H2 system was employed at ISO Torr and at 900°C in a commercial reactor. The initial stacked·inverted 3D P·MOS devices typicaJly show IIp=110 cm2N·s with good subthreshold characteristics. The first sidewaJl gate controlled diode was fabricated to quantify the surface and near bulk region of Si·SEG for use in BIT advanced devices that do not require a buried layer and are self isolating. The SEG growth conditions are related to the defects along the sidewaJls and their effect on device performance. (3:
Session Chainnan: K. Rajan, Rensselaer Polytechnic Institute, Troy, NY 12181 (2:00 p.m.)
(Invited)
ELECfRICAL TRANSPORT IN Si/CoSi 2/Si HETEROSTRUCTURES GROWN BY IMPLANTATION. Alice E Whjte M. L. Green, K. T. Short, D. Brasen, and K. Maex', AT&T Bell Laboratories, 600 Mountain Ave., Murray Hill, NJ 07974. We have used mesotaxy, a technique which involves high dose implantation followed by high temperature anneaJing, to create thin «500A.) layers of oriented single·crystal CoSi 2 buried beneath the surface of (100) silicon wafers. Several thousand angstroms of epitaxial silicon are then deposited using an RTCVD process. This results in a Si/CoSi 2/Si heterostructure which is difficult to grow by UHV deposition and reaction. Rutherford backscattering and channeling measurements in conjunction with TEM observation indicate that the Si overlayer has threading dislocations at a density of _108 /cm2 . In addition, some SiC precipitates appear at the originaJ Si surface. Nonetheless, Schottky barrier height measurements show that the lower CoSi 2/Si interface is of acceptable quality and efforts are underway to improve the upper CoSi2/Si interface in order to make vertical transport measurements. permanent address: IMEC, Leuven, Belgium (2:30
p.m.)
ELECfRICAL AND OPTICAL PROPERTIES OF EPITAXIAL ErSi 2./Si HETEROSTRUCTURES: P.A. Badoz, J.Y. Duboz, F. Arnaud d'Avitaya and I.A. Chroboczek, CNET·CNS, BP 98, 38243 Meylan Cedex FRANCE. Recent results (Arnaud d'Avitaya et aJ., Appl. Phys. Lett. 54, 2198 (1989» showed that high quality epitaxiaJ layer of erbium silicide ErSi2., can be MBE grown on (111) Si. It was further shown (Duboz et aJ. Appl. Phys. Lett. July 1989) that the silicide is semimetallic and has an unusually low (respectively, high) Schottky barrier height on n type (respectively, p type) silicon. These features make ErSi2., very attractive for microelectronic applications such as metal base transistors, MetaVSemiconductorlMetal superlattices and Si based infrared detectors. In this paper, we present novel results on the electrical, magnetic, and optical properties of ErSi2., thin films on Si and discuss measurements of perpendicular transport measurements across the recently developed ErSi,.xlSi/ErSi2., MSM epitaxial structures.
(2:50 p.m.)
~~~~~~~Ti~A:~C~E:~~T~i~ g:o:r:~UEBT~~D~~~E~I~~~~~~~I~~ ~~~:A~:Ot~~iD!T~~~TURE T. NIINA:ULSI RESEARCH CENTER, Sanyo Electric Co .• Ltd. 1-18-13. Hashiridani. Hirakata. Osaka 573, Japan We developed a new MBE growth method for fabrication of high-quality double hetero-epitaxial Sild-AIZO IS~ structures. In the present ....ork ..... e investigated the characteristics at t-.'MOSFETs fabricated on the 5il -AIZO/ 5i structures. comparing with those of 5i/MgA1 0/5i structures. 2 -AI 0 layers were transformed from MgAI 0 layers by the reaction with 7 4 Z 3 51 beams as folloW'sj MgA1204+5i......,..I-A12~,+5iOt+Hgt. Then the MaE gro .... th of 5i on the effectively restructured 1 -A1 0'"3 layer was performed at a substrate 2 temperature of 700°C .....hich was 140°C lower than the MBE growth of 5i on the MgA1 0 /Si substrate. 2 4 Tne electron field effect mobility and leakage current bet ....een sour5:e and drain of the NM05FETs fabricated on 5i1 '( -A1 0./5i structure ....ere 660cm I 2 V' 5 and 2. 8pAlflm. respectively J .... hich were excellent compared with those on 5i/MgA1 0 lSi structure. 2 4 As the result of the investigation about the difference in the characteristics of NM05FETs. it was fn.,lnd that suppression of autodoping from
t
a
p.m.)
THREE DIMENSIONAL DEVICES FABRICATED BY SILICON EPITAXIAL LATERAL OVERGROWTH; ~~, W.A. Klaasen, J. A. Friedrich, and J. L. Glenn, School of Electrical Engineering, Purdue University, W. Lafayette, IN 47906
30 p.m.)
(3:45
BREAK
p.m.)
DELTA DOPING SUPERLATTICES IN SILICON: H. P. Zeindl and I. Eisele, Fakulttit fiir Elektrotechnik, Universitat der Bundeswehr Miinchen, 08014 Neubiberg, FRG. N- as well as p~type delta doped layers with free carrier areal densities up to some 1014 cm" have been grown by adding a low temperature step to the conventional MBE deposition technique. According to secondary ion mass spectroscopy and transmission electron microscopy the dopant confinement is less than 2 nm. Quantization effects for the twCH:limensional electron gas have been verified by resonant electron tunneling and intersubband IR absorption. Modulation doped superlattices as well as n--i-p--i structures with defined electron and hole concentrations have been fabricated. The carrier density and the mobility has been determined by electrochemical C-V depth profiling and Hall measurements. Temperature dependent transport investigations of doping superlattices have been carried out. (4:05
p.m.)
DESIGN ~ FOR A NEW+GFNERATION OF si l1lE srSTEMS: P W Sullivan , 0 Williams am R A A ~iak. VG Semicon Ltd, Fast Grinstead, RHl9 lXZ, Erglam. Dept of Rlysics, Wru:wick University, Coventry, CV4 7AL, Erglam.
MBE is a powerful technology 1Nhich is beirq awlied to the fabrication of advanced devices am navel structures such as
SijSiGe superlattices. HoWeVer sane significant prcblems have emerged, such as inconsistemies in ultimate material p.rrity, the density of particle related defects and heavy metal =ntamination. Furtherm:lre previoos generation MBE equipment is i.nc:atpatible with VISI processirq. VG Semi=n has introduced a new generation si MBE product, the V90S, to address the known prcblems and satisfy a ran:Je of requirements. Technological advancements in::lu:3e: 6" wafer harrllirq withcut wafer holders, semi-automated wafer transport, a low-particle deposition environment and a new high stability, 180= electron beam evaporator. '!he design =iteria for this system are discussed am prelintinary data are presented. p.m.) In situ 5TH Characterization of epitaxial layers in a 3-inch MBE system R. Butz, H. Wagner, Kernforschungsanlage JUlich GmbH, Institut fUr Schicht- und Ionentechnik 1, Postfach 19 13, 0-5170 JUlich, FRG
(4:25
The structural characterization of evaporated layers on an atomic scale is one of the most important problems in epitaxy. In this paper an experimental set up of an STM is described which allows to work in an UHV chamber attached to a conunercial Si-MBE system. 3-inch wafers, just prepared or deposited with Si/Ge can be transferred directly into the STM chamber. Pictures of the well known 7 x 7 structure of the Si(lll) demonstrate that the stability of the STM against vibrations and thermal drift allows to achieve atomic resolution. Differences in the surface structure of thermally cleaned samples and epitaxial layers are presented.
120-Annual Meeting
METALLIZATION FOR ELECTRONICS APPLICATIONS IV: Advanced Technology for Contacts to InP Sponsored by TMS Electronic and Photonic Device Materials and Electronic Materials Committees Wednesday PM February 21, 1990
Room: California 1·11 Quality Hotel
Session Chairmen: A. Katz, AT&T Bell Laboratories, Murray Hill, NJ 07974; T. Sands, 331 Newman Srpings Road, Red Bank, NJ 07701 (1 :30 p.m.) A COMPARISON OF THE REQUIREMENTS OF ELECTRICAL CONTACTS TO SOLAR CELLS AND THOSE USED IN VLSI. ~,T.A. Gessert, Solar Energy Research Institute, 1617 Cole Blvd., Golden, CO 80401 The importance of contacts to solar cells is often underrated even though the values of specific contact resistance required are relatively modest compared with VLSI. Under solar concentration, however, the upper tolerable limit on the specific contact resistance for a solar cell scales, roughly inversely, with the solar concentration. At I sun operation, the upper limit in specific contact resistance is usually taken to be 10-2 n cm2. Consequently, at several hundred suns this value must be reduced to 10-4 n cm2 or less. This paper is concerned with the design of large area rear contacts and collector grids to the front of the solar cells. As will be pointed out, many of the requirements of the contact are similar to those pertaining to VLSI although, in some respects, they are more demanding. For example, the front collector grid must be optimally designed to achieve the ideal balance bewteen geometrical shadow loss and series resistance. Under solar concentration the width of the grid lines becomes quite fine though still coarse compared with VLSI requirements. Nevertheless, a detailed analysis of the grid design reveals that when account is taken of current crowding, the grid line width can be reduced, with a consequent reduction of shadow loss. The discussions will center on III-V materials since these are currently of great interest in photovoltaics for operation under concentrated sunlight. In particular, the material InP that will be considered is of particular interest for space application in a severe radiation environment (2:00 p.m_) SCHOITKY CONTACTS ON "PROCESSED" SURFACES OF loP: U. K. Chakrabarti, C. W. Seabury, S. J. Pearton and T. Hayes, AT&T Bell Laboratories, 600 Mountain Avenue, :J1urray Hill, New Jersey 07974. ::=In contrast to the ideal situation of Schottky contacts prepared in situ on a vacuum cleaved :::IoP, we will describe tbe electrical characteristics of metal-lnP structures which are of real world device fabrication sequences. In a typical device processing .:ondition (e.g., etching, dry or wet) the near surface region of InP is unavoidably modified .residual lattice damage, changed stoichiometry, native oxide, etc.). We will present results _m the I-V characleristics of metal-InP structures prepared on "process" modified surfaces =nd show that Schottky characteristics can be used as a powerful diagnostic tool to ~presentative
!!!!!5upplement other analytical techniques. In addition. we will present the results on deliberate
o=:hemical surface modification leading to an enhanced barrier Schottley contact lI:(an = 0.7-0.8eV and n = 1.01-1.2) on n loP. Furthermore, as a practical application of <:nhanced barrier Schottley on n loP, we will present results on a high performance _:I... =4IOmA/mm,!lm = 233mS/mm) MES-FET fabricaled by ion implantation using """'If-aligned geometry.
(2:30 p.m_) AN ORDER OF MAGNITUDE IMPROVEMENT IN THE SPECIFIC CONTACT RESISTANCE OF GOLD BASED OHMIC CONTACT TO PInGaAsP (1.1 11m) BY RAPID THERMAL PROCESS. A Applebaum D. Wolf, Microelectronics Technology Center, Rockwell International Corp., 2427 W. Hillcrest Dr., Newbury Park, CA 91320, and B. Kanack, R.L. Carter, University of Texas at Arlington, Arlington, TX 75356 The challenges for future devices in compound semiconductor metallization lie in three areas: i) the reduction of contact resistivity; ii) improved control of the metallurgy of the interface between metals and semiconductor; and iii) movement away from thermodynamically-unstable systems. These goals can be accomplished by introduction of new processes, new materials and by better fundamental understanding of the electrical and metallurgical characteristics of metal-semiconductor interfaces. In this paper, we will report an order-of-magnitude improvement in the specifc contact resistance of gold-based ohmic contacts to p-type InGaAsP 1.3 11m wavelength. A novel technique was employed using Rapid Thermal Process with a silicon susceptor, Auger depth profile results revealed an abrupt interface between the reacted metal and the semiconductor for wafers alloyed by RTP, while smeared interfaces were observed for wafers alloyed in a conventional furnace. In addition, different interrnetallic compounds were observed. Implementation of this new RTP has resulted in lower series resistance and improved device performance.
(3:00 p.m_) UPRAC'l'ORY
ODIC
COIl'1'ACTS
POR InP-BASJ!lD OP'l'OBLBCTROJlIC DBVICES
A.kat., W.C.Dautr •• ont-saitb, B.W.ir, P.M.Tho ••• , S.H.G.Cbu, B.S.Cben and L.C.ki•• rlinq. lnP-based optoelectronic devices are strongly influenced, both in short and long term performance and reliability, by the quality of their ohmic contacts. The more sensitive contact is the one which is in the immediate vicinity of the active layer, typically the p-contact, and thus much more attention has to be paid to its properties. Correct design of the ohmic contact to these devices has to seriously consider all the following issues: selection of the metallization scheme to form a pure ohmic contact with the lowest possible contact resistance, creating a very highly thermodynamically and kinetically stable metal-semiconductor system over a large temperature region, tuninq the deposition and subse-
quent thermal process in order to introduce the lowest possible stresses into the thin metallic film. Subsequently, and however as important as the previous considerations, one has to define the process sequence which will yield the optimum performance of as many properties from the above mentioned list and successfully incorporate it into the overall device fabrication process sequence. In this talk we will demonstrate this concept, using the vehicle of the refractory metallization schemes such as Pt/Ti and W systems.
(3:20 p.m.)
BREAK
(3:30 p.m.) IN-SITU RAPID ISOTHERMAL PROCESSING WITH APPLICATIONS TO InP DEVICES R. SINGH, R.P.S. THAKUR, A. KUMAR, AND J. NARAYAN' SCHOOL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCE UNIVERSITY OF OKLAHOMA; NORMAN, OK 73019 PHONE No. (405) 325-4721; FAX No. (405) 325-2894 'MATERIALS SCIENCE AND ENGINEERING DEPARTMENT NORTH CAROLINA STATE UNIVERSITY; RALEIGH, NC 27650
ABSTRACT Coupling of the rapid isothermal processor to the deposition system provides additional advantages over the case where deposition and annealing are carried out in different systems. We have used in-situ rapid isothermal processing for in-situ rapid isothermal cleaning of loP substrates, in-eitu rapid isothermal annealing of II-A fluorides and annealing of ohmic contacts. Structural and electrical characteristics of Al-U-A fluoride-InP capacitors will be presented.
(4:00 p.m.) ELECTRONIC PROPERTIES OF MIS DIODES ON n- InP CONTROLLED BY SURFACE PREPARATION, W.A. Anderson and Y.S. Lee, State University of New York at Buffalo, Center for Electronic and Electro-optic Materials, Department of Electrical & Computer Engineering, 2l7C Bonner Hall, Buffalo, NY 14260. Undoped, n-type, (100) InP wafers were utilized in studying effects of different surface preparations on electrical properties of diodes having a structure Pd/40 oA oxide/n-InP/Au:Ge ohmic contact. Surface preparation was varied utilizing KOHmethanol etch, H2 S0 4 -HF etch, proximity cap during ohmic contact heating, thermal oxide or chemical oxide. The various techniques reproducibly controlled the conduction mechanisms as revealed by current-voltage-temperature (IVT) techniques. Regions of thermionic emission, thermionic field emission, and recombination are clearly revealed. These mechanisms may be correlated with properties of the oxide determined by angle - resolved ESCA. Surface preparation techniques may be utilized in alterning the Fermi level pinning. A barrier height of 0.86 eV has been obtained for the Pd/oxide/n-InP structure.
(4:30 p.m.) THE FORMATION OF ELEVATED BARRIER HEIGHT SCHO'ITKY DIODES TO InP AND InGaAs USING THIN, EXCIMER LASER DEPOSITED Cd INTERLAYERS. T.J Licata D.V. Podlesnik, R.M. Osgood, Jr., Microelectronics Sciences Laboratories, Columbia University, New York, NY 10027, and W.K. Chan, R. Bhat, Bellcore, Red bank, NJ 07701 Much recent attention has been paid to elevating the barrier height of contacts to InP and InGaAs via the formation of a thin intermediate layer between the semiconductor and a conventionally deposited highly conductive layer. Two significant examples are the introduction of a thin Cd-based layer through Langmuir-Blodgett deposition and the formation of a MIS structure through a UV-enhanced oxidation process. Furthermore, electroplated Cd has been shown to yield an unusually high barrier height to InP of == 0.6-0.7 ev. Here, we report on the use of thin (';;IOOA) excimer laser photodeposited Cd as an interlayer between these semiconductors and metal overlayers such as Au and other commonly applied metals. Previous work has shown that deep uv pulsed and cw lasers can be used to deposit photolytically high quality metal films from organometallic gas source molecules in a non-thermal manner. Specifically, evaporated Au-coaled Cd contacts fabricated through this process have resulted in barrier heights to moderately ndoped InP of == 0.7 ev and ideality factors of == 1.10, values similar to those commonly found for widely used GaAs schottky barriers. By contrast, Au contacts without a Cd interlayer result in barrier heights == 0.43 ev. To demonstrate the applicability of this approach, n-channel, depletion mode MESFETs have been fabricated. Discussion of the process is presented emphasizing the chemical and structural properties of the active interface and identification of factors important for optimization of device structures. (4:50 p.m_) AU/PD/ZN/PD OHMIC CONTACTS TO P-TYPE INP: DOllglas G Ivey and Ping lian, Deparunent of Mining, Metallurgical and Petroleum Engineering, University of Alberta, Edmonton, Alberta, Canada, T6G 2G6; Robert Bruce, Ben Northern Research, P.O. Box 3511, Station C, Ottawa, Ontario, Canada, KI Y 4H7. Ohmic contacts to compound semiconductors, and in particular to IHype loP, can be difficult to fabricate. A successful low resistance contact (rc< lxlO- 7 n-cm 2) to p-type GaAs using a Au/Pd/Zn/Pd metallization has been reported recently (1). The performance of a similar metallization to p-type loP is reported in this study. The contact structure was fabricaled by electron beam evaporation of metal layers on Zn-doped (2x10 18 cm- 3) <001> orienled InP substrates. The resultant metallization consisted of, in order of deposition, IOnm Pd, 5nm Zn, 20nm Pd and 60nm Au layers. Zn was chosen as the p-type dopant, while Pd aided in the wetting of Zn on loP and assisled the doping process. Au provided a highly conductive, nonoxidizing protective layer. The contacts were annealed in nitrogen gas at temperatures up to
4500 C. Contact resistance decreased with increasing annealing temperature, exhibiting ohmic behaviour arler annealing in the 400-4500C range for 15 s. The change in contact resistance was correlated to the corresponding changes in microstructure by means of electron microscopy and x-ray diffraction IeChniques. 1) R. Bruce, D. Clark and S. Eicher, Low Resistance PdlZnlPdlAu Ohmic Contacts to p-Type GaAs. presented at 1989 Electronic Materials Conference. Massachusetts Institute Of
Technology, Cambridge, Massachusetts, June 21-23,1989.
Annual Meeting-121
PROCESSFUNDAMENTALSAND MATERIALS PROCESSING III: Thermal and Mechanical Manufacturing Processes Sponsored by the TMS Process Fundamentals and Synthesis and Analysis in Materials Processing Committees Wednesday PM February 21, 1990
Room: Salon H
Marriott
Session Chainnen: S.C. Jain, GE Aircraft Engines, Cincinnati,OH 45215-6301; J. McKelliget, Department of Mechanical Engineering, University of Lowell, MA 01854
samples. Greensheets of a mathematical model was kinetics, the pore-diffuthe heat-transfer effect.
*Work supported by IBM under Contract No. F472U. (3,40 p.m.)
BREAK
(3: 55 p.m.) WANTED: A CENTER OF EXCEUENCE IN HEAT TREATMENf TECHNOLOGY: AH. Sonl, M. Mehta, and R. Naraln, Dept. of Mechanical and Industrial Engineering, University of Cincinnati, Cincinnati, OH 45221-0072. This paper presents the results of the activities undertaken by the University of Cincinnati during Project CT-ll supported by the Edison Materials Technology
Center (EMlEC), Kettering, Ohio. The project examines the feasibility of establishing a Center of Excellence addressing the needs of the Nation's heat treatment industry.
A
part of the task Involved conducting an Industry survey among Ohio heat treaters
to detennine the state of the art presently being practiced by Ohio heat treaters. Ohio leads the Nation In the number of heat treatment shops, and Is thought to be
a valid sample of the state of practice In the heat treating Industry In the Nation. The results of the Industry survey indicate that there Is a clear need for a National Center for Heat Treatment Technology. The future of the heat treatment Industry Is uncertain: The heat treaters face tough competition, and need to absorb and utUlze technological advances more effectively. The major technology transfer and application needs of the heat treatment Industry are In the areas of: (I) modeling and simulation to predict part distortion. microstructures and process conditions
(2:00 p.m.)
SOLID FREEFORM FABRICATION OF POWDER MATERIALS: U. Lakshminarayan, P. Haase, E. Nuss, C. Deckard, J. Barlow, D.L. Bourell, I. Beaman and H.L. Marcus, Materials Science & Engineering, ETC 5.160, The University of Texas, Austin, TX 78712-1063.
Solid Freeform Fabrication (SFF) is a new computer integrated fabrication technique that does not require any part specific tooling. The starting material can be either metallic, polymeric or ceramic. The powder process described in this paper involves laying down a thin layer of the powder and selectively sintering the powder according to the cross sectional pattern of the part being formed. The pattern is established with a computer controlled scanning laser beam. The part is built by successively laying down a number of layers and selectively sintering them. After laser treatment, secondary processing to maximize the properties of the final product may be required. The effect of material properties including particle size and composition, the optical properties of the powder and processing parameters like the laser energy, scanning speeds and ambient temperature on the density, defmition, strength and microstructure of the final product will be discussed. Partially funded by Texas Advanced Technology Program. (2: 25 p.m.)
A TAGUCHI APPROACH FOR FORGING PROCESS DESIGN R. Srinivasan, A. Chaudhary Universal Energy Systems, Inc. In place of factorial experimentation to achieve optimal forging process design, a Taguchi loss function minimization approach is recommended with the loss function evaluated through numercial computation in the
first determined using thin piepes of industrial size were then studied and formulated by combining the intrinsic sion and mass-transfer processes, and
first design phase and experimental data in
the subsequent design phases.
This approach is based upon the
observation that the numerical computations have now become true to nature and they are most suited for the initial wide range
evaluations which may potentially stretch the limits of the metalworking equipment. The loss function itself is defined using the dynamic material modeling technique which is based upon thermodynamic extremum principles.
50 p.m.) COMPUTER SIMULATION OF GLASS DEFORMATION DURING HOT PRESSING USING A PARTICLE METHOD: Ernesto Gutierrez-Miravete, Hartford Graduate Center, 275 Windsor Street, Hartford, CT 06120, and Gary Linsey, United Technologies/Pratt & Whitney, 400 Main Street, East Hartford, CT 06108. (2:
Glass components are formed by hot pressing in a single action press. To better understand the hot pressing process, it is necessary to investigate the flow behavior of the glass inside the pressing die. Deformation problems of this kind are usually formulated via continuum physics and solved using numerical methods. A convenient alternative are the so-called particle methods. In particle methods, the system is simulated by a collection of macroparticles subject to the laws of classical mechanics. This paper presents the results of using a particle method to represent the deformation behavior of glass during hot pressing. It is shown that a combination of results from the particle calculations with simple continuum mechanics ideas provides valuable insight into the deformation process. (3:15 p.m.)
THE REMOVAL OF CARBONACEOUS RESIDUE IN GREENSHEET PROCESSING FOR MULTILAYER CERAMIC MODULE:* D. R. Wall, IBM, East Fishkill Facility, Route 52, Hopewell Junction, NY 12533, and H. Y. Sohn, Department of Metallurgical Engineering, University of Utah, Salt Lake City, Utah 84112-1183. An experimental and mathematical investigation of the removal of carbonaceous residue deposited in ceramic substrates in the fabrication of multilayer ceramic modules has been conducted. Such carbonaceous residue formed in ceramic substrates when organic binder is pyrolyzed and removed from the modules during the firing of greensheets. Intrinsic kinetics of the gasification of carbonaceous residue in a mixture of steam with hydrogen was
through the appUcation of computer technology; (2) application of real-time quality control techniques (e.g., SPC); (3) Infonnatlon dissemination on technology relating to present and emerging heat treatment processes, matertals and equipment; (4) certification, education and training of personnel.
20 p.m.) PREDICTION OF THE INFLUENCE OF WATER SPRAY QUENCHING ON THE AGE HARDENABILITY OF ALUMINUM ALLOYS. J.-S. Kim and D. R. Gaskell,
(4:
School of Materials Engineering, Purdue University, West
Lafayette, IN
47907
Local heat transfer correlations to volumetric spray flux, representative drop diameter and drop velocity in quenching
water sprays applied to aluminum alloy 2024 have been developed for temperatures up to the Leidenfrost point on the boiling curve, and the C-curve for this alloy has been determined experimentally. The cooling curves for discrete locations in various cross-sections of water spray-quenched aluminum extrusions have been calculated by the finite difference technique and combination of these curves with the C-curve permitted calculation of iso-quench factor contours in the sections. The isoquench factor maps provide a quantitative relationship between the conditions of spray quenching and the hardness which can be attained by subsequent age-hardening heat treat-
ment of the alloy. (4: 45 p.m.)
MODELING AND CAE RESEARCH ACTIVITIES AT OHIO UNIVERSITY: Jay S. Gunasekera, Moss Professor of Mechanical Engineering, Dept. of Mechanical Engineering, Stocker Center, Ohio University, Athens. Ohio 45701. This paper highlights some of Ihe current research and development activities in the areas of Modeling and CAE in the department of Mechanical Engineering at Ohio University. A number of organizations (such as NSF, USAF, Edison, GE, Pratt & Whitney, Wyman Gordon. Ladish. GM·Allison. Schultz Steel) have sponsored research projects in this area. The main thrust has been in the areas of dynamic materials modeling, computer simulation of forming processes including the development of new computer codes or enhancement of existing ones and the CAD/CAM of dies. The materials under consideration range from AI2024 with SiC whiskers, Ti4242 to Gamma TiAi. It is concluded that the CAE approach to metal forming has helped to significantly reduce the lead-time taken from concept to manufacture by eliminating the traditional trial·and·error approach, help reduce defects such as laps and hence reduce the overall cost of manufacture.
REDUCTION TECHNOLOGY VI: Plant Matters and Alumina Sponsored by TMS Aluminum Committee Wednesday PM February 21, 1990
Room: Salon J & K
Marriott
Session Chainnan: Abbott Weatherly, Real Time Engineering Inc., 116 East Main St., Goldendale, WA 98620 (2:00 p.m.) HEAT TRANSFER BETWEEN MOLTEN AND SOUD CRYOliTE BATH: Y.R. Gan, Zhengzhou Light Metal Research Institule, Zhengzhou, China, J. Thonstad and A.Solheim, Laboratories of Industrial Electrochemislry, the Norwegian Institule of Technology, N-7034 Trondheim, Norway. The heat transfer coefficient between mollen and solid cryolite was measured by recording the temperature course inside a cold sphere which was immersed
122-Annual Meeting into the me~. The data obtained indicate that the measured heat transfer coefficients are higher than what can be calculated from dimensionless equations for convective heat transfer or from measurements in low-temperature physical models. Possible reasons for this are discussed. The heat transfer coefficient between bath and side ledge in aluminium cells is estimated to be in the range of SOO - 1000 W·m- 2-K'1. (2:30 p_m_) DENSITY, HARDNESS AND THEMAL CONDUCTIVITY OF HALL-HEROULT CRUSTS: M A Llayana; L.F. Verdeja; R. Zapico; F. Alvarez and J.P. Sancho, Departamento De Ciencia De Los Materiales De Ingenieria Metalurgica, Catedra Y Laboratorios De Matelurgia Extractiva, Universidad De Oviedo, Oviedo 13 33004 Spain. Density, hardness and thermal conductivity of synthetic and industrial Hall-Heroult crusts have been determined. Little differences of the properties have been observed. Crusts characteristics seems to depend on the type of alumina: primary and secondary. (3:00 p_m_) FORMATION AND CHARACTERIZATION OF ALUMINIUM ELECTROLYSIS CRUSTS: M A Llayona; L.F. Verdeja; F. Alvarez; ~P. Garcia Coque and J.P. Sancho, Departamento De Ciencia De Los Materials De Ingenieria Metalurgica, Catedra Y Laboratorios De Matelurgia Extractiva, Universidad De Oviedo, Oviedo 13 33004 Spain. A method to prepare synthetic Hall-Heroult crusts with different types of industrial aluminas have been developed, and the temperatures of formation have been determined. Discs of about 10inm thickness were cut and the composition, porosity, capillarity, bulk density and squeletal density were measured on each disc and on industrial crusts to compare the properties of synthetic and industrial crusts. (3:30 p_m.) MODERNIZATION OF MATERIALS TRANSPORTATION IN REDUCTION LINES: I::I...L.ac!lw: and H. Medina, Gerencia De Coordination Techologica, CVG Industria Venezolana de Aluminio CA, CVG-Venalum Apdo. 312, Puerto Ordaz, Edo. Bolivar, Venezuela At Venalum's reduction plant, until recently, the transportation of alumina, additives and bath materials to the pots was via potroom multipurpose cranes and buckets elevators, resulting in ore spills causing air pollution exposure, extra cleaning and increasing operation costs. This paper deals with the characteristical features of a dense phase neumatic conveying system for the transportation of primary and secondary alumina (crushed bath and additives) from the existing storage silos and the processing plant to the pots in the reduction lines, in which results in considerable savings on maintenance and operation costs of the aluminium production. (4:00 p_m_) CONVERSION FROM OPEN SlDEWORKED TO HOODED POINT FEEDER POTS AT ALBRAS-ALUMINIO BRASILEIRO SA: J R D de Carvalho, H.R. de Freitas and J. Bicalho, Albras, Barcarena PA. This paper describes the technical studies made on pot heat balance, pot lining design and pot operating parameters necessary to modemize the no. 1 Albras potline aimed at increasing current efficiency and production, lowering specific energy consumption, reduce fluoride emissions and improving working conditions. Comments are made on the method of substituting the pot superstructure without cooling the pot, the overall work schedule and the operational results obtained so far. (4:30 p.m_) THE SELECTION OF BUS-BAR ECONOMICAL SECTION IN HIGH AMPERAGE CELLS: yao Shjhyan and He Zhihui, Gui Yang Aluminium and Magnesium, Design and Research Institute, Gui Yang, Peoples Republic of China. Although the study of bus-bar system, no matter what is the single or double end riser cell under 170KA , has been remained in many papers. With the development of IS0-300KA high amperage cell in later years, there exists mainly three problems in bus-bar design system. 1) The optimum design of magneto-hydrodynamics. 2) the most reasonable short-circuit design and 3) the most economical bus section design. The above three questions are all very complicated. Very few of this kind of report can be seen to the second and third question before 1987. The analysis and classification of bus-bar arrangements is included in this paper. Through which a new calculating method of bus-bar section has been derived and a program that it is call GYBUS has been performed. GYBUS is suitable for all kinds of high amperage cells which can be used to calculate bus-bar section, weight, voltage and current density and so on.
break pots originally designed for 150 KA. The cathode side lining was replaced with thinner silicon carbide wall and the bath composition modified to higher aluminium flouride. The combination of changes had enabled operation at 175KA with substantially improved current efficiency and a considerable reduction in carbon and energy consumption. The development effort is now directed to demand feed control, rod-stub design and an improved cathode.
REFRACTORY GOLD, SILVER & PRECIOUS METALS PROCESSING II: New Process Methods for Complex Gold/Silver Ores Sponsored by the TMS Precious Metal Committee Wednesday PM February 21, 1990
Room: Salon 1 Marriott
Session Chainnen: M. Jennings, Vice President, Echo Bay Management Corp., 5270 Neil Road Suite 101, Reno, NV 89502; M. Fuerstenau, University of Nevada-Reno, Department of Chemical and Metallurgical Engineering, Mackay School of Mines, Reno, NV 89557-0047 (2:00 p.m.) AFFECT OF SULFUR GRADE ON PRESSURE OXIDATION AUTOCLAVE SIZING: M. H. Conway, Davy McKee Corporation, 2430 Camino Ramon, Suite 100, San Ramon, CA 94583, D. C. Gale, Davy McKee Corporation, 2430 Camino Ramon, San Ramon, CA 94583. Pressure oxidation has become a widely accepted technology for treatment of the refractory gold ores. This paper examines the affects of sulfur grade in the feed on sizing of the autoclaves. Major factors which must be considered in sizing of the autoclaves include new feed rate, sulfur grade, residence time, temperature, pressure, and pulp density. In determining the autoclave size, the total amount of sulfur contained in the feed is by far the most important factor, even more important than the total tonnage of new feed. This apparent anomaly in sizing is readily explained by thermodynamic and kinetic considerations.
(2:30 p.m.)
PROCESS OPTIONS FOR REFRACTORY SULFIDE GOLD ORES; TECHNICAL, ENVIRONMENTAL AND ECONOMIC ASPECTS. A. Kontopoulos, National Technical University of Athens, GR-106 82 Athens, M. Stefanakis, METBA SA, GR-1J6 35 Athens. Refractory sulfide gold ores or concentrates require an oxidation pretreatment step to yield the contained gold amenable to the subsequent leaching step. The process options for the oxidation pretreatment can be broadly classified as follows: roasting, pressure oxidation, bacterial oxidation, chemical oxidation, each with a number of alternative suboptions. Roasting and pressure oxidation are being applied on an industrial scale; the stage of development of the other options ranges from preliminary laboratory to advanced pilot-plant. In this paper, the major technical, environmental and economic aspects of the above process options will be discussed, with specific reference to the Olympias and other concentrates. Future industrial trends will also be considered. Particular emphasis will be placed on the treatment of arsenical ores and concentrates, where safe plant operation as well as handling and disposal of the arsenic containing products and/or effluents and residues is of paramount importance. (3:15 p.m.)
GOLDOX - RECENT FINDINGS IN GOLD PROCESSING TECHNOLOGIES, J.R. Dale, Airco Industrial Gases, 575 Mountain Avenue, Murray Hill, New Jersey 07974 Oxygen is just beginning to be appreciated for its capabilitie4 to aid in the processing of gold ores. Historically, oxygen has been regarded only as an elemental gas to be used to increase the flame temperature in melting processes. Recent equipment and processing techniques involving wet oxidation processing know-how has resulted in cost effective techniques for increasing ore yields. Following extensive efforts to develop this know-how, Airco's sister company AFROX, working in South Africa, has developed and quantified the Goldox Process. The paper will review and compare this technique with some of the more recent techniques developed that utilize oxygen's chemical processing features to advantage.
(5:00 p_m_) POT RETROFIT WITH LARGER ANODES: R. Dharneja & G.S. Sachan, Dubai Aluminium Co., Ltd., P.O. Box 3627, Dubai, U.A.E.
(3:45 p.m.) ENERGY REQUIREMENTS FOR PRESSURE OXIDATION, P.G. Mason, Wright Engineers Ltd., 1444 Alberni Street, Vancouver B.C. V6G 2i4
The Anode size was increased nearly twenty percent in the Three Pot lines of DUBAL which have prebaked centre
There are now several operating Pressure Oxidation plants in the gold mining industry and several more in the design stage. Feed
Annual Meeting-123 material varies from ore containing less than 1% sulphide sulphut to concentrate containing greater than 20%. Energy is supplied to these plants in the form of steam, compressed oxygen and electrical power. This paper describes how the energy requirement relates to sulphide level in the feed material, the extent of oxidation, the retention time, etc., and proposes measures that may be taken to save energy. 14:30
p.m.)
THE USE OF FERROUS SULFATE FOR CYANIDE DESTRUCTION IN HEAP LEACH OPERATIONS: M. Drozd, Gwalia USA, Ltd., World Trade Center, 1675 Broadway, Denver, CO 80208. (ABSTRACT WILL APPEAR IN ADDENDA ERRATA)
ppm. Therefore, repetitive recycling of uranium scrap can result in significant levels (hundreds of ppm) of carbon buildup. This results in more numerous carbide inclusions which can have a deleterious effect on mechanical properties. Hafnium carbide has a very low free energy of formation compared to most other carbides, and is anticipated to be less dense than the uranium carbides that are formed during the casting process. Hence, the proper level of hafnium additions may be able to react with the carbon and "float" up, and therefore, reduce or even eliminate the carbon during the casting process. This work is evaluating the effectiveness of using hafnium additions to reduce these carbon levels in the resultant casting. The results of these findings will be discussed. *Operated by Martin Marietta Energy Systems, Inc., for the U.S.
Department of Energy under contract DE-AC05-840R21400.
SOLIDIFICATION PHENOMENA II
(3:20 p.m.) 'MO'IJ'ELI'J{fj O:r '1I...'E5l'IJWI.L 'E{51.S'IfC S'I'1I...51.f'l{ 1JfS'I'1I...f'B'll'IfO'J{S f'l{ P5l.'1('If5l.L 'P'E'l{'E'I'R..51.'IfiJ'J{ (j5l..S 'I'll'l{(jS'I'E'l{ 5l.'I\C WEL1JS: K. W. Mahin
TMS General Abstract Session Wednesday PM February 21, 1990
Wm. S. Winters. Jr., J. S. Krafcik. Sandia National Laboratories. Livermore. California 94551; T. Holden, R. Hosbons. and S. MacEwen, Chalk River Nucl. Labs. Chalk River. Ontario. Canada KOJIJO
Room: Salon G Marriott
Session Chairman: Dr. Hani Henein, Department of Mining, Metallurgy & Petroleum, University of Alberta, 606 Chemical-Mineral Engineering Building, Edmonton Alberta T6G 2G6 Canada (2:00 p.m.) ROLE OF ATOMIZING GAS ON CHARACTER OF RAPID-SOLIDIFIED MAGNESIUM ALLOY POWDERS: A. Joshi, Lockheed Palo Alto Research Laboratory, 3251 Hanover St, Palo Alto CA 94304, R. David Adamson, Tracor Aerospace Technology Ctr, Bollinger Cyn Rd, San Ramon, CA 94583 Gas atomization procedures to produce magnesium alloy powders with uniform powder particle shape, size distribution, and oxide coating thickness have been developed. High-pressure inert Argon gas was used to produce powder particles with a controlled rate of cooling. In order to model newly developed alloys and determine Rapid-Solidification effects, atomization trials with AZ9l and ZK60 alloys were conducted using Argon and Helium gases (for different rates of cooling). Powder samples were characterized using SEM, AES, DSC, and metallography to understand powder surface topography, type and thickness of oxide coating, bulk chemistry and microstructure, and to compare the different RapidSolidification cooling-rate effects of both Argon and Helium atomization gases.
[2:20 p.m.) THE
GROWTH
RATE DEPENDENCE OF MERCURY CADMIUM
TELLURIDE,
~
McKeagney, R. N. Andrews, S. D. Walck, University of Alabama at Birmingham, Department of Materials Engineering, UAB Station, Birmingham, AL 35294. The dependence of the defect density of Bridgman grown Hg Cd Te (x;O.2) upon growth rate haa been studied. 1 DisI~ca~ion etch pits and sub-grain boundaries have been counted from defect etched slices of crystalS grown at 0.3 ~m/s and 0.6 ~/s. Quantitative determinations were used to assess growth rate effect8. Crystallographic orientation information was obtained to ensure comparability of counts from different slices.
(2:40 p.m.) DIRECT OBSERVATION OF SOLIDIFICATION PROCESSES SYSTEMS. R. A Merrick and J. B. Andrews. Materials
Engineering,
Birmingham. AL 35294
University
of Alabama
IN
samples systems.
t Work supported under U.S. Dept. of Energy Contract DE-AC04-76DPOO78. (3:40 p.m.) SOLIDIFICATION CRACKING IN THE TITANIUM ALLOY WELD METALS: H. Inoue and T. Ogawa, R&D Laboratories-II, Nippon Steel Corporation, Sagamihara, Kanagawa, 229, Japan Susceptibility to weld solidification cracking of three kinds of Ti-alloys was investigated by Trans-Varestraint test. Weld crackings did not occur in Ti-6AI-4V at 0.5% augmented strain, but significantly in Ti-6AI-6V-2Sn, and Ti-15V-3AI-3Cr-3Sn, which was actually the most susceptible to cracking. Through close examination of weld solidification microstructures by tin-quenching, conspicuous segregations of V, Fe, Cr, Cu were confirmed at dendritic·grain boundaries during solidification in each alloys. In Ti-6AI-4V, these compositional microsegregations in weld metals, were readily decreased and eliminated by the occurrence of solute diffusions at very early stage of weld cooling, while that in other alloys remained till rather lower temperatures to cause crackings due to liquid films of low-melting constituents. This segregation behaviour was the most remarkable in Ti-15V-3AI-3Cr-3Sn weld metals.
SUPERPLASTICITY IN AEROSPACE V, 2nd International Symposium
Sponsored by the TMS Shaping and Forming Committee
IMMISCIBLE
Department of at Birmingham,
There has been considerable discussion in the literature on solidification processes in immiscible (hypermonotectic) alloy systems. Recent results obtained by the author on directionally solidified immiscible alloys indicate structures are often obtained that cannot be easily explained using current theory. In this investigation, solidification processes are investigated by directly observing the events taking place at the interface. This is accomplished through directional solidification of immiscible transparent analogue samples using a temperature gradient stage fitted to an optical microscope. Results will be presented showing the differences between solidification of
hypermonotectic miscibility· gap
The residual elastic strain distributions in a stationary partial penetration gas tungsten arc weld on 304L have been calculated using finite element techniques and experimentally verified using neutron diffraction. The experiment was modeled using a fully-coupled 2-D thermomechanical finite element code. called PASTA2D. Evaporation effects from the arc were included in the heat flux model, in addition to radiation and natural convection effects. However. heat transfer due to convective flows in the molten weld pool region were neglected. The mechanical response of the material during welding was modeled using a visco·elasto-plastic constitutive model. which accounted for both kinematic and isotropic hardening as a function of temperature and strain rate. The residual elastic strains calculated using this model agreed very well with the neutron diffraction measurements. The results fell within the scatterband of the experimental data for all three components of strain. i.e., radial. hoop and normal. This presentation will compare the experimental and numerical results and discuss the implications of the strain distribution on weld performance.
in high miscibility·gap and Early attempts at modeling
low the
solidification process will be discussed.
3:00 p.m.) CARBON REDUCTION IN URANIUM ALLOYS UTILIZING HAFNIUM ADDITIONS: G. Mackiewicz Ludtka, Oak Ridge Y-12 Plant*, P.O. Box 2009, Oak Ridge, TN 37831-8096. Carbon pickUp in uranium alloys melted in yttria-coated, graphite crucibles in on the order of 25 to 100
Wednesday PM February 21, 1990
Room: Garden Rooms 1-2-3
Inn at the Park
Session Chairman: T.R. McNelley, Department of Mechnical Engineering, Naval Postgraduate School, Monterey, CA 93943 (2:00 p.m.) PART SELECTION CRITERIA AND DESIGN CONSIDERATIONS FOR THE USE OF SUPERPLASTICALL Y FORMED ALUMINUM FOR AEROSPACE STRUCTURE: J. M. Story. Alcoa Laboratories. Alcoa Center, PA 15069
Superplastic forming (SPF) has been accomplished by using gas pressure to blow a hot sheet into a female or over a male fonn, in a process very similar to the vacuum forming of plastics. It is characterized by extreme ductility at high forming temperature and slow forming rates. The high ductility often allows structures with designed·in strength and stiffness to be produced as a single component. rather than as built up structure consisting of many details
124-Annual Meeting. assembled by riveting, to result in a component with equivalent functionality. Superplastic forming methods, candidate part selection crit~ria , superplastic alloys and their properties and example applications are reviewed. The differences between forming aluminum and titanium alloys and the differences between fonning for S01Jctural and non-s01Jctural applications are highlighted. (2: 30 p . m. ) FABRICATION OF STRUCTURAL COMPONENTS FROM COMMERCIAL ALUMINUM ALLOYS USING SUPERPLASTIC FORMING_ S J Hales, AS&M; W. F. James, PRC; J. M. Shinn and Thomas T. Bales, NASA Langley Research Center, Mail Stop 188A, Hampton,
VA 23665-5225.
Structural components possessing relatively simple geometries were designed and the pressure-time profiles for superplastic forming of the parts using various commercial alloys were determined. The maximum pressure limitations of the superplastic forming facility dictated that strain rates with acceptable corresponding flow stresses had to be employed. It was established that higher forming rates than the optimum strain rates selected from prior tension tests could be used, providing that a post-forming pressure cycle was imposed on the material at the completion of fonning. ModifIcation to the initial forming cycles, tool gcanetty and post-fonning procedures were made to minimize cycle time and improve final part quality simultaneously. The effects of using thermomechanically processed material which was statically recrystallized prior to forming compared to material designed to dynamically recrystallize during forming were assessed. The fabricated parts were evaluated in terms of cavitation, thickness variations, microstructural homogeneity and mechanical properties. (3:00 p.m.) THE MECHANICAL PROPERTIES OF SUPERPLASTICALLY FORMED ALUMINUM-LITHIUM ALLOYS AT CRYOGENIC TEMPERATURES: E..L.. Bradley III, R. A. Emigh, and J. W. Morris, Jr., Department of Materials Science and Mineral Engineering, University of California, Berkeley and Center for Advanced Materials , Lawrence Berkeley Laboratory, Berkeley, CA 94720.
The properties of aluminum-lithium alloys are very promising in terms of their specific strength, low temperature toughness, and their ability to be superplastically formed (SPF) . However, until recently very few data have been reported on the properties of these aerospace alloys after superplastic forming. This study examined the tensile and toughness properties of SPF alloys from several AI-Li-X systems at room and cryogenic temperatures. The alloys tested included both commercially produced aluminum-lithium alloys and new alloys produced and processed on site. The results for the SPF alloys are compared to earlier results for the wrought varieties and differences between the two forms will be discussed. (3:30 p.m.) THERHOHECHANICAL PROCESSING OF SPRAY CAST 7475: N. Ridley , Haterials Science Department, University of Hanchester and UHIST, Hanchester, HI 7HS , U. K. (Abstract not available) (4 : 00 p.m.) SUPERPLASTICITY OF MICRODUPLEX Ti-6Al-4V : Mike Meier and A.K. IllllMukherjee, University of California at Davis, Department of 1IIIIMechanical Engineering, Division of Materials Science and Engineering, Davis, CA 95616.
····The deformation behavior of Ti -6AI-4V has been investi9~ted iiiiibetw~~n ~15·C and 925·C and true strain rates from 2xlO to "",2xIO s ~3 D!1location creep was observed at strain rates ..... above 2xl0 s. At lower strain rates the deformation I l11behavior spl it at around 850·C, probably due to changes in 1IIIIbulk behavior caused by changes in the volume fraction of beta IIIlIphase . Superplastic behavior was observed above 850·C and was ""'characterized by relatively high volume fractions of beta and strain hardening beyond the UTS . AnalysiS of the strain IIl11hardening behavior, assuming grain growth as the major cause ····for hardeni ng, suggests that the apparent act i vat i on energy, -oa ' is a combination of that for the deformation mechanism and !!!grain growth and ideally would be 1.5 times that for self =diffusion . Q measured using the UTS varied from 1.3 to 1.5 IQ j • However, ~a measured from hardening rates was found to be COnsiderably higher.
A new thermomechanical treatment of high strength aluminum alloy is proposed . The new process (TMPA) include s IMP plus Artificial Aging . The difference in superplastic mechanical behavior between THPA and IMP process is compared. Adding art i ficial aging to the thermomechani cal treatment can i mprove the s uperplasticity of the alloy. The maximum superplastic elongation increases from 1000% to 2300% by adding artificial aging to THP process. The microstructure
changes
during
superp!astic
deformation shows that adding artifical aging process can improve the distribution of the second 'phase particles and the rate of grain
growth
elongation .
decreases
which
leads
to
the
increase
of
Wednesday PM February 21, 1990
Room: California lll-IV Quality Hotel
Session Chainnan: P.K .Gallagher, AT&T Bell Laboratories, MUlTay Hill, NJ 07974 (2 :00 p.m . )
(Invited)
APPLICATION OF THERMAL ANALYSIS TO PHASE DIAGRAMS, R. D. Shull , National Institute of Standards and Te chnol ogy, Bldg 223, Rm B150 , Gaither sburg, MD 20899. The e x istence of certain phase changes i n various alloy systems at elevat ed temperatures may be determine d by conventional thermal ana l y tical techniques. However, the construction of phase
diagrams for these systems also requires a knowledge of the types
of transitions and the temperatures of transition. These quantit i es are much more difficult to obtain from the measured
thermal data. Methods for performing such evaluations are presented here along with a heat flow analysis for a typical differential thermal analysis device . A br i ef description of the Kiss i nger method for determining the kinetics of a phase trans format i on from thermal analysis data will be given along with information on its limitations . Both the Ti -AI and Ti-Cu s ystems will be de sc ribed as example sys t ems where in thermal analytical techniques were especially useful in the clarification of certain portions of their phase diagrams.
(2 :30 p.m.)
Invited THERMAL ANALYSIS OF SOLIDIFICATION KINETICS, L...lL Perepezko, Univ. of Wisconsin, 1509 University Ave .. Madison, WI 53706.
Measurements of heat effects are used as a common tool in monitoring and controlling solidification processes and microstructure development. Almost every form of thermal analysis is used, but the most common is the deceptively simple cooling curve. Even when care is taken to insure an isothermal sample and uniform cooling, the interpretation of thermal signals during solidification to understand the kinetics is difficult in bulk samples. The heterogeneous nature of solidification in macroscopic samples yields conditions which preclude the isolation and separate analysis of nucleation and growth processes in a reliahle way. When bulk samples are subdivided into five droplets large undercoolings often result with a sin~e nucleation event in each droplet. Since crystallVowth is rapid, the rate limiting step is nucleation. An analysis of the crystallization rate of droplets with known size distribution provides a means to distinguish different nucleation kinetics and to interpret some of the nucleatioh catalysis mechanisms. With droplet samples other thermal measurements are also available for thermodynamic properties of stable and metastable solidification products. The support of the ARO (DAAL03-86-K-0114) is gratefu!!y acknowledged. IInvltedj
STUDIES OF METASTABLE HETALLIC PHASE FORHATION BY THERMAL ANALYSIS . z . Altounian and J.O . Strom-Olsen Ph y sics Department, McGill Uni v ersity, 3600 Uni v ersity St., Montreal, Quebec, Canada H3A 2T8.
THE EFFECT OF ARTIFICIAL AGING AN SUPERPLASTIC BEHAVIOR OF HIGH STRENGTH ALUHINUM ALLOY: Jiang Xinggang, Wu Qingling, Chui Jianzong and Ha Longxiang , Department of Hetal Forming , Northeast University of Technology , Shen Yang , P R C
of
Sponsored by the TMS Chemistry and Physics of Materials Committee
(3 :00 p.m.)
(4:30 p.m.)
observation
THERMAL ANALYSIS OF METALLURGICAL SYSTEMS n: Phase Transformations
the
Hetastable phases can have crystalline or amorphous structures. Differential scanning calorimetry (DSC), is a simple, fast and effective technique in characterizing the amorphous phase. With proper use, irreversible and reversible structural relaxation effects can also be studied. For crystalline metastable phases DSC provides important information on the nucl e ation and growth kinetics and the degree of stability. For magnetically ordered phases, the use of thermomagnetic analysis (thermogravimetric analyzer, TGA, in a magnetic field) provides fast characterization of the magnetic phase(s) present . A properly cal i brated TGA can also provide quantitative information. This technique has remarkable sensitivity, and in term s of speed and reliability easily surpasses :flrgl~~J~~cal instrument including th. electron
(3 :30 p.m.)
(Invited)
SCALING CONCEPTS APPLIED TO CRYSTALLIZATION KINETICS OF HETALLIC AND POLYMERIC GLASSES: ~ and M. Pont, Dept. of Solid State Physics, Royal Inst. of Technology, S-100 44 Stockholm, Sweden. The metastability of amorphous or glassy systems (like metallic glasses , polymers etc.) and the manner in which they return to
Annual Meeting-125 equilibrium have been commonly studied using thermal analyses techniques. Usually in analyzing the results classical Kissinger unphysical values for the activation energies, growth-velocity
THERMODYNAMICS AND PHASE RELATIONS OF EXOTIC MATERIALS IV: Experimental Studies
techniques, the crystallization phenomena in a number of metallic
Sponsored by the ASM-MSD Thennodynamic and Alloy Phases Committee
or Johnson-Mehl Avrami approaches have been adopted resulting in
prefactor etc. We demonstrate that development of order and the dynamic processes in systems quenched far from equilibrium can be described in terms of Universality and Scaling concepts. From experimental studies extensively using thermal and magnetic glasses are shown to follow a time-scaling universal function with a power-law behavior. Thus it appears that the dynamic process in the glass phase can be described in terms of computer simulated models for quenched Ising systems in the so-called conserved order-parameter case.
Invited) (4:00 p_m_) KINETICS OF CRYSTAL-TO-AMORPHOUS TRANSFORMATIONS IN METALLIC ALLOYS:* R, B, Schwarz, Center for Materials Science, Los Alamos National Laboratory, Los Alamos, NM 87545. Amorphous alloys of early- and late-transition metals can be synthesized in powder form by ball milling either (a) a mixture of elemental powders or (b) powders of intermetallics. We used differential scanning calorimetry to measure the volume fractions of amorphous phase, the crystallization temperatures, and
the enthalpies of crystallization as a function of ball-milling time. These measurements, combined with structural characterizations, show that for a given composition the amorphous alloys
prepared by the two techniques are identical.
We discuss models
for these solid state crystal-to-amorphous transformations.
*
Work supported by the U. S, Department of Energy, Office of
Basic Energy Sciences.
(Invited) (4:30 p_m.) TRANSIENT PHENOMENA FOLLOWING PRECIPITATION AND RE-SOLUTION ANNEALS IN ANI-RICH NiTi ALLOY: L. V. Meisel and P. J. Cote, Research, Development and Engineering Center, Close Combat Armament Center, Benet Laboratories, Watervliet, NY 12189-4050_ Transient phenomena following anneals in Ni50.5Ti49.5 are investigated by means of Differential Scanning Calorimetry (DSC). Precipitation anneals (350"C) raise the temperature range of the martensite and R-transformation.
Subsequent
cycling through the transformation range (-40 to 100"C) produces small changes in the martensite DSC peaks and no observable change in the R peak. Subsequent re-solution anneals (600"C) produce dramatic effects: Ms rises
approximately 20°C above its value preceding the precipitation anneal and the martensite DSC peak is very sharp and symmetric,
while the R peak is not observable_
During ensuing cycling,
the martensite peak returns to its original form and position
after about 300 cycles and the R peak appears after about four cycles and is invariant to cycling.
with a dislocation model. (5 :00 p.m_)
The data are consistent
(Invited)
APPLICATION OF DSC TO THE STUDY OF NI-TI SHAPE MEMORY ALLOYS: Wejkang Yu and T.w. Duerig, Raychem Corporation, 300 Constitution Drive, Menlo Park. CA 94025 Shape Memory Alloys have the ability to be deformed at 10Yolemperature and then revert back to their original shape upon heating. This effect involves a martensitic phase transformation in which thespecific volume change between the austenite (the high temperature phase) and the martensite (the low temperature phase) is minimal. When deformed in the martensite condition. the deformation strain can be accommodated through a mechanical twinning process. The austenite phase cannot retain these twins and upon heating must return to the origin shape. Although there is little volume change during this transformation, a Significant enthalpy change accompaniesthese solid state phase transformations. For near equiatomic binary Ni-Ti alloys, there is also a rhombohedral pre-martensitidransformation and the DSC shows a peak for each step in the austenite: (B2) ~ R-phase ~ martensite transformation. The R phase can be suppressed by adding copper. With more than about 5 at.% copper addition, the transformation changes into B2~B19~B19·. The DSC results will be compared with mechanical and electrical transformation measurements and the different phase transitions will be identified.
Wednesday PM February 21, 1990
Room: Plaza C Inn at the Park
Session Chairman: Martyn G. Adamson, Lawrence Livermore National Laboratory, Livermore, CA 94550 (2:00 p.m_) EXOTIC KINETICS OF DISORDER -) ORDER 1RANSFORMATIONS B. Fultz, 138-78. Caltech, Pasadena, Calif. 91125 Intuitively, we expect most kinetic phenomena to obey the systematics proposed by Onsager; for a disorder -) order transfonnation we expect the rate of change of an order parameter to be proportional to a mobility times the dependence of the free energy on that order parameter. Assuming a fixed mobility, we then expect the order parameter to change monotonically with time (with an exponential relaxation). I am making a theoretical study of the kinetics of ordering by a vacancy mechanism. Here the kinetic equations are inherently nonlinear, and some exotic phenomena are observed. For example, vacancies may prefer a specific sublattice, and their relaxation to this sublattice may occur on a different time scale than the ordering of the solutes themselves. Such changes in vacancy populations alter the kinetics of the disorder -) order transfonnation in non-intuitive ways. More exotic nonlinearities occur for ternary alloys.
This work was supported by the U. S. Dept. of Energy under contract # DE-FG03-86ER45270.
(2 :30 p_m.) PHASE TRANSITIONS IN "PLASfIC CRYSTALS": D. Chandra. W. Ding and R. A. Lynch, Dept of Chem. and Met. Engineering. University of Nevada, Reno, NY 89557 and J. J. Tomlinson. Oak Ridge National Laboratory, P. O. Box Y. Oak Ridge. TN 37831.
"Plastic Crystals" which have hydrogen bonded globular molecules can be used to store latent heat in high temperature gamma phases. In these materials the solid-solid phase 0
transitions occur between -33 C to ISSoC with transitional enthalpies in the range of NI0 to N76 cal/gm. The transformation temperatures can be adjusted by the addition of dopants. Binary phase equilibria between high and low temperature phases for selected pairs of polyalcohols such as pentaerythritol. pentaglycerine and neopentylglycol: anisotropiC thermal expansions in low temperature phases and isotropiC behavior in the high temperature phases will be discussed.
Thermal energy storage aspects in supercooled
phases in 2-amino-2-methyl- 1. 3-propanediol amine will be
also presented.
(3:00 p.m_) METASTABLE PHASE RELATIONSHIPS OF IOOSAHEDRAL ALLOYS: .J!,...JL Follstaedt and J. A. Knapp. Sandia National Laboratories. Division 1112. Albuquerque. NM 87185
Phases whose diffraction patterns exhibit icosahedral symmetry belong to a new class of solids termed "quasicrystals··. We have used surface modification techniques to form layers of icosahedral AI-Mn. and we have studied their microstructures with rEM as a function of concentration and temperature. In alloys formed in the solid state by ion-beam mixing we find that the minimum Mn concentration of icosahedral AI-Mn in equilibrium with fcc Al is 15±1 at.X at 150 C. By rapidly' heating a layered icosahedral alloy on sapphire with an electron beam to known temperatures. the icosahedral phase with 20 at .X Mn is inferred to melt congruently at 910±2O C, which is N30 C below the melting points of crystalline AI-Mn phases with similar compositions. The liquidus of the metastable decagonal phase was also determined. These results have enabled us to evaluate the metastable equilibria between the icosahedral phase, liquid, and other solid AI-Mn phases. Work supported by U.S. Department of Energy. DE-AaH-76PDOO789
126-Annual Meeting
TITANIUM I: Advances in Alloys/Processes Sponsored by the TMS Titanium Committee Wednesday PM February 21, 1990
Room: ParkA Inn at the Park
Session Chairman: R.R. Boyer, Boeing Commercial Airplanes, P.O. Box 3707, MS 73-43, Seattle, WA 98124 (2:00 p.m.) BETA 21S: A HIGH TEMPERATURE METASTABLE BETA ALLOY P. J. Bania and W. M. Parris, Henderson Technical Laboratory Henderson, Nevada 89015 Because of their strip producibility (cold rollability), cold formability and age hardening response, ,qIetastable beta titanium alloys are particularly attractive candidates for sheet metal applciations. However, one of the principal drawbacks to such alloys is that, because of the alloying elements used to stabilize the beta phase they are generally quite poor in oxidation resistance and are generally not useful in 'warm' applications, above roughly 600 o F. A new metastable beta alloy has been developed - designated Beta 21S - which offers the normal attributes of such an alloy, but in addition offers oxidation resistance and Significantly enhanced elevated temperature properties. The background development of this alloy will be reviewed, along with properties at elevated temperature, including oxidation resistance. (2:30 p.m.) RECENT DEVELOPMENTS IN Ti-15V-3Cr-3Al-3Sn CASTING TECHNOLOGY Robert M. McKenzie, TiTech, Pomona, CA 81769-3060 Beta titanium alloys have held promise because of their ease of reaching high aged strength levels resulting in specific strengths in excess of one million. Ti-15V-3Al-3Sn-3Cr is one such alloy. This paper explains that excellent properties can be attained in parts cast in this particular alloy. Outstanding aging response from the as-HIPped condition were found in this study. The exceptional strength attributes afforded by selected heat treatments were evaluated by measuring tensile, fatigue and fracture toughness properties. Processing parameters such as straightenability, chem milling, and vacuum degassing during aging are also addressed. The work discussed is a continuation of a previous study conducted in 1986. This study confirms the impressive properties found in this earlier work and added knowledge which should support its use as a weight saving alternative to precipitation hardened steels. (3:00 p.m.) CERMETi TITANIUM MATRIX MICRO-COMPOSITE AND "CM3C" MACRO-COMPOSITES. S.•__ .AbkQ.\OI.itz. Dynamet Technology, Inc., Eight A Street. Burlington MA 01803. The CHIP (Cold and Hot Isostatic Pressing) process for titanium blend powders to near net shape is an economically advantageous production process for many critical components. This technology forms the basis for a new approach to metal matrix composite (MMC) manufacture which overcomes many of the shortfalls (directionality, geometric, economic constraints) typically associated with other MMC materials, and more importantly MMC components. CermeTi, a new family of titanium alloy micro-composite materials using TiC particulate as the reinforcement, provides higher modulus and improved high temperature performance over the monolithic alloys while main-taining the lightweight advantage. The level of TiC reinforcement can be selected to tailor the properties to meet specific applications. The ability of these new materials to be diffusion bonded to a monolithic alloy of the matrix composition offers an additional dimension in desigo flexibility. Desigoated "Common Matrix Micro-Macro Composites" (CM3C), materials exhibiting dual properties can be produced. This technology can be extended to other matrices. and other reinforcements. (3:30 p.m.) BREAK (3:40 p.m.) FRACTURE MECHANISMS IN HIGH STRENGTH BETA TITANIUM ALLOY CASTINGS: W.J. Porter, D.J. Evans, D. Eylon, University of Dayton, 300 College Park, KL407, Dayton, OH 45469; W.J. Barice, PCC, 4600 S.E. Harney Drive, Portland, OR 97206; and R.R. Boyer, Boeing Commercial Airplane Co., Mail Stop 73-43, PO Box 3707, Seattle, WA 98124 Beta titanium alloy castings are, currently, considered as replacement to high strength corrosion resistant steel castings for airframe applications. This is, primarily, due to their high strength-to-density ratio, excellent corrosion resistance, high fatigue strength, and relatively low cost when compared to
wrought products. In the present study, the tensile and fatigue behavior of Ti-15V-3Al-3Sn-3Cr investment castings were investigated. The hot isostatically pressed (HIP' d) and solution treated and aged cast parts displayed tensile and fatigue strengths above those commonly found in wrought alpha+beta titanium alloys. The nature of the tensile and fatigue fractures were studied in detail. Grain boundary alpha phase was identified as having a major roll in determining the tensile ductility and the fatigue crack initiation. (4:10 p.m.) ISOTATIC FORGING OF P/M AND CAST TITANIUM PARTS: R. M. Conaway, 6175 Shamrock Ct Dublin, OH 43017. Hot isostatic pressing (HIP) of a wide range of cast and powdered materials is carried out to produce fully dense components. These components exhibit markedly improved dynamic properties. The HIP process is used widely in production for aircraft and other high value castings. Isostatic forging is a version of HIP where the pressures are high enough to effect consolidation or densification essentially instantaneously. Isostatic forging cycles are 100 to 1000 times faster than conventional HIP. This paper will examine the development of isostatic forging and the very favorable economic implications of this faster process. Data concerning its application to titanium P/M and titanium castings will be presented. (4:40 p.m.) Ripple Load Cracking Sensitivity in li-6AI-4V: G.A. Yoder. Office of Naval Research, Arlington, VA 22217-5000, P.S. Pao, S.J. Gill and A.A. Bayles, Naval Research Laboratory, Washington, D.C. 20375-5000 The influence of ripple loading (a small amplitude cyclic load superimposed on a large sustained load) on the stress corrosion cracking (SCC) resistance of a li-6AI-4V alloy has been investigated. Of prime importance, it has been found that the degree of susceptibility to degradation from ripple-load cracking is dependent upon microstructure. In particular, the presence of small ripple loads significantly reduces time-to-failure and threshold stress intensity in SCC-resistant beta-annealed microstructure, which consists of Widmanstatten colonies. On the other hand, for the less SCC-resistant microstructure developed from a recrystallization anneal, with an equiaxed primary a-phase morphology, the degree of susceptibility to ripple-load cracking is almost negligible. This apparent contrast between SCC and ripple-load cracking susceptibilities has also been observed in recent work with ferrous alloys. A predictive framework for susceptlbiHty to the rippleload cracking effect Is developed based on the descriptive parameters used in SCC and corrosion fatigue characterization. Critical conditions required for the occurrence of ripple-load effect will be discussed.
THURSDAY AM ADVANCED METAL & CERAMIC MATRIX COMPOSITES VI: Emerging Technologies and Applications Sponsored by the TMS Powder Metallurgy Committee, TMS Physical Metallurgy Committee, and ASM-TMS Composites Committee Thursday AM February 22, 1990
Room: ParkD Inn at the Park
Session Chairmen: Gunter E. Petzow, Max-PlanckInstitut f. Metallf., Pulvermetallurgisches Laboratorium, Reisenbergstr 5, D-7000 Stuttgart 80, FRG; Joseph Stephens, NASA Lewis Research Center, M.S. 49-1, Cleveland, OR 44135 (8:30 a.m.) CARBON FIBER REINFORCED TIN-SUPERCONDUCTOR COMPOSITES: C.T. Ho and D.D.L. Chung, Composite Materials Research Laboratory, Furnas Hall, State University of New York, Buffalo, NY 14260, USA Unidirectional and continuous carbon fiber tin-matrix composites were used for the packaging of the high-temperature 5uper-
Annual Meeting-127 conductor
YBa2Cu307_~
by diffusion bonding at 170·C and 500 psi.
Tin served as the adfiesive and to increase the ductility, the normal-state electrical conductivity and the thermal conductivity. Carbon fibers served to increase the strength and the modulus, both in tension along the fiber direction and in compression perpendicular to the fiber layers, and also served to increase the thermal conductivity and the thermal fatigue
resistance.
At 24 vol.%
fib~rs,
the tensile strength was
approximately equal to the compressive strength. With further increase of the fiber content, the tensile strength exceeded
the compressive strength, reaching 134 MFa at 31 vol.? fibers. For fiber contents less than 30 vol.%, the compressive ductility exceeded that of the plain superconductor. At 30 vol.% fibers, the tensile modulus reached 15 GPa. The tensile load was
essentially sustained by the carbon fibers.
(8:50 a.m.) TIlE MICROSTRUCfURAL STABILITY OF PLANAR CAST AI-Fe-X-Si-TiC COMPOSITES STRUCTURES: D. Skinner, M. Zedelis, Allied Signal, Morristown, NJ, WE Frazier, NADC, Warminster, PA; and M.J. Koczak, Drexel University, Philadelphia, PA 19104
The elevated temperature response of AI-Fe-X-Si-TiC (X=W,V,Cr,Mo) composites alloy microstructures are examined with regard to the combined effects of silicide and carbide reinforcements with regard to strength, modulus and coarsening response. The alloys were fabricated via an in-situ gas liquid reaction of TiC platelets followed by a planar flow casting of the AI-Fe-X-Si-TiC alloy. The microstructure combines TiC (e.g. O.5Jlm) and a silicide reinforcement e.g. 30-800m. All 3(FeX)3 Si (where X=W, Y, Cr, or Mo) to provide for a dual microstructure with exceptional strength and stability. It is anticipated that excellent mechanical properties at ambient and elevated temperatures will be realized from the superposition of 5 volume % platelet TiC and 35 volume% All 3(FeX)3 Si with corrosion resistance superior to 2000 and 7000 series alloys The authors gratefully acknowledge program support from Naval Air Development Center, Warminster, PA.
(9:06 a.m.) TRANSIENT LIOUID PHASE DIFFUSION BONDING OF ALUMINUM BASED METAL MATRIX Ca1POSITES : Sudhakar Kolli and David.W.Dickinson, Welding ~ineer~ DeparOpent. 190w 19th Avenue, The Ohio State University, Columbus, OhLO 4j2l0. Metal Matrix ~sites are attractive structural materials in view of th~ir high 9pecific strength aod high specific stiffne~s. They are also super Lor to polymer matrLx cornposltes Ln terms ot higher operating temperatures and resistance to environmental d~gradatLon. Tney are manufactured in (1) fiber (2) whisk~r and (JJparticulate reLnforced forms. The disadvantages of MMC s are their anisotropy and relatively limited material property date availa9ility. In order to obtain the full pote~tial of these materLals, they are re~uLred to have good JOLnLpg pro~r ties. F~sion W?lding of MMC s results in defects such as fLber degra9atLon, pullout, 9Rlitting, porosity an9 unwanted chemical reactLon Rroducts. SOlLa State JOLnLn£ technLqu~s are,superLor lJ1 VLew ot the lower tenIPefiilb,lres reqmred to JOJ.ll MMC s. In this paper, aspects of ut~ L$Lng transient lLquid pha$e diffusion bonding, for joining SiCn/60bl 11.1 MMC, with a silver Lnterlayer and teroa:(y aluminUIQ :int:!!rlay.;r have been investigijt<;:d. R.;sults so far Lndlcate good ]OLnts wLth small amounts ot-detorrnatLon. The results of optical, electro-microscopy and microprobe analysis will be discussed. (9:22 a.m.) FABRICATION OF MoB AND Ni COMPOSITE BY P/M TECHNIQUES K. Hamashima, N. Horie, Y. Shinozaki and M. Imakawa Research Center, Asahi Glass Co., LTD. 1150 Hazawa-cho, Kanagawa-ku, Yokohama, Japan Mixtures of molybdenum mono-boride and nickel were sintered to high density compact without pressure, in vacuum. The sintering process was investigated with X-ray diffraction, DTA and TMA. The ternary M02NiB2 phase is detected above 800·C and the sintered compact is consisted with only the M02NiB2 phase and Ni phase. The sintered compacts were examined in pol ished section and tested for bending strength, hardness, oxidation rate, and resistance to thermal shock. Thereafter, some improvement in mechanical properties of the composite were performed by the modification of sintering conditions and additional components. This composite exhibits typical features at high temperature. For example, the loss of strength and hardness on increasing temperature is relatively small, compared with other ceramics-metal composits.
(9:38 a.m.) EVALUATION OF SiC PLATELETS IN METAL MATRIX COMPOSITES. M. K. Jain, Alcan International Ltd, Jonquiere, Quebec, Canada; S. P. Tremblay, C-Axis Technology Ltd, Jonquiere, Quebec, Canada; and D. J. Lloyd, Alcan International Ltd., Kingston, Ontario, Canada. Particle reinforced aluminum composites have drawn considerable attention in recent years due to their improved mechanical properties at a lower cost. These properties can be further improved by using high aspect ratio reinforcements, such as whiskers or platelets. Considering the health hazards associated
with whiskers, C-Axis Technology Limited has recently introduced platelet shape SiC for metal and ceramic reinforcement applica-
tions. The platelets produced by Alcan's proprietary technology are 2-5 ~m thick and have a high aspect ratio. Aluminum alloy metal matrix composites have been made using both powder metallurgy and molten metal mixing methods. The mechan-
ical
properties
of
platelet
reinforced composites
will
be
presented and their behavior compared with other particulate reinforced metal matrix composites.
(9:54 a.m.) VACUUM PLASMA SPRAY OF Cb AND Ta METAL MATRIX COMPOSITES: P.D. Prichard and R.L. McCormick, Howmet Corporation, 1500 South Warner Street, Whitehall, MI 49461, 2.2. Mutasim and R.W. Smith, Drexel University, 32nd and Chestnut Streets, Philadelphia, PA
19104.
Vacuum Plasma Spray was used to consolidate alloys CI03 (Cb10Hf-l.0Ti), WC3009 (Cb-30Hf) and Ta-l0W. Plasma spray processing offers many advantages in producing near net shapes using unique combinations of refractory metals and compounds. Composite materials were produced by the simultaneous deposition of TiC particulate blended into the alloy powders. Plasma deposition parameters were selected which produced dense as-sprayed
structures (>98%).
Postdeposition thermal processes such as HIP
and sinter cycles were also explored. The thermal processing further improved the density of the product and produced a more homogeneous distribution of the TiC. The room temperature and elevated temperature tensile and impact properties were evaluated. The effects of the TiC reinforcement on the structure and properties will be discussed.
(10: 10 a.m.) POTENTIAL FOR SILICIDES AS MATRICES FOR DIBORIDE REINFORCED COMPOSITES*: K. Vedula, Department of Materials Science and Engineering, Case Western Reserve University, Cleveland, OH
44106
This paper will review the potential for using MoSi2 as a matrix material for composites reinforced with TiB2. Particulate composites were made by hot pressing blends of powders. These were then examined for compatability in the hot pressed condition, and during exposure to vacuum as well as air in temperatures ranging from 1500 to laOO°C. The results are promising in terms of compatability and oxidation resistance. The possiblity of using disilicide as an oxygen diffusion barrier between diboride reinforcement and oxide matrices for high temperature applications has also been shown to be promising. * Research supported through PRDA on Ultra High Temperature Ceramic-Ceramic Composites, Wright Patterson Air Force Base, Dayton.
(10:26 a.m.) THERMODYNAMIC STABILIZATION OF SiC IN METAL MATRICES - A NOVEl TECHNIQUE FOR COATING SiC PLATELETS WITH PROTECTIVE LAYERS; R. Riedel, J. Kunesch, G. Passing, W. A. Kaysser and G. Petzow, MaxPlanck-Institut fur Metallforschung. Institut fur Werkstoffwissenschaft, Pulvermetallurgisches Laboratorium, 0-7000 Stuttgart 80, F.R.G. The number of metal matrices which can be reinforced by dispersion of SiC fibers, whiskers or platelets is limited due to the thermodynamic instability of SiC in most systems. Typically, MSi, and MC, (M=metal) form during sintering of the metal/SiC composite. The SiC deterioration may be avoided or decreased by coating the ceramic particulates with materials that hinder the reaction of SiC with the metal. In this work SiC platelets were coated with a borane adduct compound that yields boron nitride upon pyrolysis at 1000°C in protective atmospheres. The quality of the coatings generated by the thermal conversion of the borane compound is characterized by analytical methods (IR, SIMS, chemical analysis, SEM and optical transmission microscopy). Prelimi· nary results on the sintering behaviour of the coated SiC platelets incorporated in a molybdenum matrix will be presented.
(10:42 a.m.) METAL MAlRIX COMPOSITES: TiN FILMS ASl\UCROLAMINATE REINFORCEMENTS: J H Giyens and J. M. Rigsbee, Dept. of Materials Science and Engineering, University of Illinois at UrbanaChampaign, Urbana, IL 61801. Reactive ion plated titanium nitride (TiN) was deposited onto metallic foils which were subsequently hot pressed to form microlaminate composites. The volume percent of TiN was varied and its effect upon the structure and properties of the composites investigated. Scanning electron microscopy (SEM) and cross-sectional transmission electron microscopy (XTEM) have been used to characterize the film morphology, the fracture behavior of the composites, and the metal/ceramic interfaces formed during reactive ion plating and hot pressing. The mechanical properties of the micro laminate composites were determined by standard tensile testing. Correlations between the microstructure, mechanical properties, and processing parameters will be presented.
128-Annual Meeting (10:58 a.m.)
Nitridation of Liquid Aluminum-Magnesium-Silicon Alloys by Ccrnl::ostion Reaction: H. I.eHuy arrl S. Dallaire, National Research Council of canada, Industrial Materials Research Inst., 75 de Iobrtagne, Boocherville, QJebec, Canada J4B 6Y4.
(Abstract Not Available) (11: 14 a.m.)
PROPERTIES OF EXTRUDED PM ALUMINUM POLYMER BLENDS: William E. Frazier, and Gilbert J. London, Naval Air Development Center, Aerospace Mate rials Division, Warminster, PA 18974., and Alfred R. Austen, Innovare, Inc., Bath, PA 18014. In order to reduce density, enhance damage tolerance, and provide mechanical damping, selected pol ymers were combined with aluminum alloys. This new class of materials has been termed " Hymats , " In this study,S to 20% fine po lymeric powders (e.g., PEEK) were mixed with -325 mesh aluminum powder. The polymer-meta l blends were vacuum degassed and extruded into round rod . The microstruc tures and tensile properties o f the hymats were fully cha racterized. The microstructures consist of an interpenetrating contiguous network of po lymer imbeded in an al um inum matrix. Parallel to the extrusion direction, the polymers are elongated into a filamen~l ike structure.
(11:30 a.m.) CONTINUOUS SiC FIBER REINFORCED METALS. Melvin A. Mittnick, Textron Specialty Materials, 2 Industrial Avenue, Lowell, MA Ol851. Continuous silicon carbide (SiC) fiber reinforced metals (FRM) have been successfully applied on numerous aerospace development programs fulfilling primary design objectives of high specific strength over baseline monolithic materials. This presentation will review the current state-of-the-art in silicon carbide fiber reinforced metals through discussion of their application. Discussion will include a review of mission requirements, program objectives, and accomplislunents to date employing these FRMs. (11 :46 a.m.) THERMAL SPRAY PROCESSING OF METAL MATRIX COMPOSITES: R. Tiwari, H. Herman, Dept. of Mat. Sci. & Engr., SUNY, Stony Brook, NY; K. Kowalsky, D.R. Marantz, Flame Spray Industr ies , Port Washington, NY.
Near-net shapes of aluminum based metal matrix compos ites are being fabricated by a novel thermal spray processing method. The process involves a combination of hypersonic flame and electric arc sp ra ying, in which the particulates are injected into a jet of mo lten metal produced from the arc spray ing process. This approach offers control over size and distribution of t he particulates within the matrix. Preliminary studies have shown that uniform distribution of the par t iculates within the matrix can be obtained. Properties studied include elastic properties using ultrasonic techniques, density, hardness, tensile strength and percent elongation of the composite in the Optimum age hardening as-spr~yed and heat treated condition . behavior will a lso be discussed.
AGING AND LIFE EXTENSION OF NUCLEAR REACTORS I: Pressure Vessels Sponsored by the TMS-ASM-MSD Nuclear Materials Committee ThursdayAM February 22, 1990
Room: Garden Rooms 1-2-3 Inn at the Park
Session Chairmen: J.R. Weeks, Brookhaven National Laboratory, Building 130, Upton, NY 11973; J.B. Darby, Jr., U.S. Department of Energy, ER-13, G235, GTN, Washington, DC 20545 (8:30 a.m.) THERMAL AGING CONSIDERATIONS OF REACTOR VESSEL STEELS: ~ and A. L. Lowe, Jr., Babcock and Wil cox, P.O. Box 10935, LynChburg, VA 24506-0935.
~
The second generation nuclear reactor vessels used relatively new materials for construction, i.e., SA-533 plates and SA-508 forgings. Little is known about the long term stability of these
materials. Normal operating temperatures are near that for temper embrittlement whi ch may be lowered after extended time period s. Review of current knowledge indicates a sensitivity t o chemic al composition . Cu rrent programs are designed to address these concern s and others. (9:15 a.m.) IMPROVED MODEL FOR PREDICTING J - R CURVES FROM CHARPY DATA: Ernest D Eason and E. E. Nelson, Modeling and Computing Services, 1153 Bordeaux Drive , Suite 107, Sunnyvale, CA 94089. It is nece ssary to estimate J-R data for aging and residual life studies, using data from surveillance capsule specimens that refle ct the current state of the pressure vessel steel. An improved model is needed that will predict J-R data from Charpy data, tensile properties, material composition, temperature. and fluence. This report describes Phase I of a project to devel op such a model, using innovative pattern recognition and surfacefitting codes to identify the key parameters and optimal modeling functions. A readily-available sample of data on rea c tor pressure vessel and piping steels was collected and analyzed by the pattern recognition code. A preliminary model was fitted to the data, producing a more convenient model that gives a better fit with fewer fitting constants , compared to earlier models. 00:00 a.m.) POST-IRRADIATION, ANNEALING BEHAVIOR OF PRESSURE VESSEL STEELS: G E Lucas, E. Mader, G. R. Odette. Department of Chemical and Nuclear Engineering, U. of California, Santa Barbara, CA 93106 . E'm b:ritt.lement of light water reactor pressure vessel steels can be re covered by annealing at a temperature higher than the irradiation temperature for relatively short times, and this has promis e for plant life extension . While embrittlement itself has been attributed to the evolution of a fine scale damage microst ruc ture, the nature of the behavior of t.hese defects during annealing is not well known. To investigate this, a set 'of commercial pressure vessel steels and model alloys was irradiated at several combinations of neutron flux in the range 0.3-5xl0 12 nlcm 2 -s, fluence in the range O.27-8.8x10 19 nlcm 2 and irradiation temperature in the range 27l-327·C. Following irradiation, small disc specimens were subject to postirradiation annealing under isothermal conditions at 400 and 450·C for times up to 100h. Mechanical property changes were monitored with microhardness measurements. Recovery in hardness was observed at both anneal temperatures. Residual hardness increased with increasing copper and nickel content and neutron fluen ce. Defect behavior during annealing is discussed. 00 , 45 a.m.) MATERIALS AGING AND LIFE EXTENSION IN SOVIET-DESIGNED VVER's: John R. weeks, Reactor Analysis Division, Department of Nuclear Energy , Brookhaven National Laboratory, Upton, NY 11973. Pressure vessels of the Soviet-designed VVERs are exposed to up to lOX the fast flux of the vessels of US PWRs. They are fabricated of Z 114% Cr 1% Mo steels developed for that purpose. Newer grades of steels, low in-copper and phosphorus have been developed. Considerable research has been performed on pressure vessel annealing, and routine anneals have been made of the core weld region in several operating reactors. The horizontal, stainless steel tubed steam generators have performed well in some units; however, SCC of tubing has led to plans to replace steam generators in others.
CAST SHOP TECHNOLOGY VII: Casting2XXX, 6XXX and 7XXX Alloys Sponsored by the TMS Aluminum Committee Thursday AM February 22, 1990
Room: Salon 4 & 5 Marriott
Session Chairman: W. Schneider, YAW, Vereinigte Aluminium-Werke AG, Georg-von-Boeselager-Stra~e 25, D-5300 Bonn 1, West Germany (8 :30a.m .)
EFFECTS OF CASTING PRACTICE ON MACROSEGREGATION ANO MICROSTRUCTURE OF 2024 ALLOY BILLET : R. C. Dorward and D. J. Beerntsen, Center for Technology, Kaiser Aluminum, Pleasanton, CA 94566. 2024 al loy billets were cast at various drop rates, metal temperatures and mold water flow rates . Of these variables,
Annual Meeting-l29 drop rate had the greatest effect on macrosegregation and microstructure. Fast drop rates generate deep sumps which promote the collection of "isothermal l1 aluminum-rich dendrites
at the ingot center, resulting in solute depletion and duplex, heterogeneous microstructures at this location. Slow drop rates provide more uniform microstructures and much flatter segregation profiles, in which the alloying element concentrations can actually be a maximum at the ingot center. Differential scanning calorimetry was used to correlate melting thermograms with the microstructural features of the as-cast and homogenized billets.
(8:55 a.m.) MACROSEGREGATION CHARACI'ERISTICS OF COMMERCIAL SIZE ALUMINUM ALLOY INGOT CAST BY DIRECT CHILL METHOD: Men G Chu and John E. Jacoby, Alcoa Laboratories, Alcoa Center, PA 15069. A metallurgical investigation of the macrosegregation occurring in commercial size direct chill cast AI-Zn-Mg-Cu alloy ingot is described. The extent of the chemical homogeneity from the surface to the center of the rectangular ingot produced by the bi1evel transfer method is compared to ingot produced by the level pour method. The relative severity of macrosegregation for each major alloying element is discussed. Existing metallurgical theories of macrosegregation including interdendritic liquid flow and convection in the liquid pool are used to explain the observed variation in chemical composition.
(9:20 a.m.) INVESTIGATIOOS ABOOT STARrING CRACKS IN OC-cASTING OF 6063-TYPE BILLETS. PARr I: EXPERIMENTAL RESULTS. W. Schneider, Vereinigte Aluminium-Werke AG, Georg-von-Boeselager StraBe 25, D-5300 Benn 1, West GenranY1 E. K. Jensen, Elkem Aluminium Lista, N-4551 Farsund, Norway. Influence on starting crack tendency (hot cracks in ingot center) of varying a nurrber of casting pararreters was studied by experirrents (Part I) and by rrodel calculations (Part II). Effects of varying the follCMing factors were examined in detail: casting conditions; starting block material; insulating various parts of the starting block; starting block shape. Beth experirrents and calculations p:>inted to starting block shape as a rrost imp:>rtant single factor controlling starting cracks. By using the in experirrents determined and calculated depth of the liquid pool as a criterion, a proper shape was found that practically eliminates the starting crack problems in extrusion ingots of the AA 6063 type alloys.
(9:45 a.m.) INVESTIGATIOOS ABOUT STARrING CRACKS IN OC-cASTING OF 6063-TYPE BILLETS. PARr II: MJDELLING RESULTS. E. K. Jensen, Elkem Aluminium Lista, N-4551 Farsund, Norway; W. Schneider, Vereinigte Aluminium-\'/erke AG, Georg-von-Boeselager-StraBe 25, 0-5300 Benn 1, West Genrany. Influence on starting crack tendency of varying a number of casting pararreters has been studied by experirrents (Part I) and by model calculations (Part II). Both studies p:>inted to starting block shape as a rrost important single factor in controlling starting cracks. By using the Norwegian thermal rrodel ALSIM-2 in analysing experirrental results, the variable heat transfer towards the starting block was determined initially. This made possible a satisfactory model analysis of the starting phase and likewise the formulation of a useful cracking concept. By using the calculated depth of the liquid pool as a criterion, a proper starting block shape was found that practically eliminates the starting crack problems in extrusion ingots of the AA 6063 type alloys.
(10:10 a.m.)
BREAK
(10:25 a.m.) MATHEMATICAL MJDELLING OF THERMAL STRESSES DURING D.C. CASTING OF ALUMINIUM BILLETS. H. Fjaer,Institutt for energiteknikk, Box 40, 2007 Kjeller, Norway, A. M::l, Senter for Industriforskning, Box 124 Blindem, 0314 Oslo 3, Norway. The paper presents the mathenatical rrodel ALSPEN in which the
therrna.lly induced strains and stresses being developed during direct-chill continuous casting of aluminium billets are calculated by a finite elerrent method. The metal is assumed to be an isotropic elastic-viscoplastic material with strongly temperature dependent properties. In the mathenatical rrodel the generated viscoplastic strain is treated in a 'unified' manner where lCM temperature (approximately) tirre-independent plasticity and high temperature creep occur as special cases. As a special case study we have modelled the casting of the AlMgSi-alloy AA 6063 whose material properties have first been studied in detail. As one of our results the residual stresses which arise during casting for different casting speeds are presented.
(10:50 a.m.) THE ROLE OF MANGANESE IN 6XXX ALLOYS: I. Musulin, O. C. Celliers, Comalco Research Centre, 15 Edgars Road, Thomastown, Victoria 3074, Australia. Additions of manganese are common 6XXX billet and can be useful in both the low and high strength end of the 6XXX range. During the homogenization of 6063 for instance, Mn promotes the transformation of the deleterious ~-AlFeSi phase to the more favorable a-AlFeSi phase. This can result in reduced temperatures and times required for sufficient ~-.a transformation - normally the rate determining step during homogenization. Additionally, Mn can promote a uniform Mg2Si distribution with AI6Mn acting as nuclei for Mg2Si precipitation throughout the billet. In 6XXX extrusions, smaIl levels of the Mn generally do not reduce the final mechanical properties, as long as sufficient press cooling is employed even though Mn increases quench sensitivity. Some loss of properties can result, however, when slow cooling is employed, or when Mn levels are increased.
DEFORMATION & MECHANICAL BEHAVIOR II TMS General Abstract Session Thursday AM February 22, 1990
Room: Parke Inn at the Park
Session Chainnan: G.T. Gray, Los Alamos National Laboratory, Los Alamos, NM 87545 (8:30 a.m.)
PLASTIC ANISOTROPY OF FCC METALS AND ALLOYS: M. A. Przystupa, P.O. Box 39, Redondo Beach, CA 90277; A. K. Vasudevan, Office of Naval Research, Code 1216, 800 N. Quincy St., Arlington, VA 22217-5000; W. G. Fricke, Jr., Alcoa Technical Center, Alcoa Center, PA 15069. Plastic anisotropy of FCC metals and alloys, manifested by a change of yield stress with the loading direction, can be influenced by several factors. The most important one is texture, but directional precipitates, grain shape and the size of the tensile specimen could also add to the observed anisotropy. This talk will deal with the predictions of the magnitudes of anisotropic effects from all above factors and with their interactions. The models used assumed that grains deformed with either five of less slip systems and that directional precipitates were either rigid or elastic. Results will show anisotropy resulted from rolling and annealing textures for different number of slip systems in grains and for directional precipitates on Illl}, {110} and {100} habit planes. Tensile specimen size effects will be illustrated on Barlat-Richmond tricomponent yield surfaces with traces of stress state corresponding to tensile tests for different specimen sizes. (8:50 a.m.) ELASTO-PLASTIC MODELING OF INHOMOGENEOUS DEFORMATION IN POLYGRANULAR SOLIDS,* Mark McCormack, Jean-Hubert Schmitt, and J.W. Morris, Jr., Center for Advanced Materials, Lawrence Berkeley Laboratory, and Department of Materials Science and Mineral Engineering, University of California, Berkeley, CA 94720.
A new elasto-plastic deformation model has been developed in order to investigate inhomogeneous deformation in polygranular solids. This model allows one to independently vary the contributions of grain texture, hardening, and multiple active slip systems. Simulating deformation processes with this model can provide an understanding of how these contributions can affect inhomogeneous deformation. Simulation examples, including the formation of localized deformation bands during tensile testing, will be presented on video tape.
* Supported by the Director, Office of Energy Research, Office of Basic Energy Science, Materials Sciences Division of the U.S. Department of Energy under control No. DE-AC03-76SFOOO98. (9:10 a.m.) FEM SIMULATION OF PLASTIC BEHAVIOR OF COMPOSITES. R.E.Bolmaro*, U.F. Kocks** and R.V. Browning**. * Universidad Nacional de Rosario, CONICET. Argentina. ** Los Alamos National Laboratory, Mail stop K-765, Los Alamos, NM 87545. The understanding of the plastic behavior of two phase metals and composite materials is of paramount importance in both theorethical and applied material science. 2-D FEM calculations are performed on different volume fractions of hard spheres in a soft
130-Annual Meeting matrix and vice versa. The results are compared with usual phenomenological theories and they also give clues to the understanding of development of textures in two phase materials. * Currently on CONICET fellowship at the Center for Material Science. Los Alamos National Laboratory.
(10:50 a.m.) COMPUTER SIMULATION OF IMMERSION QUENCHING THE URANIUM-0.8%TI ALLOY IN WATER: ANALYTICAL PREDICTIONS AND EXPERIMENTAL VERIFICATION: Gerard M. Ludtka,K.W. Childs,G.A. Aramayo,K.L. Luk,J.E. Park, Oak Ridge Y-12 Plant*,P.O. Box 20009, Oak Ridge, TN 378318096, and A. Salinas-Rodriguez,J.H. Root,T.M. Holden,S.R. MacEwen, Chalk River Nuclear Laboratories,Chalk River, Ontario, KOJ
lJO.
(9:30 a.m.)
TEXTURE DEVELOPMENT IN TWO PHASE MATERIALS: FREE COMPRESSION ON Ag-Ni COMPOSITES: R.E.Bolmaro* ,U.F. Kocks** and A.D. Rollett**.* Universidad Nacional de Rosario, CONICET. Argentina. ** Los Alamos National Laboratory, Mail stop K-765, Los Alamos, NM 87545. Few attempts to understand the development of textures in two phase materials have been made in the past, despite the fact that these materials constitute the majority of commercial alloys. with the goal of understanding the general behaviour of two phase materials, free compression tests have been performed on 25% Ag-Ni, 50% Ag-Ni and 75% Ag-Ni powder composites. The results are analyzed by FEM simulation and using the Los Alamos Polycrystal Plasticity code. * Currently on CONICET fellowship at the Center for Material Science. Los Alamos National Laboratory. (9:50 a.m.)
SENSITIVITY OF R-VALUES TO STRESS STATE: ~ Rollett, U. F. Kocks and M. G. Stout. Los Alamos National Laboratory, Mail Stop K765, Los Alamos, NM 87545. It appears that the r-values (plastic strain ratio in tensile test) as calculated by a polycrystal plasticity code are sensitive to the assumptions made about the stress boundary conditions. Since the rvalue is a sensitive measure of the slope of the yield surface, this st&tement is equivalent to stating that the yield surface shape is sensitive to the stress boundary conditions. This concept is explored in some detail for examples of a simulated rolling texture and for measured rolling textures in copper and aluminum.
Residual stresses on the order of the yield strength (760
mPa) can develop in the U-O.STi alloy under water immersion quenching. Analytical efforts have been under way to simulate the quenching process to predict the microstructure and residual distributions which develop after a quenching operation. One benefit of the analytical approach is that parametric studies can be run relatively inexpensively to determine the critical varia-
ble(s) which have the largest influence on the magnitude of the
quenched-in residual stresses.
These results can assist in de-
fining alloy development goals to yield a new alloy which devel-
ops less residual stress upon quenching. Several of these parametric studies will be discussed. In addition, experimental verification of the residual stress predictions of the QUENCH SIMULATOR will be provided. The experimental data were nondestructively determined using a neutron diffraction technique. *Operated by Martin Marietta Energy Systems, INc., for the U.S.
Department of Energy under contract DE-AC05-840R21400.
(11:10 a.m.) EFFECT OF HEAT TREATMENT AND UNIAXIAL COMPRESSION ON THE RESIDUAL STRESSES IN A U-0.8 WT %TI ALLOY MEASURED BY NEUTRON DIFFRACTION: Gerard M. Ludtka, Oak Ridge Y-12 Plant", P.O. Box 2009, Oak Ridge, TN 37831-8096, and A. Salinas-Rodriguez, T. M. Holden, J.H. Root, and S. R. MacEwen, Chalk River Nuclear Laboratories, Ontario, Canada KOJ IJO. The residual stress distribution in U-0.8Ti alloy clinders was studied using neutron diffraction techniques. The residual strains were evaluated from the difference in d-spacings measured in the cylinders and in small reference samples machined from each cylinder, and the stresses were calculated with the isotropic bulk values of the elastic constants for U-0.8Ti alloys. Quenching from theY field results in a biaxial stress state at every point across the wall of the tube. The radial stress was negligible and the axial and circumferential stresses, which were compressive at the surface of the cylinder and tensile in the interior, had equal magnitude. The quantitative results for the quenched-in residual stresses will be discussed along with the effects of subsequent stress leveling (0.015 permanent compressive strain) and/or aging treatments (658°K for 4 h and water quenched) on these initial stresses. *Operated by Martin Marietta Energy Systems, Inc.,
~-quenched
for the U.S. Dept. of Energy under contract DE-AC05-840R21400.
(10:10 a.m.)
FORMATION OF FIRST GENERATION MICROBANDS IN ROLLED Al +5.5 at.% Mg: D. A. Hughes, Sandia National Laboratories, Livermore, CA 94550. The deformation substructure in lightly rolled Al +5.5 at. % Mgwas ob· served using TEM to determine whether a similar mechanism for microband formation in cell forming fcc metals applied to an alloy which did not form cells. At small strains, the dislocation substructure of Al +5.5 at. % Mg consists of a uniform net of dislocation tangles. Additionally a new dis· location substructure was observed which consisted of long straight dense dislocation walls (DDW) which often, but not exclusively lay near {lll} slip planes. In an analogous manner to pure aluminum and nickel, these DDWs subdivide a grain into slightly misoriented regions in which defor· mation occurs on fewer slip systems than specified by the Taylor criterion for strain accomodation. First generation microbands consisting of paired dislocation sheets were found to form from the single DDWs by gradual glide processes without the initiation of a shear instability. At larger strains subsequent to their formation, these microbands exhibit the commonly reo ported shear displacements where they intersect a boundary. Work supported by U. S. DOE under contract No. DE-AC04-76DP00789. (10:30 a.m.)
PLASTIC fLOW BEHAVIOR IN TENSION OF ULTRA-HIGH STRENGTH, HIGH DUCIlLITY, WARM ROLLED FE-12.SMN-2.0IMO- J.lSC HADFIELD STEEL. John F. Chinella. U.S. Anny Materials Technology Laboratory, Watenown, MA. Floyd R. Tuler, Worcester Polytechnic Institute, Worcester, MA. Hadfield steel plastic flow behavior is described by strain hardening without necking to high levels of strength and strain, and is described by upward curvature strain hardening with near·identical, near-linear strain hardening rates over large ranges of deformation temperature. This plastic flow behavior is valued for severe service applications, but Hadfield steel yield stress is low. In this study, solution treated, Fe-12.SMn-2.0IMo-l.lSC Hadfield steel is wann rolled for the purpose of increasing yield strength and to retain characteristics of Hadfield steel plastic flow behavior. Experimental plastic flow behavior is evaluated in tension. The wann rolled Hadfield steel has high yield stress, moderate rates of increasing load or strain hardening, high true uniform strain and ultra-high true uniform stress. The plastic flow behavior and microstructures suggest an influence of Hadfield steel deformation TWins on dislocation slip and on plastic flow behavior.
mGH TEMPERATURE DEFORMATION,
FRACTURE AND APPLICATION OF INTERMETALLIC COMPOUNDS Ill: Selected High Temperature Intermetallics
Jointly sponsored by ASM Flow and Fracture Committee and TMS High Temperature Alloys Committee Thursday AM February 22, 1990
Room: Plaza B Inn at the Park
Session Chainnen: E.P. George, Metals and Ceramics Div., Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831-6093; N. Nazmy, Dept. of Mats. Technology, Asea Brown Boveri and Company, CH-5401 Baden, Switzerland (8:30 a.m.)
(Invited)
1000 to 1300 K Slow Plastic Compression Properties of AlDeficient NiAl, J. Daniel Whittenberger, NASA Lewis Research Center, Cleveland, OH 44135 and K. S. Kumar and S. K. Mannan, Martin Marietta Laboratories, Baltimore, MD 21227-3898 Nickel aluminides containing 37, 38.5 and 40 at pet. Al have been fabricated by XD m synthesis and hot pressing. Such materials were compression tested in air under constant velOCity conditions between 1000 and 1300 K. Examination of the microstructures of hot pressed and compression tested aluminides indicated that the structure consisted of two phases, y' and NiAl, for essentially all conditions, where y' was usually found on the NiAl grain boundaries. The stress-strain behavior of all three intermetallics was similar where flow at a nominally constant stress occurred after about two plastiC percent deformation. Furthermore, the 1000 to 1300 K flow stress-strain rate
Annual Meeting-131 properties are nearly identical for these materials, and they are much lower than those for XD m processed Ni-SOA1. The overall deformation of the two phase nickel aluminides appears to be controlled by dislocation climb in NiAl rather than processes in y' . (Invited)
(9:00 a.m.)
sic REINFORCED-MoSi2 BASED MATRIX COMPOSITES. J.J. Petrovic, R.E. Honnell, D.H. Carter, W.S. Gibbs, F.D. Gac, Materials Science and Technology Division, Los Alamos National Laboratory, Los Alamos, New Mexico 87545. sic reinforced-MoSi 2 based matrix composites have very significant potential as elevated temperature structural materials for temperatures above 1200 C, due to their combination of high melting points, oxidation resistance, thermodynamic stability, and elevated temperature ductility and strength. Fabrication of composites, various types of sic reinforcements, and alloying of the MoSi 2 matrix with other silicides will be discussed. Mechanical properties of these composites will be presented. (9:30 a.m.) HIGH-TEMPERATURE DEFORMATION IN MoSi2:* D. S. Schwartz and P. J. Meschter, McDonnell Douglas Research Laboratories, St. Louis, MO 63166. Molybdenum disilicide (MoSiZ) is a promlslng material for applications at lOOO-1600°C owing to its superior high-temperature strength and oxidation resistance. MoSi2 exhibits a brittleto-ductile transition at - 925°C and increasing ductility without loss of strength up to 1250°C. The deformation processes that occur in dense, polycrystal1ine MoSi2 subjected to compressive loading at 1000 and l200°C have been examined by transmission electron microscopy. The dominant slip planes and their associated dislocations have been determined. In addition, hexagonal dislocation networks have been observed and their constituent dislocations identified. *Research conducted under the McDonnell Douglas Corporation Independent Research and Development program. (9,50 a.m.l
(10:40 a.m.) THE EFFECT OF TiB2 PARTICLES ON THE ROOM AND ELEVATED TEMPERATURE MECHANICAL PROPERTIES OF REACTIVELY PROCESSED NiAI. D.E. Alman and N.S. Stoloff, Materials Engineering Department, Rensselaer Polytechnic Institute, Troy, NY l21S0-3590 Reactive synthesis techniques comprise blending elemental powders and then sintering to form a des,ired compound. When reactive sintering is coupled with the application of pressure via a Hot Isotatic Press, the process is termed Reactive Hot Isostatic Pressing (RHIP). Due to the large exotherm associated with the formation of NiAl, small amounts of prealloyed NiAl powder had to be added to the mixture of elemental powders to "dilute" the reaction. Monolithic NiAl and NiAl reinforced with TiB2 particles were fabricated by RHIP. Compression and tension tests were performed at temperatures between 2SoC and lOOO°C. The effects of volume fraction and particle size of TiB2 on the ductile to brittle transition temperature, fracture morphology and strength of NiAl will be discussed.
(11 :00 a.m.) Structures and Grain Morphology of Directionally Solidified Ni3Al Intermetallic Materials: Y. Kim, J.C. Earthman and E.J. Lavernia, Materials Section, Department of Mechanical Engineering, UniverSity of California, Irvine, CA 92717 This work reports on a study of the structures and grain morphology of a directionally solidified IC 396M Ni3Al intermetallic material (Ni-SCr-SAl-3Mo-0.SZr in wt. %). In directional solidification the material is drawn through a constant thermal gradient at a uniform rate. As a result of the imposed thermal gradient and the fixed rate of sample displacement, the latent heat of fusion is extracted in a steady manner through the solid. This permits the growth rate of the solid (which mayor may not be equal to the rate of displacement) and the thermal gradient to be separately controlled. Hence, using the appropriate experimental conditions, an elongated, highly directional microstructure can be readily developed. The microstructural studies are accomplished using the scanning, transmission and optical microscopes.
(11 :20 a.m.) ELEVATED-TEMPERATURE MECHANICAL PROPERTIES OF TiAl and XD' TiAl ALLOYS: B. London and D. E. Larsen, Jr., Howmet Corporation, Applied Research Division, 1500 S. Warner St., Whitehall, MI 49461.
BREAK
(10:00 a.m.) HIGH TEMPERATURE CREEP PROPERTIES AND MECHANISMS IN Fe3AI:* C. G. McKamey. C. T. Liu, and J. A. Honon, Metals and Ceramics Division, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, Tennessee 37831-6115 Iron aluminide compositions based on Fe3AI and alloyed with various other elements were tested in creep at 593"C and 207 MPa in air. These tests were constant load tests and were continued to rupture, providing creep rupture life and ductility data. It was found that the creep life of Fe3AI, which is only 2-3 h in the binary alloy, could be improved to 200-300 h by alloying with mOlybdenum or niobium. This presentation will summarize our study to date and discuss the compositional and microstructural effects on creep mechanisms in Fe3Al. *Research supported by the U.S. Depanment of Energy, Office of Fossil Energy, AR&TD Materials Program under contract DE-AC05-840R2 I 400 with Martin Marietta Energy Systems, Inc. (10 :20 a.m.) EFFECT OF L1 TRANSFORMATION ON THE CREEP STRENGTH AND1JBTT OF AL.3ZR: J.R. Dobbs; R.D. Field, G.E. Aircraft Engines, I Neumann Way, Cmcinnati, OH 45215-6301 The intermetallic compound Al3Zr could be attractive for aerospace applications becau e of its high melting temperature (l853K) and low density (4.1 g/cm ). A major drawback is, however, its lack of low temperature ductility, and little is known about the high temperature deformation or creep resistance of the brittle 0023 crystal structure of AI3Zr. In previous investigations on improving low temperature ductility, the 0023 structure was converted to the Ll2 structure of Ni3AI with the addition of Ni, Fe, Cr or Cu, but no improvement was noted. In the present investigation, AI-25Zr with the 0023 crystal structure, and the ternary AI-6Ni-25Zr with the Ll2 crystal structure were studied to determine the effect of the Ll2 transformation on the ductile to brittle transformation temperature (OBIT) and the creep strength. All material was processed by arc-melt-drop casting, and homogenization at 1560K. The OBTT was established by four point bend testing, and creep behavior was evaluated under compressive loading. Structure and deformation features are discussed.
3
High-temperature tensile and creep properties of unreinforced Ti-4SAl-2V (at.%) and XD' reinforced Ti-4SAl-2V (at.%) + 7 vol.% TiB2 were studied. Samples were tested in the as-cast plus heat treated condition, and comparisons were made to conventional wrought TiAl material. Tensile properties were studied between room temperature and 9S0°C (IS00°F). Creep properties were evaluated over a range of stress and temperature conditions. Preliminary results indicated the cast material was superior to wrought material in several instances. Microstructural analysis of the XD' material indicated the role of the TiB partic1ulate in tensile and creep deformation processes. APP-)ications for these materials in high-performance aircraft are discussed.
HYDROMETALLURGY TMS General Abstract Session Thursday AM February 22, 1990
Room: Salon G
Marriott
Session Chairman: John F. Pusateri, Director, Horsehead Resource Development, Technical Center, 300 Frankfort Rd., Monaca, PA 15061 (8:30 a.m.)
SILVER ION CATALYSIS IN LEACHING OF NICKEL MATIE: Jill Zuanei, Wen wen, Dept.of Metal Materials, Chengdu University of Science and Technology, Chengdu, Sichuan, P.R.China The kinetics of dissolution of nickel matte in the presence of mangilllese dioxide with silver ions was investigated. The catalytic silver ions enhanced the leaching rate up to a factor of 17. The effects of stirring, temperature, particle size, concentrations of Mn02 ' silver and ferric i.ons, and the ways of feeding the raw materials were studied. Solid residues at various levels of nickel extraction were examined by sm, X-ray diffraction, electron microprobe and chemical analysis. These results indicated that the dissolution reaction was topochemical
132-Annual Meeting
and fits a surtace reaction control model. The activation energy '1S calculated to be 39.15 KJ/mol. The leaching rate of Ni S was 1 2 '~ndent with stirring rate, but the reaction rate cOnstant ,"
00:10 a.m.) Solid Liquid Separation in Neutralized Acid Mine Drainage
J. Brousseau. L. Garrison, and K. Prisbrey University of Idaho. Moscow, Idaho
A dramatic improvement has been developed in solid liquid separation of yellow boy (solids precipitated from
acid
(8:50 a.m.)
APPLICATION OF SULFATE ROASTING AND LEACHING TO A COPPER-ZINC SULFIDE CONCENTRATE, C.J. Ferron, Lakefield Reseach, P.O. Box 430, Lakefield, ON, Canada KOL 2HO, J. De Cuyper, Laboratoire de Traitement des Minerais, Universite Catholique de Louvain, 2, Place Ste Barbe, 1348 Louvain-La-Neuve, Belgium. Sulfate roasting followed by leaching in a mildly acidic solution was applied to a copper-zinc sulfide concentrate. Thermodynamic principles of the sulfate roasting were reviewed; roasting in open hearth and flua-solids furnaces were compared, and the latter was selected for its better selectivity towards iron dissolution. The major parameters of the leach, such as duration, temperature and acid strength, were investigated, and recoveries of 99% and 95% were achieved for copper and zinc respectively, with low iron contamination. Various methods are proposed for the purification of the leach solution, the recovery of metal values from the purified solution and that of precious metals from the leach residue. (9:10 a.m.)
MECHANISM OF THE ANODIC DISSOLUTION OF CHALCOCITE IN HYDROCHLORIC ACID R. S. Parikh and K. C. Liddell, Chemical Engineering Department, Washington State University, Pullman, WA 99164 Chalcocite was anodically dissolved in 1M HCl using demountable mineral disks in a rotating ring-disk electrode. Disk potentials between 0.0 and 1.0 V vs. SCE were investigated, and the disk current, solid reaction products and Cu(I)jCu(II) product ratio were a strong function of potential. In order to account for the different products observed at different potentials, separate reaction mechanisms are proposed for the potential ranges 0.0 - 0.3, 0.3 - 0.6, and 0.6 - 1.0 V. The products at intermediate potentials (0.3 - 0.6 V) can be accounted for by a combination of the mechanisms for higher and lower potentials. (9:30 a.m.) THE APPLICATION OF THE FINMET PROCESS TO THE PRODUCTION OF ANTIMONY TRIOXIDE: I.D.Ewart, L.D. Jayaweera, E.E. DaviS, and R.C. Robins, FMC (Technologies) Ltd., 31-33 Smith Street, Marrickville, NSW 2204, Australia. The FINMET process which was developed by FMC (Technologies) Limited in Australia involves oxidative halide leaching with This process has been applied to various concentrates and residues for metal recovery. This paper discusses the application of the FINMET process to the treatment of antimony concentrates to produce antimony trioxide and other antimony chemicals. Leaching is carried out with ferric chloride at pH 0.5 0.7 and a temperature of 95-98°C. The antimony is removed from the leach solution by precipitation as an antimony oxychloride which is then hydrolysed to antimony trioxide. Ferric chloride is regenerated at the anode of an electrochemical cell with metallic antimony being produced at the cathode. A semi-commercial pilot plant has been operating in Sydney,
mine drain.ge by
pH adjustment).
The application is
for small acid mine drainage flows which must be reused for drill water or other mine operations under conditions where thickeners are impractical. The technique is similar to oil agglomeration of coal fines in black water. Selected surfactants are added to make the yellow boy hydrophobic. then the yellow boy is agglomerated for easy removal. By removal of the floating agglomerates. 99~ percent ot the suspended solids are removed. The remaining cloudy water turns completely clear upon filtration through a coarse filter cloth. Almost no noticeable residue remains on the filter medium. In order to produce clean. iron free, pH neutral water. we evaluated water chemistry. agitation intensity. surfactant and other process parameters.
(10:30 a.m.)
INTERFACIAL EFFECT IN CHROMIUM-ALIQUAT 336 EXTRACTION SYSTEM:Wen-Shou Chuang, Jiin-Shiung Horng & Ying-Chu Hoh; P.O. Box 3-7, Lung-Tan, Taiwan 325, Republic of China. The pendant drop method was used to measure the interfacial tensions for the extraction of chromium in basic medium by Aliquat 336 at various conditions,such as variation of extractant, aqueous feed concentration and diluent effect etc. For chromium-Aliquat 336 extraction system, the relationship between the interfacial tension (8 ) and the distribution coef~icient ( D ) can be satisfactorily expressed as 8 = D , where n is an empirical constant. Marangoni effect can interpret these interfacial phenomena. Based on the interfacial tension data,one may qualitatively predict which phase will be continuous or dispersed. (10:50 a.m.)
New Chelating Ion Exchange Resins for Industrial Use: O. Schumacher, Schering Berlin, Bergkamen, West Germany D. Gefvert, Sherex Chemical Company, 5777 Frantz Rd., Dublin, Ohio 43017 A novel class of ion exchange resins incorporating the chelating functionality of 8-hydroxyquinoline is presented. The extraction of base metals, rare metals, and precious metals is examined and compared to several commercial chelating resins. Stability in acidic and basic media is demonstrated. Resin stripping using various aqueous stripping solutions is discussed. The extraction properties of the resin are compared to the 8-hydroxyquinoline based solvent extraction reagents.
electrochemical regeneration of the leachant.
Australia since March 1989.
Some of the operating details for
this plant are discussed.
(9:50 a.m.)
THE
SHDY
OF~Gr:LOMFRAT I O~J
IN
THE
rECO~pnc, I Tl ON
PROCESS OF TilE
ALrMJ\':\TF sOltrT!fl\ 7l!flll. H"uifang lllPllgzhull ',iyhl mdals rt~'~eafl:ll ins titutp ::han:jjIl' ht'nan P. R. fhinil This papf'T is ahtJlit the aggLoDleratlon iIi tl,e d!:'I>OWpo~i tion prort'.)~, llf the ,ltll~Il::la!t' :;olutluti \Vin~ !ht, :'Hldition of :~red, Whilr limitiii19 ~he se(oJldary nucleation (d lry~;lal agglomeration shm:;:> the following lIain charactrri:)tics: 1. AgLjlolll'[(ltiun happens amung tllt'lIArtir-If'S of seed whirh are the smaltest and simi lar in the size, 2 . The changes of agglomeration lontiilions,such as temperat!HP, tillH' anli scerl amount, have much UrPater influences on the clint"'n! of large palticles than that of small llH,:"s.3. The criti.-al degree of ~jupt'rsatllration for thr agglomeration are various fnr tiifferent--sizfld trihydrate aluminium 'dH'cI paTticl('~i lbf CO~f'.:>cr thr particle:'> are fhf' higlifr the valiles ;hall be. 4. The agglo.eration, rry,tal growth and serondary nuclcation Ld.l' llatce :;irnuttaneollsly an.t the) are infllH'J1CeU OIll' anothpT thp level of the inftuence is dependent on the conditions.5.The a!Jglomeration procedures are the fOllo~iIlgl fi~fJ the ~;iDaltcst partil,le~; in the liquor agglomerate to form larger partirles, then the larger particles agglomerate to form even larger particles and so Oil, uuti 1 the critical degree of supersaturation of !he solution fOT that particle sjze is real-lied.
INTERFACES IN METAL-CERAMICS COMPOSITES VI: Mechanical Properties Sponsored by Process Fundamentals Committee and Composites Committee Thursday AM February 22, 1990
Room: Plaza D Inn at the Park
Session Chairmen: K.K. Chawla, National Science Foundation, Division Mats. Research, Washington, DC 20550; P.K .Liaw, Westinghouse Science & Technology Center, Pittsburgh, PA 15235
J
J
(8:30 a.m.) EFFECT OF PHASE MORPHOLOGIES ON THE MECHANICAL PIWf'EI{T IES OF BP.BBITT-BRONZE COMPOSITE INTERFACES. P. K. Liaw. Westinghouse Science &. Technology Center, Pittsburgh, PA 15235; Y. Ijiri, Princeton University, Princeton. NJ 08544; B. J • .Taszarek. 107 Elatan Dr., Pittsburgh, PA 51243; S. Frohlich 1231 Denniston Ave .• Pittsburgh, PA 15217; M. N. Gungor, Westinghouse Science & Technology Center; and W. A. Logsdon, Lyons Run Road, Trafford, PA 15085. Interfaces of two different babbitt-bronze composites were tested ultrasonically and then were fractured using the Chalmers test method. The primary distinction between the two composites was in the copper content. Use of less copper in the babbitt results in interfaces with higher strength, lower ductility, less cracking and less unbonded area. The differences appeared to stem from the structure of the intermetallic compounds found at the interface. namely the CU6SnS and the CU3Sn layers. The low copper composite failed within a thick. whisker-like CU6SnS layer, while the high copper one separated at the interface between a smooth CU6SnS layer
Annual Meeting-133 and the babbitt metal. The rough interface morphology seemed responsible for the low copper composite's increased strength. The <.orrelation between mechanical and ultrasonic properties was poor for the low copper composite but excellent for the high copper one. These results suggest that interface morphology can significantly affect mechanical as well as ultrasonic properties.
(9: 00 a.m.) SUBCRITICAL CRACK PROPAGATION BEHAVIOR ALONG CERAMICMETAL INTERFACES:' R. H. Dauskardt,1 B. J. Dalgleish,2 R. M. Cannon I and R. O. Ritchie, I I Center for Advanced Materials, Lawrence Berkeley Laboratory, and Department of Materials Science and Mineral Engineering, University of California, Berkeley, CA 94720; 2Materials Department, University of California, Santa Barbara, CA 93106. Micromechanisms contributing to the toughness and subcritical crack-growth resistance of ceramic-metal interfaces have been examined in the glass-copper, silica-copper and alumina-aluminum systems. Specifically, toughness, stresscorrosion and cyclic-fatigue crack propagation behavior have been studied in moist and dry environments. Interface engineering techniques, which implant pre-existing crack-like defects near the interfaces, have been shown to induce large increases in interfacial fracture toughness; such patterned interfaces similarly show far lower stress-corrosion and fatigue-crack growth rates at equivalent strain-energy release rates. Models for such behavior are presented in terms of crack bridging, crack deflection and metal-segment tearing phenomena. Supported by the U.S. Department of Energy, Basic Energy Sciences, under Contract No. DE-AC03-76SF00098. (9:30 a.m.)
MICR:JSTRUCI'URE AND PROPERITES OF ALUMINUM/STAINLESS Il\MINATE ~ITES: K.K. Chawla, Departrrent of Materials and Metallurgical Engineering, New Mexico Institute of Mining and Technology, Socorro, New Mexico 87801, USA. Laminated composites consisting of alternating sheets of commercially'pure aluminum and type 304 stainless steel were fabricated by roll bonding. The composites thus made were characterize::l microstructurally and rrechanically. Hardness and tensile properties """re determined in the as-made conJition and after isothermal exposure for different tirres at 823 K. The strength and uniform strain increase::l with the steel volurre fraction. The high temperature exposure resulted in a degradation of these properties because of the formation of brittle intermetallic oampound at the interface. It was also observed that a steel volurre fraction between 50 and 60% gave the best specific properties.
(10:00 a.m.)
BREAK
(10:15 a.m.)
SHEAR PROPERTIES OF MMC BY TORQUE TESTS K.Hashimoto,S.Sekiguchi and K.Yamada R&D LaboratoriesI, Nippon Steel Corp. Nakahara-ku Kawasaki 211 JAPAN
MMC with the following combinations: carbon fiber/AI and pcs-SiC fiber/AI were produced by squeeze casting method and examined shear properties by means of torque tests. It was found that shear stress- strain curves of CF/Al in fiber direction showed upper yielding like phenomenon, on the other hand, those of pcs-SiC/AI showed parabolic deformation behavior. These differences in shear stress-strain behaviors among MMC's are considered due to the difference in the character of interface between metal matrix and fibers. The chemical composition at the vicinity of the interface was analyzed by EPMA and AES techniques. The effects of chemical bonding strength at the interface on the both shear strength and modulus were rationalized using the finite element method calculation applying simple self-consistent model. (10:45 a.m.)
MICROSTRUCTURE AND TENSILE STRENGTH OF GRAPHITE/ ALUMINUM COMPOSITES AFTER THERMAL CYCLING, Weidong Cai, Renjie Wu, Institute of Composite Materials, Shanghai Jiao University, 1954 Huashan, Shanghai 200030 P.R. China. Studies on microstructure and tensile strength of graphite-aluminum composites after thermal cycling between room temperature and 450· C were carried out. It was indicated that the interfacial bond was obviously weakened and tensile strength changed after thermal cycling. According to the variation of tensile strength and interfacial bonding of composites, the relationship between them will also be discussed. (11:15 a.m.) THE EFFECTS OF INTERFACIAL REACTION ON THE PROPERTIES OF CARBON/ALUMINUM COMPOSITES Guoding Zhang Yeshen Chen, Renhjie Wu, Institute of Composite Materials, Shanghai Jiao Tong University, Shanghai, 200030, P. R. China
The interfacial reaction between fiber and matrix of the C/Al composites occurs during the fabrication and service at high temperature. The eondi tion of interfacial reaction in the composites relates closely to the production process of composites. This paper deals with the interfacial reaction taking place at different temperatures and influences on properties of the C/Al composites. The results of this
investigation show that the corresponding interfacial reaction occurs at different temperatures. It strengthens the bonding between fiber and matrix of the composites, and forms a brittle layer on the fiber surface. In addition, at high temperature it damages the fiber by interfacial reaction. These results obviously influence the properties of composites. The fracture of the CIAl composite under lower stress results from the stronger interfacial bonding and brittle interfacial layer even though carbon fiber may remarkably not be damaged. In order to produce high quality C/Al composites, i t is necessary to control interface reaction to obtain a suitable interfacial bonding between fiber and matrix in the composites.
LOW TEMPERATURE Si AND Si-BASED EPITAXIAL STRUCTURES: ELECTRICAL PROPERTIES AND DEFECTS IV: Si-Ge
Epitaxy by MBE
Sponsored by the TMS Electronic and Photonic Device Materials and Electronic Materials Committees Thursday AM February 22, 1990
Room: Orangewood ill Quality Hotel
Session Chairman: R. Reif, Massachusetts Institute of Technology, Cambridge, MA 02139 (8: 30 a.m.)
(Invited)
RECENT ADVANCES IN THE SCIENCE AND APPLICATION OF SILICON MOLECULAR BEAM EPITAXY: John C. Bean, AT&T Bell Laboratories, Murray Hill, NJ 07974. The aim of this talk will be to provide a perspective on recent experimental work in silicon molecular beam epitaxy. Highlighted material will include: I) The physics of strained layer GeSi, 2) A possible breakthrough in the control of epitaxial metal silicide morphology and current prospects for metal or permeable base transistors. 3) The Swan song for epitaxial insulators? 4) The persistent problem of particulates and current thinking on the design of apparatus. 5) The drive towards a gas source hybrid of eVD and MBE approaches to silicon epitaxy.
(9:00 a.m.) THE INFLUNCE OF DEFECTS ON DEVICE PERFORMANCE OF MBE GROWN SI HOMOJUNCTION AND STRAINED SII_xGEx/SI HETERO·STRUCTURERS. !1..X....XJI, G. D. Shen, M. Willander and G. V. Hansson, Department of Physics, Linkoping University, S·581 83 Linkoping, SWEDEN Several kinds of Si homojunetion and strained Si l.xGex/Si heterojunction diodes and bipola:transistors have been fabricated by a low temperature process: Si·MBE.The influence of defecl' on MBE film quality, the influnce of annealing and electron irradiation on the device perlormanc" are investigated by studing the temperature dependence of electrical characteristics, TEM and DLTS results. Growth temperature Ts has important impact on film quality. Electrical measurements show that Ts should be -8000C for Si homojunction ami 500~5500C for strained Sil_xGcx/Si hClcrojuncLion devices. However, TEM does not show more dislocations in Si films grown at -5000C than at -8000C. It is found that annealing generally
improve film quality and so the device perfonnance, such as ideality factor and breakdown characteristics, particularlly for the devices with MBE·Si buffer layers. Electron irradiation shifts the forward and reverse I-V curves of Sil.xGex/Si devices towards lower voltages. It may be due to that irradiation electrons relax the strained Sil.xGex/Si bonds easily, resulting i~ a higher defect concentration at heterojunclion inLerfaces and in the epitaxy strained layers. Electrical measurement and calculation show Lhat the defect assosiated recombination-tunneling process is important in current transport for these MBE grown Si homojunction and strained Sil_xGcx/Si heterojunction devices.
(9: 20 a.m.)
(Invited)
DISLOCATION STRUCTURES IN Ge-Si ALLOYS: R. Hull, J.C. Bean and M.L. Green, AT&T Bell Laboratories, 600 Mountain Ave., Murray Hill, NJ 07964.
The relaxation of strained GexSi t _ x alloy layers grown epitaxially upon Si
substrates is studied by in-sit.u relaxation experiments in a transmission electron
microscope (TEM). By growth of metastably strained GeSi/Si structures at relatively low crystal growth temperatures (say < ~ BOOOe), relaxation via introduction of misfit dislocations may be induced in the TEM by annealing to temperatures ~ the original growth temperature. Thus dynamic dislocation nucleation, propogation and interaction events may be observed and recorded in real time. Quantification of these processes provides the framework for a complete understanding of the relaxation process.
Relaxation processes in structures grown by Molecular Beam Epitaxy (MBE) and Rapid Thermal Chemical Vapor Deposition (RTCVD) will be compared. It is found that a significantly greater regime of metastable growth exists for MBE films. Finally, extension of these experiments to structures akin to heterojunction bipolar transistor structures will be discussed with respect to the effects of the Si capping layer and contact implantation processes.
1M-Annual Meeting (9:50 a . m.)
THE KINETI CS Of DISLOCATI ON GLIDE IN SiGe ALLOY LAYERS g Tuppen a nd C J Gibbings. British Telecom Resear c h Laboratori e s . Martles ham Hea th . Ips wi ch IPS 7RE. UK.
~
Di s loca lion glide ve l oc it y is one fact o r tha t con tro ls the rate of s train relaxation in structure s grown beyond the ir
equilibrium critical thickness (he). A si mple technique will be pre sented
[or measuring glide veloc i t i es in strained layer It has been applied t o t he SiGe materials system a nd glide velocities have bee n measur ed as a function of tempe r a ture, Ge content and layer thickness. A range of layers of const ant Ge content were grown wi th thi cknesses (h) in the range he
struc tures.
stress values (the driving force for d i sl ocation glide) without varyi ng the Ge content. Activ at i on e nergies for the glide process have been derived and these wi l l be discussed in terms
MAGNETISM SUPERCONDUCTIVITY AND THEIR INTERPLAY I Sponsored by the ASM-MSD Electrical, Magnetic, and Optical Properties Committee Thursday AM February 22, 1990
Room: Orangewood I Quality Inn
Session Chairman: R.J. Soulen, Jr., Code 6344, Naval Research Laboratory, Washington, DC 20375-5000
of their de pendence on excess stress. The stab ilising effects of a Si capping l aye r have also been examine d and these r esu l ts
will be di cussed in terms of the therma l budgets available for process lng Si/S iGe hete rojunc ti on bipo lar tra ns ist o rs.
(8 :30 a,m.)
(10:10 a.m.)
SUPERCONDUCTIVITY, MAGNETISM, AND THEIR INTERPLAY IN HIGH Tc CUPRATE SUPERCONDUCTORS. M B MilPle, U . California. San Diego. La Jolla, CA 92093.
BREAK
(10: 25 a.m •.)
DEFECT MICROSTRUCTURE IN SINGLE CRYSTAL SILICON THIN FILMS GROWN AT 150°C BY REMOTE PLASMA·ENHANCED CHEMICAL VAPOR DEPOSITION T. Hsu, B. Anthony, L. Breaux, S. Banerjee, and A. Tasch. University of Texas, Austin, TX Low temperature silicon epitaxy is critical to future generation ultra-large scale integrated circuits and silicon· based hetez:ostructures. In this study, remote plasma·enhanced chemical vapor deposition has been used to demonstrate silicon homoepitaxial films at temperatures as low as l500C, the lowest temperature reported for silicon epitaxy. The process relies on a stringent ex·situ preparation and loading of samples coupled with an in·situ remote hydrogen plasma clean. The in-situ clean is performed at temperatures at or below 300"C with a total pressure of 45 mTorr for 30 minutes at low rf powers «15 W). The growth proceeds by remotely exciting the source gas, silane, with energetic species extracted from the noble gas plasma, such as Ar metastable gas atoms and energetic electrons. Typical parameters for the growth are 200 mTorr total pressure, a flow rate of 360 sccm, and a silane flow of 0.3 sccm. Characterization of the films has been performed in·situ using RHEED and AES and ex·situ using TEM and dilute Schimmel defect etching followed by Nomarski interference contrast microscopy. The films grown at 150°C to date have a thickness of 250A - 350A and show excellent crystallinity by the above methods. (10 : 45 a .m.)
DISLOCATION CONFIGURATIONS AT AN SixGeI_x/Si INTERFACE: Krishna Rajan, f-laterials Engineering Department, Rensselaer Polytechnic Institute, Troy, NY 12180-3590 A detailed weak-beam transmission electron microscopy study of misfit dislocations at an SixGel_x/Si strained layer interface is presented. Direct evidence of dissociated misfit dislocations is provided. Constrictions in these dislocations have been observed. The dislocation configurations are similar to those observed in deformed bulk Si and Ge. An estimation of core cut-off parameters is made from the analysis of dislocation nodes. (11:05 a.m.)
(Invited)
Si,_.Ge. ALLOYS: MBE GROWTH. PROPERTIES AND APPLICATIONS S.S. Iyer, V.P. Kesan. F.K. LeGoues. J.e. Tsang. and J.L. Freeouf IBM Research Division. TJ . Watson Research Center. PO Box 218. Yorktown Heights. NY 10598. Si, _. Ge. alloys grown on Si substrates have several applications in band engineered electronic devices. TIlls talk will focus on growth and doping of such alloys using Molecular Beam Epitaxy (MBE). We discuss special considerations required for the growth of mgh (x > 0.5) Ge layers. The structural and interface quality of SijSi,_.Ge, (0 S; x s; I) heterostructures have been studied by Transmission Electron Microscopy (TEM). and optical methods such as Raman spectroscopy and and ellipsometry. In particular. for ultra·thin pure Ge layers we can identify the evolution of a character· istic bandstructure with increasing monolayer tmcl
We will discuss the interrelation of superconductivity and magnetic order in the new high Te cuprate supercon-ductors . For example. the proximity of anti ferromagnetic (AFM) ordering and superconductivity in the hole-doped superconductors La2.xMxCU04_y (M = Ca, Sr, Ba, Na) and RBa2CU307-l) (R = Y or Ln except Ce, Pro Pm. Tb) and the electron doped superconductors Ln2. xMxCU04.y (Ln = Pr, Nd, Sm, Eu; M = Ceo andLn = Pr. Nd. Sm: M = Th) suggest that a common mechanism may be responsible for both phenomena. The LnBa2Cu307.0 and Ln2.xMxCU04.y com·pounds also exhibit AFM ordering of the Ln(3+) ions which coexists with super-conductivity. The Neel temperatures TN and the magnetic specific heat anomalies of the LnBa2Cu307./\ compounds change markedly upon increase of the oxygen vacancy concentration 0, indicating that RKKY and/or superexchange interactions are involved in the magnetic ordering of the Ln(3+) ions in these materials. The upper critical field of the compound Sm2.xCexCU04_y is enhanced below TN where the Sm(3+) ions undergo AFM ordering, similar to that previously seen in the compound SmRh4B4. Research supported by US DOE Grant No. DE-FG03·86ER45230 and US NSF·Low Temperature Phvsics·Grant No. DMR84-lI839. (9: 15 a.m.) MICROSCOPIC STUDIES OF MAGNETISM IN TIfE HIGH Tc FAMILY OF COMPOUNDS USING NEUTRONS. S K Sjnha. Physics Department, Brookhaven National Laboratory, Upton, NY 11973 and Exxon Research & Engineering Co. Annadale, NJ 08801. We shall review what is currently understood about the nature of the antiferromagnetism observed by neutron scattering and other methods in the copper oxide compounds and its interplay with superconductivity. The details of the magnetic SbUctures and excitations in these materials have yielded infonnation regarding the electronic structure, the applicability of band structure calculations, and possible clues as to the microscopic origins of superconductivity, although a conclusive proof of any particular theoretical model has not yet been established. (10:00 a.m.)
BREAK
(10:30 a.m.) SUPPRESSION OF ANTIFERROMAGNETIC ORDER IN TWO DIMENSIONS. Gerald B. Arnold. University of Notre Dame, Department of Physics. Notre Dame. IN 46556. The copper oxide layer superconductors exhibit phases which possess anti ferromagnetic order and correlations within the two dimensional layers. With doping, which introduces charge into the layers, anti ferromagnetic order is suppressed and superconducting correlations grow. An accurate description of the suppression of antiferromagnetism may therefore provide an explanation of the mechanism for pairing in these materials. The importance of topological excitations in reduced dimensional magnets. the quasi·Lorentz invariance of the critical theory of antiferromagnets (in which spin wave velocity plays the role of the velocity of light) and the importance of fluctuation phenomena are all providing a system which is quite rich from the point of view of field theory. I shall review the current theoretical understanding of the suppresson of antiferromagnetic correlations in two dimensional systems, and discuss the implications of these results for superconductivity. (11: 15 a.m.) THE EFFECT OF MAGNETIC EXCHANGE FIELDS IN SUPERCONDUCTORS. W W Fuller. S.A. Wolf, Naval Research Laboratory, Washington, D.C. 20375·5000, and C.Y. Huang, Lockhead Research Center, Palo Alto, CA 94304 Usually magnetic ions in superconductors break Cooper pairs by their magnetic scattering. depressing Te. An exchange interaction, on the other hand. can either enhance or compensate for the applied magnetic field, and thus significantly modify the critical magnetic field. In the latter case. under some circumstances. the negative exchange field induced by the magnetic ions can polarize the conduction electrons and destroy superconductivity at very low applied fields. An increased applied field can then compensate this exchange field and superconductivity can reappear at a higher field. The material then remains superconducting until the orbital critical field is reached. A review of these effects in both conventional and high Te superconductors will be presented.
Annual Meeting-135
METALLIZATION FOR ELECTRONICS APPLICATIONS V: Epitaxial Heterostructures of Metal/Compound Semiconductors Sponsored by TMS Electronic and Photonic Device Materials and Electronic Materials Committees Thursday AM February 22, 1990
Room: California I-II Quality Hotel
Session Chairmen: J. Raynien Kwo, AT&T Bell Laboratories, Murray Hill, NJ 07974; J.H. Weaver, University of Minnesota, Minniapolis, MN 55455
(11: 15 a.m . ) GROWTH AND CHARACTERIZATION OF IRON ALUHINIDE FILMS ON COMPOUND SEMICONDUCTORS* A.M. Wowchak, J . N. Kuznia, N. Zielinski, P . I. Cohen, and W. W. Gerberich, Electrical Engng, and Chem. Engng . and Mat. Sci. Depts., University of Minnesota, Minneapolis, MN.
(9:00 a.m.) Epitaxial Growth of Magnetic Films on Semiconductors
G.A. Prinz Code 6345 Naval Research Laboratory Washington, DC 20375-5000 Ultra high vacuum growth techniques are now being employed to grow single crystal films of magnetic materials. This growth, carried out in the same MBE systems commonly used for the growth of semiconductor films, has . yielded a variety of new materials and structures whic~ may prove useful for integrated electronics and integrated optical device applications. Examples will be given for growth on GaAs, ZnSe and diluted magnetic semiconductors (OMS), including magnetic sandwich structures and superlatlices.
(9 : 30 a.m.)
METALLIC RARE-EARTH MONOPNICTIDE AND MONOCHALGOGENIDE TERNARY SYSTEMS, LATTICE MATCHED TO COMPOUND SEMICONDUCTORS. l::......L. Palmstr"m,
Tabatabaie
T .G.
Finstad,
and V.G.
S.
Keramidas, Road, Red Bank, N.J. 07701
Moun ier,
S . J.
Bellcore,
331
Allen,
Jr.,
Newman
N.
Springs
Thermodynamically stable epitaxial thin film metallizations, incorporated in compound semicondu cto r heterostructures, open new vistas for transport phenomena studies and n ove l device concepts. The rare-earth mono arsenides ErAs, LuAs and YbAS, recently grown on GaAs, have provided early demonstration of such feasibilities . These compounds, however, exhibit sufficient lattice mismatch to GaAs to cause difficulties at the interfaces and inhibit the study of gr ow th kinetics and overgrowth. SC 1 _ xErxAs, a te rn ary representative of an extensive family of monopnictides
and
monochalcogenides
has
been
grown
lattice
matched to GaAs, removing the mismatch hindrances . Interface chemistry is sho wn to play a dominant r ole in the gr owt h o f these compounds on semiconductors and in the subsequent semiconductor overgrowth. The growth and propert ies of these metallizatio n systems will be discussed.
(10:00 a.m.)
MBE GROWTH AND CHARACTERIZATION Of ARTlfICIALl.Y·LAYEREU MAGNETIC METAL STRUCTURES ON SEMICONDUCTORS.
R.F.C. Farrow dIBM Research Division, Almaden Research Center, 650 lIarry Roa . San Jose, CA 9-'120·6099; There has recentl:1 been an upsurge or interest in artifici~lIy· la yered magnetic metal. structures prepa red by MIJE. This interest has been stimulated by the unu,ual and potentially uscrul magnetic properties which such structures co< hibi!. The« properties include perpendicular anisotropy and giant magnctorc.,istancc in specific structures, .sorne or them grown on Ga,\, substrates. In addition, magnetic ordering in epitaxial films or ra ,'e earths difTers dramatically fr om that in bulk single crystals as a re, ult of long.range coherency strain. In this paper, recent developments in the MilE growth and characterization o f a rtificially-laye red magnetic metal Hruct.urcs will be reviewed and cxarnpics of strtJ<:lurcs hased on (ja;\s ~uhslratcs
presented. In pa rticulal . the value or a high rerfection ,uhstratc, such as GaAs . in q"entirving ',rain in such 'trm,tu re' will he be discussed. (10:30 a.m.)
BREAK
(10:45 a.m.) BURIED INTERMETALLICS IN III-V SEMICONDUcrOR HETEROSTRUCfURES: NOVEL PROPERTIES AND DEVICE PROSPECTS J P Harbison , T . Sands, C.J. Palmstrc,im, N. Tabatabaie, H.L Gilchrist, LT. Florez, T.L Cheeks, R.E. Nahory, WK Chan, and V.G. Keramidas, Bellcore, 331 Newman Springs Rd., Red Bank, NJ 07701-7040. Though metallization is used extensively in semiconductor devices, it is usually applied on top of the semiconductor as a post-growth processing step. In this talk we will concentrate on a carefully chosen metal system, NiAl, which can be epitaxially grown using molecular beam epitaxy (MBE) on (In,Ga,AJ)As semiconductor heterostructures and which does not react with the contiguous semiconductor material even at the elevated temperatUJes required for semiconductor growth (400-600"C). Thus we are able to fabricate heterostructUJes which consist of epitaxial metal layers buried within complex semiconductor structures. We will discuss some of the new device possibilities which this type of structure opens up, as well as some of the interesting properties that these buried metallic layers possess, especially in the ultrathin regime (1-5 run) where they can be describe as metallic quantum wells.
TEM, electron channeling, and RHEED are used to investigate the quality of e.pitaxial iron aluminide filllls grown on GaAs and InP substrates . These films are observed to grow layer-by· layer even when the constituents are codeposited . By growing on these two substrates, the role of lattice mismatch is examined. During growth the development of the lattice constant is measured using RHEED for films up to lOOnm thick. After growth, Moire fringe, micro-micro analytical and selected area channeling methods are applied to sort out defect, stoichiometric, and elastic strain effects . By examining overlayers in the subcritical and supercritical thickness range, data will be presented with the relative contributions due to dislocation relaxation and lattice mismatch. *Supported by the AFOSR and the Univ of HN NSF Center for Interfacial Engineering
(11 :45 a.m . ) UNDERSTANDING THE INFLUENCE OF INTERFACE PROPERTIES AND DEPOSmON PARAMETERS ON THE SELECTIVE GROWTH OF (100) AND (110) CoGa FILMS ON (IOO)GaAs. Delroy A Baugh. Young K. Kim. and R. Stanley Williams, Department of Chemistry and Biochemistry, University of California, Los Angeles. CA 90024. The codeposition of elemental Co and Ga by molecular beam epitaxy
is used to selectively prepare (100) and (110) oriented CoGa films on (100) GaAs surfaces. The propensity towards a particular orientation, (100) or (110) CoGa, is investigated as a function of the relative Co and Ga incident flux (i.e. the film stoichiometry). total deposition rate , and the surface temperature. The effects of predepositing one to several mono layers of Co or Ga on the e1ean GaAs surface prior to film growth and the thermochemical stability of the resulting films are also studied in detail. The primary diagnos tic lechniques utilized In these experiments are Auger
elec!ron spectroscopy (AES). transmission elec!ron microscopy (TEM), two theta x-ray diffractometry (XRD) and x-ray photoelectron spectroscopy (XPS). In addition, a quadrupole mass spectrometer (QMS) and a quartz crystal monitor (QCM) are used for precise online calibration of the relative and total beam flux.
136-Annual Meeting
PROCESS MINERALOGY I: Applications to Hydrometallurgy, Flotation and Liberation Sponsred by TMS Process Mineralogy Committee ThursdayAM February 22, 1990
Room: Salon H Marriott
Session Chainnen: Richard D. Hagni, Dept. of Geology and Geophysics, University of Missouri-Rolla, MO 65401; Daniel C. McLean, 3318 American River Drive, Sacramento, CO 95864 (8:30 a.m.) A MINERALOGICAL AND CHEMICAL ANALYSIS OF CARBONACEOUS ORE AT THE SANTA FE MINE, MINERAL COUNTY, NEVADA: THE K 0 - PREG ROB RELATION: W.A. Fuchs, and V.C . Schurer, Schurer Fuchs, 1555 Ridgeview Dr., 88, Reno, NV 89509
&
A study of carb onaceous ore at the Santa Fe mine (Corona Gold Inc.) shows that there is no quantitative relationship between organic carbon content and preg rob, although much of this material preg robs. However, there is a relation between K20 and preg rob. Preg rob generally occurs only in those samples with greater than 2% K 0. X-ray diffraction and chemical 2 studies show that the critical factor is the illite content of mixed (possibly interstratified) illite - smectite . App·arently illite, at a certain threshold, activates the carbon. The potential for easily monitoring preg rob using a gamma ray spectrometer is discussed.
(8:55 a.m.) MANGANESE MINERALOGY WITH ASSOCIATED COBALT AND GOLD IN THE WEST-CENTRAL ARKANSAS MANGANESE DISTRICT: W.K. O'Connor, and W.K. Collins, Albany Research Center, Bureau of Mines, Albany, OR 97321 The Bureau of Mines investigated the West-Central Arkansas Manganese District to evaluate the Mn and associated metals resource. The study included detailed petrographic and scanning electron microscopic examination on four small head samples and their concentration products from gravity separation and flotation . Manganese mineralization occurs as open-space filling, complete and partial replacement, and incrustation on novaculite fragments. Brecciated novaculite is cemented with a mixture of Mn oxides, chiefly cryptomelane and psilumelane. Cobalt is contained within the Mn, while Au grains are located within the novaculite. Chemical analyses on the head samples reported up to 25 pet Mn, 0.17 pet Co, and 0.002 oz/ton Au. Table concentration tests recovered up to 82 pet of the Mn and 81 pet of the Co at grades of 30 pet Mn and 0.19 pet Co. Novaculite products contained grades of up to 0.013 oz/ton Au.
(9:20 a.m.) CHARACTERIZATION OF THE RARE EARTH MINERALOGY AT THE PEA RIDGE DEPOSIT : C. W. Whitten, and R.J. Yancey, Bureau of Mines, Rolla Research Center, P. O. Box 280, Rolla, MO 65401 Iron ore deposits in south central Missouri have emerged as a possible resource for the rare earth elements (REE). The Pea Ridge Iron Mine has one such deposit. REE minerals at the Pea Ridge Iron Mine are contained in "breccia pipes" extending vertically through the magnetite ore body and appear to be alterations of earlier hematite. The mineralogy of these pipes is complex and varied, which could complicate the concentration .nd extraction of the rare earth minerals . Samples from the Pea Ridge Iron Mine were analyzed by transmitted light, reflected light, and scanning electron microscopy to characterize the mineralogy. X-ray diffraction was used to verify the identifications. The minerals identified were: the major ore minerals, magnetite and hematite; the rare earth phosphates, monazite and xenotime; the rare earth silicate, allanite; the minor associated minerals, caSSiterite, pyrite, and apatite; and the gangue minerals, feldspar, quartz, and actinolite.
(9:45 a.m.) THE MINERALOGICAL CHARACTERIZATION OF "SILICA" IN COPPER ANODES AND COPPER REFINERY ANODE SLIMES: T.T. Chen, and J.E. Dutrizac, CANMET, 555 Booth Street, Ottawa, Canada, KIA OGI Although copper anodes are commonly considered to be "silicafree", anode slimes recovered from electrorefining cells contain modest amounts of various silicate. The presence of "silica" complicates the processing of anode slimes and also may affect the passivation of anodes during electrorefining. Mineralogical study of as-cast copper anodes indicated inclusions of silicates preferentially concentrated near the upper (air) surface. In continuously cast anodes, Cu-Pb silicate inclus ions are relati-
vely plentiful and likely result from the reaction of the silicone oil belt dressing with PbO inclusions in the anode. In anode slimes, silica is dected in the complex Cu-Ag-Se-Pb-As oxide phase, and likely occurs as a gel. Particles of Cu silicate, Cu-Pb silicate and Ca-Cu-Ni silicate also are observed in the anode slimes, and their morphologies and composition suggest they originated in the copper anode. (10: 10 a.m.)
SOLUTION MINING OF MASSIVE SULFIDES - COLUMN EXTRACTIN TEST DATA - Daniel C. McLean. 3318 American River Drive. Sacramento, CA. 95864 Extraction Data for copper. zinc and iron was obtained from 4.5" and 10" column tests using massive sulfide ore (minus I") and natural acid mine water from Iron Mountain Mines near Redding. CA. The purpose of the tests was to determine if recycling of the mine water would produce leach liquors containing 4 g/L copper and zinc at extraction rates of 1 giL per pass thru the column using flow rates of 0.002 to 0.006 gpm/ft 2 • Testtresults exc~d~d ex~ektatioos. Concentrations grea er tnan b g(L Gu ana ~ giL ~n were obtalned at natural acid concetrations in excess of 40 giL. No reagents were required. (10:35 a.m.) LEACHING COMPOSITES OF FIVE PLATING WASTES WITH AMMONIUM CARBONATE: J.W. Keck, D. H. Carlson, C.W. Schultz, and J.Y. Hwang: Michigan Technological University, Houghton, MI 49931 and University of Alabama, Tuscaloosa, AL 35487 There are an estimated 16,000 shops in the United States which plate copper, nickel, chromium, cadmium and zinc. Sludge wastes are typically produced by precipitating the me tals from rinse solutions with lime or sodium hydroxide. While the amount of sludge generated by a single plating shop may be too small to treat economically, the total amount generated in a major city can be large and represent an economic opportunity. A composite sample obtained by blending five sludges from the Detroit area was studied t o evaluate the feasibility of leaching with ammonium carbonate. The advantages of ammonium carbonate include alkaline conditions, a recoverable reagent, little iron contamination and a history of successful commercial operation in copper scrap leaching. The results show that metal recoveries with this approach are: 99.1% Cu, 90.6% Ni, 77.9% Cr, 94.0% Zn, and 87.7% Cd. The effects of reagent concentrations, sludge age, sludge drying, and locked cycle countercurrent leaching are discussed.
:00 a.m.) EVALUATION OF A RANDOM FRACTURE MODEL FOR THE PREDICTION OF MINERAL LIBERATION: J. Bole, C. L. Lin, R. P. King, and J. D. Miller, Department of Metallurgy and Metallurglcal Engineering, 412 Wm. C. Browning Building, University of Utah, Salt Lake City, Utah 84112 (11
The simulated random breakage of irregularly shaped binary PARGEN particles has been used to establish a model for the prediction of mineral liberation. The model is based on random fracture and the conservation of interfacial area. Fractionated ore particles of fixed size and grade are being analyzed in polished section to evaluate the utility of the random fracture model for the prediction of mineral liberation. (11 :25 a.m.) INCIPIENT OVERLOAD CONDITIONS IN BALL MILL GRINDING: Y.C. Lo and C.G. Rhin, Utah Comminution Center, liS EMRO, University of Utah, Salt Lake City, Utah 84112 In recent years there has been a great deal of interest in ways to optimize the performance of tumbling mills. An important part of the performance improvements is to identify and quantify the overload conditions. This paper describes the electronic measuring system and methodology developed for detecting and quantifying the overload conditions in the tumbling mills. Both batch and pilot plant s c ale tests were perfo rmed with vary ing feed material and critical operating variables. The relationship between overload conditions and feed characteristics -and operating variables has been shown to provide a reliable basis for analyzing and predicting the overload conditions in the grinding mills. Since the success achieved in detecting and predicting the incipient overloa d condition, it is expected that this approach will reduce the risk of overloading tumbling mills.
Annual Meeting-137
REFRACTORY GOLD, SILVER & PRECIOUS METALS PROCESSING ITI: New Process Methods for Gold and Silver Sponsored by the TMS Precious Metal Committee Thursday AM February 22,1990
Room: Salon 1 Marriott
Session Chainnen: R. Bautista, University of NevadaReno, Department of Chemical and Metallurgical Engineering, Mackay School of Mines, Reno, NV 89557-0047; R. Sandberg, U.S. Bureau of Mines, Salt Lake Ctiy Research Center, 729 Arapeen Drive, Salt Lake City, UT 84108 (8:30 a.m.) LEACHING OF LOW GRADE PRECIOUS METAL ALLOYS DERIVED FROM SCRAP, D. C. McLean, 3318 American River Drive, Sacramento, CA 95864 Low-grade P.M. alloys used in this work were generated by melting incinerated P.M. scrap with copper to produce hornogenious alloys with the following range of compositions: Cu Zn
56 t o 74% Fe 0.5 " 10% Ag
3.5 to 6% Ni 1 "10% Au
3.5 to 10% 0.1 " 1.2%
Additional metals contained were Pc, Pd, Cd, In, Ph , and Sn. Bench scale teses, confirmed by pilot plant work, demonstrated that a simple sulfuric acid leach of atommetal us ing oxygen, appropriate free acid concentrasions and redox potential control at BODC and atmospheric pressure induced quantitative dissolution of most of the base metals within a 1 t o 2 hour period. Recovery of all P .M.'s in the leach residues was 99.9%. Leach liquors produced were amenable t o st and SX-electrowinning to recover acid and copper. (9:00 a.m.) PROCESS SELECTION FOR A GOLD AND SILVER REFINERY, R.J. Wesely and B. Ledeboer, Davy McKee Corporation, 2430 Camino Ramon, Suite 100, San Ramon, CA 94583 The paper describes the var~ous methods which are currently being used or considered to refine gold and silver. These methods are then compared on a technical and economic basis. Conclusions are drawn on the factors which might influence the choice of process for different feed stocks and e.conomic constraints. (9:45 a.m.) BIOOXIDATIVE CURING TO STIMULATE RECOVERIES IN PRECIOUS METAL HEAP LEACHING, Dr. Keith Prisbrey, A. Burbank, Z. Zhan, University of Idaho, Moscow, Idaho 83843 The problem is to determine the conditions necessary for improving cyanide leach extractions from low grade refractory ores by bio-oxidation of the sulfides. Batch bio-oxidation of sulfide concentrates can often break down the sulfide metal encapsulation and stimulate recoveries in subsequent cyanide leaching. However, to duplica te the process in heaps requires overcoming several problems: sulfide substrates may be too low to sustain the bacterial consortium, calcite or dolomitic minerals may prevent the optimal pH for facterial growth. and the effect of any "preg robbing'f components such as humic acids must be neutralized. We present the results of column biooxidative curing tests on several different low grade refractory gold ores and conclude that heap design must depend on "Vedos" model design parameters including "matr ic" potential or soil energy states, soil saturation or wetness points, and soil "conductivity" or transport parameters.
00:15 a.m.) GOLD LIBERATION BY BIO-OXYDATION OF SULFIDES, Pierre Ollivier, Eric DeBacker, Dominique Morin, Bureau De Recherches Geologiques Et Minieres, BP 6009, 45060 Orleans Cedex, France Bioleaching of a refractory gold ore has been carried out to implement a complete frontend treatment flowsheet for precious metals recovery by cyanidation. A bioreactor has been defined for Lab and pilo t experiments. Extrapolation has been calculated to a 500 m3. Effluent treatment and products insolubility have been studied a flowsheet could be proposed for safe disposal. The cyanidation study was designed for high and economic gold recovery.
(11 :00 a .m.) ~(N;
IN 1lE M:NrnRlN; CF A MURATILY 'IlflMlffiUC GJl1lEE IN TIE
CXIImITN CF IDRIICItRY am CRES AID aNJ
Julia R. li.rlE1 am Dr teter A. S!:e=r, Tran.so.vr1d Minirg am Mirera1s (Alstraha) Pty. ltd., 13-15 fuldNin St., Ka..da1e 61<:6 \-estern Alstra1ia . 1he grcM:h am oodatim clmacteristics of 1I1Y lBcterial rulture are ultirratelv d2termire:l l:7y t:te mture of t:te ~le ~ treate:!. 1he nrst irq:ortrot clmacteristics are lBcterial cell nnter, gnMh rates, 00igffi UJXake an rate of oodatim. Cells prese:1t in t:te solutim of 11"1 establiS'ed rulture Ioere canted directlV usirg a micrcscq:e aU related to a protein valLE,
THERMAL ANALYSIS OF METALLURGICAL SYSTEMS ill: Processing and Microstructural Correlations Sponsored by the TMS Chemistry and Physics of Materials Committee Thursday AM February 22, 1990
Room: California III-IV Quality Hotel
Session Chainnan: J.H. Perepezko, Department of Metallurgical Engineering, University of Wisconsin, Madison, WI 53706 (8:30 a.m.) (Invited) TRANSIENT THERMAL EFFECTS IN SYNTHESIS OF INTERMETALLIG ALLOYS, ~ ~ German! Materials Engineering Department , Rensselaer Polytechnic Institute, Troy, NY 121S0-3590. Several high melting temperature intermetallic compounds have been con solidated utilizing the powder metallurgy techniques of reactive sintering or reactive hot isostatic compaction. These approaches manipulate the exothermic reaction to induce a transient liquid phase. Aluminides of titanium, nickel, niobium, and tantalum have been prepared by these approaches. In some cases, compound synthesis occurs with densification, depending on the initial powder condition and reaction parameters. There are several factors involved in successful control of the transient liquid phase! including heating rate , particle size, pa rticle size ratio , powder homogeneity. green density, degassing procedure! heat of reac tion, and basic phase diagram features. Research sponsored by DARPA/ONR under Contract N00014-S6-K-0770. (9:00 a.m.)
(Invited)
CALORIMETRIC INVESTIGATION OF RECOVERY AND RECRYSTALLIZATION PHENOMENA IN METALS: Frank Haessner, Institut fUr Werkstoffe Technische Universitat Braunschweig, 3300 Braunschweig, Germany
During the annealing of cold worked metals thermal effects occur which are due to the rearrangement and annihilation of lattice defects. They can be measured by a calorimeter and allow investigation of recovery and recrystallization phenomena. In the present contribution mainly the advantages and the limitations of calorimetric methods will be discussed. To this aim the correlation between the state of the material and the calorimetric Signal is treated first. From this, consequences on the utilization of the method will be drawn. Subsequently the measuring strategy and the prinCiple of data analysis is discussed. From this, the advantages and limitations of calorimetric methods follow quite naturally. For illustration of these more or less fundamental considerations quantitative analyses of some actual calorimetric data are presented. The conclusions drawn from this analyses with respect to the processes going on in the material will be evaluated critically. (Invited) (9:30 a.m.l MEASUREMENT OF ENERGY STORAGE AND DISSIPATION BY A DEFORMATION CALORIMETRY METHOD AND THE THERMODYNAMICAL ANALYSIS OF THE DATA. D. Ronnpagel, F. Jager and Th. Pingel T.U. Braunschweig, Institut fur Metallphysik, F.R. of G. We have developed a method of deformation calorimetry which is applied to measure the energy stored and di ssipated during
138-Annual Meeting plastic deformation of metal single crystals.
Bar-shaped
forgings. Exploration of the precipitation characteristics of modified alloy 644B where Zr is replaced or supplemented by Ti or Hf was explored by DSC and TEM. Preliminary results indicate that the propensity to form 6 in Hf containing alloys is much greater than for the Zr and Ti alloys.
specimens are deformed in tension at room temperature. The evaluation of the measured temperature curves, the main
difficulty of deformation calorimetry, is performed by means
of a respons e function being determined experimentally. temperature change of the specimen can be represented mathematically in terms of this response function and
Any
subsequently be evaluated numerically. The physical differences between the stored energies measured by annealing and deformation calorimetry, the question of the most convenient thermodynamic definition of stored energy, and changes of specific heat are treated. (10:00 a.m.)
TITANIUM II: Advances in Melting Practice Sponsored by the TMS Titanium Committee
:Invited)
THE USE OF DIFFERENTIAL SCANNING CALORIMETRY FOR PRECIPITATE CHARACTERIZATION IN ALUMINUM ALLOYS WITH PARTICULAR APPUCATION TO COMMERCIAL AJ-U ALLOYS, J.M. Papazian, Grumman Corporate Research Center, Bethpage, NY 117143580 The use of differential scanning calorimetry (DSC) for characterization of the precipitate microstructure in aluminum alloys will be described. The basic technique will be reviewed, and examples of the correlation between DSC and transmission electron microscopy results will be presented for thealurninum-copper alloy 2219. DSC results from the characterization of precipitation in pure binary and multi-component commercial AJ-Li alloys will then be presented. The use of DSC for evaluation of the effects of heat treatment and deformation on the precipitate microstructure will be described and illustrated with results from alloys 2090, 2091, 8090 and IN905XL. These results will demonstrate the utility of DSC for rapid comparison of the effects of heat treatment, deformation and alloy composition on precipitation.
(10:30 a.m.)
Invited
CALORIMETRIC ANALYSIS OF PARTICLE COARSENING IN A HIGH TEMPE~lUR? AL~~ ALLOY: R. D. Doheny, Dept of Matis Engrg, Drexel Umverstty, PhlladelphIa, PA 19104, R. E. Lewis and D. L.Yaney, Metallurgy Dept.,LMSC, 3251 Hanover St., Palo Alto, CA, 94304, Y. C. Chen, Matis .Res. Lab~, Industrial Technology Institute, 195 Chung-Hfing Rd, Chutung, HSlOchu, Tlilwan 31015, Republic of China An AI-Zr-V alloy designed for high thenna! stability to 5000C was prepared by rapid soli~cation processi.ng involving melt spinning of thin ribbons, vacuum h~t degaSSl~g .and hot presslOg, followed by extrusion. The thermal stability in thiS ~ystem 15 unparted ~y slow coarsening of metastable, coherent AI3(ZrxV I-xl preCIpItates nucleated 10 the supersaturated, rapidly quenched solid solution. ~ ~oarsening beh~vior of these Ll2-type precipitates was measured by transmISSIOn electron mIcroscopy after exposures at 425, 450 and 5000c. Calorimetric measurements were also performed at the same temperatures in the attempt to ~orrelate the energy change, dF/dV, with observed coarsening rates. The expenmental results are oompared with analytical predictions for both particle coarsening and surface oxidation exothermic reactions. The results are encouraging, in the light of the real distribution of initial particle radius, rand the high sensitivity of energy released in the coarsening reaction, which is a function of r4.
(11:00 a.m.) Invited) PreCipitation of 5' Al)(Li,Zr) Phase in SiC p Reinforced AI-Li Base Composites, M. Zedalis & D.J. Skinner, Allied-Signal Inc., Metals Laboratory, P.O. Box 1021R, Morristown , NJ 07962 P:ecipitation of 5' A13(Li,Zr) phase in SiC p reinforced Al-3.1 L.- 1.0 Cu- 0.5 Mg- 0.5 Zr (alloy AL644) composites has been (DSC) and examined by differential scanning calorimetry transmission electron microscopy (TEM). Peak hardness and strength of monolithic A1644, and 5 and 15 vol. % SiC p reinforced alloy AL644 were obtained after aging at 130·C for approximately 16, 9 and 2 hours, respectively. DSC analyses of the
composite
samples,
aged
for
times
corresponding
underaged, peakaged and overaged conditions, indicate that hardness
in each of the respective samples corresponds
to
peak
to
the
precipitation of 5'. The presence and precipitation of this phase was verified by TEM. The reduced aging times to achieve peak hardness for the composite samples versus the monolithic alloy are suggested to be due increased dislocation densities in
the matrix arising from the. disparity in thermal coefficients of the AI-Li -base alloy matrix and reinforcement .
Comparison of observed and required
densities for enhanced precipitation will be made.
expansion the SiC
dislocatio~
(Invited) (11 :30 a.m _) Al3(Li,X) Precipitation Phenomena in Rapidly Solidified Al-Li, J.C. LaSalle, D.J. Skinner and V.R.V. Ramanan, Allied-Signal Inc., Metals and Ceramics Laboratory , P.O. Box 1021R, NJ 07962 Precipitatio~ of non~h:a~able 6' A13(Li,Zr) compound precipitates 1n rap1dly SOlld1f1ed Al-2.6Li-l.OCu-0.5Mg-0.5Zr (alloy 644B) has resulted in the production of an aluminum lithium alloy having highly isotropic mechanical properties in large
Thursday AM February 22, 1990
Room: ParkA Inn at the Park
Session Chairman: P.J. Bania, TIMET, Henderson Technical Laboratory, P.O. Box 2128, Henderson, NV 89015 (8 : 30 a.m.) TITANIUM BASE ALLOYS CLEAN MELT PROCESS DEVELOPMENT, Clifford E. Shamblen and Go"don B. Hunter, G. E. Aircraft Engines, 1 Neumann Way Cincinnati, OH 45215-fi301 Efforts to develop cleaner titanium base alloy materials using the Vacuum Arc Remelt (VAR) process and enhanced nondestructive inspection techniques have been in progress with However, more out partial success since the early 1970's. recent process development efforts initiated by G.E. Aircraft Engines in 1983 are presented which confirm that the capability for eliminating both Type I Hard Alpha and HOI's is best achieved via hearth refining. It is proposed that hearth refining can be achieved using either the Electron Beam Melting (EBM) or Pl asma Arc Mel ti ng (PAM) processes. Based on the excellent EBM process development results and on extensive EBM + VAR producibility demonstrations, GE Aircraft Engines has recently introduced the hearth melt plus VAR process into production for premium qual ity titanium alloy rotating disk components. (9:00 a.m.) SOME ASPECTS ON COMPUTER MODELING OF TITANIUM VAR MELTING, J. R. Faber, Timet, Henderson Technical Laboratory P. O. Bo x 212R, Henderson, Nevada QQ015 A mathematical model simulating the VAR melting process will be reviewed. This model simulates the heat transfer processes that occur in a titanium VAR melting furnace. Computer simulation studies will be presented showing the effects of melting parameters on pool characteristics. Melting parameters reviewed include the following: Melt rate, alloy composition, melting polarity, Crucible thickness, crucible diameter, and crucible surfa ce condition. (9:30 a.m.) A REDIRECTION FOR ELECTRON BEAM COLD HEARTH TECHNOLOGY, C. H. Entrekin and R. H. Harker, Axel Johnson Metals, Inc. 215 Welsh Pool Road, Lionville, PA 19353 Electron beam cold hearth refining technology was used in the 1970's and early 19~O's primarily to recycle and upgade titanium scrap . Within the last four years the process has evolved from a recycling tool to a major metallurgical refining operatlOn. Intense development efforts have been aimed at applying this technology to the improvement of fatigue properties of materials for both aerospace structural and engine components. The growth of the technology in recent years will be reviewed including improvements in hardware, beam control techniques, and the construction of a next-generation furnace . (10 :00 a.m.)
BREAK
(10:30 a.m.) PROCESS CONTROL CONSIDERATIONS IN PLASMA MELTING, M. Eric Schlienger, Retech, Inc. P. O. Box 997, Uk1ah, CA 95482 In Plasma Melting, proper torch motion control, arc gas control and withdrawal control are key factors in the production of a quality ingot. The techniques for proper control and the benefits of successful appl ication will be examined. Complex torch motion is achieved via a CAD like programming environment ~hi.ch will be reviewed. Changes in gas pressure, gas compos1t10n and gas flow rate have substantial effects on melting parameters which wiil be discussed. The path of the withdrawal ram is p~e-programmed and graphically displayed us!ng .software technlques. The use of video imaging to malntaln a constant pool height will be described.
Annual Meeting-139 (2:25 p.m.)
(11 :00 a.m.) PLASMA COLD HEARTH REFINING OF TI ALLOYS, S. H. Reichman, R. N. Jarrett, Wyman Gordon, Eastern DivisIon, 244 Worcester Street, Box 8001 North Grafton, Massachusetts 01635-g001
ROLE OF RESIDUAL STRESSES IN DAMPING BEHAVIOR OF Gr/AI COMPOSITES, S P Rawal. lH. Annstrong, and M.S. Misra, Martin Marietta Astronautics Group, P.O. Box 179, Denver, CO, 80201
This presentation will provide an update of the status of the 2 MW Plasma Cold Hearth Refining furnace being placed into production for the processing of aerospace Ti alloys at WymanGordon. The rationale for the process vis a vis new specifications for premium grade Ti alloys will be discussed, as well as the results of recent experience in the processing of Ti (6-4) through the system. Discussions relating to feedstock, melting, and subsequent ingot structure will be updated. (11 :30 a.m.) PLASMA HEARTH MELTING - A P&W PRESPECTIVE Dave Halley, Pratt & Whitney, 400 Main Street, Mall Stop 122-05, East Hartford, CT 06108
Diffusion bonded Grl Al composites exhibit an increase in inherent damping capacity as the vibrational strain amplitude increases above 20 11£. Whereas, after a cryogenic quench to relieve residual stresses in the matrix, the damping remains nearly constant at strain amplitude level between 20-300 11£. The absence of strain amplitude dependent damping is related to the reduced residual stress state near the fiber-matrix interface. Subsequently, if the specimens are heated beyond a stress free temperature and slowly cooled to room temperature, the composite exhibits an increase in damping with strain levels, similar to the as-fabricated condition. In both cases, the presence of strain amplitude dependent damping has been attributed to the existence of a similar level of residual stress state and associated dislocation density near the fiber matrix interface. (2:42p.m . )
THERMAL FATIGUE IN SiC FIBER REINFORCED ALUMINOSILICATE GLASSCERAMIC COMPOSITE. Youngman Kim, Won Jae Lee and Eldon D. Case, .
A pftot plant 750 kW Plasma Arc MeltIng (PAM) furnace has been operatfng at Pratt & Whitney for two years. The configUratIon and capabilItIes of the furnace and the operational characteristIcs of the torches wIll also be discussed. White thIs new melting method results In excellent chemical homogeneIty, Increased scrap utilIzatIon and fewer meltIng cycles for a varIety of alloys, a major objectIve is to develop advanced hearth melting practIces to produce defect-free titanIum alloys. A program is underway to qualify PAM + VAR for producIng rotor grade TI-6AI-4V ~aterial.
Department
of
Metallurgy,
Mechanics
and
Materials
Michigan State University, East Lans ing , MI 48824.
Science,
Thermal shock damage was monitored non-destructively by Young's modulus and internal friction measurements on SiC fiber reinforced aluminosilicate glass- ceramic composites. As the cumulative number of thermal shock cycle increased, the Young's modulus of the composite decreased, while the internal friction of the composite increased. Both Young's modulus and internal fricti on reached a steady state value after a sufficient number
of thermal shock cycles. Thermal fatigue damage in SiC fiber/LAS
composites is compared to previous thermal fatigue results for
pol ycrysta lline alumina and SiC whisker /a lumina composites.
(2:59p.m . )
TRIBOLOGICAL BEHAVIOR OF LAMINATED SiC/Al 2 0 COMPOSITES. 3 J. L. Bain, J. R. Fox, M. F. Amateau and J. C. Conway, Jr.,
THURSDAY PM
Enginee r ing Science and Mechanics Department, Penn State
University, University Park , PA 16802.
Laminated silicon carbide whisker reinforced aluminum oxide comp osi tes have been fabricat e d from tape cast sheets by
hot pressing. Steady-state wear behavior, produced by blockon-disk and ring-on-disk sliding under both lubricated and
ADVANCED METAL & CERAMIC MATRIX
COMPOSITES VII: Damping, Tribology and
NDC Sponsored by the TMS Powder Metallurgy Committee, TMS Physical Metallurgy Committee and TMS/ASM Composites Committee Thursday PM February 22, 1990
Clemson, SC
Session Chainnen: Toshiro Kobayashi, Toyohashi, University of Teehnology, Tempaku-eho, Toyohashi, Aiehi, 440, Japan; William E. Frazier, NADC, Structural Materials, Wanninster, PA 18974
(2 :00 p.m.)
CONTROL OF DAMPING CHARACTERISTICS OF GRAPHITE FIBER REINFORCED ALUMINUM COMPOSITES. M. J. Pechersky, R. B. Bhagat, C. A. Updike and M. F. Amateau, Applied Research Laboratory, The Pennsylvania State University, Post Office Box 30, State College, PA 16804. Recent developments in the manufacture of graphite-aluminum composi tes may allow for the separate control of damping and In this investi-
gation' Gr/Al composites were fabricated by high pressure infiltration casting (HiPIC). This processing leads to minimal fiber-to-matrix
reaction
and
the
fiber/matrix
interface
practically free from interfacial reaction products.
is
Damping
measurements were performed on the fabricated composites using a non-contact technique based on a laser vibrometer. Beam type specimens were seismically suspended at the nodal points for a particular mode of vibration and excited electromagnetically over a narrow frequency range about the resonance frequency. The
damping loss factors were obtained by the half power point method based on the measurements. It is found that the fabricated composites exhibit enhanced damping over that of the matrix and
the fiber .
Suitable mechanisms are proposed.
(3: 16 p.m.) EFFECT OF AGING ON ABRASION RESISTANCE OF A SIC WHISKER REINFORCED 2124 AL ALLOY. Aiguo Wang and Henry J. Rack, Department of Mechanical Engine e ring, Clemson University,
Room: ParkD Inn at the Park
other mechanical properties of composites.
unlubricated conditions, was examined. The effect of compOSition, ranging from 0.1 to 0.4 fibe r fraction, and fiber orien tation were determined agains t carbon graphite mating disks. A number of specimens were fabricated by gr ading compo si tion through the thickness direction t o induce hi gh compressive stresses on the s liding s urface . The effect o f these compressive s tresses on the wear mechanism was determined.
29634-0921, USA
Investigations are underway to study the response of abrasion resistance of a Siew reinforced 2124 alloy to aging treatments . The materials contain 0 to 20% Sie w in a 2124 Al matrix (p/m process), abrasion te s ts havin g been conducted by means of a pin-an-disk confi guration. Two kinds of abrasive counterparts have been used in the ab rasion tests. The abrasive resistance has been correlated with both the microstructural parameters of the composites and the external experimental conditions. Emphasis has been g iven on the in fluen ce o f aging treatments on bo th the microstructure changes of the mat ices and the ab r asion resistance of the composites.
(3:33 p.m.) THE TRIBOLOGICAL BEHAVIORS OF AI-SiC COMPOSITES IN DRY AND STEAM ENVIRONMENTS. Ram Kossowsky, Applied Research Laboratory, The Pennsylvania State Univer sity , Post Office Box 30, State College, PA 16804 and Robert Brown, Southwest Research Institute, 6220 Culebra Road, Pos t al Drawer 28510 , San Antonio,
TX 78284.
A special pin-on-disk apparatus has been constructed t o study the tribologica l characteristics of materia l s in steam environment. The pin materials s tudied were Ai-SiC composite, which was compared to aluminum al l oy , AI-bronze and nitrided
Ti-6Al-4V, all wearing against low al l oy s teel disks.
It was
in t e resting to note that the ranking of the fri ction and wear behaviors of the various materia l s varied with environment, i . e ., dry, dry steam, and lubr ica ted steam. In all cases, howeve r, the wear of AI-SiC composite was lower.
The behavio rs of the materials will be discussed in terms of deformation wear, with empha sis on fracture mechanics concepts ~
140-Annual Meeting (3:50 p.m.) SLIDING WEAR BEHAVIOR OF ALUMINUM-BASED METAL MATRIX COMPOSITES. C.P. You, J.M. Boileau, and W.T. Donlon, Research Staff, Ford Motor Company, Dearborn, MI 48121. The sliding wear behavior of several 2xxx aluminum composites with different reinforcement volume fractions and types is the subject of the present study. In particular, research aimed at developing an understanding of the micromechanisms of wear in these materials will be described. SEM and TEM techniques are used to determine the nature of the transfer layers present on worn surfaces and the sub-surface deformed microstructures. In addition, the details of wear of the sliding counterfaces are examined to provide an understanding of the complete tribosystem.
(4:07 p.m.)
WEAR CHARACTERISTICS OF OXIDE,CARBIDE AND NITRIDE PARTICLE DISPERSED NICKEL BASE COMPOSITES.M.K.Totlani, M.N.Joshi, G.P.Tiwari, and P.Ramakrishnan~B.A.R.C., Bombay 400085 &*Dept.of Met.Eng.,Indian Institute of Technology,Bombay 400076,India. Nickel base particle dispersed composites containing 2 to 12.7 wt. % dispersoid particles of A120:3, TiG" ZrO:z, Si3 N4 and SiC ttave been prepared J:iy electroaeposition. Mechanical 'properties and abrasive wear resistance of these composites have been evaluated in the as prepared ando heat-treated conditions upto a temperature of 1000 C. The UTS and Y.S of the composites in the as prepared ( & heat-treated) condition were higher of the order of 10-35% than the pure nickel depending on the dispersoid type and its content in the composite. Abrasive wear-resistance, measured as Taber Wear index were in the range of 7 to 11 in relation to 27 for pure nickel. These results have been discussed in detail. (4:24 p.m.) EROSION OF SHORT 6-A1203 FIBER ALUMINUM ALLOY METAL MATRIX COMPOSITES BY SOLID PARTICLES. Aiguo Wang and I. M. Hutchings, University of Cambridge, Department of Materials Science and Metallurgy, Pembroke Street, Cambridge CB2 3QZ, United Kingdom. Investigations are reported on the response of short ~A1203 fiber reinforced Al alloy metal matrix composites to solid particle erosion. The composites which are based on Al 6061 and contain different volume percentages of short 8-A1203 fibers were subjected to both SiC and silica particle impingements at various angles. It was found that the erosion resistance decreased with fiber volume fraction at all impact angles when SiC particles were used while a slight increase in erosion resistance was achieved at impact angles less than 30° when silica sand was used. It was also observed that under all test conditions employed the critical impingement angle at which the peak erosion rate occurs was shifted to higher values as the fiber volume fraction increases. It was suggested that short
~!~i~t~~~~f6fc:t~~~n~;saY~~y; ~~o~oi~~s~~~ti~l!~:rove erosion (4:41 p.m.) MODERN ULTRASONIC NDC FOR ADVANCED CERAMICS DEVELOPMENT, Mahesh C. Bhardwaj. Ullran Laboralories. Inc., 139 R. N. Gill Street, State College. PA 16801 USA Until recently the subject of NonDestructive Characterization (NDC) escaped the attention of materials commun~y. On the other hand, metals NDT (Testing) is not satisfactorily applicable to materials varying in compositions and microstructures. Logical developments lor NDC were, therefore imminent. This paper summarizes more than 10 years of important advancements in unrasonic NDC from our laboratory. We will present an ultrasonic classification of ceramics and microstructurally complex materials lor better understanding and applications of ultrasound·material interaction phenomenon. Reliability of NDC is established through objective·oriented and practical concepts such as. materials and applications suitable acoustics and methodologies. Simple. yet powerful NDC methods are applied for the analysis of green and sintered ceramic·metal and metal·metal compos~es. Relevance of this investigation is shown for early defect detection in green materials; and for dens~y, porosity, elastic properties. and microstructural parameters of sintered materials. Texture and density variations are highlighted by high resolution unrasonic microscopy.
AGING AND LIFE EXTENSION OF NUCLEAR REACTORS II: Other Components Sponsored by the TMS-ASM-MSD Nuclear Materials Committee Thursday PM February 22, 1990
Room: Garden Rooms 1·2·3 Inn at the Park
Session Chairmen: J.R. Weeks, Brookhaven National Laboratory, Building 130, Upton, NY 11973; J.B. Darby, Jr., U.S. Department of Energy, ER-13, G235, GTN, Washington, DC 20545 (2:00 p.m.) LIFE EXTENSION FOR NUCLEAR STEAM GENERATORS: G. J Theus. Materials Performance, ARC, Babcock and Wilcox, Alliance, OH. The problems currently limiting the performance of nuclear steam generators and sometimes resulting in their premature replacement will be described. Methods of correcting these problems and extending the life of defective steam generators will be presented. (2:45 p.m.)
REACTOR AGING EFFECTS ON STRESS CORROSION CRACKING RESISTANCE OF AUSTENITIC STAINLESS ALLOYS: S. M Bruemmer, Pacific Northwest Laboratory. Richland, WA 99352. Long-time, low-temperature service exposure can promote microchemical changes and an increased susceptibility to stress corrosion cracking (SCC) in reactor system piping and tubing materials. This phenomenon has been referred to as lowtemperature sensitization (LTS) and has been demonstrated to occur in several austenitic stainless alloys including 304 5S and Inconel alloy 600. Current understanding and direct observations of LTS will be reviewed and critical issues discussed. Initial material condition and prior/simultaneous deformation may control LTS development during reactor aging. The ability to model lowtemperature microchemical evolution and SSC susceptibility will be examined and compared to the existing data base.
(3:30 p.m.) LIFE EXTENSION AND AGING OF PUMPS AND VALVES: John H. Taylor, Department of Nuclear Energy. Brookhaven National Laboratory. Upton, NY
11973.
Life extension of active moving components such as pumps and valves, must be approached on two levels: 1) the continuous ongoing maintenance and surveillance activities necessary to mitigate aging and sustain performance, and 2) identification of the associated passive subcomponents which must be addressed to satisfy long term operability needs. Evaluation of designs and operating experience provides the key to identifying the failure modes, mechanisms, and causes of age-related degradation. Examples of aging analyses and life extension recommendations and long term needs will be presented for active components in both emergency standby systems and continuously operating systems. (4:15 p.m.)
ASSESSMENT OF THE SIGNIFICANCE OF AGING OF CONCRETE STRUCTURES IN NUCLEAR POWER PLANTS: D, J. Naus, C. B. Oland, and M. F. Marchbanks, Oak Ridge National Laboratory (ORNL), Oak Ridge, TN 37831; E. G. Arndt, U. S. Nuclear Regulatory Commission (USNRC), Washington. DC 20555. A structural aging (SAG) program was implemented at ORNL, in March 1988, to prepare an expandable handbook to provide USNRC license reviewers and licensees with the following: (1) identification and evaluation of the structural degradation processes; (2) issues to be addressed under nuclear power plant continued service reviews, as well as criteria, and their bases, for resolution of these issues; (3) identification and evaluation of relevant in-service inspection or structural assessment programs in use. or needed; and (4) issues still unresolved that need continuing or future research for development of criteria, including recommended priorities based on perceived need. Progress to date in achieving each of the SAG program goals will be presented.
Annual Meeting-141
CAST SHOP TECHNOLOGY VIn: Advances in Casting Technology Sponsored by the TMS Aluminum Committee Thursday PM February 22, 1990
Room: Salon 4 & 5 Marriott
Session Chairman: L. Katgerman, Alean International Limited, Banbury Laboratories, Banbury, Oxon OX16 7SP, England (2:00 p.m.)
VISUAL OBSERVATIONS INSIDE AN AIRSLIP MOLD DURING CASTING: J. M. Ekenes, Wagstaff Engineering Incorporated, N. 3910 Flora Road, Spokane, WA 99216 and W. S. Peterson, Consultant, 2113 East 37th Avenue, Spokane, WA 99203.
In conventional DC casting, the molten metal meniscus is visible to the naked eye. In hot-top casting, this region is masked from direct observation by refractory and molten metal. The advent of air-casting technology has made this region once again visible. This is made possible with the aid of fiber optics. Instrumentation and technique are described for making these observations. The technique is useful in gaining a better understanding of factors influencing the air-casting process. Sample observations are provided. (2:25 p.m.)
PROGRESS ON TIlE DEVEWPMENT OF A SECOND GENERATION TWIN ROLL CASTER: P. M. Thomas, Davy McKee (Poole) Limited, Wallisdown Road, Poole, Doreset, England BHI2 SAG and R. Cook and D. J. Browne, Oxford University, Department of Metallurgy and Materials Science, Parks Road, Oxford, England OXI 3PH. It is generally accepted that in conventional twin roll casting there is a significant variation in the rate of heat removal along the contact length, and it is only when a high volume fraction of solid metal is present that there is good thermal contact with the rolls. Moreover, it is this variation in heat removal that is responsible for the low productivity and limited alloy capability that characterize the process. If higher values of heat transfer could be achieved it is probable that increased productivity and enhanced metallurgical properties would result. Mathematical modelling of the process has demonstrated the possibility of significant increases in productivity as the thickness of the cast strip is progressively reduced. To investigate this and the variation in heat transfer in the contact length, DAVY MCKEE, in conjunction with OXFORD UNIVERSITY, has built an experimental caster capable of high speed, thin strip casting. The machine has a high degree of instrumentation that allows the rate of heat transfer between the rolls and the strip to be measured. This paper will describe the results of this experimental program. (2:50 p.m.)
STRUCTURE AND MECHANICAL PROPERTIES OF A STRIPCAST AI-Si ALLOY: E. Essadigi, S. Caron, F.G. Hamel and J. Masounave, Industrial Materials Research Institute, National Research Council Canada, 75 De Mortagne, Boucherville, Quebec, Canada J4B 6Y 4 1\ commercial AI-Si casting alloy containing 7.4% Si and 0.3% Mg has been investigated by using horizontal twin-roll caster. Strips of 3 to 5 mm in Ihickness and 100 mm in width were produced. Samples of the as-cast alloy were solution heat treated at 540°C and aged at 150°C for different times. Particular attention was given to the mechanical properties of the as-cast and heat treated alloy. The ultimate tensile strength (UTS), the elongation and the yield strength were evaluated with respect to the solidification structure, solution heat treatment and aging. They were compared with those obtained by different casting processes such as chill and squeeze-casting. (3:15 p.m.)
characterized and the cooling rates derived from the secondary dendrite arm spacing measurements are compared with the calculated ones.
(3:55 p.m.) 3-MOS : a general FEM code for the prediction of microstructures in castings Ph. TMvoz, M. Rappaz and J.-L. Oesbiolles. Physical metallurgy laboratory, Swiss Federal Institute of Technology, Ch. de Bellerive 34, CH-1015 Lausanne, Switzerland.
3-MOS ("Micro-/Macroscopic Modeling Of Solidification") is a computer code which was developed at the Swiss Federal Institute of Technology in Lausanne to model microstructure formation in solidification processes. While based on a standard finite element method with an enthalpy formulation, the code is able to predict columnar and equiaxed microstructures for both dendritic and eutectic alloys. For columnar solidification; microstructure formation is predicted from the speed of the calculated isotherms. In the case of equiaxed growth, a special coupling has been developed to integrate the microscopic calculation of nucleation and growth into the macroscopic heat flow calculation, including convection. Beside temperatures, fractions of solid, gradients and local solidification times, parameters characterizing microstructures such as undercoolings, grain sizes, dendritic or eutectic spacings can be obtained. Results of calculations on real castings of AI-alloys (dendritiC) and cast iron alloys (eutectiC) will be presented. (4:20 p.m.)
PHASE EQUILIBRIA AND SOLIDIFICATION OF AL-RICH AL-LI-CU ALLOYS: Sinn-wen Chen, Howard Beumler, Y. Austin Chang, MS&E Dept., University of Wisconsin-Madison, Madison, WI 53706; Men G. Chu, Alcoa Laboratories, Alcoa Center, PA 15069. The primary phases, reaction sequences and temperatures during solidification of AI-rich AI-Li-Cu alloys were studied using optical microscopy, EPMA, XRD, SEM and DTA. The eutectic temperature determined by DTA was found to be 522°C. An invariant reaction temperature at 541°C corresponded to the L+TB-e+Tl reaction. A new liquidus projection was proposed. Based on our experimental results, data in the literature and thermodynamic models of Al-Li, AI-Cu and Cu-Li, a thennodynarnic description of AI-Li-Cu ternary system on the AI-rich comer was developed. Combining the thennodynamic model and Scheil equation, quantitative solidification paths were described. The calculated amounts of the primary phases fonned during solidification were compared to experimental determinations.
IDGH TEMPERATURE DEFORMATION, FRACTURE AND APPLICATION OF INTERMETALLIC COMPOUNDS IV: Engineering Applications of Alternate Materials Jointly sponsored by ASM-MSD Flow and Fracture Committee and TMS High Temperature Alloys Committee Thursday PM February 22, 1990
Room: Plaza B Inn at the Park
Session Chairmen: Michael V. Nathal, NASA, Lewis Research Center, MS 49-3, 21000 Brookpark Road, Cleveland, OR 44135; Richard G. Menzies, GE Aircraft Engines, Mail Drop M85, Cincinnati, OR 45215
BREAK
(3:30 p.m.) NUMERICAL MODELING OF THE HEAT TRANSFER IN WIDE CONTACT ANGLE HORIZONTAL STRIP CASTING PROCESSES: S.Caron, E Essadiqi,
(2:00p.m.)
F.G.Hamel and J.Masounave, National Research Council of Canada, 75 De Mortagne Blvd. Boucherville, Quebec, J4B-6Y4.
PO'l'Em'IAL USES OF ALTERNATE MATERIALS IN LAroE GAS TURBINE OOINES: R. C. Haubert, GE Aircraft Engines, Mail Drop A311, Cincinnati, 00 45215
Several simple and useful unidimensional formulations have been used to mode lise the heat transfer in strip casting processes. However, due to the geometry of the process, predictions diverge from the experimental measurements for angles over 10 to 15 degrees. A simple and efficient variable geometry model is proposed to describe the heat transfer behavior for these larger contact angles. Experimental results obtained from A356 and 5083 aluminum alloys casted in thicknesses of 3 to 5 mm with contact angles from 10 to 20 degrees are presented along with the theoretical predictions. The calculations are in good agreement with the experimental measurements of the strip exit temperature and the temperature distribution in the rolls. The solidification microstructure was
Joint industry and government goals for the IHPTET initiative call for a doubling of aircraft engine propulsion performance by the year 2003. Achieving this goal will require significant advancements in many technology fields: aerodynamics, engine cycle design, mechanical design, analytical methods, materials, and manufacturing processes. Of these technologies, the one probably having the most profound and far reaching effect is the introduction of advanced materials which permit h:\8her cycle temperatures, reduced component weights, higher rotational speeds, and stiffer, more compact structures. H:\8h temperature titanium based alloys and metal matrix canposites will be key materials in these future IHPTET engines. This paper presents a designer's perspective on the use of these advanced materials'.
142-Annual Meeting (2 :30 p.m.) DESIGNING OF SMALL GAS TURBINE ENGINES WITH ALTERNATIVE MATERIALS . D. V. Sundberg, D. H. Comey, 111 South 34th Street, P.O. Box 5217, Phoenix, AZ 85010 . Military 1990' 8
and
commercial
and beyond
require
small
gas
increased
turbine thrust
engines to weight
for
the
ratios
and improved thrust specific fuel consumption. To accomplish these goals, the use of advanced material systems and the design methodology to use the new materials effectively and safely are required. A large increase in performance of small engines dictates that temperatures and tip speeds increase without the normal corresponding increases in size and weight. Conventional materials are no longer adequate for selected components. Advanced material systems must be considered in areas such as high pressure compressors, diffusers, combustors, turbine stators, shafting and engine static structure. Considerable effort on material systems development and design methodology i8 being funded by DoD and subsidized within small engine manufacturing organizations by IR&D.
(3:00 p.m.) FIBER REINFORCED SUPERALLOYS FOR ROCKET ENGINES: Donald W. Petrasek, and Joseph R. Stephens, NASA Lewis Research Center, 21000 Brookpark Road, Cleveland, OH 44135. High-pressure turbopumps for advanced reusable liquid-propellant rocket engines such as that for the Space Shuttle Main Engine (SSME) require turbine blade materials that operate under extreme conditions of temperature, hydrogen environment, high-cycle fatigue loading, thermal fatigue and thermal shock. Based on projections of properties for tungsten fiber reinforced superalloy (FRS) composites, it was concluded that FRS turbine blades offer 0 the potential of a several fold increase in life and over a 200 C increase in temperature capability over the current SSME blade material. Compared to the current blade material, the thermal shock resistance of FRS materials is excellent, two to nine times better, and their thermal fatigue resistance is equal to or higher than the current blade material. FRS materials had excellent low and high-cycle fatigue strengths, and thermal shock-induced surface microcracks had no influence on their fatigue strength. The material also exhibited negligible embrittlement when exposed to a hydrogen environment. (3,30 p.m.)
APPLICATION AND MANUFACTURE OF a Ti A1 COMPONENTS: E. Johansen and D. Wagner, GE Aircr&ft ~ngines, Mail Drop M85, Cincinnati,OH 45215. (Abstract not available) (4 :00 p.m.) PROCESSIIIG, PROPERTIES, AIID APPLICATIONS OF nClEL AIiD !ROIl ALlIIIIIIDF.S
V.t. Sim, Metals and Ceramics Division, Oak Ridge Hational Laboratory, P.O. Box 2008, Oak Ridge, TN 37831
Hickel and iron allDinides are a new class of alloys UDder clevelopEDt at the Oak Ridge National Laboratory. Because of recent progress in ductillzi.og these allDioides, lDaDy applicatiooa have been identified for these materials . 'l'hi.s paper describes the basis for ductilization, processing techniques, EChanical properties, and. applicatioos of nickel and iroo alUllinides. Processing techniques IIill include powder ...tall=, cooventiooal Elting, casting, defomatiOll processing, aod near-net sbaping. Properties w111 include tensile, creep, and. fatigue, which vill be ~ with currently used. materials. *Researcb spoosored by the Office of Energy Utllizatioo Research Energy Conversioo and. Utilization Tecboologies (EClIT) ~, and Office of Fossil Energy, Advanced Research and Technology Devel_t (AR&"lIl) IIaterials Progr.., U.S. Deport.ent of Energy, under contract D£-ACOS-840R21400 with Martin Marietta Energy Syst_, Inc.
LOW TEMPERATURE Si AND Si·BASED EPITAXIAL STRUCTURES: ELECTRICAL PROPERTIES AND DEFECTS V: Other Si Heteroepitaxy Techniques Sponsored by the TMS Electronic and Photonic Device Materials and Electronic Materials Committees Thursday PM February 22,1990
Room: Orangewood In Quality Hotel
Session Chainnan: J.C. Stunn, Princeton University, Princeton, NJ 08544 (2:00 p.m.)
HETEROEPITAXY IN Ag/Si SYSTEMS: Kyung-Ho Park*, Gwo-Ching Wang* and Kri shna Rajan**, *Physi cs Department, **t·1ateria 1s Engineering Department, Rensselaer PolytechniC Institute, Troy, NY 12180-3590. Ag films grown on Si(lll), Si(lOO) and Si(llO) substrates by the Molecular Beam Epitaxy technique have been characterized by X-ray pole-figure analysis, electron microscopy, meV He+ Rutherford Backscattering Spectrometry (RBS)/channeling and Secondary Ion Mass Spectrometry . Despite the large lattice mismatch (~25%) between Ag and Si, high quality epitaxial Ag single crystal films (6-10% surface minimum field) were . grown on clean Si substrates. The quality of the Ag film depends on the film thickness (with a decreasing yield as the thickness of the film increases) and the substrate orientation (with the best field in Ag(lOO) grown on Si(lOO»). Defects in the film are mainly confined within an ~1000 ~ region of the Ag/Si interface. The epitaxial Ag/Si interface was thermally stable up to at least SOO·C. The mechanisms of epitaxial growth and defect evolution in this large lattice mismatch system will be discussed. (2:20 p.m.)
PULSED UY·LASER INDUCED EPITAXY OF 51 AND 511.XGEX ALLOYS
Thomas W Sigmon and Y. Chang. Depanment of Electrical Engineering Stanford University. Stanford. CA 94305 and K.H. Weiner. Lawerence Livennore National Laboratory, Livermore, CA 94550
The selective epitaxy of Si and Si1.xGex strained layers on (100) silicon substrates is described. The process uses a pulsed XeCI excimer laser operating at "- =308 nm to melt and intennix either ex or polycrystalline Si or Ge selectively deposited on the Si substrate. Layer formation occurs by mixing in the molten phase with subsequent rapid recystallization of the film. Incorporation of conventional dopants can be performed in situ during the process for device fabrication. Selectivity is accomplished by conventional IC lithographic techniques, not mesa fabrication. Results have been obtained for poly-Si/ and aGe/(I00) Si structures. For the Si1.xCJex layers x values form 0.12 to 0.33 have been obtained so far. Structural analysis. utilizing RBS and TEM indicate excellent crystallinity for most of the layers. Electrical results have been obtained for p+·n Si1.xGex diodes which indicate near ideal characteristics. Detailed description of the process and results will be presented. (2:40 p.m.) DEFECTS IN DEPOSITED AND RECRYSTALLISED Si LAYERS; SO. Brother ton , l.R. Ayres and M .l. K ing, Philip~ Research L.aboratories, Redhill . Surrey . RHI 5HA. U.K. Electrically active defects have heen measured in MBE deposited Si, amorphous deposited and epitaxially regrown UHV evaporated Si and regrown amorphiscd silicon. (n all layers a high concentration of donor-like defects has been found at the layer/substrate interface. In the case of the amorphised films these are atrribUled to interstitial dislocation loop' . but for the other two types of film thc structu ral/c he mical nature of the defects is less clear. Experiments with MRE silicon have enabled us to rule out certain possihle impurities as responsihle fur the donors such as interstitially honded horon, oxygen thermal donors and copper. The po:-.s ible role of carbon in Ihese structures will he discussed. (3:00 p . m. ) BREAK (3:15 p.m.)
THE INITIAL STAGES OF GROWTH OF SI ON SI(111) BY SPA-LEED.
M. Horn von Hoegen l T. J. Watson Research Center, Yorktown Hgts, NY;
M. Henzler, Instltut fuer Festkoerperphyslk. Universitaet Hannover.
Germany. LEED Is used to Invesllgate the Inilial stages,.of the growth of Silicon on silicon (111) between 556K and 900K. The partial coverages e of the grOWing film (or roughness) are derIved out of the Intensity osclllalion of the 00· and 7x7-spot during evaporation. We observe a preferred growth
Annual Meeting-143 In the second layer long before the first layer Is completed. Surface defects caused by superstructure disorder on the grown layer act as nucleation center for the diffuSing adatoms, which are trapped on top of the first layer. Nearly perfect layer by layer growth takes place after completion of the first layer. The grown films show below 650K an unordered superstructure, In an Intermediate temperture range a mixture of 5x5- and 7x7-domalns and finally above 870K the perfect 7x7 -superstructure. The 5x5-superstructure Is a metastable variation of the 7x7. The average size of Islands Increases from 30 up to 40000 atoms, which reveals the perfection of the layer with Increasing temperatures. (3:35 p.m.)
ETCHING OF c-Si, a-Si:H,F AND SiO z IN SiF4 -H z PLASMAS: Y. Okada and S. Wagner, Department of Electrical Engineering, Princeton University, Princeton, NJ 08544 Crystalline silicon (c-Si) can be grown epitaxially at temperatures as low as 250 °C by plasma assisted chemical deposition from SiF4 and H 2 . At temperatures around 250°C and process pressures around I Torr, Si0 2 , produced by impurities such as O 2 and H 20, is thermodynamically stable on the growing surface; probably, amorphous silicon (a-Si:H,F) growth also occurs. Therefore, the etching of Si02 and a-Si:H,F by F· and H· radicals may play an important role in promoting epitaxial growth in this system. We report the etching of c-Si , a-Si:H,F and Si02 in glow discharges of SiF4 and H 2, the selectivities for these materials, and discuss the growth mechanism of c-Si from SiF4 and H 2. This work is supported by the Electric Power Research Institute and the Office of Naval Research. (3: 55 p.m.) EMBEDDED GROWTH AND CHARACTERIZATION OF GAA5 ON 51 BY MBE: J.B. Liang, M.Y. Kong, Institute of Semiconductors, Academia Sinica, P.O. Box 912, Beijing 100083, China. A GaAs on Si coplanar technology by MBE which is suitable on combine GaAs and Si circuits on chip is reported. The wells for embedded growth are accomplished by wet chemical etching with a controllable solution of HF HN0 3 " The quality of the embedded MBE GaAs layers are found to be comparable to that of unetched planar GaAs on Si. Low temperature Photoluminescence, Raman scattering and Double crystal X-ray diffraction are used to examine the quality of epitaxial layers, the narrow linewith of 8meV, 3 cm- l and 270 arcsecond are achieved respectively for a epilayer with 2 urn thickness. High resolution electron microscope image reveal the epitaxial relation of GaAs/Si interface on the bottom and the sidewall of the wells; the dislocation are restricted to a narrow region near the interface.
MAGNETISM SUPERCONDUCTIVITY AND
THEm INTERPLAY IT
Sponsored by the ASM-MSD Electrical, Magnetic, and Optical Properties Committee Thursday PM February 22, 1990
Room: Orangewood I Quality Inn
Session Chainnan: G.A. Prinz, Code 6345, Naval Research Laboratory, Washington, DC 20375-5000
(2:00 p.m.) REVIEW TALK ON SUPERCONDUCTIVITY AND MAGNETISM IN ORGANIC CONDUCTORS. PM Chaikin, Department of Physics, Princeton University, Princton, NJ 08544.
Before the 1980's the quasi-one-dimensional organic conductors were best known for their instabilities toward low temperature semiconducting states. Since that time there have been new calsses of organic charge transfer saIts which exhibit a wide variety of interesting ground states. Of particular interest for this symposium is the family of "Bechgaard Salts", (TMTSF12X (where X is CI04, Pf6, Re04 or a number of other anions) where the fIrst superconducting organic was found and where the fIrst organic antiferromagnet was found. In fact several of thie materials in this family are antiferromagnetic insulators at low temperature and ambient pressure and become superconducting with the application of modest pressure. The competition between these states will be discussed theoretically and experimentally. Even more interesting is the existence of a unique Field Induced Spin Density Wave Phase at high magnetic fIeld in the superconducting compounds. Recent theoretical work indicates that the high field antiferromagnetism may be the result of the superconducting pairing interactions.
(2:45 p.m.)
INTERACTION BETWEEN SUPERCONDUCTIVITY AND MAGNETISM IN SUPERLA TTICES: Ivan K Schuller, University of California-San Diego, La Jolla, CA 92093 and Y. Bruynseraede, Katholieke Universiteit-Leuven, Leuven, B3030 Belgium. Superlattices have been used as model systems to study the interaction between superconductivity and magnetism. In superconducting/normal metal superlattices, the upper critical fIeld exhibits transitions from two to three dimensions and the flux lattice changes from one to two dimensional behavior. In superconducting/magnetic superlattices the superconducting dimensional behavior is modifIed by the magnetic ordering with a region of coexistence present. Work supported by the US Department of Energy under grant DE-FG0387ER45332 (at UCSD), the Belgian Concerted Action (GOA) and the Interuniversity Attraction Poles (IUAP) (at KUL). (3:30 p.m.) SUPERCONDUCTING AND MAGNETIC INTERCALATION COMPOUNDS. A Chaiken, Naval Research Laboratory, Washington, D.C. 20375-5000.
Intercalation compounds are chemically synthesized superlattices in which the properties of a host material can be modifIed by the insertion of a second intercalant species. Both the host and intercalant may be chosen to be superconducting or magnetic, so that intercalation gives some insight into the competition between these two phenomena. One unusual type of intercalation compound is the class of superconducting graphite-based compounds formed from non-superconducting precursors. A wide variety of ordered magnetic phases is found in intercalation compounds, where the host-intercalant interaction may playa crucial role.
(4: 15 p.m.) "MAGNETIC" IONS IN REA VY FERMION SUPERCONDUCTORS. Q,R. ~, University of Florida, Gainesville, FL 32611. Heavy fermion systems have both single-ion and collective effects which compensate the local4f- or 5f- electron moment deduced from the Curie-Weiss behavior of the magnetic susceptibility in the 100-300 K regime. These compensation effects complicate the normally simple process of doping with either magnetic or non-magnetic ions. We present a study of doping in heavy fermion superconductors which begins to unravel the complexities of the effect of doping on superconducting T c and specific heat discontinuity. The question of what kind of pairing is present in heavy fermion superconductors will be reexamined in the light of our new data. Work at University of Florida supported by U. S. Department of Energy, grant DE-FG05-86ER45268.
METALLIZATION FOR ELECTRONICS APPLICATIONS VI: Advanced Technology for Contacts to GaAs Sponsored by TMS Electronic and Photonic Device Materials, Electronic Materials, and Alloy Phases Committees Thursday PM February 22, 1990
Room: California I-II Quality Hotel
Session Chainnan: K.A. Jones, Electronics Technology and Devices Laboratory, Fort Monmouth, NJ 07703; L.C. Kimerling, AT&T Bell Laboratories, Murray Hill, NJ 07974 (1 :30 p.m.)
PHASB STABILITY AT KETAL/SBMICONDUCTOR INTERFACES: ~ ~ D. swenson, C-H. Jan and F-Y. Shiau, Department of Materials Science and Bngineering, University of Wisconsin -Madison, Madison, WI 53706. The chemical stability of metal/semiconductor contacts as vell as their electrical performance are governed by the thermodynamics, kinetics and interface morphology of phase formation between the metallizing elements and the semiconductor. When the semiconductor is a compound such as GaAs, InP, InAs, etc" we are dealing with a three-component system, Since we are often concerned with having contacts of a metallic compound on a compound semiconductor, we will be dealing with a quaternary or even higher order system. The objective of this presentation is to discuss the general prinCiples governing the stability of interfaces for ternary and higher order systems. Bxamples will be given to illustrate these principles.
144-Annual Meeting (2 :00 p.m.) STABLE AND SINGLE-PHASE OHMIC AND SCHOTTKY CONTACfS TO III-V SEMICONDUCTORS T.....Sands. , J.P. Harbison, T.L Cheeks, R.E. Nahory, H.L Gilchrist, LT. Florez, R. Ramesh, V.G. Keramidas and LC. Wang', Bellcore, 331 Newman Springs Rd., Red Bank, NJ 07701, 'Department of Electrical Engineering and Computer Sciences, University of California, San Diego, La Jolla, CA 92093. The development of new compound semiconductor metaJlization systems that meet the performance, uniformity and stability criteria imposed by the increasing speed and decreasing dimensions of compound semiconductor devices is essential to continued progress in high-speed electronic, photonic and quantum-effect devices. We have applied thermodynamics combined with isoelectronic substitution of group III elements to design and fabricate thermally stable single-phase Ohmic and Schottky contacts to III- V semiconductors. In this talk we describe the fabrication, electrical characteristics, interface structure and thermal stability of epitaxial CoAl and NiAl Schottky contacts to (Al,Ga)As and Pdln Ohmic contacts to GaAs. (2:30 p.m.) ABRUPT INTERFACES WIlli NOVEL STRUCTURAL AND ELECTRONIC PROPERTIES: METAL CLUSTER DEPOSITION AND METALSEMICONDUCTOR JUNCTIONS. J H Weaver, G.D. Waddill, I.M. Vitomirov, C.M. Aldao, S.G. Anderson, and C. Capasso, Department of Chemical Engineering and Materials Science., University of Minnesota, NlinneapoJis,MN 55455 Abrupt interfaces with no substrate disruption are produced by a novel method of metal- semiconductor junction formation. This method involves the condensation of a thin Xe buffer layer on cleaved surfaces to isolate the semiconductor from impinging metal atoms. This Xe buffer layer provides a surface on which the metal atoms diffuse, nucleate, and grow into metallic clusters. These clusters are then brought into contact with the substrate when the Xe is thermally desorbed. The result is an abrupt, non-disrupted, nearly-ideal interface. Photoemission studies of AI, Ag, Au, Ga, Ti, and Co clusters grown on GaAs(110) show unique Fermi level positions -0.3 and 1.0 eV below the conduction band minimum that are nearly metal- and coverage-independent. We find no evidence for metal-induced gap states or conventional defect levels, but photoemission results show surface unrelaxation around the clusters. This unrelaxation induces the reappearance of states in the gap. High-resolution electron microscopy results for Au(clusters)/GaAs(110) show intimate contact with no intermixing. (3:00 p.m.)
NON-SPIKING OHMIC CONTACT TO GaAs BY LIMITED SOLID PHASE REACTIONS, L. C. Wang, C. C. Han, E. D. Marshall, and S. S. Lau, Department of Electrical and Computer Engineering, R-007, University of California, San Diego, California 92093. We present several metallization schemes to form non-spiking ohmic contact to GaAs. The transport mechanisms are interpreted in terms of the temperature dependence of the contact behavior (from room temperature to - 10 K). Comparisons are made with conventional alloy contacts. (3:30 p.m.)
BREAK
(3:40 p.m.) CORRELATION BETWEEN ELECTRICAL BEHAVIOR AND INTERFACIAL PHASES Au/Ge/Ni CONTACTS TO n-GaAs: T. Kim, D.D.L Chung, S. Mahajan, Department of Metallurgical Engineering and Materials Science, Carnegie Mellon University, Pittsburgh, PA 15213, J.P. Harbison, T. Sands, Bellcore, Red Bank, NJ 07701 The thermal stability of Au/Ni/Ge/Au contacts to niiGaAs has been investigated in the temperature range 250-500 C. Concomitantly, I-V characteristics of the contacts have been ascertained. The changes in electrical properties of the contacts correlate well with the formation of different phases at the contact-semiconductor interface. The lowest contact resistance is observed when NiAs grains are contiguous to GaAs. Ancillary experiments on NiAs/n-GaAs structures grown by MBE indicate that NiAs froms a Schottky barrier with GaAs and the barrier height is 0.75 eV. These results suggest that Ge has a very important role as a dopant either in GaAs or in NiAs in producing low resistance Au/Ni/Ge/Au contacts to n-GaAs. (4: 10 p.m.) MECHANISMS FOR THE FORMATION OF LOW TEMPERATURE, NON-ALLOYED AuGe OHMIC CONTACTS TO N-GaAs. M. Dornath-Mohr, M.W. Cole, R.T. Lareau, W.H. Chang, D.W. Eckart K.A. Jones, Electronics Technology & Devices Laboratory, Fort Monmouth, NJ 07703 Mechanisms for the formation of Au-Ge ohmic contacts to n-GaAs are difficult to understand, partly because a liquid is formed during a typical annealing process. This liquid destroys the existing microstructure and allows such rapid mixing that the individual steps cannot be identified. For varied compositions and annealing times, ohmic contacts were formed at 320 C, which is well below the Au-Ge (356 C) and beta-AuGa (341 C) eutectics. Optical microscopy, SEM, TEM and SIMS suggest that Au and Ge behave synergistically by acting as sinks for Ga and As respectively. A nitride cap slows the reactiQft fer the eutectie
mix. Without a cap the atmosphere acts as a secondary As sink for the Ge-poor mix. Theoretical analysis indicates that ohmic contacts are formed by a GexAs or AuxGay-GaAs junction, rather than a heavily doped GaAs-metaI junction. Less spiking occurs when metal is deposited on an oxide free GaAs surface.
(4:30 p.m.) INVESTIGATION OF THE ELECTRICAL PROPERTIES AND STRUCTURE OF AI-Ge-Ni CONTACTS TO GaAs AND Al Gal_xAs: R.J. Graham. Center for Solid State Science, Arizona State Oniversity. lempe. Anzona 85287-1704. T.B. Haddock and R.J. Roedel. Department of Electrical and Computer Engineering. Arizona State University, Tempe. Arizona 85287-5706. R.W. Nelson and P. Williams, Department of Chemistry, Arizona State University. Tempe, Arizona 85287-1604. The microstructure of AI-Ge-Ni Ohmic contacts to both n- and p-type GaAs and AlxGa J _xAs has been investigated by high resolution transmission electron mIcroscopy (TEM) and secondary ion mass spectroscopy (SIMS). Electrical assessment of these contacts shows that Ohmic contacts are formed on both n- and p-type material and that the thickness of Ge deposited and the alloying time have a large influence over the degree of Ohmicity observed, particularly in the case of n-type material. TEM shows the contact interface to be extremely flat and uniform in all cases with a continuous polycrystalline layer of AI3Ni adjacent to the semiconductor. SIMS shows that the contact components. especially the AI. diffuse into the underlying semiconductor during alloying. The possible reasons for the observed variations in Ohmic behaviour as a result of processing are discussed in terms of the microstructure. (4:50 p.m.) OPTIMIZATION
OF OHMIC CONTACTS FOR RELIABLE HETEROSTRUCTURE GaAs HATERIALS
C.S. \lUI K.K. YU, H. HU, & H. KANDER --HUGHES AIRCRAFT COMPANY MICROVAYB PRODUCTS DIVISION, TORRANCE, CA 90509 Microwave and aUliaeter wave systems requirements for GaAs based high frequency devices and circuits have led to development of HEHT structures for both low noise and power applications. The new emphasis on performance and reliability necessitates the following features for ohmic contacts on BEMT structures: good morphology of the metallization, low contact resistance, compatabili ty wi th fine line l i thography, high yield and prolonged high performance. The Ni/AuGe/Ag/Au metal system has recently been applied to GaAs heterostructures. Optimization of this metal system requires the proper choice of alloy condi tions and substrate cleaning procedures. In this study, we used different cleaning procedures and .easured the contact resistance as a function of alloy te.perature. We obtained very loy contact resistances for both low noise and power BEMTs with Rc .. 0.13+0.03 ohm-mm. The alloyed metal depths were also characterized by-SIMS analysis. Ohmic contacts with the opti.ized cleaning procedure were lDeasured for reliability performance. Ve obtained stable contacts up to 1000 hours at 200°C under full current bias operation.
PROCESS MINERALOGY IT: Applications to Pyrometallurgy Sponsored by TMS Process Mineralogy Committee Thursday PM February 22, 1990
Room: Salon H Marriott
Session Chairmen: Richard D. Hagni, Dept. of Geology and Geophysics, University of Missouri-Rolla, Rolla, MO 65401; David G.C. Robertson, Dept. of Met. Engr., University of Missouri-Rolla, Rolla, MO 65401
(2:00 p.m.) KEYNOTE GEOLOGY, MINING, AND METALLURGY OF THE DOMINICAN REPUBLIC NICKELIFEROUS LATERITE ORES: E. W. Lithgow, Falconbridge Dominicana C. por A., Santo Domingo, Dominican Republic The central Dominican Republic Ni laterite ores occur in a belt (6 by 90 km) of alpine-type serpentinized peridotites. Protores are harzburgite, serpentinite, and dunite. Laterization between Miocene and Pleistocene formed supergene oxide ore (10 m thick) types: limonitic, soft serpentine, and rocky. Ni is in serpentine group, goethite, quartz, and Fe-Mg minerals. Open pit mining is along 7 meter benches with direct excavation by hydrauliC shovel unaided by blasting, and with a stripping:ore = 1.50:1. Grade control is by roto-percussion drilling (12.5 m grid), sampling, and visual means. Ore haulage is by 40-70 ST off-highway trucks. Backfilling, recontouring, and reforestation complete mlnlng. The ore is briquetted, fed through 12 shaft furnaces, and reduced by burning naphtha. The 850°C calcine goes to the melting plant, two 60-80 megawatt rectangular electric furnaces, with ferronickel and slag tapped from opposite ends. Dephosphorizing and desulphurizing of the ferronickel occurs in the refining area prior to casting into ingots and cones.
Annual Meeting-145 (2:50 p.m.) CARBOTHERMIC REDUCTION OF DOMESTIC FERRUGINOUS MAGANESE RESOURCES: P.C. Turner, J.E. Tress, and R.H. Nafziger, Albany Research Center, Bureau of Mines, Albany, OR 97321 The Bureau of Mines, U. S. Department of the Interior, completed laboratory scale studies involving the prereduction of domestic ferruginous resources to produce an acceptable ferromanganese for alloying iron and steel. The objective was to evaluate the prereduction of such resources prior to melting as an alternative to direct smelting. As part of the Bureau's goal to recover metal values economically and efficiently from domestic resources, ferruginous manganese oxide ores were prereduced with low cost, carbonaceous reductants. The products will provide suitable charge material for the production of ferromanganese in the electric arc furnace. Greater than 95 percent reduction of the tetravalent and trivalent states of manganese to the bivalent state, which represents typical feed for the ferromanganese are furnace, were achieved using inexpensive blacksmith coal at 750°C. Previously, it was expected that tsese domestic, lowgrade ores would not prereduce below 1,050 C.
(3: 15 p.m.) ORE MICROSCOPIC STUDY OF AGGLOMERATION AND FRAGMENTATION IN COPPER FLASH FURNACE PRODUCTS: Richard D. Hagni, Dept. Geology and Geophysics, University of Missouri-Rolla, Rolla, MO 65401 Several suites of product samples from two laboratory and two commercial flash furnaces have been studied by process mineralogy techniques to document the mineral reactions that occur in falling copper sulfide particles in flash furnaces. Ore microscopy provides an excellent technique to directly observe the reaction of pyrite ___ pyrrhotite ~ magnetite ~ hematite, and chalcopyrite --. Cu-Fe-S intermediate solid solution --. bornite --.digenite --. chalcocite ___ copper + magnetite and hematite. The microtextura1 characteristics of the particles are angular, spherical, and cenospherica1, and are a result of the solid-gas and liquid-gas reactions and the evolution of SO?. Both agglomeration and fragmentation of particles has occurrea during their descent in the flash furnaces. The extent of agglomeration increases with an increase in the size of the flash furnace. This research has been supported by the U.S. Bureau of Mines Generic Mineral Technology Center for Pyrometa 11 urgy. (3:40 p.m.)
PHASE IDENTIFICATION, PHASE ABUNDANCE, AND TEXTURE ANALYSIS IN CERAMIC BUILD-UPS FROM ELECTRIC INDUCTION FURNACES WITH THE ELECTRON MICROPROBE AND CATHODOLUMINESCENCE MICROSCOPE: Musa Karakus, Richard D. Hagni, Department of Geology and Geophysics, and Robert C. Moore, Department of Ceramic Engineering, University of Missouri-Rolla, Rolla, MO 65401
Ceramic build-ups are undesirable deposits formed in or near the throats of the induction units in electric channel induction furnaces that slow down melt circulation and decrease the quality of the metal. Phases identified by cathodoluminescence microscopy (CLM) and EMPA were: spinel, forsterite, peric1ase, corundum, anorthite, hibonite, mu11ite, geh1enite, silica, montice11ite, silicon carbide, oldhamite, Ce-La phases, and glass. Each phase is readily recognized by its characteristic cathodoluminescence. Non-luminescent phases were hercynite, Fe-oxides, pyrrhotite, and metallic iron. Most build-ups are composed mainly of spinel, hibonite, corundum and peric1ase. Textures show crystals that were crystallized from the melt and fragments derived from refractories and feed materials. This research was supported by the American Foundryman's Society. (4:05 p.m.) MINERALOGICAL STUDY OF ILMENITE CONCENTRATE FEEDS AND PLASMA SMELTING TITANIUM SLAGS: S. Pignolet·Brandom, R. T. Brandom, A. D. Brent and K. J. Reid, Minerals Resource Research Center, University of Minnesota, 56 East River Road, Minneapolis, MN 55455 As part of a study to examine the potential for economic development of Minnesota ilmenite resources, ilmenite feeds and the high titanium slags generated by plasma arc smelting were studied by reflected and transmitted light microscopy, x·ray diffraction, and automated scanning electron microscopy image analysis. These techniques were used to identify the phases and to characterize the occurrences of phases containing calcium and magnesium, elements that are deleterious in the refining of the titanium slag. A discussion of this data will be presented and will include recommendations for the removal of these elements.
(4:30 p.m.)
COMPUTER PROGRAM TO IDENTIFY ORE MINERALS BASED ON ELEMENTAL CHEMISTRY AND/OR PHYSICAL PROPERTIES: ~. C. Laudon, and R. D. Hagni, Department of Geology & Geophysics, University of Missouri-Rolla, Rolla, Mo. 65401. ~MID, standing for Ore Mineral Identification, is a very fast computer program written in compiled BASICA for the IBM-PC which identifies 570 different ore minerals based on physical properties and/or elemental chemistry. The program is equally
applicable to ore microscopists or microbeam analysts. A copy of the program and datafile may be obtained by sending a self addressed, stamped envelope and a double sided, double density 5-1/4 inch or 3-1/2 inch diskette to the authors. The physical properties search is a "closeness of fit" search and is based on estimates of air reflectivity, Vickers hardness, Talmage hardness, anisotropism, mineral color, mineral tint, color of internal reflections, and tint of internal reflections. Uncertain data may be weighted low or eliminated. The chemical element search is almost instantaneous and is based on the presence of particular elements, but not on their relative abundance.
REFRACTORY GOLD, SnNER & PRECIOUS METALS PROCESSING IV: Alternate Lixiviants for Precious Metals Sponsred by the TMS Precious Metal Committee Thursday PM February 22, 1990
Room: Salon 1 Marriott
Session Chainnen: Rong-Yu-Wan, Newmont Metallurgical Services, 417 Wakara Way, Suite 210, Salt Lake City, UT 84108; M. Drozd, The Gwalia Group, World Trade Center, 1675 Broadway, Suite 2350, Denver, CO 80202 (2:00 p.m.) THE COST OF OXYGEN IN IIYDR0I1ETIILLURGICIIL PROCESSING OF REFRACTORY
GOLD SULFIDES: Gus VanWeert, RRft3, Caledon East, Ontario, Canada LONlEO.
The cost of oxygen is a major and common factor in the hydrometallurgical processes developed or proposed for the oxidation of refractory pyrites. This cost involves charges for the generation and the transfer of the oxygen to the mineral flurry.
(2:30 p.m.) mE RCITI\TING DISC S'IUDY' OF (DID DISSOIlJI'IOO WI'IH GEDBIU!: 3400: B. Pesic, Depart:ment of Metallurgy, University of Idaho, Moscow', Idaho 83843 R. H. Sezgent A. ~, Great Lakes Chanical co:rporation, West lafayette, IN 47906
am
am
'!he reactions of branine with gold were studiErl by us:in;J pure rotatin:J disc of gold. Branine was provided by usin:J GeOOran 3400, a trade name (Great Lakes) for a branine oontainin:J lixiviant. '!he fun:lamental parameters studiErl were: rotat:in;J speed, concentratioo of GeOOran, txnJentratioo of acid, am tenperature. In addition, particular attentioo was given to
the effects of various additives, cations am anions. It was found that gold dissolution rate with GeOOran 3400, for the sane concentrations of lixiviants, 2 gpl, was aba.it 5 tine; faster than with cyanide am aba.it 2 tine; faster than with thiourea.
0:15 p.m.) ADSORPTION OF GOLD FROM HYPOCHLORITE/CHLORIDE SOLUTIONS USING NEUTRAL POLYMERIC ADSORBENTS, R. Y. Wan and B. Ball, Newmont Metallurgical Services, Salt Lake City, Utah Testing of commercial ion exchange resins was carried out using synthetic gold-hypochlorite solutions. Several anion exchange resins gave good loading capacity, but most of the resins were degraded due to the hypochlorite oxidant. Polyacrylic ester neutral adsorbents appeared to be much less active with hypochlorite, and gave adsorption isotherms similar to those using activated carbon. Elution of the resin would appear to be much simpler than stripping of carbon and no regeneration would be required. The preliminary findings were confirmed using hypochlorite solutions produced from column leach tests on low grade refractory ore.
0:45 p.m.) GOLD RECOVERY FROM REFRACTORY ORES BY CARBON-IN-CHLORINE LEACH, John N. Greaves, U.S. Dept. of the Interior, Bureau of Mines, 729 Arapeen Drive, Salt Lake City, Utah 84108 An investigation undertaken by the U.S. Bureau of Mines examined the recovery of gold by chlorination of refractory carbonaceous and sulfidic ores. Comparisons were made between several treatment methods, two of which are: (1) the ore pulp is chlorinated and subsequently contacted with activated carbon (L-CIP), and (2) a carbon-in-chlorine leach (CICL) technique in which a simultaneous refractory treatment, gold extraction, and recovery with activated carbon occures. Depending on the
146-Annual Meeting ore type, gold recovery was enhanced by use of either L-CIP or CICL. Chlorination without carbon of a carbonaceous and a sulfidic ore extracted 1-. and 20-pct of the gold. respectively. By use of either CICL or L-CIP treatments, the carbonaceous and sulfidic ores achieved 90-, and 93-pet recoveries on the carbon, respectively.
(4:30 p.m.) PLATINUM GROUP METALS Eh-pH DIAGRAMS. LEACHING AND RECOVERY WITH GEOBROM™ 3400. A. Dadgar and R. H. Sergent. Great Lakes Chemical Corp .• P.O. Box 2200. Highway 52 N.W .• West Lafayette. IN 47906 Thermodynamic data are used to construct Eh-pH diagrams for the platinum group metals in the CI-H 20 and Br-H 20 systems at 25'C. In the presence of chloride ions, platinum group metals can form a series of soluble and/or solid chloride complexes, depending on the nature of each element, concentrations of chloride and metal ions, applied potential and pH values. In general, the trend for the platinum elements to form the bromide complex species seems to be stronger than that for the chloride complexes. This can be easily seen by comparing the related Eh-pH diagrams for Pt. Pd and Rh in the chloride and bromide systems. 'This report also covers the leaching of platinum from a sample of automobile catalytic converter with Geobram™ 3400 and its subsequent recovery by a combination of standard methods and a new ion exchange resin recently developed at Great .Lake~ Chemical.
THERMAL ANALYSIS OF METALLURGICAL SYSTEMS IV: Surface and Interface Phenomena Sponsored by the TMS Chemistry and Physics of Materials Committee Thursday PM February 22, 1990
Room: California ill-IV Quality Hotel
Session Chainnen: A. Joshi, Lockheed Palo Alto Research Laboratory, 3251 Hanover Street, Palo Alto, CA 94304; R.D. Shull, Metallurgy Division, IMSE, National Institute of Standards and Technology, Gaithersburg, MD 20899 (2:00 p.m.)
(Invited)
DIFFERENTIAL SCANNING CALORIMETRY STUDIES OF THIN FILMS AND MULTILAYERED COMPOSITES: E. J. Cotts, Department of Physics, Applied Physics and Astronomy, State University of New York at Binghamton, Binghamton, New York 13901. Phase and structural transformations in thin films and multilayered composites are monitored by means of differential scanning calorimetry. Solid state reactions at the interfaces between two different constituents are examined. These reactions include the formation of metastable phases such as amorphous alloys in multilayered thin-film diffusion couples of elemental metals. Phase changes are correlated with structural studies by means of x-ray diffraction and/or transmission electron microscopy. Analysis of calorimetric data results in quantitative information on both the kinetics and energetics of solid state reactions, including averaged interdiffusion constants, activation energies for given processes and the heat of formation of certain phases. Sample preparation procedure, calorimetry technique and data analysis will be described. (2:30 p.m.) Invited THERMAL ANALYSIS OF THIN FILMS. P. K. Gallagher, AT&T Bell Laboratories, Murray Hill, NJ 07974. Characterization of thin films is demanding because of the large amounts of "inert or uninteresting" substrate which generally accompany such samples. Unless one is interested in the film-substrate interaction, this extraneous material only serves to dilute the effect being measured. Consequently it is necessary to use the more sensitive techniques. Evolved gas analysis (EGA) using mass spectrometric analysis is particularly useful. If the film or product is magnetic, then an external magnetic field may be used to enhance the weight changes associated with reaction. Examples using these and other techniques will be described.
(3:00 p.m.)
(Invited)
OXIDATION BEHAVIOR OF CARBON AND DIAMOND FILMS. A. Joshi, Lockheed Palo Alto Research Laboratory. 3251 Hanover Street, Palo Alto, CA 94304. Oxidation behavior of natural diamond, graphite and plasma assisted CVD diamond and diamond· like films were evaluated in molecular oxygen in the temperature range 500 to SOO'C. Weight loss due to oxidation as a function of temperature and time were measured with high sensitivity by thermal gravimetric analysis (TGA). The oxidation rate of diamond films was found to be lower than natural diamond, which exhibited lower rates than grafoil and pyrolytic forms of graphite. SEM examination of oxidized diamond film and natural diamond surfaces provided further insights into the oxidation behavior of diamond films. In diamond films. oxidation occurred preferentially at grain boundaries. local defects and diamond-like carbon containing regions during the early stages of oxidation process. The results suggest that preferred orientation of diamond crystallites in the CVD film plays a major role on its oxidation behavior.
(3:30 p.m.)
(Invited)
OXIDATION BEHAVIOR OF BORIDES*. G.M. Mehrotra. Mechanical and Materials Engineering Department. Wright State University, Dayton. OH 45435. The oxidation behavior of hot pressed specimens of ZrB • 2 HfB2 and SiB will be discussed. The effect of incorporating 6 MoSI 2 and AIN in the matrices of these borides will also be discussed. The results obtained from characterization of the specimens using thermogravimetry. X-ray diffraction and scanning electron microscopy will be presented. *This work was sponsored by Materials Laboratory. Air Force Wright Aeronautical Laboratories. Aeronautiaal Systems Division (AFSC), United States Air Force. Wright-Patterson AFB, Ohio 45433-6533.
(4:00 p.m.)
(Invited)
APPLICATION OF TGA AND DSC TO THE STUDY OF OXIDATION BEHAVIOR AND PHASE EQUILffiRIA IN ALLOYS BASED ON TIlE MgLi SYSTEM: .s......!la.s.. Department of Materials Science and Engineering. University of Illinois at Urbana-Champaign. 1304 West Green Street. Urbana. Illinois 61801 and H. L. Fraser. Department of Materials Science and Engineering. The Ohio State University. Columbus. OH 43210-1179. There is considerable interest in the development of elevated temperature magnesium (Mg) alloys for aerospace applications. For example, Mg-Li-Si alloys are good candidates for elevated temperature applications. At these temperatures, oxidation plays a major role in detennining the usefulness of these alloys. A thennogravimetric analyzer (TGA) has been used to study the oxidation behavior of these alloys. Optical microscopy. x-ray diffraction and various surface analysis techniques have also been used to complement the TGA study. In a separate study, the potential of differential scanning calorimetry (DSC) technique has been utilized to understand the effect of Li on the metastable Mg-Si phase diagram. (4:30 p.m.)
(Invited)
KINETICS OF OXIDATION OF LIQUID METALS BY Cl. O. F: Ceramics Division, NIST, Gaithersburg, MD 20899
L. P. Cook,
A method has been developed for thermogravimetric (TGA) study of the oxidation of liquid metals by C1 2 , 02 and F 2 • Briefly, this involves creation of a metal droplet of known surface area supported by a non-wetted surface, which also serves to direct the flow of oxidant around the sample. Chlorine, oxygen and fluorine each produce reaction products of different character, which influence the TGA results. The chlorides tend to be volatile and leave the reaction site, while the oxides have low vapor pressures and form a solid product layer; the fluorides are intermediate in their behavior. Quantitative TGA data on the oxidation kinetics of liquid Al by gaseous F (as present in dissociated ClO,F) have been collected at BOO'C. Results show that oxidation is a multistage process and can be treated in terms of the classic steps of nucleation, island growth, island coalescence and film thickening. Structure of the film suggests that growth occurs at least in part by outward diffusion of aluminum. Work is under way to apply the method quantitatively to other metal/oxidant systems as well.
(5 :00 p.m.)
Invited
ANALYTICAL STUDIES OF THE PYROLYSIS OF PHENOL-FORMALDEHYDE RESINS AND C/C-COMPOSITES: C.J. Chu. M. Sampson. R.H. Hauge, and J.L. Margrave, Houston Area Research Center, Materials Science Eesearch Center, 4802 Research Forest Drive, The Woodlands, TX 773BI. Pyrolyses of phenol-formaldehyde resins and C/C-composites have been studied from room temperature up to·....... ISOO K. By using line-of-sight mass-spectrometric and FTIR-spectroscopic monitors one can get real-time data for the various molecular species evolved (~zO, CO, COZ' CH , etc.) and correlate these with the kinetics/mechanism of th~ pyrolysis processes.
Annual Meeting-147
TITANIUM Ill: Computer Modeling of Fabrication of Titanium Alloys Sponsored by the TMS Titanium Committee and TMS Synthesis and Analysis of Materials Processing Committee Thursday PM February 22,1990
Room: ParkA Inn at the Park
Session Chainnen: G.W. Kuhlman, Alcoa Forging Division, Cleveland, OR 44105; S.C. Jain, G.E. Aircraft Engines, Cincinnati, OR 45215 (2:00 p.m.) STATUS OF PROCESS MODELING APPLICATION IN FORGING INDUSTRY: S.C. Jain, G.E. Aircraft Engines, Cincinnati, OH 45215 Domestic Aerospace Vendor Supplier Industry is moving from "Trail and Error" to IIPlan, Predict and Produce". Such discipline offers near net shapes, improved part quality and reduced development, inspection and costs by making it right the first time. Smaller, faster, economical and easier to use computers have led to significant capture of CAD/CAM and process modeling tools for unit manufacturing process simulation. These technologies are used to hasten part and tooling design and to eliminate costly shop trials. Computer simulation of forging, die fill, forging defect prediction, die stress analysis and workpiece heat treating, predicting transient, nonlinear temperature and stress in the workpiece, during furnace heatup and quenching, is becoming a way of life. Engine manufacturers including GE Aircraft Engines have started requiring process modeling for selected critical forged parts from the forge companies bidding on these parts. This presentation cover the current status and application of process modeling in forging industry, its limitation, future developments, and technology needs.
(2:30 p.m.) ADVANCED CAE IN THE FORGING INDUSTRY: T.B. Furman and B. Ravi Shankar, Ladish Company, Inc., Cudahy, WI 53110 Review of advanced analytical tools for simulating fabrication processes of titanium and nickel based alloys. Improved mechanical properties and quality are attained through computer simulation. These techniques have reduced production lead-times by eliminating multiple shop floor tryouts. Forging modeling predicts metal flow, die cavity fill, forming load and the dl.stribution of stress, strain, strain-rate and temperature encountered throughout the process. Forming loads are used for non-linear, elastic-plastic die stress analyses. Computer simulations of forging processes elimination defect formation, undesirable flow patterns, and aggressively designing dies and forging processes. Heat treat modeling predicts transient temperature and stress distributions and reSUlting distortion. Film coefficients, as function of temperature and space, are estimated using a nonlinear inverse two dimensional heat transfer code. Heat-up. simulation predicts time, stresses, effects of creep, temperature distribution, strains, mechanical failures and residual stresses.
(3:00 p.m.) COMPUTER AIDED ENGINEERING APPLIED TO THE FORGING OF A TITANIUM INTEGRALLY BLADED ROTOR: J.M. Walters and R.A. Wallis, Cameron Forge Company, Houston, TX 77251. As part of the Manufacturing Technology for Advanced Propulsion Materials Program computer aided engineering (CAE) was applied to the manufacture of a titanium integrally bladed rotor. Computer models were used to develop a manufacturing process that would result in an acceptable product the first time, thus replacing the expensive and time consuming trial runs of the past.
A metal deformation model was used to predict metal flow during the forging process and the loads predicted by the model were subsequently used in the design of the dies. Thermal models were used to predict sagging of the part in the furnace during heat treatment as well as the stresses generated in the disk during quenching. The use of CAE resulted in the manufacture of five acceptable rotors without the need for a trial run.
(3:30 p.m.)
BREAK
(3:40 p.m.) PHASE SELECTION IN SURFACE MELTS OF TITANIUM INTERMETALLICS: A.J. Paul, C.G. Levi and R. Mehrabian, Materials Department, College of Engineering, University of California, Santa Barbara, CA 93106 Experimental evidence suggests that kinetic limitations to the re-growth of surface melted intermetallic compounds may result in the development of substantial supercoolings inthe liquid ahead of the interface. Thus, a competition is established between the re-growth of the original substrate compound and the nucleation of alternate phases. A one-dimensional finite difference model was developed to study surface melting under high energy beams and the ensuing rapid solidification process. Thermodynamic and kinetic relationships were introduced in a standard heat transfer formulation to describe the behavior of the liquid-solid interface during re-growth. The effect of operating parameters, like input energy and pulse length, on undercoolability and phase selection will be examined. Microstructural observations in titanium aluminides and other intermetallics will be discussed in the light of the trends established by the modelling work.
(4: 10 p.m.) PROCESS MODELING OF FORGING AND HEAT TREATMENT: T.E. Howson and H.E. Delgado, Wyman Gordon Company, North Grafton, ~1A 01536 Process modeling to validate designs for manufacture of a superalloy integrally bladed rotor (IBR) for an Air Force Manufacturing Technology Program will be described. Computer modeling covered metal flow simulation, die stress analysis, and heat treatment analysis. Metal flow simulations and die stress analysis predicted successful production of the IBR with proposed die designs and forge process. Heat treatment modeling indicated that proposed support of a thin web region of the IBR was inadequate and that distortion of the IBR geometry would occur during solution treatment. Support was redesigned and validated by modeling. Analysis confirmed stability of the IBR geometry during the oil quench after solution treatment. Physical modeling included subscale IBR's to evaluate forging parameters, and forging a wedge-shaped specimen to obtain material for evaluation of microstructures and mechanical properties. Validation of the modeling through full scale production of the IBR's will be discussed. Integration of modeling tools into production and application of the tools to forging design for superalloys and titanium alloys will be described.
(4:40 p.m.) EVALUATION OF A TITANIUM DIE FAILURE: D. F. Carbaugh and M. Ripepi, Alcoa Cleveland Works, Alcoa Technical Center, PA Traditionally the design of forging dies has depended on heuristics. These heuristics have met with limited success, especially when forgers attempt to push the limits of proven designs. In these instances, recent advances in computer aided engineering offer the forging engineer a better means to evaluate aggressive die designs both before and after the fact. This presentation focuses on the use of current analytical tools for investigating the failure of a titanium forging die. The use of a general purpose finite element code, ANSYS, coupled with a specialty non-linear finite element code, ALPID, are employed to determine the interaction between the die and the workpiece during the forging process. These analytical tools have helped to expose critical aspects of the forging process which warrant special attention. The more proactive use of these tools allows forgers to push technology due to an improved understanding. The net result of their use should provide more reliable die designs, with fewer premature die failures.
