HYDROGEN

IN M E T A L S

The problem of the action of hydrogen on the corrosion-mechanical strength of metals is a pressing one in modern technology. At present metallurgists are devoting much attention to the following questions: hydrogen embrittlement of metals, the influence of hydrogen-containing media on their properties, metallurgical and technological methods of increasing the hydrogen resistance of metallic materials, and the use of hydrogen in processes for the working of metallic materials. The ~e~ults of investigations in this direction were discussed in the Fourth All-Union Seminar on "Hydrogen in Metals" held on September 17-21, 1984, in Moscow. Below are presented the results of four papers presented in the seminar. Questions of hydrogen embrittlement were presented earlier in Metal Science and Heat Treatment No. 5, 1982 and No. 2, 1984.

NATURE OF HYDROGEN EMBRITTLEMENT OF STEEL Yu. I. Archakov and I. D. Grebeshkova

UDC 669.788

In recent years the study of the interaction of metals with hydrogen has acquired in~ creasing importance. This is related primarily to the development of new directions in science and technology and to the ever wider use of hydrogen as a basic or accompanying reactant in various production operations. The negative influence of hydrogen on the properties of metals has been known for more than 100 years. A large number [1-7] of works has been published on this question. However, at present there is not a common opinion on the mechanism of hydrogen embrittlement of metals. Apparently, this is related to the influence of a multitude of factors, the many accompanying phenomena, and the complexity and insufficient study of the individual physicochemical processes. Hydrogen reacts with and dissolves in practically all metals and alloys [8, 9]. Therefore, at present hydrogen corrosion and brittle fracture have developed into a very important problem of hydrogen embrittlement of constructional and high-strength steels and nickel, titanium, vanadium, zirconium, and other metal base alloys. The hydrogen embrittlement of metals, the same as other forms of brittle fracture, is the result of the origin and development of microcracks, which are formed as the result of the occurrence of internal stresses. The specific features of the appearance of hydrogen embrittlement are the result of the physical properties of the metals and the character of their interaction with hydrogen. The tendency of metals toward hydrogen embrittlement is determined by the following characteristics: their capacity toidissolve hydrogen and its maximum solubility, the chemical activity of the metals and other phases in relation to hydrogen, that is, the capacity toward hydride formation and failure of the carbides and oxides, and the tendency of the metal toward the occurrence and propagation of cracks. In general two forms of action of hydrogen on metals may be distinguished. I. Action primarily at below-zero and low temperatures, when visible chemical reactions do not occur, that is~ the so-called physical action of hydrogen on metals. 2. Physicochemical action, when chemical interaction of hydrogen with the different phases and the individual components of the alloy on the surface and in the volume occurs. Brittleness of the first form is related to the concentration of hydrogen in cerzain zones such as the tips of submicrocracks, intergranular boundaries, and interphase surfaces. All-Union Scientific-Research Institute for Petrochemical Processes. Translated from Metallovedenie i Termicheskaya Obrabotka Metallov, No. 8, pp. 2-7, August, 1985.

0026-0673/85/0708-0555509.50

© 1986 Plenum Publishing Corporation

555

It should be noted that brittleness may occur with low rates of deformation. of such brittleness is not completely clear. It is assumed that weakening of bonds at the points of localization of hydrogen, which eases the formation of the action of the internal pressure of hydrogen and external stresses, has no ance in this. Brittle fracture is possible both at the boundaries and within pending upon the level of intercrystallite and intracrystallite strength.

The mechanism the interatomic cracks under small importthe grains de-

In the first case in adsorption and sorption of hydrogen by the metal the application of a load is necessary for the origin and development of cracks. Under conditions of the combined action of stresses and gas, hydrogen embrittlement is not accompanied by the formation of new phases and a change in microstructure. There are very many forms of hydrogen embrittlement of metals but a necessary condition of its appearance is the simultaneous presence of hydrogen and stresses which occur under the action of various factors, including the action of pressure of the hydrogen itself. Such a form of hydrogen embrittlement is frequently reversible. In cracking of the metal hydrogen embrittlement is irreversible (Fig. i). In the second case the physicochemical action of hydrogen on the metal is accompanied by chemical interaction of the gas with the individual phases and components of the alloy (reduction of carbide and other phases, the formation of hydrides), which leads to a change in microstructure of the metal. Hydrogen embrittlement of the second type is irreversible and no heat treatment is able to restore the initial properties of the metal. In contrast to hydrogen embrittlement of the first type, in this case fracture of the metal may occur even without the application of an external load. Let us consider the mechanism of hydrogen emhrittlement at increased temperatures and hydrogen pressure as the result of decarburization and cracking of the metal. First on the surface of contact of the metal with the hydrogen, as the result of thermal dissociation the molecular hydrogen is converted to atomic. With a constant temperature, in accordance with the law of mass action the elasticity of the atomic hydrogen increases proportionally to the square of pressure. Since the rate of diffusion of hydrogen in the metal is also proportional to the square of pressure, then it may be stated that in the absence of cracking the steel is impregnated only with atomic hydrogen. The hydrogen diffuses into the steel both along the grain boundaries and through the body of the grains. Penetration of the hydrogen occurs simultaneously with partial adsorption of it by the metal. The hydrogen is concentrated in the steel in the zones with the maximum free energy, in the grain boundaries, in all of the imperfections of the crystalline lattice, etc. Simultaneously with penetration of hydrogen into the steel decarburization of its surface starts. Thermodynamic calculations confirm [I0] that with a high hydrogen pressure and temperatures of 200-600°C the equilibrium of the decarburization reaction is shifted in the direction of the formation of methane, and practically complete decomposition of the cementite occurs° The process occurs both at the external and at the internal interfaces (grain boundaries and interphase boundaries):

C+4H~CH~, FeaC+4H~CH4+3Fe. The results of metallographic and electron microscopic investigations~show that decarburization starts at the grain boundaries (Fig. 2). The product of the decarburization reaction is methane, which was established chromatographically. The size of the methane molecule (d = 0.296 nm) is quite large and such a molecule may not diffuse through the lattice of the metal. Accumulation of the products of the reaction (methane and atomic hydrogen recombining into molecules) may occur initially in the pores and microvoids of the intercrystallite layer at the grain boundaries. In contrast to high temperature decarburization of steel (at temperatures above 700°C) when the rate of diffusion of carbon is high and the methane formation process occurs at the metal--gas interface, under the conditions considered internal decarburization of the steel occurs. The grain boundaries have a number of features which, in our opinion, are responsible for their initial cracking. According to much data [li D 12] at the grain boundaries there are concentrated primarily foreign impurity atoms, including carbon and "voids," and boundary segregations are created, that is, in an energy sense the grain boundaries are the most unstable areas. Therefore, the reaction products and molecular hydrogen are concentrated in the intercrystallite layer. According to calculation results [13] under these conditions the pressure of methane may reach very high values (tens of thousands of atmospheres), as 556

fOOl~.... 6ot%

•

I

~/aOr 6.o--~ ....

8or i, 80 I"\

.q fO0

.~

60

k",- _~___ J

• 20 60 120

6OK, I 4 o L~=

2o 6o12o 2ao ITime of hydrogen im1 pregnafion, h

28o

Time of hydrogen impregnation, h

Fig. i. Change in mechanical properties of 20 steel in relation to the time of action of hydrogen at 450"C with a pressure of 20 MPa: solid lines) directly after hydrogen impregnation; broken lines) after hydrogen impregnation and normalizing°

Fig. 2. Microstructure of commercially pure iron after a long hold at high temperature and a hydrogen pressure of 20 MPa. l15x. the result of which stresses exceeding the tensile strength of the metal occur. The period of time during which localized chemical reactions and the accumulation of the products of these reactions occur but a marked reduction in the strength and plastic properties of the steel is not observed is called the induction period [14]. According to the data of autoradiographic and electron-microscopic investigations the decarburization process starts immediately in chemisorption of hydrogen by the steel. Therefore, the induction period during decarburizatien of the steel may be characterized as the time during which decarburization occurs in the surface localized volumes, but Joining of individual voids at the grain boundaries, discharge of corrosion products, and a reduction in mechanical properties of the steel are not observed (Fig. 3). In the second stage of the action of hydrogen on steel the pressure of the corrosion products, primarily methane, causes porosity and loss of strength of the grain boundaries. The development of this process leads to the occurrence of microscopic cracks and discharge of the corrosion products through the cracks from the metal. Hydrogen chemisorbed on the surfaces of the individual microvoids also initiates the corrosion cracking process as the result of the decrease in surface energy of the cracks. The origin of cracks occurs in the zone of maximum tensile stresses occurring at the tips of voids located close to the surface of the metal and at the grain boundaries. Opening up of the voids located at the surface occurs and the decarburization process is sharply 557

a t, ag.z, M P a

~]r;~%

F00

L

~oo

,"×--~-

* OL

I

0

6

:~

I -7 ~ 12 18 24

so)

0 ZO

J6

4,2

~

Sqrlest, n

Fig. 3. Influence of test time Y- st on the mechanical properties and carbon content in ~ steel at 450=C in a hydrogen medium (PH2 = 30 MPa). accelerated. Then the individual microvoids, under the high pressure of methane and hydrogen, join. With the development of cracks a fresh surface of metal appears and the molecular hydrogen obtains access to the inner surfaces of the polycrystals, as the result of wh~ich ; the surface of interaction of hydrogen with the metal increases sharply. At the same time there is a reduction in carbon content in the surface layers, and intercrystallite cracking of the metal accompanied by a reduction in its mechanical properties occurs. Subsequently, decarburization may occur both at the boundaries and in the volume of the pearlite grains with the condition of the possibility of exit of the reaction products from the metal. Decarburization of individual pearlite grains occurs in the following manner. First the pearlite grain is surrounded on all sides or only on the side closest to the front of decarburization (in relation to test conditions) by a ferrite layer. After cracking of this layer and exit of the reaction products decarburization of the pearlite grain itself starts. Gradually there is a decrease in the size of the pearlite areas within the grains and broadening of the ferrite areas. Until recently there was a generally accepted opinion that the decarburization process occurs only on the surface of the grain boundaries. In this case, as the resultof creation of a carbon concentration gradient in the microvolumes dissociation of the cementite constituent occurs within the grain and the carbon liberated diffuses toward the boundary areas, where it reacts with the hydrogen. A confirmation of this opinion was the apparent absence of cracking within the pearlite grains. However, the presence of microdispersed ferrite in the structure after tests* and certain features of decarburization of the steel under conditions of increased temperatures and hydrogen pressure are difficult to explain based on the generally accepted decarburization mechanism. These include, for example, the strong influence of hydrogen pressure on the decarburization rate, the low values of the diffusion coefficients of hydrogen at 300-500°C, and the rapid decarburization of steel under such conditions. A comparison of the activation energy of the decarburization process of 20 steel (7.2 kcal/g.atom) with the activation energy of the carbon diffusion process (20 kcal/g'atom) is an indication of the fact that carbon diffusion in the steel may not be the determining factor in its decarburization. Calculations show that the quantity of hydrogen diffusion through specimens under certain conditions is several times greater than the quantity which reaets with the carbon of the steel. The activation energy of the process of diffusion of hydrogen in s-iron is 3.2 kcal/g'atom, that is, half the apparent activation energy of the decarburization process. Apparently during decarburization not only diffusion movement of carbon in ferrite occurs but also transfer of the reaction products along certain channels in the volume of the ferrite grains to their boundaries,.~that is, the boundaries of the blocks may be those routes along which the reaction products will travel from the internal volumes to the grain boundaries. It may be assumed that in the first moments of the reaction within the grains not methane is formed but type CH unsaturated hydrocarbons, the molecules of which have a small size per*Holds of specimens at hightemperatures

558

and hydrogen pressures.

48

/I

o~+(ir;.Fe) C3+

o,~

/I 7

0

"/

2

J

#

5

5

I ! o! °

I

7 B .,o 10 II 12 13 7z~ 75 16 17 18 %£r

Fig. 4. Relationship between the hydrogen resistance of chromium steels and their phase composition (PH; = 80 MPa, t = 600°C, m = 1000-4000 h): I) resistant steels; II) with signs of surface decarburization; III) not resistant; IV) resistant (data of [6]); V) not resistant (data of [6]).

MPa

& MPa -A__I I 78e

i i 'F

I Ill

!-!! j L~

i/i

z

:-

.----+2~

TFI irF r "r-

I I

] 1

70

I

702

,I

a

703

70

z-,,h

2

7

10

103

~,h

b

Fig. 5. Stress-rupture strength of 20Kh3VMF (i, 2), 20Kh3VMFB (3, 4), 15KhI2MVF (5), 15Kh5M (6, 7), 12KhI8NIOT (8, 9) [4] steels: a) at 600°C; b, c) at 800=C; i, 3) in nitrogen; 2, 4, 7, 9) in hydrogen; 5) in nitrogen and hydrogen; 6) in air; 8) in argon; 1-5) p = 60 MPa; 6-7) 15 MPa; b) 20 MPa; c) 40 MPa; center lines of the stress-rupture strength of 12KhI8NIOT steel) data of calculation by the method of least squares; upper and lower lines) boundaries of the 95% confidence interval. mitring them to freely move along the boundaries of the blocks. At the exit to the grain boundaries, where there is an excess of hydrogen, they are hydrated to formation of methane Such a concept makes it possible to explain the influence of pressure and temperature on the decarburization process. The results of investigation of the influence of alloy elements on the hydrogen resistance of steels and alloys showed that the stability of the carbide phases has decisive value in providing hydrogen resistance of the steel. Actually cementite is the least resistant to the action of hydrogen. With the addition to steel of stronger carbide-forming elements the forces of the bond in the carbide phases change so that a more stable electron configuration is formed. For example, alloying of steel with chromium, molybdenum, titanium, vanadium, and niobium provides a sharp increase in its hydrogen resistance [15]. It has been shown [16-18] that steels with 6% Cr are resistant to hydrogen embrittlement at 600=C with any hydrogen pressure. However~ as the result of longer tests [19] it was established that at 600=C with a hydrogen pressure of 80 MPa steels containing 6% Cr are subject to decarburization and are only resistant up to 500=C. In such steels with a carbon content of 0.18-0.21%, (Cr, Fe)TCs triangular chromium carbide, in which 30-50% Fe may be dissolved, is formed. Apparently this feature of the triangular carbide lattice structure is the reason for lack of resistance of it in the hydrogen medium at 6000C with a pressure of 80 MPa (Fig. 4). Data has been obtained on the influence of chromium on the hydrogen resistance of steel [19] which shows that it is also necessary to take into consideration the absolute quantities

559

TABLE 1

i I OpreI

Medium

I

ot

6~

MPa

Argon Hydrogen Hydrogen Hydrogen

70 70 120 150

360 370 390 390

%

I

[ 34 i 31 35 35

61 61 59 57

*Preliminary stress. Note. The preliminary hold of the specimens in the hydrogen medium was 30 min.

8~X,%

400 800 72007600 r~ h

Fig. 6.

Relationship of the ,;

(Dfml--D°rig 100%)

maximum plasticity \ Dori-~ " of 12KhI8NIOT steel at 800°C to test time: I) in argon~ p = 20 and 40 MPa; 2, 3) in hydrogen, p = 20 and 40 MPa, respectively; Dorig) specimen dianeter before testing; Dfail) after failure; Tf) time until failure.

of carbon and chromium and accordingly the phase composition of the steel. ~ r o m i u m steels are resistant to hydrogen embrittlement at 600°C with a hydrogen pressure of 80 MPa when all of the carbon is combined in type M2sC6 c a ~ i d e phase. With the addition to the steel of other strong ca~ide-forming elements the c a ~ o n must be combined in type MC cavities. Consequently, the nature of the c a ~ i d e phase, which determines the strength of interatomic bonds between the metal and the ca~on, has the decisive influence on the hydrogen resistance of the steel. This is confirmed by the results of stress-rupture strength investigations of mediumand high-alloy steels. For:example, according to the data of [20] the stress-rupture strength of 20Kh3VMF steel at 600°C with a hydrogen pressure of 60 MPa is lower than in tests in nitrogen. For steels not possessing ~ufficient hydrogen resistance an irrevers~le reduction in mechanical properties and stress-rupture strength is characteristic. High-alloy steels with 12% Cr additionally alloyed with tungsten, mo~bdenum, and vanadium and also 12Kh!8NIOT steel are hydrogen resistant at 600°C with a pressure of up to 80MPa, and their stress-rupture strength in hydrogen and nitrogen (and for 12KhI8NIOT steel in argon) is the same (Fig. 5). With high temperatures and hydrogen pressures decarburization of heat resistant and high temperature strength alloys containing a sufficient quantity of ca~ide-forming elements (titanium, vanadium, chromium, molybdenum) does not occur and therefore a reduction in their stress-rupture strength in tests in a hydrogen medium is not observed. This is confirmed by the results of short-term tensile tests of 12KhI8NIOT steel in a hydrogen medium with a pressure of 50 MPa at 700°C [21, 22] (Table i). The mechanical properties of this steel in argon and hydrogen media are practically the same, that is, a hydrogen content on the order of 80-90 cmS/100 g does not influence the change in them at 700°C. However, it should be noted that in tests of stainless steels in hydrogen there is a change in the character of failure of t ~ u l a r specimens. For example, the maximum plasticity of 12KhI8NIOT steel in hydrogen is significantly lower than in argon and with an increase in 560

pressure decreases (Fig. 6). Apparently an increase in hydrogen concentration (pressure) in the metal obstructs the occurrence of relaxation processes under conditions of difficult deformation. With a comparatively low hydrogen concentration (5-10 cm3/100 g) in carbon and lowand medium-alloy steels hydrogen has practically no influence on the resistance of the material to plastic deformation (in the absence of decarburization) but sharply decreases the maximum plasticity and fracture resistance. The tendency toward hydrogen embrittlement, that is, the relative intensity of the reduction in plasticity with the same hydrogen concentration, increases with an increase in the strength of the steel. In addition to the strength this characteristic also depends upon the chemical composition and structural condition of the steel. The reduction in strength and plasticity is accompanied by a decrease in the specific work for fracture of the material. The most complex problem of hydrogen embrittlement is determination of the boundary concentration relationships of hydrogen in metal guaranteeing safe service of structures under various loading conditions. It should be noted that the phenomena of high-temperature brittleness considered here are characteristic only for conditions of static loading. The influence of cyclic loads and the character and tendency toward brittle fracture of high-strength steels and also of hydride-forming metals may be different. Therefore, despite the large number of works questions of hydrogen embrittlement in the area of high temperatures and hydrogen pressures have been insufficiently studied. Further investigations must be directed toward revealing the influence of increased hydrogen concentrations on the strength and plastic properties of metals and alloys and also toward establishment of quantitative rules determining the life of the metal in the stressed state with various hydrogen concentrations. LITERATURE CITED i. 2. 3. 4. 5. 6. 7. 8. 9. i0. ii. 12. 13.

14.

15.

P. Kotterill, The Hydrogen Brittleness of Metals [in Russian], Metallurgizdat, Moscow (1963). B. A. Kolachev, The Hydrogen Brittleness of Nonferrous Metals [in Russian], Metallurgiya, Moscow (1966). L. S. Moroz and B. V. Chechulin, Hydrogen Brittleness of Metals [in Russian], Metallurgiya, Moscow (1967). Yu. I. Archakov, The Hydrogen Resistance of Steel [in Russian], Metallurgiya, Moscow (1978). G. V. Karpenko~and R. I. Kripyakevich, The Influence of Hydrogen on the Properties of Steel [in Russian], Metallurgizdat, Moscow (1962). B. I. Sarrak, "The hydrogen brittleness and structural condition of steel," Metalloved. Term. Obrab. Met., No. 5, ii (1982). V. N. Zikeev, "The alloying and structure of constructional steels resistant to hydrogen embrittlement," Metalloved. Term. Obrab. Met., No. 5, 18 (1982). P. V. GelVd, R. A. Ryabov, and E. S. Kodes, Hydrogen and Imperfections of the Structure of Metal [in Russian], Metallurgiya, Moscow (1979). G. Alefeld and I. Felkel, Hydrogen in Metals [Russian translation], Vol. l, Mir, Moscow (1981). R. E. Bisaro and C. H. Geiger, Corrosion, 23, No. i0, 289-296 (1967). D. MacLean, Grain Boundaries in Metals [in Russian], Metallurgizdat, Moscow (1960). A. P. Gulyaev, Physical Metallurgy [in Russian], Metallurgiya, Moscow (1977). S. D. Bogolyubskii, V. I. Alekseev, I. S. Ushakov, and L. A. Shvartsman, "The relationship of hydrogen corrosion of steel to the thermodynamic activity of carbon," in: Hydrogen Corrosion of Steel and Combatting It [in Russian], Tsent. Nauch.-Issled. Inst. Inf. i Tekh.-Ekon. Issled. Neftepererab. i Neftekhim. Promysh., Moscow (1967), pp. 58-65. S. L. Skop, V. V. Ipat'ev, and V. P. Teodorovich, "The decarburization of carbon steels by hydrogen at high temperatures and pressures", Zh. Prikl. Khim., 31, No. 12, 1894-1897 (1958). Yu. I. Archakov and I. D. Grebeshkoya, "Hydrogen-resistant steels not in short supply," in: New Materials in Machine Building [in Russian], Mashinostroenie, Moscow (1964), pp. 19-43. 561

16. 17. 18. 19.

20.

21. 22.

F. K. Namnann, Stahl Eisen, 58, No. 44, 1239-1250 (1938). K. Clark, ~igh Temperature Strength Alloys [Russian translation], Metallurgizdat, Moscow (1957). G. A. Nelson, Hydrocarbon Proc., 44, No. 5, 185-188 (1965). Yu. I. Archakov and I. D. Grebeshkova, "The influence of alloy elements on the long' term hydrogen resistance of steel," in: Investigations of High-Temperature Strength Alloys [in Russian], No. i0, Izd. Akad. Nauk SSSR, Moscow (1963), pp. 305-313. N. P. Chernykh, "The influence of hydrogen on the stress-rupture strength of certain steels," in: The Influence of Hydrogen on the Service Properties of Steel [in Russian], Irkutsk. Knizh. Izd-vo (1963), pp. 22-46. L. A. Glikman and V. I. Deryabina, "Tensile tests in hydrogen and other corrosive media at high pressures and temperatures," Zavod. Lab., No. 5, 612-613 (1965). L. A. Glikman, V. I. Deryabina, and V. P. Teodorovich, "Determination of the mechanical properties of steel by short-term rupture in hydrogen at high temperatures and pressures," Fiz.-Khim° Mekh. Mater., No. 3, 71-74 (1972).

RESISTANCE OF CONSTRUCTIONAL STEEL TO FRACTURE IN HYDROGEN IMPREGNATION AND HYDROGEN SULFIDE CRACKING Eo A. Savchenkov

UDC 620.184:620.194:620.172:669.14

At present definite successes have been attained in understanding the phenomenon of hydrogen embrittlement of steel [1-3 et al.]. However, only a qualitative relationship has been established. The greater the hydrogen concentration at the crack tip, the lower the failure stress intensity factor [4]. The concept of Troiano and Oriani [5, 6] of the critical hydrogen concentration as the limit at which the metal loses cohesion strength has not made it possible to completely clarify this question since neither theoretical nor experimental methods of determination of such a critical concentration have been proposed [2, 4]. The critical activity (concentration) is a thermodynamic parameter of hydrogen, and the hydrogen resistance of a steel is determined by its structure and is not subject to a thermodynamic description. Absorbed hydrogen may be found in various structural conditions and does not have the same influence on the strength and plasticity of steel [7], The variety and number of the effects of hydrogen influence have been responsible for conflicting opinions and hypotheses on hydrogen embrittlement. Hydrogen in alloys is simultaneously a surface and a volumetric ~gent. mally only its adsorption or cohesion action is taken into consideration.

However, nor-

The study of hydrogen embrittlement involves method difficulties since the most "dangerous" is active hydrogen. The activity a ~ exp(Ap/kT) depends upon the difference in chemical potentials Ap in the presence of which hydrogen is diffusion-mobile, which leads to the appearance of both flows and nonuniformity in the hydrogen concentration. Of the many works devoted to investigation of hydrogen in metals only in a few ([3, 8-11], for example) are the mechanical properties compared with the activity (fugacity) of hydrogen. Earl~er on specimens with stress raisers we [i0, ii] obtained for the first time generalized empirical relationships of the strength S of steel (the characteristic of the metal which makes it possible to evaluate its average cohesion strength [12, 13, et al.]) to the concentration of the free C~ and combined C2 components of the absorbed hydrogen. Below are analyzedthe basic results following from these relationships. According to the data of [I0, ii] the coefficient of hydrogen decohesion (loss of strength) B = AS/S in reversible hydrogen embrittlement is determined as ~1 = ~e exp (7C1) °'s = ~e exp a? "5 = exP (a? "5 -- a ~ s ) ;

(1)

Orenburg Polytechnic Institute. Translated from Metallovedenie i Termioheskaya Obrabotka Metallov, No. 8p pp. 7-11, August, 1985. 562

0026-0673/85/0708-0562509.50

© 1986 Plenum Publishing Corporation