"The Gateway to Technical Excellence" GENERAL INFORMATION Advance Registration. All attendees are encouraged to register in advance to avoid registration delays at the Meeting. For convenience, you may charge your registration fee to your VISA or MasterCard charge account. Register in advance and pick up your Meeting program at the Advance Registration desk in the Main Lobby of the Cervantes Convention Center. Your Meeting badge, receipt, social function tickets will be mailed to you. The deadline for receipt of advance registration is Friday, September 24, 1982. Registration fees are indicated on the Advance Registration form inside. Transportation. The St. Louis airport, Lambert International, is served by American, Delta, Eastern, Ozark, TWA, Continental, Frontier, Northwest Orient, Pan American, Southern, United, and USAir carriers. It is located 15 miles (20 minutes) from downtown St. Louis. The taxi fare from Lambert International is approximately 518.00 one way. The airport limousine costs 55.00 per person. Registration. Registration for the 1982 Fall Meeting will be located in the Main Lobby of the CeNantes Convention Center. The following registration hours will be obseNed: Sun., Mon" Tues" Wed., Thurs.,

Oct. Oct. Oct. Oct. Oct.

24: 25: 26: 27: 28:

4:00 7:30 7:30 7:30 7:30

p.m. a.m. a.m. a.m. a.m.

-

8:00 4:00 4:00 4:00 10:00

p.m. p.m. p.m. p.m. a.m.

ALL ATTENDEES OF THE 1982 FALL MEETING MUST REGISTER. REFUND POLICY If circumstances require you to cancel your TMS-AIME registration, written notice must be sent to TMSAIME, 420 Commonwealth Dr., Warrendale, PA 15086, postmarked no later than October 8, 1982. All tickets for which refunds are being claimed should be returned with your request. NO REFUNDS ON REGISTRATION, SEMINARS, EDUCATIONAL COURSES, SOCIAL FUNCTION TICKETS, OR PUBLICATIONS WILL BE ISSUED AFTER THE OCTOBER 8 DEADLINE.

Hotel Accommodations. Housing accommodations have been reserved at the Radisson St. Louis (TMS-AIME & MSDASM Headquarters Hotel), and the Sheraton St. Louis (ASM Headquarters Hotel). The Cervantes Convention Center (technical sessions and registration location) is located across the street from both the Radisson and Sheraton Hotels. Limited student and government employee housing has been reserved at various St. Louis hotels (see Housing ReseNation form) at special triple/quad and regular government rates, respectively. Housing Reservation forms should be mailed to the TMS-AIME Housing Bureau NO LA TER THAN FRIDA Y, September, 24, 1982. (See the Housing Reservation form for address.)

~~t-..LL~.f.

~~

~ ~t=) '<";rOCIl'5~· ~ Extensive programming by two leading technical societies.

Technical Program. TMS-AIME technical sessions will be held in the upper level of the Cervantes Convention Center, with the Poster Session. This level is conveniently linked with the Main Lobby by a centrally placed escalator. Morning sessions will begin at 8:30 a.m. and continue until 12 noon on Monday and Thursday. Tuesday and Wednesday morning sessions will end at 11 :30 a.m. to allow attendance at the Joint TMS/ASM Distinguished Lecture in Materials and SoCiety and the ASM Campbell Lecture. Afternoon sessions will begin at 2:00 p.m. and continue until 5:00 p.m. Starting times have been scheduled for each presentation. This will enable attendees to plan their daily schedule in advance to maximize their time. The complete program with abstracts will be published in the August issue of the Joumal of Metals. Poster Session. A Poster Session will once again be held in conjunction with the

Fall Meeting. The Session, located on the Cervantes Convention Center's upper level, will begin at 8:00 a.m. on Monday, October 25, and continue throughout the Meeting during registration hours. The TMS-AIME Program Committee encourages attendees to partiCipate in this valuable part of the technical program. Therefore, in addition to Poster Session authors' individual presentation hours, the following "Official Poster Session Hours" have been established to avoid conflict with other technical sessions and social activities: Monday, October 25: 1:00 p.m. - 2:00 p.m. Wednesday, October 27: 1:00 p.m. - 2:00 p.m. Abtracts for this Poster Session will be accepted through September 10, 1982, and should be sent to: Fall Meeting Poster Session, 420 Commonwealth Drive, Warrendale, PA 15086.

WELCOMING RECEPTION. The 1982 Fall Meeting Welcoming Reception will be held Monday, October 25 in the Admiral Ballroom on the first floor of the Radisson Hotel. The reception will begin promptly at 5:00 p.m. and close at 6:00 p.m. Beer and dry snacks will be served. This reception is an opportunity for metallurgists and material scientists to renew old friendships, as well as make new friends.

TMS·AIME STUDENT ACTIVITIES Student Housing. Special hotel room rates have been established at the Marriott's Pavilion, Holiday Inn Riverfront, The Mayfair, and St. Louisian hotels. These rates include triple or quad rates. If you are a student and want to take advantage of this offer, use the Housing Reservation form. Note that it is a student housing application form. There are a limited number of hotel rooms at the student rate, so get your forms to the TMS and ASM Housing Bureau in St. Louis early. NOTE: The housing form must include the names of

all students. the Bureau '11111 not oSSIgn roommates

Student Affa irs Comm ittee. The TMSAIME Student Affairs Committee will have a lable at the TMS Registration area located in the Cervantes Convention Center. Whereas this committee is "the Students' Voice" of Ihe Society_ all students and Foculty Sponsors are ancouraged to stop by the table and dis· cuss any suggestions. charges, 01 problems relating to sludent membership in TMS-AIME. Faculty Spo nsors Coffee _The TMS·AIME Student Affairs Committee 'Hill sponsor a table 01 the TMS-AIME Authors' Coffee located in Room 274 of the Cervantes Convention Cenler from 7:15 om. untlt 8:30 o,m. daily. All TMS'AIME faculty sponsors ore urged to stop by. enjoy some coffee and danish. and discuss sludent chapter and membership pro· grams with the Student Afalrs Committee members. Student Monitors. Arrangements have been mode to allow TMS students the opportunity to defray their Fait Meeting expenses by working as Session Monitors, Each monitor wilt be responsible for not only the reporting of NV equipment malfunction_ but also fOI monitoring the number of session otlendees for each paper. These monitor positions are limited and wilt be assigned on a first· come. first-served basis. Inleresled siudents should contact by writing Tom DeSalvo. Student Monllor Coordinator. TM5-AIME Headquarters. 420 Commonwealth Drive. Warrendale. PA 15086. NO lATER THAN FRIDAY. September J. 1982. or via telephone at [412) 776·9050.

J

SPECIAL SYMPOSIA Programmed by TMS-AIME Constituent Phases in Aluminum Alloys: Mon .. a .m, & p.m. • Design Requirements for Titanium Alloys' Tues .. a .m. & p.m. • Groin Refinement in Castings & Welds: Mon .. a .m. & p,m.: Tues. a.m. • Hot Isostatic Pressing of Metal & Ceramic Powders: Mon .. a.m. & p.m. • Hydrogen in Solution in Metals & Alloys~ Mon., a.m. & p,m.: Tues. a.m. Recent Advances in Superalloy Technology' Weds_. a m. & pm.

Programmed by TMS-AIME and MSD-ASM • Chemistry and Physics of Rapidly Solidified Materials: Tues .. p.m.; Weds., a.m. & p.m .. Thurs. a.m. • Groin 80undary Chemistry & Environmental Interaclions: Tues .. p.m.: Weds .. a.m. • liquid & Solid Metal Induced Embrittlement of Metals: Man .. p.m.; Tues .. a.m. & p.m .• Weds .. a.m. & p.m .. Thurs .. a.m. • Modeling 01 the Structure & Properties of Amorphous Materials Man .. a.m. & pm. Tues .. am. & p.m. • Radratian Damage Analysis for Fusion Reactors: Man .. p.m .. lues.. a.m. & p.m.: Weds .. a.m. & p.m .. Thurs., a.m.

Programmed by TMS-AIME and ASM Mec hanic al Worki ng &. Forming Division Novel Techniques in Metal Defarmallon Testing: Mon .. a m. & p.m.; Tues .. a.m. & p.m

./

SPECIAL EVENTS TMS-AIME 1982 Fall Meellng WELCOMI NG RE CEPTION Monday. October 25. 1982 5:00 pm. - 6:00 p.m. Admiral Ballroom. Radisson SI. Louis Free Beer and Dfy Snacks

JOINT OISTINGUISHED LECTURE IN MATERIALS AND SOCIETY Tuesday_ October 26. 1982. 11:45 am Room 131.CervantesConvenlionCenler lecturer: Morris Cohen Topic: "Materialism and Meaning A Philosophical Viewpoint"

'~~.

-- SpfCI"~

OfffllZ

For more tha n a century. the symbol of personal commitment to technical excellence in metallurgy and materials science has been membership in The Metallurgical Society of AIME. The membership of the Society comprises the broadest possible spectrum o f professionals. a nd. wherever found. they are the leaders in their respective fields. Three-quarters o f TMS members have progressed Into middle or upper management; almosl two-thirds hold advanced degrees. Without question. the membership of The Metal'urgical Society represents the 'Who's Who" of our profession, The St.louis arCh islhe Gate, way to the Wesl- it symbolically can be your gateway to technical excellence. Tha Metallurgical Society eKiends to you a special invitation to join "TMS·AIME" and participate in the 1982 TMS-AIME foil Meeling and ASM Metols Cangressl Exposition. If you are planning to attend the Meeling. why not submit the Mem-

TMS-AIME - "LEADERS IN THE STATE-OF-THE-ART"

bership Application, along with your ad· vance registratiOn form. save 555.00 [more tha n TMS membership dues) and come 10 st. Louis as a membertln add ition, submission of the attached Mem-

bership Application a t this time entilles you to purchase all TMS-AIME publications at the substantially reduced member rate. which. in most cases. amoun ts to a 40-50% savings!

SEE YOU IN ST. LOUIS AS A MEMBER OF TMS-AIMEI To Apply for Membe rship: Complete the Membership Application included on the lost page at this brochure, giving full details of your education. military service. engineering experience. and any time spenl In charge of engineering work. Have Ihe compleled application endorsed by a member of AI ME who is tamillar wl1h your qualificalions and professional bock· ground.

• Pay no money now! (Dues and fees are not payable until elec1!on 10 membership. II elected. you will be billed this Fall for your 1983 mem· bership dues.) • Forward your Membership Appllcalion and Advonce Registration form 10 : The Melallurglcal Society of AIME 420 Commonwealth Drive Warrendale. PA 15086

."Mt{+

TECHNICAL SESSIONS i~~ 1982 FALL MEETING "'0('1 £"'"

The Metallurgical Society of AIME & Materials Science Division of the American Society for Metals

FlaClu re; Toughness. Elevaled Tempe,atures

F,acture: Titan ium .... lIoy.

Fraclure, Inc lusions Impu rilies Non·Met allic,

G,ain Aelinement in Cast ings and Welds

G, aln Aeflnemen t in Castings and Weld,

"

'"

G,ain Boundary Chem l.tryand Envi,onmental Infe'acHo"s

G,.ln Boundary Chemistry and Environmental Inte'aclion.

DeformaHon

Metallurgy of Wea,·AesisHng Mate'lai s

Metallu'gyof Wear·Aesisllng Mate' lais

Struclural Mate,i alS fo' Ground Transportation

Dllluslon

G,aln Aefinement In Castings and Welds

,

,

Defo,mallon

,

"

Pnase T'ansfOfmaHons I

Grain Boundary Phenomena

,

Phase Transformat ions III

O,ldation

• Nove l Techniques In Metat Deformation TeSHng

• Novel TechnIques in Meta l O
'"

"

HOI ISMlaHc PreSSing of Metal and Ceram ic Powders

'0' ProperHes

Surfaces

Design Aequlrement. 10' Titanium Alloys

Design AequlremenlS tor Tit anium Al loy.

Fr.ctu'e·.... na lytlc.1 Technique .

Delormation Pl"ocesslng

Supe,plasticity

Fracw,e Envlronmenlal Ellects

Liquid and Solid Melal Induced Emb,ili lement

Ion Implanlalion and In ' di~lon Eflect .

"

,

.... lIoy Phnn

Powder Metallurgy

Alloy Phase s

" F,lcilon

'0'

Wear

Radlallon Damage .... na lysls lor FUS ion AUClors

,

" Hyd'ogen In Solution in MetalS and .... IIOyS

,

CoMllluenl Phase.

Amo'phous

Mechanical Prope'He. Compo. lllonal Elleci.

Mulllple Loading Eftects on FaHgu e

EfteCIS ot Coatings On Supe'~lIoy Me<;hanlc. 1 P'ope,jies

Aecenl Advancn In Supe,al loy Technology

Re<:enl Advances in Supe,allay Technology

Liquid and Solid Metal Induced Emb'lltle men l

Liquid and Sol id Met al Induced Emtr'iUlemenl

Liquid and Solid Meta l Induced Emb,jlUemenl

"

'"

"

Aadlallon Dama ge An.lysls 10' Fusion Reaclors

Aadial ion Damage An.lysls to, FuSIon AeaclOrs

Aadlallon Damage .... nalysls to, F"sion Reacto,"

"

'"

"

,

,

",

,

,

St,uclural Malerial. to<" G,ound Tran.porlalion

" Time Dependent Phenomena: Creep

'0

"

Envlfonmenlal Enecls· Delorm.l lon

MeChanica l Bellavlor

Time Dependent PIIenom en., Fallgue

.... Ium lnum Alloy.

"

Phase T,ansform at ion s 11

'0' CoatIngs

HOI IscslaHe Pressing 01 Metal and Ceramic Powders

,

Fraclure Mec~anlsm.

, ,,

" "

"

"

'"

l-1ydrogen In Solulion In Mel al S and Alloy.

Hyd'ogen In Solullon in Mel als and Alloy.

"

'"

Con.1ituenl Phases

'0

.... Iumlnum Alloys

"

ChemlSUy and PhYSIC. of A.pid ly Solidified Mate,l al s

,

Overview 01 Pl"OduCl ivit v Iss ues

Pro
Alloys Phase.

,

,

" liquid and Solid Metal Induced Emb,)lli emenl

,

Aadiaiion Damage An.lySi. fo' Fus ion Reaclors

,

Ael' actory MetBl s Bnd Alloy. lor Electrical Appllcallons

G,aln Bounda ry DittuS i"" . nd Kinetics

Chem lslry and Phys ics of RapIdly So lid ified Malerlals

ChemlSlry and Phy Sics of Aapldly Solldlhed MaI8"als

"

'"

Computer Aided

,

P'oceSSlng MelhodS WorkIng/Forming

" Melting

Disloc ation BehaviOr

'0' Solldilicalion

Fractu,e; Hydrogen Embfltllement

F,acwre Mic'OSlructural Ellecl S

liquid and Solid Melal Induced Embfltllemenl

Aadialion Da mage .... naly.ls 10' Fusion AUclors

"

"

'" Grain Boundary Dlllusl"" and Klneli c.

"

• No.,1 Techniques In Melal 0 , to,m&110n I & II: P'ogram med by ASM ISee ASM P'ogr.m G,Id)

E,o.lon

Cheml . lry and PhysIc s 01 Rapid ly Solldilied Male,i als

"

Mulliple Loading ElleclS on Fallgue

,

z

en

o en en w en

oiS ~

en lZ w ~

O~

z 8 w ~

w ~

0:: LL ~

Z ~ O ~

U! -' «

o z o

:I: ~

w

•

!.

•

1 ~1

1J !,

II

"I .,.

" !jl i, 1,

!I!

. ."S:0', ~~

•

j~

,.

.!~~~ ~~ .

! '" '"j " ,

.,

1982 TMS-AIME FALL MEETING TECHNICAL SESSIONS The numbers to the left of each session title indicate the page on which that session is listed in the TMS Fall Meeting Technical Program.

Program Page 46. . Alloy Phases I . 57. . All oy Phases I I 66. .Alloy Phases III 36. . Chemistry & Physics of Rapidly Solidified Materials 47. . Chemistry & Physics of Rapidly Solidified Materials II 58. . Chemistry & Physics of Rapidly Solidified Materials III. 67. . Chemistry & Physics of Rapidly Solidified Materials IV . 48. . Computer Aided Manufacturing (GEM/G-PAC Productivity Symposium). 1. . Constituent Phases in Aluminum Alloys I. 14. . Constituent Phases in Aluminum Alloys II 1. . Deformation 14. . Deformation II . . 15. . Deformation Processing 26. . Design Requirements for Titanium Alloys 37. . Design Requirements for Titanium Alloys II 3. · Diffusion · Dislocation Behavior. 68. 38. · Effects of Coatings on Superalloy Mechanical Properties. 59. · Envi ronmen ta 1 Effects: Deformation 69. · Erosion 4. · Fracture I: Analytical Techniques 5. · Fracture II: Environmental Effects 26. · Fracture II I: Toughness, E1 evated Temperatures 39. · Fracture IV: Titanium Alloys . . . . . . . . . 49. · Fracture V: Inclusions, Impurities, Non-Metallics 60. · Fracture VI: Mechani sms. . . . . . . . 70. · Fracture VII: Hydrogen Embrittlement . 71. · Fracture VIII: r~icrostructural Effects 61. · Friction &Wear 40. · Grai n Boundary Chemistry & Environmental Interactions 50. · Grain Boundary Chemistry &Environmental Interactions II 62. · Grain Boundary Diffusion & Kinetics I. 72. · Grain Boundary Diffusion & Kineti cs II 6. · Grain Boundary Phenomena 7. · Grain Refinement in Castings &He1ds 16. · Grain Refinement in Castings &He1ds II 27. · Grain Refinement in Castings &Welds III 8. . Hot Isostatic Pressing of Metal & Ceramic Powders I. 17. . Hot Isostatic Pressing of Metal & Ceramic Powders II 8. . Hydrogen in Solution in Metals & Alloys I. 18. . Hydrogen in Solution in Metals &Alloys II

Weds AM. Weds PM. Thurs AM Tues PM. Weds AM. \~eds PM. Thurs AM Heds AM. Mon AM Mon PM Mon AM Mon PM Mon PM Tues AM. Tues PM. Mon AM . Thurs AM Tues PM. Weds PM. Thurs AM Mon AM . Mon AM . Tues AM. Tues PM. Weds AM. Weds PM. , Thurs AM Thurs AM Weds PM. Tues PM. Weds MI. Weds PM. Thurs AM Mon AM . Mon AM. Mon PM. Tues AM Mon AM. Mon PM. Mon AM. Mon PM.

Room No. 264 264 264 273 273 273 273 275 275 275 262 262 267 266 266 260 261 267 263 265 267 270 260 260 260 260 263 267 265 261 261 272 272 264 .261 .261 .261 .265 .265 .273 .273

Program Page 28.

9. 19. 29.

41. 51. 62. 73. 29.

51.

74. 30. 42. 10. 20. 31. 42. 63. 75. 10. 20. 32. 43. 33. 21. 11.

22.

33. 52. 63. 44. 23. 34. 44. 53. 64. 75. 54. 65. 56. 56. 66. 35. 45. 12. 25.

. Hydrogen in Solution in Metals

&Alloys III

Tues AM . Ion Implantation & Irradiation Effects. Mon AM. . Liquid & Sol id Metal Induced Embrittlement of Metals Mon PM. . Liquid & Solid Metal Induced Embrittlement of Metals II Tues AM . Liquid &Solid Metal Induced Embrittlement of Metals III Tues PM . Liquid &Solid Metal Induced Embrittlement of Metals IV. Weds AM . Liquid & Solid Metal Induced Embrittlement of Metals V . Weds PM . Liquid &Solid Metal Induced Embrittlement of Metals VI Thurs AM. . Mechanical Behavior &Properties . Tues A~l. . Mechanical Properties: Compositional Effects Weds AM. . Melting &Solidification. Thurs AM. . Metallurgy of Wear Resisting Materials I . Tues AM . Metallurgy of Wear Resisting Materials II Tues PM . Modeling of the Structure &Properties of Amorphous Materials Mon AM . r~odeling of the Structure & Properties of Amorphous Materials II Mon PM . Modeling of the Structure &Properties of Amorphous Materials III. Tues AM Modeling of the Structure & Properties of Amorphous Materials IV Tues PM Multiple Loading Effects on Fatigue I Weds PM Multiple Loading Effects on Fatigue II . Thurs AM. Novel Techniques in Metal Deformation Testing I Mon AM Novel Techniques in Metal Deformation Testing II Mon PM Novel Techniques in Metal Deformation Testing III. Tues AM Novel Techniques in Metal Deformation Testing IV . Tues PM Overview of Productivity Issues (GEt~/G-PAC Productivity Symposium) Tues AM Oxidation &Coatings. Mon PM Phase Transformations I Mon A.'II Phase Transformations II Mon PM Phase Transformations III. Tues AM Powder Metallurgy. Weds AM Processing Methods - Working/Forming (GEM/G-PAC Productivity Symp.). Weds PM Productivity in Extractive Metallurgy (GEM/G-PAC Productivity Symp.) . . Tues PM. Radiation Damage Analysis for Fusion Reactors I . .Mon PM . Radiation Damage Analysis for Fusion Reactors II . Tues AM. Radiation Damage Analysis for Fusion Reactors III . Tues PM. Radiation Damage Analysis for Fusion Reactors III cont'd. & IV. .Weds AM. Radiation Damage Analysis for Fusion Reactors V • .Weds PM. Radiation Damage Analysis for Fusion Reactors V cont'd. & VI. .Thurs AM Recent Advances in Superalloy Technology I. .Weds AM Recent Advances in Superalloy Technology II . .Weds PM .Weds AM Refractory Metals &Alloys for Electrical Applications .Weds AM Structural Materials for Ground Transportation I. Structural Materials for Ground Transportation II .Weds PM Superplasticity. Tues AM. Surfaces . Tues PM. Time Dependent Phenomena: Fatigue Mon AM Time Dependent Phenomena: Creep. Mon PM .

* Fi rst Floor

ii

Room No. .273 .271 .270 .270 .270 .270 .270 .270 .265 .266 .262 .262 .262 .272 .272 .272 .272 .266 .266 .120 * .120 * .264 .264 .275 .264 .263 .263 .263 .265 .275 275 271 271 271 271 271 271 267 267 272 262 262 267 265 266 266

1982 PAPER SELECTIONS The Metallurgical Society publishes hundreds of individual papers at AIME annual and TMS Fall Meetings. These papers represent up-to-the minute finds in metallurgical technology ; the AIME Paper Selection Program provides you with invaluable access to this information. To order your copies simply list the paper number and mail to : Publications Order Dept., The Metallurgical Society of AIME , 420 Commonwealth Dr. , Warrendale, PA 15086, USA or use the handy order form at the back of this issue. To request papers published in previous years list 1977, 1978, 1979, 1980, or 1981 on the order form and that list(s) will be sent to you. Ask for your free copy of our newest publications catalog too. Prepayment is required. Don't wait, send your order today!

Prices - $2.00 AIME members, $4.00 non-members A82-1 Effect of S03 on Corrosion of Process Equipment in Copper Smelters, S. Sarkar .................... .. .... .. ................ . A82-2 Refractory Practice at Ghatsila Copper Smelter, S. K. Key, S. K. Ghost, A. K. Das, and S. K. Roychoudhury

A82-24 Electrolytic Treatment of Dilute Metal Containing Solutions, R. Kammel and Hans-Wilhelm Lieber ........ ...... .. . A82-25 The Effect of Thermomechanical Processing Schedules on the Corrosion of Hydriding Behavior of ZR2.5 WT% NB Nuclear Reactor Pressure Tubing , D. O. Northwood and U. Kasasih ................ .. .............................. .. A82-26 Copper Electrowinning in Chloride Aqueous Solutions, L. Albert and R. Winand .. .. ........................ .............. .. . A82-27 Manganese Electrodeposition from Suphate, Chloride and Mixed Solutions, Including High Current Densities, G. Parissis and R. Winand ............................................ ...... . A82-28 Studies on the Preparation of Pure Niobium Oxide from Pyrochlore Concentrate, G. Gabra .... .. .. .. .... .. ...... .. .... .. A82-29 Auger Investigation of the Film Formed on 26% Cr-1 % Mo Stainless Steel in Boiling McCI2 Solutions, S. C. Tjong .................. ..... .................... ... .................. ........ .... ... . A82-30 The Critical Conditons for the Formation of "A" Segregation in Forging Ingots, H. Yamada , et al . ...... .. .. .. .. .. A82-31 The Effects of a Surface Agent, P20S , and Agitation upon the Gaseous Reduction of PbO-Si02 Melts, U. B. Pal, T. Deb Roy, and G. Simkovich .............. ............ . A82-32 A Mathematical Model for Adsorption of Silver on Active Carbon , Y. B. Hahn, K. J. Hahn, and S. J. 1m ........ .. A82-33 Residual Life Prediction for Jet Engine Rotor Disks at Elevated Temperatures, H. Ohnabe et al. .... .... ........ .. .... . A82-34 Reduction of the Iron Content of Guymine's Bayer Process Liquor, M. I. Johnson ........ ........ ............................ . A82-35 Hydrochloric Acid Leaching of Nickel Sulfide Pr.ecipitates, Mahesh C. Jha, John R. Carlberg, and Gustavo A. Meyer .. .. .. ... ...... .................. ... ......... .. ........ .... .. .. .... ... .. .. ... .. A82-36 Chloride Leaching of Copper ConcentratesPractical Operational Aspects, D. C. McLean .......... .......... .. A82-37 A Technical and Economic Evaluation of the ISA Electrorefining Process in Use at Townsville Copper Refinery, I. J. Perry, D. W. Hoey, and R. B. Ness .. .... .... ...... .. . A82-38 Separation of Nickel from Cobalt in Sulphate Medium by Ion Exchange, L. Rosato, G. B. Harris, and R. W. Stanley ........................................... ...... ...................... .... . A82-39 KHD Humboldt Wedag AG - New Process for Zinc Solution Purification, M. Esna-Ashari and H. Fischer

A82-4 Removal of Harmful Impurities from Iron , Copper, Nickel, and Cobalt Concentrates and ores, H. Tuovinen and P. Seta/a .... ....... .. ......... ... ..... .... .. ........ ......... .. .. .. ... ... ..... .. A82-5 The Use of Energy Production and Energy in Materials Production, K. M. Zwilsky and D. G. Groves ........ .... .. . A82-6 Materials for Energy Production and Energy in Materials Production , K. M. Zwilsky and D. G. Groves .. ...... .. .. A82-7 Stress Corrosion Cracking of Turbine Rotor Steels, S. Somuah, A. Boateng, and D. D. MacDonald ................ .. A82-8 Notes on the Treatment of Complex Sulfide Ores, G. H. Boe, S. Krogh, and B. Ydstie .................................... . A82-9 The Energy Balance as a Design Tool in Zinc Plants, C. Twigge-Molecey, J. Dam berger, and N. Ghatas A82-11 An Effective Viscosity Model for Gas Shirred Liquid Metal Reactors" Y. Sahai and R. I. L. Guthrie ............... .... .. A82-12 Selenium Recovery from Copper Electrolysis Slimes at Mitsubishi-Osaka Refinery, I. Fujimura and A. Katal .. ............. .... ..... .... .... .... ..... ........ ......... ..... .. .. .. ... ............. . . A82-13 The Role of Titanomagnetite Composites in Gravity Separation of New Zealand Ironsands, J. L. Watson and H. F. Law .............................. .................... ....... ............ ...... . .. A82-14 Refractory Performance in TBR Converters in the Non-Ferrous Industry, O. Porkert and H. Barthel ........ .. .... .. A82-15 Functional Technology and System-Design for the Fine Filtration of Tankhouse Electrolytes, P. Berger .. .. .. .... . A82-16 Energy and Fluid Relationships in the Tank House at the Copper Refinery, Townsville, Australia, D. Barrett and T. C. Hunter .................................. ................................ . A82-17 The Present Status of Development of the OSLLead Process. P. Fischer, W. Blum, and H. Maczek .. ........ A82-18 Alternate Manufacturing Technologies for the Production of Multifilamentary Superconducting Wire by the External Bronze Technique, B. Avitzur .............................. .. A82-19 Calculation of Ternary Free Energy of Mixing Using Binary Data, D. N. Lee ...... .... .... .... ...... .... .. .... .... .... .. .. . A82-20 Interruptible Power for Zinc Electrowinning, M. J. Agnew ....... .... ................ .... ...................... .......... ................. .. A82-21 The Critical Conditions for the Formation of Porasities in Forging Ingots, et al. ........ .......... .................... .. A82-22 Development of Electric furnace Slag Cleaning at a Secondary Copper Smelter, H. P. Rajcevic and W. R. Opie .... .............. ... ......... ... ..... .. .......... ...... ........... .......... .... ..... A82-23 Electrowinning From Aqueous Chlorides in SMM 's Nickel and Cobalt Refining Process, M. Fujimori, et al. .. ....

A82-40 Use of Alternative Fuel at Onahama, H. Kohno, H. Asao, and T. Amano .............................................. .. .... ... . A82-41 Nucoalyzer - A New Way to Monitor the Composition of Process Streams, T. Gozani, H. Bernatowicz, and D. R. Brown .. .......... ... .. .. .. .. .... .. ...... .... ....... .... .......... ............ .. A82-42 The Horne Waffle Hood, J. B. W. Bailey, W. R. Kallio, and J. A. Vogt ............................................ .... ...... .... . A82-43 Refractory Technique Improvements in Asarco Non-Ferrous Smelters, Melvin A. Sharp .......................... .. ..

Iv

TMS Fall Meeting 1

1982 FALL MEETING TECHNICAL PROGRAM WITH ABSTRACTS The Metallurgical Society of AIME (in conjunction with 1982 ASM Metals Congress)

MONDAY - OCTOBER 25, AM CONSTITUENT PHASES IN ALUMINUM ALLOYS I Sponsored by the TMS Nonferrous Metals Committee Cervante8 Convention Center Room 275 Sea.iOn Chairman: Henry G. Paris, Alloy Technology Division, Alcoa Laboratories, Alcoa Center, PA 15069.

This paper reviews the processing and properties of a recently developed series of aluminum alloys whose strength depends upon a grain size in the range 0.5 to 3 microns. The grain boundaries are pinned by a uniform dispersion of intermetallic particles, about 0 . 3 microns in diameter, that are formed by the plastic deformation' of a rapidly solidified ingot of eutectic composi tion. The eutectics ' that are used contain insoluble rod-like intermetallic phases and include the Al-fe-MIl , AI-N i , and. AI-Ca-Zn systems . The final grain ' size de pends primarily upon the ingot structure and the final annealing temperature. Grain growth appears to be governed by the diffusion controlled coarsening of the intermetallic particles; accordingly, the grain size increases (and yield strength decreases) gradually and continuously as the final annealing temperature i s increased .

MoDd8y, 'October 26, 1982

8:30

a.m.

8:30 , a.m.

11:10 a.m. THE CONTROL OF MICROSTRUCTURAL EVALUATION AND MECHANI CAL PROPERTIES IN X7 0 90- TYPE Al PI M ALLOYS BY THE CONSTITUENT PHASES Al9FeNi OR A1 9 (Fe, Co) 2: J . A. Walker, H. G . Paris, A. Hafee z, Alloy Technology Divisio n, and M. E . Dalfonso,

Fabrication Technology Di v ision, Alcoa Laboratories , Alco a

MICROSTRUCTURE OF CONSTITUENT PI1ASES IN ALUMINUM ALLOYS, Th6mas H. Sanders, Jr., School of Materials Engineering, Purdue University, West Lafayette, IN 47907. The size and size distribution of constituent phases in aluminum alloys affect strength and duct il i ty. An unders tandi ng of the micros tructure is necessary if programs ina 11 oys development are to proceed efficiently. This paper will review the techniques of microstructure analysis with illustrations from high strength aluminum alloys produced by 11M and P/M techniques.

9:10 a . m. TIlE RECRYSTALLISATION TEMPERATIJRE OF 'NO-PHASE AllJ)YS AS A FUNCIION OF TIlE PARTICLE SIZE AND SPACING. P. Cotterill,

Department of Metallurgy and Materials Technology, UnIversIty of Surrey, U. K. The paper describes recrystallisation in dispersion alloys and shows that many of apparent anomalies reflect di fferences in the methods used to describe the microstructure rather than differences in recrystallisation behaviour. It suggests that if data are presented in terms of an unambiguous description of the particle size and spacing, and the recrystallisation tempet:ature is related to that ~f relevant particle-free matrix ptJ,ase, then it is poss ib Ie to cons truct a gene ralised di agram which relates recrystallis8tion behaviour to the particle size and spacing for a wide range of ~luminium alloys; thus peru tting an interchange of cO'q)arisons over a greater range of disp.ersions than is normally possible in a single alloy system.

9:50 a.m. RECRYSTALLISATION IN AN ALUMINIl'M ALLOY CONTAINING 'NO TYPES AND SIZE OF DISPERSED PARTICLE. C. A. Romanowski, College of

Engineering, Dre xel University and P . Cotterill, Department of Metallurgy and Materials Technology, VnL verSI ty of Surtey , l' .K. Current knowledge of the recrystallisation characteristics of dispersion alloys has been largely developed on the basis of alloys containing a single type and size of dispersed particle. This paper describes the behaviour of a typical engineering alloy containing coarse intermetallic particles alongside finer precipitates . At 360 0 C recrystallisation originated at the coarser particles and went to completion b y migration through the finer particles in the matrix phase, whereas at 2700 C there was no recrystallisation and softening oc c urred by other mechanisms. These effects are interpreted in terms of the different influences of the two particle types which were examined in separate alloys each containing one o f the basic particle modes.

Center. PA

1 5069

Al P/ M alloy X7090 contains both aluminum oxide and intermetallic constituent phase AI g (Co,Fe) 2 for grain structure control . This s tudy illus trates the influer c e o f c onstituent ide ntity and volume fra c t i on o n t h e me c hani c al and c orros ion properti e s of a base X7090, Al-8.0 wt\ Zn-2.5 wt\ Mg-l.O wt\ eu composition. While increasing the volume fracti o n of c onstituent signific antly influences the amou n t o f rec rystalliz ed texture component in e x trusions , the stre ngths are o nl y modestly changed. A more subs ta n tial i n fluence o n toughness is achieved by variation in· volume fr a ction . The c athodi c polarization data sugges t a n influence on the cathodic reducti o n sites .

DEFORMATION I TMS-AIME General Abstract Session Monday, OctolMlr 26, 1982 Cervantell Convention Center 8:30 a.m. Room 262 Se.siOn Chairman: Professor Howard Kuhn, University of Pittsburgh, Department of MetallurgieallMaterials Engineering, Pittsburgh, PA 15261.

8:30 a.m. A MICROSTRUCTURAL COMPARISON OF PLANE STRAIN AND UNIAXIAL TENSILE DEFORMATION OF 70 /3 0 BRASS: * K. P . Staudhammer and M. G . Stout Lo s Alamo s National Laborat o ry, Los Alamo s , NM 8 754 5 .

We have mea s ur e d the stress - s tr a in be hav ior o f ann e al e d 70 /30 bra s s tubes 25 ~m g rain s ize tested und e r unaixial t e n s i o n a nd plane strain conditions, «(Z = - ER • E n = 0 >, t o an effective von Mises strain o f .40. Th e fl ow s tr es s in uniaxial tens i on was found to be . . . 101 hi g he r t he n that in plane strain when compare d on the ba s i s o f the von Mi ses e ff ec tiv e s tr ess- s tr a in c rit e rion. A quantitative compari s on via e l ec tron micr osc o py of twin volume fractions, twin s pacing a nd e xtend ed di s locat ion spa c in g has bee n made of th e two mi c r os tru c tur es t o ( vm = . 4 . At E v rn = .4 we f ound a twin volume fracti o n of -15%, and esse ntiall y th e s ame twin a nd e xt e nd dislocation s pacings for bo th s train s tat es . Bec au se o f th e similar mi c rostru c tur es we be lieve that the diff e r e nce in flow s tr es s r e sult s from th e tube t e xture Th ese a nd o th e r r es ult s of th ~ mic ro s tru c tural comp a ris on will be prese nt ed and di s cu ssed. ~""'Wo rk perfo rmed und e r t he au s pi ces o f th e U.S. De partment of Energ y , Offi ce o f Ba s i c Energy Sc i e nces, Di v isi o n o f Mat e rial s Sc i e nc es.

8:50 a.m. THE STRAI N RATE SENSI Tl VI TY OF TENSI LE DUC Tl LI TY OF CU AND CU-

30ZN: * P. S. Follansbee and S. S. He cker , Lo s Alamo s Nati onal Laborator y , Los Alamos , NM 87 54 5

10:30 a.m . CON!ROL OF GRAIN SIZE BY INT ERMETALLlC PARTICLES:

L. R. Morris

and David M. Moore, Alcan International Ltd ., King s ton Laboratories, Box 8400, Kings ton, Ontario, Canada, K7L 4Z4.

The s train rat e s en s i t i v ity of ten s il e ductilit y i s a n impo rtan t c o nsid e ration in formi ng ope rati o n s , ye t tr e compl e x factor s which influe nc e t e nsil e ductility ha ve limited th e pre-

2 TMS Fall Meeting dictive capability. We

have

initiated

an

investigation of

the

effect of strain rate on the tensile ductility of model FCC r.laterials to d('tcrmine whether in these simp! ified systems the factors that control tensile ductility can be isolated. In eu the

uni'form elongation is relatively independent of strain rate below a strain ratf' of ~lOO 5- 1 , but at higher strain rates, the uniform elongation increases dramatically. The factors which influence tf'nsilc ductility, including the contributions of the strain h.Jrdcning coefficient, strain rate sensitivity, and thermal effects, will be discussed in light of recent expefiments \-,lith eu and Cu-V)Zn.

h'ork perforF.lf'd--undf'r thf' aupices of U.S. Dept. of Energy.

9:10 a.m. WORKHAPCENING Afm STt,TE VP,PIA8LE APPROACH HI COPPEP A.T rWOM T(f,PERATURE:* S.-P. Haonula, f'.A. Korhonen, Che-Yu Li, Depart"'ent of ~,ateriaTs--Scieoce & Engioeering, Cornell University, Rard Ha 11, I thau, NY 14853 The Yiorkhardening characteristic, of corrmercially pure 99.99% Copper are measured in a uni-oxi"l constant displacement rate testing machine, ~onotonic loadings are conducted up to 30% of r.on-elastic strain using different c.isplacement rates, and

material state parameters are determined by load relaxation tests. In log a - log ~ representation, Copper shows at room tel'lperature almost 1 ioear behavior with the slope increesing with strain. A, corparison between monotonic and cycl ic hardening of copper is made and the significance of the experimental findings will be discussed in the tems of the state variable approach for non-elastic defomation. *Research sponsored by the Departl'lent of Energy through the r~aterials Science Division.

9:30 a.m. A !-10DEL TO DESCRIBE THE EXTRAORDINARY WORK HARDENING IN A DUAL-

PHASE STEEL. * Walter S. Owen and Ana M. Sarosiek, Department of ~aterials Science & Engineering, MIT, 77 Massachusetts Ave., Room 8-309, Cambridge, Massachusetts, 02139 It is shown that the work hardening of a dual-phase steel is much greater than predicted by continuum models or by the AshbyBrown model of dispersion hardening. A model is presented in outline in which the major contribution to the work hardening is attributed to localized plastic flow in the ferrite matrix induced by the y -+a' transformation in the second phase. It is shown that the flow curve predicted by the model is in fair agreement with that measured over a strain range from 10- 6 to macroscopic strains.

*Research sponsored by the Amax Foundation and the National Science Foundation.

concepts of cell wall "polarization ll and "memory" have been used to account for the anomalous tensile behavior of the prestrained material.

10:30 a.m. EFFECT OF

~!n

and Si ON THE PRECIPITATION OF TiC IN Ti-BEARING

HSLA STEELS: P. Plassiard, T. Chandra, M.C. Akben and J.J. Jonas, Department of Metallurgical Engineering, McGill University, 3450 University Street, Montreal, Canada H3A 2A7 The effect of increasing the Mn and 5i concentrations on the dynamic precipitation kinetics of TiC was investigated in a series of 0.17. C - 0.17. Ti steels. Following an austenitization treatment at 1260°C for 30 minutes, constant true strain rate compression tests were carried out in the strain rate range from 10- 5 to 1 5- 1 at testing temperatures of 925, 975 and 102SoC. The dependence of the peak strain on strain rate and composition was established at each of the three temperatures and the dynamic precipitation kinetics of TiC were determined in this way. Precipitation began in approximately 5 seconds in the 1% MIl - 0.257. Si steel. The results indicate that a decrease in the MIl level to 0.57. or an increase to 1.5% shifts the nose of the PTT curve to the left or to the right respectively. By contrast, an increase in Si level is equivalent to a decrease in the Mn concentration, and vice versa. This phenomenon is discussed in terms of the influence of Mn and 5i on the activity coefficients of C and N, and on the diffusivity of TL

10:50 a.m. DY/iAMIC RECRYSTALLIZATION DURING THE TRANSIENT DEFORMATION OF A VANADIUM HSLA STEEL. 1. Sakai. M.G. Akben and J.J. Jonas,

Dept. of Metallurgical Engineering, McGlll University, 3450 University Street,Montreal, Canada H3A 2A7 The shapes of dynamic recrystallization stress-strain curves were studied following an increase or decrease in strain rate after deformation into the steady state region. For this purpose, a vanadium HSLA steel was hot compressed at constant true strain rates of 5.6 x 10- 5 s- 1 to 7.4 x 10- 2 s- 1 at a testing temperature of 1000°C. After an increase in strain rate, the flow curve displayed a single peak; the peak strain in this case was smaller than that associated with the annealed structure. After a decrease in strain rate, by contrast, the flow curve displayed a mUltiple peak; under these conditions was nearly equal to that for the annealed structure. As a result, the critical condition at which the shape of the stress-strain curve changes from the multiple to the single peak type is DS1 = Dsz • (Here DSI and DS2 are the stable dynamic recrystallized grain sizes before and after the change in strain rate, respectively.) This differs from the critical condition for the annealed structure, Le. Do = 20 s ' where Do and Ds are the initial and stable grain sizes. These results are interpreted in terms of a grain size based model for dynamic recrystallization.

(.p

cp

11: 10 a.m. DEFORNATlON BEHAVIOR OF FINE-GRAINED PM/IN-100 STUDIED BY LOAD

9:50 a.m. DISLOCATION STRENGTHENING IN WARM-ROLLED IRON: * C. Persad and D.L. Bourell, Materials Science and Engineering Program, The University of Texas, Austin, TX 78712 A constitutive equation relating the room-temperature yield strength to mi cros tructura 1 parameters has been proposed. Thi s equation separates the influence of initial grain size, subgrain size and dislocation density. Extensive experimental support was sought through the use of seven initial grain sizes of Armco iron rolled at 500 0 C to true thi ckness s tra i ns rangi ng from -0.5 to-3.75. Yield strength was detennined by longitudinal tension tests. Initial grain size, subgrain size and dislocation dens ity were determi ned by opt i ca 1 and transmi ss ion elect ron mi croscopy. Elements of the proposed equation are compared to the experimenta 1 data. *This research was funded in part by NSF Grant #I»1R-8113087.

RELAXATION: t S. K. Srivastava* and A. K. Hukherjee, University of California, Davis, CA 95616 A computerized data acquisition system has been used to study load relaxation of fine-grained PM/IN-IOO. The load-drop data were gathered for 80 seconds, @ 10 readings/sec. at test temperatures from 1203°K to l278°K. The digital data were manipulated to determine various thermomechanical parameters. Strain rate sensitivities (m, a measure of superplasticity) were determined to range from 0.52 at l203°K to 0.39 at l278°K. The internal stresses were determined to vary from 4.80 MPa at l203°K to 1.75 MPa at l278°K. Internal stresses are believed to be associated with gamma prime particles. Deformation activation energy at 30 MFa was determined as 421.5 ± 21 KJ/mol, but it did not relate to any known diffusion or deformation mode. Finally, it is shown that flow stress-strain rate relationship over a wide range of strain rates could be generated using a single test.

t Financially Supported by NSF IIDMR-7727724 Presently at Cabot Corporation, Kokomo, IN 46901

:all:

11:30 a.m. 10:10 a.m. MICROSTRUCTURAL MECHANISM FOR THE ANOMALOUS TENSILE BEHAVIOR OF ALUMINUM-KILLED STEEL PRESTRAINED IN PLANE STRAIN TENSION: J. V. Laukonis, Physics Department; B. V. N. Rao, Analytical Chemistry Department, General Motors Research Laboratories, Warren, MI 48090-9055 Aluminum-killed steel subjected to a tensile deformation following a prestrain in plane strain tension (PST) exhibits increases in yield strength, an earlier onset of strain localization, and less work hardening than it would exhibit had it been subjected to an equivalent uniaxial prestrain. The effects are generally larger if the tensile axis is orthogonal to the prestrain axis. To understand the substructural origin of such anomalous b~havior, transmission electron microscopy of the steel was performed after each step of processing. It has been found that when the tensile axis is orthogonal to the prestrain axis, the dislocation cell structure of the prestrain becomes unstable and "dissolves", whereas, it is stable and reinforced if the tensile axis is parallel to the prest rain axis. These differences in substructure and the

KINETICS OF STRAIN AGlllC IN A

:~OLYBDENUM-CONTAINING

DUAL-

PHASE STEEL: M..A. Jager*, General Dynamics Co., Electric Boat DiVision, Groton, CT 063;,0; J.A. Shields, Jr.*, Clir;pax Co., Research Laboratory, Ann Arbor, MI 1.8:06. The kinetics of strain aging were invest'igated in a moiybdenumcontaininG as-rolled dual-phase steel, using the change in yield stress and LUders strain to characterize the strain aging process. 30th parameters exhi bi ted two rer;im~s of aging response wher. plotted as a function of aginG ti!":lc: a comparatively small response occurinp.; at short aeing times, and a larp;e response occuring at longer acing times. The short-time aginr; response had an activation enervs of 113-115 k,;!rrol (22-23 kcal!mol). while the long-time aging response had ar; activation energy of 152-J57 k';/ mo~ (30-3? kcal/mo:!.). The lower acti vation e':1ergy process is believed to reflect the fonnatio':1 of interstitial carbon atmospheres around dislocations, resulting :'n relatively weak pinning. The fact that the activation enerp:y is elevated above that normally observed for carbon diffusior. ir. iron is believed to be a result

TMS Fall Meeting 3 of the inf'::'ue:1ce of ~o or. the dif:"'usion of carbon. The r.igh activation enerr';Y process is believed to reCect the precipitation of carbides, resu~tinf, ir. stronp; pinning. '!'he shape of the stress-st!'ain curves observed in the investigation is consistent \o/'ith such a:1 interpretation. U:'timute tensile strenGth and unii'o::'m elonE<;ation were also monitored during the agi:1g process. but displayed considerab':'y l':1ore scatter than did the yieJd strenGth and LUders st!'ain. ':'ensil c st r-er.p;th te:1ded to increase, while uni!'orr:l elonf,ation decreased as af,inp; pror;ressed, consistc!1t with the res'cllts of other investiGations.

positions were used to demonstrate the negligible effect of the Cr concentration gradient on the diffusion of AI. Kirkendall-type porosi ty in the y+S side of the couple indicated that the intrinsic diffusivity of Al was greater than that of Cr. Small additions of Zr to the y+B alloys, which significantly improves the oxidation resistance during thermal cycling, did not affect the y /y+S interface motion but did result in a reduction of the Kirkendall-type porosity.

9:50 a.m. *Formerly Wayr.e State Uni veri sty ,Detroi t, 1·11 48102

DIFFUSION TMS-AIME General Abstract Session Monday, October 25, 1982

Cervantes Convention Center Room 260 Session Chairman: Dr. J. I. Goldstein, Lehigh University, Metallurgy Dept., Whitaker Lab #5, Bethlehem, PA 18015.

8:30 a.m.

8:30 a.m. AN ANALYSIS OF CONCENTRATION PROFILES FOR FLUXES, DIFFUSION DEPTHS AND ZERO-FLUX PLANES IN MULTI COMPONENT DIFFUSION: M. A. Dayananda, School of Materials Engineering, Purdue University, W. Lafayette, Indi ana. Concentration profiles expressed on the basis of a relative concentration variable for each component are analyzed for the determination of interdiffusion fluxes in multi component diffusion. New relations are developed to describe internal consistency among the concentration profiles of the various components. The individual concentration profiles intersect at a CORlnon cross-over composition where the reiative concentrations of all components are identical. A link is developed between the cross-over composition and the depths of the di ffus i on zone on ei ther of the Ma tano plane for a diffusion couple. The cross-over composition also corresponds to the average relative concentration of each component over the diffusion zone. The identification of a zero-flux plane from concentration profiles is also described.

ANOMALOUS DIFFUSION OF CARBON IN A 70/30 Fe-Ni ALLOY: Manfred University of Mlssouri-Rolla, MO; D. A. Powers, NRC, Was lngton, DC; and L. R. Chapman, Union Carbide Corp., Oak Ridge, TN Wut~i9,

I nterna 1 fri ct i on and magneti c d i saccanmodat i on measurements have been made to complement the available high temperature carbon diffusion data in a 70/30 Fe-Ni alloy. On the basis of all data, an Arrhenius plot of the carbon diffusivity varying over thirteen orders of magnitude can be prepared. This plot is distinctly non 1 i near. It is proposed to ana 1yze the non 1 i neari ty by assumi ng the anti symmetrical component of the self-trapping strain becomes zero as the martensite start temperature is approached. Such an ana lys is reproduces the anOf'la lous temperature dependence 9f the diffusivity. It further suggests that the sy:rrnetrical and antisy:rrnetri ca 1 components of the self-trapping energy contri bute about equally to the activation energy of interstitial diffusion in this face centered cubic alloy.

10:10 a.m. THEORY OF THERMAL DIFFUSION lN BINARY ALLOYS. * Ryoichi Kikuchi,** Takuma Ishika.,a: * and Hiroshi Sato, School of Materials Engineering, Purdue University, We:;t Lafayette, IN 47907. Problems of material transport under the temperature and the concentration gradient have been examined by the path probability method of irreversible statistical mechanics. The Onsager equation for diffusion for disordered binary alloys under the temperature and the concentration gradient has been derived from the atomistic point of view. Special attention is paid to the examination of the energy flm. involved.

8:50 a.m. INTERDIFFUSION IN Fe-Ni-Cr SYSTEM: J. G. Duh and M. A. Dayananda, School of Materials Engineerinq, Purdue UnlVersity, W. lafayette, Indiana. Ternary diffusion has been investigated in the Fe-Ni-Cr system with solid-solid diffusion couples and interdiffusion coefficients have been determined over a wide range of compositions in the y(fcc) region of the ternary isotherm at 1100·C. This investigation includes selected diffusion couples with terminal alloys characterized by similar thermodynamic activities of either Cr, Ni, and Fe for the development of zero-flux planes and flux reversals of the diffus i ng components. Ternary di ffus ion 'Pa ths are presented and described by means of analytical expressions on the basis of two important diffusion parameters, the cross-over composition of the diffusion path and the path slope at the cross-over composition. Data on interdiffus ion coeffi ci ents wi 11 also be di scussed.

9:10 a.m. DIFFUSION AND THE VACANCY-WIND EFFECT IN A CONCENTRATED BINARY

ALLPY WITH ORDER: * G. E. Murch, Chemistry Division, Argonne National Laboratory, Argonne, IL 60439. This paper is concerned with the nature of the vacancy-wind term occurring in the equation connecting the intrinsic and tracer diffusion coefficients and the chemical potential. First, a simple new expression for the vacancy-wind term is derived from the Qnsager phenomenological equations. This expression leads immeqiately to the exact binary analogue of the exact Nernst-Einstein equation derived recently by Murch for the unary system. The new expression is exploited first in a computer simulation of the random binary alloy. The results were in excellent agreement with Manning's expression which is exact for this alloy. Next, a computer simulation was performed for a binary alloy with order above and below the long range order/disorder temperature. In this case, the results were in astonishingly good agreement with Manning's expression substituted with computer simulated values of the component tracer diffusion coefficients. *Work performed under the auspices of the us Department of Energy.

9:30 a.m. INTERDIFFUSION AND INTERFACE MOTION IN Y/Y+8 Ni-Cr-Al DIFFUSION COUPLES: J. A. Nesbitt and R. ". Heckel, Dept. of Metallurgical Enqineerinq, Mlchlqan Technoloqical University, Houghton, '11 49931, A. E. Zyskowski, Martin Marietta Aerospace, New Orleans, LA 70189, and S. R. Levine, NASA-Lewis Research Center, Cleveland. OH 44135. Interdiffusion in y/y+B diffusion couples represen'tative of y+S Ni-Cr-Al oxidation coatings on Y Ni-base superalloy substrates was examined for Various y/Y+S couple combinations. The motion of the Y/y+S interface and the concentration at the interface were measured after annealing at l200°C for times up to 500 hours. Diffusion specimens consisting of y+S alloys of the same composition coupled to y alloys of different com-

*supported by the Department of Energy under the Grant No. DE-AC02-81ER10913 AOOl. ** Permanent address: Hughes Research Lcboratories, Mal1bu, ***CA 90265. On leave of absence from the Physics Department, Tokyo Institute of Technology.

10:30 a.m. THEORY OF MANY-BODY DIFFUSION BY THE PATH PROBABILITY METHOD: CONVERSION FROM THE ENSEMBLE AVER~~ING TO THE TIME AVERAGING. * Hiroshi Sato and Ryoichi Kikuchi,* School of Materials Engineering. Purdue University, West Lafayette, IN 47907. In order to evaluate the time correlation of the motion of a small member of particles necessary for the calculation of the correlation factor in tracer diffusion, the ensemble averaging process characteristic of the path probability method ha~ been converted to the time averaging pl"OCess. The examination of the meaning of tracer diffusion coefficient under a variety of experimental conditions in many body diffusion phenomena has thus been made. In particular, the percolation sensitivity of tracer d i ffu s ion in the time average will be discussed.

*Supported **

by NSF DMR 7903932.

Permanent address: 90265.

Hughes Research Laboratories, Malibu, CA

10:50 a.m. THE EFFECT OF JOULE HEATING IN ELECTROCHE!HCAL PERMEATION MEASUREMENTS: R.-W. Lin and H. H. Johnson, Dept. of Materials Sc i ence and Engl neeri ng, Corne 11 Un i vers i ty, Ithaca, NY 14853. Electrochemical charging, either galvanostatic or potentiostatic, is most comnonly used to produce the required input fugacity for hydrogen permeation measurements at low temperatures. Joule heating associated with the cathodic charging process, thouqh not well studied, was neglect~d in the past as of no siqnificant effect on the measured results. In the present work, a detailed study of the effect of Joule heatinn on the measured anodic current has been carried out as a function of charqing current density, cell geometry, catholyte resistance, and plastic defonnation. Specimen and catholyte temperatures were recorded during charginq. Results show that dependinq on the experimental conditions, Joule heatinq can have significant effect on the soecimen temperature and the measured anodic current.

4 TMS Fall Meeting ll,IO a.m. PERMEATION OF HYDROGEN THROUGH AN AMORPHOUS ALLOY, R.-W . Lin and H. H. Johnson, Dept. of Materials Science and Engineering, Corne 11 Un i vers ity, Ithaca, NY 14853. The solution and diffusion of hydrogen in amorphous metals is of both theoretical and practical interest. The result of a direct measurement of hydrogen transport from the gas phase through the metallic glass F40N!OPI4B6(Metglas 2826) with the electrochemical technique will be presented. Surface impedances to hydrogen entry and exit lYere eliminated by coating the specimen surfaces with a thin layer of palladium. Solubility was found to obey Sievert's 1aw over tge temp~rature range of 313 to 353 K and pressure range of 10 to 10 Pa. Analyses of the permeation transient and time lag showed that the diffusion behavior is in agreenient with Fick's diffusion theory . Within the concentration range stud i ed, the d i ffus i vi ty wa s found to be independent of input hydorqen concentration .

Reflection Synchrotron x-ray diffraction topography provides a new technique for fracture studies. This method non-destructively probed (001) EBZR Mo cleavage surfaces which had been fractured at 77°K. The "white" spectrum made possible simultaneous observation of topographs from a large number of hkl reflections which permitted g. b analysis . These had the special property of sampling the microstructure to a depth of;one extinction distance,IOJ..i; this layer is lost in etch-pit procedures. Subsequent electropolishing allowed step-profile mapping of the microplasticity and locked-in lattice rotations associated with crack opening. The image contrast in the vicinity of the original precursor crack is observed in all reflections and mapped to be;40\.l in radius. This corresponds to a dense tangle of dislocations to a depth consistent with theory.

Research supported by USOOE contract no. DE-AC02-80ER10740

9:50 a.m. OBSERVATIONS OF DEFECTS ASSOCIATED WITH CLEAVAGE IN Nb CRYSTALS USING X-RAY TOPOGRAPHY: A.B. Hmelo and J .C . Bilello, Department

FRACTU~E

I: Analytical Techniques TMS-AIME General Abstract Session

Monday, October 25, 1982 Cervantes Convention Center 8:30 a.m. Room 267 Session Chairman: Dr. S. R. Shatynski, Department of Materials Engineering, Rensselaer Polytechnic Institute, Troy, NY 12181.

8:30 a.m. "X-RAY TOPOGRAPHY OF FRACTURED SURFACES 1/1 REFRACTORY METALS BY A COMPUTERIZED DIFFRACTION FACILITY". John tt. Liu. Fu-Rong Chen, A. Hmelo and J.C.Bilello, Department of Materials Science & Engineering, State University of New York at Stony Brook. Stony Brook, N.Y. 11794 A study of the inhomogenous plastic strain that remains on fracture surfaces in refractory metals has been performed by x-ray topography. In the past. the setting up of such experiments was time consuming. since manual adjustment of the orientation of the specimen was required . A computerized system will be described that allows automatic, systematic searching of di ffraction peaks, as well as the measurement of the width of reflection curves. The mathematical theory governing the operation of the experiments will be illustrated by means of stereographic projections. The use of an x-ray image intensifying device that aids in the aligning and displaying of topographs will be described. Work supported by U. S. Department of ~ner9Y through Grant DE-AC02-80ER10740.

8:50 a.m. FIlACTURE PREDICTION OF FATIGUED ALUMINUM ALLOYS BY A COHPUTERAIDED X-RAY DIFFRACTION METIlOD, W. Mayo and S. Weissmann, Dept. of Mechanics and Materials Science, Rutgers University, Piscataway, N.J. 08854 The accrued microscopic damage resulting from low cycle fatigue of AA2024-T4 was determined by a novel compu ter-aided double crystal diffractometer method (CARCA). The results show a linear dependence of the X-ray rocking curve halfwidth with the fraction of fatigue life. Moreover, this effect was found to be independent of strain level or loading sequence. The information thus obtained was used as a new criterion for predicting the onset of fatigue fracture. To critically test the validity of the method, samples were subjected to a wide range of spectral loading histories. Even under such complex test conditions, the nondestructive prediction of fracture was accurate to within ± 10 percent.

9:10 a.m. X-RAY DETERMINATION OF STRAIN CONCENTRATION FACTORS AROUND HOLES AND INCLUSIONS: S. Weissmann, Z. H. Kalman, J. Chaudhuri and G. J. Weng, Dept. of Mechanics and Materials Science, College of Engineering, Rutgers University, Piscataway, N.J. 08854 Long-range. elastic strains emanating from stress raisers, such as holes and inclusions were investigated in bent crystals of silicon which functioned as a model material. The strains and strain interactions were characterized by X-ray pendellosung fringe topography and were quantitatively evaluated by X-ray intensity measurements of traverse-oscillation topographs. The strain gradient emanating from a bent specimen containing a hole was experimentally determined and the results were compared to calculations based on continuum mechanics having closed form solution. Very good agreement between experiment and theor y was obtained . The dependence of strain interaction on interflaw distance was experimentally demonstrated for specimens containing two holes.

9:30 a.m. WHITE BEAM SYNCHROTRON FRACTOGRAPHY OF (001) ORIENTED MOLYBDENUM CRYSTALS: A.B. Hmelo, Dept. Mat. Sci. & Eng . , SUNY. Stony Brook, NY. S.T. Davies. Dept. Eng., Univ. of Warwick, Coventry, UK, D.K . Bowen, Uoiv. Warwick and Daresbury Laboratory, Warrington, UK~ and J. C. Bilello, SUNY Stony Brook

of Materials Science, State University of New York at Stony Brook,

Stony Brook, N.Y. 11794 The morphology of cleavage for the (001) (100) and (001) (110) crack propagation systems was studied for Nb electron beam zonerefined crystals. Comparisons were made between single pass specimens and samples which were further ul trahigh vacuum out-gassed to total interstitial gas contents <1 ppm . Cleavage was initiated at 77°K using a spark discharge method which produced precursor cracks with crack-tips radii estimated in the few J..i m range . Defect observations were made using Berg-Barrett topography with Co ka radiation. Topographs were taken for various reflections as a function of depth below the cleavage surface. Dislocation bands of predominant screw orientatlontgross twins and microtwins of (112) (llI)character were observed. The energetics of cleavage for the two crack systems are compared relative to the associated defects generated during crack propagation. Surfaces energies of the order of 2000 ergs/cm 2 were obtained. Differences in morphology are rationalized in terms of the crack tip stress field in each case.

10:lQ a.m. SMALL-ANGLE NEUTRON SCATTERING STUDY OF CREEP CAVITATION IN 304 STAINLESS STEEL:" M. H. Yoo, J. C. Ogle, J. H. Schneibel and R. W. Swindeman, Metals and Ceramics Division , Oak National Laboratory, Oak Ridge, TN 37830

Rid~c

A small-angle neutron scattering (SANS) technique has been used to investigate the formation of creep cavities in Type 304 stainless steel. SANS measurements from the primary, secondary, and tertiary creep stages indicate that the evolution of M23 C6 occurs mostly in the primary stage. No magnetic scattering contribution from the carbides was observed. An analysis of the size distribution of grain boundary cavities was made by repeating SANS measurements after post-creep solution heat treatment. The formation of grain boundary cavities during the transient primary stage of creep deformation is discussed in view of the SANS data and TEM obse rva tions.

*Research sponsored by the Division of Material s Sciences, U.S. Department of Energy under contract W-7405-eng-26 with the Union Carbide Corporation.

10:30 a.m. GEL ELECTRODE IMAGING OF METAL FATIGUE: !I!. CRACKS IN ALUMINUM ALLOYS: W. J. Baxter, Physics Department, General Motors Research Laboratories, Warren, MI 48090-9055 A reliable method of detecting and measuring very small fatigue cracks in metals could reduce the time required for fatigue testing. Previous reports described a Gel Electrode technique recently devised for detecting and imaging fatigue cracks in aluminum tested in simple bending. In this study, this technique is shown to be applicable to testing in both bending and torsion and to high strength aluminum alloys 7075-T6. 2024-T3 and 2024-T4. Images of fatigue cracks are formed in a surface film on a gel without removing the test piece from the fatigue machine. The imaging conditions were selected to provide good spatial resolution at the expense of sensitivity. Nevertheless, fatigue cracks as short as 10 ~m in length are consistently detected and located. The images in the gel may be viewed with a pocket magnifier or an optical microscope. The now of charge during image formation under standard conditions provides a quantitative measure of crack length, which is independent of alloy composition. A crack 100 ~ m long can be reliably detected by charge now measurement; thus, this approach is not as

sensitive

8S

the information contained in the actual images.

10:50 a.m. DAMPING OF ACOUSTIC WAVES - IN METALS: R. A. Masumura, C. L. VoId, and B. B. Rath. Naval Resear c h Laboratory-:--Washington, DC 20375 and R. Plunkett. Department of Aerospace Engineering and Mechanics, University of Minnesota, Minneapolis, Minnesota The . damping coefficient, whi c h is a measure of the intrinsic acousti c energy dissipation within a solid, has been determined for several alloys systems fabricated by either conve ntional wrought and/or PM processes. The test method and equipment uses the resonant vibrating cantilever beam system. The energy losses

TMS Fall Meeting 5 are found to be dependent on local interactions between the acoustic wave and microscopic defects. An evaluation of the effects of metallurgical variables on the loss factor suggest that damping characteristics can be altered by more than two orders of magnitude by judiciously selecting the composition and thermo mechanical processing. Measurements of the loss factors for the RSP based high damping alloys suggest possible improvement over normal fabrication methods.

H20 solution does not appear to affect the rates of near-threshold crack propagation. Increasing the hydrazine level from 30 to 10 7 ppb in the same salt solution enhances the resistance to crack growth while redUCing the percentage of intergranular fracture to nearly zero.

9:10 a.m. CORROSIVE FATIGUE CRACK PROPAGATION TESTING WITH THE KRAK-GAGE IN

11:10 a.m. FAT'IGUE CRACK PROPAGATION STUDY BY ACOUSTIC EMISSION METHOD:'

C. J. Kim, G. Lil and J. Weertman, Dept. vf Materials Science and Engineering, Northwestern University, Evanston, IL 60201 Acoustic emission technique has been used to determine the apparent crack extension position in a stress cycle during fatigue. Tests were conducted for different materials. The results of these tests were used to check the validity of a recent theory of ours. The theory is based on crack tip elastic enclave model and it predicts that when the power exponent n of the Paris fatigue crack propagation equation, da/dn c c(6K)n, is approximately equal to 4 crack extension starts when the cyclic stress has almost reached its maximum value in a stress cycle. If n=2, crack extension starts when the stress is well below its maximum value. Preliminary results appear to show good agreement with the theory for n-4 type materials. Tests for nz:2 type materials will be done and results will be presented. *Research supported under the NSF-MRL program grant f1DMR79-23573 through the Materials Research Center of Northwestern University.

The application of a relatively new, thin-film ~ondable transducer, commerc ially available under the name KRAK-GAGE , was evaluated for corrosion fatigue crack propagation tests in a sea water (3.5% NaCI) environment at ambient temperature on an HY80 steel. Fatigue crack growth data generated by this method are shown to be consistent with those obtained by the compliance method and the commonly used optical/visual measurements. Thus, this new instrumentation system appears to be a valuable addition for corrosion fatigue crack ' growth testing. Furthermore, test data acquisition and analysis of the KRAK-GAGE and the compl iance method were computer automated, which results in considerable cost savings for such customarily labor intensive fatigue crack propagation experiments.

•Reg.

U.S. Trademark, Hartrun Corporation.

9:30 a.m. EFFECT OF ENVIRONMENT ON FATIGUE CRACK PROPAGATION BEHAVIOR IN TYPE 316 STAINLESS STEEL AT ELEVATED TEMPERATURES: H. H. Smith

11:30 a.m. ELECTRON CHANNELLING FROM FRACTURE SURFACES OF TI-30MO.

Kenneth

A. Peterson and William W. Gerberich. Department of Chemical Engineering and Materials Science, University of Minnesota, MinneapOlis, MN

SALT WATER: Peter K. Liaw, Westinghouse R&D Center, Pittsburgh, PA 15235, H. R. Hartmann. Hartrun Corporation, Chaska, MN 5531B,and E. J. Helm, Westinghouse R&D Center, Pittsburgh, PA 15235.

55455.

Electron channelling has been used to determine the orientations of fracture facets and the extent of damage directly from slow crack growth and fatigue fracture surfaces of Ti-30Mo. Ion milling has been used to profile for strains as a function of depth away from the crack surface . This is in conjunction with understanding deformation gradients at fracture surfaces, produced by different growth mechanisms. Possible contributing factors such as effect of microfracture times on effective stress intensity and effects of hydrogen on slip character are discussed.

FRACTURE II: Environmental Effects TMS-AIME General Abstract Session

and D. J.

Michel~

Naval Research

Laboratory~

Washington, D.C.

The influence of test environment on fatigue crack propagation rates in type 316 stainless steel was studied at 593 and 649·C in air and in vacuum. The crack propagation rate was found to be lower in vacuum than in air and the increase in propagation rate with increase in stress intensity factor was larger in vacuum than in air. The slope of the experimental relationship between da/dN and 4K was determined to be approximately 2 in air and from 4 to 6 in vacuum. Instantaneous crack closure measurements were made using potential drop techniques during crack propagation as the partial pressure of air in the test chamber was varied. The relative sizes of the plastic zones developed during fatigue were determined using hardness and thermal etching methods. The results show that the differences in fatigue behavior in air and and in vacuum were consistent with crack closure phenomenon observed for other alloys at low temperatures .

9:50 a.m.

Cervantes Convention Center Monday. October 25.1982 8:30 a.m. Room 270 Susion Chairman: Dr. Joseph M. Wells, Westinghouse R&D Center, Pittsburgh, PA 15235.

8:30 a.m. COMPUTER-CONTROL FATIGUE CRACK GROWTH RATE TESTING ON BEND BARS IN CORROSIVE ENVIRONMENT: T. R. Fabis, P . K. Liaw, Westinghouse

R&D Center, Pittsburgh, PA 15235 and T. R. Leax, Westinghouse Steam Turbine Generator Div., East Pittsburgh. PA 15112 Automated fatigue crack growth experiments were conducted on a 304 stainless steel in air and wet hydrogen at 652 KPa. The tests were performed using three point bend specimens in the frequency range from 0.1 to 10 Hz. An electrohydraulic fatigue machine was interfaced with a DEC PDP 11/34 computer to develop the fatigue crack growth rate properties. The computer can be simultaneously interfaced with several electrohydraulic machines to handle a large amount of crack growth experiments. The fatigue crack growth rate tests can be performed unattended and all of the data acquisition and analysis are automated. Thus, costly labor associated with fatigue crack growth testing can be reduced considerably. The constant 6K (stress intensity range) tests and the decreasing (or increaSing) 6K tests give consistent crack growth rate data. In wet hydrogen, increasing frequency decreases the fatigue c rack propagati,on rates.

8:50 a.m.

STRESS CORROSION CRACKING OF ADMIRALTY BRASS IN AN ACID SULFATE

ENVIRONMENT: EFFECTS OF STATE OF STRESS: W. Keith Blanchard, Donald A. Koss and Lloyd A. Heldt, Dept. of Metallurgical Engineering, Michigan Technological Universi ty, Houghton, HI 49931. Stress corrosion cracking studies have been perfonned on commercial purity Admiralty brass sheet in .1M CuS04 aqueous solution at open ci rcui t corrosion potential. Techniques used in sheet metal plasticity studies have been employed to contrast the SCC susceptibility in uniaxial tension with that of both plane strain and balanced biaxial tension. The susceptibility to SCC increases as the stress state changes from uniaxial toward biaxial tension. These results are examined in terms of the relationships between loading path, stress state. and crack morphology. This research has been supported by the Department of Energy.

10:10 a.m. EMBRI'M'LEMEm OF a-B BRASS IN AQUEOUS SOLUTIONS OF SOOIUM SULFATE,

Michael B. Hintz, Linda J. Nettleton, and Lloyd A. Heldt, Dept . of Metallurgical Engineering, Michigan Technological Uni versi ty, Houghton, HI 49931. Slow strain rate tensile tests of Ct.-B brass (42 w/o Zn) were performed in 1 N Na2S04 solutions under both open circuit and potentiostatic conditions. For open circuit testing, the electrochemical potential was measured as affected by straining. Embrittlement occurs over a specific range of pH values; area reduction at fracture decreases by as much as 70\ and a fracture mode transition from ductile to mixed transgranular/intergranular is observed. The results of the electrochemical measurements are correlated with the susceptibility to ernbrittlement.

INFLUENCE OF CORROSIVE ENVIRONMENTS ON NEAR-THRESHOLD FATIGUE CRACK GROWTH IN 403 STAINLESS STEEL: Peter K. Liaw, Westinghouse

R&D Center, Pittsburgh, PA 15235, J. Anello, \.Jestinghouse Lester Plant, Philadelphia, PA 19113, J. K. Donald, Del Research Division,

10:30 a.m.

Hellertown, PA 18055.

The near-threshold fatigue crack growth behavior of 403 stainless steel has been investigated in low 02 steam «1 ppm), high 0;> steam (40 ppm), and boiling water with NaCl (Na2S0,).

High 0,

steam slightly increases crack propagation rates in the threshold region, relative to low 02 steam. Values of threshold stress intensity range, 6K t h, slightly increase with an increase in the concentration of NaCl in the solution. Varying pH, within the ran~e

5.0 to lO.O, in a 0.1 gm NaCl plus 1.0 gm Na,SO, per 100 ml

EFFECTS OF COMPRESSIVE RESIDUAL STRESS ON CRACK OPENING BEHAVIOR IN HY-130 STEEL:' G. R. Leverant and J. E. Hack, Southwest Re-

search Institute, 6220 Culebra Rd., San Antonio, TX

78284.

The surface crack opening displacement behavior of fatigue microcracks in HY-130 steel has been compared on specimens which have no residual stresses present and ones which have a residual stress gradient in the near-surface region. Residual stresses were introduced by shot-peening. It was found that the zone of

6 TMS Fall Meeting compressive residual stress induced by the peening acted to reduce the effective ill<. on the microcracks and delay or preclude crack opening during the loading cycle. This result will be compared to similar data gathered on Ti-6Al-4V. Attempts at pre die ting the opening behavior under the influence of the residual stress gradient will also be reviewed.

The surface dislocation analysis of three dimensional distortions developed earlier for a point force will now be applied to a curved dislocation situated near a planar interface in a twophase medium.

8:50 a.m. *Work sponsored by the Office of Naval Research under Contract N00014-78-C-0674.

10:50 a.m. OXIDE WEDGING OF SURFACE MICROCRACKS IN 1IY-130 STEEL IN A SALT WATER ENVIRONMENT:* J. E. Hack and G. R. Leverant, Southwest

Research Institute, 6220 Culebra Rd., San Antonio, TX

78284.

The effect of an aqueous 3.5% NaCl solution on the surface crack opening displacement behavior of small part-through fatigue cracks in HY-130 has been studied. Measurements were made directly using a fatigue stage mounted in-situ in an SEM. It was found that oxide wedging of the cracks occurred during cycling at 1 Hz. The action of the oxide was to reduce the effective.6K rang~ on the cracks which delayed crack opening during the loading cycle. Since the size of the cracks studied restricts the maximum possible /).K to the near-threshold region, the oxideinduced reduction in /).K can arrest crack growth. The similarity of the present results to data on macrocracks and the synergistic effects of residual stress states on effective /).K will also be discussed.

* Work sponsored by the Office of Naval Research under Contract N00014-78-C-0674.

11:10 a.m. INFLUENCE OF AGING TREATMENT, LOAD RATIO AND ENvIRONMENT ON NEARTHRESHOLD FATIGUE CRACK GROWTH BEHAVIOR IN ALUMINUM ALLOYS: P.E. Bretz and A.K. Vasudevan, ALCOA, Alcoa Center, PA15069 and S. Suresh, Dept. of Materials Science and Mineral Engineering, University of California, Berkeley, California 94720.

The effects of aging treatment, load ratio and environment on the near-threshold fatigue crack growth behavior of 7075 aluminum alloy are examined in detail. The threshold stress intenSity factor range values are found to decrease with increasing aging time, load ratio and moisture content in the surrounding medium. The results suggest that environmentally-influenced near-thre-" shold crack growth in aluminum alloys is controlled by two concurrent and mutually-competitive mechanistic processes: a dominant role of crack closure (which tends to increase the threshold stress intensity factor range) and a strong embrittling effect of the moist environment (which sigIlificantly decreases the threshold stress intensity factor range).

11:30 a.m. STRESS-CORROSION-CRACKING SUSCEPTIBILITY OF ALUMINUM IN CONTROLLED ATMOSPHERE: * M. Khobaib, Systems Research Laboratories, Inc., Dayton, OR 45440; C. T. Lynch, Office of Naval Research, Arlington, VA 22217.

The stress-corrosion-cracking susceptibility of Al 7075-T6 has been studied in a controlled-atmospheric chamber using the slowstrain-rate technique. The effects of realistic atmospheres such as high-humidity air with small additions of sulfur dioxide and surface salt have been investigated. The design of the slowstrain-rate apparatus as well as the environmental chamber for controlled atmospheres will be discussed and initial experimental results presented. Most tests were conducted at 80-90% relative humidity in sulfur-dioxide concentrations of 100 and 1000 ppm. Differences in stress-corrosion susceptibility were greatest fox high-humidity, environments containing small amounts of surface salt in the presence of sulfur dioxide. The qualitiative estimates of embrittlement in the fracture surfaces were consistent with the slow-strain-rate results. These data provide the basis for" development of accelerated-corrosion tests in realistic atmospheres. *Research supported in part by USAF Contract F3361S-79-C-Sl09.

GRAIN BOUNDARY PHENOMENA TMS-AIME General Abstract Session Monday, October 25, 1982 Cervantes Convention Center 8:30 a.m. Room 264 Session Chairman: Dr. W. A. T. Clark, The Ohio State University, Dept. of Metallurgical Engineering, 116 W. 19th Ave., Columbus, OH 43210.

8:30 a.m. ANALYSIS OF ELASTIC FIELD ASSOCIATED WITH A CURVED DISLOCATION IN A TWO-PHASE MEDIUM WITH A PLANAR INTERFACE: K. Jagannadham, Engineering Science and Mechanics, Tennessee Technological University, Cookeville, TN 38501

SURFACE DISLOCATION DESCRIPTION OF THREE DIMENSIONAL DISTORTIONS IN TWO-PHASE SYSTEMS--A POINT FORCE NORMAL TO A PLANAR INTERFACE: K. Jagannadham, Engineering Science and Mechanics, Tennessee Technological University, Cookeville, TN 38501 The surface dislocation description of a three dimensional distortion in a two-phase medium with a planar interface is formulated. In particular, the continuity conditions across the planar interface are satisfied by a continuous distribution of dislocations when a point force nannal to the interface is situated in one of the phases. Furthermore, the dislocation distribution functions are derived and the traction component at the interface obtained. The availability of an alternate description of distortion interms of surface dislocations at the interface is emphasized so that it can be employed advantageously to replace the conventional elasticity methods.

9:10 a.m. CYCLIC GRAIN BOUNDARY MIGRATION DURING HIGH TEMPERATURE FATIGUE OF ALUMlNUM:* Parviz Yavari, Northrop Corporation, Aircraft Group - Manufacturing Technology, Hawthorne, California 90250, and Terence G. Langdon, Department of Materials Science, University of Southern California, Los Angeles, California 90007.

Grain boundary migration occurs in a cyclic manner when polycrystalline aluminum is tested in fatigue at high temperatures and low frequencies. This paper describes the general characteristics of the cyclic migration, and examines the influence of strain amplitude and frequency. By taking measurements of the average migration distance, it is possible to derive a relationship which describes migration in terms of frequency, strain amplitude, number of cycles and temperature. *This work was supported by the United States Department of Energy under Contract DE-AM03-76SFOO1l3 PA-DE-AT03-76ERI0408.

9:30 a.m. MIGRATION-ASSISTED DIFFUSIONAL CREEP BY BOUNDARY DIFFUSION: I-W. Chen, Department of Nuclear Engineering and Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139

A new mechanism of migration-assisted long-range diffusion due to non-hydrostatic stresses is proposed. Migration of grainboundaries under stresses may enhance the diffusional creep rate by orders of magnitude, switch the rate-controlling mechanism from boundary diffusion of the slowest species to that of the fastest species. Alloyed and dealloyed zones are left in its wake. The phenomenon predicted here has close resemblance to DIGM, and is potentially important at low temperature, stressed conditions which are relevant for several applications.

9:50 a.m. THE A~COM~'ODATION OF DEFORMATION AT GRAIN BOUNDARIES IN MOLYBDENUM: E. Kamenetzky and R. Gronsky, lawrence Berkeley laboratory and the Department of Materials Science and Mineral Engineering, University of California, Berkeley, CA 94720 The absorption of dislocations in near-coincidence boundaries and the accorrrnodation of a change in grain boundary (GB) plane in a low angle boundary were studied in TEM. Incoming dislocations di ssoci ate into di sp 1acement-sbift-comp1 ete 1atti ce di s 1ocati ons (DSCl) in a near L = 41 boundary. Trapped lattice dislocations and dissociated trapped lattice dislocations are also observed in this boundary. Three mechanisms of accommodation of a change in GB plane were studied in different sections of a [101] low angle twist boundary. A GB step shows no change in GB dislocation structure; a GB facet produces a change in dislocation line directions and a GB curvature producing an overall change in GB plane is accommodated by the emission of dislocations onto a favorable plane. These results are analyzed in terms of 0-lattice theory. This work is supported under DOE Contract # DE-AC03-76SF00098.

10:10 a.m. CARBON SEGREGATION AT AUSTENITE GRAIN BOUNDARIES IN Fe-C ALLOYS: Y. W. Lee, Revere Research, Edison, NJ 08817, J. R. Bradley, General Motors Research Laboratories, Warren, MI 48090, W. C. Johnson, Perkin-Elmer, Physical Electronics, Eden Prairie, MN 55344 and H. I. Aaronson, Carnegie-Mellon University, Pittsburgh, PA 15213.

The carbon concentration at former austenite grain boundaries in high-purity Fe-C alloys was determined as a function of austenitizing temperature and carbon content with a scanning Auger microprobe. This concentration was found to be weakly "dependent upon temperature, evidently because of both a small binding energy and the rapid diffusion of carbon in austenite during quenching. The

TMS Fall Meeting 7 data were analyzed with an adaptation of McLean's equation to interstitial solid solutions. Two different approaches to analysis of the data were employed. In one it was assumed that the data are those which actually obtained at temperature. This analysis yields the upper bound to the binding eneTgy of carbon to austenite grain boundaries . A second approach takes account of changes in the carbon concentration at grain boundaries during quenching to room temperature. This method gives a clearer indication of the range in which the binding energy probably lies. This r~search was sponsored in part by the NSF Division of Materials Research and the Army Research Office.

10:30 a.m. QUANTITATIVE ANALYSIS OF TEXTURE IN CUBIC FILMS: Satish bo, C. R. Houska t Depart;ment of Materials Engineering, VPI & SU, 201 Holden Hall, Blacksburg, VA 24061. The method of Roe and Krigbaum for determining fiber texture has been extended to thin film applications. It is often desirable to make in situ pole density tqeasurem.ents of a film on a thick substrate which does not permit data to be collected by x-ray tranS1Di~sion techniques. If the data are restricted to the range o < X < 75~, obtained by reflection, incomplete pole density plots are obtained t and it is necessary to devise a self consistent extrapolation techriique that extends the pole density data td 90 0 • This requires least squares fitting over the range from x - 0 to 75 0 and an iterative procedure for extrapolating with functions consistent with a single orientation function. The example of a 1.14 micron Mo film on a (111) SI substrate requires an expansion of the symmetry relations to order 46. Simplified methods may be obtained to describe orientation with data obtained from a conventional x-ray diffractometer. These are of value in quantitative volume fraction analyses.

covery and recrystallization is discussed . The effect of the use of glass and graphite is also analysed, and a comparison is made between the rates of static softening determined in this work and the dynamic ones obtained in earlier investigations.

GRAIN REFINEMENT IN CASTINGS AND WELDS: THEORY AND PRACTICE I: Non-Ferrous Alloy Castings Sponsored by the TMS Solidification Committee Monday, October 25, 1982 Cervantes Convention Center 8:30 a.m. Room 261 Session Chairman: G. J. Abbaschian, University of Florida, Depart· ment of Materials Science and Engineering, Gainesville, Florida 32611.

8:30 a.m. GRAIN REFINEMENT IN THE CASTING ON NON FERROUS ALLOYS - L. F. Mondolfo, Rensselaer PolytechniC lnst., Troy, N. Y., 1218-1-Substantial grain refinement in castings is obtained by enhancing nucleation so that a large numer of crystals is formed, that soon impinge on each other and prevent further growth. This enhancement comes from heterogeneous nucleation produced by nucleants either present in the melt or intentionally added. After a brief review of the present knowledge on the nature of nucleants, the techniques available for grain refinement are discussed under the headings: rapid cooling, dynamic methods, growth hindering addi tions, nucleating addi tions. denucleation .

10:50 a.m.

9:00 a.m.

DEVELOPMENT OF CRYSTALLOGRAPHIC TEXTURE IN OXIDE DISPERSION STRENGTHENED ALLOYS" M. P. Anderson. J. Y. Koo, and R. Petkovi c-Luton, Exxon Research and Engi neeri ng Company, P.O. Box 45, Linden, New Jersey 07036

GRAIN VOLUME DISTRIBUTION AS A FUNCTION OF THE FREEZING RATE IN AN ALUMINUM-ZINC ALLOY, F . N. Rhines, P. J. Laskey and H. H. Ho, University of Florida, Gainesville, Florida 32611

Microstructure and preferred orientation developed in oxide di spers i on st rengthened alloys have been exami ned as a funct ion of the alloy crystal system. The alloys studied were Fe-20Cr4.5Al-0.5Ti-0.5Y Z03 and Ni-16Cr-5Al-1Y203' The Fe - base alloy is characterized" by a deformation type texture consisting of {1l2} <110> and {114} <110> in the warm worked state. In contrast, the Ni-base alloy is characterized by a recrystallization type texture in the warm worked state. This preferred orientation is best described as a combination of a [001] fiber texture and a {100} <001> rolling texture. The origin of the differences in crysta 11 ographi c texture is di scussed in terms of the preferred deformation modes in the two alloys and the recrystallization constraint imposed by the oxide dispersoids.

The grain volume distribution in castings of an aluminum 10% zinc alloy was determined by separating the grains with gallium and measuring their volumes by weighing them individually. The distributi~n is log-normal at all rates of freezing and the standard deviation of the distribtution decreases with increasing freezing rate. At all rates of freezing, about 20% of the grains (by number) weighed less than lxlO-~ grams, even when the average graif) weight was as large as 0.22 grams. This is understood to mean that many new grains are nucleated in the interde,:dr~tic liquid late in freeZing. It correlates also with the fact that the elongation of cast metals is relatively low, it having been shown in the case of recrystallize d aluminum that the elongation is smaller the broader the distribution of grain volumes.

9:30 a.m. 11:10 a.m. RECRYSTALLIZATION IN AN AUSTENITIC STAINLESS STEEL WELD THROUGH POST-WELD THERMO-MECHANICAL TREATMENT, R. M. Nekkanti, J. Foulds, and John Moteff, Materials Science and Metallurgical Engineering Department, University of Cincinnati. Cincinnati, OH 45221. The possibility of weld metal recrystallization through small amounts of cold work and annealing is investigated for an austenitic stainless steel weldment. A weldment section comprising 316 stainless steel base metal and GTA welded 16-8-2 weld metal obtained from large diameter thin wall formed-and welded pipe was used for the study. The as-solidified stable weld metal dislocation substructure enhances the susceptibility of the weld metal region to recrystallization and an ensuing grain refinement . Cold-rolling of the section \,,1 th subsequent thermal treatment both below and above the Y-SOlvu5 effects the removal of the original weld substructure with acc ompanying recrystallization and little change in base metal microstruc ture . Importantl y , this investigation opens the possibility of achieving an improved weld-base metal compatibility through post-weld th e rmo-mechanical treatment. The authors are thankfuZ to Dr. V. K. Sikka of the

Oak Ridge Nationa Z Laboratory f or pr OViding the above specimens.

11:30 a.m. EFFECT OF Nb, V and Mo ON STATIC RECOVERY AND RECRYSTALLIZATION IN MICROALLOYED STEELS: H.L. Andrad e , M.G. Akben and J . J. Jonas, Department of ~etallurgical Engineering, McGill University, Montreal, Canada H3A 2A7. The effect of Nb, V and Mo on th e occurrence of static recovery and recrystalliz a tion after high t e mperature deformation was investigated in a s eries of microalloy e d steels. The steels had a base composition of 0 . 05% C and 1.40% !in. To this, single additions of 0.035% Nb, 0.115% V or 0.30% Mo were made. Interrupted hot compression t e sts were performed at 900 and lOOO°C and at a constant true strain rate of 2 sec -I. The load-fre e time was decreased from 5000 5 to 50 ms and the degree of static softening during this period was determined. Both graphi te and g lass were used as lubricant s . Percent softening v s . delay time curve s are present e d and the retarding effe c t of Nb, V and :-10 addition on the rate of static re-

POSSIBLE GRAIN REFINING MECHANISMS IN ALUMINUM, AS A RESULT OF ADDITION OF MASTER ALLOYS OF THE Al-Ti-B Type, L. Arnberg, L. Backerud and H. Klang, University of Stockholm, Stockholm, ~eden In recent publications the authors have described a series of experiments where the grain refining ability of master alloys produced under different condi tions are reported. The present paper is a conclus ion of the earlier work and an extension of theories of nucleating mechanisms, expecially into the very details of the peritectic reaction and the early stages of growth of a-aluminum on Al)Ti particles. These processes have been followed by thermal analysis during the solidification process and the result gives support to the theory of nucleation on Al)Ti particles surviving from the master alloy, when such alloy conditions are used to achieve grain refinement. Under special casting conditions other grain refining mechanisms may operate and an account will also be gi ven to such possibili ties.

10:00 a.m. THE CAPABILITIES AND LIMITATIONS OF AL - 6% TI GRAIN REFINERS *B. L. Tuttle, **R. Sturdevant, and ***G. M. Farrior, *General Motors Institute, Flint, HI, **Hilward Alloys, Inc., Lockport, NY, ***I. T. T., St. Louis, MO GraIn refinement during casting requires control of three phenomena: (1) c onstitution control , (2) nucleation control, and (3) grOW'th restriction control. In aluminum alloy castings the complex Al-Ti-B master alloys have traditionally been utili~ed to control all thre phenomena . In truth, the grain refining master alloy is only required to establish the nucleation control. Therefore in many applications an A.£.-6% Ti which does not contain boride particles can provide a more cost effective grain refining treatment. To better understand the capabilities and the limitations of Al-Ti grain refiners in commercial aluminum alloys, grain refining tests were performed in binary aluminum alloys containing the major alloYlng elements - Mg, Si, Cu, and Mo, and in several commercial alloys. The results indicate that in several alloy systems the Al.-Ti grain refiner produces equivalent results to the more expensive Al-Ti-B grain refiners.

8 TMS Fall Meeting 10:30 a.m. SIZE EFFECTS IN GRAIN REFINEMENT OF ALUMINUM ALLOYS: D. S. Lee

and current actual costs for HIP installations of various sizes will be compared. Finally, the possible effect on cost of future process and equi pment improvements wi 11 be di scussed.

and H. Basaran. Revere Research, Inc •• P.O. Box 1352, Edison, NJ

08837

10:00 a.m. The grain refinement performance of commercial Al-5% Ti-l%B ma8te~ alloys has been investigated. The alloys differing 1n particle size and distribution of TiB2 particles have shown a marked difference in the refinement of Al alloy ingots, espeCially in the presence of convection. Alloys containing the particle size of a few micron ranges tend to result 1n poor refinement due to suspended particles under the influence of convection. This suggests that in commercial practice, the choice of grain refiners is of particular importance.

11:00 a.m. GRAIN REFINEMENT OF ALUMINUM ALLOY INGOT CASTINGS BY MEANS OF MELT AGITATION DURING SOLIDIFICATION USING NAFHTALINE, S. E. Kisakurek, Bogazic! University, Bebek, Instanbul, TURKE-Y--

A research work has been performed to demonstrate the possibility of grain refining of laboratory scale ingot castings of binary aluminum alloys poured into chill molds containing naphtaline. The results have revealed a marked effect of naphtaline in refining the as-cast macrostructure. The method of introduction of naphtali!1e into the castings has been noted to be one of the major parameters in the process. Other parameters of importance found were the superheat of the melt above the equilibrS-um liquidus temperature and the alloy content. The nature of the solute addition has also been measured as an effective variable. It is suggested that naphtaline, which sublimes at 80°C, stimulates a crystals multiplication mechanism that resul~s from the disruption of the solid skin that forms at the initial stages of the process of solidification.

HOT ISOSTATIC PRESSING OF METAL AND CERAMIC POWDERS I Sponsored by tile TMS Powder Metallurgy Committee Monday, October 25, 1982 Cervantes Convention Center 8:30 a.m. Room 265 Session Chairman: A. H. Clauer, Battelle Columbus Laboratories, Columbus, Ohio 43201.

8:30 a.m. HOT ISOSTATIC PRESSING EQUIPMENT DEVELOPMENT: F. X. Zilll1lennan, Autoclave Engineers, 2930 West 22nd Street, Erie, Pennsylvania 16512 This paper reviews the various types of Hot Isostatic Press (HIP) equipment and its development as related to HIP applications for processing metal and ceramic materials. A review of major HIP system components such as pressure-vessel, furnace, gas handling, electrical control, and auxil iary systems are discussed for both research and production use. The successful application of graphite furnaces rated for 200°C operation at 200 MPA is also described. Concludin9 remarks will show how HIP productivity has improved in the last several years.

CERAMIC MOLD PROCESS FOR POWDER METALLURGY NEAR-NET SHAPES: V. K. Chandhok and W. L. McCollough, Crucible Inc., Research Center, P. O. Box 88, Parkway West and Route 60, Pittsburgh, Pennsylvania 15230 Fully dense Powder Metallurgy products are being considered for a wide variety of applications. A ceramic mold process, using rapidly sol idified prealloyed powders and hot-isostatic-pressing. has been developed for the manufacture of near-net shapes. The process will be described and its application to superalloy and titanium alloy aerospace components will be reviewed. Also, specialized application of the process to dual alloy structures, orthopedic prosthesis and oil country products will be discussed.

10:30 a.m. HOT ISOSTATIC PRESSING OF CERAMIC MATERIALS: R. R. Wills, L. G. McCoy, and M. C. Brockway, Ceramics and Materials Processing Section, Battelle Columbus Laboratories, 505 King Avenue, Columbus, Ohio 43201.

The application of Hot Isostatic Processing (HIP) to ceramic materials is beginning to increase rapidly. The methods of HIP processing of ceramics are discussed together with recent and past work. The potential for improved material properties is !=:onfirmed for both oxide and non oxide ceramics.

HYDROGEN IN SOLUTION IN METALS AND ALLOYS I Invited Papers Sponsored by TMS Alloy Phases Committee Monday. October 25, 1982 Cervantes Convention Center 8:30 a.m. Room 273 Session Chairman: T. J. Rowland, Materials Research Laboratory and Metallurgy Department, University of Illinois, lJrbana, IL 61801.

8:30 a.m. THE EFFECT OF IMPURITIES AND ALLOYING ADDITIONS ON HYDROGEN IN METALS:* W. D. Wilson and H. S. Daw, Theoretical Division, Sandia National Laboratories, LIvermore, CA 94550

First principles calculations of the binding of hydrogen to a number of impurities and alloying additions in f.c.c. and b.c.c. metals will be presented. The calculations allow electronic rearrangement of both the impur tty and hydrogen atom within an effective medium (local density) framework. Lattice relaxations are included for nearly a thousand atoms surrounding the impurity hydrogen complex and the energy of the system determined as a function of the separation between the hydrogen atom and the defect. Trapping and lattice location of H in th~ viCinity of impurity elements such as C. 0, and S In f .C.c. Ni and alloying eleaents such as Ti and Zr in b.c.c. V and Nb will be studied.

9:00 a.m. *This work was supported by the U. S. Dept. of Energy.

HIP FROM A PROCESSORS POINT OF VIEW: Robert Widmer, Industrial Materials Technology, Inc., P. O. Box 565, A~dover, Massachusetts 01810 Customer proceSSing of materials by Hot Isostatic Pressing requires versatility, flexibility, and strict quality control. With costly customers parts in-house for HIP, the risks of over heating, distortion, and contamination must be elimated by complete and accurate control of the process. Cases where these problems have been successfully handled are sited. On the other hand, the custom processor must be knowledgeable in the metallurgy and sc i ence of il wi de va ri ety of meta 11 i c and non-meta 11 i c materi a 1s. The particular processing problems of special materials must be identified. Extensions of applications will be related to the ingenuity of materials research workers in applying the unique pressure, temperature, and chemical variables in the HIP process. Possible new fields are cited.

9:30 a.m. ECONOMICS OF HOT ISOSTATIC PRESSING : Kenneth E. Meiners, Manager Materials Processing, Battelle-Columbus Laboratories, 505 King Avenue, Co 1umbus, Ohi 0 43201 The factors which influence the capital and operations costs of hot isostatic pressing (HIP) installation will be identified and discussed. Such factors as equipment utilization rate, product demand, equipment size, temperature, pressure, cycle time, product geometry, and product type will be described from the view point of general trends. Projected manufacturing costs

9:20 a.m. PHONON SPECTRA IN METAL-HYDROGEN SOLUTIONS: J. II. Rowe and J. J. Rush, National Measurement Laboratory, National -Bureau of Standards, Washington, DC 20234

The phonon spectrum of hydrogen in metals is a sensitive probe of the local environment of the hydrogen which can be measured directly by neutron inelastic scattering • . This technique has been used to study all three hydrogen isotopes in the group Vb metals Nb. V. and Ta. In addition, the effect on t~e hydrogen spectrum of 0, N, Ti. V. and Cr impurities in pure Nb metal has also been ~easured. By means of improved instrumentation, we have been able to measure hydrogen concentrations as low as 0.2 atomic percent (i.e., 20 ppm by weight). The results will be described in terms of the possible trapping sites for hydrogen in Nb with impurities and in terms of a possible clustering of hydrogen in pure Nb above the hydride phase precipitation boundary.

10:10 a.m. LOW TEIIPERATURE HEAT CAFACITY AND MAGNETIC SUSCEFTIBILITY: K. A. Gschneidner, Jr •• Ames Laboratory and Department of Materials Science and Engineering, Iowa State University, Ames, IA 50011

Only a few systems containing H dissolved 1n a metal have been examine
TMS Fall Meeting 9 Sc, Nb and Lu. One difficulty is the small solubility of H in metals below 7.5 DC. This Is no problem for Sc and Lu since they dissolve 'VJ5 at.% Hat 1 K. For Nb and Lu a Cp anomaly was found below 1.5 K. which is due to H tunneling associated with H-N or H-O pairs. In the Lu-H system the addition of ",1 at.X H increases both the electronic specific heat constant y (which is proportional to tbe density of states at the Fermi surface) and the Debye temperature SD (which Is a measure of the lattice stiffness) by "-'to and 10%, respectively. Additional H causes both quantities to decrease. The change in the X confirms the y dependence on the H concentration. The change in y. X and e are not well understood, but apparently are not associated with tunneling which occurs at lower concentrations <0.4 at.% H.

tRe

11:00 a.m. OPTICAL AND PHOTOEHISSION INVESTIGATIONS OF HYDROGEN IN METALS,* J.H. Weaver@ and D.J. Peterman, Synchrotron Radiation Center, University of Wisconsin-Madison, Stoughton, Wisconsin 53589

The photon Is a very flexible probe with which to examine the electronic properties of hydrogen in metals. We will review our unders tand i ng of the ro 1e of the e I ec t ron i c interact ions in hydrides and hydrogen solutions, emphasizing photoelectron spectroscopy results and optical studies, and relate our findings to the hydrogen absorption process. It will be shown that the symmetry and magnitude of the metal charge distribution at the interstice governs the metal-hydrogen interaction.

'Supported by USDOE Off ice of Bas I c Energy Sc i ences. @Oepartrnent of Chemical Engineering and Materials Sciences, University of Minnesota, Minneapolis, MN 55455 after 9/82.

ION IMPLANTATION AND IRRADIATION EFFECTS TMS-AIME General Abstract Session Monday, October 25, 1982

Cervantes Convention Center Room 271 Session Chairman: Dr. Gerhardt E. Welsch, Dept. of Metallurgy and Materials Science, Case Western Reserve University, Cleveland, OH 44106.

8:30 a.m.

8:30 a.m. EFFECfS OF ION IMPLANrATION ON TIlE FATIGUE GlARACfERISTICS OF TITANILN ALLOYS: J. Han, K. Jatavallabhula, E. A. Starke, Jr., Fracture & Fatigue Research Laboratory, Georgia Institute of Technology, Atlanta, GA 30332 and K. O. Legg, School of Physics Georgia Institute of Technology, Atlanta, GA 30332
9:00 a.m. ION IMPLANrATION EFFECfS ON THE FATIGUE LIFE OF A 4140 STEEL: p. I\eydari-Darani, K. Jatavallabhula, E. A. Starke, Jr., Fracture & Fatigue Research Laboratory, Georgia Institute of Technology, Atlanta, GA 30332 and K. O. Legg, School of Physics, Georgia Institute of Technology, Atlanta, GA 30332 One of the ways to improve the fatigue Ii fe of a structural component, is to modify the surface characteristics. A 4140 steel has been chosen in order to understand the ion implantation effects on the fatigue life' 2 Nitrogen ions have been implanted to a dose of 1019 ions/em on the steel samples. Low cycle fatigue tests have been conducted soon after implantation and after artificially aging the samples. An increase in the life time compared to the non-implanted condition is observed. The mechanisms controlling the improvement in fatigue life time are being established via electron I'licroscopy techniques.

9:30 a.m.

DEFECT PRODUCTION RATES BY ELECTRONS, IONS AND NEUTRONS IN CUBIC METALS: ~,{nstitut fur Festkorperforschunq der Kernforschungsanlage Jlilich, 0-5170 Julich, Assoziation KFA-Euratom, Gennany, B. R. Nielsen, H. H. Andersen, J. F. Bak, H. Knudsen, Institute of Physics, University of Aarhus, DK-BOOO Aarhus, Denmark, R. R. Coltman, Jr., C. E. Klabunde, J. M. Williams, Solid State

Division, Oak Ridge National Laboratory, Oak Ridge! TN, M. W. Guinan and C. E. Violet, Lawrence Livennore Laboratory, Llvennore, CA. The results on an interlaboratory program to study low temperature damage rates in dilute alloys of 300 ppm Zr in vanadium, niobium and molybdenum with electrons, light ions, fission neutrons and high energy neutrons are summarized. Additional experiments and literature data supplied complete sets of data also for the fcc metals Ai, eu and Pt. From the initial damage rates displacement functions for each material were derived which allows to reliably calculate atomic displacement rates also for particles and/or energies not employed in the program. Such calculations are given for defect production during the implantation of a-particles. The results are suggested to find practical use in defect calculations for fusion reactor first wall technology and in correlating the corresponding simulation experiments.

10:00 a.m. CONCENTRATION PROHLES AROUND INTERACTIVE SINKS DURING IRRADIATION:-;-, r'. A. Nichols, Materials Science Division, Argonne National Laboratory, Argonne, IL 60432

We have previously presented ail internally consistent model for radiatio~-induced kinetic processes such as void swelling, radia tion creep and radiation growth. The model employed a Wigner-Seitz cell around each type of sink and coupled these cells by requiring continuity in concentration and current at their boundaries. Explicit solutions for the currents entering the sinks were obtained for arbitrary sihk strengths and all interactive and competitive multi-sink effects were automatically accounted for. This eliminates the use of separate calculational models for obtaining sink strengths which may introduce inconsistencies and errors when employed in a reaction-rate-theory formalism. The nature of the solutions we obtained did not require explici t calculation of concentration profiles. Since these concentrations are required in order to form the averages needed for quantitative comparison with r:esults of other authors employing reaction-rate-theory, we have now calculated these concentration profiles. Quanti ta ti ve comparisons between our resul ts and those of others will be presented.

10:30 a.m. FRICTICN AND SOURCE HARDENING IN IRRADIATED MIUl STEEL: K. Linga Murty and Dirk J. Oh, N.C.State University, Raleigh, NC 27650-5636

Effect of neutron irradiation on tensile properties, in particu-

lar the yield and hardening behaviors was investigated on an experimental mild steel. Tensile test results on 0.05% C rimned mild steel irradiated at 80 DC were analyzed using the extrapolation method. Distinct Luders bands and lower yield stresses were noted at room temperature upto neutron fluence levels of

about 1018 n/cm 2 (> lMeV) while material irradiated to fluences in excess of 1019 n/an 2 exhibited "rounded" yield. Irradiation influenced both the friction and source tenns; while the friction hardening increased wi th fluence the SOill'Ce canponent initially increased reaching soon an apparent saturation level. These resul ts are in accord wi th those reported by Chow et al while are in conflict with those of Nichols and Harries. These findings may have a bearing on the radiation embri ttlement of

ferritic steels. Analyses of the experimental results at elevated terrperatures are in progress to evaluate the temperature and fluence dependences of ky and friction stress.

11:00 a.a. ON THE DEFINITION OF HICROHARDNESS:* Y. G. Yost, Sandia National Laboratories, Albuquerque, NH 871B5 Kicrohardness testing can be a very useful tool for studying modern 1I8teriais, but is plagued by well known experimental difficulties. Reasons for the unusual behavior of hardness data at very low loads are explored by Monte Carlo simulation. These simulations bear remarkable resbblance to the results of actual hardness experiments. The limit of hardness as load or indentation depth tends to zero is shown to depend on experimental error rather than upon intrinsic material properties. The large scatter of hardness data at very low loads is insured by the accepted definition of hardness. A new definition of hardness is suggested which eliminates much of this scatter and possesses a limit as indentation depth approaches zero. Some simple calculations are used to show the utility of this new approach to hardness testing.

* This

work performed at Sandia National Laboratories supported by the U. S. Department of Energy under Contrac t number DE-AC04-7 6DP007 89.

10 TMS Fall Meeting MODELING OF THE STRUCTURE AND PROP· ERTIES OF AMORPHOUS MATERIALS I Sponsored by Computer Simulation in Materials Science Activity of ASM and by the TMS Nuclear Metallurgy Committee Monday, October 25, 1982 Cervantes Convention Center 8:30 a.m. Room 272 Session Chairmen: Morrel H. Cohen, Exxon Research and Engineering Company, Corporate Research Science Laboratories, Linden, New Jersey 07036; N. Rivier. Blackett Laboratory, Imperial College, London SW7 2B7.

8:30 a.m. RECONCILING SHORT RANGE ORDER WITH PACKING CONSTRAINTS IN AM::>RPHOUS ALLOY MJDELLING: J.L.Finney, B.J.Gellatly, and J.P. Bouquiere, Crystallography Department, Birkbeck College, Malet St, London welE 7HX, England.

An acceptable structural model of a binary amorphous alloy must be noncrystalline & consistent with (a) known short range order (SRO) & (b) packing constraints imposed by the experimental

density. Our poor understanding of packing constraints even in

10:45 a.m. EXPLORING THE MEDIUM-RANGE STRUCTURE OF BINARY AMORPHOUS METALLIC ALLOYS AND OXIDE GLASSES: P. H. Gaskell,

Cavendish Laboratory, Madingley Road, Cambrldge CB3 OHE Knowledge of the irrmediate environment of an atom in a glass can often be inferred directly from experimental measurements and it becomes possible to dissect the structure into local clusters, with well-defined geometry and symmetry, centred on particular 'glass-forming' atoms. However, it is not clear whether definable structure persists at longer distances. The problem of definining this 'medium-range' structure depends absolutely on sound atomic models. To date, random models have received most attention, but even wi thin this concept there is scope for variations through changes in ring topology, etc. An attempt is made to calculate those properties of amorphous solids which distinguish competitive medium-range .structural concepts. Specifically, measurements of X-ray and n€utron scattering and vibrAtional r1ensities of state from transition metal-lIEtalloid alloys and silicate glasses are compared with properties simulated by computer-refined atomic models.

11:15 a.m. STUDIES OF SKORT-RANGE ORDER IN METALLIC GLASS ALLOYS:

sinqle component den!2 random packings makes difficult the

D. S. Boudreaux, Corporate Research Center, Allied Corporation,

reconciling of these two constraints in binary alloy models.

Morristown, NJ

In the two main approaches to modelling binary alloys, SRO and packing constraints have been differently treated. Boudreaux stresses packing, while Gaskell's method gives primary emphasis to SRO, packing constraints being introduced by subsequent relaxation. The relative success of these methods in reconciling SRO wi th packing constraints is discussed, using the radical plane analysis to probe the local ordering distortions. The possibili ties are discussed of two further procedures designed to reconcile SRO and packing constraints.

9:00 a.m. DENSE RANIJOM PACKING MODELS FOR AMORPHOUS RARE EARTH METALG. S. Cargill III, IBM T. J. Watson TRANSITION METAL ALLOYS:

Research Center, Yorkton Heights, NY

07960

Metallic glass alloys are now known to possess a Significant short-range order (SRO).

The kind and degree of SRO permits the

systematization of many of the properties of these materials. Theoretical methods and computer modelling have been brought to bear in the absence of traditional probes which depend on lattice periodicity. Some of the techniques to be described are: s1011lation and analysis of pair correlations, application of graphic methods, number-concentration correlations, and comparison of model simulated EXAFS to experiment. Once models of SRO have been established, it is possible to predict variations of specific quantities which are not directly measurable. It is important to develop improved methods of cataloging SRO for use in the construction of more detailed theories, e.g. in the calculation of electronic structure.

10598

Binary dense random packing models have been generated for amorphous rare earth metal-transition metal alloys over wide composition ranges. Comparisons have been made with experimentally accessible information about the atomic scale structures of these alloys, e.g. Cd-Co alloys. Effects of different generation and relaxation algorithms have been investigated. Results of these studies will be discussed and compared with results for other types of amorphous alloys.

9:30 a.m.

NOVEL TECHNIQUES IN METAL DEFORMA· TlON TESTING I: Formability Testing Sponsored by the TMS Shaping and Forming Committee and the ASM Process Modeling Activity Monday, October 25, 1982 Cervantes Convention Center 8:30 a.m. Room 120 (First Floor) Session Chairman: R. H. Wagoner, Physics Department, General Motors Research Laboratories, Warren, MI 48090·9055.

THE GEOMETRY OF CAVITIES IN DENSE RANDOM PACKINGS by H. J. Frost,

Thayer School of Engineering, Dartmouth College, Hanover, N.H. 03755

8:30 a.m.

The structure of dense random packing of spheres is analyzed in terms of the polyhedral cavities bet~een spheres, using either the polyhedra with all triangular faces, known as deltahedra, or the shapes possible with both triangular and four-edged faces. Both these classification schemes follow a network approach in which shapes are identified according to the network of connections between neighboring spheres. An alternative for defining cavity geometry is a vector approach where the shapes are distinguished by the angular distribution of the vectors from the cavity center to the neighboring spheres. The two approaches give the same identification of regular polyhedra, but may give different identifications to distorted shapes. The vector approach may be useful for describing the arrangement of metal atoms surrounding a metalloid. Cavities are also characterized by the sizes of small interstitial spheres that can be placed inside. These results are correlated with the network and vector descriptions of cavity shape.

It.

10:00 a.m.

Break

10:15 a.m. COMPUTER SIMULATION OF AMORPHOUS IRON: Masao Doyarna, Department of Metallurgy and Materials Science, Faculty of Engineering. the University of Tokyo, 7-3-1 Hongo Bunkyo-ku, Tokyo 113 Japan Assuming a pairwise interaction potential between atoms, the atomic structure of amorphous iron has been studied. Bernal holes and Voronoi Polyhedra have been classified. The connection between Bernal holes are studied. The radial distribution functions were calculated and compared with the experimental ones. Liquid iron was quenched and obtained amorphous iron. The radial distribution function obtained in this way was compared with the experimental one. A small piece of amorphous iron was pulled and displacements of atoms were studied. Radiation damage in amorphous iron was also studied.

PLANE

STRAIN

TENSILE

It.PPIt.Rlt.TUS:

Alcan International Limited, Ontario, Canada, K7L 4Z4.

~

Kingston

and

Y.

Nishikawa,

Laboratories,

Kingston,

A simple clip-on apparatus has been designed to deform a modified sheet tensile specimen In a state of plane strain. The apparatus consists of a sprung clamp having four knife-edges to prevent specimen edge pull-in. Both the plane strain limit strain and fracture strain are obtained from measurements on the fractured sample. Results compare favourably with preViously published data. In addition, the apparatus has been used to measure the plane strain formability parameters at non-ambient temperatures.

8:50 a.m. EDGE

I'ORMlt.BILITY

CHAR1t.CTERIZIt.TION

OF

SOME

ALUMINUM

L.D. Kenny and H. Sang, Alcan International Li.ited, Laboratories, Kingston, OntariO, Canada, K7L 4Z4.

ALLOYS:

ICings ton

The quality of edge surfaces of sheet -.etal can affect subsequent fonrlng. Failure in such fonalog operations aa press bending, curling, hole flanging, and flaring can occur at damaged areas produced during shearing, punching or machining of the edge. A test rig has been designed to measure the effect of edge condition on strain to crack initiation in some alullinum alloys. The apparatus subjects specimens of various edge roughness to a tensile stress gradient across the specillen width. Results of this work, together with results of hole expansion tests, show that crack initiation is very sensitive to both sheet thickness and edge condition.

9:10 a.m. TESTING LUBRICANTS FOR SHEET METAL FORMING: H. D. Nine, Physics Department, General Motors Research Laboratories, Warren, MI 48090-9055

TMS Fall Meeting 11 Lubricants control friction in sheet metal forming processes. Friction, however, is a function of variables which may be affected by the forming process itself. Examples are discussed in which sheet metal surface roughness and material properties interact with lubricant viscosity in a complex way to change friction. Based on these complexities, the import8j"lce of duplicating production conditions for a lubrication test is emphasized. A drawbead simulator test is described which has been successful in specifying lubricants for production parts in which sheet metal is drawn through drawbeads and over a die throat. It is stressed that no one lubrication test is applicable to all sheet metal forming applications. 9:30 a.m. AIDS FOR EVALUATING SHEET METAL FORMABILITY: THE LIMITING DOME HEIGHT TEST AND THE GRID CIRCLE ANALYZER: R. A. A~res, Physics Department, General Motors Research Laborator.es, arren, MI 48090-9055 Simulative tests are an irlportant tool in understanding the complex interaction of sheet metal, die tooling, and lubrication during the stamping process. The Limiting Dome Height (LDH) test simulates the common failure strain states in stampings. The LDH test ranks the stretch-formability of various sheet materials and correlates laboratory values to press performance. It is necessary, however, to conduct this test according to a published standard procedure. Key test parameters, punch rate and lubrication are varied to evaluate their effect on the LDH values. This evaluation requires measurements of strain from grid circles printed on the specimen blanks. These measurements are rapidly made by the Grid Circle Analyzer which is a computer-aided/optics system. It is concluded that the standard punch rate from 0.42 to 4.2 mm/s with no lubricant is best ror material ranking and data reproducibility. 9:50 a.m. OBSERVATIONS OF DIE-WORK INTERFACES THROUGH TRANSPARENT SAPPHIRE DIES: R. S. Rao, C. Y. Lu, P. K. Ilright, Carnegie-11ellon University, Pittsbu~15213; M. L. Devenpeck, O. RiclYllOnd, E. J. Appleby, U.S. Steel Corp., Research Laboratory, Monroeville, PA 15146.

10:50 a.m. ANALYSIS OF ANGULAR STRETCH-BEND TEST: O. S. Narayanaswamy and M. Y. Demeri, Engineering and Research Staff, Research, Ford Motor Company, Dearborn, MI 48121. A si.ple _ e l for calculating fracture strains in the angular stretch-bend test is used to analyze data relating aheet . . tal thickneas(t) and punch radiua(R) to the height of the test piece at fracture( H). Unlike H, the calculated strain at fracture(£ ) is found to be relatively independent of punch radius a;{ sheet thickness. Tberefore, E f qualifies as a . .asure ot tbe intrinSic fo ..... bility ot sheet . . tal. Fracture beigbt and location in tbe stretcb-bend test depend on lubrication and t/R ratio. Lubrication increases height

when t'racture ocours outside the punch contact area. Tbe reverse is true wben fraoture oocurs under the punoh. Tbese erracts are explained on tbe basiS of frictional constraints at tbe sheetl punch interface.

11:10 a.m. THE EFFECTS OF STRAIN HARDENING, STRAIN RATE SENSITIVITY, AND NORMAL ANISO"mOPY ON HIGH-SPEED DOME TEST RESULTS FOR PRECIPI-

TATION-HARDENED ALUMINlM ALLOY SHEETS: J.M.Story, W.H.Hunt, Jr., O.J.Lege, Alcoa Laboratories, Alcoa Center, PA 15069 Four precipitation-hardened aluminum alloys, Al-0.8 Mg-0.90 Si, Al-O.8 Mg-O.56 Si, Al-l.O Mg-O.67 Si, Al-3.5 Cu, and a dispersion-strengthened base alloy, Al-0.4S Mn-O.l Fe, are prepared with both fine and coarse grain sizes. 'ltlese alloys are hemispherical punch stretch-tested near plane strain. Mean plastic strain ratios are calculated from uniaxial-tension test results. Data from hydraulic bulge testirlg are fit to the Voce equation, and instantaneous strain hardening and strain rate sensitivity exponents are calculated. The effect of microstructure on these parameters and the effect of these parameters on the mininum dome heights obtained in punch stretching are discussed. '!he present study differs from earlier work by Ghosh, in that the minimum point on the dome height versus minor strain curve is used as a measure of stretch formability and the punch speed used in the dome testing is much greater.

Interfacial velocities at the tool/workpiece interface have been measured during drawing of tinplated mild steel strip. Techniques for measuring the displacement of the strip surface as the strip passes through the dies are described. Movies were used to observe dynamic behavior. Transparent sapphire dies will allow direct observation of the micromechanisms of interfacial behavior including flattening of strip-surface a.sperities and the resulting entrapment of lubricant. Because sapphire dies are birefringent, photoelastici ty can be a.pplied to the traverse section of the forming dies.

*Sponsored in part by the National Science Foundation.

10:10 a.m. LIMITING STRAIN TESTS FOR IN-PLANE SHEET STRETCHING: M. L. Devenpeck and O. Richmond, U. S. Steel Corporation, Research Laboratory, Monroeville, FA 15146 Mechanical tests designed to measure the stretchability of sheet materials independently of tool/workpiece interface effects are presented. The deformation in tension specimens is controlled by electron-beam-welded support plates, and the strain state is varied by using different 1ength-to-width ratios. The Marciniak hollow punch test is used to evaluate balanced b.iaxial stretching. Strain histories are obtained from photographs of the specimens, giving far more accurate values of "safe," "neCk," and "failure" strain states for generating FLD curves than customary techniques. Holographic interferometry observations of these plane specimens demonstrate the collapse of the plastic zone as necking sets in.

11 :30 a.m. IN-PLANE COMPRESSION.

LATE DRAWING IN SHEETS.

A NEW LABORATORY FORMING TECHNIQUE TO SIMU-

P. L. Charpentier and H. R. Piehler

The deformation mode or strain path imposed during sheet forming can have a strong influence on the mechanical behavior of materials. To study strain-path effects, laboratory forming techniques capable of producing deformed samples that are flat, of suff iCient size, and free from strain gradients are desirable.. A new laboratory forming technique, called in-plane compression has been developed in this stUdy to produce samples in which the stram state simulates that observed in the flange of a partially drawn cylindrical deep-drawn cup, that is, £1 <0, £2 and £3 >0. These samples are essentially strain-qradient free, and of adequate size (sheet thickness by 31 .. 8 by 76.2 mm) to permit subsequent me.asurement of the postformed mechanical properties.. For a 552 MPa (80ksi) yield-strength high-strength low-alloy (HSLA) steel, the postformed tensile properties after in-plane compression aqree well with those observed after cup drawing.

PHASE TRANSFORMATIONS I TMS-AIME General Abstract Session Monday, October 25,1982 Cervantes Convention Center 8:30 a.m. Room 263 Session Chairman: Dr. J. A. Shields, Climax Molybdenum Co., Re· -~h Lab., P.O. Box 1568, Ann Arbor, MI 48106.

8:30 a.lII. 10:30 a.m. A TAPERED SPECIMEN FOR MEASURING SHEET STRETCHABILITY: O. Richmond, M. L. Devenpeck, and H. L. Morrison, U. S. Steel Corporation, Research Laboratory 1 Monroeville, PA 15146

NON-DESTRUCTIVE DETECTION OF A PRE-aiEGA PHASE IN BETA-III T1 BY SANS: K. Fatemi, C. S. Pande, M. A. Imam, and P. G. Moore, Naval Research Laboratory, Washington, DC 20375 and H. R. Child, Solid State DiviSion, Oak Ridge National Laboratory, Oak Ridge, TN 37830

The limit strain before necking in a uniform strain field has long been recognized as a useful measure of ductility. It is the basis for forming limit diagrams. An additional measure of ductility is needed, however, to assess stretchability outside of the necked region in a nonuniform strain field. Limiting dome height of a punch stretch test is one such measure, but it is cOIT,plicated by specimen-curvature and tool-interface effects. To avoid these complications, a tapered tension specimen is proposed and analyzed. The strain in the minimum section at initiation of necking is a measure of resistance to localized thinning, and the total elongation is a measure of the ability to spread strain outside of the neck. It is demonstrated that two materials with the same limi ting local strain can have" quite different total elongations.

Omega phase precipitation in beta-III Ti-alloy was examined by small-angle neutron scattering (SANS) to investigate the correspondence between the profiles and the distribution of defects. These profiles, obtained from a series of 32 specimens prepared over an extensive range of aging temperatures and times, showed as their principal component broad peaks which followed a complex pattern of shifting towards and away from the main beam. Close examination of the peak positions suggested the existence of an early-stage, pre-omega phase transformation in this alloy. The accompanying variations in morphological and crystallographic processes in this transformation were corroborated by transmission electron microscopy (TEM), in the relevant time-temperature regime. The application of non-destructive SANS method has demonstrated its potential in revealing bulk microstructural properties which have not been previously reported in similar TEH investigations.

12 TMS Fall Meeting 8:50 a.m. CRYSTALLOGRAPHIC ANALYSIS OF HARTENSITES IN TITANIUM ALLOYS:* Masaharu Kato and K. Mukherjee, Department of Metallurgy, Mechanics and Materials Science. Michigan State University, East Lansing, MI 48824 A survey of literature indicates that there are several martensttic structures in commercial Ti alloys. These structures include a hexagonal Q I, an orthorhombic a". a face centered cubic, and a face centered orthorhombic (f. c. 0.) unit cells. We have reviewed the existing results to examine which of these structures are justifiable in terms of observed orientation relation-

ships and the applicability of the crystallographic theories of martensitic transformation. We will show that a unified crystallographic approach can be used to express orieJ'ltation relationships of the a',a" and the f.c.o. martensites. Further, it is possible to apply the strain-energy minimization criterion to predict habit planes of these structures which agree with experimental observation.

'This work is partially supproted by the U. S. Office of Naval Research under the contract No. NOOOl4 82-K-0268.

9:10 a.m.

The transient time (the time l'equired to reach the steady state nucleation rate) may be experimentally observed but the result is often found to depend on the experimental method; we show how this can be explained by considering the growth rate of postcritical nuclei. *Present address:

Cambridge, MA

10:30 a.m. THE COARSENING BEHAVIOUR OF SEMI-COHERENT PRECIPITATES: D. Banerjee*and M. L. Bhatia, Defence Metallurgical R, search Laboratory, Hyderabad-500258, India The possibility that semi-coherent precipitates may be stabilized against Ostwald Ripening is explored by analyzing the energetics of semi-coherent interfaces. It is shown that the strain energy cusps which exist at preCipitate sizes corresponding to Brooks Criterion (D=r...'b/£o) may constitute a barrier to coarsening. A criterion for stabilization which depends on the magnitude of the term.....§...... b ~ is proposed, and the criterion applied to explain I-v y the experimentally observed stabilization of spherical, semicoherent Co precipitates against coarsening in the Cu-Co system.

THE ORIENTATION RELATIONSHIPS BETWEEN BeC Mo AND FCe y IN A Ni-Al-Mo-W SUPERALLOY: D. M. Dimiduk, AFWAL/HLLM, WrightPatterson AFB OH 45433

*Now

Rapid solidification P/M techniques have been used in developing nickel-base superalloys with high molybdenum contents. Elevated temperature aging of these consolidated powders results in the precipitation of a Bee Mo + W phase. The present study centers around the orientation relationships between the Bec precipitate phase and the FCC matrix in an alloy with the following composi-

10:50 a.m.

tion:

72.6Ni-6.8AI-I4.4Mo-6.2W-.04e in weight percent.

BCe

precipitates with two different orientation relationships were observed after aging samples at 930 0 C for 250 hours. TEM studies have shown that the Nishiyama-Wassermann BCC-FCC orientat ion relationship holds for some precipitates while a less common orientation relationship, that heing the Pitsch relationship for BCC-FCC systems, holds for a majority of the Mo + W precipitates.

9:30 a.m. HARTENSITIC TRANSFORMATION IN A Ni3Sn ALLOY FROM THE D03 TYPE ORDERED PARENT PHASE TO A 2H STRUCTURE: S. Chang, Hasaharu Kato and H-r. Pak, Department of Metallurgy, Mechanics and Materials Science, Michigan State University, East Lansing, HI 48824

Research Laboratory of Electronics, M.I.T., 02139

at Carnegie-Mellon University, Pittsburgh, PA

COMPUTER SIMULATiON OF TRANSFORMATIONS IN SOLIDS:

15213, U.S.A.

D. Esterling,

George Washington Uni versi ty, Hampton, VA 23665, R. Arsenault, University of Maryland, College Park, MD 20742 and B. B. Rath and R. A. Masumura, Naval Research Laboratory, Washington, DC 20375 Nucleation and growth controlled transformations in solids have been modeled by a computer simulation method. A variety of generalized empirical forms has been assumed to describe both the nucleation and the growth processes and is consistent with experimentally observed behavior. The nucleation event is based on either a pre-existing embryo or time-dependent evolution model and is allowed to start at randomly distributed sites in a large three-dimensional block. Stereological parameters are evaluated as a function of time using statistical analysis. From these calculations, the time-dependent volume fraction of the transformed phase and migrating interfacial area are determined and related to other theoretical models and the observed behavior in

solids.

The high temperature phase of the Ni3Sn has been found to have a D03 type ordering as determined from high-temperature x-ray diffraction. The martensitic product phase has an orthorhombic symmetry and the electron diffraction pattern can be analyzed as a long-period stacking order of the 2H type. Such 2H martensites have been observed in other alloys where the parent phase has CsCl type ordering. The lattice invariant deformation has been

11:10 a.m.

analyzed as m'ainly (121)2H twinning with a few amount of (101)2H twinning.. The experimentally observed habi t plane is compared with that determined by applying the strain energy minimization criterion.

The effects of transfonnatlon cycling between the parent (P) and martensite (M) phases of selected Cy-Zn-Al alloys were studied. Us 1 ng differentia 1 scannl ng calorimetry 1 t was observed that the martensite start temperature, Ms ' progressively shifts upward w1th cyc 11 ng. Al so, the proportion of the ml cros tructure IC tually undergoing P <-+ M transfonnatlon decreases. Transmission electron microscope observations reveal dislocation substruct~res generated by martensltlc transfonnatlon and reversion.

9:50 a.m. KINETICS OF THE ORDER-DISQRQER TRMSFOR~:ATJONS. * J. Kevin McCoy, ** Ryoichi Kikuchi, * Klaus Gsch,;end and Hiroshi Sato, School of Materials Engineering, Purdue University, West Lafayette, IN 47907. The path probability method of irreversible statistical mechanics is applied to a system undergoing a first-order, order-disorder transformation. For this purpose, the path probability method has been worked out for the first til:le in a degree of approximation higher than the pair approximation. The characteristics of the order-disorder t,-ansfor",ation in the A?R type system in the two dimensional triangle lattice are ~i:lculated

TRANSFORMATION CYCLING EFFECTS IN MARTENSlTIC Cu-Zn-Al SHAPE MEMORY ALLOYS.* Jeff Perkins, Materials Science Group, Oepartment of Mechanical Engineering, Naval Postgraduate School, Monterey, CA 93940

*This research was sponsored by the 01vlslon of Materials Research of the National Science Fo~ndatlon through Grant No. lMR-81-08407.

TIME DEPENDENT PHENOMENA: FATIGUE TMS-AIME General Abstract Session Monday, October 25, 1982

a.m.

Cervantes Convention Center Room 266

to show the !lew featurc-s and the complexities \-Jhich arise. The

8:30

treatment is compared "ith the previously published derivation for the pair approximation.

Session Chairman: Dr. David J. Quesnel, Dept. of Mechanical Engin. eering, University of Rochester, River Campus, Rochester, NY 14627.

-. -*- --- "--- - - -------- -Supported by CIS, under Gront No. ltiR790:;93? c1dJress: Phy:,;cal :':etal1ul'qy Seetior:, S3.ttele Colulllbus Li:boratories, Colu,::l,us, 01: 4J201. ***Permd nent add t'PS s: Huqhes Resea reh Lit bora tori es, Ma 1 i bu. CfI **rl~esent

90265.

10:10 a.m. MODELLING OF TRANSIENT NUCLEATION IN CONDENSED PHASES: K.F. Kelton, A.L. Greer and C.V. Thompson*, Division of Applied Sciences, Harvard University, Cambridge, MA 02138 Non steady state (transient) nucleation is important in many transformations; in particular the redu tion of nucleation rates from steady state values may be crucial in the formation of some amorphous alloys. We present an exact solution of the coupled differential equations describing the non steady state regime in a one-component system at constant temperature. An alternative approach, which is mathematically more simple and may more easily be applied to the non-isothermal case and to multi-component systems, is also presented; it is shown to give time-dependent embryo populations in good agreement with the exact solution.

8:30 a.m. EFFECTS OF DYNAMIC STRAIN AGING ON CYCLIC DEFORMATION OF a-IRON: * K. Maruyama and M. Meshii, Materials Research Center, Northwestern UniverSity, Evanston, IL 60201 Cyclic deformation was studied in annealed (c + 0: 200 at. ppm) and ZrH2-purified HRC VP-grade iron polycrysta1s. Tests were carried out in a pure argon atmosphere under total strain control (strain amplitude: 10- 3 and 10- 2 ) at a strain rate of 10- 3s- 1 A large hump appeared on the saturation stress vs. temperature plot for the annealed specimens around 600 K where the stress level was I'l.early twice that of the purified specimens The purified specimen also showed a small hump near 700 K. Stress relaxation, static strain aging and strain rate dependence of stress examinations indicated that the hump at 600 K was caused by dynamic strain aging of carbon, while the hump at 700 K was due to dynamic strain aging of oxygen. *Supported by the NSF-MRL program

(Grant No. DHR79-23573).

TMS Fall Meeting 13 8:50 a.m.

10:30 a.m.

FATIGUE BEHAVIOR OF AL203 FIBER REINFORCED MAGNESIUM:. J. E. Hack and G. R. Leverant, Southwest Research Institute,

6220 Culebra Rd., San Antonio, Texas

78l84.

Tension-tension fatigue testing was performed on unidirectional A120) fiber reinforced cOUIDercially pure magnesium matrix COlEr posttes at 10 Hz and an R-ratio of 0.1. The test specimens included material of two different volume fractions of fiber and four different orientations of the fibers to the tensile axis. It was found that fatigue crack initiation in the material with the fibers aligned parallel to the stress axis was extremely sensitive to the presence of process defects. In spite of this, the endurance limit of the material with defects remained high

(- 70% UTS).

Fatigue and tensile behavior were also found to be

sensitive to the through-the-thickness orientation of off-axis fibers. Initial results of fatigue crack growth experiments will also be discussed.

*Work sponsored by the Army Research Office under Contract

DAAG29-81-K-0049.

Crack Propagation in Equiaxed 0(.) Ti-Mn Alloys: J. Park and H. Margolin, Department of Physical and Engineering Metallurgy, Polytechnic Institute of New York, Brooklyn, N.Y. 11201. Crack propagation studies were carried out on compact tension specimens, with the crack oriented in the transverse (LT) and

the longitudinal (TL) directions, of equiaxed

oi·~

Ti-Mn alloys

containing 0.4, 2.0, 3.9, 5.6, 8.0, and 10.0 Mn. The alloys were heat treated at 700 0 C to produce constant composition of phases, with vol. % of ~ varying from 1.9 to 75.97.. Textures in4("'" were determined for each alloy. From the orientation of possible slip systems with respect to the stress axis and the nominal crack propagation direction, it was possible to predict the relative rates of crack propagation in the LT and TL directions as the vol. i. of ~ increased. A Kth was found to vary with vol. i. of;! in the same way as y. s., and was determined by slip in "" , which slip contributed to the formation of a "shelf" in Stage I. This work was sponsored by the Air Force Office of Scientific Research under Grant 1179-0028.

10:50 a.m.

9:10 a.m. EFFECT OF AN IN!'ERMEDIATE THFR.'.\AL I-lEOiANlCAL TREAT/lEN!' ON FATIGUE PROPERTIES OF I/H X709l ALlJlINlJ.1 ALLOY: L9)an g and E. A. Starke, Jr., Fracture & Fatigue Researc a oratory, Georgia Institute of Technology, Atlanta, GA 30332

U'

A four- step intennediate thennal mechanical treatment (In-IT) has been conducted on an I/M X709l aluminum alloy. The pUl1JOse of using this 11MT processing is to optimize the microstructure in order to obtain a better combination of mechanical properties than those obtained for conventionally processed material. When compared with conventional processing, 11MT processing produces a fine, homogeneous grain structure, and a random texture. Low cycle fatigue and fatigue crack propagation tests have been conducted on the alloy in the peak aged condition in a dry air envirorunent (humidity < 3 ppm). Experiment results can be explained in tenns of the microstructural fea· tures, especially the refined grain size and texture.

9:30 a.m. CYCLIC DEFORMATION OF A SPINODALLY HARDENED Cu-Ni-Cr ALLOY: S. Snyder and W.A. Soffa, Department of Metallurgical and Materials En~ineerin~, University of Pittsburgh, Pittsburjith , FA 15261. The cyclic hardening/softening behavior of a spinodally hardened Cu-3INi-2.7Cr alloy has been studied under conditions of controlled plastic strain, and the slip band morphology and crack initiation behavior have been characterized and correlated with the microst ruc ture of the age hardened polycrystals. Fur thermore, the Bauschinger effect has been used to analyze the reverse strain behavior in terms of the fundamental micromechanisms of flow. This work is supported by the National Science Foundation under

grant DMR-800 5989.

Stage II Crack Propagation in Equiaxed Ti-Mn Alloys: J.S. Park and H. Margolin, Dept. of Physical and Eng. Metallurgy, Polytechnic Inst. of New York, Brooklyn, NY 11201 Crack propagation studies were carried out on 6 equiaxed Ti-Hn allOYS, with 0.4-10 Mn, in the TL and LT orientations. The y.s. of the "" phase was 1/3 that of jf and the y.s. of the alloys increased with increasing vol. % of ,I. When~• ..was plotted against ~K, nonnalized for y.s., the crack prop. curves separated out according to vol. % of,l .4K/y.s. is proportional to (7, where rf is the measured plastic zone size. A a/A N4r, data fell on separate cunes for the TL and LT directions. Deformation bands in ~ ) emanating from an interior crack surface and the extent of which defined the plastic zone size, were observed long before striations were observed. This behavior is discussed in terms of the strain partit ioning between 01 and /l phases and the Bauschinger effect.

This work was supported under Air Force Office of Scientific Research Grant #79-0028.

11:10 a.m. DIRECTIONAL MECHANICAL PROPERTIES AND FATIGUE OF A FIBER REINFORCED EPOXY COMPOSITE:. C. F. Chen, J. Hartin and K. Mukherjee, Department of Metallurgy, Mechanics and Materials

Science, Michigan State University, East LanSing, HI 48824 Directional mechanical properties and fatigue response of a gl&ssfiber reinforced epoxy composite have been investigated. the composite consists of alternate layers of woven glass fibers interposed by an epoxy matrix. Static mechanical properties, as a function of fiber orientation. have been measured in order to understand anisotropy of fatigue response. A laser int"erferometric strain measuring system is utilized which can determine

dynamic strain amplitudes in gauge lengths as small as 50

~m.

9:50 a.m.

Using this technique, anisotropy and local variations in the cyclic stress-strain curves can be determined. These results

FATIGUE DAMAGE IN NICKEL: S. Sivashankaran, and G. Welsch, I}{>partment of Metallurgy &. Materials Science, Case Western Reserve University, Cleveland, Ohio 44106.

will be presented in detail.

The accumulation of fatigue damage in flexurally fatigued nickel samples was measured by residual resistivity. Subsequent isochronal annealing treatments were performed to gain inSight into the types of lattice defects that constitute fatigue damage. Additional characterization was done by optical, scanning and transmission electron microscopy. The accumulation of dislocations and of vacancies or vancancy clusters will be discussed. This work was supported through a gran t by NASA - Lewis Research Cen ter.

10:10 a.m. FATIGUE CRACK GROWTH AND CREEP OF HYDROVAC-PROCESSED TITANIUM

ALLOYS: * P. S. Pao, R. J. Lederich, and S. M. L. Sastry, McDonnell Douglas Research Laboratories, St. Louis, MO 63166 and W. R. Kerr, Air Force Wright Aeronautical Laboratories. Dayton,

OH 4S433. Hydrovac processing. which consists of adding ~ 1.0 wt % hydrogen to titanium alloys, heat treating, and dehydrogenating, results in fine microstructures with order-of-magnitude smaller grain and subgrain sizes. The effect of hydrovac processing on the fatigue and creep properties of equiaxed Ti-6Al-4V and duplexannealed Ti-6Al-2Sn-4Zr-2Mo were investigated by comparing the crack growth rates and creep rates of hydrov2.c-pro cessed material to those of as-received material. Hydrovac-processed material of both alloys is more resistant to fatigue crack initiation but more susceptible to subsequent crack growth . The creep rate of Ti-6Al-4V is reduced by one-tliird upon hydrovac processing, but hydrovac processing does not improve the creep resistance of Ti-6Al-2Sn-4Zr-2Mo. The susceptibili ty to crack growth and creep deformation of hydrovac-processed alloys will be correIa ted with the micros true tures. *This research was co nducted under Air Force Contract No.

F3361S-80-C-SIl8.

* This

.

work is partially supported by the Ford Motor Company.

11:30 a.lII. C!lARACTERIZA:rION OF FATIGUE CRACIC GROWTH BEHAVIOR OF X'7091T7E69 EXTRUSION: S .t. Lal!!!enbeclt, J .M. COlt, Lockheed-California Canpan,y, au-b auk, CA 91520 Fracture tOll8hne88 and fatigue crack growth data on X'7091-T7E69 extruded rectangular bar are presented. The test parameters invest1
14 TMS Fall Meeting

MONDAY - OCTOBER 25, PM

(Fe,Cu) (Al,Cu)6' MgzSi and AlxSiy compounds . The positive identification of these phases by the combination of diffraction and mi croana lys i s techni ques, and the effect of heat treatment and impurity content on the chemical composition of these phases wi 11 be presented.

4:20 p.m.

CONSTITUENT PHASES IN ALUMINUM ALLOYS II Sponsored by the TMS Nonferrous Metals Committee

Control of Alloy Chemistry during Laser Welding of AL-5083: A. Blake and J. Mazumder. Department of Mechanical & Industrial Engineering, University of Illinois, Urbana , IL 61801

Monday, October 25, 1982 Cervantes Convention Center 2:00 p.m. Room 275 Session Chairman: T. H. Sanders, Jr., School of Materials Engineering, CMET Building, Purdue University, West Lafayette, IN 47907.

One of the main problems for autogenous laser weldin g of Ai-50B3 is the loss of magnes i um as plasma due to its high vapor pressure and the consequent p o rosit y . A method has been developed to co ntrol the plasma and thus control the alloy chemistry dur in .~ laser weldin g. The experiments have been carried out usin g a 10 KW CWC02 laser following a two level factorial design to co rrelat e the ef fects of laser processin g variables with the resultant microstructure. Microstructure and alloy compositions have been characterized using optical and e l ec tron optical techniques (SEMI EDAX) and electron probe microanalysis techniques. Coolin g rate s have been estimated using a 3-D heat trans fer model. Interdendritic seg regation of the intermetallics are characterized us ing SEM/EDAX. It was observed that maRnesium loss could be minimized to as low as 0.9 percent. Radiographic examinat ion and t ensile tests revealed that low porosity welds with st ren g ths at least as good as the parent material may be produced by controlling the alloy composit ion.

2:00 p.m. OPTICAL AND MICROANALYT ICAL TECHNIQUES AND TOOLS TO DETERMINE CONSTITUENT PHASES IN ALUMINUM & ALUMINUM ALLOYS:

K.

s.

Chopra ,

Elkem Metals Company, Technology Center, P.O. Box 1344, Niagara Falls, NY 14302 The object of this paper is to demonstrate how optical and a host of microanalytical tools can be used to solve phase characterization and other problems in aluminum and aluminum alloys. The paper will deal with optical and microanalytical tools and techniques used for phase characterization, crystal structure by using complimentary instrumentation such as Opti cal, Electron Microprobe, TEM and X-ray diffraction. Examples will include aluminum, binary system, ternary systems, 3000 and 5000 ser ies alloys.

2:35 p.m. THE EFFECT OF DISPERSOID TYPE, PURITY AND OXIDES ON NEAR-THRESHOLD STRESS INTENSITY, AKth' IN I/M AND P/ M 7XXX ALUMINUM ALLOYS:* M. Zedalis, L. Filler , S. 1. Kwun and M. E. Fine, Dept. of Mat. Sci. and Eng. and Mat. Res. Ctr., Northwestern Univ., Evanston, n.

DEFORMATION II TMS-AIME General Abstract Session Monday, October 25, 1982 Cervantes Convention Center 2:00 p.m. Room 262 Session Chairman: John A. Spitznagel, Westinghouse Research and Development Center, 1310 Beulah Rd., Pittsburgh, PA 15235.

2:00 p.m. A :1F.X:HANISM FOR D!
;tabul Basu, Scientist

High purity 11M 7050 aluminum alloy containing ZrAla dispersoids d emons trated in th e peak aged condition an improved 6~h when co,?pa;ed to ~he I /M 7475 alloy containing Al,. Mg" Cr dispersoid •• Th~s 1.S attr1.buted to the ability of the semi-coherent ZrAls dispersoids to plastically deform during cycling. The commercial purity variant of the 11M 7050 alloy containing constituent particles demonstrated a higher 6Kr.h than found for the high purity alloy. This is attributed to an increased homogenization of the deformati on by the constituent particles. Fatigue tests performed on air atomi.ed P/M 7091 (0.47. and 0 •.8% Co) aluminum alloys in the as-extruded and thermomechanicaily treated condition resulted in a reduced 6Ke h when compared to the ingot alloys. This response was suggested to be due to oxides introduced during air atomiz.ation of the rapidly solidified powders.

Experimental data for dislocation mobilities in z inc single crystals is analysed. A drag model is formulaterl for the low stress and high stres. ree-ions after Stage I-etage II transition. It i. found that the drae appro'iches an asymptotiC value R.t high velocities starting from an ini tially high value. The significlUlce of this value is discussed reGarding the formation of Zn-N interstitial molecules that impede the free flight of dislocations on slip systems in zinc single crystalline specimens.

* Supported by AFOSR-78-3732B

grant at tne C"lifornia Institute of Techn.Ioe:v.

3:10 p.m.

• Research sponsored by the National Science Foundation on a summer

2:20 p.m .

INFLUENCE OF CONSTITUENT PHASES IN HIGH STRENGTH POWDER METALLURGY ALUMINUM STRIP by H. Sunwoo and D. H. Ro, Technology Center, Imperial Clevite Inc., Cleveland, Ohio 44108

YIELDING BEHAVIOR OF NOTCHED BEND BARS: D. J. Alexander, J. J. Lewandowski, A. W. Thomps on and 1. M. Bernstein , pept. of Met. Eng. & Mat. Science, Carnegie-Mellon UniverSity, Pittsburgh, PA

To take full advantage of the versatile PIM process, an understa nding of the relationships between the var i ous microstructural features and processing to achieve properties in rapidly solidified aluminum alloys needs to be understood. This paper is concerned with the constituent phases formed in X7091 at high solidification rates (in excess of 10' K/sec) and their influence on mechanical properties of powder rolled strip. In some instances the formation of harmful constituent phases is Observed in coarse powder . It appears that large powder particles are generally associated with lower solidification rates and coarse particulate microstructures. There are indications of large iron, silicon and zi nc-rich regi ons ina i r atomi zed X7091 powder. Fract09raphy of fatigued specimens revealed that a small amount of inclusions appear to influence the fracture path an'd ductility of the powder rolled X7091 strip.

The yield ing of notched bend bars for a variety of materials has been studied. The plas tic zone growth was directly moni tored using a brittle surface lacquer, and compared to the load- displace ment curves. Two general types of behavior were observed : materials with a Luders strain yielded as p r e di cted by slip line field theory I with the plastic zone beginning at the notch ro ot and rapidly traveling across the specimen in a narrow band . The corresponding load-displacement curve showed a distinct "knee" at general yield . Materials without a Luders s train developed mu ch larger diffuse plastic zones which spread slowl y across the specimen . Their load-displacement curves were smooth, with a g r adually decreasing slope; no pOint of general yield could be reproducibly defined. These results will be compared to literature results, and the effect of such different behaviors on calculations of yie ldi ng and peak te'nsile stresses will be discussed. Research was sponsored by National Science Foundation, the Fannie and John Hertz Foundation and the Association of American Railroads.

3:45 p.m. MICROANAl YTICAL CHARACTERIZATION OF CONSTITUENT PHASES : M. Raghavan and J. Y. Koo, Exxon Research and Engineering Co ., P.O. Box 45, Linden, New Jersey, 07036. Microdiffraction and x-ray microanalytical methods were used to characteri ze compl ex i ntenneta 11 i c compounds in a 11 oys 7075 and 7475. Due to the co-existence of many fine constituent phases which occur in these systems, the use of high spatial resolution techniques become imperative to accurately detennine their chemical composition and crystal structures . Analysis was conducted using both thin foils and extracted residue of the second phase particles. Results show that the E phase dispersoid which was present in three morphologies (rOd, sphere and trianqle), had identical chemical composition, and its chemical formulation is sU9gested to be (AI,Cu,Zn) 19M93.2 (Cr,Til2. The other coarse constituent phases (;:.. O.5~m) were identified to be Al7CU2 Fe,

2:40 p.m. DYrlAr4ICS OF nliN BOUNDARY MOTION IN In- Tl: Manfred Wuttig and Chun-Hung Lin, University of Mi ssouri , Rolla, MO 65401. Large amplitude internal friction experiments in martensitic In-Tl alloys show the phenomenon of autmodulation. In this case, the phenomenon can be interpreted as reflecting the breakaway of twin boundaries from a point defect atmosphere. The analysis of the data in these tenns y ields an activation energy of defe ct segregation of about one half the activation energy of selfdiffusion . .

TMS Fall Meeting 15 3:00 p.m.

The conventional view of disloca~ion-hard particle interaction i8 that dislocations are repelled as a result of misfit strainll

Elastic Interactions in Polycrystalline Alpha-Brass and Their Effect on Slip: K. Hashimoto and H. Margolin, Dept. of Phys. and Eng. Meta~lurgy, Polytechnic lost. of N.Y., Brooklyn, N.Y. 11201.

and IIOdulus differences. Thlo has lead to the fonulation of IIOdela of creep in particie .yotellB which depend on clillb controlled by-pass along the stand-off path. Although the misfit _train. can be .hown to rel.,. rapidly at high


teaperatures, the relaxation of the ..,dulull effect has not been considered explicitly. It ill .hown that, in systellS where the

to the prop. limit. Orientations of grains were determined by the Laue back refl. method and slip systems were determined by trace analysis of pairs of cross slipping systems. 195 slip systems in 67 grains were determined. Of these, 118 were primary and the remainder were either cross or secondary slip. Of the 118, only 31 slip systems had the highest Schmid factor determined from the applied stress. Elastic assistance from surrounding grains were found to assist slip in a given grain. Finite element method analysis, carried out on the 2.4rrun g.s .• indicated that 95% of the slip ~ystem had the highest total resolved shear stress.

This research was sponsored by the National Science Foundation

under Grant IIDMR-08l-8849.

propo.e a baslo for the IIOdeling of creep in OS .y.t ......

4:40 p.m. YIELD STRENGTH OF SPINODALLY DECOMPOSED Fe-3Owt%Cr: * K. Park and L. H. Schwartz, Dept. of Materials Science & Engineer~ Northwestern University, Evanston, IL 60201

3:20 p.m. Slip in a Stress Gradient: Z. Wang and H. Margolin, Department of PhYSical and Engineering Metallurgy, Polytechnic Institute

of New York, Brooklyn, NY

particle .ize 10 _aall, the diffusional relaution of particledislocation repultion can be extre.ely rapid 110 that dislocations can approach the particles. Particle pinning then occurs by elimination of a dislocation seg_nt in the particle boundary. The mechanim of subsequent unpinning is examined and shovn -to require a stress similar to the classical Orovan stress. The competition between pinning and climb bypass are discussed in ter1lll of the particle dispersion, and used to

11201.

Wedge-shaped ~-brass single crystal specimens were subjected to compression loads to produce slip. Although the specimens were oriented for single slip, multiple slip was observed. The slip systems were analyzed stereographically and were compared tu a map of the resolved shear stresses calculated by the finite element method (rEM). A non-symmetric distribution of stress was found and the operating slip systems conformed to the rEM determined stresses. In two cases two systems were found to· operate on a specific slip plane. Cross slip was fuund by stress gradients. Some evidence was obtained to indicate that local stresses produced by the operation of intersec'O: ting slip systems may promise different types of cross slip. This work was supported in part by the National Science Foun-

dat ion under Grant # DMR-081-8849

3:40 p.m. Kink Band Formation in a Stress Gradient: Z. Wang and H. Margolin, Depa rtment of Physical and Eng ineering Metallurgy,

It is now well known that Fe-Cr alloys containing 20-8Owt%Cr exhibit spinodal decomposition upon isothermal heat treatment around 773 K with concomitant increases in yield strength. To study this phenomenon quantitatively, tensile test have been made of Fe-3Owt%Cr as a function of aging time up to 2000 hours, aging temperatures between 758 K-788 K and test temperatures between 77 K-473 K. It has been observed that the deformation mode is mixed slip and twinning for specimens aged 10 hours while twinning deformation dominates for specimens aged longer times over the entire test temperature range. The observed incremental yield stress is 428 MPa for a specimen aged 2000 hours at 773 K, and tested at room temperature. This is more than 3 times larger than an estimation based on the theory of Kato which attributes the incremental yield strength to internal coherency stress effects and modulus effects due to spinodal decomposition.

*

Supported by National Science Foundation Grant DMR79-lll07.

DEFORMATION PROCESSING TMS-AIME General Abstract Session Monday, October 25,1982 Cervantes Convention Center 2:00 p.m. Room 267 Session Chairman: Professor Roger N. Wright, Department of Metallurgy, Rensselaer Polytechnic Institute, Troy, NY 12181.

Polytechnic Institute of New York, Brooklyn, NY 11201. A Wedge shapedo( -brass single crystal specimen with a 14.3 taper and oriented for Single slip deformed in compression revealed a kink band. The kink band occurred in a region of

stress, determined by finite element method (FEM), which was higher than the stress in other parts of the specimen, and was confined to this region. The kink band extended from the bottom compression ~urface to the one edge. Because'the region

of the kink band deformed preferentially, the kink produced was a lead kink. The presence of a gradient in producing a non-uniform distribution of stress, cessive operation of slip at larger thicknesses cimen, as the load is increased. This behavior to promote kink band formation.

addition to promotes sucof the spealso serves

This work was supported in part by the National Science Foun-

dation under Grant # DMR-081-8849

4:00 p.m. AN ANALYSIS OF THE EFFECT OF VOIDS ON UNIFORM ELONGATION: R. Stevenson, Physics Department, General Motors Research Laboratorles, Warren, MI 48090-9055 and A. K. Ghosh Rockwell International, Thousand Oaks, CA 91360 ' Frequently, work hardening is. described using empirical constitutive equations. Implicit in this approach is the assumption that such an describes a homogeneolls material which remains homogeneous untIl It Cractures. However, the ductile fracture oC commercial metals is preceded by the nucleation and growth oC voids. In commerciaf sheet steels, the nucleation sites for voids can be isolated carbides, cementite plates in pearlite or non-metallic inclusions. The deCormation behavior oC such sheet steels is Crequently described by the empirical relation a = K En, where a = true stress, E = true strain, and nand K are material constants. In this study the effect of voids on n measured in a conventional tensile test was investigated. It was determined that if the void volume fraction varies linearly with strain, a relation which has been derived theoretically, then the true strain at maximum load (E T ) is not numerically equal to the n-value but is reduced by a quantity corresponding to the void volume fraction (C v ) (i.e., ET = n-C v )' equ~t!on

4:20 p.m. ON HIGH TEMPERATURES DISLOCATION-INCOHEHENT PARTICLE INTERACTIONS: D. J. Srolovit., M. J. Luton, and R. PetkovicLuton, Exxon Research and Engineering Company, P.O. Box 45,

Linden, New Jersey 07036

2:00 p.m. HOT TORSION OF COMMERCIAL Al-5Mg-0 .8Mn ALLOY; DYNAMIC RECOVERY AND RECRYSTALLIZATION. H.J. McQueen, E. Evangelista*, J. Bowles and G. Crawford. Mech. Eng., Concordia University, Montr~al, Qu~., Canada H3G lM8. *Inst.de Fisici, Univ. of Bologna 40126, Italy. l
2 :20 p.m. HOT l
16 TMS Fall Meeting 2:40 p.m. HOT DEFORMATION BEHAVIOR OF TYPES 304, 316, and 317 STAINLESS STEELS'. N.D. Ryan, H.J. McQueen and J.J. Jonas". Mech. Eng., Concordia University, Montr~al, Qu~. H3G 1M8 . "Met. Eng., McGill University, Montr~a1, Qu~. H3A 2A7 During hot torsion tests in the range 900-ll00°C, 0.1 to 5 s-1 on three types of austenitic stainless steel, the strain for recrysta 11 i za t i on depended on tempera ture T and s tra i n ra tet. The flow stress had exponential relationships with T and f. and was higher for the high Mo Type 317, which had a much greater activation energy. Interrupted deforma t ions were conducted wi.th equa 1 intervals of either 20 or 40 s between repeated equal strains of 0.2. From the fractional softening and comparison of the pass flow curves with that for continuous deformation, the static restoration mechanisms were found to be recovery, recrystallization and grain growth, who se rates increased with c, T and accumulated strain. The duct i 1ity de creased with the total content of Cr and Mo and did not depend on the ',C. High ductility resulted from the impeding of intergranu1ar fissuring by dynamic recrystallization. Static sof tening between passes also improved the ductility. 'Prov ided by Atlas Steels, Tracy, Qu~bec.

3:00 p.m. ON THE NATURE OF SHOCK DEFORMATION IN SI NGLE CRYSTALS OF ALUMINUM:

Narendra Dahotre and K. Mukherjee, Department of Metallurgy, Mechanics and Materials Science, Michigan State University, East Lansing, MI 48824

Nano-seconds duration high-voltage pulse discharge has been produced on single crystal s urfaces of pure aluminum. These discharge experiments have been car ried out at various overpressures of N2 atmosphere. The nature of damage is similar to that produced by a pulse laser ene rgy . The nature of the micro-crater and the associated plastic deformation modes have been analyzed by u s ing SEM and x-ray diffraction. It is observed that immediately after the spark damage, wavy deformation bands and slip traces are present. Upon storage of this damaged material, some of the internal st ress es are relieved by c rack formation. Both radial and circumferential cracks are observed and slip line density decreases. The nature of this delayed cracking and other structura l informations obtained from these experiements will be discussed.

3:20 p.m. ISOTHERMAL FORGING CHARACTERISTICS OF TRANSAGE 175 TITANIUM ALLOY : F. A. Crossley, G. L. Tingley, Lockheed Missiles & Space Co. ,P.O. Box 504, Sunnyvale, CA 94086, C. C. Chen, Chen-Tech Industries, Inc., 10 Autry, Irvine, CA 92714, and J. R. Becker, Cameron Iron Works, Inc., P.O . Box 1212, Houston, TX 77001 Isothermal forging characteristics of Transage 175 alloy (Ti-2.5 A1- 13V -7Sn-2Zr) were detenni ned over the temperature range from 650 to 815 0 C (1200 to 15000 F), employing platen speeds of 2 . 5, 13 and 25 mm/min (0-. 1, 0.5 and 1.0 inch/min) for two starting microstructural conditions. Tension test properties were detennined over the forging temperature range at constant strain rates of 0.005, 0.1, 0.5 and 1.0 min- 1 and related to hot defonnation characteristics. Structural features and room temperature tension test properties of isothennally forged material heat treated to 1240 MPa (180 ksi) minimum ultimate tensile strength are presented . Research sponsored by the Naval Sea Systems COfIIl1and and administered by the Naval Ocean Systems Cornnand .

3:40 p.m. THE CHARACTERIZATION OF PLASTIC DEFORMATION UNDER ROLLING CONTACT: C. Azuamah, V. Bhargava, C.A. Rubin, G.T. Hahn. Dept. of Mecha ni ca l & Materials Engr .• Vanderbilt Universit y , Nashville, TN 37235

Measurements of the accumulated plastic deformation in the rim of a wheel subjected to pure rolling contact have been performed on 5Omm-diameter disks of AISI 4140, AISI 1018 and Fe-5i steel at several strength levels. The work examSnes plastic strains and displacements produced by from 1 to 10 contacts at load factors 3.86< polk < 6, under essentially plane strain conditions . The variations of the net forward displacement and shear strain with depth are detected with imbedded nickle markers. A special etching technique and microhardness measurements also reve al that the extent of the plastically deformed zone below the rIm is about 4 times larger than previously reported. The results, together with conventional cyclic flow measurements. are compared with the cyclic strain predictions of the Merwin and Johnson analysis and a finite element model. The need to invoke cyc lic hardening and so ftenin g to account for the deforrna tion is examined.

4:00 p.m. MEASUREMENT OF THE FLOW STRESS IN THE FLANGE DURING DEEP DRAW: D. N. Harvey, D. H. Andersen, Materials Forming Divis10n, MTS Systems Corporation, Minneapoli~ MN 55424 and D. A. Frohrib, D. R. Riley, University of Minnesot~, Minneapolis, MN 55455.

The flO~ stress in the flange during the deep draw of a highly-cold-worked aluminum alloy, 3004-HI9, was experimentally investigated. Five sets of specimens ranging ~n diameter from 4.994 in. to 3.694 in. have been drawn into 2.810 in. diameter cups. Each draw was stopped at a depth that would leave 0.20 in. of material remaining in the flange • ... The average effective strain in the flange was deter~ined by fiducial marks and direct measurement. Calculation of the ave~ age effective stress in the flange at the instant the draw was stopped was based on punch force data. Corrections for friction and inhomogenious deformation were addressed. Comparisons were made with data from tensile and hydraulic bulge tests.

4:20 p.m. DEFORMATION AND FAILURE AT ISOLATED HOLES IN PLANE STRAIN TENS:f,0N: R. J. Bourcier,t R. E . Smelser,· O. Richmond,- and D. A. Koss, I Dept . of Metallurgical Engineering I Michigan Technological University, Houghton, MI, ·U . S. Steel Corporation, Research Laboratory, Monroeville, PA 15146.

In order to model aspects of void growth and coalescence, plane strain tensile specimens each with a hole oriented in the zerostrain direction have been tested. The study is based on the contrasting behavior of two materials: an HSLA steel, EX-TEN FSO Mod, with a relatively high rate of strain hardening, and the Ti alloy Ti-6Al-4V which possesses a much lower strain hardening exponent. The deformation of the holes and associated necking have been determined; these observations exhibit excellent agreement with predictions from a large scale elastoplastic finite-element model. The HSLA steel fails by ductile tearing across the minimum section, whereas failure of the low hardening Ti alloy occurs by a shear localization process. The implications of this work with regard to the fracture of high strength alloys will be discussed. This program is supported by the Office of Naval Research.

4:40 p.m. HIGH TEMPERATURE DEFORMATION OF CoAl-Co ELASTIC-PLASTIC AGGREGATES: M. Vedu1a, S. S. Fr,ydman, and T. H. Courtney, Dept. of Metallurgical Enqineering, Michigan Technological University, Houqhton, MI 49S31. Two-phase elastic- (the intermetallic compound CoAl) plastic (aluminum stabilized fcc Co) aggregates have been compression tested between 298 and 1373K (.18 'I'm to .82 'I'm). The volume fraction of CoAl was varied between zero and seventy-five percent which allowed for investigation of t he effect of phase morpho logy on mechanical behavior. Significant decreases in yield strength were observed at high temperature. Moreover, yield strengths at 13731< were, regardless of the CoA l volume fraction, on the order of the yield strength of the solid solution matrix. In contrast at low temperatures alloys containing CoAL are considerably stronger than the solid solution matrix. The data suggest that high temperature deformation is conttolled by diffusion processes and the elastic CoAL phase influences yielding only by a geometrical restriction of the diffusive flow pa th. At low temperatures, on the other hand, deformation behavior is related to the constraint the CoAl particles exert on plastic phase flow. The temperature variation of aggregate yield strengths is discussed in tenns of the individual p hase properties and their temperature variation. This work was supported by the National Science Foundation.

GRAIN REFINEMENT IN CASTINGS AND WELDS: THEORY AND PRACTICE II: Ferrous Alloy Castings Sponsored by the TMS Solidification Committee Monday, October 25, 1982 2:00 p.m.

Cervantes Convention Center Room 261 Session Chair17Uln: Bruce McDonald, Code 421, Materials Division, Office of Navel Research, 800 N. Quincey Street, Arlington, Virginia 22217.

2:00 p.m. GRAIN REFINING MECHANISMS: G. Amaya, J. Patchett, and G. J. Abbaschian, Department of Materials Science and Engineering. University of Florida, Gainesville, FL 32611

Grain refining is generally achieved via enhanced nucleatio n caused by the addition of a n active heterogeneous nucleating agent, or by grain multiplication which is induced by appropriate fluid flow and thermal conditions. Grain refining can also be accomplished by the addition of microalloying elements which result in the formation of a new phase particles prior to bulk solidification. To fully understand grain refinement in a casting, it is imperative to identify the rel ative importance of each of the above mentioned mechanisms . 1n the present experiments, levitation melting has been used to s t udy solidification of pure. a nd binary alloy of, Fe, Ni, Cr, and Cu. The effect of va riou s inoculants. microalloying and/or supercooling on the str u ctu re and grain size is investigated.

TMS Fall Meeting 17 2:25 p.m.

By the introduction of less than

NUCLEATION ASPECTS OF GRAIN REFINEMENT:

1. Eo Anderson, Naval Research Laboratory, Washington, D. C. 20375 and J. H. Perepezko, University of Wis co nsin-Madison, 1509 Uoiv. Ave . , Madison, WI

0.'"

of such a

potential

nucleation catalyst I significant refinement of as-cast gr~tn Size has been achieved in industrial alloys of up to 2 x 10 m dia. based on copper or nickel. Grain refinement was obtained in laboratory ingots of ferritic stainless steel.

53706 Although solidication microst ructures are often reproduci bly modified by grain refiners, little experimental evidence could be utilized to study me c hanisms of solidification catalysis in metals. In a large melt, the action 'of residual sites can

The technique is described in detail , with experimental data. The factors which at present appear to set the limits of application are discussed with brief comments on those which may govern the selection of specific strip for each application.

preclude meaningful interpretation of the effe c t of a grain refining agent. However, controlled nuc leation ci"talysis can be studied with fine droplet emulsions, where isolatIon or elimina-

tion of spurious effects can be achieved and verified.

catalysis reactions, including the undercooling for nucleation, of a series of alloy systems have been characterized by DTA, XRD, and metallography. These results are useful in an exploration of structural and chemical fa c tors con trolling solidification c atalysis. Various relationships have been analyzed including dis registry and parameters related to phase stability, electronic

effects and solubility.

4:30 p.m.

The

The support of the NSF (DHR-79-15802)

1s gratefully acknowledged.

LOCALIZED HARDENING OF S. G. CAST IRON BY LASER SURFACE MELTING: W. M. Steen, I. Hawkes and D. R. F. West, Metallurgy Department, Imperial College, London, England The metallurgical effects of surface melting of S.C. cast iron with a laser are described and discussed in relation to the mathematically predicted thermal history of the melt zone. Observations on the flow .ln the melt zone are compared to existing theories, with particular reference to the surface profile.

2:50 p.m. FINE GRAIN, INVESTMENT CAST INTEGRAL TURBINE WHEELS D.E. Macha, G.R. Cole II, Howmet Turbine Components Corporation, 699 Benston Road, Whitehall, MI 49461; and J.A . Butzer, Howmet Turbine Components Corporation, 1110 :':ast Lincoln Way, LaPorte, IN 46350 The Grainex™ casting process has been evaluated in production of integral turbine wheels, producing a uniforml y fine grain structure in Mar-M 247 alloy. Improvements in low cycle fatigue behavior derived with Grainex iH castings are related to the reduced variability in Young's Modulus inherent in the finer grain castings. Fatigue crack growth rate and tensile, creep and stress rupture properties have been evaluated for both Grainex™ 'and conventionally cast rotors. Effects of the casting process and subsequent thermal processing, including hot isostatic pressing (HIP) and post-HIP heat treatment on the microstructure and long term stability of the alloy are also documented.

3:15 p.m. STRUCTURE OF UNDERCOOLED IRON AND NICKEL-BASED ALLOYS:

D. G.

MacIsaac, Y. Shiohara, M. G. Chu, and M. C . Flemings, MIT, Materials ProceSSing Center, Cambridge, MA 02139 A series of iron and nickel base allo y drqplets were solidified in glass. Droplet diameters were varied:. from
3:40 p.m. EFFEct OF COARSENING DURING SOLIDIFICATION ON GRAIN SIZE. S. Skolianos and T. Z. Kattamis, Dept. of Metallut~y, Institute of Materials Science, Univ. of Connecticut, Storrs, ct 06268. Grains in castings are usually dendritic. Coarsening during solidification then controls final dendrite morphology, in particular dendrite arm spacing and has no effect on grain size. In grain-refined alloys grains may be dendritic or non-dendritic with spherical coring depending on nucleation rate and cooling conditions. The grain size of non-dendritic alloys, both grain-refined, as well as highly bulk-undercooled was found to be controlled by coarsening over a wide range of cooling rates reaching those that prevail during rapid solidification processing. The work was carried out on a grain-refined aluminum alloy, a grain-refined magnesium-zinc alloy and a highly undercooled iron-nickel alloy. A model was introduced. for predicting coarsening kinetics of a grain-refined magnesium-zinc alloy during solidification. This model yielded a grain diameter versus local solidification time dependence which was compared with previous experimental results.

4:05 p.m. GRAIN R11I'IIII!I1EIIT BY TIlE INTRODUCTION TO TIlE HIILT 01' OXIDE PARTICLES PRIIDISPI!RSI!D IN 1II!T.t.L STRIP. J. P. Dennison Dept. of Metallurgy & Materials Technology, Un! vers1 ty College, Swansea, U.K.

Particles of metallic oxides of controlled size and dispersion have been produced by internal oxidation of thin strip of suitable dilute alloys based on copper, nickel and iron respectively .

POT ISOSTATIC PRESSING OF METAL AND CERAMIC POWDERS II Sponsored by the TMS Powder Metallurgy Committee Monday, October 25,1982 Cervantes Convention Center 2:00 p.m. Room 265 Session Chairman: K. E. Meiners, Battelle Columbus Laboratories, Columbus, Ohio 43201.

2:00 p.m. MECHANICAL AND MICROSTRUCTURAL PROPERTY EVALUATION OF HOT ISOSTATICALLY PRESSED PLUS HOT DIE FORGED RENE' 95 POWDER: W. H. COUTS, JR. AND M. E. SCHREINER, Wyman-Gordon Company, Worcester Street, North Grafton, Ma. 01536 The mechanical and microstructural properties of HIp'ed plus hot die forged Rene' 95 powder are discussed and compared to as-HIP Rene' 95. Three heats of -150 mesh Rene' 95 powder were HIP'ed at 2050° F to yield forging preforms. The preforms were then forged to approximately 60% reduction utilizing 2050° F and 1700° F as metal and die temperatures respectively. An increase of approximately 40% in low cycle fatigue lifetime is obtained when material forged under these conditions is compared to equivalently heat treated as-HIP material. Fatigue fracture origins were found to be predominantly stage 1 faceting surrounding a nucleating foreign particle for the HIP plus forged material while as-HIP material typically exhibited a stage 2 flat fracture morphology surrounding the nucleating particle. No significant differences in tensile or rupture properties were noted for the two processes.

2:30 p.m. EVALUATION OF Ti-6AI-4V BLENDEO ELEMENTAL POWOER PLATE: P. R. Smith, Materials Engineer Metals and Ceramics Oivision, AFWAL/MLLS, WPAFB, Ohio 45433, C. M. Cooke, Metcut-Materials Research Group, P. O. Box 33511, WPAFB, Ohio 45433, and A. N. Patel, Research Scientist, Battelle-Columbus Laboratories 505 King Avenue, Columbus, Ohio 43201 Previous titanium P/M material made by blending elemental titanium and master alloys followed by cold compactioh and sintering resulted in 95-96 percent of theoretical density with tensile strengths near or equ i va 1ent to HIP' ped compacts produced from prea 11 oyed powders . Unfortunately, the residual porosity caused by high chlorine containing sponge degraded the fatigue properties relative to prealloyed powders. Thi s study i nves t i gated whether a fu 11 y dense, compos it i ona lly uniform Ti-6AI-4V plate could be fabricated from blended elemental powders using low chlorine titanium sponge powder and compared its microstructure and mechanical properties to those of plates similarly produced from prea 11 oyed powders. The pIa tes were fabri ca ted from HIP'ped billets using the same hot rolling schedules. Comparison will be made on the basis of microstructures, microhomogeneity, chemistry, and mechanical properties. Mechanical property evaluation will include: tenSile, S-N fatigue, fatigue crack growth and fracture toughness.

3:00 p.m. EFFECT OF POWDER QUENCH RATE ON THE PROPERTIES OF HIP PROCESSED FERROUS ALLOYS., J. E. Smugeresky, Sandia National Laboratories, Livermore, CA 94550, and R. M. German, Rensselaer Polytechnic Institute, Troy, NY 12181. Several pre-allayed powder-making techniques result in rapid solidification of molten droplets during fabrication. The quench rate varies not only with technique, but also with powder particle size. Several ferrous alloys containing many alloying elemen ts and processed by the r o tating electrode process, inert gas atomization and centrifugal atomization were consolidated by hot isostatic pressing (HIP). Hechanical properties, microstructure and f racture mode of the HIP powde r were irtvestigated. The effect of powder quench rate is dependent upon the alloy composi-

18 TMS Fall Meeting and the HI P parameters. The details of how the properties, microstructure and fracture mode are affected in each case will be discussed.

*This work sup port e d by US DOE, Contract IIDE-AC04-76DP00789.

ISOSTATIC PRESSING O'F PLATINUM-IRIDIUM POWDERS: D. M. Jarboe, Corporation, Box 1159, Kansas City, MO 64141.

long-range diffusion of hydrogen and (iIi) the observation of a substantial internal friction peak whose characteristics suggest it is an example of a hydrogen Snoek peak, modified by the disorder of the matrix. Hydrogenation of some glass compositions causes the peak to rise smoothly to very large values, indicating a capability for bydrogen storage simply by the formation of a concentrated interstitial solid solution.

3:00 p.m. Ali EQUILIBRIUM

S'l;UDY OF VACANCY-HYDROGEN INTERACTION IN Ni USING POSITRON ANNIHILATION SPECTROSCOPY:. M. K. Chason, R. W. Siegel, and M. J. Fluss, Materials Science DiVision, Argonne National Laboratory, Argonne, IL 60439

ABSTRACT NOT AVAILABLE

p.m. HOT ISOST,~TIC PRESSING OF NICKEL ZINC FERRITE: General Produ cts Division, Tucson, Arizona

E. Bertrand Rigby

Significant reduction in pore and void concentration of NiZn Ferrite is effected by hot isostatic pressing of presintered bricks at 30,000 psi at 1150°C for four hours. The final grain size is controlled by variation of the pre-sinter time of the cold isostatic pressed bricks. Post anneal conditions of one atmosphere pressure at 900°C in an oxygen atmosphere for 10 hours reduce stress and enhance the magnetic properties. To prevent furnace element contami na t ion from the ferrite, a protective barrier of alumina sand and titanium shavings was util ized.

HYDROGEN IN SOLUTION IN METALS AND ALLOYS II Contributed Papers Sponsored by TMS Alloy Phases Committee Monday, October 25, 1982 Cervantes Convention Center 2:00 p.m. .' Room 273 Session Chairmen: C. C. Koch, Metallurgy and Ceramic Division Oak Ridge National Laboratory, Oak Ridge, TN 37830; J . E. Hack, Southwest Research Institute, San Antonio, TX 78284.

2:00 p.m.

EFFECT OF !iJJSORBEO SPECIES ON HYDROGEN PERMEATION THROUGH NIOBIUM :* Robert Sherman and H. K. Birnbaum, Materials Research Laboratory and Department of Metailurgy and Mining Engineering, Uni vers ity of Illinois at Urbana-Champaign, Urbana, IL 61801. Surface dependent permeation experiments were performed as part of a continuing effort to understand the processes of hydrogen transportation in niboium. First, we will discuss high temperature diffusion controlled permeation experiments for input hydrogen and deuterium gas. Next, we will discuss the experimental procedur~s used with the adsorption surface and the desorption surface being the rate controlling process. As part of this work, we developed a method to ellctochemically form Pdblack on Nb surfaces to insure barrier-free entrance and exit for H in Nb. The results for as-prepared surfaces and with 0, N, Cl, CO 2 , CO and H2S input gas wi 11 be di scussed to i nvestigate the effect of adsorbed species on surface permeation processes. The role of fjr+ sputtering to clean and maintain a known surface chemistry wi 11 be mentioned. *Thi s work was supported by DOE graht DE-AC02-76ER01l98. 2:20 p.m. SOLUBILITY AND DIFFUSION OF HYDROGEN IN VANADIUM-OXYGEN ALLOYS:* D. T. Peterson and B. J. Schlader, Ames Laboratory and MSE De partment, Iowa State University, Ames, IA 5001l. The effect of oxygen concentration on the isopiestic solubility o f hydrogen in v a nadium has been me asured and does not indicate strong local trapping. The diffusion of hydrogen in these sam~ a lloys has bee n mea sured and will be discussed in connection with current trapping theories.

Hydrogen trapping at vacancies in metals urider nonequilibrium. conditions is known to have important technological consequences. A PAS stndy of high-purity (99.995 wt.%) Ni containing equilibritD concentrations of dissolved H and vacancies is now beng carried out over the temperature range 20-l420°C. The Ni sample is first heated. iIi a UHV system. to a temperature at which a signific.a nt concentration of vacancies is present in solution; subsequently. pure H is introduced into the system ai: a given pressure from a hydride source and the system reaches a new vacancy-H eqtillibrilDD. _6The ,equ_1JibrilD H concentration can thus be controlled from 10 to 10 , overlapping the range of equilibrium vacancy concentrations. Doppler-broadening measurements made before and after the introduction of the H then allow for the study of vacancy-H interaction under equilibrium conditions for the first time. *This work was supported by the U.S. Department of Energy.

3:20 p.m.

A "TRAPPING foIJDEL" OF HYDROGEN IN REFRACTORY' ALLOYS:* C. V. Owen, O. Bllck and T. E. Scott, Ames Laboratory-USDOE, Iowa State Universlty, Ames, IA 50011 Resu lts on the mechani ca 1 propert ies of V-Nb a lloys with and without hydrogen in solid solution are consistent with a "misfit" model of one of the metallic elements in the other. For instance, Nb-rich alloys, with a negative misfit, are found to be much less embrittled by H than the V-rich alloys, having a positive misfit. It is postulated that a negative misfit is attractive to the H, thus giving the grain boundaries less opportunity to reach a "critical concentration", necessary for embrittlement. The results are also in agreement with a recent ly advanced "trap theory of hydrogen embrit t lement" . The implications of these models as to the general problem of H in refractory alloys as well as additional critical experiments will be discussed. *This work was performed for the USDOE, Office of Basic Energy Sciences under Contract No. W-7405-Eng-82. 3:40 p.m.

THE INFLUENCE OF HYDROGEN ON THE BEHAVIOR OF V-10AT.% TI UNDER TENSILE LOAD*~ T.~. Row1and+ and C.V. Owen, Ames Lab., .Iowa State Univ., Ames, IA 50011. +Permanent address : Materials Research Lab. and Metallurgy Dept., Univ. of Ill., Urbana, IL 61801 We report an exploratory study of the effect of diSSOlved hydrogen on the mechanical properties of V-10at. % Ti. Tensile tests were conducted over a region of temperature which incl tided the .ductilebrittle transition. Hydrogen concentrations of 1. 1, 2.1, 4.6, and 4.7at.% were introduced directly from the gas. The data show a marked change in both the reduction in area and total elongation with hydrogen concentration at any given temperature. The elongation is reduced to roughly 1/2 its high temperature value (24%) at 200, 245, '310, and 320K respectively for the above H concentrations; a very similar pattern of dependence on H/M holds for the RA. The deb transition temperature in Kelvins is ldb=170+31.5CH for this alloy within the conc . range investigated; thus, for example, at 300K, CH:4.1at.% causes an approximately 50% decrease in RA from the high temperature value. *Work performed under USOOE Contract W7405-Eng-82. 4:00 p.m.

*Supported by USDOE, Office of Ba sic Energy Sc i ences, Div. of Materials Scie nces, under Contract No. W-7405-ENG-82.

A GEOMETRIC ~DEL FOR HYDROGEN-RICH SOLID SOLUTIONS OF HYDROGEN IN INTERHETALLIC COMPOUNDS:. D. G. Westlake, Materials Science Division, Argonne National Laboratory, Argonne, IL 60439, USA

2:40 p.m.

Many intermetallic compounds are known to react 1flth hydr~gen to form hydride phases. For some, however t HIM ratios as great as 2 can be achieved with the H in solid solution. A geometriC model, which was developed to rationalize observed stolchio1letri~s and preferred H-sites in hydrides of intermetalilc compounds, will be applied to the solid solution ZrV2Rr, where o ' x , 6. Failure to form a hydride phase at any varue of x can be attributed to a "leap-frog" effect: Before the H occupation of a given set of interstices has reached saturation, volume expansion has increased the sizes of other holes andlor lengthened H-H distances beyond their critical lI.inimlD values. Accordingly, the ratio HIM can .. jump" over those H/M values expected for the formation of stoichiometric hydrides t

HYDROGEN IN SOLUTION IN METALLIC GLASSES: B.S. Berry and W.C. Pritchet, IBM t.J. Wat~on Rjsearch Center, Yorktown Heights, N.Y. 10598

Hydrogen absorption by metallic glasses is a widespread phenomenon, which we are investigating by anelastic and dilatational measurements. The formation of an interstitial solid solution is supported by (i) the magnitude of the associated di,latation, (ii) the observation of a Gorsky effect associated with the

TMS Fall Meeting 19 which ..,uld have led to plateaus in the pressure-cOII.position-

temperature diagram. *This Work was supported by the U.S. Deparment of Energy.

4:20 p.m. HYDROGEN VAPOR PRESSURE MEASUREMENTS OVER A PORTION OF THE LUTETIUM-HYDROGEN SYSTEM: P. R. Subramanian and J. F. Smith,

*

Dept. of Materials Science & Engineering and Ames Laboratory, Iowa State University, Ames, Iowa 50011. Hydrogen vapor pressures have been measured over the Lu-H system with the Sieverts t technique and cover the composition range Lu to LuH2 and the temperature range 825-950o C. The measurements have allowed delineation of the phase boundaries for the temperature range D-lOOOoC. Combination of the present data with. earlier measurements show consistency and has allowed evaluation of the enthalpy ~nd entropy of formation of LuH2 as llH~ c -204.8±1.9 kJ/mol and !lSf = 136±2 J/mol deg. The results make dubious the. validity of an earlier suggestion of metastability for LuH2 at temperatures below 170°C.

*Research sponsored by the U. S. Department of Energy, Assistant Secretary for Energy Research, Office of Basic Energy Science, under contract No. W-740S-Eng-82.

4:40 p.m. HYDROGEN-RELATED PHASE TRANSFORMATIONS IN JlJJSTENITIC STAINLESS STEELS:* C.J. Altstetter, N. Narita and H.K. Birnbaum, Dept. of Metallurgy, University of Illinois, Urbana, IL 61801 The effect of hydrogen and stress {strai nJ on the stabi 1ity of the austenite phase in stainless steels was investigated. Hydrogen was introduced by severe cathodic charging and by elevated terrperature ~quilibration with high pressure H2 gas. Using x-ray diffraction and magnetic techniques the behavior of two "stable" type AISI 310 steels and an "unstable" type AISI 304 steel was studied during charging and during the outgassing period following charging. Transformation from the fcc y phase to an expanded fcc phase, y*, and to the hcp • phase occurred during cathodic charging. Reversion of the y* and £ phases to the original y structure and formation of the bcc a structure were examined, and the kinetics of these processes was studied. The y* phase was shown<-to l>e ferromagnetic with a subambient Curie temperature. The y... phase transition was studied after hydrogen charging in high pressure gas, as was the formation of a during outgassing. *Work supported by the Nat i ona 1 Science Foundat i on through Grant No. NSF OMR 77-09808.

LIQUID AND SOLID METAL INDUCED EMBRIT· TLEMENT OF METALS I Sponsored by the TMS Corrosion and Environmental Effects Committee and the ASM -MSD Corrosion and Oxidation Activity Monday. October 25. 1982 Cervantes Convention Center 2:00 p.m. Room 270 Session Chairman: M. H. Kamdar, ARRADCOM, LCWSL, Benet Weap· ons Laboratory, Watervliet, NY 12189.

2:00 p.m. METAL-INDUCED EMBRITTLEMENT: AN HISTORICAL PERSPECTIVE: N. S. Stoloff, Materials Engineering Department', Rensselaer POlytechnic Institute, Troy, New 'York 12181.

The history of research on metal-induced embri ttlement is reviewed, wi th the delineation of several major periods of activity. Early research often emphasize. the properties of the liquid metal environments, i. e., me I ting point, contact angle, and fluid flow. Beginning approximately in the 1950's emphasis was placed on the microstructure of the substrate,wi th models of fracture incorporating dislocation theory replacing theories based on surface energy or other thermodynamic parameters. Decohesion theories were proposed in 1965. Evidence for solid metal-induced embri tt1ement also was provided in the 1960' s, but theoretical progress in explaining this effect was slow. During the past five years attention has been refocused on the mechanism of embrittlement, with considerable controversy arising between advocates of enhanced shear, decohesion and surface energy lowering models. Accompanying this renewed interest have been reports of industrial problems related to metal environments . Recent experiments devoted to clarifying the mechanism of embri ttlement are described.

3:00 p.m. MECHANISTIC AND KINETIC ASPECTS OF EMBRITTLEMENT OF A HIGHSTRENGTH STEEL BY LIQUID I1ERCURY: S.P. Lynch, Aeronautical Research Laboratories, Defence Science and Technology Organi sation, Department of Defence, Box. 4331, GPO Melbourne, 3001, Australia The velocity of sub-critical crack growth (v) in a high-strength, tempered martensitic steel (D6ac) in a liquid mercury environment

depended ma·rkedly on the strength of the steel and the stressintensity fact;or (K). v-K curves exhibited regimes where v was strongly dependent on K and where v was indep.endent of K, and for a given K, v was greater for steels with higher strengths. Metallographic and fracto graphic observations showed that crack growth was predominantly intercrystalline; intercrystalline (and transcrystalline) facets were completely dimpled for steels with 'low' strengths but were comparatively featureless except for isolated tear ridges for 'high' strength material. The mechanisms of embrittlement and the factors controlling rates of crack growth are discussed.

3:20 p.m. SOME FRACTOGRAPHIC OBSERVATIONS OF METALS FRACTURED· IN LIQUID METAL ENVIRONMENTS, J . R. Pickens, W. Precht, and A.R.C. Westwood, Martin Marietta Laboratories, 1450 S. Rolling Road, Baltimore, MD 21227

Currently, two mechanisms of adsorption-dependent liquid metal embrittlement (LME) are considered feasible: 1) adsorption-induced reduction of the cohesive strength of the solid metal ~ and 2) adsorption aided dislocation nucleation of the crack tip, facilitating microscopically ductile crack propagation. Lynch has provided fractographic evidence supporting mechanism 2). For example, he observed microscopic dimples on LME fracture surfaces of AI-Zn-Mg alloys that macroscopically appear quite brittle, and observed high local strains near LME fracture surfaces. In the present work, fractographic studies were performed on single crystals of an Al-Zn-Mg alloy provided by Dr. Lynch, and on single crystals of zinc, wetted with various liquid metals. Dimples were observed in some cases, but not others. These, and other observations, such as dislocation densities near fractured surfaces, will be described in terms of possible mechanisms of LME.

3:40 p.m. CRACK INITIATION AND PROPAGATION IN METALS IN LIQUID MERCURY USING FRACTURE MECHANICS HETHODS: J.A. Kapp, Benet Weapons Lab, LCWSL, USA ARRADCOM, Watervliet, NY 12189

Fracture mechanics is a useful tool in the study of sub-critical crack growth. This paper presents a summary of the results of several recent fracture mechanics studies of liquid metal embrittlement. Specific topics include crack growth measurements under cyclic loading conditions in 6061-T651 alloy embrittled by liquid mercury. The effect of mean stress on fatigue crack growth in a high strength-low alloy steel in mercury is discussed. The results of crack growth studies under static loading conditions at various temperatures in both 6061-T651 aluminum and 70/30 alpha brass embrittled by mercury are also presented. Finally, the results of some experiments on mercury wetted 70/30 alpha brass and high purity zinc in Mode III loading (pure shear) are presented. The results are discussed in relation to transport mechanisms.

PAPER WITHDRAWN A FRi\CIURE MEOlANICS APPIDl\Ol TO LIQUID !\NO SOLID METAL EMlRITI'LD1ENT: H. J. Rack, Exxoo Enterprises, Greer, s. c. 29651, J. W. M.mford an'! T. J. Lutz, San:lia National Laboratories, Albuquerque, N. M. 87185

Traditionally, susceptibility to liquid/solid metal anbrittlement has been assessed by examining what influence a potentially albrittling environment has on the tensile ductility - tmperature relationship. Dtbrittlarent is associated with a decrease in tensile ductility, ccmrencing bel"", the melting point of the albri ttling species or carrier alloy. Jls the tarq:>erature increases, the tensile ductility returns to that expected for the alloy. This suggests that in engineering service anbrittle" ment should not occur as long as tmperatures exceed the regime where the ductility trough is ooserved. IEcent ooservations with AISI 4130 in eutectic Pb-Sn solder indicate that this approach is non-=nservative. These results she>; that while the ductility of 4130 in eutectic solder exhibits the classical ductility trough, the fracture tougtmess does not. This presentation will ronsider these results and relate them to mechanistic and engineering evaluations of liquid/solid metal enbri ttlarent.

4:00 p.m.

EFFECTS OF ALLOY ADDITIONS TO TIN ON THE METAL INDUCED EMBRITTLEMENT OF HIGH-STRENGTH STEEL: M. W. Mucek, Standard Oil Company (Indiana), P.O. Box 400, Naperville, IL 60566, and N. N. Breyer, Illinois Institute of Technology, Chicago, IL 60616. The effects of compos1.-ti:onal changes to the liquid metal on the brittle-to-ductile recovery temperature, TR , of high strength steel wetted with pure tin, tin bismuth, and tin-antimony alloys were investigated. The embrittlement of AlSI 4140 steel heat treated either to a nominal strength of 200 ksi (1397 MPa) or 140ksi (965 MPa) was observed as a loss in ductility and true fracture strength for tensile specimens tested at elevated temperatures when wetted with pure tin and tin alloys. No recovery of ductility was encountered for the higher strength tensile 'specimens tested up to the tempering temperature of the steel whereas as at the lower strength level a recovery of ductility

20 TMS Fall Meeting was found at temperatures which depended upon the composition of the embrittlement agent. The data were examined to reconcile the results with previously reported similar experiments.

4:20 p.m.

EMBRITTLEMENT OF 4340 STEEL BY LIQUID LEAD AND ANTIMONY AND LEADANTIMONY SOLUTIONS: M. H. Kamdar, ARRADCOM, LCWSL, Benet Weapons Laboratory, Watervliet, NY 12181. Single edge notched fatigue precracked steel specimens were tested in fatigue in high purity (99.9999 Pb) liquid lead, lead-antimony solutions and liquid antimony at various temperatures. Steel is embrittled by high purity lead and lead 3W% antimony solution at temperatures of up to 1200-F. This temper.IJ.ture is approximately 400·F higher than the d-b transition temperature for crack initiation in steel in high purity lead and lead-O. 3W% antimony solutions reported in the literature. The susceptibility to embrittlement of steel by liquid lead remained unchangedwhen 3W% antimony was added to liquid lead. However, this embrittlement susceptibility increased some two to three folds when the specimens were tested in 25 and 50W% antimony-lead solutions. The embrittlement susceptibility increased by several orders of IIlilgnitude when the speciIlens were tested in liquid antimony at approximately l300°F. These results will be presented and possible reasons for their occurrence will be discussed.

MODELING OF THE STRUCTURE AND PROPERTIES OF AMORPHOUS MATERIALS II Sponsored by the Computer Simulation in Materials Science Activity of ASM and by the TMS Nuclear Metallurgy Committee Monday, October 25, 1982 Cervantes Convention Center 2:00 p.m. Room 272 Session Chairmen: H. S. Chen, Bell Laboratories, Murray Hill, N.J. 07974; S. Takeuchi, The Institute for Solid State Physics, The University of Tokyo, Roppongi, Minato-ku, Tokyo 106, Japan.

2:00 p.m. TOPOLOGICAL STRUCTURE OF GLASSES: N. Rivier, Blackett Laboratory, Imperial College, London SW7 2BZ, U.K. The topological or configurational degrees of freedom in continuous, random struc tures like glasses or supercooled liquids, can be identified as uniterrupted line defects of a particularly elementary kind. The line defects and their consequences (condensation of free defects at a temperature T where the viscosity diverges and the topological entropy isOnegligible, tunneling modes at low temperatures where the defects are: frozen in) will be exhibited in discrete models (continuous random networks or random close packings) and in continuous, macroscopic models (continuum elasticity theory - in which they are disclinations - or gauge invariant field theory) where their existence i~ associated with the overall, yet non-constructive, homogeneity of the glass.

2:30 p.m. TOPOLOGY OF DISORDER IN AMORPHOUS MATERIALS. M. Kleman, Laboratoire de Physique des Soli des ; Bat. 51 0; 91405 Orsay, France. Ideal amorphous systems, like continuous random networks and dense random packings, can be classified as specific mappings of ordered lattices in spaces of constant curvature onto the usual 3-dimensional euclidean space. Mappings of a spherical space introduce naturally surface defects, while mappings of spaces of negati ve curvature introduce line defects (disclinations). Some review will be given of this topology and some consequences which can be expected from this model concerning elementary excitations will be presented.

3:00 p.m. AMORPHOUS STRUCTURES DESCRIPTION STARTING FOR ORDERED STRUCTURES IN CURVED SPACES. THE TOPOLOGICAL DEFECTS INDUCED IN THE REAL SPACE: J. F. Sadoc, Laboratoire de Physique des Solidest Universite de Paris-Sud, Centre d'Orsay, Batiment 510, 91405 Orsay, France The local order of an amorphous structure can be easily defined and used to define as structural unit. The packing of structural units leads to an amorphous structure if it is impossible to arrange it in a regular way in order to fill all of space. Nevertheless a regu;1ar arrangement can be obtained if the space is curved. This allows the description of the order in amorphous structure and the definition of topological defects arising where the curved space is forced into the real euclidean space. Defects can be of different types: -2.D defects: internal surfaces -l.D defects: disclination lines The occurence of defects is necessary, but a competition between defects of different types leads to different structures with the same local order.

3:30·p.m. - Break

3:45 p.m. LOCAL STRUCTURAL FLUCTUATION AND PROPERTIES OF AMORPHOUS METALS: T. Egami and V. Vitek, Dept. of Materials Science and Engineering, University of Pennsylvania, Philadelphia, Pa. 19104 A large number of the physical properties of amorphous metals are governed not only by the average structural parameters such as the density and the average interatomic distance, but are strongly affected by the degree of local structural fluctuations. The relevant local structural fluctuations, however, are not purely topological fluctuations, but geometrical fluctuations or distortions of the environment of each atom. Such fluctuations can not be adequately described by the hard sphere DRP model, and the use of the soft sphere model is essential. We present a method to characterize these fluctuations i tl terms of the local stresses and strains, and apply the method of the analysis of a DRP structure. We show that such a view point leads to a satisfactory elucidation of the structural relaxation, glass transition, mechanical deformation and stability of amorphous alloy.

4:15 p.m. STRUCTURE-PROPERTY RELATIONSHIPS IN METALLIC GLASSES: W. L. Johnson, Keck, Laboratory, California Institute of Technology, Fasadena, CA 91125 The structure of metallic glasses has been described in terms of continuous random packings of atoms and more recently in the terms of packings of molecular units. There is recent evidence that some metallic glasses may consist of still larger structural units or domains with two or more distinct types of domains in a given metallic glass. In this later picture, the metallic glass would be inhomogeneous on a scale of distance comparable to the domain size. The above structural models have important implications for the electronic, magnetic, and superconducting properties of metallic glasses. Several examples of such structure-property relationships will be given.

NOVEL TECHNIQUES IN METAL DEFORMATION TESTING II: Stress/Strain Measurements Sponsored by the TMS Shaping and Forming Committee and the ASM Process Modeling Activity Monday, October 25, 1982 Cervantes Convention Center 2:00 p.m. Room 120 (First Floor) Session Chairman: S. M. Wolf, U.S. Department of Energy, Division of Material Science, Washington, D.C. 20545.

2 :00 p.m. NEW PROCEDURES FOR DETERMINING STRESSES WITH X-RAYS:* I.C. Noyan, Dept. of Materials Science & Engineering, The Technological Institute, Northwestern University, Evanston, Illinois 60201 There is a major assumption in the classical determinations of stress by x-ray methods: in the surface layers penetrated by the x-ray beam only a bi-axial stress tensor is postulated. ' This has recently been disproved. Normal (0" 3) and shear (0" ,a ) of appreciable magnitude in the direction of the surfa~~ no~i6al h,ave been detected in ground iron and mild steel. Theory, as we11:as experiment, shows that if shear stresses a 3' a 2 are present in the surfa:e la~ers, lattice spacings "d l l measared at positive and negat1ve t1lts (w) of the sample to ~~e x-ray beam are different, causing "splitting" in lid" vs sin.2*. When only a is presen in tbe near surface region, curvature occurs in t~~ lid" vs sin. relationship, but no II V"splitting". The "linearity" of lid" vs s102 is not a good measure of the dimension of the stress tensor existing in the surface layers. Analysis of lid" VB sinat with very small curvature (the correlation coefficient of a straight line fit being .99 or better) may lend to errors of up to 50 pet . in the surface stress when the contribution of 0'33 is neglected. Also, when steep stress gradients exist for any component of the stress tensor, the ~ range usually employed [ljIe (0, 45 0 )J may produce appreciable errors, even if the average value of a is taken into account in the calculations. Simple methods have 5~en deti\ed which will minimize these errors, and these have been tested by computer simulation.

i

5

*

*This research was support'ed by ONR.

2:20 p.m. RESIDUAL STRESS DETERMINATION IN INCONEL-600 TUBES USING ELECTROCHEMICAL MACHINING": Raghavan Srinivasan and Craig S. Hartley, Department of Materials Science and Engineering, State University of New York, Stony Brook, N.Y. 11794. A new technique, based on the Sachs boring out method, for determining residual stresses in thick wall tubes is described. The inner surface of a tube of Inconel-600 is machined using ECM techniques to ensure uniform removal of material. Strains measured on the outer surface, using foil strain gauges, are used to calculate 'the variations of axial , circumferential and radial stresses with position along the thickness of the tube, using

TMS Fall Meeting 21 Sachs I equations. Electro-chemical machining ensures stress-free removal of material. This technique is, therefore, mOre accurate than earlier mechanical machining techniques. ~ Results of stress distributions in cold-worked and annealed Inconel-600 are presented.

*Sponsorship of the work by the National Science Foundation, Metallurgy Program and provision of laboratory facilities by the Department of Nuclear Energy, Brookhaven National Laboratory, is hereby acknowledged. 2:40 p.m. ACOUSTOELASTICITY MEASUREMENTS OF SPECIMENS WITH CONTROLLED RESIDUAL STRESSES * Wolfgang Sachse and R. H. Lance, Department of Theoretical and Applied Mechanics, Cornell University, Ithaca, New York - 14853. Based on the acoustoelasticity phenomenon, precISIon ultrasonic wavespeed measurements are potentially a viable means for determining the residual or applied stresses and their distribution in a part. However, even in specimens possessing essentially two-dimensional stress states, difficulties arise in implementing such techniques. These are associated with the measurement resolution of the signals' time-of-flight, amplitude and spatial features. In addition, there are material microstructurerelated wavespeed effects and superimposed plastic deformation effects. This paper summarizes recent work which has overcome, at least in part, some of these difficulties. The results of experiments will be described in which acoustoelasticity measurements are made on shrink-fit specimens of aluminum and mild steel possessing various microstructures and fabricated with different interference fits to result in various twodimensional elastic/plastic stress distributions. Comparison is made between the results of the ultrasonic measurements and those obtained from a destructive technique.

* This work was supported by the National Science Foundation through a grant to the Materials Science Center at Cornell University. 3:00 p.m. A MINICOMPUTER-CONTROLLED LASER-BASED SURFACE STRAIN MEASUREMENT TECHNIQUE: W. N. Sharpe, Jr., Mechanical Engineering Department, Louisiana State University, Baton Rouge, LA 70803 Surface strain can be measured by monitoring the interference patterns emanating from two reflective identations illuminated with a laser. The indentations are applied with a Vicker'S microhardness tester. Fringe monitoring is accomplished with a minicomputer-based system which provides a strain output that can be used for data acquisition or control. The technique can measure over short gage-lengths (-100 micrometers), into the plastic strain region, at temperatures up to 732"c on superalloys, and at sampling rates of 10 per second. This paper briefly reviews the measurement system and describes recent strain-rate controlled tests on a superalloy at 650"c. 3:20 p.m. SOME OPTICAL TECHNIQUES OF DISPLACEMENT AND STRAIN MEASUREMENTS ON METAL SURFACES: F. P. Chiang, Department of Mechanical Engineering, SUNY at Stony Brook, Stony Brook, N.Y. 11794. In the pa~t decade many new optical techniques of experimental stress analysis have been developed, some of which are directly applicable to surface deformation measurement in metals. The paper reviews three groups of these techniques, namely: the moire methods, the laser speckle methods, and the white light speckle methods. The moire methods are not new, but their full potentials have not been realized. Emphasis will be on the methods I ability to measure large in-plane and out-of-plane deformations. Laser speckle methods were mainly developed in the seventies. The so-called laser speckles, created when an optically rough surface is illuminated by a coherent laser beam, are the result of random interference among reflected wavelets. They can be used as gauging elements for displacement and strain measurements. White light speckles are artificially created, or naturally present, surface random patterns. They can also be used for displacement and strain measurements. Advantages and disadvantages of these techniques will be compared.

3:40 p.m. DETERMINATION OF SMALL SURFACE DISPLACEMENT CHANGES USING MODIFIED OPTICAL INTERFEROMETRY TECHIQUES: J. Wayne Jones, Department of Materials and Metallurgical Engineering, University of Michigan, Ann Arbor, MI 48109 Two techniques for measuring out-of-plane displacement changes ,resulting from in-plane deformation of materials containing geometric continuities such as notches or cracks is described. Both techniques use standard optical measurement tool. In one technique, however, analog and video recording equipment is coupled with the interferometric system to allow determination of displacement changes as a function of load, strain or time. In the second technique, interferograms are used to produce moire grids which are then combined as an optical filtering system to allow measurement of minute displacement changes which occur over rela-

tively large areas. The use of both modified interferometric techniques is illustrated by the determination of the out-ofplane displacement field of a propagating fatique crack. It is shown that displacement changes of 3 x 10- 7 m can be easily made using either technique. With "real-time" measurements, cyclic plastic deformation fields surrounding crack tips is determined.

4:00 p.m. HOLOGRAPHIC INTERFEROMETRY FOR EVALUATING LARGE DEFORMATION IN SHEETS: M. L. Devenpeck, U. S. Steel Corporation, Research Laboratory, Monroevi:).le, PA 15146

The technique of holographic interferometry, whi"ch enables the observation of ,minute motion of a body, has been applied to observing the deformation of sheet tensile specimens during mechanical testing. Apparatus and procedures are described, and examples using two modes of generating holographic interferograms are given. Concomitant examination enables the observation of deformation as it occurs relative to some reference state, but requires interruption of the deformation at the reference state. A do'uble-exposure technique allows continuous deformation of the tension specimen, but requires the preselection of two obser:vation times with the deformation occurring between them being fixed and observable only after the test is completed. The patterns of the interferograms correspond to contours of equal rate of thickness change from 2 x 10-6 to 200 x 10-6 with a resolution of 0.1 ~m over 40 cm2 • The retreat of the plastiC zene ioto a narrow ne~k is illustrated.

4:20 p.m. AN OPTICAL CORRELATION TECHNIQUE FOR DEFORMATION TESTING: Michael C. Montpetit, David Sigler, and ~;. L. Haworth, Department of Metallurgical Engineering, Hayne State UnlVersity, Detroit, MI 48202. Optical correlation measurements can be used to evaluate elastic and plastic strain in metals. In this technique, a selected area of the metal surface is illuminated by coherent light. The optical correlation intensity (Ic) is then measured by transmitting light scattered fran the surface through a holographic filter, in which infonnation about the s~rface topography at an earlier time is stored. Topographic changes arising from rigidbody displacement, elastic strain, or plastic strain cause corresponding changes in Ie: The changes arising from rigid-body motion can be compensated tor, yielding a corrected correlation intensity I C • In this paper we present an analytical model of the holograBhic reconstruction process and show how this can be extended to predict I~ as a function of strain in uniaxial tension for various degrees ot surface roughness. The optical correlation intensity was measured as a function of elastic and plastic strain for an age-hardened aluminum alloy; the experimental results are compared with those of the foregoing analysis. The results demonstrate that the relationship between correlation intensity and strain depends strongly on the initial surface roughness of the materi al under test in the case of elastic defonnation, but only weakly in the case of plastic deformati on. *This work is supported by the U.S. Anny Research Office.

OXIDATION AND COATINGS TMS-AIME General Abstract Session Monday, October 25, 1982 Cervantes Convention Center 2:00p.m. Room 264 Session Chairman: Dr. J. L. Smialek, NASA Lewis Research Center, MS 49-1, 21000 Brookpark Rd., Cleveland, OH 44135.

2:00 p.m. EARLY STAGES OF OXIDATION OF. Fe-Cr ALLOYS:* P.A.Labun, K. Kuroda G.Welsch and T.E.Mitche11, Department of Metallurgy and Materials Science, Case Western Reserve University, Cleveland, Ohio 44106 The growth of oxides on Fe-Cr alloys with Cr contents of 3, 9, 12 and 18wt. % have been investigated both in-situ and ex-situ wit~ e1ectr~~ microscopy. Oxidations were carried out at 700800 C in 10 torr of air or 20 torr of oxygen for 5 seconds, 5 minutes and 30 minutes. A Philips EM400T electron microscope equipped with an energy dispersive x-ray detector, a secondary electron· detector and a back-scattered electron detector were used to study the microstructure and microchemistry of the oxide scales. Fe 301t containing trace amounts of Cr was observed to nucleate on the Fe-3% Cr specimens while M30 .. (M = Fe,Cr) and/or Y-M203 nucleated on the higher Cr specimens. The initial scale (5 seconds) formed at 20 torr on the higher Cr-al1oys consists of transient a-Ml03 oxides that are enriched in Fe.

*

This research was supported by a grant of MRL-NSF.

22 TMS Fall Meeting 2:20 p.m. OXIDATION OF A FE-3% CR ALLOY AT HIGH TEMPERATURE: * A. 1. Kahveci, and Gerhard Welsch, Department of Metallurgy & Materials Science, Case Western Reserve University, Cleveland, Ohio 44106. Theooxidation kinetics of an Fe-3% Cr alloy were investigated at 800 C in dry air at atmospheric pressure. Weight gain was continuous ly recorded by a Caho 1000 electrobalance. X-ray diffraction and scanning e lectron microscopy were used to examine the microstructure of the oxide scale cross-sections and of the oxide su rf ace morphology. Microchemical investigation of fractured cross-sections was carried out by EDAX. The alloy displayed two distinct oxidation kinetics: initial fast oxidation following a parabolic rate law and a subsequent slower growth rate that was less than parabolic. Four-layer oxide scales were formed con-

sisting of a (Fe,Cr)304 spinel layer, a (Fe,Cr)Z03

~~~~u~t~r~I~~na ~i~~~t,+_5S ~o~:: ~~~f toS de~~~a~~e~ni~~~!!~~ :;gain with an increase in the concentrllio6 -of Ni3+XS2o The activation energy for conduction also follows the same- pattern. The type of predominant carriers was determined by the hot point method. The conductivity changes from predominantly p type for pure NiO to predominantly n type for compositions equal to or greater than Smole% Ni 3+ v S2. The sharp decrease in conduction in NiO containing Smole ~f Ni 3+xS2 has been explained by postulating the localized trapping of troles at the interface by the electrons from Ni3+xS2. At higher concentrations of Ni3+xS2' conduction is predornirnrntly due to the electrons from Ni3+:S"') phase. Pure Ni3+xS2 is found to have a very high conduQ.6.fv~ty. Change over ~~~e~~~~gg~C~tO~i t~ ~ 2h~t~~te~O~~~~t~~ge b~~tvtgr 2~~lrid at Ni3+xS2o This is muc~igher than predicted~y the percolation theory. The possible reasons are discussed u

layer

and two outer Fe 2 0 3 layers. The layers vary in morphology and in composition. The outer two Fe 2 0 3 layers are clearly distingUished by their different morphologies. Details of the growth of the oxide layers and their morphologies as a function of oxidation time will be discussed.

*This

research was sponsored by a NSF-MRL contract . A. 1. Kahveci acknowledges support through a stipend by the Government of Turkey .

2:40 p.m. THE EFfECT OF ALUMINIDE PROCESSING VARIABLES ON OXIDE STRUCTURE* D. H. Boone, N.. L. Lambertson, Materials and Molecular Research Division, Lawrence Berkeley Laboratory, University of California, Berkeley, California 94720 Recent studies have shown differences in the surface structure of protective oxides formed on aluminide coated superalloys. These differences and resulting differences in protectivity are

related to the type of coating, inward and outward; method of CVD deposition, in and out of the pack; and the presence of additional elements and particles both from the coating process and from the substrate itself. A study was initiated, using the deep e tching technique, to examine the differences in the oxide structure at the oxide-coating interface produced by these coating processing variables. Ini tial results of this study wi 11 be presented. *Work supported by Foss il Energy Office, D.O.E.

3:00 p.m. LOW TEMPERATURE (700°C) CORROSION TESTING OF MODIFIED ALUMINIDE COATINGS: * D. H. Boone, B. R. Rose, Department of Mechanical Engineering, Naval Postgraduate School, Monterey, Calif 93940 The protection provided by simple aluminide coatings on super-

alloys can be rather limited in low temperature (700°C) hot corrosion (LTHC) type environments. MCrAIY type overlay coatings are often used, therefore, to provide the required protection on critical gas turbine airfoils operating in this regime. Intere s t continues for reasons of economy and simplicity in the possibility of modifying the composition and/or structure of the aluminides for improved 700°C hot corrosion resistance. Results of an initial study of the effect of various addition elements such as Si, Cr, and Pt on the LTHC resistance of aluminide coated nicke 1 base superalloys wi 11 be presented.

4:00 p.m. STRUCTURAL VARIATIONS IN PLATINUM MODIFIED ALUMINIDE COATINGS* L. J. Purvis, D. H. Boone, Materials and Molecular Research Division , Lawrence Berkeley Laboratory, University of California, Berkeley, California 94720 Aluminide coatings have been the mainstay of protection systems on superalloys used in gas turbine s . Interest in improving their performance· has led to us~ of addition elements such as t.he noble metals, platinum, rhodium, etc. The mechanism(s) by which improvements are confered or the specific range of improvements are not understood and form the basis for a research effort at

LBL

and the Naval Postgraduate School.

Results of the

initial

study of the structural modification possible in the basic platinum modified aluminide, phases present and distribution, will

be presented. 'Work supported by Fossil Energy Office, D.O.E.

4:20 p.m. THE SUBSTRATE Hf EFFECT IN ALUMINIDE COATINGS : J. R. Exell, D. H. Boone, Department of Mechanical Engineering, Naval Postgraduate School, Monterey, California 93940 The effect of substrate Hf level and aluminide coating type on oxide s tructure and protectivity was studied as part of an ongoing investigation of critical factors affecting coating protectivity. Both inward and outward type coatings were applied to IN 738 modified with 0, I and 2%Hf. Initial furnace testing was conducted under conditions producing low temperature hot corrosion attack. Hf , was found to be a beneficial substrate

addition to IN 738, a larger effect being found with inward type coat ings.

4:40 p.m. OXIDATION AND SULFIDATION RESISTANCE OF CHROMIUM FREE STAINLESS STEEL SUBSTITlTfES IN THE IRON-MANGANESE-SILICON-NICKEL SYSTEM: M. J. McNallan, J. M. Oh, Y. H. Chung, Department of Materials Engineering, University of Illinois at Chicago Circle, P .O. Box 4348, Chicago, I L 60680 The high concentration of world chromium reserves in politically unstable regions has led to an interest in developing chromiumfree corrosion resistant metals. Silicon is potentially an attractive substitute for chromium because it is common, inexpen-

sive and forms a protective SiOZ film in both high temperature , and aqueous environments, but most iron-silicon alloys are too

*This work supported in part by the Office of Naval Research

3:20 p.m. CORROSION RESISTANT OXIDE SCALES ON CllS ALLOYS: T.A.Ramanarayanan, M.Ragha van and R,Petkovic-Luton, Exxon Research and Engineering Company, Corporate Research-Science Laboratories. Linden, NJ 07036 The presence of rare earth oxide dispersoids in Fe based alloys has been s hown to improve the adherence of..(-A1203 sca les formed on such alloys during ox idati on in dry oxygen in the temperature range, 1000-1200 C. The A1203 scales ha ve a fine-grained co lumna r morphology with mic r ovoids dispersed along ox ide grain boundaries. The scale is s hown to grow by inward oxygen transport via the ox id e grain boundary. The yttria dispersoid particles, upon incorporation into the alumina scale transform to yttrium alum·inurn garnet (YAG). The ox id e scale saturated with YAG shows improved properties. The mechanistic aspects of oxide scale growth and dispersoid incorporation into the sc ale are discussed with attention to the influence of the environmental oxygen press ure.

3:40 p.m. ELECTRICAL CONDUCTION IN TWO PHASE NICKEL OXIDE-NICKEL SULFIDE

MIXTURES V.. BD Tare/J . B.Wagner,Jr., Center for Solid State Science Ar izona State Universi ty, Tempe, AZ 85287 Electrical Conductivity of sintered pellets of NiO-Ni3+xS2 mixtures of various compositions has been mei'isured as a f'tIIlction of temperatures in the range of 6000 C to BOOoC . Electrical conducti vity of pur e NiO initia lly heated in a known sul fur vapor at 9500C for 150 hours has b een also measured as a function of temperature ..

brittle to be formed by processes other than casting. A number of formable, austenitic alloys have been produced by alloying iron and silicon with manganese and nickel. The resistance of several such alloys to attack by oxygen and sulfur dioxide in the temperature range of 600°C to 900°C is described and compared with the performance of conventional stainless steel .

PHASE TRANSFORMATIONS II TMS-AIME General Abstract Session Monday, October 25, 1982 Cervantes Convention Center 2:00 p.m. Room 263 Session Chairman: Dr. J. A. Wert, Rockwell International Science Center, 1049 Camino Dos Rios, Thousand Oaks. CA 91360. 2:00 p.m. LOCAL AT~C ARRANGEMENTS IN THE SOLID SOLlTfION, AI-!. 7 AT PCT Cu, AT 790 K:* E. Matsubara and J. B. Cohen, Dept. of Materials Science & Engineering, Technological Institute, Northwestern University, Evanston, Illinois 60201 While much research has been directed to understanding the nature of the plate-like G. P. lones that form after quenching in this system, little is known about the solid solution at the usual solution temperature, prior to quenching. In this study, diffuse x-ray scattering was measured at 7900 K in an extended volume in reciprocal space, on an absolute scale, and separated into the contribution due to local order and that due to atomic displacements. The results indicate that there is clustering of Cu atoms even at this high temperature and at this low CU concentration.

TMS Fall Meeting 23 A computer simulation was made with the x-ray results, wilha threedimensional 32,000 atom model, and this was compared to a nearlyrandom array. There is a clear tendency in the alloy for the Cu atoms to segregate on to {lOO) planes; about 30 pct of the Cu atoms are arranged in this manner. Fur"t hermore, the density of these planar configurations is sufficient to serve as nucleii for zones upon quenching. Information will also be presented on the interatomic displacements. *'l1tis research was sponsored by the Materials Science Division of NSF.

2:20 p.m. FACETING OF GP ZONES IN AI-Ag ALLOYS: K. B. Alexander, F . K. LeGoues, H. L Aaronson and D. E. Laughlin, Department of Metallurgical Engineering and Materials Science, Carrtegie-Mellon University, Pittsburgh, PA 15213. GP zones in Al-Ag alloys were shown by TEM to be faceted spheres. Facet planes are {Ill} and {lOO} at lower aging temperatures and primarily {Ill} at higher temperatures. Prominence of faceting dec~eases with rising temperature. These observations were interpreted on the assumption that they represent reasonable approximations of equilibrium shape. A discrete lattice plane analysis recently employed to deduce the equilibrium shape of precipitates in a regular solution miscibility gap formed the basis of the interpretation developed for the present results. Because of the asymmetry of the coherent miscibility gap in Al-Ag, however, the regular solution model was replaced with one of non-regular type. This modified analysis successfully accounts for the wider range over which faceting occurs in Al-Ag alloys relative to that predicted by the regular solution modeL This research was supported by grants from the NSF Division of Materials Research.

2:40 p.m. PRECIPITATE GROWTH IN AL-L1 ALLOYS, James H. Kulwicki and Thomas H. Sanders, Jr., School of Maten a 1 s Eng i neeri ng, Purdue University, West Lafayette, IN 47907. The Al-Li is an excellent system to model the coarsening behavior during artificial aging. The microstructure of an Al-Li alloy when solution heat treated and aged belo., 6 ' (Al,Li) solvus is characterized by a homogeneous distribution of coherent spherical 6' preci pi ta tes. Once the nuc 1ea t i on process has occurred, the growth rate appears to be governed by the transport of solute. Con?equently the particle size/distribution continuously shifts to larger sizes following a t' 3 relationship. Particle size distribution functions were determined at a variety of aging conditions. The experimental distribution functions will be related to the distribution functions generated by current models. NAVAIR Contract No. NOOOI9-81-C-0471

3:00 p.m. DISCONTINUOUS COARSENING OF LAMELLAR CELLULAR PRECIPITATE IN AN AL-29 AT. PCT. ZN ALLOY",: R.A. Fournelle and C.P. Ju, Department of Mechanical Engineering, Marquette University. Milwaukee, WI 53233 The morphology and growth kinetics of cellular precipitation and

size distribution of the beryllides in the alloys were determined and their effect on the various phase transformations studied.

3:40 p.m. STRESS-INDUCED Y ~a TRANSFORMATION AND CHANGE IN MAGNETIC PROPERTIES OF Fe PARTICLES IN A Cu-Fe SYSTEM: Can Toksoy, G. Huang, W. Pratt, J .C.L. Pak, Hasaharu Kato and K. Mukherjee, Department of Metallurgy, Mechanics and Materials Science, Michigan State University, East Lansing, MI 48824. The small (200 to 2,000 i.. diameter) y-Fe particlea in a Cu matrix can be transformed to CI-Fe martensitically if external stress is applied. In this study, Single crystals of Cu-l%Fe with spherical Y-Fe particles were deformed in tension and a relationship between particle Size, amount of deformation and transformation was investigated by utilizing tranmission electron microscopy. As a supplement to the TEM studies, a superconducting suaceptometer (SQUID) was used to measure magnetization of the specimens containing Y aOO/or a. particles over the temperature range of 5 K to 300 K. These studies are aimed at understanding the magnetic interactions between small particles of ferro-, para- and antiferro magnetic states. The role (if any) of magnetoelastic energy on the stabilization of the metastable Y phase was also examined.

4:00 p.m. METASTABLE Y-Fe COHERENT LAYER DEPOSITION ON SINGLE CRYSTAL Cu SUBSTRATE: J.C.L. Pak, H-r. Pak, Masaharu Kato and K. Mukherjee, Department of Metallurgy, Mechanics and Materials Science, Michigan State University, East Lansing, MI 48824. It is well known that coherent Y-Fe clusters can be precipitate in a single crystal Cu matrix. This metastable v-phase can subsequently be transformed to the equilibrium Cl-pbase via a martensitic reaction induced by an external stress. The stability of the y-phase at room temperature is believed to be associated with an excellent lattice parameter maching between f.c.c. Cu and f. c. c. y-Fe. Coherent "V-Fe layers up to about lsA thick have been produced by vapor deposition of Fe on Cu and it has been suggested that coherency depends on crystal orientation and the film thickness. We have found that y-Fe could be stabilized in layers as thick as 100,1. and this stabilization is independent of the substrate orientation. From this results, the nature of a coherency can be estimated and can be compared with the martensitic transformation in the bulk alloy of Cu-Fe. TEM and related studies will be presented.

4:20 p.m. AN ELECTRON MICROSCOPY INVESTIGATION OF THE LONG-TERM AGING CHARACTERISTICS OF "HASTELLOY". ALLOY X: H. M. Tawancy, and S. J. katthews, Technology Department, Cabot Corporation, 1020 W. Park Avenue, Kokomo, IN 46901 Electron m"iCr05copy was used to characterize the precipitates that form in HASTELLOY alloy X after aging for up to 16;000 hours at various temperatures. The corresponding room temperature tensile properties were also determined. Aging at 811 K (lOOO"F) caused the precipitation of H23C6 carbides. A mixture of M23C6 carbides and sigma-type phases was observed after aging at both 922 K (1200"F) and 1033 K (1400"F). After aging at 1144 K (1600"F), the precipitates found were H23C6 carbides and mu phase.

discontinuous coarsening of the cellular precipitate in an AL-29

6

at. pet. Zn a1 0y has been investigated at temperatures ranging from 50 to 250 C by light, electron microscopy and x-ray diffraction. At all aging temperatures, the alloy was observed to decompose completely by a cellular precipitation reaction which resulted in a fine lamellar structure of aluminum rich and zinc rich solid solutions. The first cell lamellar structure was subsequent 1y decomposed at all ag j n9 temperatu res by a second eel lular or discontinuous coarsening reaction. The discontinuous coarsening reaction occurred at a much slower rate than the first cellular reaction and resulted in a much coarser lamellar structure. Analysis of the kinetics of both the cellular reaction and the discontinuous coarsening reaction suggests that }"hey are controlled by grain boundary diffusivity. "This work was supported by NSF Grant No. DHR-7915218.

.HASTELLOY is a registered trademark of Cabot Corporation.

RADIATION DAMAGE ANALYSIS FOR FUSION REACTORS I: Damage Production Sponsored by the TMS-AIMEIASM Nuclear Metallurgy Committee Cervantes Convention Center Monday, October 25, 1982 Room 271 2:00p.m. Session ChaimUln: D. Parkin, U.S. Department of Energy, Office of Basic Energy Sciences, Washington, DC 20545.

3:20 p.m. METASTA8LE PRECIPITATION IN TERNARY Cu-Ni-Be AND Cu-Co-Be ALLOYS: A. Guha, Brush Wellman, Inc., 17876 St. Clair Avenue, Cleveland, OH 44110, K. B. Alexander and D. E. Laughlin, Carnegie-Mellon University, Department of Metallurgical Engineering and Materials Science, Pittsburgh, PA 15213. The metastable precipitation reactions were studied in a CuNi-Be (0.41 w/o Be, 1.90 w/o Ni, 0.10 Co, balance Cu) and a Cu-Co-Be (0.60 w/o Be, 2.53 W/o Co, 0.01 Ni, balance Cu) alloy by transmi ss i on electron mi croscopy and d i fferenti a 1 scann i ng calorimetry, and compared to the binary Cu-Be alloy. The physical properties, e.g., tensile strength, hardness and electrical conductivity, were evaluated as a function of the precipitation hardening temperature and microstructure. From these results the effect of nickel on the G.P. solvus was determi ned and compared to tha t of cobalt. The G. P. sol vus was estimated to be slightly higher in the nickel containing alloys than in the alloys containing cobalt. The composition and the

2:00 p.m. THE ROLE OF RADIATION DAMAGE ANALYSIS IN THE FUSION MATERIALS PROGRAM:. D. G. Doran, Westinghouse Hanford Company, Richland, WA 99352 The objective of radiation damage analysis is the prediction of radiation effects. To accomplish this requires an understanding of relevant damage mechanisms and how they are influenced by environmental (e.g., primary knockon atom spectrum, defect production rate, temperature, stress) and material variables, and a knowledge of property-property correlations. Specifically, the strategy from the beginning of the fusio"n materials program has been to obtain engineering data 1n fission reactors, the only large volume irradiation facilities, and to obtain damage analysis data in both high energy and fission energy facilities for use in guiding application of the engineering data. Since neither type facility accurately simulates fusion reactor flux-spectra, an essential element of the strategy is the development of models for

24 TMS Fall Meeting property changes in which differences in the radiation environment can be taken into account. Because of space limitations in accelerator-based high neutron facilities. such as FMIT. special emphasis is placed on modeling microstructural changes and on relating microstructure to macroscopic properties. *Research sponsored by the U.S. Department of Energy.

2:20 p.m.

DEFECT PRODUCTION IN ENERGETIC DISPLACEMENT CASCADES:* M. Guinan and J. H. Ki nney, Lawrence Li vennore Nati ona 1 Labora tory, Li vennore, CA 94550 A review of the experimental data on total defect production and disordering in metals at 4.2K will be presented. Results from electron, ion, and reactor neutron irradiations will be compared to the results of fusion neutron irradiations. The dependencies of total defect production and disordering on primary knock-on energy, which are deduced from the experimental results, will be interpreted in terms of theoretical models of the displacement process. In particular, the results of fully dynamic computer simulations will be employed to illustrate the role played by various cascade processes in defect production and disordering. *Work performed under the auspices of the U. S. Department of Energy by Lawrence Livennore National Laboratory under Contract W-14D5Eng-48. 2:40 p.m. EXPERIMENTAL DETERMINATION OF THE ENERGY DEPENDENCE OF DEFECT PRODUCTION,. K. L. Herkle, Wayne E. King, Argonne National Laboratory, Argonne, 11 60439 and A. C. Baily, K. Haga, H. Mesh!!, Northwestern University, Evanston, IL 60201. A knowledge of primary damage production as a function of recoil energy is essential for predicting defect production in radiation environments of practical interest. Ion damage studies have In the past been important for understanding the observed deviations in defect production relative to the predictions of the modified Kinchin-Pease model. The potentialities for deriving experimental damage functions from a combination of electron and ion damage experiments will be discussed in the light of recent advances in determining the threshold energy surface and the low energy behavior of the damage func t ion in copper •. *Research supported by the u.s. Department of Energy and the National Science Foundation.

3:00 p.m. THRESHOLD ENERGY SURFACE AND FRENKEL PAIR RESISTIVITY OF Cu.* Wayne E. King and K. L. Merkle, Argonne National Laboratory, Argonne, IL 60439, and M. Meshii, Northwestern University, Evanston, IL 60201 In-situ electrical-resistivity damage-rate measurements in the high voltage electron microscope have been used to study electron irradiation induced defect production in copper single crystals at T < 10 K. Analysis of the directional and energy dependence yields a threshold energy surface that is significantly different from those of previous investigations: two pockets of low threshold energy centered at (100) and (110) surrounded by regions of much higher threshold energy. The corresponding damage function exhibits a plateau at 0.65 Frenkel pairs. The present results imply a Frenkel pair resistivity for -+().6 ) 0-4 Cu of ( 2.75 -0.2 xl \J-cm. *Work supported by.the National Science Foundation and the U.S. Department of Energy.

3:20 p.m. MOLECULAR DYNAMICS SIMULATION STUDIES OF LOW-ENERGY DISPLACEMENT CASCADES IN Cu.* Wayne E. King and R. Benedek, Argonne National Laboratory, Argonne, IL 60439 Molecular dynamics computer simulations have been employed to study several properties of low energy « 500 eV) displacement cascades in Cu, inc Iud ing replacement sequences, threshold energies, and the damage function. The sialU1ations are based on the Gibson II potential in an initially static lattice. The collisional and cooling phases of the cascade are correlated to different behavior of tille dependent properties such as the number of Frenkel pairs and distribution of atom kinetic energies. The anisotropy of the threshold energy is compared with results of recent single-crystal electron irradiation experiments. Replacement sequences and the damage function were analyzed from events corresponding to 18-34 PKA directions. The damage function exhibits a plateau at 0.5 Frenkel pairs (25-150 eV)j at higher recoil energies the onset of multiple defect production is much slower than predicted from the modified Kinchin-Pease model. *Work supported by the U.S. Department of Energy.

3:40 p.m. COMPUTER SIMULATION OF DEFECT BEHAVIOR

UNDER FUSION IRRADIATION

ENVIRONMENTS: T. Muroga and S. Ishino, Department of Nuclear Engineering, University of Tokyo, Japan

The critical problem in damage simulation is how these data derived by different particle irradiations are correIa ted to those Wlder fusion environments. The typical correlation factors to estimate the microstructural evolution are (1) The probability to escape recombination of Frenkel pairs, (2) Cascade relaxation to two-dimensional clusters, (3) Effect of dose rate and (4) Effect of helium. In the present study, cascade-anneal calculations are carried out to evaluate the effect of those correlation factors using codes MARLOWE and DAIQUIRI and their modifications. We have carried out three kinds of calculations in iron lattices. (1) Cascade-anneal calculation with different mass of PKA. (2) Defect drifting near dislocations after cascade production. (3) Cascade-overlap calculation. Defect survival ratio is fOWld to increase as decreasing the mass of PICA in both cases of after close-pair recombination and after thennal annealing. At moderate temperature, vacancies are accumulated near dislocations. Cascade-overlapping is fOWld to decrease the defect survivability. In addition, we have calculated in aluminum lattices the role of helium in vacancy clustering and found strong dependence of its effect on temperature, interstitials and the mobility of small clusters. These results well corresJ;X>nd to the experimental data and will be helpful to establish the correlation methodology .

4:00 p.m. DEFECT PRODUCTION IN SIMULATED CASCADES: CASCADE QUENCHING AND SHORT-TERM ANNEALING,* H. L. Heinisch, Westinghouse Hanford Company, Richland, WA 99352 Defect production in high energy displacement cascades has been modeled using the computer code MARLOWE to generate the cascades and the stochastic computer code ALSOME to simulate the cascade quenching and short-term annealing of isolated cascades. The quenching is accomplished by using ALSOME with exaggerated values for defect mobilities and critical reaction distances for recombination and clustering, which are in effect until the number of defect pairs is equal to the value determined from resistivity experiments at 4K. Then normal mobilities and reaction distances are used during short-term annealing to a point representative of Stage III recovery. The number of free interstitials after shortterm annealing is fairly insensitive to the quench model, but the number of free vacancies depends strongly on the clustering criteria used during the quench. Effects of cascade interactions at low fluences are also being investigated. *Research sponsored by the u.S. Department of Energy.

4:20 p.lII. PRECIPITArE RESOLUTION DUE TO HIGH ENERGY COLLISION CASCADES: Phillip Chou and N. M. Ghoniem, School of Engineering and Applied Science. UnlVersity of Califo"nia, Los Angeles, CA 90024. A new Monte Carlo computer code, TRIPOS, for the Transport of .!.ons in POlyatomic §..olids is presented in this paper. The code is applied to the problem of precipitate dissolution due to collision cascades. Precipitate stability is important for the high temperature strength and microstructure stability of irradiated alloys. PKA ahd recoil atom motion is numerically simulated in order to determine the sputtering rate of precipitate atoms into the matrix, and hence the dissolution rate. Benchmark calculations show agreement between TRIPOS code results arid other similar codes for single specie materials. On the other hand, the uniqueness of the method is shown by its application to the internal sputtering of MC-type preCipitates in steels.

4:40 p.m. ION MASS EFFECTS ON BOMBARDMENT-INDUCED DISORDERING OF y' Ni3Si. S.P. Lamond and D. I. Potter, Metallurgy Dept. and Institute of Materials Science, University of Connecticut, Storrs, CT 06268 The kinetics of the order-disorder transformation occurring in y'NiJSi during ion bombardment were investigated as a function of ion mass at temperatures from 25°C to 400°C. Changes in the long range order were monitored in-situ by measuring surface infrared emissivity changes. Microstructural changes were investigated by T.E.M. Ion bombardment produced disordered zones in an otherwise ordered matrix. The zone size and number densi~y varied with bomba~dment conditi0!ls, e.g., a z~ne size of :::80 A typical for 1 MV Kr ions and 100 A for 3 MV Ni ions. The relationship between the logarithm of fractional emissivity change and bombardment time was linear. The slopes of these curves are equal to the product of the zone production rate and the average zone size, I.e., the volume fraction transformation rate or rate of change of the degree of long range order. This disordering rate is expected to be proportional to the displacement or damage rate. Calculations made using current damage codes showed this was the case. Research supported by NSF Grant DRMB006084.

TMS Fall Meeting 25 TIME DEPENDENT PHENOMENA: Creep TMS-AIME General Abstract Session Monday, October 25, 1982 Cervantes Convention Center 2:00 p.m. Room 266 Sel!8ion Chairman: Dr. A. K. Mukherjee, Dept. of Mechanical Engineering, University of California-Davis, Davis, CA 95616.

2:00 p.m. P~RTlCLE COARSENING EFFECTS IN OXIDE OISPERSION STRENGTHENED SYSTEMS: Dayid J. Sro10yitz, and R. PetkoYic-Luton, Exxon Research and Engineering Co., P. O. Box 45, Linden, NJ 07036

In ODS alloy systems alloy st rength is i l1"4>arted by a fi ne dispersion of an essentially inert oxide particles. The specific oxide is chosen so as to resist classical Ostwald ripening. Howeyer in the presence of particular environments and alloy matrix species, chemical reaction can occur at the particle interface leading "in situ" particle coarsening, thus the n4mber of particles per unit yo1ume does not change, but the yo1ume fraction increases. This scenario is examined in the 1 i ght of a mode 1 for creep ina OS systems and the resultant mechanical response described. It is shown that as particle size and particle spacing changes with long exposure to high temperatures, the creep strength shows transitional behavior where it at first decreases, then holds and finally increases.

for the most likely mechanism is given on the basis of enhanced boundary damage. Final rupture for the stress ratio of unity was always parallel to the tubing axis suggesting that inclusions may have played an important role in the cracking process. Resea:r-ch

sponsored by the Department of Energy under contracts L>i th the Unio'n Carbide Corporation and the University of Cincinnati.

3:20 p.m. STATE VARIABLE APPROACH TO TRANSIH!T CREEP DEFORMATION It! STRESS DIP EXPERIMENT: 'LA. Korhonen and Che-Yu Li, Dep<>rtment of Materials Science & Engineering, Cornell University, Bard Hall, Ithaca, NY 14853 Stress dip experiments have widely been used to explore the internal stress state of a creeping material. Recently there has been presented some criticism as to the possibility to measure the internal stress directly by the method, but it has been pointed out that the deta from stress dip experiments can be used to generate parameters for a state variable model, verified already by other means. In this contribution the predictions of a state variable approach to stress dip expPriments are explored in detail and compared to previously published data . . The state variable model will be shown to give a consistent de~cription of the processes i nvc 1ved duri ng the trans i ent deforma t i on, and the predictions and the model parameters cOillpare well with the data obta i ~ed by other means. *Research sponsored by the Depart!'lent of Energy through the Materials Science Division.

3:40 p.m. 2 :20 p.m. ASSESSMENT OF CAVITY GROIHH AND CREEP IN COPPER: M. D. Hanna, Michiqan Technological University, Department of Metallurqica1 Engineerinq, Houghton, MI 49931, and G. ,I. Greenwood, University of Sheffield, Department of Metallurgy, Sheffield, United Kingdom. Recent theoretical analyses of cavity growth indicate that various mechanisms are possible and may be combined in specific ways but there remains only little experimental support. Further experimental studies are reported here based on strain measurement and accurate density detennination. To avoid continuous nucleation cavities "ere first introduced above a critical size either by fonning steam bubbles or by first prestraining at higher stress levels. Evidence is obtained of diffusion controlled cavity growth at low stresses and intennediate temperatures with a progressive transition at higher stresses towards a mechanism controlled by dislocation !'lovement. Where 1rain boundary diffusion controlled growth predominates, the fractional specimen volume increase is equal to creep strain. In this reqion a comoarison is made of the creep to accommodate cavity growth with that of Coble creep when cavities are not present. "hen cavity qrowth is entirely by dislocation movement, as in room temperature defonnation, the fractional volume increase in the cavities is apDroximately equal to the strain.

Examination of the Subgrain Size and Stress Exponent After Stress Reduction During Creep of Aluminum:* Mahmoud S. Soliman, Timothy J. Ginter and Farghalli A. Mohamed, Mechanical Engineering, University of california, Irvine, CA 92717. Etch-pit (EP) technique and transmission electron microscopy (TEM) have been used to investigate the effect of stress reduction on subgrain size during high-temperature creep of aluminum. Based on a very detailed examination of substructures along with mechanical data, two observations are made. First, under, steadystate creep conditions, there is no significant difference between the stress exponent inferred from stress reduction tests and from constant stress tests. Second, after deformation at the reduced stress, the 5ubgrain size approaches the steady-state subgrain size at the reduced stress. Problems ariSing from the influence of the type of technique used to measure subgrain size are also discussed. *Supported by the National Science Foundation under Grant No. DMR 80-25820

4:00 p.m. DIFFUSIONAL CREEP OF TWO-PHASE ALLOYS: I-W. Chen, Department of Nuclear Engineering and Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139

2:40 p.m. ON THE NATURE OF THRESHOLD STRESS FOR DISLOCATIO~ CREEP IN SINGLE CRYSTAL MA956: T. C. Tiearney, Jr. and R. Petkovic-Luton, Exxon Research and Engineering, Corporate Research Science Labs., Linden, New Jersf!Y 07036 To further our u~1derstanding of the origin of the threshold stress for dislocation creep in ODS alloys, experiments were carried out on the single crystals of MA956 in the temperature range 10501200°C at strain rates between 10- 6 and 10- 1 s-l where plastic flow occurs by the generalized motion of dislocations through the particle strengthened :-:latrix. Evidence is presented for a presence of truly athermal component of the low stress which is in the viCinity of 59MPa at 1050°C. The strain rate sensitivity and temperature dependence of the thermal component of the stress is consistent ....·ith that obtained with Fe-20 Cr alloys. In MA956, the threshold stress corresponds closely to the computed.Orol,.,lan bowing stress. This indicates that in this alloy system threshold stress is generated by a simple dislocation - particle interaction and that there is no first order contribution from the solutes. The experimental results are discussed in terms of various models for the particle stress and the therwal stability of particle spacing and siie distribution.

3:00 p.m. CREEP OF AISI 304 STAINLESS STEEL THIN WALL TUBING SUBJECT TO MULTIAXIAL STRESSES AT 1l00oF (593°C), Ed Rosa, Albert C. Wang, and John Moteff, Materials Science and Met-a-iiurgical Engineering Department, University of Cincinnati. Cincinnati, OH and Robert w. Swindeman • Metals and Ceramics Division, Oak Ridge National Laboratory, Oak Ridge, TN.

Optical and Electron Microscopy of failed thin wall tubing after creep under internal pressure and axial loading have been carried out. Preliminary analysis of the stress and strain distributions. using existing models, has been used for correlation with the quantitative microstructural data. While an apparently good correlation between strain rate and effective stress was observed, the rupture times showed perceptible reduction as the ratio of circumferential to axial stress approached unity. The explanation

An analytic solution of diffusional creep of two-phase alloys exhibiting superplasticity is reported. Enhanced diffusion and interface migration are found to be the inherent nature of the dispersed, two-phase duplex alloys under the condition of cUffusional creep. Microstructural evidence supports our conclusion that the unique thermochemistry and kinetics features in two-phase alloys are the key elements for superplas tici ty, Application to alloy designs is discussed.

4:20 p.m. DIFFUSIONAL CREEP IN Ni-207. Cr AND THE INFLUENCE OF Zr*: J. H. Schneibel and G·, F. Peterson, Metals and Ceramics Division, Oak Ridge National Laboratory, Oak Ridge, TN 37830

Nabarro-Herring creep and Coble creep were studied in highpurity Ni-20 wt. % Cr and Ni-20 wt. % Cr-O.ll wt. % Zr by measuring the extension of spirals fabricated from these alloys. The results are compared to available data on the lattice and grain boundary diffusivities in Ni-20% Cr. Coble creep is found to be slowed down significantly by trace additions of Zr and this is explained in terms of a lowering of the grain boundary diffusivity by segregated Zr. The implications for the beneficial effect of Zr in Ni-base superalloys are discussed.

*Research sponsored by the Division of Materials Sci'ences t U.S. Department of Energy under contract W-7405-eng-26 with the Union Carbide Corporation.

4:40 p.m. INTERNAL STRESSES IN Ai-Cu SOLID SOLUTION ALLOYS: * P. Godavarti and J. Weertman, Department of Materials Science and Engineering, Northwestern University, Evanston, 1L 60201

internal stresses during steady state creep of Al-Cu solid solution alloys were measured over a range of composition. stress and temperature, using the stress drop technique. Results indi-

26 TMS Fall Meeting cate a low internal stress at high applied stresses. The interna1 stress is independent of temperature but depends on the eu content of the alloy. The friction stress is proportional to the alloy composition. Results of the internal friction measurements of these alloys wi 11 be presented and the relative contributions of the friction stress and the internal stress to the creep hehavior of the alloy will be analyzed.

* Supportec by NSF Grant IIDMR79-12136.

TUESDA Y - OCTOBER 26, AM

DESIGN REQUIREMENTS FOR TITANIUM ALLOYS I: Non-Aerospace Applications Sponsored by the TMS Titanium Committee Tuesday, October 26, 1982 Cervantes Convention Center 8:30 a.m. Room 266 Session Chairmen: J. C. Chesnutt. Rockwell International Science Center, Thousand Oaks. CA 91360; R. W. Stusrud. Detroit Diesel Allison Division. General Motors Corp., Indianapolis. IN 46206.

e:30 a.m. DESIGN CONSIDERATIONS FOR SELECTION OF TITANIUM I.N INDUSTRIAL APPLICATIONS: Charles S. Young, Titanium Industries, 17 Industrial Road, Fairfield, NJ 07006 The initial use of titanium in industrial applications was based primarily on corrosion resistance of the corrmercially pure grades. Recently, grade 12 titanium has seen increasing usage due to its combination of corrosion resistance and strength at elevated temperatures. This has allowed designs to become more cost effective and has increased the U6e of titanium. In the future. other titanium alloys, such as 6Al-4V, 3Al-2.SV. and 38-6-44, will see usage in industrial applications needing their unique properties. Other alloys, may in the future, be designed to provide combinations of properties to meet certain industrial requi rements.

9:00 a.m. mE POTENTIAL FOR TITANIUM IN AUTOMOTIVE APPLICATIO~S.

A. M. Sherman, Engineering and

Re~earch

Ford flotor Company. Dearborn. MI

Staff, Research,

9:30 a.m. THE APPLICATION OF TITANIUM IN HIGH PERFOR!1ANCE RACING CARS R. W. Stusrud, Detroit Diesel Allison Division, General Motors G;";P:-:-?:Q.Box 894, Code T27, Indianapol is, IN 46206. Derrick Walker, Penske Racing 366 Penske Plaza,. P.O. Box 801, ReadIng. PA 19603

The performance capabilities of Titanium and its alloys have long been recognized by the aircraft industry, and as a result it has found wide spread use. In raCing cars, where there is also a premium on alloy performance, titanium alloys have found application. This presentation will examine some of the applications of titanium in racing cars. Selection criteria will be discussed, along with the alloys selected, the reasons for selection, and the metallurgical processing used to develop the desired properties. Both chassis and engine applications will be covered. The philosophies of the car builders, with respect to the use of titanium will be covered, along with the reasons for excluding titanium from some applications or the total exclusion by some car builders.

Break

Titanium alloys (Ti-6AI-4V generally) are of interest for the long, la tter stage blades of low-pressure s team turbines, because of their high fatigue strength and resistance to pitting corrosion in the acid chloride environment at BO°C, characteristic of the transition stage where the incipient condensation from dry to wet steam occurs in the L-l row of fossil~ turbines. Ti-6Al-4V has about one-third higher fatigue strength in air than the common 13Cr martensitic steel blades. The l3Cr steel pits readily in acid chloride condensates and loses about two-thirds of its fatigue strength, whereas Ti-6AI.;..4V is immune to pitting and maintains about 907. of its fatigue strength in air. In the last stage enough steam has condensed that corrosion fatigue is not limiting, but resistance to water droplet erosion is important. Because of its higher fatigue strength and lower density Ti-6AI-4V would be useful in blades longer than can be produced in 13Cr steel.

10:40 a.m. TITANIUM FOR NAVAL STRUCTURES: R. W. Judy. Jr •• Laboratory, Washington, D. C. 20375

Naval Research

Naval vehicles operate in a very aggressive, hostile environment and have extremely rigorous perfonnance requirements, compared to most applications for titanium. Complicating this scenario is the necessity to build and operate vehicles at the lowest possible cost. Titanium alloys offer many advantages to naval construction; foremost among these are their low density and their resistance to corrosion. The factor preventing wide use of titanium is cost both material cost and fabrication cost. A titanium pressure hull for the ALVIN submersible has demonstrated that long-tern, maintenance-free operation can result trom use of titanium. A second pressure hull is under construction; this hull will enable the SEACLIFF to operate at 20,000 foot depths. General Navy design practice for all structures considers a myriad of interrelated factors involved in the design, fabrication, operation and maintenance of military ships. Crack growth and fracture are considered under the broader umbrella of a Structural Integrity plan. How the characteristics of titanium fit" into this fornat and methods developed for evaluation of titanium alloys will be discussed.

FRACTURE III: Toughness, Elevated Temperatures TMS-AIME General Abstract Session Tuesday, October 28, 1982 Cervantes Convention Center 8:30 a.m. Room 260 Session Chairman: Dr. F. S. Pettit, MetallurgicallMaterials Engineer' ing Dept., University of Pittsburgh. Pittsburgh. PA 15261.

48121

Although high cost and limi ted availability have severely restrained consideration of ti tanium usage in production automobi les there arc certain components for which this material, from a design viewpoint, may be particularly suited. These applications include suspension springs and such engine components as valves, valve springs and connecting rods. The possible technical advantages offered by the use of titanium in these applications include reductions in vehicle weight and improvements in performance and fuel efficiency. In particular, titanium is an ideal spring material due to its combination of high strength and low density and elastic modulii. Therefore, springs are probably the most promising automotive application of titanium. In this talk the design and testing of prototype suspension springs made of g-titanium will be described and the additional research and development necessary before use in production applications will be discussed. Al though the technical advantages noted appear to he substantial, the cost penal ties associated with ti tanium remain too large to consider substitutions at this time.

10:00 a.m.

10:10 a.m. APPLICATION OF TITANIUM ALLOYS IN LOW PRESSURE STEAM TURBINE BLADING: R. 1. Jaffee, Electric Power Research Institute, Palo Al to, CA 94304 -

8:30 a.m. INDENTATION AND FRACTURE TOUGHNESS OF wC-Co ALLOYS: G. A. Sargent, H. Conrad and Y.W. Shin, Metallurgical Engineering and Materials Science Department, Uni versi ty of Kentucky, Lexington, Kentucky 40506

Experiments have been undertaken to relate the hardness of we-co alloys to microstructure and to evaluate resistance to crack growth by means of the Palmquist te!?t. It was found that the crack resistance parameter, GIS' measured by the Palmquist test had a simple dependence on hardness. It was also found that there was a linear relationship between GIS and the mean free path of the cobalt,).. Furthermore a linear relationship was found between GIS and the bulk fracture toughness, GIC '

8:50 a.m. THE EFFECT OF AUSTENITIZING TEMPERATURE ON THE FRACTURE TOUGHNESS OF A 9Ni-4-Co-0-2C STEEL*, T. F. Klimowicz and M. Perra, Sandia National Laboratories, Livemore, CA 94550

A study has been made of the effect of austenitizing temperature on the Charpy impact energy and the fracture toughness (KIC) of a 9Ni-4Co-0.2C steel. In the as-quenched condition, it was found that increased austenitizing temperatures significantly increased both Charpy energy and fracture toughness (KIC). These improvements in toughness were accompanied by increases in both the critical fracture strain and the characteristic fracture distance. In material tempered at 565 0 C, austenitizing temperature was found to have no significant effect on the Charpy impact energy and only a small, but nonetheless beneficial, effect on the sharp notch fracture toughness (KIC). In these samples, fracture was controlled by cementite precipitates produced during the tempering process. Thus, the microstructural changes produced by an increased austenitizing temperature were outweighed by those which took place during the subsequent tempering operation. Hence, for these samples, an increased austenitizing temperature produced no large property changes. *This work supported by the DOE under contract #DE-AC04-76DP00789.

TMS Fall Meeting 27 9:10 a.m. EFFECT OF SULFUR ON THE ROOM TEMPERATURE FRACTURE TOUGHNESS OF 12 wt% CHROMIUM STEELS:*

V. Sriniv asan, currently Senior N.R.C.

Associate, AFWAL/MLLM, Wright-Patterson AFB OH 45433; V. P. Raghupathy, National Aeronautical Laboratory, Bangalore, India; and H. Krishnan, The Corporate R&D, Bharat Heavy Electricals Ltd., Hyderabad t India Fracture toughness was evaluated as a function of sulfur c ontent in 12 wt% chromium steel with sulfur level ranging from 50 to 350 ppm. The steel wi~h ,the lowe st sulfur was prepared through electroslag refining technique, while the others through conventional melting process. A new procedure was appl i ed to evaluate fracture toughness at the onset of first c·rack extension. Fracture toughness first increases significantly and then reaches a plateau with decreaSing sulfur content. Correlation between fracture toughness, ductility and microstructural parameters such as volume fraction, size of sulfides, etc., was found to be best represented by the modified Kraft relationship.

Research carri ed out at the National Aeronautical Laboratory with the financial support· from Bharat Heavy Electricals, Ltd.

9:30 a.m. RATIONALIZING TOUGHNESS IN PEARLITIC STEEL: D. J. Alexander, and I. M. Bernstein, Dept. of Met. Engineering & Mat. Science, Carnegie-Mellon University, Pittsburgh, PA 15213 The toughness of pearlitic steel has been analyzed and compared to the Ritchie-Knott-Rice model for brittle fracture. The fracture toughness, yield strength, wo rk-hardening exponent, and cleavage fracture stress were measured, and used to calculate critical distances , for a variety of microstructures. The critical distances obtained were then compared to those microstructural features· expected to exercise control over the fracture process. The results show, that for pearlitic steels, the RKR model does not provide a satisfact'ory explana tion of the observed toughness variations . The possible reason s for this, as well as the applicability of other models, will be discussed. This research was sponsored by the Association of American Railroads.

The modes of crack initiation and propagation of several nicke1base superalloys have been examined after fatigue and creepfatigue loading at 650°C. In fatigue, crack initiation occurred at surface irregularities or, frequently i.n the higher strength alloys, at inclusions or pores. These defects were usually located at the surface, except for tests at low strain ranges where larger, buried defects often initiated failure. Initial fatigue crack propagation was transgranu1ar, but in those alloys with grain sizes $2011m it quickly became intergranu1ar. This transition was environmentally assisted and did not occur for subsurface cracks until they broke through to the atmosphere. In the creep-fatigue cycle, which included a 15 minute tensile dwell, crack initiation and propagation were both intergranu1ar in all the alloys. Correlations with the alloys' fatigue performance are discussed .

10:50 a.m. CREEP AND CREEP-FATIGUE CRACK GROWTH IN SEVERAL HIGH TEMPERATURE ALLOYS: K. Sadananda and P. Shahinian, Code 6393. Materials Science and Technology Division, Naval Research Laboratory, Washi ngton, DC 20375 Crack growth behavior under static load was studied in the temperature range of 540 to 8700 C in several wrought superalloys, Incone1 600, Incone1 625, Incone1 X-750, Inco10y 800, Hastelloy X, HS 25 and Nimonic PE16. Among these no crack growth was observed in Incone1 600 and Hastelloy X. Specimens instead deformed continuously giving rise to crack tip blunting. Specimens of these two alloys "!j{e also tested under cyclic load and cyclic with hold time at 650 C to see how creep during the hold affects the fatigue crack growth. Among the other a1l8Ys, creep crack growth ~s observed in Incone1 625 at 650 and 760 C but not at 540 and 870 C where crack tip blunting occurred without any crack growth. Crack growth was sensitive to the air environment in some alloys while in others it was not. The crack growth behaviors of these alloys are compared and the results are discussed in terms of the role of microstructure and creep properties of materials in contro111ng behavior.

11:10 a.m. DISLOCATION STRUCTURES IN COPPER FATIGUED AT ELEVATED TEMPERATURES:* H. Shirai and J. R. Weertman, Dept. of Mats. Sci. & Eng . and Mats. Res. Center, Northwestern University, Evanston, IL

60201.

9:50 a.m. EFFECTS OF PRECIPITATION HEAT TREA1MENT ON '!HE MICROSfRUCTIJRE, TOUGHNESS, AND STRESS CORROSION CRACK PROPAGATION RESISfANCE OF ALThllNUM ALLOY 2020: . J. G. Rinker, Metallurgy Department, School of Chanical Engineenng, Georgia Institute of Technology, Atlanta, GA 30332 and T. H. Sanders, Jr., School of Materials Engineering, Purdue University, West Lafayette, IN 47907 It has been previously recognized that underaging high strength aluminum alloys provides greater toughness than overaging to the same yield strength . It has al so been established, for AI-Cu alloys , that maximum susceptibility to s tress corrosion c racking (SCC) occurs in underaged tempers. In this study, it was found that an underaged temper with a yield strength 1 2% lower than the peak aged condition had 80% higher plane strain fracture toughness (~, SL orientation) and 300% higher crack propagation resistance (ill': , TL orientation) . The SCC r esistance was also very good; plateau velocities for short transverse, doub Ie cantilever beam samples exposed an meter/ aqueous 3.5% NaCl solution were on the order of 10second. The aforementioned maxinrum in SCC susceptibility was found to occur at a very early stage in the precipitation hardening process .

i8

Dislocation structures have been observed in polycrystalline copper fatigued over a range of strain amplitudes and temperatures. The corresponding cyclic stress-strain curves have been measured and compared with the dislocation structur-e s. The early stages of PSB formation have been examined and inciusion theory used to estimate stress concentrations in front of PSB nuclei. Such concentrations may play an important role in the propagation of PSB' s once nuclei have formed.

* This

work is sponsored by the MRL program of NSF, grant

DMR79-23573 .

GRAIN REFINEMENT IN CASTINGS AND WELDS: THEORY AND PRACTICE III: Welding Sponsored by the TMS Solidification Committee Tuesday, October 26, 1982 Cervantes Convention Center 8:30 a.m. Room 261 Session Chairmen: S. A. David, and J. M. Vitek, Metals and Ceramics Division, Oak Ridge National Laboratory, Oak Ridge, TN 37830.

10:10 a.m. LOW VELOCITY PUNCTURE RESISTANCE OF IIATERIALS: STATE-OF-THE-ART, PROBLEMS AND PReSPECTS. M.R. Krishnadev*, M.C. Cheresh**.

8:30 a.m .

* Dept of Metallurgy, Laval Univ. Quebec, GIK 7P4, Canada. **Effects Technology,Inc., 5383 Hollister Ave., Santa Barbara,

GRAIN REFINEMENT DEPENDENCE ON THE SOLIDIFICATION AND SOLID STATE REACTION IN ELECTRO SLAG WELDS, S. Venkataraman, J. H. Dev1etian, W•. E. Wood, and D. G. Atteridge, Oregon Graduate Center, 19600

California

93111

Development of puncture-proof containers for transportation of hazardous materials (chemical and nuclear waste, corrosive liquids and gases) are becoming increasingly important in view of the disastrous consequences of an accidental spill-over or crash. Hence, a study has been undertaken to evaluate the PRB of HSLA Steel, Stainless Steel, and composite materials via a specially designed Drop Tower Testing System. Hardness, tensile, and instrumented impact tests have been carried out to establish the basic mechanical properties under dynamic condition of loading. The influence of strain rate, temperature and matrix structure on fracture and PRB are 'discussed. Practical implications of the r e sults for material selection and areas of material development and characterization necessary for a rational approach to container design are discussed.

NW Walker Road, Beaverton, Oregon

97006

A new method to substantially refine the as-deposited grain size

of electros1ag (ES) weld metal has been achieved without the necessity of external vibration. A ten-fold reduction in columnar (prior austenite) grain width has been obtained in consumable

guide ES welds deposited on 2 in. thick A588 plate by shrouding the guide tube with a fused quartz sleeve". A 10% increase in voltage was required with the quartz shrouded guide tube to maintain adequate sidewall fusion. The quartz shrouded guide tube produced a very sharp thermal gradient in the slag pool generating an intense stirring action. This resulted in not only a uniformly refined. grain structure, but also consistent Charpy impact toughness throughout the weld fusion zone. The limited segregation due to the intense stirring action in the weld pool led to the absence of noticeable solute band formations and weld centerline cracking when welded under severe restraint.

10:30 a.m.

9:00 a.m.

FATIGUE CRACK INITIATION AND PROPAGATION IN SEVERAL NICKEL-BASE SUPERALLOYS AT 650°C: J. Gayda and R. V. Hiner, NASA Lewi s Research Center, Cleveland, OH 44135

MODIFICATION OF WELD FUS ION ZONE GRAIN STRUCTURE IN THORIUM DOPED IRIDIUM ALLOYS* S. A. David and C. T. Liu, Metals and Ceramics Division, Oak Ridge National Laboratory, Oak Ridge, TN

28 TMS Fall Meeting The arc and laser weldability of two Ir-D.3 % W alloys containing

60 and 200 wt ppm Th have been investigated.

The Ir-0.3 % W alloy

containing 200 wt ppm Th is severely prone to hog cracking during gas tungsten arc welding. However, the alloy can be welded with a continuous wave high power C02 laser system and an electron beam. (EB) welding system. Successful laser and EB welds without hot cracking have been attributed to the highly concentrated heat source available in th ese sys tems and the refinement of the fusion zone g rain structure. Efforts to refine the fusion zone structure during gas , tungs ten arc welding of Ir-O. 3 % W alloy containing 60 wt % ppm Th were successful. Transvers e arc oscillation during welding has been found to effectively refine the fusion 20ne structure. Microstructural refinement of the fusion 20ne in these alloys will be discussed in the light of weld puddle geometry, welding speed and the turbulent action within the weld puddle. 'll'Research sponso r ed by the Division of Advanced Systems and Materials Production, U.S. Dept. of Energy, under contract W-7405eng-26 'with the Union Carbide Corporation.

9 : 30 a.m. INOCULATION OF SIMILARCOMPOSED MIG WELDING OF NODULAR CAST IRON WITH SMALL GRAIN SIZE U. Draugelates, ' Professor and Head, Department of Welding and Production Engineering, Technical University Claustahal, D-3392, Claustahal- Zellerf eld, Fed. Rep. of Germany.

Application of basic research results on the effect of "inoculating e l ements and alloys on the deoxidation or inoculation of nodular cast iron weldings will be discussed. The influence of welding current, - veloc ity etc. on the structure of similarcomp6sed M IG weldings will be shown. Development of special cored wir es with optimized properties referred to perfect spherodial g raphite crysta ll ization, joint structures and welding technology, a nd influence of in situ heat treatment with different sources of heat during MIG welding on s tructure and grain size of nodular cast iron will be demonstrated.

10:00 a.m. EFFECT OF DIFFERENT IIETHODS OF DEOXIDATION AND INOCULATION ON NUCLEATION DURING SIHlLAR WELDING OF NODULAR CAS T IRON Dr. Vollanar Neubert, Department of Welding and Production Engineering, Tec hnical University Claustahal, 0-3392, Claustahal Zellerfeld, Fed. Rep. of Germany

The effect o f inoculation elements and alloys on the deoxidation and inoculation of sperodia1 graphite cast irons in similar weldings was investigated. The inoculation in melts of nodular and austenitic nodular cast iron, influence of elements like LI, CA .... etc., and alloys or mixtures of inoculates on the structure of various nodular irons were studied. Influence of weld pool temperature, time and cooling rate will be discussed. Determination of optimized combination and amount of inoculating elements for the application at similar welding of nodular cast iron will be presented.

10:30 a.m. THE INFLUENCE OF PROCESS PARAMETERS AND SPECIFIC ADDITIONS ON EPITAXIAL GROWTH IN MULTIPLE PASS WELDMENTS: M.S . Misra, D.L. Ol son, and C.R. Edwards, Center for Welding Re"Search. Colorado School of Mines, Golden , Colorado, 80401.

Epitaxial growth in multiple pass weldments promotes long, directional gra ins and concern over potential reduction in mechanical integri t y. The influence of process parameters and specific additions on the nucleation and growth processes in multiple pass weldments was stud i ed. The va riou s mod els to exp l ain the varia tion s i n epitaxial g rowth will be discussed.

11:00 a.m. GRAIN REFINEMENT OF WELDS IN LIGHT METALS USING PULSED CURRENT, W. Simms a nd T. W. Eagar, Department of Materials Science and Engineering, Massachusetts I nst i tute of Technology, Cambridge,

HYDROGEN IN SOLUTION IN METALS AND ALLOYS III Invited Papers Sponsored by TMS Alloy Phases Committee Tuesday, October 26, 1982 Cervantes Convention Center 8:30 a.m. Room 273 Session· Chairman: D. G. Westlake, Materials Science Division, Argonne National Laboratory, Argonne, IL 60439.

8:30 a.m. ELECTRICAL RESISTIVm AND PIUSE TRANSFORMATIONS: P. Vajda and J.N. Daou, E.R.A. NQ 72{) du CNRS, BBt. 350, Univereite Paris-Sud, F-91405 Oraay, France The electrical resistivity is very sensitive to all types of phase transitions and remains therefore a powerful tool for the study of hydrogen-induced transformatio!18 in metals. Special attention i. given in this talk to the systems RE-H(D) (RE = Lu, Tm, and Er) possessing a stable
9:20 a.m. THERMODYNAMIC STUDIES OF HYDROGEN IN PALLADIUM AND IN ALLOYS OF PALLADIUM . O. J . Kleppa, James Franck Institute, Departments of Cherni stry and Geophys i ca 1 Sciences, The Uni vers ity of Chi cago, Chicago, IL 66037. We have studied solutions of hydrogen in palladium and some of its alloys using a combined calorimetric-equilibrium approach. Our data provide further insight regarding the dependence of the thermodynami c properti es on temperature and on alloy compos ition. A special advantage of the calorimetric-equilibrium approach is that it provides reliable information on the partial entropy of hydrogen . It is now well establ ished that hydrogen atoms dissolved in palladium occupy octahedral interstitial sites. The new entropy data indicate that in some substitutional alloys the hydrogen atoms are distributed in a non-random manner among these interstitial sites. Quite recently we have also studied solutions of hydrogen in Pd3Mn, an alloy which undergoes an order-disorder transformation near 800 K. The results of this investigation will be reviewed. 10:10 a.m. DIFFUSION, THERMOTRANSPORT AND ELECTROTRANSPORT OF HYDROGEN IN METALS : D. T. Peterson, Ames Laboratory and Department of Materials Science and Engineering, Iowa State University, Ames, IA 50011

The interesting and important phenomena of hydrogen movement in metals will be surveyed with particular attention to the interplay o f theory and experiment. Hydrogen concentration gradient , diffusion in pure metals has been the subject of an enormous Inumber of investigations but there are still gaps in the experimental observations and the theoretical understanding is not entirely clear. The experimental studies on electrotransport and thermotransport of hydrogen are much more limited and the theoretical interpretations have serious difficulties even in pure metals. The transport behavior of hydrogen in both substitutional and interstitial alloys is even less well characteri2ed but has already shown fascinating differences from pure metals. *Supported by USDOE, Office of Basic Energy Sciences, Division of Materials Sciences, under Contract No. W-7405-ENG-82.

MA 02139

11:00 a.m. The pulsing parameters of a series of gas tungs ten arc welds have been varied to assess the effect on grain structure. The power supply consists of an anal og transistor bank capab le of pulsing to 20 kHz. Resu lt s indicate that under certain specific conditi ons the columnar grai n structure i s broken up "resulting in grain refinement. Trials on pure aluminum and titanium a lloy will be r eported.

THE INFLUENCE OF HYDROGEN ON THE MECHANICAl PROPERTIES OF METAlS:· H. K. Birnbaum, Department of Metallurgy and Mining Engl neeri ng, Un i vers Ity of III i noi s at Urbana-Champai gn, Urbana, IL 61801. The effects of hydrogen on the deformat i on and fractu re of meta 1s wi 11 be revl ewed wi th an emphas is on recent advances in Most of the talk wi 11 address mechani st I c understandi ng. fracture in systems in which hydrides can not form under the condi tl ons of fractu reo We wi 11 di scuss the role of H in affecting the plas- tic deformation of the metals and how the enhanced plasticity is related to the fracture. The role of grain boundary chemistry on the fracture will be discussed. Results of recent IWger and SIMS studies of S, C1 and H segregation at boundaries wi 11 be related to the mechanism of H embritt1ement in Ni. We will also discuss the recent in situ HVEH envi ronmenta 1 cell measurements of hydrogen enhanced dislocation motion and fracture in Ni.

*This work was supported by the Office of Naval Research and the National Science Foundation.

TMS Fall Meeting 29 LIQUID AND SOLID METAL INDUCED EMBRIT· TLEMENT OF METALS II Sponsored by the TMS Corrosion and Environmental Effects Committee and the ASM-MSD Corrosion and Oxidation Activity

10:50 a.m. TIlE ROLE OF CHEMICAL BONDING IN GRAIN BOUNDARY EMBRITTLEMENT*: C. L. Briant and R. P. Messmer, General Electric Corporate ResearCh and Development, Schenectady, NY 12301. It is well established that impurity elements segregated to grain boW\daries often cause brittle, intergranular fracture.

A SURVEY OF LITERATURE ON LIQUID METAL EMBRITTLEMENT

The questiap of why these elements cause embrittlement has remained elusive. This paper presents results of fully quantum mechanical cluster calculations which address this question. It will be shown that strong embrittling atoms draw charge from the neighboring metal atoms. Consequently, charge is removed from the metal-metal bonds and they are weakened. Cohesive enhancers do not draw charge from the metal atoms and therefore do not weaken the metal-metal bonds. They form rather homopolar bonds with the metal atoms and thereby provide an added increment of bonding at the grain boundary.

M.G.Nicholas, Materials Development and Metallurgy Divisions, AERE, HARWELL. ABSTRACT: This report presents a summary of the published technical literature on the phenomenon of Liquid Metal

'Research partially sponsored by the Office of Naval Research.

Cervantes Convention Center Tuesday, October 26, 1982 Room 270 8:30 a.m. Session Chairman: N. S. Stolofl', Materials Engineering Department, Rensselaer Polytechnic Institute, Troy, NY 12181.

8 :30 a.m. REVIEW PAPER:

Embrittlement (LME) and is intended to be a reference guide for plant designers, operators and engineers. A general descriptIon of the characterisation and prerequisites for LME is presented

followed by a systematic summary of the published information on the embrittlement susceptibility of stressed non-ferrous and ferrous materials. Particular attention has been paid in defining the combination of materials and the conditions under which LME has or has not been observed. It is emphasized that LME is not confined to embrittlement by liquid metals per se but may also be promoted when a solid metal is stressed whilst in contact with a dissimilar solid metal or in the presence of a metallic vapour.

11:10 a.m. LIQUID METAL INDUCED SLOW CRACK GROWfH IN A/oIlRPHOUS METALS: P. Xiao, C. Yang, M. E. Glicksman and N. S. Stoloff, Materials Engineering Department, Rensselaer Polytechnic Institute, Troy. NY 12181. Liquid metal environments can induce delayed failure as well as fracture under cyclic loading of amorphous metals . Fatigue lives of FeBSi in Hg- In at room temperature are shown to be dependent

upon R (omin/omax) ratio .

Delayed failure experiments in 1iquid

metal environments have been carried out on Fe40Ni3SPISMo4 as

well as Fe40Ni40B6P14'

9:30 a.m. FACTORS ALLOWING BRITTLE CRACKING RELATED TO CHEMICAL ALTERATIONS AND KINETICS. James R. Rice, Division of Applied Sciences, Harvard University, Cambridge, MA

02138.

This is a revi ew of th~oretical wo rk on factors favoring brittle versus ductile response of lattices and interfaces to stress concentrations, and on thermodynamic estimates and kinetic considerations related to local chemical alterations that allow brittle response of normally ductile solids. The models so far most developed assume that embrittlement is synonymous with cleavage-like bond breaking, and s eek conditions, phrased in terms of local chemical composit i on, under which cleavage requirements will be met before plastic relaxation can occur. But there is evidence that chemical alterations of plastiC flow properties have not yet been considered adquately, in relation to ductile rupture mechanisms, and, a l s o, that there may in some cases be a continuous transition with, say, local potential of an environmental speCies, from fully ductile crack advance by sliding-off to fully brittle advance by cleavage.

Fractographic feature s resulting from

cyclic loading are compared to those produced in delayed failure experiments; similarly J observations of shear band development under cyclic and static loading are compared. The results are discussed in terms of decohesion and enhanced shear models of embri ttlement.

MECHANICAL BEHAVIOR AND PROPERTIES TMS-AIME General Abstract Session Tuesday, October 26, 1982 Cervantes Convention Center 8:30 a.m. Room 265 Session Chairman: Professor John Kallend, Department of Metallurgical and Materials Engineering, Illinois Institute of Technology, Chicago, n.. 60616.

8:30 a.m.

10:00 a.m.

DEFORMATION BEHAVIOUR OF HSLA STEELS AND ULTRA HIGH STRENGTH STEELS THROUGH ADVANCED ~IATERIAL TESTING TECHNIQUES:

,HEORY OF EMBRITTLEMENT BY CRACK TIP CHEMICAL REACTIONS : R. Thomson, ·Nationa1 Bureau of Standards, Washinqton, DC 20234 : A oen-

M. R. Krishnadev*, L. R. Cutler *t M. C. Cheresh**, D. R. Ireland**, K.Q. Lew**, and J.E. Gre en**. * Dept. of Metallurgy, Laval Univ. Quebec, GIK 7P4, Canada **Effects Technology, Inc., 53 83 Hollister Ave ., Santa Barbara , California 93111

eral theory of embrittlement by crack tip chemical attack which will be reviewed. It will be shown how the problem of fracture decomposes into three main terms: (1) external work; (2) work of the material operating in its linear regime; and (3) work done against nonlinear bonds at the tip of the crack. The crack tip processes, though taking place in a region of atomic dimensions determines the entire system response, and the theory shows how to make this decomposition. The theory not only shows the relation to the intrinsic surface energy of the solid, but indicates how activated processes at the crack tip will lead to slow crack growth. Examples from work on glass which demonstrate the general

principles will be given.

The effect of plasti..:ity will then be

developed along the lines of previous work to show that under certain circumstances, sharp cracks can underlie a more ductile general fracture process. The general ideas of dislocation screenLng will be developed and a theory for fracture in a homogeneous material will be presented including the effect of disloc.ations in fracture. A one dimensional model will be developed from which an overall fracture criterion can be obtained and a unified picture of a simple form of fracture in a homogeneous material mn be furmulated .

10:30 a.m. INTRINSIC DUCTILITY IN MATERIALS: Robb Thomson, NBS. Washington DC and lng-Hour Lin, NBS. Boulder,

CO A crack in a material may operate dislocations either by means of intrinsic processes at the crack tip or by activating external sources. Modern amendments to earlier estimates of criteria for intrinsic brittleness will be reviewed and assessments attempted. These are: (1) Are there ways of lowering the activa~ion energy

of emission for materials with high Peierls energy?

(2)

Are

there ways of cutting off the emission of dislocations after emission of several dislocations have taken place? (3) Does chemical attack enhance brittleness or ductility? (4) What are the prospects for combined cleavage and dislocation emission?

The study deals with the u se of advanced materials testing te chniques, na mely , (i) Instrument e d Impact (i1) Instrumented Deep Penetration (iv) Microhardne ss (v) Acousti c Emission for Evaluating the De f o rma tion Behavi our of HSLA Stee ls and Ultra hi.gh Strength St ee ls as a Func t i on o f Strain Rat e and Temperature. Ductility, fracture initiation and propagation r e sistance, penetration r e sistance and flow be haviour have b een evaluated and exploited in defining the processing and s e rv i c e parameters. Areas ( e xp e rimental as well as theoretical) whe r e further work is neede d to improve furth e r the versat i lity and capabilities o f these te chn i ques t o meet the material charac t e rization need s of tomo rrow a r e als o pointed out .

8:50 a.m. THE UTILIZATION OF CHANNELING MODE SEM IN HOT WORK METALLOGRAPHY H_.J. Mcqueen, Mechani ca 1 Engi neeri ng, Concordi a Uni vers ity, Montrea 1, Que. H3G 1M8* In channel i ng mode, SEM can revea 1 differences of ori enta t ion less than 1 0 and with a resolution of regions as low as 1.5 "m. This capability suits it for application to hot worked microstructures which, through rapid cooling to prevent static recrystallization, retain the subgrain structure resulting from either dynamic recovery or dynamic recrystallization. The SEM further provides a large field of view at moderate magnifications (-2000x) from e1ectropo1ished surfaces of bulk specimens. When applied to hot worked aluminum it revealed the substructure distinctly enough to enable quantitative determination of the subgrain size as a function of working conditions with an accuracy comparable to TEM. In heavil y worked copper, it revea 1ed the dynami ca 11 y recrys ta 11ized grains with their retained substructures and was also able to distinguish statically recrystallized grains. However, in comparison to TEM the technique is unable to reveal the structure of

30 TMS Fall Meeting the sub-boundary or the precise misorientation. *The technical assistance of W.B. Hutchinson of Aston University, Birminsham, U.K., is gratefully acknowledged.

9:10 a.m. THE EFFECT OF ELASTIC ANISOTROPY ON THE (HALL-PETCH) P0.j,YCRYSTAL GRAIN SIZE DETERMINED STRESS Itj.TEN~ITY: C. S. Pande , R.I'. Arms t rong*, and R. A. Hasumura, Naval Resea rch Laboratory, Physical Metallurgy Branch,Washington, DC 20375 and *University of Maryland, College Park, MD 20740

Current interest in an accurate assessment of the Hall-Petch relationship between the fracture stress, T and grain size t over a wide range of grain sizes (especially the strengthening benefits of producing ultrafine grain sizes) had led ,to further examination of the effect of elastic anistroPy.l?~ the Hall-Petch sl ope values (in the plot (7/C) vs (t/b) now known as s tress intensity). An analytical description is given using dislocations queueing characteristic for a pile up locked at a grain boundary by the long range repulsion due to elastic anisotropy added to the local obsta c le forming the pile up. The results indicate that the power of grain size is now function of the elastic anistropy parameter. The results are put in the form of stress intensity dependence on elastic anistropy and compared with previous results obtained numerically.

9:30 a.m. MECHANICAL BEHAVIOR OF Cu-l0Ni-6Sn SPINODAl AlLOY:' S. Shekhar, A. Vilassakdanont and K. N. Subramanian, Department of Metallurgy, Mechanics and Materials Science, Michigan State University, East Lansing, MI 48824

Large-grain samples of Cu-lONi-6Sn were obtained by strain annealing and other crystal growing methods. Hechanica1 tests were carried out with solution-treated and aged specimens using a micro-tensile testing device. In order to understand the agehardening mechanism, thin films of deformed specimens were s tudied by transmission electron microscopy to analyze the dislocation structure . In-situ deformation was also performed inside the electron micros~ *This research was partially supported by the u.s. Department of Energy under the contract No. DE-AC02-8lER10942.

9:50 a.m.

*

THE INITIATION OF PLASTlC FLOW IN A DUAL-PHASE STEEL. Ana M. Sarosiek and Walter S. Owen, Department of Materials Science & Engineering, MIT, 77 Massachusetts Avenue, Rm. 8-309, Cambridge, Massachuset ts, 02139

It is shown experimentally that the elastic limit of a dualphase steel containing 27 v/o of 0.4 w/o martensite is substantially lower than the elastic limit expected from a knowledge of the elastic limits of the constituent phases. However, the difference between the measured and expected flow stresses disappears at only 5x10- 3 plastic strain. The difference in elastic limit is greatly reduced by tempering. These effects are interpreted to be clear evidence of localized internal stresses.

*Research sponsored by the Amax Foundation and the National Science Foundation.

10:10 a.m. SINTERED Ti-6Al-4V ALLOY: AN ELECTRON MICROSCOPY INVESTIGATION OF THE MICROSTRUCTURE AND THE DEFORMATION MODES, Y. T. Lee and G. Welsch, Department of Metallurgy and Materials Science, Case Western Reserve University, Cleveland, 'Ohio 44106.

A Ti-6AI-4V alloy sintered to '" 99% theoretical density from an elemental powder blend was investigated by optical, scanning, and transmission electron microscopy, and 1;>y X-ray energy dispersive analysis. The alloy's microstructure was nearly equiaxed, or lamellar with a small lamella length/width aspect ratio. Two types of pores were found. They are macropores at grain boundaries or triple points, and micropores wi thin a.-grains. Chemical homogeneity was achieved within individual a- or S-grains, however, slight compositional fluctuations were found in different grains. The deformation behavior of a-grains was investigated by electron microscopy on thin foils prepared from tensile deformed and cyclically deformed specimens. The TEM results show tensile deformation to occur by homogeneous slip on both prismatic and pyramidal slip planes of the a-phase. In cyclic deformation, on the other hand, heterogeneous slip on the basal .plane was found associated with a relatively large micropore. The micropores of 5. 0.5 }..1m size act as dislocation pinning sites. This research was supported in part by an Air Force grant of Wright Patterson AFB and by Imperial Clevite Corporation.

10:30 a.m. THE DEPENDENCE OF THE CONSTANT STRUCTURE STRAIN-RATE SENSITIVITY EXPONENT, N, ON MICROSTRUCTURE AND TEMPERATURE:' M. E. Kassner &

K. W. Mahin, Lawrence Livermore National Laboratory, Livermore, CA 94550 The constant-structure, strain-rate sensitivity exponent for materials is defined as N=[ln(i:11e'2)]/ln(ol/o2)]. We determinecl the values of N for A1, Ni and types 304 and 21-6-9 austenitic stainless steels over the temperature range of 0.2-0.8Tm. The dislocation substructures were varied from the annealed to the steady-state (large deformation) condition. We found that N is independent of dislocation substructure. For the alloys and metals studied, tti:8 for temperatures >0.6Tm. For temperatures <0.6Tm, N appears to increase to an approximately constant value between 20 and 40, depending on the alloy. These observations, together with knowledge of the variation of the steady-state stress exponents (n) with temperature, suggest that power law breakdown, as classically defined, is independent of any consideration of the dislocation microstructure. *Work performed under the auspices of the U.S. Department of Energy by LLNL under contract No. W-7405-ENG-48.

10:50 a.m. EFFECT OF DEFORMATION MODE ON SURFACE OXIDE SOFTENING IN TUNGSTEN AND MOLYBDENUM:* J.E.Talia, T.Tottori and R. Gibala, Case Western Reserve University, Cleveland, Ohio 44106.

The effect of surface oxide films on the mechanical behavior of tungsten and molybdenum at temperatures T ~ 0.15 Tm has been investigated. Measurements have been made on single crystals and on polycrystalline materials in tension and in compression. Oxide films of the order 40-100 nm in thickness cause softening of. the materials, i.e., a reduction in flow stress and an increase in ductility. The softening is interpreted in terms of the generation and motion of large densities of mobile non-screw dislocations from the oxide-metal interface. Tensile deformation produces larger softening than compression for tungsten, whereas the opposite is true for molybdenum. This effect is consistent with the nature of the oxide growth stresses on the respective metals and the role of anion versus cation diffusion in determining these stresses. * Research supported by the Division of Materials Research, U.S.National Science Foundation, Grant No. DMR· 78-05713.

METALLURGY OF WEAR RESISTING MATERIALS I Sponsored by the TMS Physical Metallurgy Committee TuellCiay, October 26, 1982 Cervantes Convention Center 8:30 a.m. Room 262 Session Chairmen: J. D. Ayers, Physical Metallurgy Branch, Naval Research Laboratory, Washington, DC 20375; K. J. Bhansali, Metallurgy Division, National Bureau of Standards, Washington, DC 20234.

8:30 a.m. WEAR IN STEELS: A. E. Miller, Dept. of Met. Eng. & Mat. Sci., University of Notre Dame, Notre Dame, IN 46556 The influence of the chemistry and microstructure of a variety of steels on their resistance to adhesive, galling and abrasive wear is reviewed. Emphasis is placed on the features of cOlTlllOna1ity between the wear processes and the mechanics involved during the interrogation of the wear surface microstructures by the i nteracti ng interfaces. The responses of mi cros tructures to imposed surface stresses is discussed in terms of the influence of high strain rate and fracture toughness. Correlations between machinabil ity criteria and laboratory wear test results are drawn and the use of machining residual stress criteria for the design and selection of alloys with resistance to abrasive wear is presented.

9:15 a.m. WEAR BEHAVIOR OF IRON AND STEEL CASTINGS FOR THE MINING INDUSTRY: D. L. Albright, Research Laboratory, Climax Molybdenum Company, Ann Arbor, Michigan 48106

Typical applications for wear-resistant castings in the mining industry include both high- and low-stress abraSion, as well as gouging abrasion. Within each class of applications, the microstructure, hardness and carbon content of candidate materials are predominant factors in determining their suitability. However, there continues to be substantial difficulty in predicting laboratory tests. As an example, the use of iron and steel castings as ball mill liner materials is one important high-stress abrasion application where the combination of wear resistance and toughness which i -5 required of the liners varies with their location in the mill, the mill dimensions, the mill operating conditions and the grinding media. Thus it is very important to carefully account for these parameters when evaluating the mill performance of various liner materials. With this accounting in mind, the results of laboratory wear tests on specimens from a series of liner materials will be compared with the liner wear behavior during mill service, and correlations based upon microstructure, hardness and composition will be presented.

TMS Fall Meeting 31 10:00 a.m.

METALLURGICAL FACTORS AFFECTING WEAR OF COBALT AND NICKEL BASE ALLOYS: K. J. Bhansali, Metallurgy Ojvision, National Bureau of Standards, Washington, D.C. 20234 Cobalt and nickel base wear resistant alloys are used in a number of different service conditions. Their resistance to different forms of wear in combination to corrosion is a function of the alloy microstructure. Typically, microstructure is comprised of hard precipitates likes carbides and/or borides in an alloyed nickel or cobalt matrix. Impact resistance of these alloys is extremely poor .owing to a large amount of hard precipitates. Hence a large majority of cobalt and nickel base wear resistant alloys are used to hardface iron base structural alloys, whi ch is also economi ca lly preferred. A vari ety of commercial alloys exist in which amount and types of hard precipitates as well as matrix composition is varied. Effects of alloy microstructure of many of the commercial alloys on different forms of wear with specific service examples are di scussed.

10:45 a.m. EFFECTS OF COMPOSITION, MICROSTRUCTURE AND SOME SERVICE CONDITIONS ON THE WEAR OF CEMENTED CARBIDES: J. Larsen-Basse, Dept. of Mechanical Engineering, University of Hawaii at Manoa, Honolulu, HI 96822 A number of different mechani sms account for the wear of cemented carbides. These mechanisms vary with service conditions and with composition and microstructure. The main mechanisms are brittle fracture, fatigue and plastic deformation, assisted to various degrees by corrosion and/or interdiffusion. Each of these maJor mechanisms may act both at the macroscopic level and at the microscopic level. At the macroscopic level a major part of the component is affected by the wear event; examples are gross brittle fracture, edge spall ing and hot plastic deformation. At the micr'oscopic level individual carbide grains or binder areas are affected and tota 1 wea r is an accumul at i on of many different wear events; examples are preferential binder removal by surface extrusion or corrosion, followed by fatigue fracturing or up_ rooting of individual carbide grains, removal of carbide grains at edges and grinding marks due to binder deformation, and localized brittle spalling. Effects of carbide type and binder metal and of microstructural parameters on these mechanisms are discussed.

MODELING OF THE STRUCTURE AND PROPERTIES OF AMORPHOUS MATERIALS III Sponsored by the Computer Simulation in Materials Science Activity of ASM and by the TMS Nuclear Metallurgy Committee Tuesd..y, October 26, 1982 Cervante" Convention Center 8:30 ...m. Room 272 S~88ion Chairmen: Y. Waseda, Department of Materials Science and Engineering, University of Pennsylvania, Philadelp~ia, ~A 19104; ~. Kl~man, Laboratoire de Physique des Bolides, Umverslte de PansSud, 91405 Oraay, France.

8:30 a.m. THE NATURE OF THE GLASS TRANSITION: Morre1 H. Cohen, Exxon Research and Engineering Company, Corporate Research Science Laboratories, Li~den, New Jersey 07036 A brief review of the theory of the glass transition recently developed by Grest and Cohen (Phys. Rev. B20, 1277, 1979) will be given. Emphasis will be on the following topics: resolution of K..'iuzman's entropy crisis, fitting the viscosity data, the relation between glas!=, transition temperature and cooling rate .• dispersion of relaxation rates, and the relation Letween low temperature tunneling states and the glass transition. All of the above will be discussed in terms of the free volume model. It will then be shown how the free volume model has counterparts for glasses for which the free volume model taken literally is inappropriate, e.g., network glasses.

9:00 a.m. UNCOMMON LENNARD-JONES GLASS TRANSITIONS. Rodney Cotterill and Jens Ulrik Madsen, Department of Structural Properties of Materials, The Technical University of Denmark, Dk-28oo Lyngby, Denmark. The molecular dynamics technique has been used to study two uncommon glass transitions of Lennard-Jones matter. A threedimensional system was subjected both to isothermal change of density and isochoric change of temperature. The pair distribution function, and also a special form of triplet distribution function were monitored throughout the computer simulations. E8.ch of ~he above treatments was found to yield a glass transition, and both these uncommon transitions shed new light on the general question of phase changes in simple matter.

9:30 a.m. STRUCTURE OF LIQUID AND AMORPHOUS ALLOYS: J. Th. M. De Rosson, Department of Applied Physics, Materials Science Centre, Rijksuniversite1t Groningen, Nijenborgh 18, 9747 AG Groningen, The Netherlands The structure of a liquid as well as of an amorphous binary alloy is descriQed by three partial structure factors and thus three independent measurements are required for a complete analysis .. In this contribution, we first show experimental results for l~­ quid alkali alloys obta~ed by X-ray and neutron diffraction which provide independent information. Since a third source of information ~s still n~eded, computer simulation experiments have been performed for a few compositions for which a procedure has been developed to obtain concentration dependent interatomic potentials. To compare general features of the ,structure of the amorphous state with those ~f the liquid state, experimental work has been carried out on Pd-Si-Cu using transmission electron microscopy and fiel~-ion microscopy. The latter provides some information about higher order correlation functions. A model for the amorphous binary alloy is then compared with a moqel for th~ liquid binary alloys.

10:00

a.m.

AMORPHOUS-CRYSTALLINE INTERFACES IN TETRAHEDRALLY COORDINATED MATERIALS:. F. Spaepen, Division of Appliec\ Sciences, Harvard Universtty, Cambridge, MA 02138 The selection of an appropriate potential for relaxation of continuous random network models " is critically reviewed in the light of recent measurements of the molar heat of crystallization of amorphous Si, "which contrary to expectations from elastic constant scaling, turns out to be smaller than that of amorphous Ge. The results of relaxation pf amorphous:-crystalline interface models, both planar and ledge-type, are reported and the implications for crystal growth and nucleation are discussed. *This research was supported in part by the National Science Foundation Wlder grant DMR80-20247.

10:30 a.m.

ELASTIC AND DYNAMICAL PROPERTIES OF METALLIC GLASSES J.Hatner:t Institute for Theoretical PhYSiCS, Technica University, Vienna, Austria In the first part of this paper a calculation of the atomic structure, the equlibrium denSity, the equation·of state and of the elastic properties of ~lassy Kg Zn and Ca~g alloys is presented, based on effectite ~olume-d'perldent interatomic pair potentials and volume energies derived from an ab-initlo pseudopotential theory and a cluster-relaxation technique. The second part is devoted to the dynamical properties. Calculations of the local and of the total vibrational densities of state and of the dynamical structure tactorshave been performed, both usi~g the recursion and the equation~of-motion techniques. The theoretical results show good agreement with inelastic neutron scattering measurements. The origin and character of the low-frequency modes typical for the amorphous state are discussed. 11:00 a.m. ELECTRONIC STRUCTURES IN AMORPHOUS METALLIC ALLOYS: Takeo Fujiwara, Institute of Materials Science, University ~ '~'sukuba, Ibaraki 30S~ Japan Recently, concentrated investigation has been focused on the electronic structures and properties in amorphous metallic alloys. The observed radial distribution functions of varj.ous metallic alloys show a quite common feature but the atomic scale st1;'ucture may be quite diffet:ent. The photoemission spectra have been observed in 3d-transition metal-metalloid alloys and trans.,.. ition metal-transition metal alloys. These spectra show a strong d-p and dod hybridization. Self-consistent calculation of the electronic structure is important to understand these spectra. We will present the se1fconsistent calculations of the electronic structures in these amorphous metallic alloys (TM-M and TM-TM), by using the LinearMuffin-Tin..,.Orbital method in the Atomic-Sphere-Approximation applied to the constructed relaxed D~P systems.

32 TMS Fall Meeting NOVEL TECHNIQUES IN METAL DEFORMATION TESTING III: Analytical Techniques Sponsored by the TMS Shaping and Forming Committee and the ASM Process Modeling Activity Tuesday, October 26, 1982 Cervantes Convention Center 8:30 a.m. Room 264 Session Chairman: C. S. Hartley, State University of New York, Dept. of Materials &ience and Engineering, Stony Brook, NY 11794.

1l:30 a.m. MATERIAL ROTATION EFFECTS IN TENSION-TORSION TESTING: E. W. Hart, Cornell University, Ithaca, NY 14853 The most usual method of biaxial testing of materials is the testing of thin-walled cylinders under combined tension and torsion. Although considerable testing has been carried out in this mode, there is an important aspect of the test that has been ignored in almost all investigations. This is that the shearing deformation associated with the torsion is that of simple (engineering) shear rather than pure shear. This means that for each increment of shear strain there is also an incremental rotation of the material elements of the test specimen. An adequate analysis and representation of tension-torsion data cannot be properly carried out without taking account of the rotations. A method for such analysis will be shown. The analysis will be applied to some recent test data for nickel. It will be shown further that, with proper measurement and analysis, tension-torsion data can be employed to determine the values of several significant deformation parameters with good precision.

8:50 a.m. OCCURRf:~CE OF PLASTlC INSTABILITY IN LARGE STRAIN TORSION TEStING OF CYLINDRICAL BARS: C.R. Canova, E.F. Rauch, J.J. Jonas, Depa.rtment of Metallurgical EngIneering, McGill University, 3450 L'nivcrsity Street, ~ontreal, Canada H3A 2A7

The occurrence of plastic instability in torsion is analyzed in terms of the growth of geometric and metallurgical defects. The influence of temperature non-uniformities on the development of strain gradients is also described. The treatment is based on a simpl ified flow law in which the flow stress is uniquely prescribed by the strain, strain rate and temperature. The conditions under which the radial temperature gradient can be neglected are derived. Experiments of this type conducted on 304 stainless steel at room temperature are reported. A more detailed analysis, in which the radial temperature gradients are no longer negl igib1e, is also described. Experiments of the latter type were carried out on Ti-6242 at 850°C and 1O- 2 s- 1 • The values of the strain and temperature gradients predicted by the analysis are compared with the results obtained in the two types of test and the shortcomings of the treatment are discussed.

9:10 a.m. STRAIN

IlISTRIBlJfIO~

IN TUBE DRAWN AND IIYDROSTATICALLY EXTRUDED Elife Unal and Craig S. Hartley, Depart-

·\I.U~IINU~I ANIl COPPER*:

ment of ~Iaterials Science-a-nd Engineering, State University of ~C\·; York. Stony Brook, N. Y. 11794. lleformat ion processes, in general, develop anisotropies related both to changes in grain boundary netv.·ork and to reorientation of grains caused by plastic deformation. Grain strain measures the deformation of grain boundary network. Use of stereological parameters characterizing the network of grain boundaries of single phase metals enables us to find a quantitative relationship between the gratn strain and bulk strain. Study shows that grain stra in, defined in terms of mean 1 inear intercepts as Et

Q,n(N~iiN~t.

varies I inearly with bulk strain, defined in termS of

changes in dimensions of workpiece as [i

= Q,n(d~/d~).

Stereological measurements done on copper and aluminum specimens deformed by tube drawing and hydrostat ic extrusion processes wi 11 be used to explore further the grain strain vs. hulk strain relationship.

*Support of this research by the National Science Foundation, Metallurgy Program, is hereby acknowledged.

9:30 a.m. LO~D P[LAXATION STUDI[S OF DHOP~ATlON M[C~ANISMS AT HfV~T[O HMP[PATUP[S: ~.A. Korhonen. R.L. Keusseyan. L.X. Li. and Ctle-Yu Li. Department of t·1aterials Science and Engineering, Cornell University, It~ 14853

Load relaxation test generates deformation data which is generally represpnted in the fom of logarithmic stress 109arithmic strain rate curves. The cl'iaracteristics of these curves have been used extensively at low hOr.'lOlogous temperatures for identifying grain matrix controlled deformation processes. At temperatures near and above onp half of the absolute melting temperatures grain boundary slidin9 and thermally indicpd structural changes. including grain boundary migration, will occur. The shape of load relaxation curves has been found to reflect the various processes involved. For example. a sigmoidal shappd curve is seen for Nickel and its alloys when the contribution of grain boundary sliding 1s significant. Thermally induced structural chonges can be shown to produce a sufficiently high strain rate sensitivity index required by structural superplasticity. This report will be concerned mainly with mechanisms. which are important at elevated temperatures. Discussion will be made of the uses of load relaxation data to determine the deformation and materiell parameters required for the char~cterilation of these processe~.

9:50 a.m. UNIAXIAL STRA[N HARDEN[NG AT LARGE STRA[N IN SEVERAL SHEET STEELS; A. K. Sachdev, Metallurgy Dept. and R. H. Wagoner, Physics Dept., General Motors Research Laboratories, Warren, Michigan 48090 The detailed strain hardening characteristics of four steels have been measured in uniaxial tension for strains beyond the uniform elongation, i.e., in the diffuse neck. Tensile tests were carried out on photogridded samples so that true strain, strain rate, and stress could be determined for small elements up to the fracture strain. Analysis of these data showed that a two-piece function could satisfactorily model the strain rate versus st~ain behavior in tension. Using this function along with an equation relating strain-rate sensitivity (m-value) and strain rate allowed accurate determination of the strain hardening exponent in the diffuse neck. [n each steel, the work-hardening exponent increased for strains beyond the uniform elongation. This effect, coupled with the measured increase of strain-rate sensitivity with strain rate helps to understand why steels in general have a forming limit strain which is higher than that expected from uniform elongation measurements.

10:10 a.m. Compression Testing and Finite EleMent ~adelling of Strain-Rate Dependent Buckling of Al Honeycomb. M.M. Hvatum, Advanced D~velopment, Texas Instruments Inc., Attleboro, MA, D. K. Roylance, MIT, Cambridge, MA Compressive yield st~ength and modulus have been derived via compression testing and finite element modelling for honeycomb structures of several geometries. It has been found that compressive yield strength depends on the buckling mode induced in the flanges of the honeYCOMb. The buckling modp in turn is influenced by the geometry and constraint conditions of the structure as well as the bulk material properties. The relative influence of material properties vs geometr,f and constraint conditions on the buckling mode and yield stress is discussed for strain rates of .002 in/sec (Instran loading) to 100 in/sec (impact loading).

10:30 a.m. An Analytical and Experimental Study of Notched Stretch Flanges: N.-M. Wang, Mathematics Department, General Motors Research Laboratories, Warren, MI 48090 This paper presents a mathematical model of a stretch flanging operation in sheet metal fanning for flanges with a notch-shaped cutout. By assuming that during the flanging operation, all intermediate sheet surfaces are developable, the associated mathematical problem is equivalent to one that involves only stretching in a plane, and can easily be solved by an appropriate finite element method. To validate the model as well as the finite element calculations, flanging experiments using specially designed notched specimens of aluminum-killed steel and IN-X80 high strength steel, were carried out. The strain distributions along the notch free edge were measured and compared to the corresponding model results. For the peak strain along the notch which governs the flange formability, the calculated results differ from test data by within 10%. Based on the model, the effects of various geometrical parameters on the peak strain were calculated.

10:50 a.m. AN EVALUATION OF CAM PLASTOMETER TEST DATA IN TERMS OF ROLLING MEASUREMENTS: C.L Ruddle, W.A. Pollard and A.F. Crawley, CANMET, Dept. of Energy, Mines and Resources, Ottawa, Canada.

True stress data determined from cam plastometer hot compression tests of a plain carbon steel were correlated with mean yield stresses computed from measurements of hot rolling tests at equivalent deformation conditions. Stresses measured in rolling deformation were larger than in cam plastometer tests by up to 2X for plate reduction schedules, but approached agreement for sheet reduction schedules. The variance of the cam plastometer data with the rolling measurements at larger gauges (slab to plate thicknesses) was examined in terms of rolling theory. and in terms of the nature and distribution of deformation relative to thic~ness in the rolling pass. This study concludes that the cam plastometer test as a means of deformation simulation for predicting rolling forces and powers is most accurate for thin gauges where uncertainties caused by deformation inhomogeneity have least effect on yield stress calculations.

11:10 a.m. AN ANALYSIS OF NONUN[FORM AND LOCALIZED DEFORMATION IN DUcrILE SINGLE CRYSTALS D. Peirce, R.J. Asaro and A. Needleman, Division of Engineering, Brown University, Providence, RI 02912 The nonuniform and localized deformations of ductile single crystalS subj ect to tensile loading are analyzed numerically. The crystal is modelled by a rate independent, elastic-plastic relation based on Schmid I s law which precisely accounts for lat-

tice rotations.

Both self-harderting and latent hardening of the

slip systems are included in the model. Initial imperfections are specified in the form of slight thickness inhomogeneities and the calculations follow the crystal deformation through diffuse necking and the formation of shear bands. The pattern of shear bands depends on the initial imperfection, but, independent

TMS Fall Meeting 33 of the particular small imperfection, the material planes. of the bands are inclined at a characteristic angle to the slip planes. Also, the lattice misorientation across the shear band, which is such as to cause geometrical softening of the bflnds, is not sensi ti ve to the imperfection form. The numerical results are compared to experimental results for aluminum-copper alloy and pure copper single crystal s .

11 :30 a.m. A MODIFIIm TORSIONAL SPLIT HOPKINSUN BAR FUR THl!: CYCLIC LUADING OF MATERIALS AT DYNAMIC STRAIN RATES*, A.M. Eleiche and M.M. ElKady, Dept. of Mechanical DeSign &: Production, FacUlty of l!.'ngineering, Cairo University, Culza, ~ypt, ilxperimental data on material behavior under cyclic loading at dynamic (impact) strain rates are lacking in the literature,because a reliable testing technique for that purpose is still nonexistent. The torsional split Hopkinson bar apparatus, used hitherto for determining mechanical response under 'constant strain rate or strain-rate jump conditions, can be easily modified to do the job successfully. The development of 01Jr apparatus is described with particular emphasis being made on its cyclic loading capability, whereby thin-walled tubular specimens of short gage length are 5ub.jected to success1 ve torsional impacts. .c2ch impact loads the specimen at a dynao1ic constant strain rate, followed by unloading. After a maximum dwell-time of 5 minutes, the next impact reloads the specimen at the same rate but in the reverse direction; and ' so on until fracture. Results of tests performed on an aluminum alloy and a mild steel are also presented. *Partly sponsored by the European Research office of the US Army.

(I0:10 a.m.) TECHNOLOGY -- A KEY TO PRODUCTI VITY: Dr. Bani R. Banerjee, Assi stant Di rector of Research, Ingersoll-Rand Co. Japan's massive leap to th~ forefront of manufacturing efficiency has amazed and staggered the leading industrialized nations. The Japanese ach i evemen t combi ~es effecti ve work-force management, a longer-tenn outlook in corporate cost-benefits calculations, and the Careful nurturing and development of middle management. In order to counter this threat of global business comoetition, U.S. industry must do many things in parallel. Primary amon~ these is the need to innovate and to effectively implement these innovations into the manufacturing methods and processes, and, of course, the key to suCh innovation lies within the realm of techno logy .

(10:40 a.m.) ROLE OF R&D PRODUCTIVITY: Herbert I. Fusfeld, Director, Center for Science & Technology POllCY, NYU, Graduate School of Business. Technical change is a major force for increa·sing industrial productivity. It is therefore critical that we focus our efforts to derive the greatest number of options most relevant for our economic and social system in the shortest time from whatever technical resources are made available. This not only calls for attenti on to tradi ti ona 1 research management wi thi n a 1aboratory, but must emphasize strategic research planning within the industrial system and more effective interactions among technical advances throughout the international structure of science and technology. The higher costs for experimental research and the se 1ecti ve shortages of personne 1 i ntens ify the need to improve R&D proQucti vity.

(11:10 a .m. )

OVERVIEW OF PRODUCTIVITY ISSUES: Session I of GEMlG·PAC Productivity Symposium Sponsored by the TMS-A/ME Governmental, Energy and Minerals Committee and the ASM Government and Public Affairs Committee Tuesday, October 26, 198~ Cervantes Convention Center 8:30 a.m. Room 275 Session Chairman: Dr. Bani R. BaneJjee, Ingersoll-Rand Research Center, P.O. Box 301, Princeton, NJ 08540.

PRODUCTIVITY ISSUES -- FACETS FOR ENGINEERS : Harris M.. Burte, Director , Metals & Ceramics Division, AFWAL, Materials Laboratory. There is significant diversity of opinion as to causes and appropriate remedies for productivity problems in the United States. Engineers can provide valuable, objective input to such debates. However, if they are to speak authoritatively and credibly, they must exert the effort to dec9ffipose complex issues and to define meaningful questions that they can respond to within the limits of their professional competence. A list of such questions will be proposed.

PHASE TRANSFORMATIONS III TMS -A/ME General ·A bstract Session

(8:30 a.m.) Keynote Presentation: Productivity Through Motivation: McCon~, Chainnan & CEO, Worthington Industries.

John H.

Mr. McConnell will discuss the decline in productivity in the U.S. during the last 30-40 years. The theme will be that the i1merican work ethic which helped make our country grow is not dead. People still want to work and will take pride in their jobs. The problem is management uses old tired methods which are not effective today. People are more educated, more infonned and management needs to implement different, more modern methods to motivate employees. Mr. McConnell will illustrate specific methods his company has utilized to attain productivity levels far above industry medians.

Tuesday, October 26, 1982 Cervantes Convention Center 8:30 a.m. Room 263 Session Chairman: A. Guha, Brush Wellman, Inc., Alloy Research & Development, 17876 St. Clair Ave., Cleveland, OR 44110.

8:30 a.m. AN AEM STUDY OF FERRITE GROITH IN TIlE Fe-Ni-P SYSTEM C. Narayan and J. I. Goldstein, Dept. Met. & Mats . Engrg., Whitaker Lab 115, Lehigh University, Bethlehem, PA 18015

An AEM study of intragranular ferrite growth in Fe-Ni-P alloys was conducted to examine the equilibrium and paraequilibrium

(9:00 a.m.) Competing Internationally: Lloyd Reuss, General Manager, Buick Division, General Motors. In order to be successful in the markets of the 80's, in which consumers become increasingly sophisticated, quality conscious, individualistic and more affluent, i1merican industry must become a true international competitor. Coopeting on an international basis will require greatly increased cooperation between government, business and labor, and a high measure of innovation in products, processes and procedures, and in people management .

models that have been proposed to explain the growth of ferrite from austenite. Alloys containing 5..;8 wt70 Ni and 0.1-0.8 wt% P were homogenized at 11000C for 5 days. The heat treatment consisted of austenitizing at 900 D e for a few hours and slow cooling through the two phase (0' + V) region and fina 1ly quenching to room temperature. Thin foils were prepared and Ni concentEation profiles across the a/v interfaces we:re measured with a SOA probe. Interfacial Ni contents agree well with those predicted by the Fe-Ni-P ternary isotherms at the last temperature before the quench. Elec~ron diff~action studies indicate that ferrite grows with a Kurdjumov-Sachs orientation rela~ionship with the paren.t austenite grain. A numerical model based on growth c ontrolled by long range volume diffusion of Ni fits the meas~red penetration curves indicating that, under controlled cooling conditions, ferrite grows under full equilibrium conditions.

(9:30 a.m.) HUMAN RESOURCES AND LABOR MANAGEMENT RELATIONS: Dunlop, Economics Deot., Harvard University.

Prof. J. T.

A wide range of private and public human resource policies impact on productivity -- education, training, safety and health, coopensation arrangements·, supervision and adjustments to changes in employment and unemployment. The quality of relations between 1abor and management and arrangements for work-pl ace i nteracti ons have significant consequences for productivity. (10:00 - 10:10 a.m.) - Break

8:50 a.m. THE BAINITE AND PEARLITE REACTIONS IN A HYPEREUTECTOID Fe-C-Mn ALLOY: D.S. Sarma, J . T. Song, I.M. Bernstein & H.I. Aaronson, Dept. of Met. Eng. & Mat. Science, Carnegie-Mellon University, Pittsburgh.

PA 15213 A microsc opy study has been conducted of the bainite and pearlite reactions at 200°-650°C in an Fe-l.35% C-2% Mn a lloy. The principal difference~ between hypoeutectoid and hype reutectoid bainite are that the upper bainite morphology rema ins dominant to considerably lower temperatures and also that the ferrite and carbide appear to grow synchronously rather than succeSSively ~n the hype r-

34 TMS Fall Meeting eute c toid range. The Bagaryatskii orientation relationship between ferrite and carbide operates exclusively in the bainite range; the Pits c h and other relationships are also pr~5ent in the case of pearlite. Reasonable agreement is found between calcula ted and meas ured lengthening rates of up per bainite whether this structure i s modeled as isola t e d ferrite plates or as pearlite. This has compounded the problem of identifying the fundamental differences b etwee n pearlite and h ypereutectoid upper bainite when both are regarded as eutectoid decomposition products. This r esearch is being supported by the NSF - funded Ce nter for the Joining of Materials.

9:10 a.m. MICROANALYSIS OF DUAL PHASE STEELS, P.A. Wyc1iffe, Physical Meta 11 urgy, Rockwell I nterna ti ona 1 Sci ence Center, Thousand Oaks, CA 91360 Manganese concentrati on profil es, which are produced at the austenite/ferrite interface during the intercritica1 annealing of an Fe-0 . 08C-l.OMn dual phase steel, were studied using an HB5 scaning transmission electron microscope (STEM). Observed concentrati on profil es showed good agreement wi th profil es predi cted us i ng a diffusion controlled local equilibrium model for austenite growth . Measurement of low concentrations ( . 1wt%) using energy dispersive X-ray analysis (EDAX) in STEM is very sensitive to background subtraction. The error in the estimate of background is reduced by performi ng ali nea r regress i on us i ng data from a number of spectra to determine the background shape. This results in accuracy which is improved relative to background estimates derived from a single EDAX spectrum. Research sponsored by Natural Sciences and Engineering Research Counci 1 of Canada and Dept. of Energy Mines and Resources, Canada, CANMET.

9:30 a.m. RESISTOMETRIC STUDY OF IRON-BASED ALLOYS NITRIOED BY CONSTANT ACTIVITY AGING:' rL M. Yang and A. D. Krawitz, Dept . of Mech. Engrg. and Matls. Sci., Rice University, Houston, TX 77001. Resistivity measurements of Fe-V-N were made as a function of V content (1.1, 2.2 and 3.3 a/o V), temperature (500, 550 and 570·C) and nitriding time (up to 150 hrs.). Resistivity measurements were also made for Fe-Mo-N as a function of Mo content (2.2 and 3.0 a/o Mo), temperature (SOD and S50·C) and nitridi ng time (up to 300 hrs.). The 1arges changes in resistivity observed are due to local atomic displacements rather than valence effects. The N and solute form clusters. Cluster size increases with temperature; number of c 1 us ters increases wi th solute content. For the same temperature and solute content, clusters are larger in Fe-MoN than Fe-V-N. In Fe-Mo-N, clusters overage prior · to saturation with N. Upon reduction in hydrogen, excess N in ·the.Fe matrix leaves rapidly; remaining N leaves slowly due to cluster growth and an attendant loss of N from the strain fields around the clusters as the equilibrium precipitate is approached. 'Research supported by Research Corporation, Rice University and the University of Missouri Research Reactor.

9:50 a.m. SPINODAL DECOMPOSITION IN IRON-CHROMIUM:· J. C. LaSalle and L. H. Schwartz, Department of Materials Science and Engineer:ing, Northwestern University, Evanston, Ii.. 60201

Spinodal decomposition in an iron-30% chromium alloy has been studied using SANS. The alloy was aged at temperatures near 50a D e for times ranging from 10 minutes to 1 hour. It was found that exponential growth occurred for aging times less than 40 minutes over a broad range of composition wavelengths centered about a wavelength of maximum growth 30R. The exponential growth indicates the presence of a ti1l1e regime in this alloy where the effects of Bronian motion and nonlinear free energy terms do not overlap, allowing the application of the linear theory of Cahn. Analysis of the data using the linear theory resulted in values of K, the gradient energy, and f " , the 2nd derivative of the free energy with respect ~o composition of 1.6 x 10- 6 ~ and -3.4 x 10- 9 .£!:&3' respectively. Further analysis indic~~es that the coherenEmspinodal fpr this alloy lies at 550 0 C in agreement with the tl)eoretical estimates of Nishizawa.

at

*

occurring in two ternary alloys (Fe-llCo-33Cr and Fe-10.5Co-28Cr). Th i~ work is supported by the National Science Foundation under a University/Industry Cooperative grant, mm.-so 22225.

10:30 a.m. ATOM PROBE FIELD ION MICROSCOPY STUDIES OF PRECIPITATION IN THE IRQN-CHROMIUM SYSTEM: S.S. Brenner, M.K. Miller, U.S. Steel Research Laboratory, Monroeville, PA 15146 and W.A. Soffa, Dept. of Metallurgical and Materials Engineering, University of Pittsburgh, Pittsburgh, PA l526l. Fiel~-ion microscopy (FIM) and atom probe (AP) microan~lysis have been used to study the kinetics and morpholo gy of the precipitation reactions occurring within the miscibility gap at low temperatures in the Fe-Cr system. A series ' of alloys ran gi ng from 1475 At . % Cr have been s tud ied to exp lore the various reF-imes of precipita t ion (classic~l nuclea tion/ spinodal decomposi tion) a!,d to characterize the coarsening kinetics as a function of composition and temperature. These fine-scale precipitation processes play a n important role in the embrittlement of corrnnercial high-chromium fe rritic stainless s t eel s , and this investigation has shed new light on the nature of the a I reaction which is r espo nsible for the well-known '~4 70° embrittlement". This work is supported b y the National Science Foundat ion under a Un i versity/Industry Cooperative grant, or-fR.-80 22225.

10:50 a.m. KINETICS OF 475"C EMB!UTTLEMENT OF FeCrAl STEELS: W. S . Spear, D. H. Palonis, Metallurgical Engineering Division, University of Washington, Seattle, WA 98195 The effec t of ternary addi tions of aluminum (up to 6 wt. %) upon the kinetics and magni tude of the 475°C embrit tlement reaction (at precipitation) in fe rriti c stainless steel s cO!'taining 15 to 18 wt. % Cr is be in g investiga t ed b y use of elec trical re sis tance mea surements and transmission e l ectron microscopy. Microhardness and tensi le t esting along with a SEM inves t iga tion of fracture behavior are being employed to relate the changes in microstructure with the <;lssociated changes in mechanical properties. Resistome tric st udies of the kinetics of microstructural instabi lit y indicate that aluminum additions promot~ the tendency of chromium atoms to c lus t er at t empera t~re s in the vicinity of 475°C. An incre ase in the magnit ude of embritt lement associated with aluminum add itions is a ttribut ed to the restriction of dislocation cross slip and increa~ed coherency strains accompanying ex I p r ecipitation.

11:10 a.m . MARTENSITE INDUCED BY DEFORMATION IN AISI 201 STAINLESS STEEL: Franci sco Boratto, Fundacao Centro, Techo 1ogi co de Mi nas, •Gera is 2000 Av. Jose Candido da Silveira, Cx. Postal 2306/CEP 30000, Be10 Horizonte, Minas Gerais, Brazil. Non-linear least square optimization tecniques were used to analyze four empirical equations that describe stress-strain curves of AISI 201 stainless steel tested at 273 0 K. The volume fraction of martensite induced by deformation, VFM, was calculated from the empirical equations and the results were compared with saturation magnetization measurements. The equation that gave the best fit was a = (0 0 + K,n ) (1 + VFM) + C VFM~ with the volume fraction of martens He given by VFM ~ A, S/(1 + A,S). The constants 00' K, n, C, D, A and B were cOO1~Llter optimizprl . The saturation magnetization
1 -

Exp

{1

-

B (1

-

Exp

(- at)

)n'}

and VFi~ qiven by os/0 100 I·lith computer yielding the value °100' the sa turation maqnetization for 100% martensite .

10:10 a.m.

RADIATION DAMAGE ANALYSIS FOR FUSION REACTORS II. MicrQstructural Evolution-Theory Sponsored by the TMSIASM Nuclear Metallurgy Committee

ATOM PROBE FIELD ION MICROSCOPY STUDIES OF PRECIPITATION IN IRONCOBALT-CHROMIUM ALLOYS: 5.S. Brenner and M.K. Miller, U.S. Stee l Research Laboratory, Monroeville, PA 15146 and P. Camus and W.A. Soffa, Department of Metallurgical and Materials Engineering, University of Pittsburgh, Pittsburgh, PA 15261.

Tuesday, October 26, 1982 Cervantes Convention Center 8:30 a.m. Room 271 Session Chairman: D. G. Doran, Materials Technology, Westinghouse Hanford Company, P.O. Box 1970, Richland, WA 99352

Supported by National Science Foundation DMR79-lll07

~ series of ductile permanent magnet alloys based on the Fe-C~-Cr ternary system have been developed. The addition of Co raises the miscibility gap of the b i nary Fe-Cr system, and the ternary miscibility gap is exploi ted to produce a fine-scale, twophase mixture resul ting in magnetic propertie~ simi1~r to those· of th e well-known Alnico alloys. The permanen t magnet properties and the mechanica l properties are very sensitive to both the amplitude and wavelength of the compositional modulations which evolve during phase separation. In this investigation field-ion microscopy (FIM) and atom probe (AP) microanalysis have been used to characteriz e the kinetics and ItK>rphology of the precipitat ion processes

Recently

8:30 a.m. FACTORS RELEVANT TO THE DEVELOPMENT OF FISSION-FUSION CORRELATIONS F . A. Garner, H. R. Brager a nd F. M. Mann, Westinghous~ Hanford Company, Richland, WA 9935 2 Development of predictive equations for changes in the p r operties of st ruc tural materials is r ~q uired for design of fusion reactors. This r equ ires that one idep.tify the cons e qu ences of differe11ces in neutron spec tra and operational variables between anticipated

TMS Fall Meeting 35 fusion environments and the fission environments from which the data are derived. The list of relevant variables now includes not only differences in atomic displacement characteristics and helium generation but also differences in solid transmutants and operational history. The development of fission-fusion correlations also requires that our perception be correct of the damage mechanisms operating to produce the fi ssion data. Results obtained in recent fission i rradiation studies have led to a reformulation of the parametric dependence of some property changes. It appears, for example, tha t the dependenc e of swelling on t emperature and neutron flux is not as pronounced as previously anticipated. This affects strong ly the prediction of helium's role in macroscopic property changes. Res earch spon so r e d by the U.S. Department of Energy .

8:55 a.m. QUANTITATIVE MODELLING OF SWELLING IN M316 STEEL: R. Bullough S.M. Murphy, T.M. Quigley and M.H. Wood, Theoretical Physics Division, AERE, Harwell, Didcot, Oxfordshire, OXII ORA, U.K.

The chemical rate theory is widely used to describe the microstructural evolution in a material during irradiation. We describe recent improvements to the theory that increase its predictive capability. The.incubation dose prior to the onset of void swelling is modelled by allowing partition of the gas between the various sinks in the microstructure. New dislocation and void sink strengths have been derived incorporating the field effects. Also, some treatment of microchemical effects is now possible. Such improvements to the theory have been incorporated into a new FACSIMILE computer code designated VSS. The new code has been successfully employed to model void swelling during HVM, VEe and fast reactor irradiation of 316 steel.

9:15 a.m. ASPECTS OF THE MECHANISMS BY WIIICH HELIUM INFLUENCES RADIATION EFFECTS IN METALS AND ALLOYS:* L. K. Mansur and W. A. Coghlan, Metals and Ceramics Division, Oak Ridge National Laboratory, Oak Ridge, TN 37830

In fusion and .f ission reactor environments helium is produced in structural materials by neutron-induced transmutatioh. reactions, while in charged particle irradiations helium is often introduced by deliberate injection. It has been found that helium modifies radiation effects on dimensional and phase stability as well as mechanical properties. For example t dose dependence of swelling, temperature dependence of swelling, cavity size distributions, response of microstructure to pulsed irradiations', and pha'se evolution and its interaction with swelling can be affected. The theory of radiation effects may be used to understand a number of the interrelated reaction pathways by which helium may modify the effects of displacement damage. Mechanisms are described based on both theoretical developments and on experiments designed to investigate these phenomena. Several areas of current understanding are highlighted. Areas for future research are discussed. *Research sponsored by the Division of Materials Sciences, U. S. Department of Energy under contract W-740S-eng-26 with the Union Carbide Corporation.

9:40 a.m. THRESHOLD CONCENTRATIONS FOR HELIUM EFFECTS IN IRRADIATED METALS: W. G. Wolfer, Fusion Engineering Program, Nuclear Engineering Department, University of Wisconsin, Madison, WI 53706 Helium produced by nuclear reactions or implanted by ion-bombardment of solids is known to cause embrittlement, assist in void nucleation and growth, fom bubbles, and cause blistering. Each of these effects occurs under certai n condi tions deternli ned by temperature, materials parameters, and a minimlJll concentration of helium. Small amounts of helium are already effective in void nucleation, but larger amounts are required to affect void growth. Bubble growth demands high helium concentrations at temperatures below half of the melting point, and modest amounts above this temperature. Finally, blistering occurs only after helium concentrations have been reached of 0.1 to 0.6 atomic fractions. In order to provide an overview for the occurrence of the various helium effects, "helium-maps" are presented in variable spaces fonned by temperature, (He/dpa) ratio, and voidage. Results are presented for nickel and stai nl ess steel s.

10:05 a.m. HELIUM PARTITIONING BETWEEN VOIDS AND DISLOCATIONS: Barbara Okray Hall, Westinghouse Research and Development Cen-;;;:Pittsburgh, PA 15235.

A comprehensive model, based on the rate-theory formalism, has been developed for describing the interaction of mobile helium with void and interstitial-loop distributions in irradiated metals. Both types of point defect and the evolving loops and voids are assumed to prov.ide saturable and reversible trap sites for interstitial helium. Detrapping by thermal emission and by irradiation displacement are included, with the various heliumtrap binding energies taken as adjustable parameters . To avoid computational limitations, differential equations have be en

developed for the moments of the cavity and dislocation distributions 1:.1 the two-dimensional size spaces that include both point defect and helium number. Concentrations of vacancies, interstitials, interstitial and trapped helium, small clusters and first and second moments of the void and loop distributions have been calculated for a variety of irradiation conditions .

10:25 a.m. HELIUM DIFFUSION IN IRRADIATED MATERIALS: N. M. Ghoniem, F. Sharafat, School of Engineering and Applled Science, University of California, Los Angeles, CA 90024, J. Williams and L. K, Mansur, Oak Ridge Natl. Lab., P.O. Box X. Oak Ridge, TN 37830. A detailed computer model is constructed for the kinetic interaction between hel ium and various microstructural components as well as point defect clusters during irradiation. The theoretical model is based on conservation equations for helium, vacancies, interstitials and their clusters. Coupled with this detailed description we also present analytical approximations of the comprehensive description. The approximations, guided by numerical results, give simple expressions for the effective diffusion coefficient of hel i um migration in irradiated materials. The model . is applied to steel irradiated in both HFIR and EBR-II as simulation facilities for fusion reactors. The model is also correlated to high temperature thermal desorption experiments on nickel.

10:45 a.m. COHPOSITIONAL CHANGES IN Fe-Cr.Ni ALLOYS UNDER PROTON BOMBARDMENT AT ELEVATED TEMPERATURES : N. Q. L8IIl, Materials SCience Division, Argonne National Laboratory, Argonne, IL 60439. USA

Compositional redistribution in austenitic Fe-Cr-Nl alloys during energetic proton bombardments has been calculated 8S a function of time, temperature, and ion energy, using a kinetic model of radiation-induced segregation recently developed for ternary alloys. The effect of spatial nonuniformlty in damage rates was taken into account and found to have a pecul tar effect on the redistribution of the alloying elemehts. Enrichment of Ni occurs at the bombarded surface and in the bulk beyond the damage range, whereas Fe and Cr are enriched In the peak-damage region at the expense of their depletion in regions near the surface and behind the damage range. .These local compositional changes can lead to austenite-ferrite phase transfomation, since the composition of local regions of the alloy is shifted into different parts of the composition-space of the multlcomponent phase diagram.~ *Work supported by the U. S . Department of Energy.

11:05 a.m. THEORY OF PULSED IRRADIATION MICROSTRUCTURES: * E. P. Simonen, Battelle Pacific Northwest Laboratory, Richland, WA 99352

Void and vacancy annealing affect the development of pulsed irradiat ion microstructures . The average void size during pulsed irradiation is influenced by irradiation-driven growth and by thermally-driven coarsening and dissolution. The critical radius for void growth during annealing can be much larger than during irradiation; consequently, pulsed anneals can have significant effects on microstructure in metals at about one-half their melting temperature. Pulse annealing periods comparable to the vacancy relaxat io n t i me can result in a significant discontinuous g rowth histor y of small aggregates. The perturbations are greatest between three-tenths and one-half of the melting temperature of metals having a low sink density.

*Research supported by the Office of Basic Energy Sciences, U. S. Department of Ener gy

SUPERPLASTICITY TMS-AIME General Abstract Session Tuesday, October 26, 1982 Cervantes Convention Center 8:30 a.m. Room 267 Session Chairman: Professor William R. D. Wilson, Dept. . of Mechanical and Nuclear Engineering, -Northwestern University, Evallston,IL60201.-

8:30 a.m . EFFECT OF HYDROSTATIC PRESSURE ON SUPERPLASTIC BEHAVIOR OF 7475 Al : A. K. Ghosh, Rockwell International Science Center, 1049 Cami no Dos Ri os, Thousand Oaks, CA 91360 The influence of superimposed hydrostatic pressure on the superplastic flow and cavitation behavior of a fine grain 7475 Al has been studied. Results are presented to indicate the influence of pressure on the stress/strain/strain-rate behavior of superplastic aluminum, and dynamic grain growth as a function of imposed pressure .. The implications of these results on the cavitation and fracture behavior, and observed tensile elongations are di scussed.

36 TMS Fall Meeting 8:50 a.m.

10:10 a.m.

SUPERPLASTI C PROPERTI ES OF THERMOMECHAN I CALL Y PROCESSED HIGH STRENGTH ALUMINUM ALLOYS, A. K. Ghosh, C. H. Hami lton, Rockwell International Science Center, P.O . Box 10B5, Thousand Oaks, CA, and M. Kumar and J. Waldman , U.S. Anny ARRADCOM, Dover, NJ

STRAIN DEPENDENT CONSTITUTIVE RELATION FOR THE SUPERPLASTIC DEFORMATION OF A MICRODUPLEX STAINLESS STEEL, B. P. Kashyap and A. K. Mukherjee, Division of Materials Science, Department of Mechanical Engineering, University of California, DAVIS, CA 95616 A microduplex stainless steel (25.7% Cr, 6.6% Nil of elongated grains is investigated over a strain tate range of 2xlO-6 to 2xlO- 3s-l and the temperature range of 700 - 10000C by differential strain rate and constant strain rate tests. Flow stress is found to depend on the strain up to about 30% strain under many combinations of strain rate and temperature. Within these conditions, strain rate sensitivity and activation energy associated with the constitutive relation for superplastic deformation have been evaluated as a function of strain. Strain dependent flow stress has been further used to predict instantaneous microstructural state. The possibility of incorporating strain in the constitutive relati on is discussed. It is noticed that optimum co nditions of strain rate and temperature reduce the influence of strain on mechanical behavior, probably through microstructural changes which respond to the equal but opposite effect of strain hardening and strain softening. a~!~t w~rgE I1T03~79}jig5~B. Division of Materials Science, DOE/OBES,

It has been found that conventionally processed high strength aluminum alloys are not superp1astically fonnab1e, primarily because the grain size is too large for them to · exhibit superplasticity. Recently developed thennomechanica1 processing methods were used to refi ne the gra ins i z~ in the fo 11 owi ng alloys: 7475 A1 produced from ingot, 7075 A1 produced from powder, and X7091 produced from powder. Superpl asti c eva 1uati on of these materials showed that while grain refinement was possible with all three materials, that 7475 A1 alloy exhibited the most superplastic behavior, and tensile elongations as high as 600 to 1200% were observed. The P1M 7075 and X7091 alloys exhi bited tens il e elongations of less than 300%, and were much less fonnab1e t~an the 11M alloy. The reason appears to lie in the coarser recrysta 11 i zed grai n size in thermomechanica lly processed P/M alloys as well as the oxide inclusions.

9:10 a.m.

10:30 a.m.

ENHANCED SUPERPLASTICITY AND STRENGTH IN MODIFIED Ii -6%Al-4%V ALLOYS, J. A. Wert, Rockwell International Sc i ence Center, Thousand Oaks, CA 91360; N. E. Paton, Rockwell International, 600 Grant Street, Pittsburgh, PA 15219

ON NEED OF PRIOR DEFORMATIONS FOR INVESTIGATING STEADY-STATE FLOW BEHAVIOR OF A SUPERPLASTIC Fe-Cr-Ni ALLOY, B. P. Kashyap and A. K. Mukherjee, Division of Materials Science, Department of Mechan ical Engineering, University of California, DAVIS, CA 95616 Superplastic behavior of IN-744 microduplex steel was investigated by differential strain rate and constant strain rate tests mostly at 1000oC. Differential strain rate tests in the order of i ncreasing strain rates between lxlO-5 to 2xlO- 3 s - l for repeated cycles and a constant strain rate test at lxlO- 4 s-1 were conducted. This investigation on different initial microstructures obtained after annealing of the as worked material for a to -30 hour s at lOOOo C show strain hardening fullowed by either steady stat e or strain softening in flow behavior depend ing on the annealing time. Parameters of the constitutive relation for superplastic deformati o ns are estimated for both initial and final microstructures. Interdependence in the magnitude of these parameters due to microstructural instability is discussed. A comparison is made between static and dynamic annealing to obtain the subsequent steady state flow behavior. This work is' supported by Division of Materials Scinece, DOE/OBES, Grant # DE AT03-79ER 10508.

Ti-6%Al-4%V exhibits optimum superp1astic properties at approximately 927°C. Alloy additions designed to· lower the optimum superp1astic fonning temperature must: (1) stabilize the a phase and (2) accelerate the rates of creep processes that contribute to superplastic deformation. Fe, Co and Ni are a-stabilizing elements that have rapid diffusion rates in a-Ii. Several modi fied alloy compositions have been prepared with additions of Fe, Co and Ni to the Ti-6%A1-4%V base alloy . All of the modified alloys display extensive superp1asticity at B15°C. The superplastic properties at B15°C modified alloys with additions of either 2% Ni or 1% Fe + 1% Co are nearly identical with the superp1astic properties of the base alloy at the conventional fonning temperature of 927°C. Furthennore, the room temperature strength of the modified alloys is higher than that of base alloy due to solid solution strengthening by the a-stabilizing additions. Sponsored by AFOSR, Contract No. F49620-BO-C-0030

10:50 a.m. 9:30 a.m. EVALUATION OF SUPERPLASTICITY OF TITANIUM ALLOYS BY USING A 'III!HISPHERE TEST' K.C. Wu, Northrop corpotation, Aircraft Division, Hawthorne, California 90250 Superplasticity of metals was discovered over half a ce,ntury ago. However, i t is only recently that superplssUc forming has been used for fabrication of aircraft structures with significant COlt and weight savings. There are several methods which are used to determine superplasticity of materials which is characterized by the strain-rate senatti vlty index m. All of these methods use uniaxial tensile tests to determine the m value, while in actual forming operations, sheet metal 1s subjected to biaxial stresses. A 'hemisphere test I was developed .to evaluate superp1asticity under a biaxial stress condition providing superior results to those obtained from conventional uniaxial tensile tests. The 'hemisphere test' is described in this paper. The 'hemisphere test' provides a simple and quick method for determining the relative formability between various alloys and various heats of the same alloy by comparing the Umes to form a sheet into the shape of a hemisphere at a predetermined constant pressure. The shorter the forming Ume, the better the formability. It is very useful as a material acceptance test method for material that will be superplssUcally formed. This method also can be used to determine the optimum forming parameters by measuring the forming time required to reach the shape of a hemisphere at various temperatures and gas pressures. This paper describes the use of this method to obtain m values for Tl-6Al4V and Ti-15V-3Cr-3Sn-3Al alloys. The maximum value of m for Tl-6Al-4V at 927C was 0.66 at a strain rate of 3.5xI0- 4 sec- I and the ...:d. . . . . .1... of m for Ti-15V-3Cr-3Sn-3Al at 816C was 0.45 at a strain rate of 2.47xlO-3 sec-I.

9:50 a.m. NON-STEADY TO STEADY STATE SUPERPLASTIC DEFORMATION OF A 25.7 Cr6.6 Ni STAINLESS STEEL, B. P. Kash.vap and A. K. Mukherjee, Division of Materials Science, Department of Mechanical Engineering University of California, DAVIS, CA 115616 A microduplex stainless steel containing 25.7% Cr, 6.6% Ni, and 66.324% Fe was investigated under the condition of a non-steady state in flow behavior. Differential strain rate tests conducted to collect stress (a) - strain rate (e) data indicate that the behavior depends on the dire ction of strain rate change, amount of strain involved, the range of st rain rates covered, and the preceding tes t hi stor y of the specimen. Changes in strain rate sensitivity and flow stress at 800 and 10000C for repeated number of cycles during the differential strain rate tests will be discussed. With increasing amount of deformations the effect of strain on flow stress is reduced and i sotropic strength unique to the strain rate is attained. Explanation for such transition is suggested . This work is supported by Division of Materials Science, DOE/OBES, Grant No. DE AT03-79ER10508.

CAVITATION DURING SUPERPLASTIC DEFORMATION:' D.S. Wilkinson and C.H. Caceres, Department of Metallurgy and Matenals SCIence, McMaster University, Hamilton, Ontario L8S 4Ml Canada

The accumulation of cavitation damage during superplastic deformation has been studied in a copper alloy, Coronze 638. The rate of damage accumulation with strain shows two distinct stages. Below a critical strain EC I damage (essentially the volume fraction of clivi ties) accumulates slowly . For strains above ECI the level of damage increases much more rapidly. The critical strain is insensitive to strain-rate over a large range, but is influenced by the inclusion content of the material. These data, along with others that will be presented, lead to the conclusion that damage is nucleated at inclusions and grows slowly until cavi ties are large enough to link up with one another. Damage then coalesces rapidly, leading to failure. *Research sponsored by the Natural Sciences and Engineering Research Council (Canada) and by a fellowship to one author (CHC) from th'e Consejo Nacional de Investigaciones Cientificas y Tecnicas de la Republica Argentina.

TUESDAY - OCTOBER 26, PM CHEMISTRY AND PHYSICS OF RAPIDLY SOLIDIFIED MATERIALS: I Sponsored by the TMS Chemistry and Physics of Metals Committee, TMS Corrosion and Environmental Effects Committee, and ASM-MSD Phase Transformation Activity Cervantes Convention Center Tuesday, October 26, 1982 Room 273 2:00 p.m. Sessicn Chairman: B. J. Berkowitz, Exxon Research and Engineering Co., P.O. Box 101, Florham Park, NJ 07932.

2:00 p.m. RAPID SOLIDIFICATION-AN OVERVIEW, B. H. Kear, Exxon Research and Engineering Co., Linden, NJ 07036 The recent surge of activity in rapid solidification provides a good example of how advances in processing can catalyze new ideas and activities across the broad spectrum of structure, properties and performance of materials. Although much of the activity has been directed towards achieving high cooling rates from the pre-alloyed liquid state, the emphasis has now shifted towards understanding

TMS Fall Meeting 37 the many and varied phenomena characteristic of rapid solidification. In particular, rapid quenching gives rise tohigh degrees of undercooling, so that nucleation and growth kinetics, and interfacial effects, exert a profound influence on the resulting microstructure. Larg~ degrees of undercooling promote compositional uniformity. In the limi t, massive (segregationless) solidification can occur, and in certain alloy chemistries metallic glasses are formed.

4:10 p.m. RAPID SOLIDIFICATION OF HIGHLY UNDERCOOLED ALLOY DROPLETS-PART I, Y. Shiohara, M. G. Chu, K. C. Flemings, Massachusetts Institute

of Technology, Cambridge, Massachusetts 02139 Using classical nucleation theory and a regular solution approximation, calculations are presented on effects of experimental variables on maximum undercoolings obtainable in pure metal and

alloy droplets. Variables examined include cooling rate before nucleation, droplet Size, alloy composition, and assumed composi-

tion of the critical nucleus. Pure tin and tin-rich tin-lead

2:30 p.m. SURFACE MODIFICATION BY THE LASER AND ELECTRON BEAM GLAZING. P. R. St.rutt, B. G. Lewis, and Mohan Kurup, Dept. of Metallurgy, University of Connecticut, Storrs, CT 06268. The formation of modified surface layers is readily facilitated by a rapidly scanning laser or electron beam. In using this technique on iron-base alloys (with. a carbon content in the range 0.25 to 2.5wt.%) it has been shown that homogeneous layers may be obtained having highly refined, and in some cases unique, microstructures. These features contribute to increased mechanical strength, wear, and corrosion resistance. Experimental work and a detailed heat flow analysis has been made to determine the criteria for obta:!-ning large area surface layers by a scanning electron beam without microstructural modification due to overlapping passes. The theoretical model considers the effect of oscillatory beam frequency, scanning width, power density and translation velocity. Process parameters may be selected to control the (i) amount of alloying element segregation, (ii) car-

alloys are used as examples for calculation. Heat flow and solidification of the droplets during and after recalescence are then considered, assuming the droplets are suspended in a fluid medium.

Solution of this problem requires assumptions regarding solidification kinetics.

Two cases considered are (1) thermodynamic

equilibrium at liquid-solid interfaces during solidification above the solidus, and (2) maximum thermodynamically permitted .composition of the solid phase forming.

Possible implications of effects of volumetric changes during solidification are also discussed.

4:30 p.m. RAPID SOLIDIFICATION OF HIGHLY UNDERCOOLED ALLOY DROPLETS-PART II M. G. Chu, Y. Shiohara, M. C. Flemings, Massachusetts Institute

of Technology, Cambridge, Massachusetts

bide morphology and distribtuion, and (iii) martensite and

Calculations are pre-

sented using tin-lead alloys as example, with oil as the fluid medium. Implications of the nucleation and heat flow calculations for structure of the solidified droplets are discussed.

02139

bainite distribution.

3:00 p.m. COMPOSITION ANALYSIS IN RAPIDLY SOLIDIFIED POWDERS VIA SCANNING TRANSMISSION ELECTRON MICROSCOPY, John B. Vander Sande ,Department of Materials Science and Engineering, M. I. T. ,Cambridge, MA 02139 As a benefit of the rapid solidification of alloys, more refined microstructures are developed. These more homogeneous microstructures must be interpreted if the fundamentals of solidification at rapid solidification rates are to be understood and if the microstructure inherited from solidification is to be deconvoluted from the post-consolidation microstructure. Fortunately instrumentation capable of high spatial resolution for compositional analysis is available. In scanning transmission electron microscopy a fine beam of electrons is used to produce a high resolution image as well as exciting signals from the sample which allow for compositional analysis. The techniques of energy dispersive X-ray analysis, electron energy loss spectroscopy, and

microdiffraction will be highlighted in this presentation.

These

techniques have been used to elucidate the microstructure of rapidly sol~dified stainless steel powders which has, in turn, led to a more thorough Wlderstanding of the solidification behavior of these alloys at rapid solidification rates.

Experimental work is described on undercooling and structure of tin-lead droplets, emulsified in oil. Undercoolings obtained range up to about 100°C; droplet diameters are in the range of ID-ZOllm and smaller. Cooling rates before nucleation range from about .04 K/ s to cooling rates approaching those of gas atomization processes. The higher cooling rates are obtained by wheel

casting and by liquid quenching.

Structures obtained in this

work depend strongly on undercoo1ing, cooling rate before nucleation, droplet size, alloy composition, and rate of heat extraction after nucleation. These structures are compared with those anticipated from the analyses of Part L

DESIGN REQUIREMENTS FOR TITANIUM ALLOYS II: Aerospace Applications Sponsored by the TMS Titanium Committee Tuesday, October 26, 1982 Cervantes Convention Center 2:00 p.m. Room 266 Session Chairmen: R. W. Stusrud, Detroit Diesel Allison Division, General Motors Corp., Indianapolis, IN 46206; J. C. Chesnutt, Rockwell International Science Center, Thousand Oaks, CA 91360.

3 :30 p_.m.

2:00 p.m.

RAPIDLY SOLIDIFIED Fe-Si BASED Cll.YSTALLINE RIBBONS, N.

STRUCTURAL APPLICATIONS OF TITANIUM ON THE B-1B STRATEGIC AIRCRAFT. G.R.Keller and J.J. Olristiana, Rockwell International, North American Aircraft operatlOns, El Segundo, CA.

Tauya, K. I. Arai, Research Institu'te of Electrical Communication, Tohoku University, Sendai, Japan

lilny rapidly solidifed crystallized ferromagnetic ribbons were found to have excellent mechanical and magnetic properties, and for example, Sendust ribbon was already used as an

audio magnetic recording head. High silicon iron ribbons with grain oriented (l00) [Okl] texture .have coercive force lower than 60 mOe and nearly zero magnetostriction, and large scale experiements are in progress. In iro'!l-cobaltsilicon alloy, the simultaneous small anisotropy and magnetostriction is considered and the similar decrease of the coerci ve force is expected when the same texture can be developed in rapidly solidified ribbons. As a result, on

Titanium alloys comprize 18% of the structural weight of the B-IB. The applications all utilize the major characteristic of titanium that makes it attractive for aircraft design i.e. its use results in low structural weight to achieve one or more of the following properties: high fracture toughness and good fatigue crack growth resistance; high strength and/or stiffness; good retention of properties at elevated temperature; and excellent corrosion resistance. This paper describes four structural applications on the B-lB and relates each application to the properties of titanium that influenced its use.

the ribbon surface, highly grain oriented (l00) [Okl] texture was appeared by annealing in a low vacuum. ab about

12000 C, and the saturation magnetiziation of Fe69Co13SilS ribbon is 14.S kG.

3:50 p.m. PRODUCTION OF ALTERNATIVE PHASES DURING LIQUID-TO-SOLID NUCLEATION IN SMALL DROPLETS~ Thomas F. Kelly, Department of Materials Science and Engineering, M.I.T., Cambridge, MA 02139 More than one distinct solid phase is sometimes observed to crystallize from small dorplets of liquid. For example, both fcc and bcc ha ve been observed in many dropl et-processed i ron-base alloys. Calculations of nucleation temperatures of fcc and bcc show a transition from fcc to bcc rrucleation with decreasing nucleant potency (increasing liquid supercooling). It appears that bcc is present as an alternative solid phase that nucleates only at large liquid supercooling. Guidelines are developed from nucleation theory which describe some necessary thermodynamic conditior5 for the appearance of alternative nucleation phases. Examples of the potential microstructural importance of alternative nucleation phases, especially from rapid solidification processes, are presented. *Financial support provided by the Office of Naval Research, Contract number N00014-81-K-0013.

2:30 p.m. DESIGN AND VERIFICATION ANALYSES ON TITANIUM STRUCTURES: J. B. Chang Rockwell International, North American Aircraft Operations, P.O. Box 92098, Los Angeles, CA 90009 Due to its high strength-to-weight ratio, high toughness, good corros i on res i stance, and excellent hi gh temperature performance, the use of titanium alloys for primary structural components such as the wing-carry-through structure of an advanced aircraft was a natural consequence. Because of the durabil ity and damage tol erance requi rements, fati gue crack growth ana lysi s needed to be perfonned. This paper describes the procedures of performing such analysis. Examples of the correlation study conducted on the spectrum test data are also presented in this paper.

3 :00 p.m. - Break

3:30 p.m. DESIGN REQUIREf.1ENTS FOR TITANIUM ALLOYS IN AIRCRAFT GAS TURBINE ENGINES: T.K. Redden and R.A. Sprague, Material and Process

38 TMS Fall Meeting Technology Laboratories, Aircraft Engine Business Group, General Electric Company, Cincinnati, Ohio 45215 This paper presents a discussion of the current state of the art of the app1 ication of titanium alloys in aircraft gas turbines as well as possible future directions for development. These applications are viewed with respect to the specific design properties required for critical rotating engine components such as fan and compressor disk and blades. Titanium alloy requirements are next measured against the properties of current and developmental alloys, and competitive materials, to highlight areas requiring improvement to extend their use in engines. Emphasis will be placed on 1 imitations imposed by fatigue and fracture mechanics considerations.

4:00 p.m. APPLICATIONS OF TITANIUM ALLOYS IN DESIGN OF SMALL GAS TURBINE ENGINES: Robert F. Kirby, G'arrett Turbine Engine Company, P. O. Box 5217, Phoenix, AZ 85010. Titanium alloys are used extensively in the design of gas turbine eng i nes. Des i gn requ i rements for these alloys vary wi th the specific app1 ication, with certain of these requirements being unique to small gas turbine engines. Types of alloys used and the criteria for component design are discussed along with vari ous problems encountered in opt imi zi ng des i gns . The options available for manufacture of components (forging, casting, and powder metallurgy) are discussed in terms of their impact on design requirements, particularly with regard to fatigue and fracture toughness properti es. Fi na lly, antici pated future design requirements are discussed and correlated with material requirements.

2:40 p.m. PHYSICAL AND MECHANICAL PROPERTIES OF AN ALUMINIDE COATING: H. Kaufman, General Electric Co., Thomson Laboratory, Aircraft Engine Group, Lynn, MA 01910 Oxidat-ionl corrosion resistant coatings based on NiA1 are widely used on many auperalloys. In order to calculate the effects of the coatings on the mechanical behavior of their substrate, some physical and mechanical information for the coating itself is required. This work presents data for Codep B-1 coating compositions including coefficient of thermal expansion, thermal conductivity, specific heat and modulus of elasticity from room temperature to about 2000'F (1093'C). The strain capability (strain to crack) of Codep B-1 coating on two alloys (Rene I SO and Rene' 125) on thin flat and round bar specimens was determined at room temperature to about l800°F (982°e) at tensile test and creep test strain rates.

3:00 p.m. HIGH TEMPERATURE FATIGUE FAILURE CHARACTERISTICS OF AN ALUMINlDE COATED DS SUPERALLOY: P. K. Wright, General Electric Co., Aircraft Engine Group, Mail Drop M-S7, Evendale, Ohio 45215

4:30 p.m. TITANIUM PRODUCT DESIGN - A PRODUCER'S VIEWPOINT: Harry Rosenberg, TIMET, 400 Rouser Road, Pittsburgh, PA 15230

cate the trend towards defining a coating system for integral nozzles with improved ductility while retaining the oxidation! corros ion protection of the baseline aluminide. The performance! ranking of these coatings has been described in a paper to be published. This paper discusses the results of the metallurgi.cal characterization of the candidate coatings and the mechanisms of coating degradation in thermal fatigue and oxidation/ corrosion and its effects on the superalloy substrate.

w.

Jus t as performance requi rements have become more demandi ng, so also have the property demands on titanium products. In many cases, these demands have been met by minor compositional or process adjustments. In others, new materials have been required. To date, materials and product engineers have been equal to the task. This paper discusses some of those events in light of an evolution in design philosophy.

EFFECTS OF COATINGS ON SUPERALLOY MECHANICAL PROPERTIES Sponsored by the TMS High Temperature Alloys Committee Tuesday, October 26, 1982 Cervantes Convention Center 2:00 p.m. Room 267 Session Chairmen: C. G. Beck, Westinghouse R&D Center, 1310 Beulah Rd., Pittsburgh, PA 15235; J. H. Wood, General Electric Company, 1 River Road, Bldg. 53, Rm 316, Schenectady, NY 12345.

2:00 p.m. THE EFFECT OF OVERLAY COATINGS ON HIGH CYCLE AND THERMAL FATIGUE: D. A. Woodford, General Electric Company, Corporate Research and Development, Schenectady, NY 12308 Gas turbine buckets may fail by several separate or synergistic modes. Complex failure mode analysis is particularly important for coated alloys where some properties may be enhanced and some degraded. These points are illustrated with recent data on high cycle and thermal fatigue. In the former case, using a high strength eutectic superalloy as a substrate and several plasma sprayed overlay coatings, it is shown how failure of the composite is controlled by the properties of the coating; in a load controlled test the coating is actually strain cycled and responds according to its cyclic strain capability. In the latter case, thermal fatigue crack initiation in a cast Rene' 80 substrate is strongly influenced by the coating composition. Although mechanical compatibility may influence the response, evidence is presented which points to environmental considerations, and particularly the grain boundary penetration of oxygen, as being the dominant factor leading to crack initiation.

2:20 p.m. CHARACTERIZATION OF COATING DEGENERATION IN THERMAL FATIGUE AND OXIDATION/CORROSION:SubhashK.Naik, Avco-Lycoming,Materials Technology Laboratory, 550 So. Main Street, Stratford, Connecticut 06497

The severe cyclic thermal exposure associated with the hot section of the advanced turbine engines causes therITlal fatigue cracking of the coated superalloy cOITlponent (megral nozzle vanes) with premature failure. Cyclic burner rig tests were conducted on a variety of developmental coatings (aluminides, plasma spray. plasma spray + alurn.inide) in an effort to indi-

Fatigue cracking of environmentally protective coatings in a low ductility manner at low temperatures has been _recognized. At higher temperatures these coatings become quite ductile, and coating cracking characteristics are significantly different. This high temperature fatigue regime has been examined more thoroughly for a Codep A1uminide coated DS superal10y, Rene SOH, and the interaction between high and low temperature failure modes during thermal mechanical cycling is described. Factors leading to coating crack development and the extent to which these cracks influence subsequent fatigue in the substrate will be discussed.

3 :20 o.m. -3 :40 p.m. - Break

3:40 p.m. THE EFFECT OF PROTECTIVE COATINGS ON THE HIGH TEMPERATURE PROPERTIES OF y'-STHENGTHENED Ni-BASE SUPERALLOY: R. Castillo and K. P. Willett, Turbine and Generator Division, Westinghouse Canada Inc., Ontario, Canada. The present study examined the influence of chrome-aluminide coatings on the creep and stress rupture properties of a wrought creep resistant nickel-base superalloy used 'in highly stressed gas turbine blade applications. Creep and stress rupture tests were conducted on coated and uncoated wrought bars in the fJilly heat treated condition in the temperature range of 750-800°(:. The tests showed that the application of the chrome-aluminide coatings caused a marked deterioration in rupture strength and ductility. Time to rupture and ductility were tabulated and are presented as log-log plots of stress and reduction in area versus time and as Larson-Miller parametric curves. Light optical microscopy revealed that creep cracks generally initiated at oxide spikes in surface connected grain boundaries. Masking procedures used, to protect the turbine blade roots during coating of the Ni-base supera1loy did not affect the rupture strength or rupture ductility. The mechanical behavior in the coated creep resistant alloy was correlated with the microstructure and is discussed in terms of possible controlling processes.

4:00 p.m. COATING SUBSTRATE DIFFUSION INTERACTIONS AND THEIR EFFECTS ON MECHANICAL PROPERTIES: R. Darolia, General Electric Co., Aircraft Engine Group, Mail Drop M-85, Evendale, Ohio 45215 Diffusional interactions between turbine airfoil alloys and their protective coatings are becoming increaSingly important due to higher" operating temperatures and the delicate balance required of some key ·elements in advanced turbine airfoil alloys. The environmental and mechanical properties of the coating as well as those of the substrate are affected by such interactions. This presentation will describe _the degradation of the stress rupture properties of a directionally solidified eutectic alloy due to coatings resulting, from coating!substrate inter-diffusion. The results are compared with the coating effects on a conventionally cast superalloy where. the coating was shown to enhance the stress rupture capability of the alloy by providing environmental protection.

TMS Fall Meeting 39 4:20 p.m. YTTRIUMIZING - A WAY TO IMPROVE HIGH TEMPERATURE PERFORMANCES OF ALIlMINlDE COATINGS: D. C. Tu, Institute of Materials Science

and Engineering, ~ational Sun Yat-Sen Univ., Kaohsiung, Taiwan, Republic of China The beneficial role of Y in high-temperature protective coatings has been previously recognized. Y-containing MCrAIY overlay coatings are currently applied by PVD or plasma spray where high cost and line-ot-sight requirements restrict their application to simple geometries and expensive parts. In this study, the feasibility of adding Y into aluminide coatings by conventional pack cementation processes has been investigated experimentally and by thermodynamic calculation. Yttriumizing was performed by two different duplex processes. In the first process, Ni, Nil5%Cr and Udimet 500 samples were yttriumized first in a NH 4Cl activated pack followed by aluminizing in various Al activity packs. The second process was the reverse of the first one. Electron microprobe analysis confirmed the presence of Y in coatings formed by either process. The first process produced coatings with higher Y concentration. The deposition rate of Y was adversely affected by substrate (or coating) Al concentrations. The concentrations and distributions of Y in the coatings were analysed and compared with theoretical predictions. The oxidation test results are also discussed.

4:40 p.m. THE ROLE OF CERIUM ON REFINING, CREEP AND EMBRITTLEMENT BEHAVIOR

OF A SUPERALLOY: M. C. Carroll, F. Consandey and J. K. Tien, Henry Krumb School of Mines, Columbia University, New York, NY 10027; C. E. Maurer, Special Metals Corp., New Hartford, m' 13413 This study investigates the role of various amounts of Ce addition ranging from 0 to 0.12 wlo Ce on refining of 0 aI'd S i:1 the melt as well as the effect of residual Ce on tensile and creep properties of an iron-nickel base superalloy U-901. The kinetics of refinement are presented. The reduction if! both 0 and C content in the alloy as well as residual Ce content result in an improvement on the creep ductility of the alloy. These improvements in properties will be discussed with respect to the refining process of 0 and S on Ce addition and the role of residual Ce on structure and microchemistry observed at grain boundaries. An optimal residual Ce content of O. 016 ~do is found for which ductility is maximum. At higher Ce content, however, eutectic Ni5 Ce-type phases are found at grain boundaries which are responsible for the embrittlement. The role of Ce on the well-established 1000°C oxidation embrittlement aspect of superalloys is also examined.

FRACTURE IV: Titanium Alloys TMS-AIME General Abstract Session Tuesday, October 26, 1982 Cervantes Convention Center 2:00 p.m. Room 260 Session Chairman: Dr. H. L. Marcus, Materials Science & Engineering Dept., University of Texas at Austin, Austin, TX 78712.

smaller than in press forged alloys. Such small subgrains may be advantageous ~n retarding the fatigue crack growth rate.

*

NRC Fellow, on leave from Kobe Steel, Kobe, Japan.

2:40 p.m. CORRELATIONS OF FRACTURE FEATURES TO FATIGUE CRACK GROWTH BEHAVIOR OF A TITANIUM ALLOY: ~,'-letcut-Materials Research Group, P.O. Box 33S11 , WPAFB, Ohio, 4S433, F. H. Froes, AFWAL/Materials Laboratory, WPAFB, Oh i 0, 4S433, and G. R. Yoder, Naval Research Laboratory, Wash i ngton, O. C., 2037S. The effect of mi cros tructure, strength and oxygen content on fat i gue crack growth rate of CORONA-S alloy (Ti -4. SA l-SMo-l. SCr) was i nvestigated for two oxygen levels (0.10 and 0.19wt% O2 ), two volume fract ions of primary alpha, and two strength 1eve 1s for a tota 1 of eight conditions. In spite of the large difference in the oxygen levels, tensile strength levels (1l2Ksi-166Ksi) and microstructures, the fatigue crack growth rates were almost identical for all conditions. The objective of this fractographic study was to investigate the reason for this behavior. It was found that the fracture behavior of all eight conditions was almost identical, consisting of faceted fracture at the lower 6K region and extensive branch crack i ng at the higher 6K 1eve 1 regi ons. The faceted fracture features were found to be related to groups of alpha plate ali gned in the same di rect ion. The secondary cracking was found to be mainly along the interfaces of those plates. Thus, in this case, mi cros tructura 1 features are more important than strength or oxygen level in deterrning the FCGR.

3:00 p.m. THE TEMPERATURE DEPENDENCE OF SUSTAtNED LOAD CRACKING IN Ti -6211 :

C. 1. Hoffmann, J. E. Cox, and R. U. Judy, Jr., Naval Research Laboratory, Physical Metallurgy Branch, Washigton, DC 20375 An experimental program was undertaken to evaluate the effect of temperature on the sustained load cracking behavior of a Ti-6211

alloy

(Ti-6%Al-27.Cb-l%Ta-l%Mo).

Crack growth

rates

in

air were measured by compliance methods on side grooved WOt specimens over the temperature range of -20°C to 70°C. The titanium alloy being tested consisted of a Widmanstatten microstructure which had been charged with hydrogen to concentrations of approximately 250 and 1,000 ppm of hydrogen. The influence of temperature on the sustained load crack growth rates at the two different levels of hydrogen concentration Is examined in order to define the mechanisms and rate controlling parameters involved. Differences in the observed crack growth rates are evaluated in terms of the possible thermally activated processes. The results of the current study are compared with similar published studies on' the temperature dependence of sustained load cracking in other titanium alloys.

3:20 p.m. LOW CYCLE FATIGUE BEHAVIOR OF STA Ti-10V-2Fe-3Al WITH AND WITHOUT BETA FLECK MICROSTRUCTURAL HETEROGENEITIES: J. E. Costa,

2:00 p.m. FATIGUE OF TITANIUM ALLOYS: M.R. Mitchell and J. C. Chesnutt, Rockwe 11 I nternat i ona 1 Sc i ence Center, Thousand Oaks, CA 91360 Considerable fatigue data have been generated recently for titanium alloys, both at this laboratory and others. Low-cycle fatigue (LCF) and high-cycle fatigue (HCF) data for Ti-6Al-4V, Ti-4.SA1-SMo-1Cr (CORONA-S), Ti-2Al-12V-2Sn-6Zr (Transage-134) and Ti-lOV-2Fe-3Al will be reviewed and compar.ed. Both stresslife (S-N) and strain-life (~ - 2N f ) data will be presented,

where available. The effect ~f microstructure on fatigue behavior of each alloy will be discussed, and compared to microstructural effects on fracture behavior, such as fatigue crack propagation and fracture toughness. This work was supported in part under Air Force Office of Scientific Research contract, F49620-80-C-0030, Naval Air Systems COlMland contract NOOOI9-79-C-OS40, and Office of Naval Research contract N00014-79-C-OS67.

Carnegie-t:ellon Univ., Pittsburgh, PA. 15213 and J. A. Hall, TIMET, Henderson, NV. 89015 The influence of beta flecks on low cycle fatigue (LCF) behavior of the near beta alloy, Ti-IOV-2Fe-3AI, in the solution treated and aged (STA) conditions was investigated. Both fieck containing and fleck free materials were obtained from the same heat in order to minimize macroscopic chemistry differences. Pancake forgings were prepared from both kinds ef material and then heat treated. Tensile and smooth LCF test results failed to distinguish the flecked material. Scanning electron fractography and detailed metallography in the fracture initiation zone of LCF specimens revealed some subsurface crack initiation associated with grain boundary alpha. It is concluded that grain boundary alpha contributes to the life limit, in low cycle fatigue of STA Ti-10-2-3, and that the beta flecks do not directly provide init iat ion sites, however an indirect potential cont ribut ion is identified. A discussion of the thermal and thermomechanical origins of grain boundary alpha films is presented.

3:40 p.m. 2:20 p.m. FATIGUE PROPERTIES OF COMMERCIAL TITANIUM ALLOYS AT ROOM TEMPERATURE AND ELEVATED TEMPERATURES: s. Fuj ishi ro and T. Matsumoto * Wright-Patterson AFB OH 45433.

Effects of metallurgical variables on fatigue properties of Ti alloys including T1 62425, Ti 6246 and 1M! 829 have been studied at room temperature and elevated temperatures. The alloys were either hammer forged or .isothermally press forged in the beta or beta + alpha field and were subsequently heat treated at supraor sub-transus temperature to produce various microstructures. In general, the fatigue crack growth rate for the transformed beta microstructure is much slower than that for the equiaxed primary alpha plus beta microstructure on all the alloys investigated. The alloy that was hammer forged at the supra-tranSlls temperature provided the slowest FCC rates. Transmission electron microscopic studies revealed that in the hammer forged alloy dislocations were more tangled and the subgrain sizes were much

SINTERED Ti-6Al-4V ALLOY:

A FRACTURE SURFACE CHARACTERIZATION,

C. Welsch, and R. Borath, Case Western Reserve University, Cleveland, Ohio 44106, and Deutsche Forschungs- und Versuchsanstalt fur Luft und Raumfahrt, 5000 Kaln 90, W. Germany. A mechanically weak area in the sintered alloy was opened by fracture in liqUid nitrogen. A number of features in the fracture surface were analyzed by scanning electron microscopy and by scanning Auger electron spectroscopy. The results will be discussed. This research was supported by Deutsche Forschungs- und Versuchsanstalt fur Luft- und Raumfahrt.

4:00 p.m. FATIGUE CRACK PROPAGATION OF AN 0. + B TITANIlN ALIDY: S. Shih, F. S. Lin and E. A. Starke, Jr., Fracture & Fatigue Researcn--

40 TMS Fall Meeting Laboratory, Georgia Institute of Technology, Atlanta, Georgia 30332 The fatigue crack propagation behavior of Ti-6Al-2Ml-lTa-0.8/-b having a Wictnanstatten structure has been studied. The slip distance was changed from the" + B colony size to the alpha platelet size hy transforming the beta to martensite. This modification significantly increases the strength with no loss in ductility. The improvement in strength can be explained in terms of the Hall-Petch relationship. Concomitant with the strength improvement there is an improvement in fatigue crack propagation resistance. This property change can be explained in terms of closure effects and modification of the crack growth mechanisT.l.

4:20

p.m.

Har

DUCTILITY LOSS IN THE ALLOY TITANIlJl 6211: D. Bowden, F. S. Lin and E. A. Starke, .Ir., Fracture and FatIgue Research Lahoratory, Georgia Institute of Technology, Atlanta, r.A 30332

Many alpha + beta titanium alloys show significant high temperature ductility loss in certain heat treated conditions. For the alloy Ti-6A1-2Nb-lTa-O.8~1o (Ti-62ll) this ductility dip occurs in the temperature range 750-850·C. In this investigation, the ductility of a variety of microstructure, produced hy lahoratory heat treatments will be examined from room temperature to the heta transus temperature. lIot ductility will he cOITelated with microstructure, deformation behavior, and fracture characteristics with the goal of understanding the metall urgica] factors involved in the observed hot ductil i ty loss. 4:40

p.m.

THE INFLUENCE OF HYDROGEN ON THE MULTIAXIAL DEFORMATION OF BETA AND ALPHA-BETA Ti ALLOY SHEET:

p.m.

IMPURITY SEGREGATION AND HYDROGEN EMBRITTLEMENT OF Fe and Ni:* R. H. Jones, S. M. Bruemmer, M. T. Thomas, and D. R. Baer, Pacific Northwest Laboratory. Richland, Washington 99352

The results of experiments to evaluate the relative effects of S, Sb and P segregation on the fracture mode and ductility of Fe and Ni at cathodic potentials will be presented and discussed with respect to H embrittlement processes. Grain boundary compositions were determined by AES. fracture mode and ductility with straining electrode tests and intergranular crack growth rates and fracture thresholds with O.IT - CT samples tested at cathodic potentials. The relative effects of segregated Sb, Sand P can be expressed by the following relationships for equivalent grain boundary S concentrations: 1) Cs eq(Fe) = 5C sb + Cs + (lp Cp 2) C:q(Ni) = O.5C sb +C s +O.07C p The effect of S in Fe and Ni was very similar, Sb was lOx more effective in Fe than Ni while P had a small effect in Fe and Nt. *Research supported by the Office of Basic Energy Sciences of the U. S. Department of Energy.

2:40 p.m. INTERGRANULAR HYDROGEN EMBRITTLEMF.NT OF Ni: EFFECTS RELATED TO INTERCRYSTALLINE MISORIENTATION:· M. B. Hintz and I.. A. Heldt, Dept. of Metallurgical Engineering, Michigan Technological University, Hou~hton, MI 49931.

The effects of intercrystalline misorientation and r;rain boundary chemistry on intergranular hydroeen embrittlement susceptibility have been measured for Ni bicrystals. The obser'/ations are discussed in terms of the misorientation dependence of interfacial energy. *Research sponsored by the U.S. Department of Energ,y-.

Barbara K. Lograsso and D. A.

KOss, Dept. of Metallurgical Engineering, Michigan Technological University, Houghton, MI 49931. The sensitivity of two Ti alloys in sheet form to hydrogen embrittlcment under multiaxial loading conditions has been examined. Uniaxial tensile and punch-stretch tests have been performed on Ti-30V containing '\40 or "-'2000 wt ppm H as well as Ti-6AI-4V with either "-'70 or "-'250 ppm H. Tests of Ti-30V in uniaxial tension indicate that H causes a small decrease of the elongation to failure but an increase in the true strain to failure at the fracture surface. Preliminary fracture limit diagr::'ms indicate that I unlike the behavior of commercially pure Ti I there is no pronounced effect of the stress .state on the hydrogen embrittlement of either Ti-30V or Ti-6AI-4V at the hydrogen levels/microstructures examined. This program is supported by the Office of Naval Research.

GRAIN BOUNDARY CHEMISTRY AND ENVIRONMENTAL INTERACTIONS I Sponsored by the TMS Corrosion and Environmental Effects Committee Tuesday, October 26, 1982 Cervantes Convention center 2:00 p.m. Room 261 Session Chairmen: Russell H. Jones, Battelle-Northwest Laboratory, P.O. Box 999, Richland, WA 99352; Lloyd A. Heldt, Michigan Technological University, Houghton, MI 49931

2:00

2:20

p.m. REVIEW PAPER:

GRAIN BOUNDARY CHEMISTRY AND ENVIRONMENTAL INTERACTIONS IN NI-BASE ALLOYS: R.M. Latanision, H.H. Uhlig Corrosion Laboratory, Hassachusetts Institute of Technology, Cambridge, MA 02139

This presentation will focus largely on work performed in this laboratory in order to better understand the intergranular hydrogen embrittlement of cathodically charged nickel and nickel-base alloys. It has been observed that intergranular separation occurs over distances in bulk specimens that are far larger than the depth of penetration of hydrogen expected from lattice diffusion. There is indirect evidence to suggest that grain boundary segregation of metalloid elements such as P, S, Sb. Sn, etc., may play a role in hydrogen absorption at grain boundaries. Such elements are known to be poor electrocatalysts for the hydrogen evolution reaction and, therefore, their presence at grain boundaries has led to the proposal that the accumulation of such elements may stimulate local absorption. Historically, however, there has been little success in demonstrating that absorption and transport may occur preferentially at grain boundaries. In this presentation. our recent efforts to detect grain boundary transport by electrolytic permeation techniques and Ion Microprobe AnalYSis will be described. Dislocation transport will also be considered.

3:00

p.m.

[FHCT OF GRAIN ROUNlJARY CHEMISTRY AND HYDROGEN ON I~TERCRANUrAR FRACTURE OF IRON* K. S. Shin and M. Meshii, Department of

Materials Science and fngineering, Northwestern University, Evans tOll, I L 60201 It is known that grain boundary chemistry and hydrogen can significantly affect grain boundary strength.

The mechanism of hydro-

gen-induced intergranular fracture and the role of grain boundary chemistry, however, still remain unclear. In this study, the grain boundary strength of iron was quantitatively correlated to sulfur concentration at the gra in boundary by Auger electron spectroscopy after in-situ fracture. The effect of hydrogen on grain boundary strength was examirled at various grain boundary sulfur concentrations and hydrogen contents. It was found that the grain boundary strength decreaspd significantly with increasing grain boundary concentration of sulfur and hydrogen charging current dens i ty. The partia 1 recovery of the hydrogen charging effect took place at room temperature.

In the light of

present observations, the mechanism of intergranular fracture due to hydrogen and the role of impurity will be discussed. *This research was supported by the U. S. Department of r:nergy.

3:20

p.m.

ATOMISTIC STUDIES OF GRAIN 80UNDARY SEGREGATION: * ,9ui-Jir. Wang, A. P. Sutton and V. Vitek., D.!pt. of Materials Science, University of Pennsylvania, Philadelphia Pennsylvania, 19104 The relationship between the atomic structure of grain boundaries and the propensity to segregation has been investigated using a computer model. The calculations have been performed for Cu-Bi, Cu-Ag and Au-Ag systems. It has been found that in the case of small concentration of the segregant, the segregation is very selective and the segregation propensity is principally control-led by the local structure while practically independent of the chemical and phYSical properties of the solute and solvent. The possibility of the increase of the concentration' of the segreg .... ant in the boundary, is, however, determin~d by the latter. This is connected with possible increase or decrease of available segregation sites with increasing concentration of the segregant associated with decrease in selectivity of the segregation sites and substantial changes in the grain boundary structure. *Research supported by DOE Contract 110. DE-AC02-79ER70429.

3:4<)

p.m.

THE EFFECTS OF IMPURITY SEGREGATION ON INTERGRANULAR EMBRITTLEMENT AND CAVITATION* I-Wei Chen, Massachusetts Institute of Technology, Cambridge, MA 02139

The effects of the equilibrium impurity segregation to the grain boundary on intergranlllar failure are considered. It is found

TMS Fall Meeting 41 that the partitioning of impurities between the fractured surfaces has an important influence on the failure nucleation process by lowering the activation energy barrier. J'he magnitude of the above thermodynamic effect is comparable to the chemical effect on the cohesive strength which is commonly explained in terms of interatomic bonding. Effects of alloying element and propensity for interfacial segregation will be discussed in this light.

examined. It 1s shown that the larger the grain size and the greater the Bi content in the range from 0 to 0.01%, the more severe is the LME due to Hg, and an additional increment of severity is shown when the Bi is allowed to segregate to the grain boundaries. The segregation increment reaches a maximum and remains constant when there is sufficient Bi present to saturate the grain boundaries. The results are interpreted in terms of the bond-breaking model of !.ME, modified to include grain boundary segregation effects.

*Research supported by the U.S. DOE under Contract No. EG-77-S02-4461.

3:00 p.m. EMBRITTLEMENT OF 4340 GUN STEEL BY COPPER:

R.M. Fisher, A.Szirmae

U.S. Steel Corp. Research Lab, Monroeville, PA 15146 and M.H.

4:00 p.m.

Kamdar, LCWSL, USA ARRADCOM, Watervliet, NY 12189

l

XPS AND lIES STUDIE OF THE BONDING OF S tIO Ni SURFACES:* T. Miyahara, \ D. Reed and H. K. Birnbaum, Materials Research Laboratory * and Department of Metallurgy and Mining Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801. We have examined the bonding of S to Ni surfaces using XPS and Auger methods to determine whether segregated solutes such as S alter the nature of the metal bonding at the grain boundaries. The S was added to the clean surfaces of Ni by segregation from the solid solution or by interaction with H2S gas •. Pure Ni, NiS and Ni 2S were used as reference materi a 1. Measurements of the S 2p, Ni 2p and Ni va 1ence band spectra suggest that S adsorbed on Ni did not exhibit significant d band bonding in contrast to the ionic compounds NiS and Ni 2S. The temperature dependence of the adsorbed S concentration was determined to obtain estimates of the bonding enthalpy to the Ni surface. *Thi s work was supported by DOE contract DE-AC02-76ER01l98.

During firing, the bore surface of cannon tubes is subjected to brief but damaging exposure to high temperatures, and stresses in an extremely reactive atmosphere. The surface is partially coated with copper ev~porated from the rotating copper bands. The deterioration of the surface is mainly due to intense carburization by the hydrocarbon-rich propellant gases which results in the formation of molten Fe-C, causing erosion and cracking in a chec1~er­ board pattern. With some propellants, very sharp cracks extend some 400 pm below the surface into the martensite matrix region and contain Cu, S, and sometimes Al right up to the tip of cracks. This region is well below the melting temperature of any of the embrittling agents. Copper is known as an important liquid-metal cracking agent but is not normally associated with embrittlement at lower temperatures. This example and its laboratory analogs will be discussed in terms of current models of liquid and solid metal embrittlement.

3:20 p.m. REDUCTION IN LIQUID MERCURY EMBRITTLEMENT OF MONEL 400 BY GRAIN BOUNDARY PHOSPHORUS. A. W. Funkenbus ch • United Technologies

4:20 p.m. SCANNING !llJGER AND SIMS STUDIES OF H, S, AND Cl SEGREGATION AT GRAIN BOUNDARIES IN NICKEL:* Hiroshi Fukushima and H. K. Bi rnbaum, Materials Research Laboratory and Departmeii"tOT Metallurgy and Mining Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801. The segregat i on of solutes at gra in bounda ri es is often s tudi ed by fractu ri ng the spec i mens along the bounda ri es and app lyi ng surface analysis techniques to analyze the compositions of the surfaces thus produced. While this technique is suitable for britt 1e so 1i ds it requ i res that the fracture occu r along the interface and that the surface composition thus produced be characteristic of the interface composition prior to fracture. We have developed techniques to study segregation at grain boundaries without requiring fracture along the interfaces. These methods utilize high resolution Auger and SIMS techniques which will be described. We will discuss recent results on the segregation of S, Cl and H at boundaries in Ni to illustrate the technique. *Thi s work was supported by NSF Grant DMR 77-09808.

LIQUID AND SOLID METAL INDUCED EMBRIT· TLEMENT OF METALS III Sponsored by the TMS Corrosio.n and Environmental Effects Committee and the ASM-MSD Corrosion and Oxidation Activity Tuesday, October 26, 1982 Cervantes Convention Center 2:00 p.m. Room 270 Session Chairman: N. N. Breyer, Department of Metallurgical and Materials Engineering, Illinois Institute of Technology, Chicago, IL 60616.

Research Center, East Hartford, CT. 06108, and L.A. Heldt and D.F. Stein, Michigan Technological University, Houghton, MI 49931. The influence of grain bO',lndary phosphorus concentration in Monel 400 on intergranular embrittlement by liquid mercury was investigated. Auger analysis of intergranular fracture surfaces showed that phosphorus segregated to grain boundaries and that its concentration depended upon alloy concentration and heat treatment. Enbrittlement by liquid mercury was reduced by increased grain boundary phosphorus concentration; tensile strength and elongation increased and intergranular fracture decreased. Similar reductions in the severity of hydrogen embrittlement were also measured and attributed to increased grain boundary phosphorus concentration. Results are examined in terms of addititive and interactive mechanisms for reduced embrittlement. It is suggested that phosphorus segregation reduces eoorittler concentration by improving atomic packing along grain boundaries.

3:40 p.m. MISORIENTATION DEPENDENCE OF LIQUID METAL INDUCED INTERGRANULAR FRACTURE OF ZINC BICRYSTALS: •• Watanabe. S •. Shima and S.

Karashima. Department of Materials Science, Faculty of Engineering Tohoku University, Sendai, Japan. The present investigation has been made to reveal the effect of grain boundary structure on the bahaviour of liquid metal induced intergranular fracture and the mechanism of it. Misorientation dependence of liquid gallium induced intergranular fracture was investigated using tilt and twist zinc bicrystals with different misorientation angles. Fracture tests were carried out with a lever type testing machine and the fracture stress and strain were determined. Intergranular fracture surfaces were observed with a SEM and the results were analysed crystallographically. Both tilt and twist zinc bicrystals, the fracture stress and strain depend strongly on the misorienta.tion. In a high angle region beyond 20° the fracture stress decreases with increasing the misorientation angle having a cusp at the angle of about 56° which is close to the r9 near-coincidence orientation.

2:00 p.m. REVIEW PAPER: SOVIET DEVELOPMENTS IN LlQUlD-METAL-EMBRITTLEMENT

A.R.C. Westwood and J.R. Pickens, Martin Marietta Laboratories, 1450 S. Rolling Rd., Baltimore, MD 21227 Recent results from scientists in the Soviet Union will be briefly reviewed, including: Yuschenko and Shchukin's computer modelling of the influence of embrittler atoms at crack tips; Glikman's results and ideas on the nalure of liquid metalenhanced creep; Popovich's views on the role of adsorptionenhanced flow in crack propagation; and Bryukhanova 1 s observat ions of enhanced mach in ing ra tes via incorpora t ion of embri t tling particles in otherwise conventional cutting fluids.

2:40 p.m. GRAIN SIZE, GRAIN BOlI:iDARY SEGREGATION Al'lD MERCURY INDUCED EHBRl1-I'LEMENT OF COPPER: W. R. \iarke, Standard Oil Co.

(Indiana). Naperville, 111-:-66566,-------a-nd U. N. Nanda, General Motors - Electromotive Div., LaGrange, Ill. 60525. The interact ion among grain size, grain boundary sf>grf'~ation of bismuth and mercury induced embrittlement of copper is

4:00 p.m. LIQUID METAL GALLIUM-INDUCED INTERGRANULAR FRACTURE OF ALUMINIUM BICRYS,TALS: T. Watanabe, S. Shima and S •. Karashima. Department

of Materials Science, Faculty of Engineering, Tohoku University, Sendai, Japan. The effect of grain boundary structure on liquid gallium induced intergranular fracture has been studied of aluminium bicrystals containing low angle, coincidence, and high angle random boundaries. They were wetted by liquid gallium and deformed under static or dynamic loading. The misorientation dependence of fracture behaviour has been investigated. Accoustic emission during intergranular fracturing has been also studied in order to obtain some basic information on crack propagation along the different types of grain boundaries. It has been found that the fracture stress is lower for high angle random boundaries than that for coincidence ones. The time to fracture depends strongly on the type and misorientation of grain boundaries, and it increases with increasing the deviation angle from the exact coincidence orientation relationship. Several stages of accoustic emission during fracture has been found which is also dependent on the type and misorientation of grain boundaries.

42 TMS Fall Meeting 4:20 p.m. INTERGRANULAR FRACTURE CAUSED BY LIQUID GALLIUM IN POLYCRYSTALLINE BETA BRASS WITH BCC STRUCTURE: T. lIatanabe, M. Tanaka and S. Karashima, Department of Materials Science, Faculty of Engineering, Tohoku University, Sends!, Japan.

In order to relate the structure of grain boundaries and their fracture behaviour, liquid gallium induced intergranular fracture has been studied in coarse-grained polycrystals of beta brass with bee structure. ECP technique was applied to determine the type and misorientation of grain boundaries contained in polyc rystals. In-situ observations of fracture process were also made to make clear crack propagation process along different types of grain boundaries . It has been found that the low angle or coincidence boundaries are very resistent to lntergranular fracture and a crack was preferentially formed at a high angle random boundary and then propagated choosing grain boundaries having less ordered structures . The frequency of fractured grain boundaries of specific types ( low angle, coincidence, and high angle random) was quantitatively determined.

METALLURGY OF WEAR RESISTING MATERIALS II Sponsored by the TMS Physical Metallurgy Committee Tuesday, October 26, 1982 Cervantes Convention Center 2:00 p.m. Room 262 Session Chairmen: K. J. Bhansali, Metallurgy Division, National Bureau of Standards, Washington, DC 20234; J. D. Ayers, Physical Metallurgy Branch, Naval Research Laboratory, Washington, DC 20375.

40,000 psi). Bearing failure is caused by overheating and design charts are based on temperature limitations. High conductivity of the copper base alloys is useful in this ·context. Metallographic analysis of wear s c ars and debris indicates wear proceeds by metal transfer and localized large strain proccesses. The metal transfer observed does not degrade the bearing performance. Ex tremely fine debris (200.!. to 500.!.) combines with liquid phases in the lubricant to form a gel. High resolution STEM and diffra c tion analysis indicates the small particles are metal and are n o t oxides. The structure of the transfer layers as seen by Transmission Electron Microscopy suggests the layers are agglomerations of the very fine wear debris. No evidence has been found of subsurface fracture in the near surface regions in the wear zones.

4:15 p.m. CARBIDE INJECTED T1 AND Al ALLOYS FOR WEAR APPLICATIONS: J. D. !
A study of the wear resisting behavior of aluminum and titanum base alloys containing particles of TiC or WC introduc ed by the laser melt / particle inje ction process suggests that these materials may have signifi cant potential for use in light weight machinery. The carbide parti c les were injected, by a stream of helium gas, into shallow melt pools produced on the alloy samples as they were traversed under a multikilowatt laser beam which was focused to a diameter of about 2 mm. Repeated overlapping melt passes produced surface lay e rs approximately 1 mm deep containing 15 to 65 volume percent of c arbide. Ground surfaces were subjected to abrassive and particulate erosion wear testing. Kineti c coefficient of friction measurements were made on ground and polished surfaces. Test results are correlated with carbide type, size, and volume fra c tion.

2:00 p.m. WEAR RESISTANT VAPOR DEPOSITED COATINGS: H. E. Hinterman, Laboratoi re Suisse De Recherches Horlogeres, Neuchatel , Switzerland

This paper considers the friction and wear behavior of hard coatings of TiC, TiN, SiC, A1203, some borides of refractory metals against hardened steel, rubies and themselves, respectively at room temperature in humid and dry air. It also considers the beha vior of TiC at 350°C a helium atmosphere (the c onditions in the He-loop of a nuc lear reactor). The adhesion o f TiC and Ti N on cemented carbide, hardened steel, and some s pec ial ~ i or Co-alloys is dis c ussed. The 105s of adhesion and the faIlure mode a re determined b y the scratch test method. This method, whi c h employs a scratch test instrument re c ently developed at LSRH, is desc ri bed. Indust rial applica t ions for which these coatings have proven useful include deep drawing tools, bearings, and general wear parts. The paper discusses some microstructural considerations of the hard coated parts.

2:55 p.m. THE tl ETALLUR GY OF METALS AND ALLOYS FOR IMPROVED BEARING WEAR RESISTA~CE: D. H. Buckley and K. Niyoshi, NASA-Lewis Research Center, Cleveland, OH -4'4C35

The wear of metals and alloys is strongly dependent upon many properties. ~ith elemental metals tensile str e ngth, shear s trength,. cr ystal structure, orientation, transformations, and latti c e parameters are found to e ffe c t wear behavio r . In general, h igh ideal shear strength is avantagous except where texturing and preferred orientation takes plac e at the interface such as in hexagonal metals and shear readily occurs along basal planes. Wear, for example, with cobalt in the hexagonal fonn with basal texturing is markedly less than with that same metal in the cubic f o rm. Variations in orientation also effect wear behavior of elemental metals. Small concen t rations of alloying elements can significantly alter alloy surfac es and dominate not orily the wear mec han i sms but the quantity of wear observed. Elements observed to effect wear of alloys b y this me c hanism in c lude carbon, sulfur, phosphorus, boron and nitrogen. Second phases such as laves, sulfides, c arbides and silicides improve markedly the wear resistance of alloys. The presence or absence of c rystallinity also has an effec t. Metallic glasses offer greater wear resistance in the amorphous than in the crystallic state. The introduction of enviro nmental constituents with alloy components is ext remel y important in wea r behavior. Further, the adapt Ion and r e actions of lubri c ants and lubricant additives with the elemental metal or components of alloys can vary markedly and ac cordingly influence observed wear behavior.

3:30 p.m. WEAR PROPERTIES OF HEAVY LOADED COPPER BASE BEARING ALLOYS: W. A. Glaeser. Battelle Columbus Labs, 505 King Avenue, Columbus, OH 43201

Copper base materials (copper-tin-lead and copper-aluminum) are used for slow mOVing, h£'avily loaded, grease lubrica ted bearings in heavy construction machinery and a irframe bearings . These materials have sufficient ductility to conform under contact stress and spread the contact are a and also have suf f icient yield strength to resist crushing under heavy bearing stresses (up to

MODELING OF THE STRUCTURE AND PROPERTIES OF AMORPHOUS MATERIALS IV Sponsored by the Computer Simulation in Materials Science Activity of ASM and by the TMS Nuclear Metallurgy Committee Tuesday, October 26, 1982 Cervantes Convention Center 2:00 p.m. Room 272 Session Chaimum: J. C. M. Li, Department of Mechanical and Aerospace Sciences, University of Rochester, Rochester, N.Y. 14627; G. S. Cargill, III, IBM T. J. Watson Research Center, Yorktown Heights, N.V. 10598.

2:00 p.m. ATOMIC SCALE DEFECTS IN AMORPHO US SOLI DS: P. Chaudhar i. IBM T. J. Watson Research Cent e r, Yorktown Heights, ~y 10598

Using static and dynamiC c omputer simulation procedures we have investigated the stability of point and line defects in ;::::morphous solids. We find that the vacancy cannot be stabilized in an amorphous s olid characterized by a Lennard-Jones pair potential but can be in one characterized by a Keating potential. The former potential simulates a dense packed me tal-like solid, and the .latter a covalently bo nded silicon-like solid. Edge and screw dislocations can be s t a bilized in an amorphous solid. Motion of dislocations can lead to an an i sotropic change in the density of packing in an amorphous solid. This is in contrast to crystals where simple dislocation glide preserves the local density.

2:30 p.m. MODELLING THE KINETICS OF SHEAR TRANSFORMATION IN METALLIC GLASSES: A. S. Argon, Department of Mechanic a l Engineering, Massachussets Institute of Technology, Cambrid g e, MA 02139

The disordered Bragg bubble raft has been developed as a quantitatively accurate analog me dium for modelling the shear transformations in metallic glasse s without directional bonds. In such bubble rafts, shear transformations and their energetics can be studied Quantitatively by means of a generalized interbubble potential and force law. These studies have reproduced the experimentally measured distributions of both the activation free energies for tr.,ansformations and free volume. They have also demonstrated the existence of an inverse correlation between the level of initial local free volume and the activation energy of a transformation. This has provided a better understanding of struc tural relaxat ions in all types of glas s es undergoing physical aging, and even of the nature of the glass transition process itself .

3:00 p.m. SIMULATION OF DEFOR.'ofATION AND ANNEALING IN METALLIC GLASSES: Shin Takeuchi, Koj i Maeda and Shinj i Kobayashi, The Institute for Solid State Physics, The University of Tokyo, Roppongi, Minato-Ku, Tokyo 106, Japan

Using central force potentials, realistic models of metallic

TMS Fall Meeting 43 glasses of elemental iron and -Cu S7 Zr 43 alloy have been constructed. These models have been deformed in pure shear in computer to reveal elementary deformation processes in the amorphous structures. Using the molecular dynamics method, the model of CU S7 Zr 43 has been annealed at various tempe-ratures to clarify the atomistic processes occurring during structural relaxation. The above results have been discussed in terms of the structural inhomogeneIty in the amorphous structure.

3:30 p.m. - Break

3:45 p.m. MODELS FOR ISO-CONFIGURATIONAL FLOW AND VISCOSITY RELAXATION IN METALLIC GLASSES: S. S. Taao and F. Spaepen, Division of Applied Sciences, Harvard University, Cambridge, MA 02138

Isoconfigurational flow in PdCuSi and PdSi shows non-Arrhenian behavior, which can be explained by fast J reversible atomic rearrangements during temperature cycling. Partial redistribution of the free volume is a possible- model for these rearrangements. At temperatures well below Tg, the Viscosity has been observed to increase linearly with time during annealing. Near Tg, we have now observed a non-linear viscosity increase: the rate slows down as the metastable equilibrium state is approached. We have fit this behavior to hoth a unimolecular and a bimolecular reaction model, with the latter one giving the better fit. Since the equilibrium viscosity was used as a fitting parameter ,we could extract its value without actually reaching the equilibrium state, thus circumventing the crystallization problem. Our results indicate that earlier equilibrium viscosity measurements may be an order of magnitude to low because of insufficient relaxation.

4:00 p.m. MICROMECHANISMS OF PLASTIC DEFORMATION IN METALLIC GLASSES, Vitek, D. Srolovitz* and T. Egami, Department of Materials Science and Engineering, University of Pennsylvania, Phildelphia, PA 19104. * now at Exxon Research, Linden, N.J. 07036

v.

The plastic deformation of a model monoatomic metallic glass has been. studied using computer simulation~ The strain was built up gradually and complete stress-strain curve has been obtained. The atomic level deformation processes have been analyzed continuously during straining. It has been found that large local inhomogeneous atomic movement, associated with the plastic flow, always occurs in the vicinity of regions of large atomic level shear stress. On the other hand, no correlation with atomic level hydrostatic stress and thus with local density fluctuations, has been found. The regions of large local shear stress are farmed during loading and appear to act primarily as stress concentrators analogous to shear cracks. Hence, the pr:incipal microscopic mechanism of flow appears to be formation of shear defects in the vicinity of which a viscous flow develops owing to the stress concentration.

4:15 p.m. RELAXATION-TIME SPECTRA FROM RELAXATION RESPONSE CURVES OF AMORPHOUS METALS, J. R. Cost, Los Alamos National Laboratory, Los Alamos, N~1 87545 To date it has not been possible to obtain directly the relaxation time spectrum which controls the rate at which a relaxation process takes place. Instead, it was necessary to assume a model for the form of the spectrum and then force fi t the data to thi s model. The present investigation uses computer-aided analysis to directly obtain the unique relaxation time spectrum from the relaxation response curve without introducing assumptions concerni ng the spectrum. The method is shown to have good va Ii di ty and accuracy, even for response data with experimental Gaussian errors of up to 5%. Also, since it is capable of resolving relatively fine structure in the spectra, new information can be obtained concerning the kinetics of relaxation processes. Results will be discussed for the amorphous Fe40Ni40Pl4B6 alloy in which the relaxation (as measured by small cli~nges in electrical resistance) caused by atomic reordering following a temperature change is shown to be controlled by a rather complex relaxation time spectrum having four jis~inct peaks.

NOVEL TECHNIQUES IN METAL DEFORMA· TION TESTING IV: Automated Testing Sponsored by the TMS Shaping and Forming Committee and the ASM Process Modeling Activity Tuesday, October 26, 1982 Cervantes Convention Center 2:00 p.m. Room 264 Session Chairman: J. F. Thomas, Wright State University, Material Science and Engineering Program, Dept. of Engineering, Dayton, OH 45435.

2:00 p.m. NEW METHODS FOR BIAXIAL TESTING OF TUBING: N. S. Cannon and C. W. Hunter, Hanford Engineering Development Laboratory, Richland, WA 99352

Anticipated and hypothetical transient events in fast breeder reactors can lead to biaxial loading of the tubular fuel pin cladding. To help determine the mechanical properties of cladding under these conditions, two new types of controlled biaxial strain-rate systems were developed which feature remote operability for testing of irradiated specimens in hot cells. Both systems incorporate computers for data acquisition and test control, and use internal argon gas loading of specimens to produce a constant diametral strain-rate. Tempera,ture, diameter and pressure are measured, and accuracies are on the order of ±2°C, ±1.3 ).lm and ±O.S MFa, respectively. Type 1 tests are performed at constant specimen temperature, ranging from 350 to 1300°C, while strain-rate is controlled at a constant value selected between O.OOl%/s to O.l%/s. Type 2 test specimens are heated through a dual temperature ramp; one rate at 25°C/s and a second at 5S0°C/s. At the initiation of the 550°C/s ramp, the specimen is pressurized at rates to 100 MFa/s until inelastic strain is detected; then the specimen pressure is controlled by a series of high speed, high pressure (1 ms, 170 MFa) valves to produce relatively constant diametral strain-rates to 50%/s. Data from both irradiated and unirradiated specimens have been correlated for inclusion in reactor transient computer codes.

2:20 p.m. AIlTOMAT1I'IG A HOT 'tORSION MACHINE TO SIMULATE ROLLING AND EXTRUSION. J.J. Mills, Walter Merriam & Ken Nielsen, Martin Marietta Labs., 1450 S. Rolling Rd., Baltimore, Maryland 21227

A cotmnercial hot torsion machine, designed by Bossard, was interfaced to a mini-computer through a CAMAC Crate. This machine operates with a lOKW DC motor and an electromagnetiC clutchbrake to rapidly apply the motor torque to the specimen. Torque and axial force are measured using a two-axis piezoelectric transducer. The mini-computer a) controls motor speed,times of opening and closing the clutch, including stopping on specimen break and resets the transducer charge amplifiers, b) acquires torque, axial force and specimen temperature data and c) computes the shear stress and strain from the acquired. data. A detailed description of the system and some results on aluminum alloys will be presented.

2:40 p.m. BIAXIAL DEFORMATION TUBE TESTING:* M. G. Stout and S. S. Hecker, Los Alamos National Laboratory, Los Alamos, NM 87545 Knowledge of the" mechanical response of materials subject to multiaxial loading has become increasingly important because of stringent service requirements and sophisticated tools for analytic elastic-plastic stress analysis. Closed-loop, servo-hydraulic computer-controlled testing machines provide the capability of measuring complex multiaxial material properties - using tubular specimens. Experiments presently possible include investigation of strain hardening for different stress states, transients following stress path changes, multiaxial plastic instability and fracture, large strain torsion and automated yield surface determination. Some important considerations in tube testing are geometric instability, test control and the testing mode. For example, there are differences in sheet versus tube diffuse instability data which result purely from specimen geometry. One must also consider which is the most effective means of controlling the experiment; load, strain or displacement signals can all be used. Insight must be used in selecting testing mode. Tensioninternal pressure tests provide information both different than and complementary to tension-torsion experiments. . Sponsored by the Department of Energy, Office of Basic Energy Sciences, Division of Materials Science.

3:00 p.m. AN EXPERIMENTAL FACILITY FOR PRECISE MECHANICAL PROPERTIES MEASUREMENTS AT ELEVATED TEMPERATURES: E. T. C. 110 and S. R. MacEwen

A test facility consisting of an environmental chamber, a remote extensometer and a miniature three-zone furnace has been developed for use with a computerized model 1116 mstron. Experiments can be carried out in either vacuum or inert gas, in a stainless steel chamber. '!be temperature of the specimen is controlled by three 125 ).ID. diameter thermocouples which are spot-welded or laser-welded to the test specimen. OY'er the 2.5 cm guage length of the specimen the temperature is constant to ± 0.5 K, for temperatures up to 1000 K. The remote extensometer allows a standard clip-on extensometer to be used to monitor strain, and provides continuous output with a resolution that can be as good as ± 1 x 10- 5 , depending on the characteristics of the clip guage. Examples of tensile tests, creep tests, stress relaxa-tion, and stress-change tests using specimens of Zr-base alloys will be presented.

3:20 p.m. MEASURi:NG BACK STRESSES IN COMMERCIAL ALLOYS AT ELEVATED TEMPERATURES:* W. B. Jones and R. W. Rohde, Sandia National Laboratories, Albuquerque, NM 87185

44 TMS Fall Meeting A back stress has been measured in 316 stainless steel and Alloy 800 at 922k with the strain transient dip test. These measurements have been conducted on both a servo-hydraulic test machine and a specially designed deadweight dip test machine. The short time «3s) transient response for these alloys if; complex but this does not prevent the unambiguous determination of a back stress. There is close agreement between the raw data from the two different test machines despite the inferior step loading capability of the servo-hydraulic machine. TWo methods of identifying the back stress' are compared: a) by the presence of zero strain rate after the load decrement anq b) by usin9: maximum change in strain after several decrements as suggested by Blum. The latter t e chnique was found to provide a more precise definition of back stress. While servo-hydraulic machines have been found to be deficient for transient tests in pure metals, they perform adequately for such experiments in conunercial alloys.

3:40 p.m. DETERMINATION OF MATERIAL PARAMETERS IN CONSTITUTIVE EQUATIONS: C. S. Hartley, Mechanical. Engineering D"e partment, Louisiana State University, Baton Rouge, LA 70803 Determination of material parameters appearing in constitutive equations requires interactive testing methods in which the specimen response controls the testing environment in order to reduce the number of independent variables in the I!lechanical test. Closed-loop testing under displacement or load. control can produce test environments in which only the total true strain, total true strain rate, total ext~nsion rate, load sustained by the specimen or average stress on the minimum cross-section remain constant while other variables change. This work presents analyses of tests, illustrating the .relationships among measured quantities and derived quantities appearing in a typical incremental constitutive equation. Use of the procedures described leads to more accurate determinations of the material parameters by eliminating the need for intermediate calculations of stress, strain and strain .rate. The analyses also present the form of the correction required for elastic strains in each case. An example of a data analysis procedure designed to deter'Dine material parameters in two types of constitutive equations illustrate an application of this approach.

2:00 p.m. PRODUCTIVITY ADVANCES IN SPECIAL METALS: Joe Klein, Manaqer of Mill Operations, Cabot Corporation, Kokomo, Indiana.

2 :45 p.m. PRODUCTIVITY ADVANCES IN THE UNITED STATES STEEL INDUSTRY: Harry paxton, Vice-President Research, U.S. Steel Corporation, Pittsburgh, PA.

3 :30 p.m. PRODUCTIVITY ADVANTAGES ASSOCIATED WITH MINI-STEEL MILL OPERATIONS: John A. Savage, Manager Personnel Services, Nucor Corporation, Charlotte, NC.

4:15 p.m. PRODUCTIVITY ADVANCES IN MINING AND PRODUCTION OF NON-FERROUS MATERIALS: Jack Goff, Executive Vice-President, AMAX, Inc., Golden, CO.

RADIATION DAMAGE ANALYSIS FOR FUSION REACTORS III. Microstructural Evolution-Experiment Sponsored by the TMSIASM Nuclear Metallurgy Committee Tuesday, October 26, 1982 Cervantes Convention Center 2:00 p.m. Room 271 Session Chairmen: G. R. Odette, Department of Chemical and Nuclear Engineering, University of California, Santa Barbara, CA 93106; L. K. Mansur, Metals and Ceramics Div., Oak Ridge National Lab., P.O. Box X, Oak Ridge, TN 37830.

4:00 p.m. PLANE STRAIN TENSILE TESTING AT ELEVATED TEMPERATURES: S. Baik and R. Raj, Department of Materials Science and Engineering, Bard Hall, Cornell University, Ithaca, NY 14853 Fundamental aspects of mul tiaxial fracture can bE! studied by a comparison of results from mechanical testing under specific stress-states, e.g. simple uniaxial tension, plane-strain tension, hi-axial tension, and testing within a variable hydrostatic pressure. In this paper, a comparison of measurements from simple tension and plane-strain tension will be reported. The material was Ti-6242 (in the B annealed condition) and the temperature and strain-rate range were similar to hot-working condi tions •

4:20 p.m.

HYDRAULIC School of IN 47907, Division,

BULGE TESTING AT CONTROLLED STRAIN RATE, J. E. Bird, Materials Engineering, Purdue University, West Lafayette, and R. F. Young, Engineering Properties and Design Alcoa Laboratories, Alcoa Center, PA 15069

Using servo-control and electronic feedback, hydraulic bulge tests can be run at ~train rates that are controlled throughout a test. This technique permits flow stress, strain hardening, and strain rate sensitivity to be measured in sheet materials up to the onset of localized necking under conditions of constant true strain rate The power of the technique wi 11 be illustrated by several new experimental results: a family of flow stress curves obtained at constant strain rates of either I, 10, or 100 times 2.4 x .10-·s-'; flow stress response after strain rate jumps of ±10 times or ±100 times between these rates; and strain hardening during these tests up to the onset of localized necking. The technique appears especially powerful for studying mechanical behavior of alloys that display negative strain rate sensitivity in uniaxial tensile tests.

PRODUCTIVITY IN EXTRACTIVE METALLURGY: MELTING, REFINING, CASTING: Session II of GEM/G-PAC Productivity Symposium Sponsored by the TMS-AIME Governmental, Energy and Minerals Committee and the ASM Government and Public Affairs Committee Tuesday, October 26, 1982 Cervantes Convention Center 2:00 p.m. Room 275 Session Chairman: George R. St. Pierre, The Ohio State University, 2041 North College Rd., Columbus, OH 43210.

Abstracts for the presentations in this session will be printed in the Addenda to the Pocket Program distributed at the meeting.

2:00 p.m. VOID GROWTH IN A PULSED IRRADIATION ENVIRONMENT:* J. L. Brimhall, E. P. Simonen, L. A. Charlot, Pacific Northwest Laboratory, Richland, WA 99352 Pulsed fusion irradiations could induce microstructures that are significantly different from steady irradiation microstructures. Void annealing during each pulsed annealing period can affect irradiation-induced swelling. In the present investigation, void growth has been studied in Ni by TEM after ion irradiation in a pulsed mode. Previous results have shown that large voids are resistant to anIlealing during pulsed operation and hence, pulsing had no measurable effect on swelling. Microstructures with much smaller void size and higher void density than studied previously have been irradiated in a pulsed mode. The pulsing conditions have been chosen to verify theoretical models of void growth and swelling. Criteria for predicting effects of pulsing on irradiation microstructures will evolve from these studies.;'

*Research supported by the Office of Basic Energy Sciences, U.S. Department of Energy.

2:20 p.m. EFFECTS OF PULSED DUAL-ION IRRADIATION ON PHASE TRANSFORMATIONS AND MICROSTRUCTURE IN Ti-MODIFIED AUSTENITIC ALLOY:. E. H. Lee, N. H. Packan, and L. K. Mansur, Metals and Ceramics Division. Oak Ridge National Laboratory, Oak Ridge, TN 37830 The response of metals and alloys to pulsed irradiation is a topic of scientific interest that is of relevance to fusion reactor materials development. Recent research on radiation effects in complex alloys has shown that phase instability as well as microstructure is sensitive _to irradiation variables. Motivated by these results we have initiated a program to study the effects of pulsed irradiation and helium injection. Irradiations with 4 MeV Ni++ ions and 0.2-0.4 MeV He+ ions were carried out at 950 K to doses of I, 10, and 40 dpa with a helium injection rate of 20 appm/dpa. Four irradiation conditions were used: (a) continuous Nt ion beam, (b) pulsed Nt ion beam, (c) continuous dualion bombardment, and (d) pulsed dual-ion bombardment. The pulsed mode consisted of chopping the beams synchronously in a 60 s on/60 s off cycle. The results are discussed in terms of mechanisms of phase instability and mlcrostructuraJ. development. Implications of these results for future research are described. *Research sponsored by the Division of Materials Sciences, U•. S. Department of Energy under contract W-7405-eng-26 with the union Carbide Corporation.

2:40 p.m. DEPTH DEPENDENT DAMAGE PROFILE IN STAINLESS STEEL IRRADIATED WITH HE-IONS: K. Shiraishi and K. Fukai, Division of Nuclear Fuel

TMS Fall Meeting 45 Research, Japan Atomic Energy Research Institute, Tokai-mura, Ibaraki-ken 319-11, Japan

Type 31g stain1e~s steel was irradiated with 24 M<;V He-ions up to 3 X 10 ions!= at ambient temperature. Hardness peak occurred at a distance of 105 ~m from irradiated surface in depth dependent micro-hardness measurement and black dots were seen in the depth ranged from 106 to 114 ~m with the number density peak at 110 ~m in cross sectional microstructure observatione The damage profile is in fairly good agreement with the calculation by means of the extended E-DEP-l computer code; displacement damage peaks at 108.0 ~m and iojected helium distributes in the width of 7.6 ~m with a peak at 108.3 ~m. A pair of parallel lines of 8 ~m apart at a distance of 107. 3 ~m from the irradiated surface was revealed on the side surface of the specimen by chemical etching after postirradiation annea1iog for 1 h at 750·C. The lines are appeared to arise from large Frank loops formed in narrow bands adjacent to heavily damaged region. Heli\DD bubb lee were observed in the range of 106 to 116 ~m from the irradiated surface in the specimen.

3:00 p.m. EFFECT OF HELIUM PRE INJECT ION AND PRIOR THERHOMECHANICAL TREATMENT ON THE MICROSTRUCTURE OF TYPE 316 55:* A. Kohyama+'++ G. Ayrault+, A. P. L. Turner+ and N. Tgata++, ~terials Science Division, Argonne National Laboratory, Argonne, IL 60439, ++Department of Materials Science, University of Tokyo, Bunkyo-ku, Tokyo 113, Japan Samples of 316 55 were preinjected with 15 appm of heli= either hot (650 0 ) or cold (room temperature) and irradiated with 3 MeV Ni+ ions to a dose level of 25 dpa at 625 0 C in order to test the validity of helium. preinjection as a means of simulation of transmtitant helilDD. production. Results for preinjected single-ion irradiated samples were compared to samples irrsdiated with 3 MeV Ni+ and Simultaneously injected with helium at a rate of 15 appm He per dpa of displacement damage (dual-ion irradiated samples). Preinjected samples exhibited bimodal cavity size distributions. Pre injected samples of solution annealed and solution annealed and aged material showed lower swelling than dual-ion samples, but 20% cold worked samples showed greater swelling. *Work supported by the U. S. Department of Energy and US-Japan FUsion Cooperation Program.

3:20 p.m. INFLUENCE OF HELIUM INJECTION SCHEDULE AND PRIOR THERMOMECHANICAL TREATMENT ON THE MICROSTRUCTURE OF TYPE 316 55:* A. Kohyama+'++, G. Ayrault+, and A. P. L. Turner+, %terials Science Division, Argonne National Laboratory, Argonne, IL 60439 and +tDepartment of Materials Science, University of Tokyo, Japan The influence of different helilDD. injection schedules on microstructure development in Ni+ ion irradiated 316 55 at 6250 C is discussed. Injection schedules were chosen to (1) approximate the MFR condition and (2) mimic the mixed-spectrum reactor condition. Dual-ion irradiation to 25 dpa produced strongly bimodal cavity size distributions in solution-annealed and solution-annealed and aged samples, whereas Single-ion irradiation followed by dual-ion irradiation to the same dose produced a cavity size distribution with a substantial component of intermediate-size cavities. fualion irradiation produced only very small cavities in 20% cold-worked material, while Single-ion followed by dual-ion irradiation produced some intermediate-size cavities and greater swelling. *Work supported by the U. S. Department of' Energy, and US-Japan fusion Cooperation Program.

3:40 p.m. MICROSTRUCTURAL/MICROCHEMICAL COMPARISON OF AI&I 316 IREADIATED IN HFIR AND EBR-II:* H. R. Brager and F. A. Gamer, Westinghouse Hanford Company, Richland, WA 99352 In an attempt to predict the influence of helium on macroscopic property changes of structural metals in fusion environments, comparisons. have been made of the microstructural evolution of cold-worked AISI 316 irradiated in the EBR-II and HFIR reactors. While displacement rates in these two reactors are comparable, there exists large differences in neutron spectra and in the generation rates of both helium and various solid transmutants. A series of papers have been published which address the microstructural and microchemical differences that occur in the two environments. In some areas the previously available specimens have been inadequate to unequivocably answer all questions relevant to fusion reactor design. In particular there are two divergent interpretations concerning the influence of helium and coldwork on the steady state swelling rate. New data from a solutionannealed specimen help to provide a more definitive description of the role of helium on swelling and microstructural development.

characterized Fe-15Cr-25Nl alloy, comparative irradiations were performed using fast neutrons, 1.0 MeV electrons, 0.75 MeV protons and nickel ions at three incident energies, 2e8, 3.5 and 5.0 MeVe In the original publication neutron data were not available above 14 dpa. Neutron-in'duced swelling and microstructural data are now available to doses as large as 60 dpa. Comparison of these data with the charged particle data have brought to l!ght some important observations of relevance to charged particle simulation studies. In particular, it appears that self-ions are only 20-25% as effective as fast neutrons in creating voidage. Helium/ dpa ratios for charged particle simulation of fusion environments should probably reflect the lower effectiveness of ion-produced displacements. *Research sponsored by the U e S. Department of Energy.

4:20 p.m. EVOLUTION OF CAVITY SIZE DISTRIBUTIONS IN DUAL-ION IREADIATED AUSTENITIC STAINLESS STEEL AND Fe-Cr-Ni TERNARY ALLOYS: J. A. Spitznagel, B. O. Hall, We Je Choyke, and J. N. McGruer*,-Westinghouse R&D Center, Pittsburgh, PA 15235, *Universlty of Pittsburgh, Pittsburgh, PA 15260. Two important design parameters for fusion reactor first wall structures are the incubation period for the o:nset of steadystate swelling and total swelling, both obtainable from the cavity size distribution function N(r,t). Information regarding microstructural stability during temperature and damage rate transients and the effects of helium to dpa ratio and microchemistry on swelling is also contained within the distribution function. We present experimental data for the fluence and temperature dependence of the first four moments of the function N(r, t) in dual-ion bombarded single phase high purity Fe~Cr-Ni alloys and polyphase austenitic stainless steel. The results are compared with comprehensive rate-theory model predictions for the effects of helium, displacement rate, and -temperature on moments of the distribution function. Discrepancies are in many cases attributable to local inhomogeneities not treated by the modele

4:40 p.m. ATOM PROBE FIELD ION MICROSCOPY OBSERVATIONS OF HELIUM-VACANCYSOLUTE CLUSTERING IN Fe-Cr-Ni TERNARY ALLOYS: J. A. Spitznagel, Westinghouse R&D Center, Pittsburgh, PA 15235; M. K. Miller and Se S. Brenner, UeS. Steel Research Laboratory, Monroeville, PA 15146; W. J. Choyke, Westinghouse R&D Center. The potential usefulness of austenitic stainless steel for fusion reactor first wall and blanket structural applications is greatly enhanced if the alloy(s) exhibit only low swelling similar to the transient or "incubation dose" regime observed in fission reactors. There has been considerable speculation on the relative roles of helium and local microchemistry in establishing the duration of this transient. We are presently applying dual-ion irradiation and atom probe field ion microscopy (APFIM) techniques to produce and image small helium bubbles in the size range beneath the visibility limit of the transmission electron microscopee Time-of-flight mass spectrometry is being used to measure local chemical compositions at the bubble/matrix interface as a function of helium concentration, displacements per atom and bubble size in 12Ni-15Cr-73Fe and 30Ni-l5Cr-55Fe alloys. We report on progress to date.

SURFACES TMS~AIME

General Abstract Session

Tuesday, October 26, 1982 Cervantes Convention Centel 2:00 p.m. Room 261 Session Chairman: Dr. R. A. Petkovic-Luton, Exxon Research & Engi neering, P.O. Box 45,· Linden, NJ 07036.

2:00 p.m. DECONVOLUTION OF AES SPECTRA: M. Baron, Westinghouse R&D Center, 1310 Beulah Road, Pittsburgh, PA 15235. A mathematical method is developed which enables the deconvolution

of experimentally obtained AES spectra, to yield true concentration profiles versus depth in alloys. The method assumes an experimentally determiried broadening function which can be measured by profiling a sharp interface between the materials of interest. The method can be used for on-line analysis of AES spectra to yield true composition versus depth profiles. The most general case where the depth resolution depends on sputtering depth is treated in detail.

*Research sponsored by the U e S. Department of Energy.

2:20 p.m. 4:00 p.m. CORRELATION OF CHARGED PARTICLE AND NEUTRON-INDUCED RADIATION DAMAGE: THE ADIP EXPERIMENT REVISITED, * F. A. Garner, Westinghouse Hanford Company, Richland, WA 99352 In 1976 the results were published of an ambitious U.S. intercorrelation experiment designated the Alloy Development Intercorrelation Program (ADIP) Experiment. Using a simple well-

MONTE CARLO STUDIES OF SURFACE SEGREGATION IN BINARY ALLOYS: H. Ke Chang and Je Ke Lee, Dept. of Metallurgical Engineering, Michigan Technological Uni versi ty, Houghton, Michigan 49931 Equilibrimn surface segregation of substitutional solute atoms is studied for the eu-Ni and Cu-Zn (a.-brass) binary systemse The atomic interactions are represented by the· Johnson-type potential functions e The Metropolis-type Monte Carlo method is applied to

46 TMS Fall Meeting determine the surface segregation, with which surface free energy is computed. The (100), (110), and (Ill) surface orientations are studied. For the Cu-Ni system, the surface segregation predicted by the Monte Carlo method is found to be in good agreement with recent experimental data. It is also found that the Cu segregation can decrease the surface free energy of the Cu-Ni binary system by 25% at 750 K. In the Cu-Ni system, the concentration profile of the Cu segregation decreases monotonically, whereas in the a-brass system, the concentration profile of the Zn segregation shows an oscillatory behavior. This work was supported by the Division of Materials Research of NSF under Grant 7805741.

2:40 p.m. PHASE DIAGRAM OF THE 2-DIMENSIONAL Ni (111) -S SYSTEM FROM LEED AND AUGER SPECTROSCOPY: R. Ramanathan and J .M. Blakely, Materi-

als Science

&

Engg., Cornell University., Ithaca, NY 14853

Using Low Energy Electron Diffraction and Auger Electron Spectroscopy the ordered phases in the 2-dirnensional Ni (Ill) -S system have been examined. The results are presented in terms of a I temperature vs. coverage I phase diagram. At relatively low coverages a well ordered Ni(lll) p(2x2)S structure was observed. This undergoes a reversible transformation as temperature is cycled. At an intermediate narrow coverage, a two phase structure was observed for the first time in this system. This structure consists of Ni (Ill) p(2x2) S and complex pattern similar to the one reported by other investigators. The reversible transformation temperatures of these structures were evaluated. At high coverages only a complex LEBO pattern was observed. This structure undergoes a reversible transformation at high temperatures (Le., >300·C). Based on our experimental data we have attempted to solve some of the discrepancies reported in earlier work.

3:00 p.m. MODEL OF THE SURFACE OF PdSOSi20 METALLIC GLASS:" W. Kowbel and W. E. Brower, Jr., Engineering Mechanics and Materials Department Southern Illinois University, Carbondale, IL 6290l. Surface properties of metallic glasses are critical to such surface related phenomena as their high corrosion resistance and their catalytic selectivity and activity. On crystalline surfaces the adsorption energy of adatoms has been observed to be higher in the vicinity of a step than in the middle of a terrace, indicating the importance of the coordination numbers of .surface sites for catalytic behavior. A model is needed to characterize the surface of metallic glasses in terms of their distribution of coordination numbers. We propose a coordination polyhedra surface model similar to the bulk model proposed by Gaskell. We assume no surface segregation and no reconstruction· (only relaxation) of the surface layer of the glass. Based upon this model the radial distribution function (RDF) and the partial (Pd-Pd) RDF were determined for the surface atoms of amorphous Pd80Si20' The distribution of coordination numbers indicated that the surf ace of metallic glass is very rich in low coordination sites. *Work sponsored by U.S.D.O.E.

~:?O

p.m.

DIRECT IMPLANTATION OF ALUMINUM INTO NICKEL. M. Ahmed, D.Potter, Metallurgy Department and Institute of Materials Science, UniverSi ty of Connecticut, Storrs, CT 06268 and B. Cordts, VarianExtrion, Gloucester, MA 01930. Surface chemistry changes and subsequent phase t.l:ansformations were investigated following the direct implantation of 180 kV Al+ions into nickel substrates. The implanted aluminum concentration, determined using Auger analysis, increased linearly with dose up to 1.8xl0 18 ions/cm Z reaching -:: . 70 at%Al, and did not change appreciably beyond this dose. The aluminum distribution was flat to within ±l5% of the average implanted concentration and extended to depths of ::::0 .1 jJm. Phase transformations occurring in these spec imens during ag ing up to 60 min. at 600"C were investigated using TEM. The phases observed depended on iinplantation dose and included y and y ' -Ni 3Al (0.6xl0 18 ions/cm 2 ), Ni2Al and B'-NiAl (1.2xl0 18 ions/cm 2 ) and Ni2Al or Ni2A13 (1.8 and 2.4xl018 ions/ cm 2 ). The phases observed in implanted specimens will be discussed in terms of those expected under equilibrium conditions. Research supported by NSF Grant DIIR 8006084.

fracture plane is located at a lesser depth than the peak of the ion distribution, and correlations between fracture plane location and failure mode are the same. Observed effects of fluence and external load are interpreted in terms of phenomenological models.

4:00 p.m. OPTICAL PROPERTIES OF BLACK CHROMIUM THIN FILMS: * P. Godavarti, Department of Materials Science and Engineering, Northwestern University, Evanston, IL 60201 and K. 1. Vasu, Department of .Metallurgy, Indian Institute of Science, Banga10re 560012, India This paper represents a systematic effort to characterize the optical properties of Black Chromium films on galvanized steel substrates viavis the electroplating variables. Optimum conditions for the best selectivity have been obtained. These results and those of large collector configurations based on this study will be pres~nted.

* Research

was done at Department of Metallurgy, Indian Institute of Science, Bangalore 560012, India.

4:20 p.m. Diffraction Methods for Measurement of Stresses at Adhesive/Adherend Interfaces of Lap Joints: C, S. Barrett and P. Predecki, University of Denver Research Institute, Denver, Colorado 80208. Stresses at adhesive/adherend interfaces become measurable by diffraction with suitable choices of materials, using extensions of methods we have employed for stresses on the outer surfaces of adhesive lap joints. For example, the stress distribution over the AI/epoxy interface of a Be/epoxy/Al lap joint under load is revealed by a CuKa diffracted beam entering and exiting the joint through the :3e adherend, thereby making available a comparison with theoretical predictioris. Stress distributions averaged throughout the thickness of the adhesive layer may also be accessible by modifications of this method.

Research supported by Army Research Office Grant DAAG 29-8l-K-0150.

4:40 p.m. AIR OONTAMINATION STUDIES ON TITANIUM ALLOYS:

R. N. Shenoy and

J. Unnam, Materials' Engineering Department, Virginia Polytechnic

Institute and State University, Blacksburg, VA

24061

Titanium is an excellent candidate for thermal protection systems in hyper-volocity vehicles. Oxidation and oxygen embrittlement are the primary considerations in its potential use at temperatures above 1000°F. Air-contamination studies have therefore been carried out on a commercial purity Ti (A5S) and a 6-2-4-2 alloy in the temperature range of 1l00-1400'F. Oxidation kinetics have been determined by thermograv.i metric method, and oxidized specimens have been studied by XRD, SEll, AES, and metallographic techniques. Oxygen depth profiles have been deduced from X-ray intensity bands and compared with those obtained frota AES; these In turn have been related to hardness depth profiles. Residual mechanical properties have been related to the degree of air-contamination. Oxidation mechanism in the temperature range of llOO-1400'F will be discussed.

WEDNESDAY - OCTOBER 27, AM ALLOY PHASES I TMS -A/ME General A bstract Session Wednesday, October 27, 1982 Cervantes Convention Center 8:30 a.m. Room 264 Session Chairman: Dr. J. F. Smith, Iowa State University, 124 Metallurgy Bldg., Ames, IA 50011.

3:'10 p.m.

8:30 a.m.

INTEGRATED LATERAL STRESS EFFECTS IN HELIUM-IMPLANTED ALUMINUM, MOLYBDENUM, AND SILICON: M. Santella, W. J. Choyke, J. R. Townsend, R. B. Irwin, University of Pittsburgh, Pittsburgh, PA 15260; and B. O. Hall, J. A. Spitznagel and N. J. Doyle, Westinghouse R&D Center, Pittsburgh, PA 15235.

Binary Coherent Phase Diagrams for the BCC Lattice with Pairwise Ordering (1st-Neighbour) and Clustering (2nd-Neighbour) Interactions: II. Yamauchi and R.I. Chittim, Dept. of Engineering IIsterials, University of Windsor, Windsor, Ontario, Canada N9B 3P4.

Implantation of energetiC ions into metals for modification of surface sensitive properties or to study irradiation effects introduces a volume change per implanted atom. In general the dopant atoms serve as centers of dilatation. The resulting ir.tegrated lateral stress may affect the microstructure and result in mechanical failure of the implanted layer at high fluences. In this study tensile and compressive loads have been applied to helium implanted fcc and bcc and diamond cubic single crystals to superpose externally-applied stresses and integrated lateral stresses resulting from ion implantation. The failure in these quite dissimilar materials exhibit several common features: The

Coherent phase diagrams in the first-(Vl) and second~neighbour interaction (V 2) model are calculated for a binary system in the bcc lattice in which VI Is of ordering tendency and V2 is of clustering tendency. There is only one ordered phase, i.e., B2, in the ground state. The calculation is performed by means of the cluster variation method in the tetrahedron approximation. The general features of the present result are qualitatively in agreement with the previous result by Ino which was obtained by use of the Bragg-Williams approximation. The present result also suggests possible observation of spinodal modulation of the long-range order parameter in the B2 phase.

TMS Fall Meeting 47 8:50 a.m. APPLYING THE CLUSTER VARIATION METHOD TO PREDICTING YIY' PHASE FIELD IN SUPERALLOYS: J.R. Barefoot, R.N. Jarrett, J.M. Sanchez and 1.K. Tien, Henry Krumb School of Mines, Columbia University. New York, NY 10027. Superalloy design and re-design (element substitution) efforts can become less tedious and less costly if a predictive method can be developed to determine the y/y' phase fields, i.e. y' volume fraction as a function of the multicomponent composition. In the past, the cluster variation method has been successfully used for binary alloys in which the precipitated phase is coherent with the matrix phase. We are extending this method for application to the multicomponent coherent y' /y nickel-base superalloys. It will be shown that the cluster variation method can accurately describe the equilibrium (incoherent) Yh' phase fields in the binary Ni-Al phase diagram. We have also computed the y' /Y phase field, as a function of temperature, for the Ni-Cr-Al ternary phase diagram. A reasonable fit results between the calculated and the experimental diagrams. The modelling of the six component Ni-Cr-AI-CoMo-Ti base superalloy is also underway. The effect of Ni substitution for Co will be discussed. *Research supported by NASA under grant NASA NAG-3-57.

Bi (II). The heat of fusion of Bi (II) is 6.80 cal/gm. compared wi th 12.92 cal Igm. for the equi 1 i bri um Bi (I). Further observations on the effect of pressure onthe nucleation of Bi (II) hi gh 1 i ght the ro le of undercoo 1 i ng. The support of the ARO (DAAG29-80-K-0068) is gratefull y acknowl edged.

10:30 a.m. MICROSTRUCTURE STUDIES OF Sf'JTTER DE~SITED Culry METALLIC ALLOYS. S. M. Shin, M. A. Ray, J. E. Greene, and J. M. Rigsbee, Oepartment of Metallurgy and Mining Engineering, University of Illinois, Urbana, IL 61801. Metastable CuxCry alloy thin fi lms with three different ~ compos i ti ons (CuO.75CrO.25, and CUO.45CrO.55 and CUO.2CrO 8) were produced by sputter deposltlon onto glass substrates. ror each composition, three different substrate bias voltages were tried (OVs, 30Vs and 60Vs) to study the mixing effect. Study of the films by TEM, in-situ annealing, x-ray diffraction and DTA revealed that substrate bias voltage and film composition affected the crystal structure, grain size and distribution and preferred ori entat ion.

10:50 a.m. 9:10 a.m. PHONON DISPERSION AND THE THERMODYNAMIC PROPERTIES OF A Au-Ni ALLOY:* T. B. WU, J. B. Cohen and W. Yelon, Dept. of Materials Science & Engineering, The Technological Institute, Northwestern University, Evanston, Illinois 60201 Phonon dispersion for Au-40 at pct Ni has been measured above the miscibility gap (", 10230 K). The nearest-neighbor force constants are considerably reduced compared to those for the pure components, whereas those for more distant neighbors are enhanced, implying an increased long-range phonon-electron interaction in the alloy. The overall effect is a structural softening which accounts quantitatively for the published excess entropy. For phonon modes near the Brillouin zone boundary there is an additional softening, which appears to be due to shielding by electrons (from the 3d band of Ni) in the viCinity of the Fermi surface. These effects are more difficult to detect after quenchir& due to the decrease of phonon lifetimes resulting from distortions associated with solute clustering.

HYDROGEN STORAGE WITH PSEUOOBINARIES OF THE TYPE ZR(B x Bi_x)2, B "" CO, FE and B' = MN, CR: Douglas G. Ivey and Dr. D.O. Northwood, Department of Engineering Materials, University of Windsor, Windsor, Ontario, Canada, N9H 3P4. A literature survey, undertaken by the authors has indicated that certain pseudobinaries of the type Zr(B x Bi_x)2, where B, H' = Co, Mo, Fe, V, Mn, Cr and 0 S x ~ 1, have the potential to absorb large quantities of hydrogen up to a hydrogen to metal ratio of greater than 1. 0 and thus may have application as hydrogen storage materials. In the present study, pseudobinary compounds of this type, with B Co, Fe, B' = Mn, Cr and x - 0.4, 0.5, 0.6, have been investigated for their potential as hydrogen storage materials. These alloys exhibit two phase microstructures, identified as the cubic and hexagonal Laves phases. Hydrogen absorption increases the lattice parameters in all cases without changing the crystal structure. Hydrogen capacities and stabilities decrease with 'x', with maximum capacities approaching 1. O. Inoomplete desorption has been observed in all instances and is in the 6S-80~ range of the quantity absorbed. :::t

*This research was supported by NSF.

11:10 a.m. 9:30 a.m. ANALYSIS OF TIIO PHASE EU'rECTOD ';'RA.VSFORIfJ.TION:· Rahul BI\su,Scientist

GTRF.,POlI 7575,Bangalore 560f/5,India. A theordtical JOlOdel is formulated to describe solid state eutect~id tr'Ulsfol"'1"l8.tion. <)olutions incorpoT.'tting appropriate boundary condi tio!1s A.rc deriv~d. for transient and st~ady stA.tes. It iR shown that the transi~nt solution de8crib~8 build up of th~ interfacial boundary layer, whereas the steady state solution extends into the phase.The pr:~sence of this bound'iry layt.ir with co~iderf\hle concentration and tc:nperature cro.d:ents sUbgests a solid state anal':>gue to consti tution-

~~::-~~~!!~-!~~-!~-!~~-~~~~~~-!~-~~!!~-~::~~!~~~~!~~~:-

*

::ese.1.rch 10ne at

~:9.terials

Science depart=ocnt, UCLA.

9:50 a.m. PR<;i)IC7IO:-: OF 1''1': "ITl'CTIC TEMP,HATUR': AND CO''l'OsrroN OF AlA A-NI:. ~ahul Basu,Scienti9t,C:TH~,POB 7575,HAArrRlor~ 56CYl75,Inoia Solut,ion of th~ thcrmonynamie: equilibrium cOllA-tions for the AITaNi t<:'rnary .1.t .... arious temp'1uent eutectoid forJJllltion of g'Jilnrna orirae. • R.ef;earch done fit the Inriian Institt,te of Sci~nce,.)llngalore as r.SI~) Pool ~ci~mtist.

Ii

MONITORING HYDROGEN IN IRON USING IR AND X-RAY ANALYSIS: E. Armacanqui, Y. Chang and J. P. Wallace, Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, MN 55455 Information about the dissolved hydrogen within iron which is a result of free corrosion has provided data which is substantially different than what is found in electrolytically charged iron specimens. The near surface (1000A) concentration of hydrogen seems substantially greater than what is expected from bulk concentrations. Also, step changes in the metals lattice parameter as a function of temperature indicate a series of possible phases are forming in this system. A comparison is made between the near surface state of stress between a freely corroding surface and the electrolytically charged surface to find an explanation between the behavior difference in the two systems.

*Present address:

Casting Analysis, Inc.; Route 1, Box 16; Fort Defiance, VA 24437.

CHEMISTRY AND PHYSICS OF RAPIDLY SOLIDIFIED MATERIALS: II Sponsored by the TMS Chemistry and Physics of Metals Committee, TMS Corrosion and Environmental Effects Committee, and ASM-MSD Phase Transformation Activity Wednesday, October 27, 1982 Cervantes Convention Center 8:30 a.m. Room 273 Session Chairman: R. O. Scattergood, North Carolina State University, Dept. of Materials Engineering, Raleigh, NC

10:10 a.m. Metastable Phase Formation in Highly Undercooled Bi Droplets, D. LaCourt, J,S. Paik and J.H. Perepezko, Univ. of WisconsinMadison, Dept. of Met. and Min. Eng., Madison, WI 53706 The undercooling range of fine (10-20~) Bi droplets in emulsion form has been extended to a maximum equivalent to 0.4lTm. At the maximum undercooling limit a fraction of the droplet sample crystallizes as a metastable phase. The presence of a metastable phase is revealed upon heating by the appearance of a melting enaothermat 174°C in additlon to the melting at 271°C for the stable phase. The melting at 174°C is close to the melting point of the Bi (II) phase based upon the metastable extension of the Bi (II)-liquid phase boundary to 1 atm. A combination of separate and independent measurements of the pressure dependence of the metastable phase melting temperature with calorimetric heat of fusion data has been used to identify the metastable phase as

8:30 a.m. OSTIolALD RIPENING OF RAPIDLY SOLIDIFIED SOLID-LIQUID MIXTURES, P. W. Voorhees, and M. E. Glicksman, Rensselaer Polytechnic Institute, lotlterials Engineering Dept. Troy, N.Y. 12180. A new series of Ostwald ripening and two phase mixtures has been developed which explicitly accounts for the diffusional interaction between the dispersed coarsening second phase. This theo.ry predicts the morphology of coarsening solid/liquid mixtures in terms of time invariant distribution of inter facial curvature. The theory also predicts a dependence of the time invariant interfacial curvature distributions and ripening kinetics on the volume fraction solidified. Experiments were

48 TMS Fall Meeting performed to measure the response of an ultra-precise thermal probe immersed in the coarsening rapidly solidified solid/liquid mixture over a wide range of fraction solids. Through this theory it is now possible to interpret the experimental results to gain a deeper insight into the nature of Ostwald ripening following rapid solidification. A discussion of the theoretical predictions along with a comparison to experimental results will be given.

9:00 a.m. MATHEMATICAL MlDELLING AND EXPERIMENTAL STUDIES OF RAPID SOLIDIFICATION PHENOMENA IN PLASMA SPRAYING, J. Szekely, N. El-Kaddah and J. M::Kelliget, Dept. of Mit. Sci. Eng., M.I.T., Cambridge, Massachusetts

Apart from E phase there have been no reported new crystalline phases observed in rapidly solidified steels. Interest in steels has therefore concentrated on features such as small grain size, reduction in segregation distance j increases in solute supersaturation, and phase stability. structural data will be presented on a number, of rapidly solidified high speed steels and model steels. It is not possible to obtain an p.omogenous structure in high speed steels and observation of precipitation reactions has therefore been undertaken in the simple model steels. Scanning and transmission electron microscopy have been used to monitor the nature of chromiun rich and varadium rich carbides.

10:50 a.m.

02139

By combining heat and fluid flow phenomena in the 'plasma region and carier gas - solid heat transfer with conductive heat transfer in the substrate, a mathematical model has been developed to represent the thermal history of plasma sprayed coatings. The theoretically predicted temperature profiles in the substrate are compared with measurements. Some general conclusions are also drawn regarding the limits on thermal histories tbat may be attained in plasma spraying operations.

9:30 a.m. Fe-B GLASSES FORMED BY PICOSECOND PULSED LASER QUENCHING:' C.J. Lintand F. Spaepen, Division of Applied Sciences, Harvard University, Cambridge, MA 02138· Picosecond pulsed laser beams can be used to melt surface layers (- 1000 fl.) of metals, which subsequently cool very rapidly (up to 10 13 K/s). In this work, the glass formability of Fe-B alloys, with 0-24 at.% B, was explored. 0 Thin compositionally modulated Fe/Fe76B24 films (- 1000 A) were produced by bias d. c. sputtering. The modulation wavelengths were short enough (16-35 J\) to ensure good mixing during the very short melt time induced by the psec laser pulses. The samples were irradiated with 0.8 J/cm 2 , 30 psec Nd:YAG laser pulses (>-=1.06 um). The resulting structures were examined by transmission electron microscopy. Our results showed that, by this technique, Fe-B alloys could be quenched into glasses with a B content as low as 5 at.%, which is significantly lower than the lowest B content (12 at. %) of Fe-B glasses fonned by conventional splat-quenching techniques. The implications of these results for the thennodynamics and kinetics of glass formation will be discussed. t Exxon pre-doctoral Fellow. This research was supported in part by the National Science Foundation, Materials Research grant, DMR-80-20247.

9:50 a.m. COARSE INTERMETALLICS IN RAPIDLY SOLIDIFIED A1-Co ALLOYS, R. K. Garrett, Jr. and T. H. Sanders, Jr., School of Materials Engineering, Purdue University, West Lafayette, IN 47907 During the solidification process, coarse, primary intermetallic particles nucleate and grow, the size and distribution of the precipitates are controlled by composition and solidification rate. The presence of these coarse precipitates adversely affects ductil ity and fracture toughness of wrought alloys fabricated from particulate containing these particles. This paper will discuss the mechanism of nucleation and growth of these intermetallics and the relationship of the surrounding microstructure to these particles.

10:10 a.m. STRUCTURE DEVELOPMENT IN MELT-SPUN RENE 9S RIBBONS: S.c. Huan~, General Electric Corporate Research and Development, Schenecta y, NY 12301 The structure of melt-spun Rene 95 ribbons is strongly influenced by the ribbon-wheel contact, which controls both the nucleation catalysis and the solidification heat transfer. Microcrysta11ine ce11s of 'V(). 2 IJIIl diameter were observed near the bottom surface of the ribbOns. Every cell might represent a single nucleation event. Larger cells were also observed, but they were usually divided into subcells (also of 'V().2 11m diameter) by a network of carbide particles. The subcells are believed to be formed by the instability breakdown of the growing solid-liquid interface. The ce11s then competed in growth, coarsened, and finally developed into branched dendrites of "-'2 IJIIl diameter near the ribbon top surface. TEM metallography wi11 be presented to show the structure development and the effect of various ribbon-wheel contact characteristics on the resultant ribbon structure will be demonstrated.

10:30 a.m. PRECIPITATION AND PHASE STABILITY IN SOME RAPIDLY SOLIDIFIED STEELS. J. V. Wood and J. V. Bee* , Department of Materials Science, Faculty of Technology, Walton Hall, Miltin Keynes, England - *Department of Metallurgy, University of Witwatersrand Johannesburg, South Africa.

TRANSFORMATION CHARACTERISTICS OF RAPIDLY SOLIDIFIED SHAPE-MEMORY ALLOYS. J. V. Wood, Department of Materials Sctence, Faculty of Technology, The open University, Walton Hall, Milton Keynes, England.

A range of shape-memory alloys (Cu-Al, Cu-Zn-AI, CU-Ni-Al, Fe-Mn etc) have been melt spun or melt extracted and their transformation temperatures characterised in a differential scanning calorimeter. In all cases these temperatures (martensite and austernite start) are different from samples heat treated in the solid state followed by quenching into water. The nature and magnitude of this shift is dependent on both alloy content and process variables. After ageing in the austenite field (yet avoiding any precipitation) the transformation temperatures tend to revert to values typical of solid state treated samples. The kinetics of this recovery process and associated structures (as moni tored by transmission electron microscopy) will be reported in order to demonstrate how a stable fine grained shape memory alloy can be processed.

11:10 a.m. CHEMISTRY AND PHYSICS OF RAPIDLY SOLIDIFIED MATERIALS, R. Becker, G. Sepold, Bremer Institut fur angewandte Strahltechnik, BIAS, Ermlandstr. 59, D-2820 Bremen 71, Germany, and P. L. Ryder, Universitat Bremen, FB Physik, D-2800 Bremen, GermaI1y. Since laser glazing has been proposed by Breinan and Kear it has been shown, that laser glazing is applicable to s7veral amorphous alloys with critical cooling rates up to 10 k/s. In this report results will be presented on the system NiNb. In a first step amorphous surface layers were produced by overlapping laser traces on crystalline NiNb-substrates. The influence of different laser-conditions (continuous-pulsed etc.) will be discussed. Emphasis will be given to the crystallization kinetics under the therma conditions of overlapping laser traces. Furthermore, NiNb layers were produced on steel-substrates. Some results will be reported.

COMPUTER-AIDED MANUFACTURING: Factory for the Future: Opportunities for Automatics in Forge Shops - Session III GEM/G-P AC Productivity Syniposium Sponsored by the TMS-AIME Governmental, Energy and Minerals Committee and the ASM Government and Public Affairs Committee Wednesday, October 27, 1982 Cervantes Convention Center Room 275 8:30 a.m. Session Chairman: Vijay Tipnis, President, Tipnis Software Systems, P.O. Box 42001, Cincinnati, OR 45242.

Abstracts for the presentations in this section will be printed in the Addenda to the Pocket Program distributed at the meeting.

(8:30 a.m.) CAD/CAM for Forge Shops:

Tay1an Altan, Battelle Columbus Labs.

(9:05 a.m.) Imp1ementinq CAD/CAM in a Forginq Company: Gordon Company.

Sanjay Shah, Wyman-

(9:40 a.m.) Aerospace Forginq Technology:

Charlie Chen, Chen-Tech Industries.

TMS Fall Meeting 49 9:30 a.m.

(10:15 a.m.) The Application of Robots in Modern Forging Companies: Shu 1tz S tee 1 Company.

A. Vinter,

(10:50 a.m.) Application of Robots on Forging Equipment: TRW, Inc.

\lilliam Co~nyn,

FRACTURE V: Inclusions, Impurities, Nonmetallics TMS-AIME General Abstract Session Wednesday, October 27, 1982 Cervantes Convention Center Room 260 8:30 a.m. Session Chairman: Dr. S. M. Copley, Dept. of Materials Science, University of Southern California, Los Angeles, CA 90007.

The effect af P additian an creep af l%CrMaV steels. Jin Yu and H.J.Grabke, KAIST Seaul, Karea and MPI fur Eisenfarschung,Dusseldarf, Germany. The 1% CrMaV steels with different P c~ntents(O.006, 0.045,and 0.1 %) were creep testzd at 546 C within the stress range af 230 and 350 N/mm • The minimum creep rate was faund to. increase with the P additian,which was partly caused by cavity farmatian. Hawever, the additian af P hindered the farmatian af VC and Ma C, but favared the precipitatian of (Fe,Cr)1c, The effect af P an caarse cementite precipitatian was also. reflected in the decrease af the frictian stress with the P cantent,as measured by the stress drap test. In additian to. thase, the presence af P was found to. increase the density af 2MnS particles. Applicatian af stresses up to. 350 N/mm did nat have significant influences an the equilibrium segregatian level af P. Hawever, cavity surfaces shawed much higher segregatian level af P than grain baundaries.

9:50 a.m. 8:30 a.m. Experimental Characterization of Inclusions and Impurities in Plates and Welds of Reactor Pressure Vessel Steels: Ofelia C. de Hodgins, IBM, Dept. A43, Poughkeepsie, NY 12602, Bryce L. ShrlVer, Dept. of Nuclear,Engrg. and Engrg. Physics, University of Virginia, Charlottesville, VA 22901, and Kenneth R.'Lawless, Dept. of Materials Science, University of Virginia, Charlottesville, VA 22901. A basic understanding of the nature of inclusions and impurities for A533-B steel plate, BP; weld, GW; and weld-plate, GP is needed to answer the radiations sensitivity of the steels for a projected life-time of 40 years, designed to operate at 550°F (288°C) . Microtensile specimens were fractured at 20°C. The fracture surface of the BP materi~l showed manganese sulfide inclusions, the weld fracture surface showed manganese-rich regions under scanning conditions. These regions represent manganese-sulfide inclusions. The weld-plate materials showed manganese-rich regions, however, the inclusions under electron-microprobe analysis were found to be sulphur inclusions. The matrix of the plate and the weld showed ~u and Zn; the weld-plate showed a very low Cu signal. The failure mode of these microtensile specimens was, in the three cases, primarily transgranular cleavage. Auger studies showed fine precipitates of Mo, P+Mo, S+Mo, Mo, and Mn for the plate and the weld. The weld material also showed Cr. Fine precipitates of Mo and Mn were found for the weld-plate steel.

8:50 a..m. MECHANISMS OF ALUMINUM NITRIDE-INDUCED INTERGRANDLAR FRACTURE IN LOW ALLOY STEEL CASTINGS. N.H.Croft, Dept. of Materials Scienl,;.e and Mineral Engineering, University of California, Berkeley, Ca 94720

One form of intergranular embrittlement which Is occasionally observed in aluminum-killed steel castings results from the precipitation of aluminum nitride (AIN) on the as-cast grain boundaries. Experimental work on a O.30%C,1.50%Mn steel containing various levels of' A1 and N has confirmed the existence of two idiomorphic fo.rms of AlN:- dendritic and plate-like. The way in which each of these contributes to the micromechanisms of fracture at different temperatures has been· determined by fracto graphic examinations of Charpy, tensile and fracture toughness test specimens. A simple model will be presented which describes the temperature dependence of the inter granular fracture process, and the effects of some processing variables will be discussed.

FRACTURE THRESHOLDS AND INTERGRANULAR CRACK GROWIH RATES OF Fe AND Fe + Sb AT CATHODIC POTENTIALS:* R. H. Jones, S. M. Bruemmer, M. T. Thomas, .and D. R. Baer, Pacific Northwest Laboratory, Richland, Washington 99352 Sub-critical crack growth rates and stress intensity thresholds, KtH, have been measured for Fe with 0.3 mono layers of sulfur and an Fe + Sb alloy with 0.3 mono layers of sulfur plus 0.06 monolayers of antimony at their grain boundaries. Fracture tests were conducted using D.IT compact tension specimens in a solution of IN H2S04 at cathodic potentials ranging from -0.6V to -1.25V (SCE). Fracture thresholds for the Fe and Fe + Sb alloys were 2 MParm at -O.6V (SCE) and 1.5 MParm at -1.25V (SCE). The fracture threshold and cathodic potentials were linearly related which is consistent with the relationship KTH a. - log PH reported for gaseous hydrogen embrittlement of steels. The eff~ct of impurity segregation and cathodic hydrogen on intergranular crack growth rates and fracture thresholds of iron will be discussed.

* Research

sponsored by the Office of Basic Energy Sciences of the U. S. Department of Energy.

10:10 a.m. EFFECT OF Co ADDITIONS ON THE CORROSION BEHAVIOR OF POWDERALUMINUM ALLOYS F. W. Vahldiek, AFWAL Materials Laboratory, WPAFB, OR, 45433; M. Khobaib, Systems Research Laboratories, Inc., 2800 Indian Ripple Road, Dayton, OH 45440 Some qualitative studies have been conducted on the corrosion behavior of powder AI-alloys (with and without Co additions). Standard immersion tests were carried out in mixtures of Hi acetic acid, 1M acetic acid + 1M NaCl, O.lM acetic acid + O.lM NaCl, and EXCO solution. The influence of Co additions on the pitting potential was also determined. The EXCO immersion data showed a decrease in corrosion with increasing Co content (ranging from 0 to 0.8%). Corrosion-fatigue crack-growth studies in O.lM NaCl on these alloys showed considerable scatter in the data. Optical and SEM studies were conducted on the fracture specimen. The scanning electron micrographs show mixed-mode behavior of the crack propagation. These results, along with the electrochemical data, will be presented~

10:30 a.m. BINDER DEFORMATION AND BEHAVIOR IN WC-CO AND we-(eO,NIl CERMETS:* C. Vasel, A. Krawitz, University of !4issouri, Columbia, MO 65211, E. Drake, Reed Rock Bit Co., Houston, TX 77001 and E. Kenik, Oak· Ridge National Laboratory, Oak Ridge, TN 37830. The mechanical response to monotonic and cyclic conditioning (1 to 3 GPa) of the binder phase of WC-(Co,Nil cermets with 0, 15 and 30

wlo Ni was studied using neutron diffraction, TEM, SEM and meas-

9:10 a..m. DEVELOPMENT OF DISPERSOID-FREE ZONES DURING CREEP OF OXIDE DISPERSION STRENGTHENED ALLOYS: D. J. Srol ovitz, M. J. Luton and R. Petkovi c-Luton, Exxon Research and Engi neeri ng Co., P.O. Box 45, Linden, New Jersey 07036 Experiments were perfonned on pseudo-bi crystals of an Fe base ODS alloy, where the grain boundaries were subjected to creep under different shear to nonnal stress ratios. It was found that dispersoid free zones develop in the vicinity of grain boundary steps on those boundaries which are loaded in shear. Two essentially distinct mechanims of denudation are proposed. The fi rst ari ses by diffusi onal accommodation of grai n boundary sl i di ng in stepped boundari es whereas the second is a result of a particle drift in the stress field generated by the step. The first mechanism requries a significant grain boundary sl i di ng whi 1 e the second can produce denudation even in the absence of any sliding at the boundary. It is shown that, in the limiting case of bulk diffusion control, the two descri pt ions are equi va 1 ent. The two mechani sms are di scussed in the light of results of detailed particle distribution analysis performed on as-crept samples.

urements of fracture toughness. The metastabl e FCC bi nder phase initially present partially transformed to HCP. The amount transformed increased with load level and Co content, i.e., Ni stabilized the FCC phase. As HCP formation decreased, dislocations and twinning became the prevalent modes of deformation. The thermal residual stresses present in the binder of as-sintered material decreased from about 66011Pa to zero wi th i ncreas i ng severi tyof load. Sample densities initially increased due to the stress relaxation then decreased due to formation of microcracks at the highest load levels. A model for this behavior is presented. *Research sponsored by Reed Rock Bit Co. and the ORAU SHaRE program.

10:50 a.m. The Applicability of Fracture Mechanics Methodologies to SiCw Reinforced 6061 Aluminum: J. M. Cox, S. L. Langenbeck, LockheedCalifornia Company, P.O. Box 551, Burbank, California 91520. The application of fracture mechanics methodologies to 20 wlo silicon carbide whisker (SiC w) reinforced 6061 aluminum sheet is

50 TMS Fall Meeting discussed. Full scale fracture mechanics specimens were machined and tested. The stress intensity factor, K, was shown to apply to this material. Constant amplitude fatigue crack growth and R-curv(' tests were accomplished satisfactorily. From the constant amplitude crack growth data, spectrum crack growth is predicted and the confirming tests conducted. SEM fractography of typical fracture surfaces are shown. In addition, metallography and baseline materials data are present{'d characterizing the material.

evaluated were: (a) chromium carbide precipitation and the associated chromium depleted layer and (b) segregation. An improvemen t in SCC resistance in Inconel Alloy 600 correlated with the presence of a semi-continuous -grain boundary carbide precipitate. Grain boundary chromium depletion and phosphorus segregation were considered to be of secondary importance. There was a smaller range of grain boundary microstructures in Inconel Alloy 690 and it was therefore more difficult to correlate SCC performance with microstructure in this alloy.

11:10 a.m.

9:45 a.m.

FRACTURE OF LiF BICRYSTALS, Jong Lee and K. N. Subramanian, Department of Metallurgy, Mechanics and Materials Science, Michigan State University, East Lansing, Mt 48824 Bicrystals of LiF having til t and twist boundaries were grown by Czochralski method. Using these specimens the role of relative crystallographic and grain boundary orientations on crack propagation was investigated. Etch-pit studies were carried out to reveal the dislocation distribution in the region where the crack interacts with the grain boundary.

GRAIN BOUNDARY CHEMISTRY AND ENVIRON· MENTAL INTERACTIONS II Sponsored by the TMS Corrosion and Environmental Effects Committee Wednesday, October 27, 1982 Cervantes Convention Center 8:30 a.m. Room 261 Sessicn Chairmen: Russell H. Jones, Battelle-Northwest Laboratory, P.O. Box 999, Richland, WA 99352; Lloyd A. Heldt, Michigan Techno· logical University, Houghton, MI 49931.

EFFECT OF SULFUR AND PHOSPHORUS SEGREGATION ON THE INTERGRANULAR FAILURE OF NICKEL AT ANODIC POTENTIALS: * S. M. Bruemmer, R. H.

Jones, D. R. Baer, M. T. Thomas, Pacific Northwest Laboratory, Richland, Washington The effectivenness of Sand P in promoting the intergranular failure of Ni was found to be dependent on the potential region of the polarization curve in which the test was performed. Sulfur segregation prompted IG corrosion during static tests in IN H2S04 at both active and transpassive potentials, while P-rich grain boundaries were only corroded at transpassive potentials. Straining electrode tests retarded the IG corrosion of Ni specimens with a grain boundary S coverage of 13% of a monolayer. No ductility loss was observed at potentials well into the transpassive region. However, Ni-P specimens with a grain boundary P coverage of 13% of a monolayer, exhibited IG cracking and a sharp ductility loss at both passive and transpassive potentials.

'*

Research supported by the Office of Basic Energy Sciences of the U. S. Department of Energy

10:10 a.m. EFFECT OF CHLORINE ON SULPHUR SEGREGATION IN 321 STAINLESS

STEEL B. Ratnaswami, S. J. Thorpe, and K. T. Aust, Department of Metallurgy and Materials Science, University of Toronto, Toronto Ontario, Canada M5S IA4

8:30 a.m. SENSITIZA TION OF AUSTE,NITIC STAINLESS STEELS. I. CONTROLLED PURITY ALLOY~ C. L. Briant, R. A. Mulford and E. L.. Hall, General Electric Company, Corporate Research and Development Center, Schenectady, NY 12301 This paper presents a study of the effects of carbon, nitrogen, molybdenum, and manganese on the sensitization of high-purity austenitic stainless steels of composition similar to type 304. The modified Strauss test (ASTM-A262-E), the oxalic acid etch test (ASTM-A262A) and analytical electron microscopy were used to determine the degree and nature of sensitization in the steels. The alloy compositions are considered in. terms of effective chromium content, and from plots of this parameter versus sensitization time a strong effect of carbon content is seen. Additions of N, Mo, and Mn are found to delay sensitization at any given carbon concentration. Strong evidence is presented that N retards the nucleation and/or growth of carbides. Molybdenum increases the ease with which the steel passivates, thus requiring more chromium depletion before sensitization will be detected. Beneficial effects of Mn are primarily seen at low sensitizing temperatures, where it assists nitrogen in slowing carbide formation. This work was partially supported by EPRI Contract 1574-1.

8:55 a.m. TIME - TEMPERATURE - SENSITIZATION DIAGRAMS OF SULFUR AND PHOSPHORUS DOPED TIPE 304 STAINLESS STEEL: Steven Dany1uk and Ira

*

Wolke, Materials Engineering Department, University of Illinois, Chicago, IL

60680

Ten vacuum remelted heats of a type 30455 have been doped and recast with 0.045, 0.065, 0.087 and 0.;!5 w/o sulfur, and 0.040,

0.070, 0.090 and 0.12 w/o phosphorus. Time - temperature - sensitization (TTS) maps versus doping level have been determined for these heats by oxalic acid ASTM A262A test of Jominy endquench samples and Charpy load-temperature impacts tests. The doping level will be correlated with the microstructure of the base materials, the cooling rate as obtained by observing the ditch structure at ,the grain boundaries, and Charpy impact energies and frac ture surface morphology.

Investigation of the susceptibility of a Ti stabilized austenitic stainless steel to stress corrosion cracking (SCC) in argon, synthetic flue gas and synthetic flue gas with 0.2 VolA: HCl at 813 0 K was performed. The slow strain rate technique was used over the range of strain rates (£) of 10- 4 to 10- 6 s-1 No cracking was observed in either argon or synthetic flue gas environments. Auger analysis of TiC precipitates revealed high concentrations of S associated with the carbide. No S was detected in the matrix or grain boundary. The addition of a chloride environment promoted intergranular SCC at [; = lO-6 sec -l. Chlorine was found to accumulate at the metal oxide interface. Sulfur was partioned between the TiC carbide and the grain boundary. The results will be discussed in terms of the models of stress corrosion cracking of austenitic stainless steel.

10:35 a.m. GRAIN BOUNDARY SEGREGATION AS RELATED TO STRESS-CORROSION CRACKING OF A1-Zn-Mg ALLOYS: J.R. Pickens, L. Christodou10u,

G.D. Davis and J.A.S. Green, Martin Marietta Laboratories, Baltimore, MD 21227 Increasing evidence suggests that the mechanism of stresscorrosion cracking in Al-Zn-Mg alloys involves a synergistic effect of both localized dissolution and hydrogen embrittlement. As a result of the intergranular nature of such failures, the structure and composition of both the grain boundary interface, and the adjacent region, are expected to have a significant inf luence on cracking. Segrega t ion phenomena of the grain boundary region have been studied to better understand the operative fracture processes. Auger electron spectroscopy, plasmon energy loss, and x-ray microanalysis techniques have been employed and results to date suggest that failure is influenced by elemental Mg segregation, and grain boundary preCipitate size and distribution. The implications of these findings to the overall sec mechanism will be discussed.

10:55 a.m. nit ROLE OF MAGNESIUM SEGREGATION IN THE INTERGRANULAR STRESS CORROSION CRACKING OF ALUMINUM ALLOYS

*Work sponsored by the American Iron and Steel Institute, Grant No.

62-440.

9:20 a.m. CORRELATION OF STRESS CORROSION CRACKING (SCC) PERFORMANCE WITH GRAIN BOUNllARY MICROSTRUCTURE IN INCONEL 600 AND INCONEL 690:

G. P. Airey, West inghouse Researc h and Development Center, Pittsburgh, Pennsylvania 15235 The stress corrosion cracking (SCC) resistance of Inconel Alloy 600 and Inconel Alloy 690 in pure water and deaerated caustic environments has been correlated with grain boundary microstructure. A range of grain boundary microstructures was introduced by annealing in the 593 to 871°C (1100 to 1600 0 F)

tempera ture range.

Gra in boundary microstruc tural features

G. M. Seamans, N. J. H.

Holroyd and e. D. S. Tuck, Alcan International Limited, Southam Road, Banbury, Oxon, England OX16 7SP There is increasing evidence that susceptibility to stress corrosion cracking of certain aluminum alloys (e.g. Al-Zn-Mg, AlZn-Mg-Cu and AI-Mg) is due to magnesium segregated to grain boundaries. The presence of magnesium at the grain boundary enhances reactivity in aqueous environments and promotes hydrogen entry and its intergranular diffusion. Accumulation of atomic hydrogen results in a reduction of cohesive strength of the boundary and stress corrosion crack propagation if stress is applied either subsequent to or concomitant with environmental reaction. Prolonged environmental reaction of unstressed specimens results in the production of a grain boundary layer of magnesium hydride. In the present work it is proposed to critically review the published data on magnesium segregation and to analyze the proposed mechanisms.

TMS Fall Meeting 51 LIQUID AND SOLID METAL INDUCED EMBRIT· TLEMENT OF METALS IV Sponsored by the TMS Corrosion and Environmental Effects Committee and the ASM-MSD Corrosion and Oxidation Activity Wednesday, October 2:1, 1982 Cervantes Convention Center 8:30 a.m. Room 270 Session Chairman: P. Gordon, Department of Metallurgical and Materials Engineering, Illinois Institute of Technology, Chicago, IL 60616.

in alloys containing liquid Bi inclusions and least severe in alloys containing liquid Pb inclusions. The calculated reducti on ~ in fracture surface energy depends upon both the di hedra 1 angle of the liquid inclusion at the grain boundary and the extent to which the alloying element segregates to the boundary. It is proposed that appreciable grain boundary segregation occurs in systems of this kind only if the dihedral angle is small, and the implications for LME of a possible relationship between wetting angle and segregation are discussed.

10:40 a.m. ZINC EMBFITTLEMENT OF STAINLESS-STEEL WELDS: J.M Johnson, M.R. Berry and J. Gutzelt, Standard Oil Company (Indiana), Naperville. Illinois 60566

8:30 a.m. REVIEW PAPER: SOLI D-METAL- INDUCED EMBRI TTLEMENT OF METALS: P. Gordon and A. Druschitz, Metallurgical and Materials Engineering, Illinois Institute of Technology, Chicago, IL 60616. It 1s now recognized that the phenomenon referred to in the past as liquid metal embrlttlement (LME) takes place not only when the embrittler 1s in the liquid state but when it 1s 1n the solid state as well - sol1d-metal-induced embrittlement (SHIEl. In this paper the published evidence with respect to SMIE is reviewed and the various ideas which have been advanced to explain both LMIE and SMIE are discussed.

9:20 a.m. SOLID CADMIUM EMBRI'ITLEMENT OF TITANIUM ALLOYS: D. A. Meyn, Code 6312, Naval Research Laboratory, Washington, D. C. 20375 Solid cadmium in intimate contact at 148 C (300 F) causes cracking under sustained tensile stress in Ti8AI-IV(an alpha alloy), Ti-3Al-13V-IICr(a beta alloy), and Ti-6Al-4V(an alpha-beta alloy). Cracking was both intergranular and transgranular(cleavage)in all three alloys. Cadmium contact did not accelerate fatigue crack initiation at moderately high stresses at frequencies of 1-1/2 and 30 cps; fretting was of greater importance. The hydrogen content of the alloy had no effect on cracking. The engineering significance of solid cadmium contact cracking depends on the application and design details since the cracks grew to less than 0.050 inch under the conditions employed.

9:40 a.m. DELAYED FAILURE OF STEEL DUE TO SOLID METAL INDUCED EMBRITTLEHENT: J. C. Lynn, .Am.sted Research Laboratories, Bensenville, Ill. 60106, and W. R_ Warke, Standard Oil Co. (Indiana), Naperville, Ill. 60566. Sooae experiaental results on the ....britt1 ...ent of .uSI 4140

steel by In, Cd, Sn, Pb, and Zn are presented, including failure times as a function of stress and temperature, and minimum .tresses for failure. It is shown that crack.

Extensive cracking of Type-30D Series stainless steel piping during construction of a petrochemical plant was attributed to zinc embrlttlement. Based on laboratory testing and metallurgical examinations, it Is shown that a combination of several factors was responsible. Piping was contaminated by paint overspray and splatter when zinc-rich primer wss applied to structural steel components. Piping was also contaminated by airborn zinc-sulfIde ore dust from a nearby ore unloading facility.

11:00 a.m. METAL-INDUCED EMBRITTLEMENT OF STEELS: Dr. Yukiteru Asayama, Mitsubishi Heavy Industries, Ltd., Nagoya Aircraft Works, 10, Oyecho, Minato-Ku, Nagoya, Japan Delayed failure of steels intimately contacted wIth lower melting metal such as cadmium, zinc, indium or brazing filler metal occurred at elevated temperature under sustained stress. The delayed failure took place with zinc and cadmium in the solid, as well as in the liquid state, but did not take place with indium in the solid state. Transport mechanisms of the embrittlers, therefore, may differ with kind of the lower melting metal. Delayed failure of steel brazed with filler metal was estimated to be caused by cadmium contained in the filler metal from activation energy. It is assumed that the intermetallic compounds formed by diffusion of zinc or cadmium embrittle the steels, and that in zinc the zinc-iron compound was formed in steel matrix and in cadmium the cadmium-nickel compound was formed along grain boundary. In the case of indium plated steel, the embrittlement mechanism is not clear.

MECHANICAL PROPERTIES: COMPOSITIONAL EFFECTS TMS-AIME General Abstract Session Wednesday, October '1/1, 1982 Cervantes Convention Center 8:30 a.m. Room 268 Session Chairman: Dr. Joseph M. Rigsbee, Dept. of Metallurgy and Mining Engineering, University of Illinois, Urbana, Dlinois 61801.

propagation due to solid metal induced embrittlement in these

systems Is lntergranular along prior austenite grain boundaries and appears to be controlled by a thermally activated process,

8:30 a.m.

believed to be the multilayer surface .elf diffusion of the embrittling specie.

GRAIN BCXJNIJARY SEGREGATIOO OF Cu, Sn AND Sb IN C-Mn STEE!.'l AT 900 o C: W. T. Nachtrab and Y. 1. Chou, Lehigh University, Bethlehem, PA 18015

10:00 a,m.

The segregation of Cu, Sn and Sb to austenite grain boundaries at 900 0 C has been investigated in several cormnercial heats of C-Mn steels. Segregation was studied on in situ fractures using a scanning Auger microprobe (SAM). The specimens for microanalysis were prepared in a manner such that the prior austenite grain boundaries could be exposed by fracturing at room temperature in the UHV chamber of the SAM unit. Initial bulk concentrations ranged between 600 to 2600 ppm Cu, 50 to 360 ppm Sn and 8 to 35 ppm Sb. Significant enrichment of Cu, Sn and Sb was detected along the austenite grain boundaries. The grain boundary concentration of Cu and Sn was found to vary depending upon the initial bulk concentration while the average concentration of Sb at the grain boundaries was found to be approximately 1 at .'Yo for all of the heats studied. For heats in which a significant level of Cu segregation was detected, a relationship of at.% Cu = at.i'. (Sn +Sb) at the -austenite grain boundaries was observed. Deformation at 900°C prior to fracture in lfi{V was found to be necessary to promote segregation. Samples that were annealed at 900°C but not hot worked did not exhibit evidence of Cu, Sn or Sb segregation. These results have been interpreted in terms of the effects of deformation on s.egregation kinetics.

EFFECTS OF METAL INDUCED EMBRITTLEMENT IN STEEL: D. Bhattacharya, Inland Steel Research Laboratories, East Chicago, Indiana 46312 The important role metal induced embrittlement plays on the processi ng and properti es of free-machi ni ng steels is explored. Results of inclusion characterization, high temperature mechani ca 1 testi ng, electron mi crofractography, Auger spect roscopy, and scanni ng Auger mi croscopy are revi ewed. It is estab 1i shed that, in 1eaded and tell uri zed free-machi ni ng steels, embrittlement by tellurium and liquid tellurium compounds, causing severe loss in ductility in the temperature range of ro 11 i ng, is the mechani sm of hot-cracki ng duri ng roll i ng of these steels. Metal induced embrittlement by low melting additives such as lead and bismuth also cause ductility losses at temperature ranges of machining and a strong case is made for a mechanism of machinability-improvement based on embrittlement by the free-mac hi ni ng addi t i ves.

10:20 a.m. INTERGRANULAR EMBRITTLEMENT OF ALUMINUM ALLOYS BY LOW-MEL TINGPOINT LIQUID INCLUSIONS. M.C. Roth, T.A. Jones, G.C. Weatherly and W.A. Miller, Department of Metallurgy and Materials Science, Universlty of Toronto, Toronto, Ontario, Canada M5S lA4. It has been shown that a sudden drop in ducti11ty occurs when

binary, two-phase alloys containing Bi, Cd or Pb inclusions in Al are heated. The embrittlement is clearly associated with the melting of the inclusions and with their fractional coverage of the aluminum grain boundaries. It was found that a fracture surface energy model of internal liquid rretal embrittlement (LME) accounts for the observation that embrittlement is most severe

8:50 a.m. EFFECTS OF ADDING BORON AND COOLING RATE ON MECHANICAL PROPERTIES IN.HSLA LINE PIPE STEELS: T. W. Kim, J. E. Kim, Pohang Iron and Steel Research Lab., Pohang, Korea; Y. G. Kim, Dept. of Mat. Sci. &: Eng., School of Engineering, Korea Advanced Institute of Science and Technology, Seoul, Korea Mechanical properties of HSLA line pipe steels, Fe-O. OBC-l. 3MnO. 03Nb-O. 06V-0. 2Mc-0. 2Ni, with and wi thout boron (10 ppm) were investigated under various cooling conditions. The steels were produced by using either conventjonal rolling or controlled roll-

52 TMS Fall Meeting ing, followed by various coolings. The cooling methods employed were air cooling (1. 3 Oct s), compressed air cooling (4.1 Dcl s) , oil cooling (9.5 °C/s), and water spray cooling (13.6 0C/s). It was found that both strength and impact toughness were enhanced with the increasing cooling rate. The conventionally rolled material with water spray cooling exhibited high strength (0.2% YS 77 kSi, UTS 102 ksi) and good impact toughness (145 ft-lb at - 20 DC). Mechanical property requirements of "X-70 gradel! line pipe steel can be met by employing conventional rolling for the boron added steel, resulting in high productivity of the steel. Mechanical properties and related microstructures will be presented.

9:10 a.m. INFLUENCE OF SUBSTITUTING Ti FOR V ON MECHANICAL AND RECRYSTALLIZATION BEHAVIOR IN A HSLA STEEL: J. W. Hong, Y. G. Kim, Dept. of Mat. Sci. & Eng., Korea Advanced Inst. of Sci. &Tech., Seoul, Korea; J. M. Han, Daewoo Shipbuilding Co., Seoul, Korea; K. B. Kang, Research Lab., Pohang Iron &Steel Co., Pohang, Korea. 9:30 a.m. INFLUENCE OF COMPOSITION AND STRUCTURE ON THE MECHANICAL PROPERTIES OF ALUMINIUM FOIL: Ake Karlsson, Granges Aluminium, 8-612 21 Finspong, Sweden.

The mechanical properties of aluminium foil are of importance both in manufacturing of the foil and for the end use. In this paper the dependence of the mechanical properties of thin foil (~14 pm) on composition, o~ grai~ si~e and on size ~d.distr~­ bution of precipitated partIcles 18 dIscussed. CompOSl.tlons wIth

Fe+Si content ranging from 0.4 to 1.2 weight per cent and Fe/Si ratios from 1.2 to 8 have been studied. Foils made from continuously cast strip and from conventional, hot rolled strip are compared. The high solidification rate at strip casting gives a decreased particle size which on annealing can cause coarse grains and unfavourable crystallographic texture in several common foil alloys. In these cases the resulting mechanical

the contribution of the mobile dislocation density and veloCity product to the strain rate. Solutes that effectively increase the density of mobile dislocations generated from grain-boundary sources should decrease the magnitude of Luders strain.

10:50 a.m. DEVELOPMENT OF LOW ALLOY AND STAINLESS STEEL PIPES PRODUCED BY UOE PROCESS; K. Hirabayashi, T. Taira, Fukuyama Research Labs, T. Ishihara, Fukuyama Works, NIPPON KOKAN K.K. 1 Kokan-cho, Fukuyama, Hiroshima-Pref. 721, JAPAN

Welded steel pipes for use in low temperature or high temperature service, conventionC?lly produced by press bending or bending roll forming process, have been developed by the UOE process. The purpose of this paper is to describe the results of research and development in the productio~ of 3.5% nickel, 9% nickel and stainless steel pipes for low temperature service, carbon steel pipes and 2'\Cr-Ul0 steel pipes for high temperature service. The manufacturing procedure has been established in a mill production test. Manufacturing of the above steel pipes by UOE process is made possible by synthetically applying such results as the developement of special welding consumables, MIG+SAW welding process and application of pipes QT technique developed by NKK. The steel pipes produced and shipped have good mechanical properties, well satisfying the intended target. They can meet the increasing demand on steel pipes for special purpose along with the future growth caused by changes in energy demand and resources.

ll:10 &.11. PROTECTIVE OXIDE CHARACTERIZATION OF Mn MODIFIED ALUMINIDE COATINGS:* D. H. Boone, R. L. Lambertson, M. J. Barber, Materials and Molecular Research Division, Lawrence Berkeley Laboratory, University of California, Berkeley, CA 94720.

properties are low. By a high temperature (about 600°C) treatment

of the cold rolled cast strip the structure can be modified so that the foil properties equal those of conventionally produced foil.

'9:50 a.m. EFFECT OF COMPOSITION ON THE MECHANICAL BEHAVIOR OF CHROMlUMFREE Mn STEELS: S. R. Schuon, NASA Lewis Research Center, Cl eve 1and, OH 44135 Recently, interest has been generated in the substitution of Mn and Al for Ni and Cr in austenitic stainless steels to form less expens i ve alloys and reduce dependence on chromi um. Although the resulting alloys are similar, their mechanical behavior is not directly analogous. A systematic study has been made on the effect of nickel, molybdenum, aluminum, titanium, and silicon levels on the deformation behavior of an Fe-25(20)Mn-Al-2C alloy. Without the addition of Mo, eutectic carbides did ~ot form. Carbide-free alloys formed significant amounts of martensite on deformation at l450°F, while much less martensite at room temperature. Alloys with carbides had similar l450°F tensile strengths and stress rupture 1i ves regardl ess of percent a1umi num or volume of carbides. Substitution of silicon for aluminum reduced room temperature tens il e strength. The effects of compos i ti on and the mode of failure are discussed.

10:10 a.la. DISPERSOID-FORMING ELEMENT ADDITIONS TO POWDER PROCESSED Al-Cu-Li AND Al-~n-Mg-Cu-Li ALLOYS:' P. J. Meschter, S. M. L. Sastry, and J. E. 0 Neal, McDonnell Douglas Research Laboratories, P. O. Box 516, St. Louis, MO 63166.

The introduction of incoherent dispersoids into Al-3Li has been shown to significantly improve ductility without sacrificing strength. The nonshearable dispersoids force Orowan bypassing during deformation and thus reduce the severity of planar slip induced by the presence of the coherent, shearable 6' ( AI 3Li ) phase. In the present investigation, the effects of additions of dispersoid-forming elements such as Ti, Ni, Fe, or Co to powder-processed Al-Cu-Li and Al-Zn-Mg-Cu-Li alloys have been evaluated. Microstructures of dispersoid-containing alloys in the solution-annealed and aged conditions have been determined and related to changes in mechanical properties. The effects of powder-processing parameters on microstructure and mechanical properties will be discussed.

* This

research is sponsored by the Air Force Office of Scientific Research under contract no. F49620-79-C-0039.

Recent interest in the improvement and expanded use of aluminide coatings has resulted in the development of alternate deposition techniques and modified compositions. One such system, the Allison Electrophoretic Process, AEP, involves the electrophoretic deposition of an aluminium rich coating bisque, containing MIl, to the component followed by a high temperature application cycle. Extensive testing has shown the resulting coating to be equivalent to other standard aluminide coatings. However, questions still arose .as to the specific effects of the Mn on the behavior of the protective oxide. The results of a study of the oxide structure as effected by the presence of Mn will be presented.

'This study was supported by Air Force Contract F3361S-78-C-Sl0l, under the direction of S. G. Lee

POWDER METALLURGY TMS-AIME General Abstract Session Wednesday, October 27, 1982

Cervantes Convention Center Room 265 Session Chairman: William R. Pfouts, Kennametal, Inc., Latrobe, PA 15650.

8:30 a.m.

8:30 a.m. EXPLOSIVE CONSOLIDATION OF RAPIDLY SOLIDIFIED ALUMINUM ALLOYS:' T. C. Peng, S. M. L. Sastry, and J. E. O'Neal, McDonnell Douglas Research Laboratories, St. Louis, MO 63166. Rapidly solidified 2024Al and 2024Al-3Li alloy powders were explosive-consolidated into 2. 5-cm and 5-cm diameter cylinders. Explosive-compactions were analyzed in terms of detonation-, shock-, and expansion-wave theories. The compacts were examined by optical metallography, scanning electron microscopy, and scanning Auger microscopy and were tes ted for mechanical properties. The effects of post-consolidation heat treatment on microstructural changes within the powder particles and the bonding between powder particles were investigated.

*This research was conducted under the McDonnell Douglas Independent Research and Development program.

10:30 a.m. EFFECT OF SOLUTES ON LUDERS STRAIN IN LOW-CARBON SHEET STEELS: Hsun Hu, U. S. Steel Corporation, Research Laboratory, Monroeville, PA 15146 The effects of phosphorus «0.092%), tin «0.164%), and silicon «0.7%) on the magnitude of Luders strain in low-carbon sheet steels has been studied. At equivalent grain sizes the Luders strain was effectively reduced by the P or Sn additions. However, additions of Sn or Si to a P-containing steel produced very little additional effects. The magnitude of Luders strain, as influenced by grain Size, is a linear function of d- l / 2 • This relation is similar in· form to the Hall-Petch relation for yield or flow stress. The effect of P or Sn on this relation is to decrease the slope of the plot. These observations have been analyzed on the basis of Liiders strain in relation to the band-front velocity and

8:50 a.m. EFFECTS OF MICROSTRUCTURE ON TENSILE PROPERTIES OF PIM ALUMINUM ALLOY X7091: Young-Won Kim J Metcut-Materials Research Group, P. O. Box 33511, Wright-Patterson AFB, OH, 45433. The effects of microstructure and aging treatment on the tensile properties have been investigated for a high strength PIM aluminum alloy, X7091. Samples with widely varying grain sizes (2 to 1000um) were thermomechanically prepared starting from as-extruded fine-grained material. Following various aging treatments (under-, peak-, and oV,er-agin 9), the specimens were tensile tested at a strain rate (1O"3s -1) at room and elevated temperatures. The tensile properties, especially yield strength, were related ,to grain size, in association with the aging response as observed by hardness measurements and transmission electron microscopy.

TMS Fall Meeting 53 9:10 a.m. AGE-HARDENING OF RAPIDLY SOLIDIFIED ALUMINUM-RICH Al-Zr-B and AlZr-Si ALLOYS: Young-Won Kim, Metcut-Materials Research Group, P. O. Box 33511, Wri ght-Patterson AFB, OH, 45433; -S. H. Whan~, F. Dabkowski and B. C. Giessen, Institute of Chemical .Analysls, Northeastern University, Boston, MA, 02115. Two ternary aluminum alloy systems have been investigated for their aging behavior following rapid solidification. Alloy systems studied are Al-xZr-y (B or Sil, where x varies from 5 to II wt% and y from 0.75 to 12 wt%, and were prepared from1he melt by a rapid solidification technique. The as-quenched, thin body samples were heat-treated i sochrona lly as well as i sotherma lly in vacuum. . Precipitation reactions and microstructural changes occurring during the heat treatments were observed and analyzed using the scanning transmi ss ion· electron mi croscope. Concurrent changes of mi crohardness and Young's modulus were measured to relate to the precipitation reactions observed. The results were compared with those of Al-xZr binary alloys.

9:30 a.m. MAGNETIC PROPERTIES OF p/M Fe-S1 SHEET MADE BY ROLL CONSOLIDATION, E. C. Oren, Homer Research Laboratories, Bethlehem Steel Corp., Bethlehem, PA 18016

A series of Fe-Sl alloys containing up to 4.6% 51 were made by a roll consolidation process. The process includes blending Iron and ferrosl1icon powders, canning, followed by hot consolidation on a rolling mill . Full density sheets with final thickness In the range from 0.018 to 0.026" have been produced. The AC and DC magnetic properties were determined on annealed Epstein strips_ The AC core losses were dependent on final annealing temperatures and the mean-free-path between microstructural inclusions. The relationships between the structural and magnet ic propert ies will be discussed .

9:50 a.m. FRICTION IN METAL POWOER: K. E. Amin, Bendix Advanced Technology Center, Columbia, MD. 21045 A number of techniques have been developed to measure both the static and dynamic friction in metal powder with particle sizes in the range 5 to 250 11m. Measurements were conducted in the temperature range 25 0 to 2000 C. Friction measurements were. found to be sensitive to the technique used. Results show that both surface morphology and chemistry of the powder particles significantly affect their frictional properties. Data will be analyzed in relation to predominant friction mechanisms.

10:10 a.m. EFFECT OF FOREIGN PARTICLES ON CRACK INITIATION IN P/M TITAtnUM: S. W. Schwenker, F. H. Froes, AFWAL/MLLS, Wright-Patterson Air Force Base, Ohio, 45433; and D. Eylon, Metcut-Materials Research Group, P.O. Box 33511, Wright-Patterson Air Force Base, Ohio, 45433. Powder Metallurgy has become a viable method for producing full density net shape titanium alloy components. By Hot Isostatic Pressing (HIP) of Rotating Electrode Process (REP) powder, it is now possible to produce compacts with static properties at least as good as wrought alloy. However, powder contami nat i on can reduce the high cycle fatigue strength of PIM compacts below that of wrought material. In many aerospace related applications, inferior fat i gue performance cannot be to I erated, therefore, a thorough i nvestigation into the nature of contamination related fatigue crack initiation was carried out. Precision sectioning, scanning electron mi croscopy and ana lys i s by x-ray energy d i spers i on methods were util ized to fully characterize the nature of the crack origins. These results were correlated to the fatigue life data to identify defects and contaminants most detrimental to the fatigue strength. The interaction of contaminants with the alloy microstructure and its effect on the fatigue life of the P~1 compacts will be discussed in detail. It will be demonstrated that by using clean powder and careful control of the microstructure, that the fatigue behavior of the PM components can be raised to levels at least equivalent to wrought material.

10:30 a.m. DI SPERSION STRENGTHENING OF TITANIUM ALLOYS:' S. M. L . Sastry, T. C. Peng, and J. E. O'Neal, McDo nnell Douglas Re sea r c h Laborat ories, St. Loui s , MO 63166 and L. P. Beckerman, Air Force Wright Aeronautical Laboratories, Wright Patterson Air Forc e Base I Dayton, OH 45433.

Production of dispersion-strength e ned T1 alloys b y conventional ingot casting and pro c essing meth o d s has been only pa rtially s uc cessful bec ause additions of lar ge a mounts of dispersoidforming element s r es ult in small volume fraction s of c oarse, equilibrium, c onstituent parti c l e s, which are ineff ec tive for dispersion stre n g thening. A rapid s olidification t e chnology (RST) approach wa s used in the pr e s e nt investigation to determine the feasibility o f producing thermally stable, fin e , incoherent dispersoids in a Ti matrix. Ti-RE (RE = Y, Er, Ce , Cd, La. Dy, and Nd) alloys were !"apidly solidified by electr on beam melting a nd splat quen ching. The rapidly so lidi'fied spec imens annealed at 700-850°C c ont a in homongeneous distribution s of fin e « 0.2 iJ m diam) dispers o ids with interpartic le spacings of 0 . 1-0 . 5 ~m. The

effects of dispersolds on deformation, recov e ry. and recrystallization and strengthening of Ti were determined. *This research is sponsored by AFWAL-ML und e r Co ntract no . F33615-81-C-5011.

10:50 a.m . THE EFFECT OF DENSITY, CtrlTlNG SPEED AND FEED RATE ON THE DRILLABILITY OF PIM STAINLESS STEEL Ao'lD COPPER STEEL MATERIALS G. W. Halidin, A. W. Abduljabbar and M. F. DeVries, Department of Mechanical Engineering, University of WisconSin-Madison, Madison, WI 53706 The drillability of 3166 s tainless steel and 5% copper, 0.9% car-

bon steel powder metallurgy materials was inve stigated in terms of drilling thrust force. Drilling performance was studied for a range of relative densitie s of 70 to 99 percent of theoretical and the thrust force was found to include with density according to a power law relationship. Thrust force was also found to increase with feed rate and decrease with cutting speed, also according to a power l aw relation. Strength, impact, hardness and microstructure properties were evaluated as a function of density and related to the drilling behavior. It was also fOlUld that for any given dens ity level, the drilling thrust increa s ed with de'creasing pore size.

11:10 a.m. POWDER FABRICATION OF HASTELLOV D-TYoE ALLOYS: S. M. Soltesz and R. W. Heckel, Department of Metallurgical Engmeering, Michigan Technological University, Houghton, MI 49931. P/M alloys of nominal compositions Ni3Cu, Ni3Si3Cu, Ni9Si3Cu, and NillSi3Cu were fabricated to essentially full density. Powder consolidation was accomplished by extruding cans of blended Ni, Cu, and Ni-Si master alloy powders. Heat treatments in the range of 950 oC-IlOOoc were carried out to produce microstructures of a (Ni-rich solid solution) , (Ni)Si), a-y (Ni 5Si 2 ), and a-6-y . Corrosi~n tests and room temperature tensile tests were used to evaluate the alloys relative to cast Hastelloy alloy D.

a-a

RADIATION DAMAGE ANALYSIS FOR FUSION REACTORS III: Microstructural Evolution-Experiment (continued) Sponsored by the TMSIASM Nuclear Metallurgy Committee Wednesday, October 27, 1982 Cervantes Convention Center 8:30 a.m. Room 271 Session Chairmen: G. R. Odette, Department of Chemical and Nuclear Engineering, University of California, Santa Barbara, CA 93106; L. K. Mansur, Metals and Ceramics Div., Oak Ridge National Lab., P.O. Box X, Oak Ridge, TN 37830.

8:30 a.m. :rE"'~ERATURE ~PENDENCE Of RADIATION DAMAGE IN HIGH PURITY AHD itElllJ4 DOPED AlUMINIJ4 IRRADIATED WITH AlUMINIJ4 -IONS: S. K. Mclaurin, G. l. Kulcinski, and R. A. Dodd, Univ. of WisconsinMadison, Madison, WI 53706.

The Illter1als for fusion reactors will be subjected to an environ!lent of f1 uctuati ng teIIIperatures, stress and transllluted hydrogen and he 11 U1D gases. He 1 i U'8 gas atollls in pure a1 uRli num serve as lIucleation sites for the fOnRation of cavities, while pure alU'8inUll containing no hel1U11 gas does not swell or f.om cavities when 1 rrad1ated up to 20 dpa. Samples of he 11111 doped a 1ulli mil containing 0.1, 1.0 and 10 at . ppm of helium have been irradiated at temperatures ranging from 2SoC to 125°C at doses of 2 to 20 dpa. The effect of hel1U'8 concentration and the temperature variation on the density and size of the cavities is discussed and for pure alullinuIII the cavity size decreases, and the number density increases with decreasing temperature. The effect of fluence from the self ion irradiation in pure aluminum at a given temperature is to increase the size of the cavities I1nearly in volume, yielding a linear swelling rate.

8:50 a.m. HELIUM EFFECTS IN RELATION TO VOID FORMATION IN Cu, Ni, AND Cu-Ni ALLOYS. S. ZinkIe, K. Farrell*, R. A. Dodd, and G. L. Kulclnskl, Unlv . of Wisconsin-Madison, Madison, WI 53706. It Is known that voi d formation is easy in pure copper and pure nickel, yet difficult In Cu-NI alloy which form a continuous solid solution. It has been postulated that this difference between the pure metals and the alloys might be due to difficulties in helium bubble formation, and the present experiments were designed to investigate whether this is the case, and to explore possible reasons for voi d suppress i on. 01 scs of Cu, Cu-20 Ni, 3Cu-50 NI, Cu-BO Ni, and NI have been injected wi th 0.2-0.4 MeV He at 0 . 65

54 TMS Fall Meeting Tm to ,a level of 200 appm He. The specimens were held at 0.65 Tm for 1 hr after Irradiation, so that dlspl acement damage structure would tend to be eliminated, and bubble development depend on thermal vacancies. Bubble parameters for the various material s will be compared, and the kinetics of bubble growth after further annealing wi11 be discussed. *Oak Ridge National Laboratory, Oak Ridge, TN

37830.

9:10 a.m. LOW-TEMPERATURE MOBILITY OF JON-IMPLANTED HELIUtl IN NICKEL D. B. Poker and J. M. Williams, Solid State Division, Oak Ridge National Laboratory,* Oak Ridge, Tennessee 37830 The release of He after 35 keV 'ion-implantation into Ni at 18 K was measured between 18 and 300 K. The release occurred in two stages, a sharply peaked release at 50 K, fo11owed by a more diffuse release extending from 90 to 300 K. For He doses in the range of 5 x 10 14 to 1.4 x 10 16 cm- 2 the fraction released was 1.5% in the first stage and 4% in the second. At higher doses, from 1.5 x 1016 to 1 x 1017 cm- 2 , the fractional release in each stage approximately doubled. The activation energy for the first stage release was 0.11 ± 0.02 eV for a11 doses. These results wi 11 be di scussed in terms of interact ions between He atoms and radiation-produced point defects. *Operated by Union Carbide Corporation under contract W-7405-eng26 for the Division of Materials Research, U. S. Department of Energy.

9:30 a.m. EFFECT OF RADIATION ON PHOSPHORUS SEGREGATION:* J. L. Brimhall, D. R. Baer, R. H. Jones, Pacific Northwest Laboratory, Richland, Washington 99352 Radiation can have a pronounced effect on the segregation of certain elements to grain boundaries and interfaces which in turn can degrade mechanical properties. Radiation was found to enhance segregation of phosphorus to free surfaces in austenitic stainless steel and PE-16 but not in a ferritic stainless steel HT-9 at 500 and 600°c. Segregation was determined by AES sputter profiling of ion bombarded specimens. Recent results have shown measurable phosphorus segregation in 316 SS at doses as low as 0.01 dpa. Similar experiments have failed to show any radiation enhanced segregation in Ni + 0.04%P or Fe + 0.04% alloys after irradiation to doses of 3 dpa at temperatures of 450°C to 700°C. The results J which show that segregation is a complex phenomenon, will be discussed in terms of the interaction of phosphorus with other elements as well as with radiation produced defec ts. *Research sponsored by the Office of Fusion Energy of the U. S. Department of Energy.

involves averaging over space and time, may not always adequately describe processes such as irradiation creep, bubble nucleation, etc. Any given point in space will in reality not see a continuous uninterrupted flow of point defects (as inherent in the rate theory) but instead will be subject to intermittent arrival of interstitials and vacancies. Mansur et. al. have developed a numerical procedure for calculating the cascade induced fluctuations,. In this paper we present a simpler approach, which gives rise to analytical expressions for the vacancy and interstitial concentrations at a given point in space. The advantage of this approach is that the expressions for the fluctuatingCv(t) and Ci(t) can be easily integrated over time (which is important, for example, for calculating dislocation climb distances), and this analytical method may be more amenable to the inclusion of mutual recombination. *Research sponsored by the National Science Foundation under grant number CPE-8025300.

10:35 a.m.. NON-EQUILIBRIUM SEGREGATION AND IRRADIATION CREEP IN CONCENTRATED ALLOYS,* I-Wei Chen, Massachusetts Institute of Technology, Cambridge, MA 02139 Under intense irradiation, non-equilibrium segregation at the steady state can be very prominent, in both its magnitude and its spatial extent, in a concentrated alloy such as stainless steels of the fusion first wall. The effects of segregation on irradiation creep therefore need to be assessed. These phenomena are treated using a unified irreversible thermodynamic approach, which leads to simple solutions for an otherwise complex system. Physical parameters, i.e., mobilities, required for the implementation of the analysis are well-defined operationally, and their total number is considerably less than that of the atomistic approach formulated for dilute alloys. Prospects for controlling segregation and creep by the alloy design will be discussed.

*Research

supported by the National Science Foundation under Grant No. DMR 7824185.

10:55 a.m. EFFECTS OF MICROSTRUCTURE ON LIGHT ION IRRADIATION CREEP IN Ni: * C. H. Henager, Jr., E. P. Simonen, E. R. Bradley, Pacific Northwest Laboratory, Richland, WA; and R. G. Stang, University of Wa~hington, Seattle, WA. Irradiation creep tests using 1~ to 17 MeV deuterons were conducted on 99.992% Ni specimens having different initial microstructures. The microstructures were obtained by thermalmechanical treatments of cold-rolled Ni which produced different initial dislocation configurations. Transmission electron microscopy (TEM) was used to characterize the initial and radiation induced microstructures. Sink densities, sizes and configurations were characterized. A climb-glide creep model and the TEM data were used to predict experimental creep rates. The model was based on reaction rate equations and detailed dislocationobstacle interactions.

9:50 a.m. IRRADIATION- INDUCED SEGREGApON OF SUBSTITUTIONAL SOLUTES IN DILUTE NIOBruM-BASE ALLOYS: B. A. Loomis, and S. B. Gerber, Materials Science Division, Argonne NItional laboratory, Argonne, Illinois 60439 The depth dependence of irradiation-induced segregation of substitutional solute atoms in several dilute r.."'-base alloys was determined in the near-surface region of irradiated specimens. The Nb-base alloys for this study contained either 2.3 a/o Mo, Ta, V Ti, Zr, Hf, Fe or Ni and were irradiated at 1250 K with 3.0-MeV 5~Ni+ ions to a damage concentration of approximately 50 dpa. The Auger electron spectroscopy technique ·was used to determine the dependence of the substitutional-solute atom concentration on depth in the irradiation-damaged layer. The magnitude of the irradiation-induced enrichment or depletion of substitutional solute atoms in the surface layer of the specimens shows a dependence on the solute diffusivity in Nb. TIle exceptional resistance to void swelling of a Nb-Mo alloy can be correlated with the substantially greater surface enrichment of Mo solute. *Work supported by the U. S. Department of Energy.

IV. Irradiation Creep and Defect Production Session Chairman: R. Bullough, Theoretical Physics Div., AERE, Harwell, Didcot, Oxfordshire, OXll ORA, U.K.

10:15 a.m. MODELING CASCADE INDUCED POINT DEFECT FLUCTUATIONS:' H. Gurol, Science Applications, Inc., La Jolla, CA 92037 High energy neutron irradiation gives rise to displacement cascades, which are produced randomly in both space and time, even under steady irradiation. There is growing speculation that the rate theory formulation of the point defect concentrations, which

11:15 a.m. EFFECTS OF 14 MeV NEUTRON IRRADIATION ON CXEEP OF NIOBIUM AND NICKEL:* W. L. Barmore, A. W. Ruotola, E. L. Raymond, A. K. Mukherjee, Lawrence Livermore National Laboratory, Livermore, CA 94550 Flux, stress and temperature effects on creep of niobium and nickel were observed in situ at the RTNS-II 14 MeV neutron source. Results are used in developing predictive models explaining creep under high energy neutron irradiaton and are compared with creep results from lower energy neutron irradiation. Creep tests were done onNb and Ni near .3 Tm and stresses to 200 MP a in high vacuum using a digital computer for control and data acquisition. Cyclic Flux tests produced dramatic changes in creep rate. This creep behavior is rationalized in terms of point defect partitioning in the crystal structure. Analysis of creep and stress relaxation under steady state flux gives activation volumes and activation energies indicating that intermediate temperature, thermally activated mechanisms are rate controlling with apparent activation energies for creep being stress dependent. *Work performed under the auspices of the U. S. Departmen t of Energy by LLNL under contract no. W-7405-ENG-48.

RECENT ADVANCES IN SUPERALLOY TECH· NOLOGY I Sponsored by the TMS High Temperature Alloys Committee Wednesday, October 27, 1982 Cervantes Convention Center 8:30 a.m. Room 267 Sessiqn Chairmen: J. H. Wood, General Electric Co., Bldg, 53, Rm, 316, Schenectady, NY 12345; C. G, Beck, Westinghouse R&D Center, 1210 Beulah Rd., Pittsburgh, PA 15235.

TMS Fall Meeting 55 tl:30 a.m. DEVELOPMENT OF MATERIALS AND PROCESS TECHNOLOGY FOR DUAL ALLOY DISKS, C. S. Kortovich, TRW Inc., 23555 Euclid Avenue, 44117 A study was conducted to develop techniques for the preparation of dual alloy disks. Preliminary evaluations on four material combinations were conducted on HIP consolidated and heat treated cylindrical and plate shapes in terms of elevated temperature tensile, stress rupture and low cycle fatigue properties. These evaluations indicated that the pre-HIP AF-115 rim/loose powder Rene' 95 hub combination offered the best overall range of mechanical properties for dual alloy disks with hub temperature capability of 900 0 F and rim capability of l4000 F. The feasibility of this dual alloy concept for the production of more complex components was demonstrated by the scale-up fabrication of a prototype CFM"56 disk made from this AF-115/Rene' 95 combination. The results of the preliminary evaluations as well as the mechanical property characterization of the large disk including tensile, stress rupture and low cycle fatigue properties will be presented. This concept is being extended to include the use of iron-base superalloys in the hubs of dual alloy disks. Preliminary property determinations have been made. • Work performed under NASA Contract NAS3-2l35l

8:50 a.m. THE EFFECT OF FORGING CONDITIONS AND HEAT TREATMENT ON THE STRUCTURE AND PROPERTIES OF ANi-BASED SUPERALLOY. F. E. Sczerzenie, C. J. Stelma, and G. E. Maurer, Special Metals Corp., New Hartford, NY 13413; G. Hoffman, Kelsey-Hayes, Utica Div., Utica, NY 13503; F. Cosandey and J. K. Tien, Henry Krumb School of Mines, Columbia University, New York, NY 10027

Hot tensile tests were done to define the strain rate sensitivity and the optimum conditions for isothermal forging of the Ni-base alloy Udimet 720. Conventionally forged and isothermally forged Udimet 720 pancakes were heat treated to different strength levels to evaluate the suitability of the alloy for turbine disk applications. Variations in strength, stress rupture life, and creep resistance are explained in terms of matrix and grain boundary structures.

9:10 a.m. THE DEVELOPMENT OF A LOW COST TURBOCHARGER ALLOY: R. C. Benn, INCO-Sterling Forest, Suffern, NY 10901; J. R. Mihallsln, Howmet Turbine Component Corp., Alloy Div., P.O. Box 371, Dover, NJ OlBO\. Industry sources indicate that by the mid-1980 t s the market for automotive turbochargers will have increased to several million units per year to support the need for acceleration power in down-sized engines. Present materials such as alloy 713 and the current workhorse alloy for automotive turbochargers, GMR 235, are considered too expensive in the long term. A replacement. lower cost alloy, IN-6225. has been developed by Inca in cooperation with Howmet Turbine Component Corp. IN-6225 has lower cost, equivalent or better mechanical properties. and lower density than GMR 235. These properties have been confirmed in a conunercial 1,814 kg (4,000 lb.) hE!8t of the alloy made by Howmet· under their designation "Turbo-2S". In addition, IN-6225 has greater stability, better oxidation resistance. but slightly lower sulfidation resistance, than GHR 235 in burner rig tests. This well-balanced combination of properties in IN-6225 makes the alloy attractive to interested turbocharger manufacturers.

9:30 a.m. THE DESIGN OF WROUGHT NICKEL-BASE ALLOYS FOR ADVA~CED HIGHTEMPERATURE GAS-COOLED REACTOR (HTGR) APPLICATIONS: * ~ Johnson and T. A. Lechtenberg, General Atomic Company, San Diego, CA 92138, and L. D. Thompson, Southwestern Community College, Chula Vista, CA 92010

The utilization of the high-temperature gas-cooled reactor (HTGR) for advanced or process heat applications will require the use of wrought components operating at temperatures up to 1000°C for times approaching 30 years. Alloys for such components must withstand the corrosive effects (carburization and oxidation) of the impure helium environment and maintain adequate elevatedtemperature strength. Commercially available wrought alloys have been found to be seriously limited for such applications because of their inherently poor resistance to corrosion in impure helium. This paper describes the development of a series of Ni-Cr-Mo-WAi-Ti-Zr-C alloys specifically formulated for such applications, presents the results of tests to evaluate tile corrosion resistance, mechanical behavior, and thermal stability of these alloys in simulated reactor helium, and contrasts their behavior with that of several commercially available wrought alloys currently being considered for such applications. *Research funded by The Department of Energy.

9:50 a.m. THE EFFECT OF PROCESSING AND THERMAL EXPOSURE O~ THE TENSILE PROPERTIES OF RAPIDLY SOLIDIFIED ALLOY 185: D. B. Miracle, K. A. Williams, and H. A. Lipsitt, Metals and Ceramics Division, Afi.lAL Materials Laboratory, Wright-Patterson AFB OH 45433

Rapid solidification, with cooling rates of ''-'10 4 to 10 6 °C!sec,

has the advantage of producing material with greater homogeneity than cast pieces, whose cooling rates are <10 3 °C!sec. The objective of this study was to investigate the influence of differences in cooling rate, within the range of 10 4 to 10 6 °C!sec, on stock extruded from powder. Powder particle sizes ranged from <53 microns to "v250 microns, representing a change in cooling rate of "v one order of magnitude. Powder particles were consolidated by extrusion at two extrusion temperatures. The effect of long time exposure at high temperature on equiaxed and directionally recrystallized material were also determined. Data were obtained by microstructural characterization and testing in tension. The results showed no effects on properties due to differences in cooling rate or extrusion temperature as long as comparable microstructures were tested. Long time exposure at high temperature was found to degrade the properties of equiaxed material at 1400°F and below, due to cellular coarsening of the grain boundaries.

10:10 a.m. FATIGUE CRACK GROWTH IN HIP-RENE 95: D. J. Duquette, S. Golwalker and N. S. Stoloff, Materials Engineering Department, Rensselaer Polytechnic Institue, Troy, NY 12181

Fatigue crack growth rates have been measured on HIP-Rene 9S compact tension specimens at high temperatures in argon and in air. The d. c. potential-drop technique was used to monitor crack length. The influence of frequency, temperature, holdtime and environment on rate of crack growth has been determined. Typical crack growth rates from "vlO- 8m/ cycle to 10-Sm! cycle were observed, over a range of stress intensities from "vIS MPam l !2. Fracture characteristics were examined by SEM and metallographic techniques. Low frequencies, oxidizing atmosphere and high temperatures lower the fatigue threshold and increase the crack growth rate. Hold times at peak load tend to increase both the fatigue threshold as well as the crack growth rate.

10:30 a.m. THE NONDESTRUCTIVE EVALUATION OF HOT-CORRODED IN-l38: A. M. Beltran and M. Fishman, General Electric Co., Schenectady, NY 12345; K. E. Nichols, Watervliet Arsenal, Watervliet, NY 12189

This paper describes the development of a nondestructive test technique utilizing magnetic induction to assess the degree of IN-738 degradation due to high temperature hot-corrosion attack. Ni-base superalloy IN-738 is rendered nonmagnetic by virtue of its Cr concentratidn. The formation of Cr203 surface oxide and erS internal sulfides during exposure to high-temperature sulfidation environments generates surface layers of increasing magnetic strength which can be detected by a magnetic induction probe. Results are presented describing the relationship between magnetic response, microstructure, and alloy concentration gradients for IN-738 exposed in small combustion burner rigs as well as gas turbine blades removed from corrosive field service.

10:50 a.m. THE EFFECT OF STRESS COARSENED y' ON THE CREEP PROPERTIES OF SINGLE CRYSTAL ALLOY 143: D. D. Pearson, United Technologies Research Center, Silver Lane, East Hartford, CT 06108

The effect of creep stress and crystal orientation on the y' coarsening behavior in directionally solidified single crystals of alloy 143 (Ni-14.3Mo-5.8Al-6Taw/o) at high temperature (1311k) has been investigated. As previously observed in alloys exhibiting precipitate/matrix lattice misfit, the y' coarsens into a lamellar array perpendicular to the stress axis when the [100] orientation is parallel to the stress axis. The lamellar structure in the present alloy is unusual in that it is very continuous and closely spaced SlJ) • Optimal structures were observed in materials having an initially fine y' distribution formed upon cooling from the homogenization or solution temperature. The degree of improvement in creep life for the stress coarsened alloys was nearly 4-fold relative to non-aligned structures in the same alloy. The ability to, form the struct':lre is dependent on a large negative misfit, a oY < aoY , which is a departure from conventional alloy development trends and suggests a new avenue for improvine. the creep strength of superalloys. The presentation will cover factors which promote the structure as well as a discussion of the crystal anisotropy observed in both structure and creep response.

«.

11:10 a.m. DEVELOPMENT OF NAF lOl-T SINGLE CRYSTAL ALLOY, C. S. Kortovich, TRW Inc., 23555 Euclid Avenue, Cleveland, OH 44117, B. M. Guenin, Bendix Advance Technology, Columbia, Maryland, R. L. Dreshfield, Materials Division, NASA-Lewis Research Center, 21000 Brookpark Road, Cleveland, OH 44135. C. A. Lombard, U.S. Air Force, ASD! YZD, y!right-Patterson Air Force Base, Dayton, OH 44135

A study was conducted to evaluate a chemical composition modification of single crystal NASAIR 100 alloy cast in the form of TFE731 turbine blades by the exothermic pack process. The alloy. named NAF 101-T. included a reduction in Cr from 9% to 6%, a Ta increase from 3.3% to 67., a Ti increase from 1.2% to 1.5% and a W decrease from 10.5% to 9.5% compared to NASAIR-IOO. The

56 TMS Fall Meeting results indicated tha't the stress rupture capability was equivalent to that demoristrated by the NASAIR-IOO alloy _and there was no evidence of microstructural instability at 180QoF. Tensile properties over the range room temperature -2000°F were generally equivalent for both alloys except for the results at room temperature and 1800°F. where the NAFIOI-T yield strength was approximately 10-15% below that of NASAIR-IOO. ~ertormed

under NASA Contract NAS3-21821

REFRACTORY METALS AND ALLOYS FOR ELECTRICAL APPLICATIONS Sponsored by the TMS Refractory Metals Committee Wednesday, October 27, 1982 Cervantes Convention Center 8:30 a.m. Room 272 Session Chairmen: H. J. Kim, GTE-Gibson Electric, Delmont, PA 15626; S. M. Tuominen, Climax Molybdenum Co., Ann Arbor, MI 48105.

8:30 a.m. REFRACTORY METALS AS ELECTRIC CONTACTS:

P. G. Slade,

Westinghouse R&D Center. Pittsburgh, PA

15235

Refractory metals have found wide usage as electrical contacts. This paper will discuss their properties and will look into the reasons why refractory metals are usually combined with good conductors such as Cu and Ag for contact applications. The use of refractory metal contacts- will be reviewed in three parts. The first will discuss the problems of carrying steady state current, in other words, the effects of surface structure, of surface properties and of oxidation. The second part will consider the possible changes in surface structure when these materials are used in switching normal currents. This involves the kinetic chemical surface changes which result from low current arcing and the effect of these changes on the contacts' performance. Finally, these materials will be examined for very high current uses. Here their high vaporization temperature helps to prevent severe contact erosion. Their hardness and brittleness also reduces the tendency to weld. This review will conclude with a brief discussion of possible future developments in this class of contact.

9:00 a.m. A CORRELATION OF MATERIAl TOUGHNESS, THERMAL SHOCK RESISTANCE, AND MICROSTRUCTURE OF HIGH TUNGSTEN, SILVER-TUNGSTEN COMPOSITE :l'IATERIALS, Go J. Witter, Pansteel, Inc., North Chicago, IL 60064

The electrical erosion resistance of composite silver-tungsten materials seems to improve with finer dispersions of the silver and tungsten; however, these materials with the finer dispersions exhibit a higher tendency to fail by catastrophic crack propagation during arc erosion testing. Tungsten-silver materials were made by two different powder metallurgical techniques which resulted in alloys having a wide latitude of structural variation in terms of volume fraction, particle sizE, and distribution. A correlation of Glc' the critical strain energy release rate, was made with the quantitative metallographic parameters I,3Ag, the mean intercept length within the silver phase; C, the contizuity of the tungsten pha~e; and the volume fraction of the silver phaseo A term composed of these three parameters was found to have a good correlation with GIc for all the alloys. A study of the fracture surfaces of each alloy by energy-dispersive X-ray analysis shows the silver content of the fracture surfaces decreases as the toughness of the alloys decreases.

V"

Two phase W-Ag composite material is a major material system which has been used for electrical contact components in household circuit breakers and large industrial circuit protection devices. The metallurgical degradation porcess occurring under severe electric arcing condition has not been fully characterized due to complex interrelated phenomena unique to contact operation. Besides composition, the major variable has been identified to be microstructure, which controls ablation, phase redistribution, oxidation and thermal stress crack propagation dictating the performanc~ level and the life of device. A discussion relating these properties with microstructural parameters and methods of manufacture will be presented.

10:30 a.m. RUTHENIUM-BASE METAl COMPOSITES;

POTENTIAl CONTACT MATERIAlS:

M. L. Green, Bell Laboratories, Murray Hill, NJ 07974, F. Bader Bell Laboratories, Columbus, OR 43213 and E. Coleman, Bell Laboratories, Murray Hill, NJ 07974 By warm rolling mechanically alloyed Cu and Ru powders, we have created a novel type of electrical contact material consisting of a submicron distribution of Ru in Cu. Because these elements are mutually insoluble, mechanical alloying is one of the only ways of creating such a non-equilibrium microstructure; subsequent processing of these powders must preserve the metastability. Removal of surface Cu from the rolled strip gives rise to a "sandpaperll structure in which the hard, refractory and conductive Ru particles, now exposed on the surface, serve as the electricai contacts, whereas the Cu matrix supports these particles and provides electrical continuity. The effects of mechanical alloying time, powder annealing time and temperature, and rolling schedule will be discussed. A Cu-15v/o Ru sample, processed to 99% + density, has exhibited a low (2m-ohm) and stable contact resistance even after exposures of up to six months in air ambient, and up to one week in a sulfidizing atmosphere at 70o F.

11:00 a.m. THE EVAlUATION OF AG/DIBORIDE COMPOSITES FOR USE IN ELECTRICAl CONTACT APPLICATIONS: R. Y. Lin and J. Gustafson, GTE Labora-

tories, inc., Waltham, Massachusetts

STRUCTURAL MATERIALS FOR GROUND TRANSPORTATION I Sponsored by the TMS Structural Materials Committee Wednesday, October 27, 1982 Cervantes Convention Center 8:30 a.m. Room 262 Session Chairman: Richard G. Connell, Jr., Department of Materials Science and Engineering, University of Florida, Gainesville, FL 32611.

9:30 a.m.

8:30 a.m.

P','lPERTY DEVHOPMENT IN ACTIVATED AND LIQUID PHASE SINTERED RE:·V\ CTORY ~fETALS:* R. M. German, P. Zovas and C-J. Li, Materials Engineering Department, Rensselaer Polytechnic Institute, Troy,

Laboratories, Warren, Michigan 48090

New York, 12181.

Several enhanced sintering densification techniques are available for the refractory metals. In most engineering applications enhanced densification is beneficial to the final properties. This presentation will provide a review of the activated and liquid phase sintering techniques by developing several important topics. The thermodynamics and kinetics will be detailed in light of heterophase transport models. Processing sensitivities will be linked to these models. New experimental data on property development in sintered compacts will be discussed to illustrate the powder-processing-property relations. Current research efforts on Mo and W will provide property data and demonstrate the microstructure manipulation efforts in progress. The presentation will conclude by linking the sintering models to the -property optimizat ion efforts.

02254

The suitability of Ag/diboride composites for contact applications has been assessed. Three compositions each of Ag/ZrB2 and Ag/HfB2 materials of densities better than 96% of -theoretical have been prepared. A significant improv~ment in oxidation resistance and a consequent lower contact resistance has been noted for the Ag/ diborides as compared to Ag/W. The measured resistivities of the diborides are acceptable for contact applications. The mechanical integrity of the material may be degraded by the less than optimum wetting between Ag and the diboride phase. Switching test show the interfacial voltage drop to be lower and the arc erosion rate to be higher for Ag/ diboride contacts than for Ag/W contacts.

DUAL PHASE STEELS, M. S. Rashid

Metallurgy Dept, GM Research

Dual phase steels are characterized by a microstructure consisting of 75-85 vol % ferrite with the remainder being a 1Dixture of 1Dartensite, bainite, and retained austenite. They have better strength-ductility combinations than ferrite-pearlite steels of equivalent tensile strength and are presently being used commercially r Publi-cations and research on these unique steels have increased exponentially since their inception in the mid-1970 t s. This paper will briefly review their historical development and present status of the various approaches that have been used to produce the steels. Also current understanding of the phase transformations that produce the dual phase microstructure and the deformation mechanisms in these steels will be discussed. Present understanding of relationships between composition, microstructure, and tensile properties will be reviewed and some trends in developmental work will be presented.

*This work supported by the U. S. Army Research Office and the Engineering FOWldation.

9:15 a.m. 10:00 a.m.

THE USE OF HIGH STRENGTH STEELS IN AU70MOBILES, R. G. Davies.

ELECTRIC ARC DEGRADATION OF W-AG CONTACTS AND MICROSTRUCTURAl

IMPLICATIONS: Pennsylvania

R. J. Kim, Gibson Electric Div. GTE, Delmont, 15626

Engineering & Research Staff, Research, Ford Motor Company, Dearborn, Michigan, 48121-2053 The application of high strength sheet steels in the automobile

TMS Fall Meeting 57 has been accelerated by the demand for lightweight fuel efficient vehicles. The types of HSS will be reviewed and their properties discussed in the light of the above application. In addition, some of the rate limiting steps in the :ncreased use of HSS will be reviewed along with the outlook for. the wider application of these materials.

10:00 a.m. - Break

10:15 a.m. MATERIALS FOR COMPONENTS OF AUTOMOTIVE POWERTRAINS, C A Stickels Engineering and Research Staff, Ford Motor Company, P.O. Box 2053, Dearborn, MI 48121

Automotive transmissions and drive trains include many parts such as shafts, gears and bearings which are manufactured to very close tolerances and are highly stressed in service. For automobiles and light trucks parts are relatively small, so high hardenability steels are 5eldom required. Operating temperatures rarely exceed 150 C and 1 because parts are well lubricated, corrosion is almost never a problem, Selection of materials and manufacturing processes is determined by the specific application, with a certain amount of give-and-take necessary between designers, manufacturing and materials engineers to arrive at a serviceable part readily manufactured in large volumes. Materials and heat treat processes used for some shafts, gears and bearings will be described.

11:00 a.m. EMERGING MATERIALS FOR ARMY USE IN GROUND VEHICLES. Robert W. Lewis, US Army Materials and Mechanics Research Center, Arsenal Street, Watertown, MA 02172.

In order to meet the ever increasing demands placed upon its ground vehicles - ... demands of increased speed, loads, mobility and durability -- in environments of increasing rigor -- high and low temperatures and the nuc lear, biological, chemical (NBC) environment -- under increasingly sophisticated threats -ballistic, mine blast, NBC -- the Army is turning more and more to advanced materials to counter these concerns. Materials that offer weight savings, reduced maintenance costs, enhanced durability and increased strength and stiffness are a must. These materials need to be producible, processable, and reliable and should reduce U.S. dependancy on strategic materials. The paper will describe the potential for new ceramics, elastomers, metal alloys and composites to satisfy these conditions and discuss their relative advantages and disadvantages from the perspective of the Army environment.

WEDNESDAY - OCTOBER 27, PM

ALLOY PHASES II TMS-AIME General Abstract Session Wednesday, October 27, 1982 Cervantes Convention Center 2:00 p.m. Room 264 Session Chairman: Professor F. X. Kayser, Iowa State University, Ames Laboratory, Ames, IA 50011.

2:00 p.m. NICKEL-ALUMINUM-MOLYBDENUM PHASE EQUILIBRIA: D. B. Miracle, M. A. Stucke, D. M. Dimiduk, L. E. Matson, H. A. Lipsitt, K. A. Lark, D. Zabierek, and V. Srinivasan, Metals and Ceramics Division, AFWAL Materials Laboratory, Wright-Patterson AFB OH 45433

Recent work on alloys based on the Ni-Al-Me system has brought to light several inconsistencies with proposed phase diagrams for this system. Existing diagrams have been based on computer models or incomplete data. The intent of this effort is to produce an experimentally validated ternary equilibrium phase diagram for the Ni-AI-Me system. Specimens for this task were produced by both conventional casting techniques and powder metallurgy techniques. The temperatures studied to date include 1700°, 1900°, 2000°, and 2100°F for times up to 2000 hr. Phases were identified using an electron probe microanalyzer and x-ray diffraction. Results thus fa1' show significant deviations from the proposed phase diagrams existing in the literature. In particular, a Class II four-phase reaction y + a ~ y' + 6 has been shown to occur between 2000°F and 2100QF.

2:20 p.m. PHASE BOUNDARY DETERMI NATION IN Ni -Cr-Mo SYSTEM: .M. Raghavan*, R. R. Jo\Jeller*, G. A. Vaughn"" and S. Floreen"**, Exxon Research and Engineering Co.", Linden, NJ 07036, Exxon Production Research Co~any**, Houston, TX 77001, International Nickel Company""", Ster1 i ng Forest, NY 10901 Phase boundaries at 850 and 1250°C in the nickel rich portion of the Ni-Cr-Mo system were detennined by x-ray microanalysis and microdiffraction. Analysis of thin, bulk and extracted residues of the heat treated samples showed that inspite of some discrepancies, the phase boundaries at 1250°C detennined in the present investigation are in reasonable agreement with the published results. The intermetallic precipitates were detennined to be 6, P and a by convergent beam diffraction. At 850°C, the solubility of Cr and Mo in the gamma phase was found to decrease, and it was also observed that the elevated te~erature P phase i ntermeta 11 i c was no longer stable and transformed to a ~ phase (A7B6) compound. The composition of the various phases and accuracy of x-ray microanalysis for phase boundary detenni nat i on will be di scu ssed.

2:40 p.m. ELEMENTAL PARTITIONING IN EUTECTIC SUPERALLOYS: S. W. ~ang, E. F. Koch & E. L. Hall, General Electric Company Corporate esearch & Development, Schenectady, NY 12301 Ni-base, tantal
3:00 p.m. CRYSTAL STRUCTURE OF ALLOYED Ni3AI:" M.J. Kaufman, 0.0. Pearson-and H.L. Fraser, Dept. of Metallurgy and Mining and the Materials Research Laboratory, University of Illinois, Urbana, IL 61801 A recent paper has descri bed the app I i cat i on of convergent beam electron diffraction to the detection of symmetry changes in a recrystallized Ni-base supera11oy. The phases y and Y' in the microstructure were found to exhibit tetragonal distortions from the expected Ll~ cubic structure. It was also proposed that this distortion was aue to sub-lattice ordering, and suggested that this tetragonal form of y' may well represent the lowest energy configuration for this phase. In the present study, attempts have been made to separate out the effect of strain by making observat ions both on a s i ng1 e crysta I of y' a II oyed with i atomi c percent Ta, and on y' in a ternary eutectic alloy which has been subjected to very large stresses during therrna1 processing. The results of this study will be discussed with references to possible sublattice ordering. -Thi s work was supported in part by the US DOE under contract DE AC02 ER01l98 and the United Technologies Research Center. -*United Technologies Research Center, E. Hartford, CT 06108

3:20 p.m. DISPERSOID STABILITY IN INCOLOY* ALLOY HA 956: ~K..:.. .i<.-aEl$. and J. M. Davidson, Inca Research & Development Center. Sterling

Forest, Suffern, :-;Y 10901 INCOLOY* alloy MA 956 is an oxide-dispersion-strengthened (ODS) iron":'base alloy produced by mechanical alloying. The nominal composition is Fe-20%Cr-4.5%AI-O.5%Ti-0.5I.Y .. 03(in wt.%). This alloy combines strength at high temperatures with excellent resistance to oxidation, carburization and hot corrosion. High temperature strength of the alloy is mainly due to the presence of very fint.> stable yttrium oxide dispersoids. In the present study, the thermal stability of the oxide dispersion (982°Cl200°C) has been investigated after long-term, high temperature exposure. Coarsening of the dispersoid was measured by TEM as a function of exposure time and temperature. Extensive particle size analysis has been done to determine the kinetics of coarsening. This information has led to an understanding of the ratelimiting process. The phenomenon has been also correlated to the room temperature and elevated temperature mechanical properties of the alloy after exposure. In addition, the effects of alloy composition on the thermal stability of the oxide dispersion will be discussed. ------*Trademark of the Inca fami 11' of companies.

3:40 p.m. CIiARACTERISTICS A.~ll TIiERMAL STABILITY OF DISPERSED PHASf.S IN RAPIDLY SOl.lDIFIEll TlTANIU~I-RARE EARTH ALLOYS:' P • .I. Ml'schter, J. E. O'Neal, and S. M. 1.. Sastry, McDonnell Douglas Research -

Laboratories, P. O. Box 516, St. Louis. MO 63166.

58 TMS Fall Meeting ~~ il' r () strtl ct lJ rl'S (I f

ti taniu m/ rar f' - ('.1T t h (RE Y, Er, Ce , Gd. or Nd ) . l ll ll \' flab ,s f(l rnw d h\' C10ctTon- hC'.1m-m('·lting / spl a t-quC'nching have ht'l'n l-li;lLl, ' tt'T i Zl'd us ing SEH, TEl-I. and STEM te c hniqu E' s. The crysI :tl stnwtllT l'S ;\ nd cllt'mic;ll comp os i t ions of di s p C'Tso ids in the ! l.i Kl'S 11 ;\v(, IH '(' n d C't prrnin l' d h y t- l E' ctr o n diffra c ti on . energy disp lor s i Vl' X- r ,l~' ;m al y s is. ;lI1 d sc a nn ing Au ge r mi c r osco p y ( SAM). As ;-o;pLlI - qut.'nclll'd flab,s h a ve a homog(· n t.'o u s di s tribu tion o f fi n e (") () - :,.'()I) nm d i mr:) . Ta n ' - ea rth-ri c h d i spe rs o i ds. Rt;>s i s t a nce t o disp l' r so i d cO;I Tscn in g du ri ng .1 nnt'.1 1 ing of t lH~ fl ake'S a t 700- B70oe is ;\ss (l c Litl. ·d . .,. i t ll fn rrnati(l n or retP o t ion of a T i- ~F. -O - C phase, 0::

\,'IWrl';ls l'qui I ih ri um RE203 disp f'rso ids coarsen r e l.1 t i v e l y rapidly . rh~' microst'rtl c tllr l' ;'Ifu>r annl';'!l i ng H.b ovt" thC" h (' til tra nsus temperat un' (900-JOOO °<:) i s det€'rmioc,d b y the interacti o n o f disp€'rsoid c(l arsl ' nin~, di ssn illt i n n. ilnd rt'p r t>ci pitation upo n qu e nc hing. Ohs l' r v;l tinns pf ph ase st ahil i ty }md1:'r v.1ri ous a nn e.1 lin g c o nditi o n s ....' i 1] 1)(' romp,l t (,d ....· i rh p r cd i et i ons o f .1 th e rmod y nam i c mod el o f l i - RE- ll s\'st('ms . "'Til i s res l' ilr eh is sponso re d h y AFWAL-ML u nder Co nt r a e t no. F3 361 5Hl - C- \Ol l.

2:00 p.m • MECHAN ICAL PROPERTIES OF RAP IDLY QUENCHED METALS AND ALLOYS:

*

J.C .M. Li, Department of Mechanical Engineering, University of Rochester, Rochester, NY 14627. Mechanical properties of amorphous and semicr y stal1ine metals and alloys obtained by rapid quenching will be reviewed in c luding their elastic, anelastic, p lastic and fracture properties . MacroscopiC behaviors 1n stress relaxation, ro l 11ng and bending, wire drawing, fracture and fatigue, temperature, pressure, and strain rate effects, annealing and hydrogen embrittlement etc. will be discussed as well as microscopic observations of shear bands and fracture surfaces. In addition, the effect of annealing on mechanical properties and the effect of deformation on other physical properties will be presented . Low temperature and high temperature deformation mechanisms will be speculated . The Question of the existence of dislocations in amorphous metal s will be dis cussed as well as some atomistic models showing the local conditions needed for shear nucleation.

4:00 p.m. *Work supported by NSF through DMR78-12807. AGE HARDENI~ G OF RAPIDLY SOLIDIFIED Ti-Zr-B and Ti- Zr-Si ALLOYS Y.Z . Lu an d ~,~ , in s titut e o f Chemical Anal ys is, ~o rth e il s t cr n Un i ve r s it y , B o~ t o n, MA 02 115; and Y.W . Kim, Metcut Hat e ri a l s Resea r c h Gr oup, Wri g ht Pa tt e rson Air Fo r ce Bas e, OH 4 543 3

~le c hanical pro p e r t y and micro st ru c ture of rapidly q uenched Ti '] C_xZrj C Bx a n d Ti 9 C;-xZr~ JSi x • wh e re x := 0 - 10 at %, were s tudied as fun c ti on s of aging co nditions. The s a mple s of these all o y s pr e p a r e d fro m the melt b y r a pid quen c hing t ec hniques were a nn ealed bo th i soc h ro nally an d i s o t h e rmall y at t e mp e ratures betwe en 400-10 00 0 C . Mechani c al p r o perties s uch as hardness, Youn g's modulus a nd du c tilit y of t he a nnealed s a mp le s were i n v estigat e d . Als o , the micros tru c ture of the sampl es were s tud ied by TEM . The TEM study i s f o cussed on th e f ormation and mo rphology o f v c r y fine precipit a t e s resulting from me talloid addition. Ov e r a ll, age hardeni ng e ffects for th e e xtended solid so lut ion of me tall o ids and th e r e l a tionships betwee n mechanical p r o pert y and mic r os tr u cture o f th es e alloys will be discussed .

The local electronic structures of a variety of amorphous metallic alloys have been calculated from first principles, by the SCF-Xa-SW cluster molecular-orbital method . The resulting electronic structures are in good quantitative agreement with, and provide a chemical bondin9 interpretation of, available photoelectron spectra for these aroorphous alloys. A molecularorbital model of glass formation and melting has also been developed. This roodel suggests that glass and liquid structures are roore the result of electronic than kinetic factors, and that melting is an electronically driven phase transition . *Research sponsored by the Nat10nal Science Foundation, Grant No . DMR-78-24185 and by the U. S. Army Resea rch Offi ce under Grant No. DAAG-29-80-K-0088 .

4: 20 p.m . I ~ FLCENcr:

2 : 30 p.m. ELECTRONIC STRUCTURE OF METALLIC GLASSES* K. H. Johnson, F .A. Leon, and M.E. Eberhart, Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139

OF HYDRO GEN ADDITION S ON DIFFUSION BONDING OF TITANIUM

Al.LOYS:* R. J . Le derich and S . M. L. Sastry, McDo nn e ll Douglas R(>search i..1b o r a t o ries, S"t. Loui s , Mi s souri and W. R. Kerr, Air For c e ~\'righ t Ae r o na utical Laborato ri e s, Dayton, Ohi o 45433. Th e effe c ts of add iti ons o f u p to 0 . 6 wt % h y dro ge n o n the diffus i on bo nda hi U ty of fine-g r ain ed e quia x ed Ti-6Al- 4V a nd duple x:lnn (' al l,d Ti- 6A I- 2Sn- 4Z r- 2Mo we r e de t e rmined . Th e in teg rit y o f h(lnds wa s l' va 1 U,J tl'd b y o p t i ca 1 mi c roscopy . Hy dro ge n a d'd i t ions I nwer the h L'ta - tra ns us of the a.1 l oys and produ ce va r y ing proportions of the a l pha a nd beta ph a s es . The increas ed a mounts of b(,ta pitas.: and th (' c onsequf'nt hi g he r diffusion r a t es i n hydrogenc nntainjng al1r1yS resul t in sh o rt e r times and l owe r t e mperatures a nd pT(' SS llT(~S for dif f usi o n bo ndin g . Co nditi o n s fo r diffusion bo nd i n g an d r et('nt i on of fin C" - g r ained e q uiaxed a lph a +b e ta mi c roS trt lC' t llrt' i n t Il(' a l loy s wp r e d e t e rmined and we r e co rrel a ted with ;l l pha - bl-'t a vol um(' f r act i ons and gra i n s izes. :"This t=-ps t';lr c h

W.1S

·c-~nducled und e r Air Force Co ntr ac t no .

F1 " hl',-HO - C-S lI R.

4:40 p.m. FIBER- MATRI X RE ACTION PRODUCT ANALYSIS IN B-B4C and B-SiC RU NFORCED Ti-6AI-4V,*, C. G. Rhodes and R.A. Spurling, Physical '~e t a ll ur gy Gr oup, Rockwell International Science Center, rh'us and Oak s , CA 91360 continuous fib e r boron is one of the candidates for reinforcing ,; tanium alloys to produce high modulus composite material. The ,"," on fibers are coated with 84C or SiC to reduce reaction kine';cs between fiber and matri x. However, a reaction zone does torm between coated fibers and matrix in reinforced Ti-6Al-4V consolidated at 927°C. In this work, composites of Ti-6Al-4V reinforced with B-B4C and B-SiC fibers have been studied by transmission electron microscopy to identify reaction products that form during consolidation at 927°C . Three reaction products were found in B-S i C rei nforced t itani um, whereas two reaction products were found in B-B4C reinforced 6-4. The reaction products and their growth characteristics at 927°C will be described. *Research supported in part by Air Force Wright Aeronautical Laboratori es .

CHEMISTRY AND PHYSICS OF RAPIDLY SOLIDIFIED MATERIALS: III Sponsored by the TMS Chemistry and Physics of Metals Committee, TMS Corrosion and Environmental Effects Committee, and ASM-MSD Phase Transformation Activity Wednesday, October 27, 1982 Cervantes Convention Center 2:00 p.m. Room 273 Sesswn Chairman: R. M. Latanision, H.H. Uhlig Corrosion Laboratory, M.LT., Cambridge, MA 02139.

3:00 p.m. EFFECTS OF CAST-IN DEFECTS ON THE MAGNETI C P ROPERTIES OF Allied F'eS2BI2s~6 META.~LI C GLASS: ,:;N",-._ _Dc.e,-C,-r,-i"s,-,t",o.cf.=a~r o. Co rporatlon, Parslppan y, NJ 07 054

Defects in melt-quenched FeS28l2si6 metallic glass ribbon have b e en studied. Sca nnlng elect ron microscopy and scanning Auger microanaly t i c al techniques we re employed to characteri z e the substrat e contact surface of the ribbon which was c omprised of a r o ugh area interrupte d by smoo t h, depress e d p ockets. The r oug h areas c o ntain e d c opper rich clusters whi c h e x tended to a de pth o f approximatel y 5 nm. At d e pth s gre ater than 10 nm, the Fe Band S i c o ncentrations were equa l t o their bulk v alu es . Th e depres sed po c kets were free o f c o p per. However, a boron depl e t e d z o ne extended approximat e ly 250 nItl into the bulk·. Ribbon samples containing such surface inhomogeneitie s generall y d o not respond t o field annealing t e chniques for enhancing magnetic pro per ti e s . It wa s f ound that the deliteri o u s e f fect- on properties could b e tr aced t o the boron de pleted zone . I

3:20 P·m. INFLUENCE OF RAPID SOLIDIFICATION ON TIlE MEOlANlCAL PROPERTIES OF AUSTENITIC MANGANESE STEELS, W. B. F. Mackay and R.

W. Smith, Dept. of li!:tallurgical Eng., Queen's University~ Kingston, Ontario . Hadfield's Steel (Fe-1.3%-14r.fotl) work hardens rapidly under

heavy impact but abrades easily when small loads are applied. In this research programme, various combinations of alloying additions and post-solidification heat-treatment are used to produce the super-fine carbide distribution necessary to obtain good abrasion resistance. However, conventional melting practice tends to result in c oarse carbides being form e d. To overcome this, con ventionally cast alloy blanks were subjected to sursface melting using laser and electron beam melting te c hniques in which the local heat input was well characterized. The influence of these treatments upon the nucleation, subsequent growth and influence of the resulting carbide distribution on the work hardening/fracture properties of the alloys will be reported.

3 :40 p.m. TITANIUM CARBIDE DISTRIBUTION IN NEUT RON IRRADIATED, RAPIDLY SOLIDIFIED AUSTENITIC STAINLESS STEELS OF VARYING TITANIUM AND CARBOr~ CONTENT. D. Imeson, C.H . Tong, J.B. Vander Sande, O.K . Harling, Department of Materials Science and Engineering and ,~uclear Reactor Laboratory, r~IT, Cambridge, MA 02139. Several titanium containing austenitic stainle ss steels, ba s ed on the prime candidate alloy for path A of the fu s ion reactor alloy

TMS Fall Meeting 59 development program, and prepared by rapid sol i dification techniques have been irradiated to moderate dose (up to 20 d.p.a.) in the High Flux Isotope Reactor. In addition simultaneous iron and helium ion irradiations to high dose have been carrled out. It is shown that very small titanium carbide particles precipitate out during irradiation at 500°C and above, in association with dislocations. The effect on the void or helium bubble population produced during irradiation, compared to the effect of pre-existing, larger particles produced by heat treatment pnor to irradiation is discussed, as is the effect of increasing the Ti C content. Rapi d soli difi cat i on process i ng provi des a means of controlling, and in particular increasing, the amount of TiC and its distribution. 4:00 p.m. STRUCTURAL" RELAXATION OF NEUTRON-IRRADIATED Fe40N140B20 METALLIC GLASS, B. Taloul, A. Kursumovic and M. G. Scott, School of Engineering and Applied Sciences, University of Sussex, Brighton, BNI 9QT, England t

Some recent work indicates neutron irradiation enhances free voblme 1n Fe40Ni40B20 metallic glass and properties such as denslty, OJ.rle temperature, and crystallization temperature are affected by irradiation. Metallic glass ribbons of composition Fe40N142B20' which had been relaxed previously, were irradiated to 11gh energy neutron fluxes (>1 MeV) in the range of 6-7xlO 9 n/cm 2 under non-oxidizing conditions. The changes in the structure of the metallic glass by neutron irradiation have been studied and compared with unirradiated samples by using differential scanning calorimetry. stress-relaxation tests. and density measurements. Stress relaxation has been measured as a function of temperature and time for unirradiated and irradiated ribbons.

4:20 p.m. ACCELERATED AGING BEHAVIOR OF AN IRON BASE AMORPHOUS ALLOY FOR 60 Hz APPLICATION A. Datta & R. J . Martis, Metglas Products, Allied Corp. Parsippany, NJ, U.S.A. 07054

The aging behavior of an Fe B13 5i amorphous metal alloy has been studied in the temJ~ra1:url range 250-375°C. Field annealed toroidal cores, excited at 60 Hz, were isothermally aged in nitrogen. Magnetic properties namely 1.4T core loss and exciting power and Hc were monitored as a function of time and fitted to Arrhenius type rate equations. In the low temperature regime (175-2S0°C), it was found that the straight line extrapolation of the high temperature data grossly overestimates the magnitude of the change in magnetic properties . Transmission electron microscopy and x-ray diffraction of the aged samples did not reveal any bulk structural degradation. However, the surface layers of samples, aged at elevated temperatures OSOOC), were found to consist of a-Fe crystallites and exhibited extensive Band Si oxidation. The major aging mechanism at elevated temperatures (350°C) was found to be associated with metalloid (B, 5i) oxidation which, in turn, catalyzed 0. -Fe precipitation.

ENVIRONMENTAL EFFECTS: DEFORMATION TMS-AIME General Abstract Session Wednesday, October 27, 1982 Cervantes Convention Center 2:00 p.m. Room 263 Session Chairman: Ram R. KOSBowsky, Westinghouse Research and Development Center, 1310 Beulah Rd., Pittsburgh, PA 15235 . .

A model for the growth of hydrogen attack cav1ties 1n steel s is presented. I~ the model, closely spaced cavft1es on isolated grain facets grow by grain boundary dtffus10n, where as the strain due to the cav1ty growth 1s accomnodated by creep in the surrounding uncavftated region, following the concept of constrained diffusive growth suggested by Dyson. The model predictions are compared with the incubation stage hydrogen attack kinetic data on plain carbon and 2 1/4 C... _ I Mo steel. The effect of carbides of dfffering stabflities which may result from variations in thermal and mechanical treatments on hydrogen attack in 2 1/4 Cr - 1 Mo steels is conSidered. The model predicts that the cavity growth fs severely constrained under conditions of low methane pressure, low temperature and small fraction of cavitated grain boundaries. It is shown that applied stress will significantly increase the cavity growth rates and hence, wfll promote hydrogen attack at lower temperatures and hydrogen pressures. The Hel son type curves predicted by the model are compared with the standard API Hel son curves.

*Research sponsored by the Department of Energy under contract ET -78-S-01-3153. 2:40 p.m. EFFECTS OF WElDING AND TEMPERING ON THE HYDROGEN ATTACK OF 2-1/4 Cr - 1 Mo PRESSURE VESSEL STEEL: T. A. Parthasarathy and P. G. Shewmon, Department of Metallurgical Engineering, The Ohio State University, Columbus, Ohio 43210 (DOE Sponsored)

The effect of welding on the hydrogen attack resistance of quenched and tempered 2-1/4 Cr - 1 Mo steel was studied by exposure to hydrogen at 20.3 MPa and 480 0 to 580°C. the resulting expansion was measured using a sensitive capacitance dilatometer. The base metal was found to expand at a rate of ""10-8 hr-l at 500°C and exhibited an activation energy of roughly 207 kJ/mo1e between 500· and 590·C. The weld metal expanded at "10-7 hr- l at 500·C but as the temperature was increased (in steps) to 570°C, the expansion rate approached that of the base metal. To study the effect of the extent of tempering on the hydrogen attack of the steel, samples from the base metal were austenitized at 950"'C for 5 hra, quenched ('USOoe/min) and tempered at various temperatures for different times. The expansion reates of the samples were measured at 550°C, at different pressures of hydrogen. At 550°C and 2°63 MPa of HZ the as-quenched structure expanded at a rate of ""10hr-l. Tempering at 650°C for 1. 5 hrs decreased the expansion rate to "10-7 hr-1.

3:00 p.m. HYDROGEN-ENHANCED FATIGUE CRACK-GROWTH IN Ti-6Al-2Sn-4Zr-2Moo.15i : * P. 5. Pao, McDonnell Douglas Re&earch Laboratories, St. Louis, MO 63166.

The kinetics of fatigue crack-growth for as-received and 530 wppm hydrogen-charged Ti-6Al-2Sn-4Zr-2Mo-0 .lS1 (Ti-6242S) in vacuu~ were determined from 290 K to 750 K. .crack-tip stress intens1ty factor amplitude (6K) was used as a measure of crack driving force. Internal hydrogen at a level of 530 wppm increased the fatigue crack-growth rate in Ti-6242S, particularly below 4Z0 K. The temperature dependence of the internal-hydrogen-enhanced fatigue crack-growth of Ti-62425 can be divided into two r~gions. Below 420 K, internal hydrogen strongly influences the fat1gue crack-growth rate, and this influence decreases as the temperature approaches 420 K. Above 420 K, the fatigue crack-growth rate of hydrogen-charged Ti-62425 remains 30% higher than that of the as-received Ti-6242S, and the rate is controlle d by a thermally a c tivated process having an apparent activation energ~ of 8.5 + 0.6 kJ fmol. These resul ts will be qiscussed in relat10n to the possibie cracking mechanisms and rate-limiting processes. *This research was conduc ted under the McDonnell Douglas Independent Research and Development program.

2:00 p.m. THE ROLE OF HYDROGEN IN THE STRESS CORROSION CRACKING OF X-7090 P/M ALUMINUM: R.E. Swanson, I.M. Bernstein and A. W. Thompson, Dept. of Met. Engineering & Mat. Science, Carnegie-Mellon University, Pittsburgh, PA 15213

The resistance of high-strength aluminum alloys to environmentally assisted fracture may be controlled by anodic dissolution, hydrogen embrittlement, or some combination of both, and is known to depend on metallurgical variables such as composition, microstructure, grain shape and texture, and thermal treatment. Test methods are described in which only hydrogen is present in the fracture process, and other methods in which the hydrogen contribution 1s minimuzed. Application of the technique is shown for various microstructures of the Al powder alloy X-7090 to identify the correlations between microstructure and environmental resistance. This work is sponsored by the Air Force Office of Scientific Research.

2:20 p.m. A CONSTRAINED CAVITY GROWTH MODEL FOR HYDROGEN ATTACK IH STEELS*: S. S. Vagarali and G. R. Odette, Department of Chemical and Huc' ear Engineering, University of California, Santa Barbara, CA 93106

3:20 p.m. THE EFFECT OF INTERNAL HYDROGEN ON FATIGUE THRESHOlDS OF HIGH STRENGTH STEEL:' Kh1efa A. Esaklul and William W. Gerberich, Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, MN 55455.

Recent results on the effect of internal hydrogen on fatigue threshold of low strength steel have indicated that internal hydrogen enhances the propensity of inter granular fracture near threshold, different from the behavior in a hydrogen atmosphere. Data on the effect of internal hydrogen on fatigue threshold and the crack growth rate near threshold in high strength steels are limited and different from the behavior in a hydrogen atmosphere. The effect of internal hydrogen concentration on the fatigue behavior of AISI 4340 steel was also studied for a range of hydrogen concentrations and prior austenite grain sizes. The results are discussed in terms of the effect of hydrogen concentration and microstructure in relation to hydrogen embrittlement and fatigue threshold mechanisms. The results are interpreted in terms of a model for the fatigue threshold in relation to static-threshold and measured mechanical parameters. *5~~K~o~_;~~~h8Z~0~orrosion Center of the U. of MN, DOE contract

60 TMS Fall Meeting 3:40 p.m.

2:00 p.m.

THE ROLE OF PRE-EXISTING MICROVOIDS IN THE HYDROGEN ATTACK OF A 2-1/4 Cr - 1 Mo STEEL:* Hugo Lopez and Paul Shewmon, Department of Metallurgical Engineering, The Ohio State University, Columbus, Ohio 43210.

DISCRETE DISLOCATION ANALYSIS OF STEADY STATE VACANCY CONCENTRATIONS AROUND A PLASTIC CRACK: K. Jagannadharn, Engineering Science and Mechanics, Tennessee Technological University, Cookeville, TN 38501

Kinetics measurements on the hydrogen attack of steels indicate that there is a negligible. nucleation barrier for the formation of methane bubbles. A reasonable explanation to account for these indications is to assume that microvolds which are· present in the material act as nucleation sites for methane generation. This work presents evidence of the existance of microvoids whose sizes range between 0.01-0.15 l-Im in 2-1/4 Cr - 1 Mo using a twostage replicas. Three regions of the steel were investigated (base metal, weld metal and heat affected zone). Cavities were found to exist at carbide tips mainly along grain boundaries. The importance of these voids in the resistance of the steel to hydrogen attack is discussed.

The discrete dislocation method of analyzing the crack configuration will be employed to determine the chemical potential associated with a plastic crack and with a lattice dislocation representing the plastic zone. The steady state diffusion equation will be so;Lved for the vacancy concentration around a plastic crack. Crack growth rates at different temperatures and loads will be obtained through the discrete dislocation analysis.

Research sponsored by the Department of Energy.

4:00 p.m. THE EFFECT OF SHOT PEENING ON FATIGUE 01' NICKEL IN HYDROGEN: * C. Verpoort, C ..Garcia, D.J. Duquette and N.S. Stoloff, Rensselaer Polytechnic Institute, Materials Engineering Department, Troy, New York 12181 High cycle fatigue tests have been conducted on 99.99% Nickel in argon and hydrogen gas at 2SoC. ''Fatigue lives were an order of magni tude less in hydrogen' gas when compared with argon gas. At high magnification fatigue striations were noted on the grain boundaries. Precharged specimens tested in low-cycle fatigue showed rapid hardening and very short fatigue lives. After surface treatment by shot peening, fatigue lives increased significantly both in hydrogen and in argon, and fatigue lives in the two atmospheres became more nearly equal. However, no change in crack path was noted after shot peening. The various interactions among hydrogen content, residual stress state, and dislocation distributions will be discussed in order to explain the effects of shot peening. *The authors are grateful to the National Science Foundation for support under Grant No. DMR 80-09241.

4:20 p.m. PLASTIC DEFORMATION AND HYDROGEN TRANSPORT IN METALS: R. -Iv. Lin and H. H. Johnson, Dept. of Materials Science afld Engineering, Cornell University, Ithaca, NY 14853. The electrochemical measurement technique has been used to study the effect of plastic deformation on the transport behavior of hydrogen in pure nickel and SAE 1020 steel near 300 K. The investigation was carried out in two parts. First, the effect of hydrogen charging on the measured anodic background current :::uring plastic deformation was studied. In order to avoid Joule hpClt ing effects associated with electrochemical charging, a gas pha'>02 charging method was also used. Secondly, the response of the steady state hydrogen penneation current to plactic deformation was monitored. Effects of strain rate and specimen surf;l.~e preparation on the measured anodic current were also studird. The role of dislocation transport of hydrogen in nir:ke 1 and steels will be discussed.

:4C p.m. EFFECTS OF ENVIRONMENT ON FATIGUE OF IRON:

D. Ma lumdar and Y. W.

'::;flUng, Dept. of Materials Science & Eng., Northwestern Univer..;.

si t.y, Evanston, Ill., 60201 Samp1.-:. of commercially pure iron were fatigued inside an ultra t1ish vacuum (UHV) scanning Auger microprobe (SAM) chamber. The en'.'U-'':'nment inside SAM was varied from UHV to different pressures of oxygen and water vapor. Surface deformation and microcrack initiation were observed under a standard SEM. Fatigue in UHV produced a rumpled surface (without prominent slip lines) with intergranular cracks. The removal of the natural oxide from the surface by ion sputtering did not change the postfatigue appearance of tne surface. Fatigue in oxygen produced well-developed slip steps with cracks along slip lines. Reasons for the observed dependence of fatigue behavior on the environment will be explored. Effects of water vapor on fatigue will also be presented.

FRACTURE VI: Mechanisms TMS-AIME General Abstract Session Wednesday, October 27, 1982 Cervantes Convention Center 2:00 p.m. Room 260 Session Chairman: Dr. A. F. Giamei, United Tech. Research Center, East Hartford, CT 06108.

2:20 p.m. ORIENTATION DEPENDENCE OF VOID GROWTH IN THE REGION OF THE PLASTIC ZONE AHEAD OF A CRACK TIP: K. Jagannadham, Engineering Science and Mechanics, Tennessee Technological University, Cookeville, TN 38501 The influence of crack tip stress field on the nucleation and growth of micro-voids in the region of the plastic zone ahead of a crack tip are analyzed employing discrete dislocation method. Nucleation of micro-voids is considered at different orientations to the crack so that a favorable orientation that leads greatest decrease in energy is obtained.

2:40 p.m. DIFFUSIVE GROWTH OF GRAIN-BOUNDARY CAVITIES: I-W. Chen, Department of Nuclear Engineering and Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 The latest status of the diffusive growth of grain-boundary cavities is critically examined, focusing on the new creep cavitation experiments in Cu, stainless steels and superalloys at MIT and els..ewhere. In particular, an attempt is made to resolve the discrepancies between experiments and theories of the growth of crack-like cavities. The effects of grainboundary sliding on diffusivities and cavitation are considered. A new model of the diffusive growth of cavities which accounts for the above effects is presented and shown to be in agreement with experiments.

3:00 p.m. THE RELATIONSHIP OF SURFACE MICROPLASTIC DEFORMATION TO CRACK INITIATION IN AL 606l-T6 and AL 22l9-T85l:* W.L. Morris and M. R. James, Rockwell IntI. Science Center, Thousand Oaks, CA. Localized plastic strains exceeding 0.3% develop during fatigue within individual surface grains in alumintun alloys loading elastically. These strains are accommodated by elastic deformation in surroundins. grains, and have been measured over 20 }lm gauge lengths by using mica flakes as references to measure surface displacements. Models relating the peak values of localized plastic strain to onset of· crack initiation in slip steps in Al 6061-T6 alloy, and at constituent particles in Al 22l9-T85l are discussed. The role of atmospheric humidity in altering crack initiation in the two alloys is interpreted using the models. In 2219, predicted and measured numbers of constituent particles fractured are reduced in humid as compared to dry air, as a consequence of reduced microplastic deformation of the surface in humid air. In 6061 crack initiation requires the formation of a critical slip step height. 'The important effect of humidity in this alloy is apparently to alter the height of the step require.d for crack fonnation. *Partial Support from ONR N00014-79-C-0334

3:20 p.m. SULFUR AND Ni + 0.15% Metals and Oak Ridge,

ANTIMONY SEGREGATION TO CREEP CAVITY SURFACES IN Sb*: C. L. White, J. H. Schneibel and R. A. Padgett, Ceramics Division, Oak Ridge National Laboratory, TN 37830

A high purity heat of Ni+0.15% Sb (containing <5 wt ppm S) and a commercial purity heat of the same nominal composition were creep tested at 600°C and 100 MFa. Optical metallography of gage sections from failed creep specimens revealed intergranular cracking and cavitation. These internally cracked gage sections were fractured again at low temperature inside a scanning Auger electron spectroscopy (AES) system. The fracture paths were not completely intergranular, but they did pass through crt;:ep cavities. Sulfur and antimony were found to be strongly segregated to internal cavity surfaces. AES analyses of specimens from these heats will be compared with previously reported results from Ni + 1% Sb and Ni + 1% Sn alloys. *Research sponsored by the Division of Materials Sciences, U.S. Department of Energy under contract W-7405-eng-26 with the Union Carbide Corporation.

TMS Fall Meeting 61 3:40 p.m.

2:00 p.m.

FATIGUE CRACK PROPAGATION IN MODE III: EFFEtTS OF CRACK SURFACE FRICTION: H. Neyab-Hashemi and F. A. McClintock, Dept. of Mechanical Engineering, M.I.T., Cambridge, MA 02139. and ~
TI£ INflUENCE Ck STACKING FAUL T ENERGY AND RESIDUAL STRESSES INDUCED BY SLIDING FRICTION: J. J. Wert, S. A. Singerman, S. G. Caldwell, Dept. of Mechan1cal and Materials Engineering Vanderbilt University, Nashville, ~ 37235

An analysis is made of fatigue crack propagation in Mode III (aoti-

plane shear) in a 4140 steel (tensile strength 590 MJ'a) . Although Mode III growth rates are found to be related to stress intensity range (6K 111 ) for conditions of small-scale yieldiny, no such correlation is found for crack advance under elastic-plastic and fully plastic conditions. A new parameter for torsional crack growth, t~e plastic strain intensity, ( 6 r TI1 ), is developed and shown to provide a unique description of Mode III crack growth behavior in the presence of extensive plasticity. provided a mean load reduces "torsional crack closure" from friction between mating crack surfaces. A micro-mechanical model for such Mode III cyclic crack gro'Wth is presented and shown to be consistent 'Wi th a crack advance per cycle being proportional to the range of plastic strain intensity (~rIII) or the cyclic cra ck tip displacement ( 6CTDIII) .

4:00 p.m. MICROMECHANISMS OF CRACK GROWTH RETARDATION FOLLOWING OVERLOADS: PART 1. NEW INTERPRETATIONS: S. Sure8h, Department of Ma terial. Science and Hineral Engineering and Lawrence Berkeley Laboratory University of California, Berkeley, California 94720 Although the phenomenon of crack growth retardation or arrest due to load excursions during fatigue is experimentally well documented, the micromechanisms underlying such transient effect are not yet clearly understood . In the present work, it is reasoned that crack closure due to residual tensile stretches in the wake of the advancing crack is not the primary mechanism for retardation. A new mechanistic interpretation based on a ''Micro roughness" model is presented to account for the inconsistencies in the existing models. It is shown that the ''Micro-roughness'' model, in conjunction with arguments based on residual compressive stresses and crack branching, rationalizes a number of hitherto unexplained experimental observations cited in the Ii tera ture.

It 15 well known that stacking fault energy can strongly influence the defonnation behavior of pure metals and alloys. The addition of aluminum to copper results in a reruction in the stack i ng faul t energy whi ch in turn enhances pI anar sli p (1. e. , restricts cross slip) and thus should influence the level of reSidual stress induced during sliding contact. Using the x-ray "two exposure" technique, the residual stress present in the surface has been measured parallel and perpendicular to the sliding direction and cOft1lared with the observed wear rates for a series of Cu-Al alloys. It was found that the residual stress Increased with decreasing stacking fault energy In a manner analogous to the measured wear rates. The authors gratefully acknowledge support from National Science Foundation under Grant No. DI'R-8017301 and the SHaRE Program at ORNL.

2:30 p.m. AN ELECTR~ MICROSCOPY STUDY OF SURFACE DEFOR""TlON AND SUBSURFACE DEFECT STRUCTURES INDUCED BY SLIDING FRICTION: S. G. Caldwell, S. A. Singennan, J. J. Wert, Dept. of Mechanical and Materials Engineering, Vanderbilt University, Nashville, ~ 37235 In recent years, the Importance of both initial and evolved nearsurface ml crostructure has recei ved cons i derab 1e attenti on in trlbology literature. The Al-Cu binary has been chosen as a model system for a series of friction and wear experiments. The primary focus of this study has been the correlation of plane-strain surface defonnation produced In unlubrlcated sliding to the resultant subsurface defect structures. Specimen characterization has been perfonned by a combination of light and electron optical methods aimed at correlating the macro and micro-scale features of the wear tracks. Wear phenomena observed In the primary and terminal solid solution ranges for the Al-Cu system have been analyzed In terms of alloy SFE, mechanics of deformation, precipitate structures, and surface oxide films. The authors gratefully acknowl edge support from the Natl ona 1 Science Foundation under Grant No. DMR-8017301 and the SHaRE Program at ORNL.

4:20 p.m. MICROMECHANISMS OF CRACK GROWTH RETARDATION FOLLOWING OVERLOADS: PART II. IMPLICATIONS: S .Suresh, Department of Materia18 Science and Mineral Engineering and Lawrence Berkeley Laboratory, Univer8i ty of Cal ifornia, Berkeley, California 94720 Quantitative analyses are presented for estimates of reductions In effective stress intensity factor range arising from possible crack branching, residual compressive stresses and premature closure due to micro-roughness in the post-overload zone. It is shown that such post-overload phenomena a8 microstructure-sensitive crack growth, absence of striations, abrasion, retarded crack advance well beyond the elastic-plastic boundary and substantial growth attenuation in plane strain can be rationalized in terms of the suggested interpretations. The implications of the model are discussed with the aid of extensi:ve experimental evidence available in the literature on fractography, electron channeling patterns, in-situ SEH analyses and attenuated crack growth data following load excursions and in light of the measurement of fatigue crack growth rates especially at low stress in tensi ties.

4:40 p.m. ON THE RELATIONSHIP OF STRETCHED ZONE SIZE AND COD IN Jic TESTS: C.M. Gilmore, V. Provenzano, F.A. Smidt, J.R. Hawthorne, Thermostructural Materials Branch, U.S. Naval Research Lab, Washington D.C., 20375 The nature of the mi croscopi c processess occurri ng at the crack tip during crack initiation and the early stages of propagation were investigated by detai led characterization of the stretched zone in J-i ntegra 1 specimens (£f HT -9 steel of 12.7 and 2.5 om thickness tested at 25 and 200 C. The topography of the crack tip region was mapped by detai led sterographic measurenents on scanni ng electron mi crographs of the regi on contai ni ng the stretched zone. The measured J values and the stretched zone dimensions for each test conditib'h were then compared using the relationship of Knott, J, = mo 6, Theory predicts values of m of about two for specimen~ undef JI1ane strain conditions and lower values of m for thin specimens under plane stress conditions. The current experiments showed good agreenent )ii th theory for the thi cker specimens but lar~er values of m, near 6.0, for the thin specimens. A rationa e for this lack of agreenent and an analysis of the crack tip deformation processes are presented.

3:00 p.m. AN X-RAY DIFFRACTION ANALYSIS OF WEAR INDUCED SUBSTRUCTURAL CHANGES IN Cu-Al ALLOYS: S. A. Singerman, S. G. Cal dwell, J. J. Wert, Department of Pecha", ca I and Materi a 1sEngi neeri ng, Vanderbilt University, Nashville, ~ 37235, D. K. Chaudhur·i, Dept. of Mechanical Engineering, Tennessee State University, Nashville, ~ 37203 Several copper-aluminum alloys have been examined using x-ray Ii ne broadenl ng techni ques to he I p characterl ze the effect of stacking fault energy (SFE) in the wear process_ This study involved surface deformation by an unlubricated slider on a reciprocating wear machine in a predominantly argon atmosphere. The resul ti ng wear scar was exami ned wi th x-ray diffracti on techniques, and the characteristic f.c.c. peaks were analyzed usi ng the method of mul ti pI e orders to determi ne the stacking fault probability, the root mean square strain, and the effective particle size of each sample. The x-ray determined parameters will also be correlated with the wear rate. The authors gratefully acknowledge support from National Science Foundation under Grant No. DMR-8017301 and the SHaRE Program at ORNL.

3:30

p~m.

ORIENTATION ANALYSIS IN SUBSTRUCTURES DEVELOPED DURING SLIDING:* P. Heilmann, t W. A. T. Clark, and D. A. Rilney , Metallurgical Engineering, The Ohio State University , Co umbus, OH 43210 . We report STEM measurements on longitudinal sections of OFHC copper blocks subjected to unlubricated sl iding for different times. A well-defined cell structure develops very early (1 pass on an LFW block-on-ring test machine); this cell layer extends further below the interface for larger sliding distances. Kikuchi patterns obtained from a large number of individual cells have been analysed. An interactive computer program allows rapid and accurate determination of the relative orientation of each diffracting region. The results show that plastic deformation causes large rotations near the sliding interface; these decrease roughly exponentially below the surface. There is no evidence of a stable preferred orientation of surface cells . Instead, the lattice in each cell continues to rotate as sliding continues. This suggests that similar effects may be important during other deformation processes which involve large plastic strains.

FRICTION AND WEAR TMS-AIME General Abstract Session WedDellday, October 27, 1982 Cervantes Convention Center 2:00 p.m. Room 265 Session Chairman: Dr. David L. Bourell, Dept. of Mechanical Engineering, Taylor Hall, Room 167, University of Texas-Austin, Austin, TX 78712.

*Research sponsored by The National Science Foundation under Grants No. DMR-7805719 and DMR-8120675 . The authors are also grateful to Oak Ridge National Laboratories for use of STEM facilities as part of ORNL's SHaRE program. t Now at BASF AG, VMT/EP Magnettechnik, 0-6700 Ludwigshafen, West Germany.

62 TMS Fall Meeting 4:00 p.m. FRICTIONAL BEHAVIOR OF INVAR ALLOYS: Z. Eliezer and R.Y. Lee, Oepartment of Mechanical Engineering, The Universi ty of Texas, Austin, Texas 78712. Austenitic iron-nickel alloys in the composition range of about 30-40% Ni (the Invar range) exhibit many anomalies: a very low coefficient of thermal expansion at about 36% Ni, low electrical and thermal conductivity, sharp changes in saturation magnetization and hyperfine field, etc . Heavy cold-rolling of a 36% Ni Invar alloy results in a decrease in magnetization and magnetic hyperfine field and an increase in electrical resistivity and x-ray line-width . These changes can be explained by the transition of iron atoms from a high-spin high-volume configuration to a low-spin low-volume configuration. The possible effect of such changes on the friction and wear behavior of Invar alloys will be discussed.

GRAIN BOUNDARY DIFFUSION AND KINETICS I: Basic Kinetic Processes Sponsored by the ASM-MSD Atomic Transport Activity Wednesday, October 27, 1982 Cervantes ConventionCeilter 2:00 p.m. Room 272 Session Chairman: R. W. Balluffi, Dept. of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139.

2:00 p.m. STRUCTURE OF GRAIN BOUNDARIES IN METALS, V. Vitek, Department of Materials Science and Engineering, University of Pennsylvania, Philadelphia, PA. 19104 The geometrical classification of tilt boundaries in terms of variation of the misorientation (and coincidence) for a fixed mean boundary plane and variation of the mean boundary plane for a given misorientation will be presented first and relationship of this concept to boundary decomposition and faceting discussed. It will then be shown that the atomic structure of general, low coincidence, symmetrical and asymmetrical boundaries with the same mean boundary plane can be described in a given misorientation range as consisting of units of two "favored" boundaries bounding this range. Application of this prinCiple to twist boundaries will also be discussed. The relationship of the decomposition model to the dislocation models of grain boundaries will be elucidated and implications for the study of properties of grain boundaries discussed.

4:15 p.m. MECHANISMS OF GRAIN BOUNDARY MIGRATION AND POINT DEFECT SOURCE/ SINK ACTION: A. H. King, Dept. of Materials Science & Engineering, State Univ. of New York, Stony Brook, N.Y . 11794 Modern theories of atom-by-atom processes occurring in grain boundaries (e.g. the jumping of atoms across the boundary and the absorption or emission of point defects) suggest that these processes may occur at sites on line defects within the boundaries. Some general features of the possible types of line defects will be described and it will be shown that they have the properties of dislocations and of steps of the boundary plane. Motion of the line defects thus causes translations of the two grains with respect to each other as well as migration of the grain boundary. The linking of these processes through the structure of the line defects produces many interesting consequences, which will be catalogued and discussed briefly.

LIQUID AND SOLID METAL INDUCED EMBRIT· TLEMENT OF METALS V Sponsored by the TMS Corrosion and Environmental Effects Committee and the ASM -MSD Corrosion and Oxidation Activity Wednellday, October 27, 1982 Cervantes Convention Center 2:00 p.m. Room 270 Session Chairman: L. F. Coffin, Corporate Research & Development Center, General Electric Co., Schenectady, NY 12201.

2:00 p.m. LIQUID METAL EHBRITTLEMENT OF ZIRCALOY-2 BY CADMIUM AND OTHER ELEMENTS: \.]. T. Grubb, General Electric Co., Schenectady, NY

In water cooled nuclear reactors, fission products are formed inside tubing made of Zircaloy. For this reason, it is important to determine their effects upon its mechanical properties. Besides the well documented stress corrosion cracking of Zircaloy oy iodine, liqUId metal embrittlement was considered possible. Therefore, a num~¥-r ,',r metals ~ alone and dissolved in cesium (an abundant fissioo product), were surveyed for their effect upon the mechnical properties of Zircaloy tensile samples. Of these, only cadmium caused severe embrittlement. This new LME system was studied with regard to the state of cadmium, liquid, solid, or in cesium solution; the temperature; and the Zircaloy texture. The results are reported in terms of constant extension rate load vs time measurements and fractography of the fracture surfaces. Cesium with a liquid range from 28 - 7D5 0 C is an interesting sovlent metal for LME studies.

2:30 p.m.

2:45 p.m. STRUCTURE AND ENERGIES OF POINT DEFECTS IN GRAIN BOUNDARIES: P. D. Bristowe, Massachusetts Institute of Technology, Room 135157. Cambridge, MA 02139 The equilihrium structure of vacancies and interstitials in grain boundaries has been invest igated by computer molecular stat ics employing pairwise potentials for metals. The point defect induced displacement field in the boundary has been classified in term s of a number of variables including metal type, boundary type, degree of lattice coincidence, and choice of boundary site. Although the defects did not lose their identity in the boundary they sometimes induced relaxations which were large and widely distributed. Point defect binding energies have been estimated and the relevance of the resul ts to the mechanism of boundary self-diffusion discussed.

3:30 p.m. COMPUTER MOLECULAR DYNAMICS SIMULATION OF GRAIN BOUNDARY DIFFUSION: Sidney Yip,* Massachusetts Institute of Technology, Cambridge, MA 02139; Thomas Kwok and Paul S. Ho, IBM Watson Research Center, Yorktown Heights, NY 10598 Application of the computer molecular dynamics simulation technique to diffusion kinetics and grain boundary stability at high temperatures will be discussed. Results on vacancy migrations in a core of bcc iron 1:=5 tilt boundary simulated over a range of temperatures will be used as illustration. In this study, the data at intermediate temperatures, about two-thirds melting, provided detailed evidence that vacancy exchange is the dominant mechanism for fast self-diffusion along grain boundaries. Thermal activation of one type of boundary interstitial was observed and the interstitials found to be mostly short-lived and immobile: The molecular dynamics results show clearly strong correlation between diffusion and boundary structure, with the specific information on vacancy jumps between different boundary sites conveniently summarized in a transition probability matrix. *Work supported by the U. S. Army Research Of fice.

SOLID CADMIUM EMBRITTLEMENT OF ZIRCALOY-2. PART I: TENSILE FRACTURE PHENO!'IENA AND MECHANISMS : ~Gangloff; Exxon Res e arch and Engineering Co" CR-SLj P.O . Box 45; Linden, New Jersey 07036 Textured Zircaloy-2 was embrittled severely by solid Cd during continuous, plane strain loading of sheet specimens at 575K. Cd assisted cracking was transgranular along near-basal planes 1n co ntrast to the ductile dimpled morphology observed for fracture in vacuum. StreSSing parallel to the maximum concentration of basal poles re s ulted in brittle fracture at the material yield strength. Increased plast ic st rains were requi red for embrit t lement as the tensile axis was rotated perpendicular to the basal poles. The strain to fracture of the suscept ibIe texture increased .:IS the strain rate increased, suggesting a transport limited embr!.ttIement process. Resistance to brittle fracture was increased significantly for recrystallized Zr-2 and for pure zirconium, consistent with previous studies. Copper additions to the solid cadmium reduced the degree of embrittlement for Zr-2, while influencing only slightly the activity of Cd in the Cu-Cd alloy. The results of t e nsile fracture experimfi>nts contributed to understanding of the behavior of barrier modified Zircaloy tubing for nuclear fuel containment.

3:00 p.m. SOLID CADMIUM EMBRITTLf.HENT OF ZIRCALOY-2. PART II: SIHULATlON BEHAVIOR OF BARRIER TUBING; ~_Ganglof f; Exxon Research and Engineering Company; CR-SL; P. O. Box 45, Linden, New J e rsey 07036 The solid Cd embrittlement resistance of copper plated and iodide zirconium lined barrier tubing was superior to that of Zircaloy-2 based ory the expanding mandrel simulation. Zr-2 was embrittled severely by slow strain rate localized deformation with either solid Cd or molten Cs-Cd at 575K. Brittle cracking was characterized by a three-fold reduction in local fracture strain compared to argon results, and by a transition from ductile dimpled rupture to transgranular cleavage and fluting. For zirconium lined 2r-2, fracture strains and morphologies were equivalent for the solid Cd and Ar envi ronments. Bri t t Ie fracture was observed for the Cs/Cd envi ronment, albeit at large plastic strains. For copper plated Zr-2, equivalent high fracture strains were observed for inert and solid cadmium environments independent of barrier thickness, defects and

TMS Fall Meeting 63 Cu-Zr ir.terfacial phases a The resistance of Cu barrier tubing was attributed to alloying between Cd and Cu which altered the Cd supply kineticsa Molten Cs acted as a carrier for Cd and exacerbated embrittlement when the integrity of the cop'Per barrier was breached.

3:30 p.m. EIIBRITILEMENT OF IRRADIATED ZIRCALOY BY CADMIUM AND IODINE: S.B. Wisner and R.B. Adamson, C.E. Co, Vallecitos Nuclear Center, Pleasanton, CA 94566 Zircaloy-2 used as cladding material for nuclear fuel, in instances is known to develop cracks by the phenomena called p~llet-cladding interaction or PCL The laboratory simulation tests under PCI conditions indicate that cracking is enhanced by irradiation and the presence of nuclear fission products iodine or cadmium. This paper investigates embrittlement by liquid cadmium and compares it with iodine stress corrosion cracking. Plane strain specimens of neutron irradiated Zircaloy2 were tensile tested at 337·C in argon, liquid cesium, liquid cadmium dissolved in cesium (Cs/Cd), and gaseous iodine. Argon and cesium caused large failure strains and ductile fracture surfaces. Iodine and Cs/Cd both strongly embrittled the Zircaloy with low ductility and fracture by brittle inter and transgramiar mode. Cadmium is more aggressive than iodine, resulting in lower fracture loads and a notch condition sensitivity. Elimination of the irradiation induced defects by thermal annealing ' results in relative immunity of Zircaloy-2 to LME by cadmium. ' The results are discussed in terms of possible embri ttle mechanisms.

4:00 p.m. EVALUATION OF ZIRCALOY CLADDING CONCEPTS IN CESIU>,\-CADMIUM BY THE LOCALIZED DUCTILITY METHOD: L.F. Coffin, Corporate Research &: Development, GE Co., Schenectady, NY 12301 The General Electric Co. has undertaken an extensive program to determine the causes of and possible solutions for cracking in Zircaloy nuclear fuel cladding, culminating in the development of a PCI (pellet-clad-interaction) resistant cladding with a metallurgiGally bonded liner of highpurity zirconium. This paper reports on one of the investigations undertaken in which the localized ductility method was employed to determine the resistance of a number of composite fuel cladding concepts to cesiumcadmium liquid metal embrittlement. The localized ductility method subjects a small section of an actual cladding to a hoop stress in the environment and temperature of interest, the stressing being performed statically or at constant strain rate on either defected or defect-free tubing. Cs-Cd liquid metal exposure was found to cause significant embrittlement in some instances. With respect to Zr-2, an embrittlement threshold was noted, depending on the amount of strain hardening prior to or during the test. The best overall performance found was that of the crystal-bar liner. This was attributed to its low strength (Jess than the embrittlement threshold) and greater notch ductility.

4:30

p •••

DEVELOPIlENT OF ZR-BARRIER FUEL: H. S. Rosenbaum, J. S. Armijo, J. H. Davies and R. P. Tucker, Nuclear Engineering DiviSion, General Electric Co., San Jose, CA 95125

Water cooled nuclear power reactors which are fueled with Zircalay-clad urania, are subject to fuel failures by the mechanism, I1pellet-cladding interaction ll (PCl). Such failures occur after substantial burnup and depend on power history. Available data indicate that PCI is either stress-corrosion or liquid metal emhrittlement of the Zircaloy fuel cladding by one or more of the fission produces in the fuel. The Zr-barrier fuel WaS designed to resist failure by PCl. In this fuel design the conventional Zircaloy cladding is replaced by a special cladding whose inner surface has a layer of pure zirconium metallurgically bonded to a substrate of Zircaloy. A testing program which includes power ramps in a test reactor has shown that Zr-barrier fuel is substantially more resistant to PCl than is conventional fuel. Based on encouraging test results to date, a large scale demonstration of Zr-barrier fuel has been started in a commercial power reactor.

MULTIPLE LOADING EFFECTS ON FATIGUE I Sponsored by the ASM Flow and Fracture Activity Wednesday, October 27, 1982 Cervantes Convention Center 2:00 p.m. Room 266 Session Chairman: J. C. M. Li, University of Rochester, Rochester, NY 14627. .

With higher OKbase levels, associated with plane stress condi ... tions, the amount of delay increases sharply with ~Kbase' In this investigation it is seen that a third regime exists carre'" sponding to crack growth rates in the threshold regime. At these lower AKbase levels, much more delay was encountered with decreasing ~Kt,asea Since the slope of the da/dn VB IlK plot in this region is significantly higher than at the intermediate growth rate levels, a small change in the effective ~K level of the crack tip will have a major impact on da/dn and hence overall life.

2:30 p.m. WHERE IN A STRESS CYCLE DOES A FATIGUE CRACK BEGIN TO PROPAGATE?: Johannes Weertman, Department of Materials Science and Engineering and Materials Research Center, Northwestern University, Evanston, IL 60201

In previous papers (see Fatigue and Microstructure (1979) American Society for Metals, p. 279-306) we gave a rational for the claim that the power exponent of the Paris equation of fatigue crack growth will have different values depending on the point in the str~s~ cycle the fatigue crack begins its incremental growth each cycle. If incremental growth starts only when the cyclic streliS" has almost reached its maximum value the power exponent shoula be approximately equal to four. If the growth occurs at a stress level well below the maximum value the power exponent shou\'Ci be approximately equal to two. In this paper the double slip plane crack model of lng-Hour Lin, Robb Thomson and ourself (submi'tted for publication), which from qualitative arguments led to, a fourth power Paris equation, is studied to determine where "in the stress cycle incremental fatigue crack growth starts. At the time of the writing of this abstract no definite results have been obtained.

3:00 p.m. EFFECT OF VARIABLE AMPLITUDE LOADING ON THE FATIGUE LIFE OF Ti6AI-4V. Shan Zhi Liu and A. J. McEvily, Dept. of Metallurgy, University of Connecticut, Storrs, CT 06268.

A study of the effects of variable amplitude loading on the fatigue life of Ti-6Al-4V has been carried out. Particular emphasis has been placed on the effect of high stresses on high cycle fatigue behavior, particularly in the region below the nominal 10 7 cycle fatigue strength. A strong history effect is observed which is associated with mobile dislocations introduced at the high stress levels. Planarity of glide promotes this history effect. Electron microscopy is used to study the variation in dislocat~on density with stress amplitude. The results are discussed in terms of current theories of load-interaction effects.

3:30 p.m. PLASTIC ZONE STRAINS AND STRESSES DUE TO AN OVERLOAD IN 7075-T6: D. L. Davidson and J. Lankford, Southwest Research Institute, San Antonio, TX 78284 Material response due to a single overload is examine
PROCESSING METHODS - WORKINGIFORMING - Session IV GEMIG-PAC Productivity Symposium Sponsored by the TMS-AIME Governmental, Energy and Minerals Committee and the ASM Government and Public Affairs Committee Cervantes Convention Center Wednesday, October 27, 1982 Room 275 2:00 p.m. Session Chairman: S8.I\iay Shah, Wyman·Gordon Company.

2:00 p.m.

Abstracts for the presentations in this session will be printed in the Addenda to the Pocket Program distributed a~ the meeting.

EFFECT OF OVERLOADS rn FATIGUE CRACK PROPAGATION BEIlAVI(l! IN AN EXTRUDED AWMINUM ALLOY, R. S. Vecchio, R. W. Hertzberg, Lehigh University, Bethlehem, PA; G. Miller, K. Donald, Del Research, Hellertown, PAt and R. Jaccard, AluBuisse, Zurich, Switzerland.

2 :00 p.m .

Fatigue crack growth retardation due to overloads in an extruded Al alloy is examined as a function of the baseline stress intens'ity. Previous studies at Lehigh University have shown that crack growth delay can be divided into two regimes according to the prevailing stress intensity level. Overloads applied at low to intermediate K levels, corresponding to a predaninantly plane strain condition, have a modest effect on the amount of delay.

RECENT DEVELOPMENTS IN FORGI NG TECHNOLOGY: A STATE-OF-THE-ART REVIEW: T. Altan, Senior Research Leader, and R. Raghupathi, Senior Research Scientist, Battelle Columbus Laboratories, Columbus, OH.

64 TMS Fall Meeting 2:35 p.m.

3:00 p.m.

HOT DIE AND ISOTHERMAL FORGING OF DISKS AND AIR FRAME COMPONENTS: J. D. McKeogh, Manager, Advance Products Development, and S. Shah, Research Project Engi neer, Wyman-Gordon Company, No. Grafton, MA.

THE DEPENDENCE OF HIGH TEMPERATURE MECHANICAL PROPERTIES OF AUSTENITIC STAINLESS STEELS ON IMPLANTED HELIUM: Herbert Schroeder, Peter Batfalsky and Josef Rothaut, IFF der KFA Julich, D-5170

3: 10 p.m.

High temperature helium embrittlement effects on the creep properties of AISI 316 SS (s.a.) and DIN 1.4970 SS (c.w.+a.) have been investigated. The generation of helium due to (n,a) nuclear reactions in a fusion reactor envifonment has been simulated by homogeneous helium implantation at a cyclotron. The creep rupture tests with various applied tensile stresses have been carried out at 1023 K (316 SS) and 1079 K (1.4970 SS), respectively,with four differently treated sets of samples: 1) Un implanted controls; 2) after room temperature implantation of 100 appm He; 3) after implantation of 100 apJ:S He at test temperature; 4) creep tested at high temperature during implantation ("in-beam") with implantation rates of 10 - 100 appm He/h . In contrast to the ductile behaviour wi th transgranular fai lure of the unimplanted controls, all He-implanted samples showed brittle, inter granular early failure. The embrittlement effect was enhanced in the high temperature implanted samples. The creep results are correlated to heliun bubble size distributions and densities from extensiv TEM work.

NEAR NET RADI AL FORG I NG OF TUBULAR COMPONENTS: V. Fencl. Manager of R&D, Grotnes Metalforming Systems, Inc., Chic~ 3:45 p.m. ISOTHERMAL ROLL FORGING OF COMPRESSOR BLADES: AN AUTOMATED PRODUCTION PROCESS: F. K. Rose, Research Staff Specialist, and J . J. Anderson, Research Engineer, Solar Turbines InterNational, San Diego, CA. 4 :20 p.m.

MODELS FOR THE COMPACTION OF METAL POWDERS IN HOT ISOSTATIC PRESSING: Prof. W. R. D. Wilson, Dept. of Mechanical and Nuclear Engineering, Northwestern University, IL.

RADIATION DAMAGE ANALYSIS FOR FUSION REACTORS V. Correlation Models and Mechanical Properties Sponsored by the TMSIASM Nuclear Metallurgy Committee Wednesday, October 27, 1982 Cervantes Convention Center 2:00 p.m. Room 271 Session Chairmen: R. H. Jones, Pacific Northwest Lab., P.O. Box 999, Richland, WA 99352; W. G. Wolfer, Nuclear Engineering Department, University of Wisconsin, 1500 West Johnson Drive, Madison, WI 53706.

Ji.i.lich, F. R. Germany, Association KFA-Euratom

3:20 p.m. HELIUM EMBRITTLEMENT OBSERVATIONS IN 9Cr-IMoVNb AND 12Cr-IHoVlo/ R. L. Klueh and J. M. Vitek, Metals and Ceramics STEELS:

*

Division,

O~RfcJge

National

IRRADIATlON EMBRTTTLE)IE!';T MECHANISMS A;W PARA.'IETERS IN STAINLESS STEELS FOR FUSION APPI.lCATIONS:* G. R. Odett e and G. E. Lucas

Irradiat ion redu ces the fracture resistence of both austenitic and fcrritic stainless steel alloys. However, quantitative measures of embrittlement are a function of a number of intrin s ic and extrinsic variables. At one extreme, constrained yielding and unique triaxial crack tip fields, the parameter of interest is fracture toughness; toughness can describe the load (below yield) at which fracture initiation takes place as a function of flaw size. Critical stress (cleavage) and critical strain (microvoid coalescence) fracture mode models are used to calculate the effect of irradiation indu ced changes in strength, work hardening and , !t:, "t i I il)" p;lram('lers. In the case of unconstrained yielding, t;i\..: influence of stress-state, non-unique fields an-d micromechanical processes on post-yield ductility is of interest. This case is "modeled us ing cont inuous mechanics and micromechanical damage models. The influence of severe flow localization on these embrittlement phenomena is discussed.

"'Researc h sponsored by the Department of Energy.

2:20 p.m. THE EFFECT OF HELIUM ON GRAIN BOUNDARY MICROSTRUCTURES AND DEFORMATION AND FRACTURE:* G. R. Odette and P. J. Maziasz

Helium has a strong effect on some mechanical properties of metals, particularly at elevated temperatures where it readily diffuses to grain boundaries, resulting in enhanced intergranu]ar failure and reduced creep strength and ductility. One mechanism believed to be responsible is the formation of grain boundary helium bubbles and stress induced growth of these bubbles. Other possible factors include the effect of helium on: boundaryene"rgy and diffusivity; boundary precipitates; and matrix microstructure and strength. This paper reviews microstructural and mechanical property effects of both implanted and neutron induced helium in nitic stainless steels; emphasis is on mixed spectrum (high helium) and fast spectrum (low helium) fission reactor results. Correlation models based on physical mechanisms are described and compared to the existing data. *Research sponsored by the Department of Energy.

2:40 p.m. PRECIPITATION OF HELIUM I N MJ:.'rALS AND ITS EFFECT ON HIGH-';'HTERA-

TIJRE EMBRITTLEt-1FNT: H. Trinkaus, Kern forschungsanlage Jul ich, Institut fur Festkorperforschung, Postfach 1913, D-5170 Ji.i.lich. High-temperature embrittlement of metals containing He is due to the formation of He bubbles on grain boundaries. Recent progress in the understanding of the behaviour of He and of the nucleation and grmrth of He bubbles in metals is reported. The results are applied to the modelling of various types of c reep tests and of recent high temperature fatigue experiments. The stress and temperature dependence of the time to rupture in the creep tests and the observed transition of the fatigue failure mode from transgranular to intergranular at a certain fatigue frequency and temperature is discussed.

Oak Ridge,

Tennessee

Small amounts of helium (15- 40 at. ppm) are known to cause embrittlement in a large number of alloys when tensile tested at elevated temperatures. To determine the effect of helium on ferritic steels of interest for fusion reactor applications, tensile specimens of normalized-and-tempereti 9Cr-1HoVNb with 0 and 2% Ni and 12Cr-1MoVlo/ with 0,

2:00 p.m.

Laboratory,

37830

I, and 2% Ni were irradiated in the

High Flux Isotope Reactor (HFIR) at -SSoC. Transmutation helium up to -SO at. ppm was produced by a two-step reaction of seNi with thermal neutrons. No embrittlement was detected when tested at 700°C. Studies reported in the literature indicate that relatively large decreases in total elongation occur for types 304, 304L, 316L, Qnd 316H austenitic stainless steels containing 15--40 at. ppm helium. Although 700°C is well above the temperature at which the ferritic steels will be used, the fact that no embrittlement occurs may indicate that helium will also have less effect in these ste~ls at applicable temperatures. .Research sponsored by the Office of Fusion Energy, U. S. Department of Energy, under contract W-7405-eng-26 with Union Carbide Corporation.

3:40 p.m. A TEM METHOD FOR ANALYZING LOCAL STRAIN FIELDS IN IRRADIATED MATERIALS: J. I . Bennetch and W. A. Jesser, Department of

Materials Science, Thornton" Hall, University of Virginia, Charlottesv; lie, VA

22901

Of great interest to the field of fracture mechanics is the

shape of strain fields in front of crack tips. In irradiated materials, cavities which naturally form as a result of rai~i­ ation provide convenient internal markers. If a miniaturized irradiated tensile sample is pulled in-situ in a transmission electron microscope (TEM), both the relative displacement of these cavities and their distortion in shape provide information on localized strain on a small microscopic level. In addition, the TEM method allows direct correlation of active slip systems with crack propaciation characteristics. To illustrate this method a strain fi~ld map was constructed about a crack propagating in a helium irradiated 316 stainless steel sample containing large cavities. Research sponsored by the Department of Energy, Office of Fusion Ener9Y.

4:00 p.m. THE EFFECT OF NEUTRON AND He-IRRADIATION atl THE CRACK MODES OF TYPE 316 STAINLESS STEEL: T. Hanamura and W. A. Jesser, Department of Materials Science, Thornton Hall. University of Virginia, Charlottesville, VA 22901 From in-situ HVE~·I observations of fracture mechanisms in unirradiated neutron-irradiated and He-irradiated SS 316 specimens, using a video recorder. crack directions, a, and crack opening · displacement vectors were measured during crack propagation. As a crack propagates the mode of cracking may change many times. The % Mode I cracking was plotted against the crack direction. n. From this data it was found that Mode II does not occur independently since when Mode II occurs a Mode I crack always precedes it. Grain boundary sliding may be initiated when a f10de I crack intersects a grain boundary at the appropriate orientation and can be suppressed by He-irradiation and neutron irradiation. There was a tendency of crack propagation: Grain boundary sliding dominates when n is close to O. Crystallographic holes resulting from slip make "Dragon's Teeth" at the crack flanks when n is close to 45 ° . Straight and sharp crack propagation occurs when, is close to goo. Sponsored by the Department of Energy, Office of Fusion Energy.

TMS Fall Meeting 65 4:20 p.m. CRACK PROPAGATION CHARACTERISTICS IN NEUTRON IRRADIATED TYPE 316 STAINLESS STEEL MICROSPECIMENS: R. D. Gerke and W. A. Jesser, Department of Materials Science, Thornton Hall, University of Virginia, Charlottesville, VA 22901 In order to better r elate the macroscopic mechanical ' behavior of irradiated alloys to their associated microstructural condition, neutron irradiated microscpecimens were tensile tested at 25-600 0 C in a quantitative load elongation stage while under continuous observation in a high voltage electron microscope (HVEM). The microtensile specimens, 40 ~m thick, of type 316 stainless steel were irradiated at ambient temperature to a fluence of lxl022n/m2 in the RTNS II facility. A graph of cumulative crack length plotted against total elongation has a slope which increases as specimen ductility decreases. Also , high crack propagation velocitie s are accompanied by rapidly changing crack direction and conversely slow velocities correspond to slowly changing directions. One effect of neutron irradiation is to enhance these cha racteri s i tcs. Research sponsored by the Department of Energy, Office of Fusion Energy.

4:40 p.m.

*

THE DEVELOPMENT OF SMALL SPECIMEN MECHANICAL TEST TECHNIQUES, G.E. LUCAS, Department of Chemical and Nuclear Engineering, Uni versity of California Santa Barba ra, Santa Barbara, CA 93106 The current program plan for the development of materials for f usion reactors requires testing candidate materials in both fission reactors and high energy neutron sources. Because of the volume limitations of available f a cilities, both current and near term, and because of the relatively large number of materials and test conditions that will need to be explored, it is essential that test techniques be developed to extract mechanical property information from small volume specimens. A variety of such test techniques are under development at the Universi ty of California Santa Barbara. These include instrumented microhardn e ss, bulge, shear punch, indentation creep and load relaxation and miniaturized fracture tests for obtaining strength, ductilit y , timedependent flow, and fracture behavior information on specimens as small as TEM dis c s. *Re search sponsored by the Department of Energy Office of Fusion Energy.

RECENT ADVANCES IN SUPERALLOY TECH· NOLOGY II Sponsored by the TMS High Temperature Alloys and Heat Treatment Committees Wednesday, October 27, 1982 Cervantes Convention Center 2:00 p.m. Room 267 Sessicn Chairmen: J . H. Wood, General Electric Co., Bldg. 53, Rm, 316, Schenectady, NY 12345; C. G. Beck, Westinghouse R&D Center, 1210 Beulah Rd., Pittsburgh, PA 15235.

PAPER WITHDRAWN AN ANALYSIS OF THE CREEP BEHAVIOR OF NICKEL-BASE ALLOYS: O. Y. Chen and D. N. Duhl, Pratt & Whitney Aircraft, 400 Main Street, East Hartford, CT 06108 . The development of high strength turbine airfoil materials has resulted in a need for a better understanding of the creep mechanicms occurring in nickel-base superalloys. As part of a NAVSEA sponsored program evaluating corrosion resi s tant single c rystal superalloy s, the creep properties of a series of e xperimental alloy s, with a wide ran ge of y' volume fractions from 3Ov/o to 55v/o and a range of y / y f mismatch b etween 0% to 0.5% have been extensively documented. The creep behavi or of these alloys will be discussed in terms of primary and secondary creep properties and their dependence on test conditions, thermal history a nd microstructure. The extent of primary creep was found to be directly proportional to the size of the y t as well as the test stress and inversely proportional to the volume fraction of y t . The second s tage creep rate was found to be controlled by the y /y t lattice mismatch and the volume fraction of y t •

2:00 p.m. EPFBCT OF MAJOR ELEMENT CHEMISTHY ON THF. HEAT TREATMENT WINDOW OF SINGLE CRYSTAL ALLOYS: M J Goulette and R P Arthey, Materials Engineering Group, Rolls Royce Ltd, PO Box 31, Derby DE2 BBJ, England. The introduction of economic production methods for gas turbine components in single crystal form allow s ne" scope for alloy development. In order to take full advantage of the new technology such alloys need to have a pOSitive temperature interval between the gamma-prime solvus and the incipient mGlting point, known as the heat treatment window. It is generally found that strengthening element addit i ons have the effect of reducing the heat treatment window. In the current

investigation the heat treatment window is shown to be significantly dependent on the time at temperature. The heat treatment windows of a series of 24 alloys, in which the composition is changed in a systematic manner, have been examined to show the effects of different contents of aluminium, ti tanium, tantalum, vanadium, niobium, tungsten and molybdenum. This examination provides valuable information for the design of new alloys for single crystal applications.

2:20 p.m. INFLUENCE OF Co, Ta, AND W ON THE CREEP RESISTANCE OF NICKELBASE SUPERALLOY SINGLE CRYSTALS. M. V. Nathal and L. J. Ebert, Dept. of Metallurgy and Materials Science, Case Western Reserve University, Cle,!,eland, OR 44106 The purpose of this study was to determine the influence of Co, Ta, and W on the microstructure and mechanical properties of Mar-M247 Single crystals. A matrix of alloy compositions was based on Mar-M247 stripped of C, B, Zr, and Hf. Co was varied between 0 and 10 w/o, Ta between 0 and 3.5 w/o, and W between 10 and l;3 w/o. Tensile and creep':"'rupture tests were performed at 1000 C.' The microstructures of the alloys were studied using phase extraction, X-ray diffraction, optical and electron microscopy, and differential thermal analysis. Special emphasis was placed on the stress dependencies of the creep rates and rupture lives. The creep resistance of the alloys will be discussed in relation to TCP phase formation, and also to the volume fraction, composition, and coarsening rates of the y' phase.

2:40 p.m. DEVELOPMENT OF THE CM SX SERIES* OF SINGLE CRYSTAL ALLOYS FOR ADVANCED TECHNOLOGY TURBINE COMPONENTS: K. Harris, G. L. Erickson, R. E. Schwer, Cannon-Muskegon Corp., P. O. Box 506, Muskegon, MI 49443 A series of four complimentary single crystal alloys has been developed from the MAR-M-247 composition, with the objective of providing high strength, good castability, practical solution heat treatment ranges and stable microstructures. Alloy CM SX-l is intended for turbine vanes because of its exceptional oxidation resistance. eM SX-2 and eM SX-3 are turbine blade and vane alloys, with CM SX-3 showing improved coated oxidation and corrosion resistance. eM sx-4 is designed for "super-strength". Foundry performance characteristics studied include freckling sensitivity, spurious grain formation, microporosity, and alloy/ ceramic core reactions. Practical solution heat treatment ranges (difference between y' solvus and inCipient melting temperatures) have been established and vary from 35°F to 55°F. Optimized solution heat treated microstructures for each alloy are presented. Extensive machined-from-blade (MFB) mechanical property work and alloy stability investigations are reported. Environmental evaluations using both bare and coated single crystal specimens, subjected to separate cyclic/dynamic oxidation, and corrosion testing in burner-type rigs are also reviewed. *Patents pending

3:00' p.m. EFFECT OF COBALT ON BEHAVIOR OF EQUIAXED MM-247, J. F. RADAVICH, Purdue University, West Lafayette, Indiana 47907, J: McLaughlin, Teledyne CAE, 1330 Laskey Road, Toledo, Ohio 43612, C. S. Kortovich, TRW Inc., 23555 Euclid Avenue, Cleveland, Ohio 44117 A study was performed on the mechanical property behavior of equiaxed cast MM-247 alloy as a function of cobalt content. Tensile and SIR characterization was conducted on CTS test bars containing Co at the 10% (standard leveJ), 5%, 0% levels. Casting parameters were selected to simulate conditions obtained in blade and hub configurations. The mechanical property results as well as results of microstructural/fractographic analyses of failed test bars will be presented.

3:20 p.m.-3:40 p.m. - Break

3:40 p.m. THE EFFECT OF COBALT ON THE STRUCTURE, PHASE CHEMISTRY AND PROPERTIES OF WROUGHT Ni-BASE SUPERALLOYS*: R.N . Jarrett and J . K. Tien, Krumb School of Mi nes, Col umbi a Un i vers i ty, New York, NY 10027. We report on the progress of the research program aimed at understanding the role of cobalt in nickel-base superalloys. The three systems discussed, Waspaloy, Udimet 700 and Nimonic 115, are representative of Ni-Cr-Co-Al-Ti-Mo superalloys strengthened by a heavily alloyed matrix, coherent y ' precipirates and carbides at the grain boundaries. These alloys differ in the amount of y ' -Waspaloy with 'C 20%, Udimet 700 with 'C 45% and Nimonic 115 with 'C 55% y '. Accordingly the way cobalt (or substituting for cobalt) affects the y ' solvus temperature and the chemical partitioning in each alloy is different. Using the results obtained for the three systems a generalized understanding of the role of cobalt is discussed. Microstructure and in-situ and extracted phase STEM micro-analysis results will be used to explain cobalts effect on mechanical properties. *Research supported by NASA under grant NASA NAG 3-57.

66 TMS Fall Meeting 4:00 p.m.

3:30 p.m.

THE EFFECTS OF THERMOMECHANICAL PROCESSING ON THE STRUCTURE AND PROPERTIES OF "HASTELLOY'" ALLOY X: D. E. Fluck, M. F. Rothman, and J. J. Simmons, Cabot Corporation, 1020 W. Park Avenue,

Engineering - New Products, FGE Cars, Inc., Alexandria, VA.

Kokomo, IN 46901

The structure and mechanical property response of a nickel-base superalloy, HASTELLOY alloy X, to variations in thermomechanical processing treatment has been studied. The nature of laboratory hot-rolled and cold-rolled material structure and properties has been found to be significantly affected by prior processing path. Among some of the variables examined in the study were: roI-ling and annealing temperatures and times, total reductions taken between anneals, heating and cooling rates, and sequencing of various steps. Means for optimizing mechanical properties and structure through manipulation of thermomechanical processing procedures will be discussed.

LIGHTWEIGHT CRYOGENIC RAIL CAR, L.L. Malo, Assistant Manager,

22314 This paper outlines the development of rail car design from the wooden structures of the 1800' s through steel and into state of the art aluminlDD extrusion and polyurethane foam techniques. Significant advances have been made in environmental control methods from cooling with ice to mechanical refrigeration to the cryogenic systems now under development. These accomplishments have been made possible through finite element design methods, infrared scanning techniques and development of a unique test facility that can accurately control the internal and external temperature and hlDDidity of rail cars.

*HASTELLOY is a registered trademark of Cabot Corporation.

THURSDAY - OCTOBER 28, AM

4:20 p.m. MTCROSTRIICl1JRAL ASPECTS OF TIlE THEm-1AL TREA'IMENT OF TNCONEL ALLOY X-7S0. D.P. Ravipati, P.J. Ryan, M.G. Hehsur and J.J. Moore, ~.1incr.:ll Resources Research Center, Uni versi ty of 1-11 Tlnesota, 56 East River Road, Minneapolis, MN 55455. High strength age-hardenable Ni-Cr-Fe alloy, Inconel X-7S0, is used extensively in hath PWR and BWR as bolts, pins, spring components, etc. However, this alloy has experienced some IGSCC in service. It has been reported that the SCC resistance of Inconel X-750 can be improved by thermal treatment in the carbide preclpitation regime of 1300°F-1600°F. Therefore, the microstructural changes that occur when Inconel X-7S0 is thermally treated at l300°F, for up to 200 hours following a 20-minute anneal at l800 o r: have been studied in order to provide better understanding of the factors controlling the IGSCC resistance of thermally treated Inconel X-7S0. A wide range of grain boundary microstructures was obtained. Grain boundary chemistry and identi fication of the various carbides were conducted using scanning transmission electron microscopy and scanning Auger microscopy. Resul ts confirm the presence of grain boundary clrromiull depletion that varies in extent with time at l300°F.

STRUCTURAL MATERIALS FOR GROUND TRANSPORTATION II Sponsored by the TMS Structural Materials Committee Wednesday, October 27, 1982 Cervantes Convention Center 2:00 p.m. Room 262 Session Chairmen: K. K. Sankaran, McDonnell-Douglas Corporation, St. Louis, MO 63166; W. M. Griffith, AFWALIMLLS, Wright-Patterson Air Force Base, OH 45433.

ALLOY PHASES III TMS-AIME General Abstract Session Thursday, October 28, 1982 Cervantes Convention Center 8:30 a.m. Room 264 Session Chairman: Dr. V. K. Sethi, Institute for Mining and Minerals Research, University of Kentucky, Lexington, KY 40506.

8:30 a.m. OBSERVATION OF STRUCTURE AND MAGNETIC DOMAIN OF CoBASED AMORPHOUS ALLOYS BY FEM: S. Takayama, T. Matsuda, A. Tonomura, N. Osakabe and H. FUjiwara, Central Research Laboratory, Hitachi Ltd., Kokubunji, Tokyo 185, Japan Observation of both structure and magnetic domain of Co based amorphous alloys has been performed by a newly developed field emission electron microscope. Magnetic domain structure has been studied by both Lorentz microscopy and electron holography. Employing the latter new technique which is recently developed in Hitachi Ltd., We have succeeded in directly observing magnetic lines of force inside and/or outside of the ferromagnetic amorphous samples. Using the same electron microscope, structure of the as quenched samples and those annealed below crystallization temperature has also been investigated by a high resolution lattice image technique. It is revealed from these observations that some tiny regular atomic arrangements exist in the present as quenched amorphous structure.

2 :00 p.m.

8:50 a.m.

REVIEW OF MACHINING AND FABRICATION ISSUES RELATED TO STRUCTURAL

SllJDIES OF THE ACTIVE CORROSION OF AfoflRPHOUS FexNi (SO-x)P14B6 ALLOYS: T. P. Moffat and B. D. Lichter, Department of f.'echanical and Matenals Englneering, Vanderbilt University, Nashville,

MATERIALS FOR GROUND TRANSPORATION, K. Subramanian, International Harvester Company, 7 South 600 County Line Road, Hinsdale, Illinois

60521

Materials with high strength to weight ratio such as HSLA steels, aluminum alloys and fiber reinforced composites are finding increasing use in equipment for ground transportation. Fine grain structure, second phase dispersions such as carbides or intermetallic compounds and high strength fiber, present in these materials, while enhancing the .strength also offer inherent resistance to essential process steps such as machining, welding or joining. A review of the limitations imposed by these operations in the application of light weight high strength materials will be followed by current approaches and possible solutions. Examples from automotive, truck and equipment industries will be used to illustrate the above issues.

2:45 p.m. ALUMINUM STRUCTURAL HATERIALS FOR GROUND TRANSPORATION

APPLICATIONS, W.H. Hunt, Jr., M.E. Fogel and R.W. Rogers, Jr., Alcoa Center, PA 15069 The primary technical advantages for aluminum usage in ground transportation applications lie in its high strength to weight ratio and excellent corrosion resistance. Manufacturers have taken advantage of these traits for decades in automobiles, railroad cars, trucks, and trailers with highly satisfactory experience. This paper will concentrate on the current efforts to optimize alloy properties for a variety of applications including auto body sheet, bumpers, and structural applications as well as in railroad cars. It will stress the necessity of using a process whereby the microstructural features desired for a particular application are identified, after which alloy composition and process are developed to obtain this microstructure in the most cost-effective manner.

TN

37235

An attempt has been made to correlate the active corrosion rates with the passivity parameters of metallic glasses. The anodic polarization characteristics of several FexNi (SO-x)P14B6 alloys have been studied as a function of pH and cr concentration in H2S~-HCl mixtures. Utilizing atomic and optical ai?sorption _ spectroscopy, the concentrations of Fe++, Ni++, P0 4=, and BO/ in the electrolyte have been monitored as a function of current density, time and solution parameters. ESCA and SEM/EDS were used to establ i sh post-corrosi on surface compositi on changes, which included the formation of both pre-passive and passive films in some instances.

9:10 a.m. PREPARATION OF IRON BASED AMORPHOUS POWDERS BY GAS ATOMIZATION E. Coleman and M. L. Green: Bell Laboratories, Murray Hill, NJ 07974 Amorphous powders of Fe69Si16B15, FenP18C5, Fe40Ni40B20 and (FeO.8NiO.2)7SP16Si4CZ were prepared by ultrasonic gas atomization. Analysis of the fines (-400 fraction) of these powders show them to be spherical and partially amorphous. The Fe69Si16 B15 alloy is apparently one of the better glass formers, as practically the entire -400 fraction is amorphous. On the other hand, for the alloy (FeO 8NiO 2)78P16Si4C2 only those particles less than _20~ were amorphous: Thermal properties of these materials were measured by differential scanning calorimetry. Their crystall ization behavior will be compared with those of ribbons of the same composition prepared by the spin quenching technique. Properties related to quenching rates, particle size distribution, compositional effects and thermal stability will be discussed.

TMS Fall Meeting 67 9:30 a.m. MOSSBAUER STUDY OF SPLAT QUENCHED FeaOB20 FORMED BY SHOCK TUBE AND ROTATING DRUM: E.E~ Alp, R. Wappling*, M. Saporoschenko*, W.E. Brower, Jr., Engr. Mech. and Matl's and'Oept. of Phys. and A[t., S.I.Uo, Carbondale, IL 62901

The method of forming a metallic glass can have a strong effect on the as-quenched structure and on the subsequent crystallization kinetics. Metglas 2605, splat quenched via the rotating drum (40 wm thick, 10 6 DC/sec cooling rate) technique is compared utilizing Mossbauer Effect to this same FeSOB20 alloy remelted and splat cooled via the shock tube method (2-10 'Wm thick, 10 8 DC/sec cooling rate). The Mossbauer spectra show a difference in relative line intensities and line widths of shock tube and Metglas 2605 samples. This difference could be due to the fact that· Metglas 2605 exhibits some as-quenched partial crystallinity as indicated by XRD. In Metglas 2605, it is also observed that the orientation of magnetic moments is more planar compared to shock tube sample. Crystallization temperatures and the amoimts of crystallization were deter~ined by DTA and DSC scans, respectively. Mossbauer spectra of isothermal vacuum aged specimens were analyzed to determine both the nature of metastable phases and the crystallization kinetics.

9:50 a.m. DECOlf'DSITlON OF SULFIDES IN COAl:* B. H. Tseng, K. C. Hs ieh and C. A. Wert, Materials Research Laboratory and Department of Metallurgy and Mining Engineering, University of Illinois, Urbana, IL 61801. We have made an investigation of the effect of heating on decompos it i on of mi nera 1sin coa 1. Pri nci pa 1 attent i on has been pa i d to the su lfi des because of thei r techno 1ogi ca 1 importance PJ:evious investigations of decomposition of sulfides by Mossbauer spectroscopy have shown that the temperature range of importance is 325°C to 450°C. We also fi nd that in ou r TEM observations. Using Illinois coals 5 and 6, we have heated specimens at these temperatures in a vari ety of atmospheres. Changes in crystal type and chemical composition of the sulfides have been followed by TEM examination, electron diffraction and x-ray emmission microscopy. Models of the process of desulfurization are being developed. *Research sponsored by the Department of Energy.

10:10 a.m. THERMAL ANALYSIS OF A REFERENCE COAL: K.M. Simon and W.E. Brower, Jr., Engineering Mechanics and Materials, Southern Illinois University, Carbondale, IL 6290l.

Reproducibility of thermal analysis data for the characterization of a reference Illinois 116 coal is shown. A reference coal is established to provide a time invariant supply of coal samples with minimal sample to sample variability. Differential Thermal Analysis (DTA), Differential Scanning Calorimetry (DSC) , Differential Thermogravimetric Analysis (DTGA), and Gas Chromatography (GC) were employed as sensitive characterization techniques. The reference coal was collected, ground to -20m in argon, riffled into storage vials, and sealed in an argon atmosphere. An atmosphere of flowing air shows DTA endotherms at l15 0 C and 450 0 C + 2. 5 0 C, whereas, with helium atmosphere the same endotherms occur at l07 0 C and 437 0 C + 2.5 0 C, respectively. The magnitude of the enthalpies associated with the thermal peaks is analyzed by integrating the corresponding transitions in DSC scans. DTGA interfaced with an integrator yields a quantitative time-temperature profile of gas evolution. A Ge, also interfaced with the DTGA, yields evolved gas analysis, which is correlated to the DTA and DSC results.

10:30 a.m. SELECTION OF APPROPRIATE ALLOYING ELEMENTS FOR DEVELOPMENT OF S(11E ALUMINUM AND MAGNESIUM ALLOYS WITH PREDETERMINED PROPERTIES: M. Libennan, Steel Heddle Mfg. Co., P. O. Box 1867, Greenville, SC 29602

An experimental study of ahnninum and magnesium alloy properties was conducted to establish the relationships between properties and parameters of appropriate phase diagrams. Selection of alloying elements for magnesium alloys for general use and for aluminum alloys with high mechanical and conductive properties was done by using all elements of a periodic table for analysis. A method for determining the minimal m.nnber 0 f alloying elements to achieve optimal alloy properties is suggested. An example of alloy development with predetermined properties is shown utilizing a combination of experimental design methods and methods discussed above.

steel, were carried out. Then, an alloy with 17%-19% Ni, 12%-13% Co, 4%-5% Mo, 1.4%-1.7% Ti and iron vas developed. Magnetic properties of this alloy are characterized as follows: residual induction (Br) exceeds 1.2T, coercive force (BHc) is 4000A/m - 4400A/m. Mechanical properties are as follows: tensile strength is about 1800N/mm2 , elongation is 12%, reduction of area is about 30%. The efficiency of the high speed hysteresis lOOtor used this semi-hard magnet alloy achieved 80%.

CHEMISTRY AND PHYSICS OF RAPIDLY SOLIDIFIED MATERIALS: IV Sponsored by the TMS Chemistry and Physics of Metals Committee, TMS Corrosion and Environmental Effects Committee, and ASM-MSD Phase Transformation Activity ThUI'llday, October 28, 1982 Cervantes Convention Center 8:30 a.m. Room 273 Session Chairman: B. H. Kear, Corporate Research Science Laborato· ries, Exxon Research and Engineering Co., P.O. Box 45, Linden, NJ 07036.

8:30 a.m. CORROSION RESISTANCE OF RAPIDLY QUENCHED ALLOYS: R.M. Latanision, E.R.C. Johns, S.X. Zhang, and A. Saito, H.H. Uhlig Corrosion Laboratory, Massachusetts Institute of Technology, Cambridge, MA 02139.

Attention will be directed in this presentation to the general and localized corrosion characteristics of glassy transition metalmetalloid as well as metal-metal alloys. Likewise, consideration will be given to the chemical properties of some crystalline alloys produced by rapid solidification techniques. Not all glassy alloys are corrosion resistant, but some have remarkable inertness in contrast to alloys produced by traditional techniques. Environmentally-induced embrittlement has been observed in glassy alloys and will be discussed. Experience with rapidly quenched, crystalline aluminum and iron-base alloys suggests that these materials are not innnune to general or localized corrosion or to stress corrosion cracking but such alloys appear to be more- resistant in most cases than their counterparts produced by conventional metallurgical approaches.

9:00 a.m. POI,DER METALLURGY PRODUCTION OF NICKEL CATALYSTS, F. D. Lemkey, United Technologies Research Center, East Hartford, CT 06108 Improved Raney nickel catalysts are provided by the rapid so 1 i difi cat i on of a range of a 1umi num-ni cke 1 alloys. Alloys containing from 6.2 to 28.4% nickel may be rapidly solidified to produce a highly active Al3Ni precursor. Compositions containing from 28.4 to 42% nlckel may be rapidly solidified to produce precursors consisting of Al3Ni2 fibers surrounding the shell of AI3Ni. This latter class of materials may be heat treated to produce 100% of the active Al3Ni orecursor material.

9:30 a.m. AMORPHOUS METAL COATINGS AS CORROSION BARRIERS: R. A. Anderson, Corpora te Research, The Foxboro Co., Foxboro, MA 02035; E. A. Dobisz, J. H. Perepezko, R. E. Thomas and J. D. Wiley, Materials Science Center, University of Wisconsin-Madison, Madison, WI 53706. Rapi dl y quenched amorphous meta 1 s genera 11 y exhi bit superi or corrosion resistance in comparison with crystallized structures. The appl ication of l~m thick amorphous Ni-40 a/o Nb sputter-deposited coatings as protective barriers on polycrystalline metallic subs tra tes has been i nves ti gated. The as-depos i ted amorphous fil ms have good adhesion and show at least an order of magnitUde improvement in corrosion protection compared to polycrystalline coatings in HCl solutions. Anneal ing treatments below the glass transition temperature have been found to enhance further this protective behavior of amorphous films. Structural examinations show that when corros i on of amorphous meta 1 coated samples occurs, the initiation site is associated with surface features such as pits or hi 11 ocks whi ch were a 1 ready present in the uncoa ted substra te. Consequently, in order to achieve the full benefits of these protective coatings, careful attention must be paid to substrate preparation.

9:50 a.m. 10:50 a.m. ISOTROPIC SEMI-HARD MAGNET ALLOY (Fe-Ni-Co-Mo-Ti) WITH VERY HIGH TENSILE STRENGTH: H. Horie, T. Kobayashi, M. Asada and I.Suzuki TOSHIBA R&D Center, TOSHIBA CORPORATION

Isotropic semi-hard magnet alloy with very high tensile strength has been required for high speed hysteresis motors used in the textile industry. To obtain such properties, research on chemical compositions and production methods of alloys, based on maraging

AES and XPS Ana lys i s of Corros i on Layers Formed on Amorphous FeNiCrPB:* n.R. Baer and M.T. Thomas, Battelle Pacific Northwest Laboratori es, Ri ch l,and, WA 99352 Corrosion layers formed on Fe32Ni26Crl4Pl2B6 in air, ina Na2S04 solution and at 10-5torr oxygen pressure have been examlned by X- ray photoelectron spectroscopy (XPS) and Auger electron spectroscopy (AES) to determine the similarities and differences of films formed in different ways and to learn about the roles

68 TMS Fall Meeting of phosphorus and chromi um in fil m forma t ion. Chemi ca 1 states of the elements in films formed at different potentials in a 1/2 M Na2S04 solution (pH 5) were determined by XPS. A progress.ion of cheffilcal states was observed that could be related to elemental stability diagrams. Comparison of XPS and AES data showed that adequate analysis of corrosion films must involve some form of depth profile and an 'ideal' profile of the native oxide on the amorphous alloy has been generated using both XPS and AES data. 'This work was sponsored by'the Corporate Technical Development Division of Battelle Memorial Institute .

field recently developed by Mori and Kato. It was found that the yield stress at 0 K in this calculation is determined only by the stress field component varying along the motion of the dislocation. Moreover, similar results in dislocation shapes, activation energy and activation vo lume were also obtained when compared with the one-dimensional case. *Research supported by the U. S. Department of Energy under Contract No. DE-AC02-8lERl0942.

8:50 a.m. 10:10 a.m. THE CHARACTERIZATION OF METASTABLE Fe-Cr-C and Fe- C METAL POWDERS PRODUCED DIRECTLY FROM THE MINERAL CONCENTRATE USING NOVEL PLASMA REDUCTION TECHNOLOGY. J.J. Moore and P.J. Ryan, Mineral Resources Research Center, Uni versi ty of Minnesota. 56 East River Road, Minneapolis. MN 55455; J. M. Sivertsen. Chemical Engineering and Materials Sciences, 421 Washington Avenue S.E .• Minneapoli s , MN 55455.

Most rapidly solidified metals are produced by processing extracted an d refined metals with the necessary chemistry to provide th e required metastability on rapid quenching. Thi s pape·r de s cribes the production of metastable Fe-C and Fe-Cr-C metal powders produced directly from the mineral concentrate by a combined in-flight reduction and rapid solidification process achieved in a sustained shockwave plasma (SSP) reactor wi thin 100 millisecond s . The characterization of these me tastable powders is examined using optical, electron and Auger microscopy and the potential for producing increased degrees of metastability discussed .

10:30 a.m. INFLUENCE OF SOLID/LIQUID INTERFACE 'lYPE ON SOLUTE TRAPPING DURING RAPID FREEZING, N. W. Blake and R. W. Smith, Department of ~tallurgical Engineering, Queen's University, Kingston, Ontario. Under normal casting conditions the solid/liquid interface is often assumed to be at local equilibrium. where the chemical potentials of the atomic species are equal in the solid and the liquid phases. At high cooling rates, as are obtained in the rapid solidification processes, departures from local equilibrium. may occur and solute trapping can result. However, in most alloy systems solute trapping can be accounted for by comparing the solute concentration obtained in the solid with extnsions of the solidus and liquidus lines below or above the equilibrium eutectic or peritectic temperatures, respectively. To avoid this problem, alloys from systems with retrograde solidus curves were rapidly solidified in a melt spinning device and the extensions of solid solubility were examined.

10:50 a.m. MICROSTRUCTURE AND SOLUTE DISTRIBUTION OF HIGHLY SUPER())OLED Se-Ni ALLOYS, D. D. It:Devitt, G. J. Abbaschian, Dept. of Mlterials Science and Eng. University of Florida, Gainsville, FL 32611.

ELASTIC CRACK;

INTERACTION hTWEEN A SCREW DISLOCATION AND SURFACE N. G. Chu, Bell Laboratories, Murray Hill, NJ 07974

S~

An analytical solution for the two dimensional elastic problem of a screw dislocation near a mode III surface crack in the presence of an external stress has been derived. The dislocation-crack interaction has a negative singularity at the tip of the crack iJ;respective of the applied stresses. Therefore, in contrary to the usual expectation, dislocations can not be generated at the tip of the crack under any applied stresses, instead they have to be nucleated or generated from pre-existing sources at some distance from the tip of the crack, a conclusion in agreement with Lils result using dislocation modelling of the crack tip stress field. In the case that the applied stress is in the direction of driving a dislocation into the medium, a me.tastable position exists beyond which the dislocation will be repelled by the crack. Thus a dislocation free zone is expected in the immediate vicinity of the tip of the crack. The size of the dislocation free zone decreases with increasing applied stresses. The conditions for the generation of a single or two dislocations and dislocation dipole in front of the surface crack are examined. The effect of the dislocations on the stress intensity factor and the crack driving force are presented. The study is further generalized to the consideration of a two screw dislocation system.

9:10 a.m. ACTIVATION ENERGY FOR DISLOCATION MOTION IN A ONE-DIMENSIONAL PERIODIC FIELD:* Masaharu Kato, Department of Metallurgy, Mechanics and Materials Science, 1-1ichigan State University, East Lansing, MI 48824 Recent theory by Mori and Kato for the asymptotic form of activation energy for the double-kink formation in a dislocation in a one-dimensional periodic field is further discussed in the light . of experimental data available in literature ~ It has been found that the experimental temperature dependence of the critical resolved shear stress, eTc' can be explained very well if we use the previously obtained analyt ical form of the activation energy F* which is proportional to (1 - (Tela: )5/4, where (T. is the Peierls stress. Moreover, the analyt~cal form has been found to give reasonable values for the parameters involved in the activation process. Comparison with other theories and the further development of the present theory will also be discussed ~

* Research supported by the U. S. Department of Energy under Contrac t No. DE-AC02-8lERl0942.

9:30 a.m. The effect of supercooling on the solidification microstructure of levitation melted Fe-2St Ni samples will be discussed. STEM analysis of solute redistribution in the structure will be given. It is shown that this structure of highly super-cooled samples consists of a spherical morphology which has resulted from fragmentation of dendrites that formed during initial stages of solidification. Each spherical element contained at least one region with high solute concentration formed by partial solidification below the solidus temperature of the alloy. This region is then surrounded by a thin shell of lower solute concentration beyond which the nickel concentration increases to a maximum at the periphery of the spherical element.

DISLOCATION BEHAVIOR TMS-AIME General Abstract Session Thursday, October 28, 1982 Cervantes Convention Center 8:30 a.m. Room 261 Session Chairman: Barbara O. Hall, Westinghouse Research & Development Center, 1310 Beulah Rd., Pittsburgh, PA 15235.

8:30 a.m. DOUBLE-KINK FORHATION OF A DISLOCATION IN TWO-DIMENSIONAL PERIODIC FIELDS:* Tong C. Lee and Masaharu Kato, Department of Met all urgy, Mechanics and Materials Science, Michigan State University , Eas t Lansing, MI 4882 4 Thermally activated dislocation motion in a two-dimensional periodic fiel d, s uch as that in spinodal alloys, is discussed. The dis location sha pe, activation ene r gy and activation volume asso ci ated with the double-kink for mati on were stud ied for various two-dimensional periodic stress fields. The results were analyzed and compared with the case of a one-dimensional periodic

CROSS-SLIP FROM (Ill) ONTO (010) AND FROM (010) ONTO (111) IN Ni3Ge SINGLE CRYSTALS : H-r. Pak, Department of Metallurgy, Mechanics and Materials Science, Michigan State University, East Lansing, MI 48824. Ni)Al-type intermetallic compounds have been known to hav e the anomalous temperature dependence of the yield strength: the yield strength increases with increasing temperature, rea ching a maximum. In the present study of s ingle -crys tallin e Ni3Ge def ormed in ten sion , cro ss-slipping from (111) onto (010) was found at l ow temperatures where the stre ngth a nomaly took place. On the other hand, at high temperatures where the normal tempera ture dependence of the yield strength occurred, reverse crossslipping from (010) onto (111) wa s found. From microstruct ural observations on those two cross-slips, the origins of the anomalous temperature dependence (at low tem peratur e) and the normal temperature dependenc e (at high temperature) of the yield strength are di scussed.

9:50 a.m. SLIP SYSTEM-TENSILE AXIS ORIENTATION REIATIONSHIP IN NIOBIUM SINGLE CRYSTALS - 1. THEORY* G. W. Bang and M. Me s hii, Materials Res earch Center and Department of Materials Science and Engineering. Northwestern Un i vers ity, Evanston, I L 60201

The sl ip systems most likely to operate were determined as a function of tensile axis or ie ntation by calculating the force on screw dislocations on various slip planes. Co-planar double slip and two crite ria ap propriate for niobium Single crystal were assumed. It was fo und that the standard s ter eographic triangle could be divided into four reg i ons of differe nt s lip modes. The meaning of the criteria · and the implication of the results wi ll be exa min ed in th is ta lk . *Supported by the U. S. Department of Energy.

TMS Fall Meeting 69 10:10 a.m.

EROSION TMS-AIME General Abstract Session

SLIP SYSTEM-TENSILE AXIS ORIENTATION RELATIONSHIP IN NIOBIUM SINGLE CRYSTALS - II. COMPARISON WITH EXPERIMENTAoL RESULTS*

G. W. Bang and M. Meshii, Materials Research Center and Department of Materials Science & Engineering, Northwestern University,

Evanston, IL

Thursday, October 28, 1982

Cervantes Convention Center Room 265 Session Chairman: Dr. Hans Conrad, Head, Dept. of Materials Engi· neering, North Carolina State University, Raleigh, NC 27650.

8:30 a.m.

60201

Slip system predictions from the calculations reported in the preceding talk can provide the expected tensile yield stress dependence on axial orientation. Predictions for niobium single crystals will be compared with experimental observations. Agreement between the predicted and the observed is generally excellent. A few discrepancies with results reported in the literature will be identified and possible reasons for this speculated. The effects of solute atoms will also be examined

8:30 a.m. MECHANISM OF EROSION OF STEELS IN NON-AQUEOUS LIQUID SLURRIES* A. V. Levy, P. Tom, Materials and Molecular Research Division Lawrence Berkeley Laboratory» University of California, Berkeley,

in light of the present study.

California *Supported by the U. S. Department of Energy

94720

The direct impingement of a solid particle-liquid Slurry from a jet on flat specimens of low allay and stainless steels was used to determine the mechanism of surface degradation by erosion. Microscopic anjllysis of the surface and cross sect~ons of eroded specimens of six steel compositions used in chemical process plant piping applications was related to the test conditions used. Variations in velocity; impingement angle; particle composition, size and shape; and liquid composition, viscosity and lubricity were used in the test. It was determined that the mechanism is one of platelet formation and removal by impacting particles, very similar to the mechanism of removal in solid particle-gas erosion.

10:30 a.m. 'l'IIEORETICAL PLASTIC STRESS-STRAIN BEHAVIOUR OF A FACE-CElfTERED-CUBIC POLYCRYSTAL FOR SLIP AND TWINliiNG DEFORMATION MECHANISMS: D. L. A. Bla~hon, DO'LAB, Lyon, France A large number of metals which twin during tensile and compressive deformation exhibit a very desirable cornbinaison of high strengh and ductility. This effect is similar to the one produced by a strain-induced martensite deformation in TRIP steels. The work-hardening behaviour of f.c.d metals is influenced by twin boundaries. The efficiency of twin boundaries as obstacles to the propagation of slip between parent and twin has long been recognized.

*Work supported by Fossil Energy Office, D.O.E.

9:00 a.m.

To date, no theories have been presented to predict the theoretical plastic stress-strain behaviour of f.c.c metals for slip and twinning deformation mechanisms. The aim of this presentation is to elaborate such a theory, extended to accomodate the tension and compression case; then, to be tested to predict the simulation of.the rolling and shear texture of fac'e-centered-cubic metals.

SLURRY EROSION OF ALLOY STEELS AT ELEVATED TEMPERATURES* A. V. Levy, M. Shaw, f-1aterials and Molecular Research Division Lawrence Berkeley Laboratory, University of California, Berkeley

California

94720

The effect of temperature on the erosion of several steel alloys in solid particle-solvent slurries was determined. The alloys ranging from 1020 carbon steel through 21, 5, 9, 12% Cr ferritic steels to several austenitic stainless steels, varied somewhat in their response to the test temperature but followed the same general pattern. The effect of variations in the test fluids and the testing conditions of velocity and solids loading on the erosion process at elevated temperatures was determined. Microscopic examination of the eroded surfaces and cross sections was used to determine how the mechanism of erosion varied with temperature.

10:50 a.m. ANISOTROPIC ELASTICITY SOLUTIONS FOR STRAIGHT DISLOCATIONS AT GRAIN BOUNDARIES IN MATERIALS : D. L. A. Blachon, DO'LAB Lyon, France The effect of elastic anisotropy on the stability of straight dislocations at grain boundaries in materials has been determined. Solutions are found for the stress and displacement fields due to infinite straight dislocations for welded or freely slipping interfaces. It is found that the results are in qualitative agreement with earlier isotropiC results.

'Work supported by Fossil Energy Office, D. O. E.

9:30 a.m. 11:10 a.m. THE EFFECT OF SOLUTE CLUSTERS ON SCREW DISLOCATIONS IN B.C.C. ALLOYS,

*

D. M. Esterling, Joint Institute for Advancement of Flight Sciences, The George Washington University, NASA Langley Research Center, Hampton, Virginia 23665, R. J. Arsenault, Chemical Engineering Department, University of Maryland, College Park, Maryland 20742 The combined phenomena of solid solution strengthening and weakening

are almost unique to b. c. c. alloys. This combined phenomena can result from (self) interstitial atoms as well as substitutional atoms. A discrete b.c.c. lattice was produced and the interactions between randomly located solute atoms and a screw dislocation were determined. The activation energies for double kink and late:t;al motion of single kinks were calculated for various solute concentrations and size differences. The effect of solute atoms on the motion of a screw dislocation has been examined. The results indicate a mechanism of solid solution weakening. The classical model is not capable of explaining the experimentally observed strengthening, even when nearest and next-nearest neighbor solutes were considered.

*

Research supported by the National Science Foundation.

11:30 a.m. SENSITIVITY OF BCC AND FCC DISLOCATION CORE STRUCTURES TO CHOICE OF INTERATOMIC POTENTIAL:* D. M. Esterling, Joint Institute for Advancement of Flight Sciences, The George Washington University, NASA Langley Research Center, Hampton, Virginia 23665, R. J. Arsenault, Chemical Engineering Department, University of Maryland, College Park, Maryland 20742 The dislocation core structures in an fcc and bcc lattice were modeled via computer simulation assuming variety of interatomic potentials. Relatively minor changes in the potential lead to dramatic changes in the core structure (for bcc metals) or the stacking fault widths (for fcc metals). This sensitivity suggests some caution should be applied to the literal interpretation of many simulation results.

*Research supported by the National Science Foundation.

ELEVATED TEMPERATURE EROSION OF DUCTILE ALLOYS IN NEUTRAL GASES* A. V. Levy, J. Patterson, Materials and Molecular Research I1i vision, Lawrence Berkeley Laboratory, Uni versi ty of California,

Berkeley, California

94720

The erosion of aluminum and steel alloys at elevated temperatures to 0.8 homologous temperature was determined. 1100-0 aluminum, 1020 carbon steel and 310 stainless steel were tested. The effects of velocity and impingement angle on erosion rates were established using alumina particles of 24011m dia ave. carried in gas streams of argon and nitrogen. The rates of erosion measured were related to material properties and the mechanism of erosion. SEM analysis of the eroded surfaces and cross sections of t.he specimens were used to determine the mechanism of erosion and the presence of any corrosion products.

*Work supported by Fossil Energy Office, D.O.E.

10:00 a.m. VELOCITY DEPENDENCE OF THE EROSION RATE IN SI*: R. O. Scattergood, North Carolina State University, Raleigh, N. C. 27650 and J. L. Routbort, Argonne Nat. Lab., Argonne, IL

60439. The velOCity dependence of the solid-particle erosion rate of (111) surface Si crystals was measured for a range of particle

types (AI Z03 , SiC and Si0 2) and sizes (40 )lm-270

)lrn).

Power-

law veloclty dependence was found in all cases, but the velocity exponent n decreases with increasing particle size. Current theories predict a constant n, independent of particle size. The results obtained here suggest a general trend where n decreases as particle size increases. One possible interpretation of this effect is that the contact-loading conditions change from dynamic to quasi-static as particle size increases, and this could be due to localized heating at the impact site.

*Work supported by U. S. Dept. of Energy.

70 TMS Fall Meeting 10:30 a.m.

9:10 a.m.

CHARACTERIZATION OF SURFACE AND BULK DAMAGE OF Ti-6Al-4V ALLOY SUBJECT TO PNEUMATIC-EROSION AT MODERATE TO HIGH-TEMPERATURES, Eduardo Rosa and John Moteff, Materials Science and Metallurgical Engineering Department, University of Cincinnati, Cincinnati, OH 45221.

Experimental data correlating particle velocity (v) and erosion rate (e) for Ti-6Al-4V show a straight line type relationship on a log-log basis. It 1s interesting to note that as the temperature increases, d.lne/dlnv decreases and the best fit lines for different temperatures seem to converge to a common point. This could be indicating that a single damaging mechanism prevails at very high particle velocities. A metallurgical investigation was undertaken, emphasizing ductility aspects as gleaned from Vickers indentation data and semi quantitative analysis of surface features observed on the eroded specimens in order to characterize the most likely damaging mechanism. Metallurgioal evaluations

Were supported by the DOE-Fossil Energy Program through the Pi ttsburgh Energy Teohnology Center (Contraot DE-FG22-81PC40794). Erosion tested samples Were supplied by Professor Tabakoff of the Aerospaoe and Applied Meohanios Dept. of the Univ. of Cinoinnati.

HYDROGEN EFFECTS ON CLEAVAGE FRACTURE OF PEARLITIC lOBO STEEL: J. J. Lewandowski, A. W. Thompson, Dept. of Met. Engineering & Mat. Science, Carnegie-Mellon University, Pittsburgh, PA 15213

The effects of microstructure and precharged hydrogen on the cleavage fracture of fully pearlitic 1080 steel have been studied using notched bend bars. Samples were tested in 4-pt. bending over a range of temperatures to determine critical fracture stresses for charged and uncharged specimens. Points of crack initiation were located in unbroken bars and correlated with microstructural features. Fractography of broken bars revealed localized plastic tearing at the notch root in prechC!rged samples tes ted at room temperature. Cleavage fracture was predominant in all bars tested at lower temperatures. The effect of microstructure and hydrogen precharging on crack initiation. fractography, and critical fracture s"tress will be discussed. This work was jointly sponsored by the Fannie and John Hertz Foundation and the National Science Foundation.

9:30 a.m. 11:00 a.m. ENERGY CONCEPTS APPLIED TO PNEUMATIC EROSION OF AISI 1010 CARBON STEEL AT ROOM TEMPERATURE: J. Foulds, E. Rosa and J. Moteff, Materials Science and Metallurgical Engineering Department and V. Tabakoff, Aerospace Engineering and Applied Mechanics, University of Cincinnati, Cincinnati, Ohio 45221.

Plates of AISI 1010 carbon steel were rolled to degrees of cold work in the range 10 to 50%. Tensile specimens were then machined and tested at cross head speeds of .02 in/min and 20 in/min. Charpy impact specimens were also made to characterize energy absorbed at each condition of cold work. Pneumatic erosion specimens were then subjected to erosive wear by impinging Si0 2 particles at 600 ft/sec average velocity and at angles of attack in the range 30 to 90°. Results were then analyzed in light of a model implementing material toughness obtained from stress-strain curves, impact energy and microhardness measurements obtained on a cross section of the "eroded foils.

THE EFFECT OF HYDROGEN ON CRACK INITIATION IN BAINITIC AND MARTENSITlC STEEL MICROSTRUCTURES.* W. J. Sisak and A. W. Thompson, Dept. of Met. Engineering & Mat. Science, Carnegie-Mellon University, Pittsburgh, PA 15213

The effect of hydrogen on crack initiation in a high-strength, high-toughness steel was studied. The behavior of isothermally transformed bainitic microstructures was compared to that of quenched and tempered martensitic microstructures, using a 9 Ni4 Co steel containing about 0.35% C. Four-point bend tests were conducted on cathodically-charged samples of the two microstructures at approximately equivalent yield strengths. Crack initiation loads and origins, as well as fracture paths, were compared and an attempt was made to correlate these results to microstructural variables. TEM, X-ray diffraction and light microscopy were utilized in characterizing the microstructures. *This work was supported by the National Science Foundation.

WOJtR ou.ppoJrted by VepaJttmettt 06 EYlVLgy (VE-F6ZZ-81PC40794) and Nuny RuelVtch

06Mce

(VAAG Z9-8Z-K-OOZ9).

9:50 a.m. HYDROGEN EMBRITTLEMENT OF ZIRCALOY-2 SHEET UNDER MULTIAXIAL

FRACTURE VII: HYDROGEN EMBRITTLEMENT TMS-AIME General Abstract Session Thursday, October 28, 1982 Cervantes Convention Center 8:30 a.m. Room 263 Session Chairman: Dr. D. A. Koss, Dept. of Metallurgical Engineering, Michigan Technological University, Houghton, MI 49931.

8:30 a.m. HYDROGEN PERMEATION AND HYDROGEN ASSISTED FATIGUE CRACK GROWTH IN Fe-Ce ALLOYS; R. E. Stoltz, F. Heubaum, Corporate Research Science Laboratories, Exxon Res. & Eng. Co., Linden, NJ 07036 and A. S. Nagelberg, Sandia National Labs, Livermore, CA 94550. The effect of two levels of cerium additions to annealed polycrystalline iron, 0.03 and 0.06 wt.%, on hydrogen permeation and hydrogen fatigue resistance was studied. Permeation lag time measurements USin9 low pressure hydrogen gas as the input source were used to determine trap energies and densities. In the pure iron and the Fe1:.e alloys, the lag time data could be fit as a function of PH-I 2 with a single binding energy of 53.2 kUMol_H (12.7 kcal/mo1 2 H). Trap densities varied from - 1 x 10 cm 3 for iron to 3 x 10 15 and 4 x 10 16 cm- 3 for the 0.03 and 0.06 wt.% Ce alloys. TEM observations showed that a fine dispersion of precipitates were present in some regions of the Ce-bearing alloys. Occasionally, a recrystallizeo diSlocation substructure was associated with these precipitates. Results of fatigue crack growth rates as a function of hydrogen pressure will be presented.

8:50 a.m. THE EFFECT OF HYDROGEN ON Tl CODE 12, Tl-O.3MO-0.BNI*: N. R. Moody and S. L. Robinson, Sandia National Laboratories, Livermore, CA.

Ti code 12 is the prime candidate overpack material for nuclear waste cannisters in salt repositories. It has excellent resistance to corrosion and hydrogen uptake in NaCl solutions. However, previous tests in Na/K/Mg salt solutions indicate that hydrogen embrittlement may be a problem. A study was initiated to examine the effects of hydrogen, introduced by gas phase charging, on Ti code 12. The effects of hydrogen on both fracture modes and mechanical properties has been determined as a function of hydrogen concentration. Charging temperature has a strong influence on the precipitation of Ni-rich grain boundary particles and increased hydrogen levels altered the amount of 6 phase present. These effects will be discussed in conjunction with fracture modes and mechanical properties. *Thi5 work supported by U. S. Department of Energy. DOE, under

contract number DE-AC04-76DP007 89.

LOADING CONDITIONS: Fan Yunchang and Donald A. Koss, Dept. of Metallurgical Engineering, Michigan Technological Uni versi ty, Houghton, MI 49931 The influence of hydrogen on the ductility of Zircaloy-2 sheet deformed under multiaxial states of stress has been studied. Results from specimens containing 20, 315, and 615 ppm H indicate that hydrogen causes only a small loss of ductility in simple tension. However, the extent of hydrogen embri ttlement increases as the stress state changes from uniaxial toward biaxial tension. In balanced biaxial tension, there is a drastic loss of ductility at high hydrogen contents. In order to understand this behavior, the critical strain for hydride fracture as a function of loading path is also examined. Implications of these results with regard to the mechanism of hydrogen embrittlernent are discussed. This program is supported by the Department of Energy.

10:10 a.m. HEAT TREATING EFFECTS ON ENVIRONMENTAL HYDROGEN EMBRITTLEMENT OF NICKEL:* G. R. Caskey, Jr., Savannah River Laboratory, E. 1. du Pont de Nemours & Co., Aiken, SC 29808

Environmental hydrogen embrittlement of Nickel 200, Nickel 301, and Beryllium Nickel was evaluated by measuring fracture strain of tensile specimens in hydrogen. Hydrogen damage was comparable in all three alloys when in the annealed condition, although absolute values of ductility differed. Aging of Nickel 301 increased susceptibility to hydrogen damage. In contrast, graphitization of Nickel 200 and agin~ of Beryllium Nickel to maximum hardness had little effect on hydrogen damage.

* Work done under Contract No. DE-AC09-76SROOOOl with the U.S. Department of Energy.

10:30 a.m. AN ANALYSIS OF METHANE BUBBLE NUCLEATION IN 2-1/4 Cr-1 Mo STEELS DURING HYDROGEN ATTACK:* D. Stone, Z.J. Su, J. Wanagel, Che-Yu Li. Department of Materials Science & Engineering, Cornell Univeristy, Ithaca, NY 14853 When a carbon containing metal is exposed to high pressure hydrogen at elevated temperatures, internal damage occurs in the form of decarburization and methane bubble formation at the grain boundaries. In this work, recent data on methane bubble nucleation in 2-1/4 Cr-1 Mo steels, as a function of exposure time and conditions of temperature, hydrogen pressure and appl ied tensile stress, are analyzed. We have employed a model based on the classical nucleation theory. This nucleation theory has been found to be able to account for the data trends for the bubble number densities when the exposure conditions of hydrogen pressure and temperature are varied, but is unable to predict numeri ca lly the experi menta 1 results. An alternate formul at i on

TMS Fall Meeting 71 is proposed for the nucleation model. This new formulation can be used to incorporate various mechanisms as suggested by the experimenta 1 observations. * Research sponsored by the Department of Energy under Union Carbide Subcontract #7963.

10:50 B..m. HYDROGEN-INDUCED FRACTURE IN IRON S INGLE CRYSTALS AT 200 K* C. G. Park and M. Meshii, Materials Research Center and Department of Materials Science and Engineering, Northwpstern Univprsity, Evanston, 11 60201

The effect of hydrogen on fracture was investigated in pure iron single crystalS with cathodic charging during tensile deformation at 200 K. Crystallographic and fractographic studies were carried out in the specimens with various tensilp axis orientations including r001J and [OllJ. Hydrogen charging promoted quaSi-cleavage fracture at the expense of ductile fracture with decreas iog strain rate and increas iog current dens i ty. The quaSi-cleavage fracture surface was found to be a maximum resolved shear stress plane with significantly increaspd planarity. Annealing above 125 C restored ductile fracture again, while the hydrogen-induced hardening effect recovprs at a higher temperature. Hydrogen charging alao caused a cleavage fracture along (001) plane accompanied by twin formation in [OOlJ specimens.

FRACTURE VIII: Microstructural Effects TMS-AIME General Abstract Session Thursday, October 28, 1982 Cervantes Convention Center 8:30 a.m. Room 267 Session Chairman: Fred A. Smidt, Jr., U.S. Naval Research Lab., Washington, D.C. 20375.

8:30 B..m. THE EFFECT OF MICROSTRUCTURE ON 650°C FATIGUE CRACK GROWTH IN P/M ASTROLOY: ~ and R. V. Miner, Materials Division, NASA Lewis Research Center, Cleveland, OH 44135 The effect of microstructure on 650°C fatigue crack growth was studied in a. systematic fashion in a P/M nickel-base superalloy Astroloy. Crack growth tests were performed in air at 20 CPM using a modified compact tension specimen in which crack length was monitored with a potential drop technique. Key microstructural variables examined included grain size, distribution and size of the gal1l11a prime phase, and grain boundary morphology. The results showed significant changes in both the rate and mode of crack growth associated with variations in microstructure.

*This work was supported by the U. S. Department of Energy.

8:50 B..m. 11:10 B..m. THE EFF ECT OF HYDROGEN ON TRE FRACTURE OF ELECTRICAL GRADE IRON: *T. Gabb and T. Shelton, Materials Group, Mechanical Engineering Department, Louis iana State Uni vers i ty, Baton Rouge, LA 70803 Hydrogen-charged tens ile spec imens of Carpenter e 1ec trical grade iron were pulled at a slow strain rate until fracture. Hydrogen concentr:ation'was varied by employing different electrolitic current densities and charging times. The fracture surfaces were analyzed topographically and quantitatively with scanning and transmission electron microscopes and an SEM-integrated X-ray detector system. Although the fracture mode remained microvoid size and spacing, macroscopic fracture surface topography, and inclusion distribution were found to change with increasing hydrogen content. Microvoids were found to nucleate preferentially at specific inclusions which are believed to be the most effective hydrogen traps.

*Research sponsored by the Office of Sea Grant NOAA Grant No. NA-79AA-D-00128

11:30 B..m. FATIGUE INDUCED CHANGES IN ~1EAN SURFACE STRESS:* M.R. James and W.L. Morris, Rockwell Intl. Scienee Center, Thousand Oaks, CA. Shallow compressive residual stresses are shown by x-ray diffraction analysis to develop on the surface of aluminum alloys fatigued in tension-tension loading at stress amplitudes below the bulk yield strength. This development does not take place in fully reversed loading. These changes are a consequence of localized microplastic deformation within individual grains which can exceed 0.3% even if the subsurface rema i ns e 1as ti c. I n an A1 7075-T6 alloy the change in surface mean stress is shown to increase with alloy grain size. Effects of load ratio, environmental humidity and internal hydrogen content in Al 2219-T851 are reported. These results are rational ized in terms of the response of surface microplasticity to alloy grain size and to loading and environmental parameters. Rates of propagation of short cracks are reduced by development of a mean compressive stress. Predicted and measured rates of propagation in 7075 and 2219 are compared to demonstrate the appl ication of a model developed to relate the magnitude of microplastic deformation of the surface to growth rate. *Parba I support from ONR N00014-79-C-0334

11:50 a.m.-12:10 p.m. HYDROGEN TRAPPING IN SEVERAL Fe-Si BINARY ALLOYS - MJRPHOLOGICAL CONSIDERATIONS. Pithuk Keattipun and T. C. Shelton, Mechanical Engineering Department, Louisiana State University, Baton Rouge, Louisiana 70803. Silicon is found in the microstructure of these steels in solid solution and in the form of silicates. In this study the effect of silicon and silicate distribution on hydrogen trapping is being investigated. Several binary high purity Fe-Si alloys ranging in silicon concentration from 0 to I wtlo have been prepared. Multiple transient electrochemical permeation experiments have been carried out to study reversible and irreversible trapping. Preliminary results indicate that silicates act as irreversible hydrogen traps and that localized cracking can occur at the silicate-ferrite matrix interface. Optical and scanning electron microscopy has been llsed to characterize the alloys in terms of the grain size, number and shape of silicate inclusions and the extent of internal cracking. The results from permeation experiments will be discussed and the effect of silicates on the failure of electrical grade iron will be discussed. Sponsored by the Office of Sea Grant NOAA Grant IINA-79AA-D-00128.

A GEOMETRIC MODEL FOR FRACTURE SURFACE ROUGHNESS-INDUCED CRACK CLOSURE DURING FATIGUE CRACK GROWTII:* S. Suresh and R.O. Ritchie, Dept. of Materials Science and Mineral Engineering, University of California, and Lawrence Berkeley Laboratory, Berkeley, CA 947i.0' Mechanisms for fatigue crack closure under plane strain conditions have recently been identified at near-threshold stress intensities in terms of effects of excess corrosion deposits or fracture surface roughness in promoting premature closure of the crack. In this work, a geometric model is presented for crack closure induced by fracture surface roughness. The model specifically addresses the contribution from both Mode I and Mode II crack tip displacements in addition to considering the nature of the fracture surface morphology. Implications are discussed in light of the roles of grain size, microstructure, and crack size in influencing near-threshold fatigue behavior in engineering alloys. *Sponsored by U. S. Department of Energy under Contract AC03-76SF00098

No.DE~

9:10 a.m. aI MICROSTRUCTURAL CONTROL OF NEAR-'nIRESHOLD FATIWE CRACK GRJlmI IN 7000-SERIES ALUMINlPl ALLOYS: G. R. Yoder, L. A. Cooley and T. W. Crooker, Code 6384, Naval Research Laboratory, Washington, D. c. 20375.

Fatigue crack growth rate data for ?OOO-series aluminum alloys can be approximated with a multilinear form, when plotted in conventional logarithmic coordinates over a sufft'ciently broad spectrum of il K. Each transition point in the growth-rate curve appears to be associated with a specific microstructural feature that can serve as a ·barrier to slip-band transmission, in accord with the cyclic plastic zone model of fatigue crack growth. Evidence strongly suggests that transition to the threshold for fatigue crack growth is controlled by dispersoid particles.

9:30 B..m. THE EFFECT OF GRAIN SIZE ON SHORT CRACK GROWTH:' A.K. Zurek. M.R. James and W.L. Morris, Rockwell IntI. Science Center, Thousand Oaks, CA. The influence of grain size on growth rates of surface cracks 20-500 flm in length was studies in Al 7075-T6 prepared in 12, 40 and 130 lim grain sizes and fatigued in fully reversed loading. Interruption of short crack growth by grain boundaries leads to an observable ac.cumuladon of crack tips near the boundaries. The role of the boundaries in early growth was investigated by measuring propagation rates as a function of cyclic stress intensity for selected peak tensile stresses. For cracks with lengths comparable to the grain size, growth rate decreased with increased cyclic stress amplitude. This effect of stress amplitude is more pronounced in the 130 11m material, a consequence which is modeled by treating the effect of grain size on plastic zone size and the resulting crack closure stress. Other mechanisms which effect early growth in these materials are also discussed. *Thi-S- work was supported by Rockwell International IR&D funds.

9:50 a.m. Tim STRENGTH AND DUCTILITY OF POLYCRYSTALLINE NiAl '*

D. Barker, E.M. schulson, Thayer School of Engineering, Dartmouth College, Hanover, N. H. 03755 The B2 aluminide, NiAl, has been multiply extruded and recrystallized to produce polycrystalline aggregates of controlled grain size. Material of different grain sizes has been deformed in

72 TMS Fall Meeting tension and in co~pre~sion at temperatures from 20 to 400o~ at a strain rate of 10 /I S 1. Fracture surfaces have been examl.ned by scanning electron microscopy. At room temperature fracture under tension coincides with yielding; the failure stress obeys a HallPetch relationship; and fracture occurs through a mixture of cleavage and intergranular cracking. At 295°C, tensile ductility is obs erved in aggr egates of grains finer than about 10 l1m in diameter. The yield strength again obeys Hall-Petch behavior and fracture occurs as noted above. At 400°C, ductility is again observed, now wi thin even t!h.e coa rser-grained aggrega tes (120 llm) and increasing to over 40% elongation in the finer-grained

(_ 8 wm) material; correspondingly, the failure mode chan ges from brittle to ductile as the grain size decreases. The results will be discussed in terms of mechani sms of plastic flow and fracture.

*Sponsored by the National Aeronautical Space Administration, through Grant NAG-3-13.

10:10 a.m. MICROSTRUCTURAL EFFECT,S ON THE STRESS CORROSION CRACKING SUSCEPTIBILITY OF HIGH STRENGTH STEELS, F. Gutierrez-Solana, 1. M. Berns tein and A. W. Thompson, Dep t. of Met. Engineering & Ma t. Science

Carnegie-Mellon University, Pittsburgh, PA 15213 Stress-corrosion cracking characterization studies were done on four different high-strength steels, HY130, 300M, D6ac and H-13, in 3~ NaCl, as a function of microstructure and strength level. The microstructures studied incl\lded tempered martensite and bainite; in addition the role of retained austenite was specifically investigated. Crack growth rate tests were performed on DCB samples under constant displacement, with the crack growth rate measured dire ctly and the stress intensity being obtained from changes in the specimen compliance. The data suggests that the major effect of a changing mi crostructure, at co nstant strength level, was in the valu e of the crit ical stress intensity, where in general K ISC increased with the amount of retained austenite and for welltempered microstructures. The results obtained were correlated with scanning and transmission electron microscopy analysis to elucidate the influence of the microstructural features and composition on the stres s co rrosion process.

10:30 a.m. EFFECT OF MICROSTRUCTURE ON HYDROGEN INDUCED CRACKING OF AN HSLA STEEL: M. F. Stevens and I. M. Bernstein, Dept. of Met. Engineering

and Mat. Science, Carnegie-Mellon Uni versi ty, Pittsburgh, PA 15 213 Work has continued on the hydrogen embrittlement susceptibility of a Ti-bearing HSLA steel having a microst ructure consisting of TiC particles in an aci cular ferrite matrix, as revealed by sustained load and displacement Mode I cracking tests on ca thodically precharged and Cd-plated double can tilever beam (DCB) specimens. These tests enabled us to obtain directly changes in crack length (velocity) with stress intensity, da /dt vs. K I . Room temperature results indicate that microstructural variations in the size and distribution of the TiC particles brought about by differing tempering temperatures can alter both the threshold stress intensity and the stage II crack velocity through changes in either the cracking mode or the extent of localized plasti ci ty. These resu lts will be dis cussed in terms of plastic zone formation and work hardening properties and will be correlated to parallel fractographic and electron micrographic analysis. In addition, previous precharged smooth tensil e results will be compared to the crack growth data to help clarify the role of local stress state on hydrogen embrittlement susceptibility. Support through Office of Naval Research.

tungsten single crystals at room temperature has been measured by crack pro~~gation and arrest techniques. An upper limit of 5.2 ± loB Jm is found for G . The partition of the available elastic energy at the crack t~p for atomic bond breaking, dislocation activation, and the kinetic energy of the crack is discussed. The dislocation activity at the initial crack tip and along the path of crack propagation, assessed by chemical etching and x-ray diffraction techniques, provides a means of stress relaxation during the crack initiation and propagation process . Work supported in part by U. S. Department of Energy through Grant DE-AC02-BOER10740.

11:30 a.m. "CRACK PROPAGATION IN TUNGSTEN BI-CRYSTALS": John M. Liu and Bing-Whey Shen, flepartment of Materials Science & Englneering, State University of New York at Stony Brook, Stony Brook, New York 11794 Crack propagation experiments were performed in electron beam zone refined tungsten bi-crystals of controlled twist misorientation about [l00}. An initial crack was generated in the vicinity of the grain boundary by en electric spark discharge. Bicrystals having an angle of misorientation less than eight degrees failed crystallographically, at a toughness value substantially lower than that of single crystals. Bi-crystals having an angle of misorientation larger than eight degrees failed intergranularly when a crack existed initia21y in the grain boundary. A cohesive energy of 1.2 ± 0.4 JIm is extracted from the data for these (l00) grain boundaries of large twist misorientation. The causes of the brittleness in these bicrystals have been discussed in terms of the restriction of plastic deformation in the crack tip region due to the presence of the grain boundary. ~/ork was supported in part by U. S. Department of Energy through Grant DE-AC02-BOER10740 . 11:50 a.m.-12:10 p.m. THERMAL FATIGUE RESISTANCE OF IRON BASE OXIDE DISPERSION STRENGTHENED ALLOY SHEET MATERIAL: O. Prabhakar and J. Unnam,

Materials Engineering Department, Virginia Polytechnic lnst 1. tute and State University, Blacksburg, VA 24061. Thermal fatigue resistance of iron base oxide dispersion strengthened (ODS) alloy sheet was compared with two ODS-nickel base alloys and one nickel-base superalloy. The number of thermal cycles for crack initiation in short cycles (total cycle time: 2 min) for ODS-Fe base alloys was found to be 340 compared with 3720 and 4850 for two ODS-Ni base alloys thus confirming the poor

thermal fatigue resistance of ODS-Fe base alloys. On the other hand, long cycle (total cycle time: 40 min) experiments showed that the number of cycles for crack nucleation for ODS-Fe base and ODS-N1 base alloys are of the same order of ma.gnitude. This observed difference in the behavior of ODS-Fe base alloy when cycle duration is altered is attributed to precipitation and dissolution of second phase particles.

GRAIN BOUNDARY DIFFUSION AND KINETICS II: Diffusion-Induced Grain Boundary Migration Sponsored by the ASM-MSD Atomic Transport Activity Thursday, October 28, 1982 Cervantes Convention Center 8:30 a.m. Room 272 Session Chairman: J. R. Manning, Metallurgy Division, National Bureau of Standards, Washington, DC 20234.

10:50 a.m. FRACTURE OF ANi-BASE SUPERALLOY SINGLE CRYSTALS:

D. L. Anton,

United Technologies Research Center, East Hartford, CT

06108

8:30 a.m. DIFFUSION-INDUCED GRAIN BOUNDARY MIGRATION IN COPPER SYSTEMS,

Single crystal specimens of a model Ni-Base superalloy were machined so as to propagate a c rack in two directions through the lattice. Instrumented impact testing was accomplished after which SEM. fractography was performed to evaluate ..he temperature and directional ef fects upon load to failure, fracture energy and crack propagation path. A method for c haracterizing a classical plain crack in a lattice was developed and used to evaluate the fractographic and instrllmented impact results. Cracks were observed to follow {Ill} crysta llographic planes primarily· with the local st ress state dicta ting crack path topography . On a macroscopic scale, brittle cleavage appeared to have occurred while micros co pic ductile tearing was observed. Temperature was fo und to greatly affect the local stress state with elevated temperatures increaSing the plane stress region of the fracture surface.

11:10 a.m.

Diffusion-induced grain boundary migration (DIGM) is a recently recognized phenomenon in which the application of a diffusion gradient to a polycrystalline specimen causes migration of grain boundaries. Rapid transport along these migrating boundaries can produce significant composition changes in regions through which the boundaries have passed. This effect occurs in many alloy systems below half the melting temperature, wher e normal lattice diffusion is negligible. Experimental observations of DIGM are reported for Cu-Zn and Cu-As systems where alloyed regions were formed by dif fusion into pure copper of Zn or As fro m the vapor. Electron microprobe area scans showed that only regions traversed by the migrating grain boundaries became alloyed and sharp concentration steps appeared at initial and final boundary positions. Comparisons are made with observations in other metallic systems and the results are discussed in terms of proposed mechanisms. *Research supported by the Army Research Office

"DYNAmC FRACTURE RESISTANCE AND CRACK ARREST IN TUNGSTEN SINGLE CRYSTALS". John M. Liu and Bing-Whey Shen, Department of Materials Science & Engineering, State University of New York at Stony Brook, Stony Brook, N. Y. 11794 The resistance to dynamic crack propagation,

D. B. Butrymowicz, J. W. Cahn and J. R. Hanning, National Bureau of Standards, Washington, DC 20234

G~,

on {loa} in

9:05 a.m. THE MICROSTRUCTURE OF DIFFUSION ZONES AS RESU LTING FROM BOUNDARY MIGRATION AND NEW RELATED PHENOMENA.

F.J.A. den Broeder , Philips Research Laboratories, 5600 MD,

TMS Fall Meeting 73 Eindhoven, The Netherlands, and S. Nakahara, Bell Laboratories, Murray Hill, N.J. 07974. New examples of diffusion couples exhibiting diffusion-induced grain boundary migration (DIGM) were found to be Cu-Ni, Au-Ni and AI-Cu. For these systems the occurrence of DIGM and its characteristics, as boundary velocity and composition of alloyed zone, were investigated as a function of diffusion parameters. In particular for Cu-Ni couples, the microstructures in and aroWld alloyed zones, the shapes and structures of migrated b01U1daries and the dislocation structures ahead and behind these bOlmdaries, were studied by transmission electron microscopy. It is found that DIGM may give rise to several new phenomena. Experiments underway to possibly ob:;erve DIGM L1 some non-metallic systems will be reported as well. The results of this study, in particular the explanations of the various new microstructural effects, will be used to critically examine and modify existing theories of the mechanism and the driving for)ce of DIGM.

9:40 a.m. EXPERIMENTAL INFORMATION RELATING TO THE DRIVING FORCE OF DIGf1.

Mats Hi 11ert, Div. Physical Metallurgy, Royal Technology, 5-100 44 Stockholm, Sweden. The experimental information attempt to define the nature various kinds of information range where DIG~1 occurs, the termination and reversion of discussed.

Institute of

on DIGM will be reviewed in an of the driving force. Among the to be discussed are the temperature migration rate and the initiation, DIGH. Related phenomena will also be

10:15 a.m. AN EXAMINAT ION OF THE NATURE OF "CHEMICAL FORCES" FOR GRAIN BOUNDARY MIGRATION K. Tashiro and G. R. Purdy Department of Metallurgy and Materials Science, McMaster UniverSity, Hamilton, Ontario, Canada. Several types of general i zed "forces" for grain boundary migration are considered, and an attempt made to define an appropriate local potential for each of the different cases. The dissipative processes within chemically driven boundaries are evaluated on the basis of several simple model s, and the results compared with experiment. It is shown that more than one model can account for the observed coupling of diffusive flux and boundary migration.

10:50 a.m. THE EFFECT OF INTERNAL STRESS AND TIME ON THE DEVELOPMENT OF DDGM LAYERS P. G. Shewmon and Yu, Zong-Sen, The Ohio State University. Department of Metallurgical Engineering, Columbus. Ohio 43210. A DDGM reaction allows the development of a thin alloyed layer over the entire surface of a metal in a relatively short period of time. In the Fe-Zn sys tern an entire 40 llm thick Fe foil can be transformed to alloy layer in a day while in a 1 em thick Fe piece under the same conditions no reaction is found on the flat surfaces in over a week. This i~interpreted as an indication that the boundary diffusion coefficient of Zn, Din' is much greater than that for iron. DFe . To study this matter further 120 llm thick iron sheets were reacted at 600°C. Here nucleation was slow but follOWing an incubation period a shallow layer formed over the entire sheet in about twelve hours. This layer then increased in thickness at a rate roughly linear with time. There is evidence of lateral expansion in the sheet which fits the assumption of Din »DFe' However, the mechanism of atomic transport that leads to the thickening of the alloyed layer is unclear.

LIQUID AND SOLID METAL INDUCED EMBRIT· TLEMENT OF METALS VI Sponsored by the TMS Corrosion and Environmental Effects Committee and the ASM·MSD Corrosion and Oxidation Activity Thursday, October 28, 1982 Cervantes Convention Center 8:30 a.m. Room 270 Session Chairman: R. P. Gangloff, Exxon Research and Engineering Company, CR-SL, Linden, NJ 07036.

8:30 a.m. EMBRITTLEMENT OF ENGINEERING MATERIALS DURING HIGH-TEMPERATURE FATIGUE IN A LIQUID-LITHIUM ENVIRONMENT: D. L. Hammon, D. K. Matlock, and D. L. Olson, Department of Metallurgical Engineering~ Colorado School of Mines, Golden, CO 80401 The effects of a liquid-lithium environment on fatigue crack propagation in Types 304 and 304L stainless steel and a 2~Cr­ IMo steel are reviewed. Crack growth rates in a lithium environment were found to be higher than in an argon environment from 473 to 973K. Depending on the test conditions, the enhanced

growth rates were attributed to lithium corrosion attack at the grain boundaries or to liquid metal embrittlement (LME). At high temperatures, crack propagation was enhanced by lithium corrosion which promoted intergranular fracture in 304L stainless steel. The appearance of large, cleavage-type facets in the 2~Cr-lMo steel and sharp, apparently crystallographic steps in the stainless steels suggests that crack propagation was enhanced by LME mechanisms at lower temperatures. A strainrate-induced embrittlement model was proposed to explain the promotion of embrittlement by high loading frequencies at temperatures above those normally associated with lME.

3:50 a.m. LIQUID METAL EHBRITTLEMENT OF AISI 316 STAINLESS STEEL BY Te-Cs MIXTURES: M.G. Adamson & W.H. Reineking, General Electric Co., Vallecitos Nuclear Ctr, Pleasant CA 94566 & S.Vaidyanathan & T.A. Lauritzen, GE Co, Sunnyvale, CA 94086 From an extended series of laboratory screening tests utilizing a novel controlledenvironment ring crush technique, evidence has been obtained that 20% cold-worked AISI 316 stainless steel, currently the preferred fast reactor fuel is embrittled under certain conditions by simulated fission produ~t mixtures containing both tellurium and cesium. The embri ttlement process is shown to exhibit many of the characteristics of 'classical t LHE, and the embrittling agents - Te and Cs - appear to function synergistically in the temperature range 450 to 700·C. Results that will be presented and discussed in detail include the effects of temperature, coating/wetting conditions, chemical variables (Cs:Te ratio, oxygen activity,etc), fsst neutron irradiation, and alloy thermomechanical condition, on the embrittlement phenomenon. The laboratory results will also be discussed in the context of observed cladding embrittlement in certain irradiated fuel pin tests and post-irradiation transient loading tests on (fuelled) cladding specimens.

9:15 a.m. EMBRITTLEMENT EFFECTS OF LIQUID ALKALI METALS ON STRUCTURAL MATERIALS:* K. Natesan, Materials Science Division, Argonne National Laboratory, Argonne, Illinois 60439 Interactions between alkali metals and structural materials are o£ considerable interest in view of the role of liquid alkali metals as heat-transport media in a number of advanced energy generation concepts. In general. exposure to alkali metals per se does not result in embrittlement of the structural metals, but the transfer or partitioning of interstitial elements such as oxygen, carbon, and nitrogen between the liquid and the solid can lead to significant degradation of the mechanical properties of the structural materials. The present paper will examine the partitioning of several interstitials between Group IA liquid alkali metals and a number of structural materials such as austenitic stainless steels and refractory metals/alloys, with emphasis on the mechanical properties of the materials.

*Work supported by the U. S. Department of Energy.

9:35 a.m. Internal Friction and Grain Boundary Sliding of Aluminum wetted with Liqui,d Gallium, Hiroyuki ICHINOSE* and Takahiro TA,KASHIMA**, *Technical Research Center, Nippon Kokan K.K., 1 Kokan-cho, Fukuyama 721, Japan, **Central Research & Development Laboratory, Kobe Steel, Ltd., 1 Wakinohama-cho, Fukiai-ku, Kobe, Japan Internal friction measurements were carried out on polycrystalline aluminum specimen wetted with gallium during heating. Two peaks were observed at 100°C and 360°C, respectively. The high temperature peak is due to the grain boundary sliding in aluminum, as reported in literature. The low temperature peak is apparen·t one. As grain, boundary diffusion proceeds during heating, grain boundary sliding contributes to internal friction, hence the latter value increases. Above lOO°C, however, vacancy motion in aluminum is activated and gallium atoms in grain boundary diffuses into aluminum matrix. Thus the internal friction value decreases, leading to the apparent peak. Holding the specimen at a temperature between 100°C and l40°C, the decrease in internal friction value is observed. From the results, the activation enery of gallium for volume diffusion is estimated to be 18.4 Kcal/mol.

9:55 a.m. Grain Boundary Diffusion and Static Fatigue of Aluminum by Liquid Gallium, Hiroyuki ICHINOSE* and Takahiro TAKASHIMA**. *Technical Research Center, Nippon Kokan K.K., 1 Kokan-cho, Fukuyama 721, Japan, **Central Research & Development Laboratory, Kobe Steel, Ltd., 1 Wakinohama-cho, Fukiai-ku, Kobe, Japan Diffusion of gallium into grain boundaries of randomly oriented aluminum bi-crystals was examined at 32°, 42° and S2°C. Diffusion rate became faster as the misorientation in tilt part increased, but did not dependent on the total misorientation between [100] directions or the misorientation in twist part. The obtained activation energy was 8 to 10 Kcal/mol. To confirm this observation, further experiments were carried out on tilt and twist grain boundaries. On tilt boundaries, diffusion rate increased as the tilt angle increased, and the activation energy was again 8 to 10 Kcal/mol. Contrary to this, diffusion was very slow to twist boundaries. Assuming that gallium diffusing zone

74 TMS Fall Meeting is acting as a crack in static fatigue test, the time to failure is related to the stress in the form of t 'V 0-'+ . Thi s relation holds, if the misorientation in tilt part is not small.

10 : 15 a.m . MEASUREMENT OF TOPOLOGY AND SIZE DISTRIBUTION OF GRAINS IN ALUMINUM BY PENETRATION WITH LIQUID GALLIUM : F. N. Rhines and A. B. Gokhale, Department of Materials Science and Engineering, University of Florida, Gainesville, Florida, 32611.

A polycrystalline aggregate, when separated into its component grains, yields important statistical and topological information about the structure which can be correlated with physical properties. Grains are classified by weight and their indiv i dual numbers of facets, edges and corners can be determined by d i rect observation. Polycrystalline aluminum can be s eparated into its component grains thruugh grain boundary penetration by liquid gallium. A method is Clutlined in detail to carry out such separation. The measurements are reproduCible and the error related to the interaction of gall i um with aluminum is found to be small.

9:10 a.m. THE PRODUCTION OF RAPIDLY SOLIDIFIED Ti ALLOY POWDERS BY LASER SPIN ATOMIZATION:* D.G. Konitzer, M.J. Kaufman and H.L. Fraser, Dept. of Metallurgy and Mln1ng and the Materials Research Laboratory, Uni vers i ty of III i noi s, Urbana, IL 61801 The Laser Spin Atomization Device has been designed and constructed for production of R.S. powders. A high power Laser 8eam is directed onto the free end of a consumable alloy ingot which is rotated at - 35,000 rpm. Liquid droplets are forced from the molten end of the ingot, and are convectively cooled by jets of He gas. Powders produced include Ti-6Al-4V, Ti3A.l, and TiAl. The collected powders are typically in thesrange 50-300 I'm and ca 1cu 1at ion predi cts cool i ng rates of - 10 °K/s for thi s material. The powders have been characterized using scanning Auger electron spectroscopy coupled with Ar ion sputtering, optical metallography, scanning electron microscopy and analytic transmission electron microscopy. microstructural modifications are effected through rapid solidification and these will be discussed with reference to heat flow models. *Thi s work was supported by the US DOE under contract DE AC02 ER01198.0

9:30 a.m. 10:35 a.m . THE HARDNESS OF TIN ALLOYED ~JTH TRACES OF ZINC AS RELATED TO THE SURFACE TENSION OF THE METAL IN THE LIQUID STATE : Abol-Hassan K. Abdel-Aziz, Professor of Metallurgy, Chemical Engineering Dept., Alexandria University, Egypt, and A. r~. Aref, previously postgraduate student at Alexandria University, Egypt. The effect of addino traces of zinc to tin on the surface tension of the latter in the liquid state and on the hardness and the grain size in the solid state was experimentally studied. The surface tension and the hardness increase, while the grain size decreases by rise of the zinc content. The results are explained by the theories of liquid and solid structures.

MELTING AND SOLIDIFICATION TMS-AIME General Abstract Session Thursday, October 28, 1982 Cervantes Convention Center 8:30 a.m. Room 262 Session Chairman: R. H. Hopkins, Westinghouse Research & Develop· ment Center, 1310 Beulah Rd. , Pittsburgh, PA 15235.

8:30 a.m. IN-SITU RAPID SOLIDIFICATION STUOIES OF AN Al-Ge EUTECTIC ALLOY:* M.J. Kaufman and H.L. Fraser, Department of Metallurgy and Mi ni ng and the Materi a 1s Resea rch La boratory, Un i vers i ty of Illlinois, Urbana, IL 61801 Submicron powders of an Al-Ge eutectic alloy have been produced by el ectrohydrodynami c atomi zati on**. After bei ng introduced into an electron microscope, the electron beam has been used as a local heat source, and powders have been both annealed, yielding a two phase equilibrium microstructure, and melted. The molted droplets can then be rapidly solidified by introducing an opaque stop into the beam path. The resultant microstructure of the resolidifed powders is mixed, consisting of a crystalline cellular phase based on Al and an amorphous phase which is apparently rich in Ge. These materials have been studied using both x-ray microchemical analysis and convergent beam electron diffraction, and the results of this work, will be discussed with reference to current heat flow models. *Thi s work was supported by the US DOE under contract DE AC02 76ER01l98. **The powders were kindly supplied by Phrasor Scientific Inc.

8:50 a.m. THE STRUCTURE OF FINE TiC-ZrC POWDERS PRODUCED BY SPARK MACHINING OF METALLIC ALLOYS: C. S. Pande and J. D. Ayers, Physical Metallurgy Branc h, Naval Researc h Laboratory, Washington, DC

20375 Fine spherical powders (size 1-40 ~m) of TiC, ZrC and TiC-ZrC a lloys were prepared from the elemental metals and their alloys by spark mach i ning them in kerosene, and analyzed using x-ray diffraction, SEM and TEM.

It was found that some carbide powder particles contained sub-micrometer s cale carbides dendrites in a metal matrix, but under appropriately c hosen processing conditions the powder is f ound to be a single phase carb i de with a carbon to metal ratio of unity. Under these conditions the rapid quenching during spark machining is able to produce a single carbide phase in a lloys which would exhi bi t two c arbide phases (both wi th rocksalt structure but with different lattice parameters) if they were produced by slow cooling. Further, heat treating these metastable alloys generated fine two phase dispersion within the c a rbide powder.

LASER SURFACE MELTING OF A Ti-Al-Y ALLOY:* D.G. Konitzer and H.L. Fraser, Department of Metallurgy and Mining and the Materials Research Laboratory, University of Illinois, Urbana, IL 61801 A Ti-8Al-4Y alloy has been arc-melted, and subsequently prepared for laser surface melting. Surface melting was achieved by traversing mounted specimens at - IDem/sec under a 2 kw laser beam. The microstructures of the as-cast and surface melted regi ons ha ve been cha racteri zed by opt i ca 1 metallography, scanni ng electron microscopy and analytical transmission electron microscopy (i nc 1udi ng x- ray mi crochemi ca 1 ana lys is and convergent beam electron diffraction). The as-cast material consists of three phases, name ly the a-Ti mat ri x, an i nterdendri tic prec i pi tate and a finely dispersed phase (typically 50 nm in diameter). 80th of these latter phases have been identified as being Y20 3 , this being established using a combination of the analytical transmission electron microscopy techniques. The effect of rapid solidification following laser surface melting is to cause a refinement of both the dendritic structure and the finely dispersed phases. *Thi s research has been supported by the US DOE under contract DE AC02 ER01l98.

9:50 a.m. Effect of Process Parameters on Solute Distribution During Laser Surface Alloying:* T. Chande, and J, Mazumder. Dept , of ~'ech, Eng, Univ. of Illinois at Urbana-Champalgn, Urbana, IL 61801 The effect of process parameters on solute distribution during laser surface alloying of nickel on steel (AISI1020) is being investigated using a 23 factorial experimental design technique. The experiments have been carried out using a 10KW CI, C02 laser. Nickel distribution has b~en experimentally determined by electron probe micro analysis (EPMA) technique. An attempt has been made to establish quantitative relationship between the process parameters and nickel distribution. Many features of the mechanism of the 1aser surface all oyi ng have been exp 1a i ned wi th the help of a dimensionless parameter (LASAP) developed during this study. A theoretical study of the associated transport phenomenon has also been carried out. It was found that effective mass transfer coefficient during laser surface alloying could be ten order of magnitude higher than atomic diffusion coefficient for 1 iquid metal. *The research is supported by American Iron and Steel Institute.

10:10 a.m. USE OF SOLUBILITY DATA TO PREDICT THE EFFECTS OF AL AND N CONTENTS ON THE SUSCEPTIBILITY OF STEEL CASTINGS TO INTERGRANULAR EMBRlTTLEMENT. By N.H.Croft, Dept. of Materials Science and Mineral Engineering, University of California, Berkeley , Ca 94720

The phenomenon of lntergranular IIRock Candy" Frac ture, (I. F.) , in aluminum-killed low alloy steel castings results from the precipitation of aluminum nitride on the as-cast grain bounda ries. Although the morphology of the AlN precipitates and associated fracture mechanisms are well understood, inst;ances of I.F continue to occur and are responsible for the rejection of l a r ge tonnages of steel due to either cracking in the mould or premature failure in service. This paper describes a model based on solubility data for AlN which examines the relative import ance of Al and N contents in determining the amount of AlN available for grain boundary precipitation. Since Al is usually added in stoichiometric excess of N, it is demonstrated that variations in the Al content do not significantly affect the amounts o f AIN formed. The sources of Al and N in steel c a stings will be considered, and the effects o f deoxidation pra ctice, macro- a nd micro-segrega tion will be discussed.

10:30 a.m. DENDRITE REMELTING-COOLING OF A SUPERHEATED MELT J. Lipto n, Dr. W. Heinemann , ProL W. Kurz

The prima ry macrostructur e o f s o lid metals, particularl y t he columnar to eq uiaxed tran s itio n is predomina n tly influenced by

TMS Fall Meeting 75 level of superheating in the melte Therefore the way, how superheat is released during the solidification may be considered as one of the key elements in the field of solidification research. Calculation of heat transfer through the thermal boundary layer have shown, that the predicted cooling rates are generally too low compaired with experimental results. This discrepancy between measured and calculated cooling rates can be explained by remelting of free crystals. An experimental set-up was designed to prove this hypothesis, using organic liquid instead of metal for the study of solidification. Results from these experiments and calculation show that the presented model of particle remelting is sui ted to explain differences in solidification structure of metals cast with different superheats.

10:50 a.m. SOLUTE REDISTRIBUTION ANALYSIS IN UNIDIRECTIONAL SOLIDIFICATION: M. Ba!?aran. Revere Research, Inc. J Edison, NJ 08837

Solute redistribution in directional solidification of binary alloys was analyzed by Ficks' second law which was solved implicitly by the finite difference numerical method. The analysis shows the dependence of solute redistribution, the amount of eutectic and secondary phases. homogenization and solution treatments on local solidification time, dendrite coarsening and alloy composition. The developed program can be used for all binary alloys by choosing the appropriate alloy characteristics and the type of operating dendrite coarsening.

11:10 a.m. SOLIDIFICATION INTERFACE STRUCTURE IN ALUMINUM AND ALUMINUM ALLOYS WITH AND WITHOUT ELECTROMAGNETIC STIRRING:* D. C. Kunerth and J. P. Wallacet, Department of Chemical Engineering and Materials Science, University of Minnesota. Minneapolis, MN 55455

In situ eddy current measurements were made to detennine the difference between solidification interfaces with and without electromagnetic stirring of the liquid. Materials examined include aluminum, aluminum-copper, and aluminum silicon. Data collected correlates time, temperature, and the eddy current response to the physical conditions within the melt. Observations indicate measurable differences in structure and thermal ;;radients between stirred and unstirred conditions. Thermal .mdercooling is also measured.

~Research

supported by Magnetic Analysis Corporation; Mount rernon, New York. 'Present address: Casting Analysis, Inc.; Route 1, Box 16; Fort Defiance, VA 24437.

stress intensity factor before crack propagation occurs. For distinction, this stress intensitf range is called fatigue tolerance range and denoted by t.K _ Experimental results will be presented to support this concept of a fatigue tolerance range. Usi ng thi s concept of t.KT, one can measure the openi ng stress intensity factor directly. Measurement techniques and preliminary results will be described.

9:00 a.m. AllALYSIS OF SEQUENCE EFFECTS UNDER VARIABLE AMPLITUDE LOADING ON THE BASIS OF RECENT CRACK PROPAGATION MODELS: H. Nowack, R. Marissen and K.H. Trautmann, OFVLR, Institute fUr WerkstoffForschung, Fatigue Branch, 0-5000 Koln 90. W. Germany. The fatigue crack propagation under variable amplitude loading is si gnifi cant 1y i nfl uenced by load sequence effects. The hi story dependent nonlinear damage accumulation in the crack stage has to be accounted for in fatigue crack growth prediction methods. The causes of sequence effects are mechanical and microstructural. The predominant mechanical causes are known from experimental and continuum mechanics analyses of the stress strain field around cracks. Because of the high expenses of such cycle by cycle analyses various simplified prediction methods have been proposed. From investigations where the microstructural crack propagation mechanism and the fracture surface morphology were varied by the thermomechanical treatment of the material further essential microstructural causes were identified. The signifjcance of the individual causes can be evaluated on the basis of simple lowhigh-low loading histories, where compressive peak loads are included. The prediction capability of presently available crack growth models is discussed.

9:30 a.m. MICROSTRUCTURAL CONTRIBUTIONS TO THE FATIGUE CRACK PROPAGATION BEHAVIOR UNDER VARIABLE AMPLITUDE LOADING CONDITIONS: K_ Schulte, H. Nowack, DFVLR, Inst. f. Werkstoff-Forschung, 5000 Koln 90, w.Germany; G. Lutjering, Techn. Hochschule Hamburg-Harburg, 2000 Hamburg, W.-Germany.

During fatigue crack propagation under service loads, pronounced sequence effects, can be observed. However, up to now only little knowledge exists how far this observed behavior is influenced by the microstructure of a material and the resulting microstructural fatigue crack propagation mechanisms. A test series is reviewed in which with different aluminum alloys and heat treatments it was possible to vary the types of fatigue crack propagation mechanisms. Additional attention was given to the influence of the environment on the fatigue crack propagation behavior. Tests were conducted in vacuum and, as an example of aggressive environments, laboratory air and a 3.5% NaCl-solution. An attempt will be made to present an understanding of the observed behavior by combining continuum mechanics descriptions with the microstructural properties of the aluminum alloys.

11:30 a.m. A MODEL OF THE HEAT AND FLUID FLOW IN RAPID SOLIDIFICATION OF METALLIC GLASSES,* S.C. Megli, D.L. Eddingfield, W.E. Brower, Jr.

Small amounts of surface crystallinity, detected by TEM diffraction, have been shown to enhance the catalytic activity of amorphous splat cooled (shock tube) PdSOSi20' Glassy splat cooled (rotating drum) FeSOB20 has been shown by XRD to be consistently top surface crystallized. The location and amount of these crystalline regions is governed by the local cooling rates encountered during the splat cooling process. A mathematical model of the heat and fluid flow during splat cooling in a hammer and anvil apparatus is solved. The model uses a "squeeze film" pressure variation within the sample, and allows for nonNewtonian heat flow from the sample. The sample is considered to become locally rigid when the viscosity corresponds to the glass transition temperature (10 14 poise). The results of the model are compared to those of Kroeger and co-workers, who assumed Newtonian cooling. Our non-Newtonian model is extended to the shock tube and rotating drum splat cooling techniques,

~b~~~ ~~6g;gr~aO~y th~s~oS~~~rt~~gt e~?i~~~r~~~

crystallization .

MULTIPLE LOADING EFFECTS ON FATIGUE II Sponsored by the ASM Flow and Fracture Activity Thursday, October 28, 1982 Cervantes Convention Center 8:30 a.m. Room 266 Session Chairman: G. Lutjering, Techn. Hochschule Hamburg-Harburg, 2000 Hamburg, W. Germany.

10:00 a.m. EFFECTS OF OVERLOADING ON THE PLASTIC ZONE PROPAGATION IN IMPRESSION FATIGUE, S.N.G.Chu, Bell Laboratories, Murray Hill, N.J. 07974 The overloading effects in fatigue crack propagation is important in aircraft design considerations. There has been controversial as to whether the overloading effects are due to the crack closure or they are the dynamic behavior of the plastic zone associated with the tip of the crack. A test called "Impression Fatigue" was designed to study the dynamic behavior of the plastic Zone propagation under cyclic load. In this test, a cylindrical punch with a flat end is pressed into the material from the surfac'e under cyclic load. The punch creates a plastic zone which propagates into the material usually without a crack. Yet, similar overloading effects in the plastic Zone propagation was observed. The results suggest that the load interaction effects in fatigue crack propagation were due to plastic zone properties rather than crack closure.

RADIATION DAMAGE ANALYSIS FOR FUSION REACTORS V. Correlation Models and Mechanical Properties (continued) Sponsored by the TMSIASM Nuclear Metallurgy Committee and ASM Electrical, Magnetic, and Optical Phenomena Activity

THRESHOLD RANGE AND OPENING STRESS INTENSITY FACTOR IN FATIGUE: G. Marci and H. Doker, DFVLR, Institut fUr Werkstoff-Forschung, Ow. of Fracture Mechanics and Mechanical Behavior, 0-5000 Kol n 90

Thursday, October 28, 1982 Cervantes Convention Center 8:30 a.m. Room 271 Session Chairmen: R. H. Jones, Pacific Northwest Lab., P.O. Box 999, Richland, WA 99352; W. G. Wolfer, Nuclear Engineering Department, University of Wisconsin, 1500 West Johnson Drive, Madison, WI 53706.

The threshold range t.Kth in fatigue, as usually measured, is depending on the mean load, i.e-. the R-ratio. The use of such a threshold range in design of fatigue loaded structures is, at 1east, questionabl e. Prior work by the authors defi ned a threshold range which is independent of the R-ration, namely, a stress intensity range which has to be surpassed above the opening

YIELD STRENGTH-MICROHARDNESS COMPARISONS IN 14 MeV NEUTRON IRRADIATED METALS:' E. R. Bradley and R. H. Jones, Pacific Northwest Laboratory, Richland, Washington 99352

8:30 a..m.

8:30 a.m.

76 TMS Fall Meeting The fluence dependence of the yield strength and microhardness in vanadium and titanium have been compared. A consistent correlation between microhardness and yield strength was not obtained. Foil and wire specimens irradiated at ambient temperature in the RTNS II facility to five fluence levels; ranging from 1 to 20xl0 21 m- 2 were evaluated. The observed differences between yie ld strength and microhardness can be attributed to the effects of neutro n irradiation on strain hardening which strongly influences microhardness but not yield strength. The present results suggest meaningful yield strength-microhardness correlations may be limited to specific materials and irradiation conditions. Furthermore, th e effect of irradiation on strain hardening needs to be considered to correctly interpret and utilize correlations between yield strength and microhardness. *Research sponsored by the U. S. Department of Energy under contract DE-AC06-76RLO-1830

Using a continuous-slowing-down, random amorphous material model, we have studi ed displ acement cascades in a number of diatomic materials. The results presented will elucidate the effects of atomic mass, recoil energy, displacement energy, capture energy and material stoichiometry on the numbers of vacancies and interstitials produced in displacement cascades. Material s for which the ratio of the heavy to 1 ight mass is 1ess than about four show somewhat different displ acement cascade characteristics than those for which the mass ratio is greater than four. An understanding of dtspl acement cascades in polyatomic material s is important to studying radiation effects in ceramic and insulator materials projected for use in fusion reactors and has appl ication to engineering alloys. *Work supported by the U.S. Department of Energy .

10:15 a.m. 8:50 a.m. A CORRELATION OF HARDNESS AND YIELD STRENGTH FOR FUSION REACTOR MATERIALS:* N. F. Panayotou and H. L. Heinisch, Westinghouse Hanford Company, Richland, WA 99352

The correlation of simply measured properties with properties of engineering interest is an objective of the fusion materials development program. A simple correlation between the change in hardness and the change in 0.2 percent offset yield strength as a function of fluence, previously reported for copper and annealed type 316 stainless steel irradiated in RTNS-II. has been observed for irradiated nickel and 20 percent cold-worked type 316 stainless steel as well. All specimens were irradiated at 25°C to a 14 MeV neutron flux of up to 1'10 18 n/cm 2 . The same correlation was also obtained for unirradiated type 316 stainless steel which was strengthened by cold working. The theoretical basis for a correlation will be discussed. Preliminary data obtained from elevated temperature high energy neutron irradiations will also be presented.

THE INTERRELATION OF MACROSCOPIC PROPERTIES AND MICROSTRUCTURE OF RADIATION DAMAGED SiC INCLUDING EFFECTS OF HELIUM: J. C. Corelli, C. W. Lee, J. A. Lazzaro, and J. G. Hoole, Rensselaer Polytechni c Institute, Troy, NY 1218l. The fracture strength, swelling, electrical and thermal conductivities, and microstructure were studied for reaction bonded siliconized SiC and alpha phase sintered SiC after reactor neutron irradiation up to fluences corresponding to 'V18 dpa. The effects of radiation-induced helium gas on the pro~8rtiif were simulated by studying samBles dOQed with enriched B, B, and natural boron using the 1 B(n,a) 7Li reaction as a helium source. The results lead to an overall consistent model of the damage mechanism in SiC and give a good indication of what to expect in a fusion environment. The helium has a small but detectable degrading effect on the fracture strength. The thermal conductivity recovers completely in sintered SiC after annealing to l500 0 C. Helium must tie up vacancies and interstitials since it inhibits growth of dislocations. The swelling can be related to the decrease in thermal conductivity due to increased phonon scattering.

*Research sponsored by the U. S. Department of Energy.

10:35 a.m. 9:10 a.m. IRRADIATION-INDUCED BEHAVIOR OF YIELD STRENGTH OF SOLUTION ANNEALED 316 STAINLESS STEEL:* R. L. Simons, Westinghouse Hanford Company, Richland, WA 99352

Materials irradiation experiments cannot be easily designed to separate damage rate and spectrum effects. Structural components in fast breeder reactors and magnetic fusion energy devices can benefit from damage rate effects because lower damage rates are expected to produce lower irradiation-induced property changes. ':'ensile property data on solution annealed 316 stainless steel irradiated in several reactor or neutron sources with various damage rates have been analyzed with a modified first order damage produc tion model. Data on maximum change in yield strength (~YSmax) show good correlation with this model. The effect of damage rate is a two fold reduction in 6YSmax per factor of ten decrease in damage rate at temperatures below 'V300°C. Extrapolation of the model to higher temperatures shows a three fold Sensitivity to damage rate. However there are no data to substantiate the high temperature predictions. The correlation sh ows that most in or near core data obtained from EBR-II irradiat:.'..ons will show a 10-25% apparent scatter which can be attritubed to damage rate effects. *Research sp onsored by the U. S. Department of Energy .

9: 30 a.m . - Break

VI. Damage Mechanisms in Nonmetals Session Chairman: M. W. Guinan, Lawrence Livermore National Lab., P.O. Box 808, Livermore, CA 94550.

9:50 a.m. DISPLACEMENT CASCADES IN POLYATOMIC MATERIALS*, Don M. Parkin and C. Alton Coulter, Los Alamos National Laboratory, Los Alamos. New Mexico 87545

DEBONDING OF EPOXY FROM GLASS IN IRRADIATED LAMINATES. * C. E. Klabunde and R. R. Coltman, Jr., Solid State Division, Oak Ridge National Laboratory, Oak Ridge, TN 37830. Glass-fabric-filled epoxies irradiated at 4.7 K and examined at room temperature by 20X stereomicroscopy showed an internal flaw structure which increasingly filled the sample as the v-dose was increased. These flaws were determined to be areas where the plastic had debonded from the glass fibers. The extent of this process correlated well with the dose-dependent loss of mechanical strength. A possible mechanism is suggested for debonding in terms of free atom diffusion. New experiments are al so suggested to clarify further the failure mechanism. *Research sponsored by the Division of Materials Science, c. S. Department of Energy under contract W-7405-eng-26 Ivith the Union Carbide Corporation.

10:55 a.m. THE STRENGTH OF G-10CR AND G-llCR EPOXIES AFTER IRRADIATION AT 5 K BY GAMMA RAYS.* R. R. Coltman, Jr., and C. E. Klabunde, Solid State Division, Oak Ridge National Laboratory, Oak Ridge, TN 37830. . New data from three gallIna-ray irradiations at 5 K are combined with previous results to provide complete curves of flexural and compress i ve strength versus dose for G-10CR and G-ll CR gl ass fabric filled epoxies. Tests were made at 77 K following warmup to 307 K. The doses required to produce 25% losses in strength were 2.1 MGy in flexure and 3.0 MGy in compression for G-10CR andl~.7 MGy for fl exure and compress i on in G-ll CR. An accompanyi ng B fi ss i on fragment dose in G-lOCR had 1 ittl e effect upon fl exure strength . A possible mechanism of failure is the irradiationinduced debonding between glass fibers and the resin. *Research sponsored by the Division of Materials Science, U. S. Department of Energy under contract W-7405-eng-26 with the Union Carbide Corporation.