Search Documents

By R. G. Treuting, W. C. Ellis

D. Whitwham, M. Mouflard, and P. Lacombe (British Council Research Fellow, Labratoire du Professor Chaudron, Vitry-sur-Seine; Inqenieur de Recherches, Labratoire de Vitry; and Maitre de Recherches, Labratoire de Vitry, Professeur Adjoint de Metallurgie, Ecole des Mines, Paris)—The authors' work is of great interest in bringing out the geometrical relations existing between twinned lattices, and for the notion of a coincidence super lattice developed from this. We should like to point out that a similar idea was put forward by Friedel8 in his book on the subject of lattice merohedrism "merihedrie réticulaire:"—When two crystals A and B are twinned with respect to a mirror plane, there exists a common superlattice, "réseau periode," which is independent of the contact surface between the two crystals. The volume V of this superlattice is a simple multiple of the volume v of the lattice of crystals A, or B, (Fig. 5). This "multiple cell" is defined by the twin plane (p q r) and a row of atoms [h k l] normal to this plane, the base being taken as the smallest unit (two-dimensional) of this plane. Two cases must be considered. First when the multiple cell, as defined, has no atom site at its center nor at the center of its faces other than (p q r), v the twinning index is defined by —. In the second v case, when one of the two above conditions does not v apply, then the twinning index is V 2v A more intuitive definition of the twinning index would be the proportion of atoms of C, which would coincide with those of C1 when the C, lattice is imagined to extend on the other side of the contact surface. In the case of cubic crystals, the twinning index

Jan 1, 1952

By C. E. Birchenall, F. N. Rhines, L. A. Hughes

A. J. Shaler—I should like to congratulate the authors on the presentation of this paper, which we have been awaiting a long time. The view they have taken of the sintering process, namely that voids are comparable to particles of a second phase capable of dissolving to a slight extent in the metal phase, is an interesting but dangerous one. Pines has given a first-order solution of the problem and Has predicted a decrease in the concentration of pores near the surface. The authors of the present paper have experimentally demonstrated the second-order effects, that is, the diffusion of the space phase between voids of different sizes, and also the movement of atoms from highly curved parts of a void to less curved regions of the same void. I have recently calculated the rate of vapor-phase transport within odd-shaped pores in copper at 850 °C and found that probably at that temperature volume flow is the more rapid shape-changing mechanism. The authors of the present paper have found that at 1000°C, vapor-phase transport is dominant. There is no disagreement here, I think. The dominant phenomenon can easily be different' in different temperature ranges and with different pore shapes: I used a cylinder going to a sphere, not a tricuspid going to a "triple cuspidor." But there is some danger in the other art of the explanation, that is, in the study of the diffusion of vacancies from a pore to its neighbors and to the outside. A vacancy is very different in dimension from an atom. When it moves, it carries a field of elastic strain. This has to be taken into account. A concentration gradient of vacancies can cause them to diffuse. But in the absence of such a gradient, they can be made to diffuse by a field of stress. What happens in some of our experiments, and it is most noticeable in the shrinkage of synthetic cylindrical pores, like the tubes we have been working with, may be broken down into the following steps. (1) The presence of a surface establishes a vacancy concentration gradient, as Dr. Kuczynski and the present authors have shown. (2) Vacancies then diffuse in order to decrease this gradient. (3) In so doing, they cause an elastic field to be built up in such a direction as to oppose the diffusion, so that it slows down until it reaches a steady-state rate. (4) The steady-state flow is due to the fact that the surface tension acting on curved surfaces applies a stress to the system which itself mechanically causes vacancy diffusion. What I have said is that the diffusive types of flow of the authors and of Dr. Kuczynski are transient phenomena which soon stop if the pores are large. We are then left with a mechanical problem only, and our researches have not yet disclosed just what the mechanism of flow is; it may be a slip process, or it may take place by a vacancy diffusion of the type Nabarro has de-

Jan 1, 1951

By J. H. O’Dette

BLISTER formation on wrought aluminum products has been a matter of concern to aluminum fabricators for many years. The fact that blisters make their appearance during the final stages of fabrication means that much time and effort have been expended before it is evident that the material must be scrapped. This and the loss of production are ample reasons for embarking on a research program to find out more about blistering and possible methods of control. Many excellent papers on this subject have appeared in the technical literature in the last few years. In one of the more recent papers, Kostronl gives a fairly complete review of the problem of gas in aluminum and blistering in general. From the work of the various investigators, a general mechanism for the formation of blisters has evolved and may be summarized briefly as follows: Hydrogen gas is usually present in the cast ingot, in the atomic or ionic state, in excess of the solid solubility. At room temperature the gas atoms are relatively immobile, but at elevated temperatures they become mobile and diffuse to free surfaces where they associate to form molecules. If these free surfaces are discontinuities within the ingot then the associated gas molecules collect and build up considerable pressure until equilibrium is reached with the dissolved gas around the discontinuity. Since the solubility of the gas is proportional to the square root of the partial pressure it is evident that large pressures may be built up at the discontinuities. When the metal is rolled to sheet the pressure of the gas in these, discontinuities is sufficient to deform the surface of the sheet and form a blister, particularly when the strength of the material is reduced during the annealing. Although the mechanism of the formation of blisters was fairly well known, the actual impor- tance of the individual roles played by the gas and the discontinuities was not so well established that positive methods for control of blistering were at hand. It was the object of the investigation to learn more about the nature of the discontinuities giving rise to blistering and to develop methods of control. The investigation was carried out in three stages as follows: a) A study of the more commonly used methods of removing hydrogen from aluminum, such as fluxing and holding, and their effects on blistering. b) Location of the discontinuities giving rise to blistering. c) A study, by indirect methods, of the movement of gas in the metal and its effect on blistering, and the development of methods of blister control during fabrication. Removing Gas and Reducing Blistering by Fluxing or Holding-—The work of removing gas and reducing blistering by fluxing or holding was carried out on metal of Alcan 2s (1100) composition with an artificially high gas content. The high gas content was obtained by introducing water into the melt prior to casting; the water reacted with the aluminum forming A12O3 and hydrogen. The procedure was to raise the gas content by introducing water into an aluminum melt and using half the melt to cast one ingot immediately. The remaining metal was treated again with water to make up for the reduction in gas content during holding, while the first ingot was being cast, and then given a fluxing or holding treatment and cast into an ingot. Samples for gas determination were cast from the molten metal before and after the gassing operation; before, after, and during fluxing, and while casting. Samples were also cut from sections of the ingots. Gas determinations were made on these samples using the well known vacuum fusion technique. All ingots were hot rolled at 500°C to 0.250-in. slab which, in turn, was cold rolled to 0.040 in. sheet. The sheet was coiled and annealed for 2 hr at 500 °C and then cut into 5-in. squares which were inspected for

Jan 1, 1958

By R. G. Gifkins

Mean boundary migration depended linearly upon creep extension under various conditions. Prolonged annealing or air-casting increased the proportion of immobile boundaries at a given stage of tests. Grain growth during annealing followed a power law. The linking of migration and sliding in creep and models for migration are examined; activation energies are calculated from these models. Segregation of oxygen to boundaries appears to be a controlling factor. THE idea that boundary migration and boundary sliding during creep are linked has often been suggested,'-5 especially when evidence has been obtained for an alternation between the two processes. In simple bicrystals sliding can occur without any migration6 but in more complex bicrystals7 it is observed. McLean' and Couling and Roberts 5 made measurements of both sliding and migration; McLean concluded that there was no linking of the two processes whereas Couling and Roberts supported the idea. McLean' also developed a model for migration during creep and annealing. The present work was initiated as part of a study of the recrystallization of lead during creep9 but has been extended to examine some aspects of the mechanism of migration. MATERIALS AND METHODS Specially refined high-purity lead* was used; the total of impurities, excluding oxygen, was 0.0008 wt pct. Oxygen was removed as far as possible by melting in vacuo, treating at 550" to 600°C with hydrogen and then casting at 450°C under a vacuum of 10-2 mm Hg; the oxygen was then believed to be 0.00005 wt pct. Specimens were made from extruded strip and then annealed at the required temperature in an air oven, controlled to 5 lac. They were then polished in a mixture of acetic acid and hydrogen peroxide. At this stage some specimens were etched to give grain contrast and their grain sizes determined by the intercept method. Migration of boundaries during creep was measured on the surface of polished (not etched) specimens, directly loaded at room temperature (22°C) or 50°C (52"C) to initial stresses of 300 to 700 1b per sq in. Twenty six vacuum-cast hydrogen-reduced specimens were used and, in addition, two specimens were from material air cast after hydrogen reduction ("partial" air cast) and two from air-cast material not treated with hydrogen ("full" air cast). Duplicates were made in the majority of tests. At intervals through the creep tests, specimens were unloaded and the migration measured, normal to the direction of migration at the point of measurement, for each boundary along standard traverses. Usually between 110 to 180 boundaries were measured at each of some twelve stages of the tests. Migration was measured on the projected image (X100) using prism illumination. Fig. 1 illustrates the method of measurement and shows the kind of variation of migration found from boundary to boundary and along particular boundaries. Fig. 2 gives an example of more extensive migration. The mean value of distance migrated was generally reproducible to *10 pct but occasional points gave a scatter of *20 pct. The correctibn applied by McLean8 to allow for the contribution sliding makes to the apparent distance of migration could not be used, for the sliding at most boundaries was too small to measure with the equipment available. There were two reasons for supposing the correction unnecessary in the present work;

Jan 1, 1960

By E. M. Baroody

MUCH information on intermetallic diffusion has been obtained in experiments in which two volumes, initially at different uniform concentrations, are maintained in contact at an interface which is plane, or nearly so. Concentration-distance curves are obtained after suitable times, and analyzed on the basis of a one-dimensional form of Fick's law In early work the linear equation to which Eq. 1 reduces when D is independent of concentration was used. However, when the initial concentrations c, and c, are appreciably different this simplification is often unjustified, and the suggestion by Matanol in 1933 of a way of avoiding it was welcomed. Matano's paper was based on work done as early as 1882 by Boltzmann.2,3 In 1938 Rhines and Meh14 reported an extensive investigation of systems in which the diffusion coefficient depends strongly on concentration, and included a detailed description of the Matano-Boltz-mann method. To a large degree, this description represents the formal method as it is understood and applied today, most later discussions having emphasized thermodynamic and mechanistic aspects of diffusion. Exceptions are the comments of R. V. Churchill, contained in a paper by Ham, Parke and Herzig,5 and the papers by Hall8 and by Kirkaldy.7 More discussion appears necessary since some aspects of the problem remain obscure. Better computational procedures are also needed. The common method, involving graphical determination of slopes and areas, is inconvenient and makes the estimation of errors difficult. Background of the Matano-Boltzmann Method Eq. 1 is to be considered along with initial and boundary conditions. Take the origin of x at the

Jan 1, 1958

By M. Cohen, C. Wagner, J. E. Reynolds

If the total volume of a diffusion couple changes during the diffusion, the measurement of distance becomes ambiguous. Use of distance parameters as suggested by Hartley and Crank is discussed. For small concentration differences, the standard form of Fick's second law is retained with conventional length units and the in- terdiffusion coefficient D. It is shown how D can be calculated from experiments involving large concentration differences and analyzed in terms of distance parameters. 'The merits of incremental diffusion couples involving small concentration differences are emphasized. THE conventional form of Fick's second law for one-dimensional diffusion of component B in a binary solution of A and B is where c, is the concentration of component B; x, the distance; t, the time; and D, the interdiffusion coefficient which may depend on ca. Hartley and Crank' have pointed out that Eq. 1 applies only if the total volume of the diffusion couple does not change during the diffusion process; otherwise, the definition of x becomes ambiguous. No change in total volume takes place if the specific volume of the solution is a linear function of the weight fraction Wb, or if the molar volume is a linear function of the mol fraction N. In some alloys, however, these relationships do not hold, and then it is necessary to introduce a modified measure of distance, 4 in order to retain the general form of Eq. 1. The application of Hartley and Crank's suggestions to diffusion in alloys has been discussed by Wagner.' Relation Between Measure of Distance and Units of Concentration The definition of [ depends on the choice of units for the concentration c,, as follows: 1—If the concentration ca is measured as the weight fraction W, (or weight percent), [ = 5, has to be defined so that equal increments of contain equal increments of mass m of the solution, e.g., dm = pSdx = Sdw [2] where p is the local density of the solution, and S is the cross-sectional area in sq cm which is taken to be constant in this problem of one-dimensional diffusion." * da Silva and Mehla have demonstrated the constancy of the cross-sectional area in the usual type of diffusion couple. However, experiments involving diffusion into thin wires4, 5 may exhibit changes in cross-sectional area, and therefore are excluded from the present treatment. Thus and the unit of is gram per sq cm instead of cm. 2—If the concentration c, is the mol fraction N, must be defined so that equal increments of contain equal numbers of mols of solution, e.g., dn= (S/V)dx= Sd [41 where V is the local molar volume of the solution. Thus dt, = dx/V [51 and the unit of is mols per sq cm. 3—If, as may be advisable in the case of interstitial solutions, the concentration C is measured in terms of the molar ratio of components B and A (c,, = Y, = nH/nA, where n and nl are numbers of mols of B and A), = Y has to be defined so that equal increments of contain equal numbers of mols of A, e.g., dnA= (NAS/V)dx = Sdf [6]

Jan 1, 1954

By D. Turnbull, J. C. Fisher

ACCORDING to the "reaction-path" modell,2 of martensite nucleation, the shear angle of the embryonic martensite plate must be treated as a variable, and included in any calculation of nucleus critical size. Also, as can be deduced from this model, the interfacial free energy between austenite and martensite does not reach its final value until the shear is completed. It is zero for zero shear angle. However, in order to account for the kinetics of the martensite transformation, some sort of interfacial energy barrier appears to be necessary even with the reaction-path model, for otherwise the volume and the energy of formation of the critical size nucleus both collapse to zero.3 Cohen independently suggested that surface energy could be incorporated into the reaction-path model, with the overall free energy of a martensite embryo being a function of its volume and shear angle.' It is possible to estimate the energy associated with the formation of a critical-size martensite nucleus starting with the reaction-path model and including a surface free-energy barrier. As the dependence of interfacial free energy upon shear angle is unknown, a simple type of dependence will be assumed, with the belief that the true dependence would not lead to appreciably different results. Consider the work required to form a lenticular martensite plate with radius r, thickness t, and shear angle 8. There are three contributions; one being the interfacial free energy, one being the free energy change in the martensite plate, and one being the free energy increase in the surrounding austenite. The interfacial free energy u is assumed to depend upon the shear angle 0 according to the relationship s=s0(?/?0)n [1] where 8, is the equilibrium shear angle and n is an exponent that may lie in the range 0 n 2. The work required to form the interfaces of a martensite plate then is W. = 2pr² s0(?/?0)n [2] The free energy change per unit volume of martensite is composed of two parts, one the ordinary volume free energy ?f1. which is negative, and the other the elastic strain energy G?m²/2, where G is the shear modulus and 7, the shear strain relative to the martensite structure. This expression for the strain energy is valid only when the shear strain ym, is sufficiently small that the martensite is within its linear elastic range. There is no doubt that ym, lies beyond the linear elastic range for embryos that are considerably subcritical. However, for critical nuclei it will be shown that ym, is 1.5 pct or less, within the linear elastic range of martensite. For embryos of nearly critical size, then, the strain energy of the martensite is correctly given by G?m²/2. The shear strain in the martensite is ym, = 8, — 8, and the work required to form the strained martensite is Wm --= (pr²t/2) [?fv + G(?O - ?)²/2] [3] The free energy change in the austenite is entirely that due to elastic distortion. The elastic strain is not uniformly distributed in the austenite, being large near the martensite plate and small elsewhere. Approximately, however, the energy corresponds to a uniform shear strain ya= (?t/2)/r [4] throughout the volume 4pr³/3 surrounding the plate. The work required to strain the surrounding austenite then is Wa = (4pr³/3) (G?a²/2) = (G?²/6) prt² [51 For simplicity, the same shear modulus G is assumed for each structure. The total free energy for forming a plate then is W = W3 + Wm + Wa. = 2pr² s0 (?/p?0)n + (pr²t/2) [?fr+G(?0-?)²/2] + (G?²6) prt2 [6] This expression is correct for nuclei and for embryos of nearly critical size, where, as will be shown, the strain energy in the martensite is correctly given by the expression G (? — ?)². Having W as a function of r, t, and 8, as in Eq. 6, there is a saddle-point where W has a stationary value, W subsequently decreasing indefinitely as the nucleus volume increases along the reaction path. The stationary value of W is the energy of the critical nucleus. The critical nucleus has radius, thickness, and shear angle such that ?W/?r - awlat: = ?W/?p? = 0. Performing these differentiations and calculating the critical nucleus energy, W* = [8192p(G?/6)²;s/27 ?fv4] [7] where a= (?/?0)3n+1[l +G(8"-8)'/2af.]' [7a] and where 8 is to be determined from the equation (1 + 3n/4) + G8(6O - (9)/[Af. +G(6>o-6>)72] = 0 [8] For ?f, near —200 cal per mol or —10" ergs per cc, and 8, near 1/6, as for iron-base alloys, Eq. 8 gives ?0 - ? ~ - (4 + 3n) ?f1./4G0O [9] as the difference between the equilibrium shear angle and the actual shear angle for a critical nu-

Jan 1, 1954

By J. W. Clark, C. T. Sims

Wrought chromium-base alloys containing yttrium, cubic monocarbides of the Ti(Zr)C type, and similay alloys containing manganese and rhenium have been melted and fabricated. Strength has been studied by hot hardness and elevated-temperature tensile and rupture measurements, low-temperature ductility by tensile testing, and surface stability by oxidation testing. In additiod, studies have been conducted of the carbide stability, and of aging behavior. The carbide dispersion generates effective elevated-temperature strength, which is further enhanced hv strain-induced precipitation. The dispersion exhibits classical dissolution and aging response. The ductile-to-brittle transition temperature of these alloys is above room temperature. The alloys reported show fairly good oxidation resistance, but nitrogen contamination can cause fortnation of a hard Cr2N layer under the oxide scale. Manganese does not appear to be a promising alloying element in chromium. In the years 1945 to 1950, the metal chromium was considered as a possible base for alloy systems due to its considerably higher melting point than superalloys, its low density, its high thermal conductivity, and its apparent capacity for strengthening. However, this interest in chromium was short-lived. It was found difficult to melt and cast, to be exceptionally sensitive to the effect of minor imperfections, to have a lack of ductility at both room and elevated temperatures, and to be subject to a deleterious effect of alloying elements upon the ductile-to-brittle transition temperature.' Since then, chromium, as a practical alloy base, has remained virtually unstudied. Further, purposeful ignoring of chromium has been promoted by statements that its bcc structure would not allow it to be strengthened to useful values, when compared to the "austenitic" alloys.2 Recently, a new look has been taken at chromium-base alloy systems. Study of the literature will show that chromium, providing some of its disadvantages could be eliminated or minimized, actually has a rather attractive potential as an alloy-system base. Analysis of rather scattered data suggests that chromium is quite capable of being strengthened to high levels. Also, significant strengthening of its two sister elements in Group VI-A, molybdenum and tungsten, has been demonstrated in a number of commercial and exploratory alloys. Chromium should be similar. Since chromium does not readily form a volatile oxide like tungsten or molybdenum, it offers a much higher probability of giving birth to alloy systems with useful oxidation resistance. Concerns about possible high elemental vapor pressure have been mitigated by recent data.3 In addition, the physical properties exhibited by chromium are attractive for application as a high-temperature structural material. For instance, its thermal conductivity varies from 49 to 36 Btu-ft/hr-sq ft-°F over its range of usefulness (which is two to four times higher than most superalloys), its density is about 7.2 g per cc (20 pct less than most nickel-base alloys), its coefficient of thermal expansion varies from 4 to 8 x 10-6 per OF, and it has a relatively high modulus of elasticity, approximately 42 x 10' psi.4 Alloying studies on a chromium base in the past have usually encompassed rather sweeping solid-solution alloy additions for strengthening. This is not consistent with contemporary alloying practice in Group VI-A. For instance, molybdenum, also in Group VI-A, is primarily alloyed for strength improvement by use of heat-treatable carbide dispersions.5 Chromium and molybdenum are similar in their chemical activity and other properties. Thus, strengthening of chromium by carbide dispersions was studied. Chromium-base alloys are plagued with room-temperature brittleness, although high-purity unal-loyed chromium can be made ductile.4,8 Use of yttrium as a scavenger has done much to improve ductility and resistance to nitrogen embrittlement in chromium systems,7 so it was utilized in this program. It has also recently been found8 that small rhenium additions (1 to 5 pct) create improvement in the ductility of Type 218 tungsten wire. This is apparently related to the remarkable effect of rhenium additions near its terminal solid solubility in all Group VI-A metals.9'10 Investigation to establish if dilute concentrations of rhenium would also be effective in chromium appeared to be logical for this program. Since rhenium is too expensive to be practical in alloys for application as structural components, ductility improvements through solid-solution alloying were also sought by substitution of manganese for rhenium; manganese, like rhenium, exists in Group VII of the periodic system. The optimum amount of carbide dispersion for chromium-base alloys was obtained by analogy with molybdenum. Strengthening in molybdenum is achieved by use of Ti-Zr carbide dispersions. A

Jan 1, 1964

By John C. Sawyer

A brief review is given to show that catastrophic oxidation of stainless steel in the presence of lead oxide is similar in many respects to catastrophic oxidation as caused by MoO3 and V2O5, Experimental data are presented to show that lead oxide accelerates the normal oxidation process of stainless steel by chemically altering the normally protective oxide film formed on the base metal. Data are given to show that similar reactions occur during oxidation of stainless steel in the presence of MOO3,and V2O5. THE mechanism by which pure metals oxidize at high temperatures has been the subject of intensive research for many years and much has been accomplished towards understanding the processes involved. The mechanisms by which alloys oxidize have also been investigated, but due partly to the fewer number of investigations and partly to the complexity of the oxide systems present, available knowledge is not as extensive as with pure metals. When oxidation processes are further complicated by atmospheres contaminated with sulfide, halide, or oxide vapors, the knowledge of fundamentals becomes meager or altogether lacking. Thus, it is not surprising to find that the catastrophic oxidation of stainless steels in metal oxide vapors of MOO, and V2O5 is not a well understood phenomenon despite the fact that a number of investigations have been undertaken. The literature in dealing with the subject of catastrophic oxidation is primarily concerned with the effects of MOO3 and V2O5, but does mention that other metal oxide vapors, one of which is lead oxide, appear to promote a similar type of accelerated oxidation. While this latter effect is virtually unknown, the deleterious effect of lead oxide is of considerable commercial importance, as it is believed to be one of the fundamental reasons for exhaust-valve failure in gasoline engines. For this reason studies have been in progress for many years attempting to better understand the mechanism involved. Modern exhaust-valve alloys"2containing 20 to 25 pct Cr plus nickel and/or manganese are in effect high-grade austenitic stainless steels exhibiting ex- cellent oxidation resistance in air at exhaust valve operating temperatures of 1100° to 1700°F; yet these same alloys are prone to corrode during engine service because of the conditions attending their operation. These conditions are such that the valves become coated with deposits of complex lead compounds mainly derived from gasoline additives of tetraethyllead, ethylene dibromide, and ethylene dichloride. Literature searches3-7 and discussions with various investigators show that considerable difference of opinion exists as to the nature of the lead-bearing deposits found on the surface of exhaust valves and the atmospheres to which they are subjected; yet, all seem to generally agree that corrosion, a gradual transformation of alloy to nonmetallic products of corrosion, is a consequence of the presence of lead compounds. This majority opinion seems to be substantiated by laboratory engine tests which show that corrosion does not occur to any appreciable extent when lead-free fuels are used. The first observations of catastrophic oxidation reported in the literature were probably made by pfeil,8 but it was Leslie and Fontana9 who undertook the first detailed study of the phenomenon. Their work was chiefly centered on the acceleration of oxidation which took place when stainless steels containing more than 3 pct Mo were heated in air with restricted circulation. These investigators reported that the presence of other oxides near the surface would further accelerate the rapid oxidation which occurred. While they recognized that the effect was attributable to the presence of molybdenum, it was Rathenau and Meijering" who emphasized the fact that the presence of molybdenum in the alloy was incidental to the presence of MOO3 vapor in the atmosphere surrounding the specimens. Evans,11 as well as Cunningham and Brasunas 12 studied similar

Jan 1, 1962

By A. E. Gorum, J. Washburn, E. R. Parker

Experimental evidence was obtained in support of the idea that cleavage fracture can be initiated by dislocation pile-up. The high ductility of MgO crystals when tested in bending comPared to their relative brittleness in tension was explained on the basis that large pile-ups are more likely in a tension specimen. CLEAVAGE fractures have been a problem of major concern for many years. A number of large steel structures have failed in a brittle manner, usually suddenly and without warning; often the failures occurred for no obvious reason and at nominal stress levels well within the generally accepted range. A great deal of data have been collected about factors that contribute to failures,1 but the physics of the fracture process has never been revealed. cottrel12 has recently developed a theory of brittle fracture in ductile materials which was based upon the concept that dislocations moving on two intersecting planes can unite to form a crack nucleus. This theory accounts for the known effect of grain size and for the effect of testing temperature. There has, as yet, been no experimental evidence published that would substantiate this or the earlier theories based on stress concentrations due to piled-up dislocations. The study reported herein was conducted to obtain information that would either substantiate or disprove the

Jan 1, 1960

By Nicholas J. Grant, David S. Bloom

An investigation of the Cr-Ni system has revealed that there exists a eutectoid transformation at about 1180°C. 'The transformation in chromium implied by the eutectoid reaction was detected at about 100°C below its melting point. Associated with the eutectoid transformation, a metastable transition structure has been found. THE Cr-Ni phase diagram is shown in current literature to be a simple eutectic system in which the high chromium end of the diagram is incompletely determined, Fig. 1. The last investigation of this system was undertaken in 1936.' Because of limitations in metal purity, especially of chromium, and temperature measuring techniques, the diagram should be accepted with some reservations, as was indicated by Hansen.V or examwle.-, the wresence of the compound Cr²Ni, with a tetragonal structure," remains unverified and incompatible with the simple eutectic diagram. The work reported here began because of discrepancies that appeared in experimental studies in the binary alloys Cr-Ni and in the ternary system Cr-Mo-Ni. The existing binary Cr-Ni diagram failed to provide a complete explanation of observations made by microscope, X-ray, and thermal analyses. Experimental Methods The materials employed were of a higher degree of purity than were available to previous investigators. The Mond nickel used had a nominal purity of 99.87 pct, while the electrolytic chromium analyzed 99.03 pct with the major impurity being 0.5 to 0.7 pct oxygen with some iron. All alloys were melted in a vacuum furnace to remove dissolved gases, sufficient carbon was added to the charge to combine with all the oxygen as CO. Final analysis determined the carbon as about 0.02 pct and the oxygen around 0.06 pct or slightly more. Stabilized zirconia crucibles and thermocouple protection tips were used throughout and contributed no detectable impurity to the melts, as long as the oxygen in the chromium was kept low. Heats averaged about 250 g, which permitted about 2 in. of the thermocouple protection tips to be submerged. The temperature drop across the thermocouple protection tip was determined to be about 3°C, a minor source of error. The alloys, after removal from the vacuum furnace, were placed in a furnace operating in a purified argon atmosphere for the thermal analyses. An- nealed wolfram-molybdenum thermocouples provided the primary means of temperature determination, though a Pt-Pt-10 pct Rh couple was employed for a check when possible. A sensitive, high-speed Leeds and Northrup Speedomax recording potentiometer indicated the thermocouple output. X-ray diffraction patterns were obtained primarily on a Norelco unit, which permitted diffraction angles only up to 45" (28 = 90"). In general, either rotated or stationary solid samples were used, because it later evolved that many samples were susceptible to transformation resulting from the cold working involved in pulverizing. It was also noted that rotating a sample sometimes suppressed weak lines, so, usually, samples were not rotated. Later a high temperature camera was adapted for use at temperatures up to 1400 °C on the Norelco unit utilizing both solid and powdered samples. Unless otherwise noted, all X-ray patterns were made with a chromium tube and a vanadium filter. Further-

Jan 1, 1952

By J. J. Gilman

Examination showed that characteristic cleavage step patterns are observed on the cleavage surfaces of undeformed, slipped, bent, twinned, compressed, and indented zinc crystals; and the effect of temperature is discussed. Dimples were seen to produce cleavage steps in a treelike pattern in otherwise undeformed crystals. The steps seem to originate when cracks intersect screw dislocations. IT has been known for a long time that the path of fracture in polycrystals may be discontinuous (see Jaffe, Reed, and Mannl for review). Recently, Kies, Sullivan, and Irwin2 have proposed, and given evidence, that crack propagation is discontinuous within individual crystals as well. Other evidence has been given by Low.' When discontinuous cracks within a crystal join together to make a macrocrack, the lamellae between each set of two cracks are torn somewhere, forming small cliffs. These cliffs appear as lines when the cleavage surface is observed microscopically.4,5 The lines have been called vein, tree, and riverlike markings by various authors, and they have sometimes been mistaken for fissures. The descriptive term cleavage steps is used in this paper. Cleavage steps vary in height over a wide range of values, from molecular dimensionsG to lor. and larger. Kies, Sullivan, and Irwin,2 as well as George,' have shown that the gross cleavage step patterns for plastics, polycrystalline metals, and for mono-crystals are sometimes similar. Thus, they depend mostly on the mechanical variables that prevail during cleavage and are relatively insensitive to the structure of the material. For example, parabolic markings2,7,8 sometimes result when cracks open up ahead of, and not coplanar with, the main crack front. If the advance crack has the same velocity as the main crack, their intersection line is a parabola, otherwise it is a hyperbola or an ellipse. The patterns are strongly affected by differences in crack velocities. This results in chevron patterns which point to the place of origin of the main crack. It is the purpose of this paper to demonstrate the existence of a mechanism of cleavage step formation which is a continuous rather than a discontinuous process. Also, certain characteristic step patterns are described, and the strong effect of temperature is shown. The specimens were zinc monocrystals (grown from 99.999+ pct pure metal). These were cleaved at room temperature and at — 196°C. Results and Discussion Cleavage step patterns are highly variable from point to point on a given specimen, as well as from one specimen to another. Although the patterns shown in the photographs are typical, they have been selected for graphic illustration. Figs. la and lb compare undeformed crystals that were cleaved at —196 °C and room temperature, respectively. Cleavage at room temperature (Fig. lb) resulted in a higher density of high steps (dark black lines) and enhanced the visibility of the fine background markings. Deformation by simple slip caused no marked change in the step patterns until the glide strain reached about 1.0. But, as Fig. lc shows, the density of high cleavage steps was greatly increased by large glide strains. Corrugations lying perpendicular to the slip direction may also be seen in Fig. lc. These are caused by deformation bands. The cleavage resistance of the crystal of Fig. lc was very high compared to undeformed crystals (estimated by the force on a needle required for cleavage). Striking and varied cleavage step patterns were observed on bent crystals. Two characteristic patterns that were observed on crystals bent at 25°C, and cleaved by reverse bending at —196°C, are shown in Figs. 2a and 2b. The first, Fig. 2a, consists of V-shaped lines similar to the parabolas of other materials2,7 Fig. 2b shows a pattern that is the equivalent of Fig. la, consisting of faint background lines with a few higher step markings. Cleavage of bent crystals at room temperature resulted in Figs. 2c and 2d. Now, the cleavage step lines show a strong tendency to follow one of two perpendicular paths. In Fig. 2c (bent once), many of the cleavage step components that lie parallel to the bend axis are assembled into irregular lines. In Fig. 2d (bent twice), the cleavage steps again tend to consist of two perpendicular components, but neither of the components is assembled into lines. Also, the step density is higher.

Jan 1, 1956

By A. E. Dwight, P. A. Beck

A search was made for ordered phases of AB3 composition in various combinations of Ti- and V-group elements with CO-and Ni-group elements. Three new TiNi3-type phases ZrPd3, HfPd3, and HfPt3 were found. In addition, twelve new Cu3Au-type phases TiRh3, Tilr3, ZrRh3, Zrlr3, HfRh3, Hflr3, VRh3, VIr3, CbRh3, Cblr3, TaRh3, and TaIr3 were also found. The lattice contraction for all three structures considered increases with increasing relative size of the A atom, suggesting that it is mostly the A atom that cont~acts. It was noted that, for the same radius ratio the lattice contraction is larger when the A atom is Ti than if it is Cb OY Ta. It is well known that elements or disordered solid solutions, which have structures with spherical close-packing, may be described in terms of various stacking schemes of close-packed layers. Stacking of the type abab corresponds to the hexagonal close-packed structure, and stacking of the type abcabc corresponds to the face-centered-cubic structure. In binary alloys of the first, second, and third long period transition elements various AB3 phases occur with close-packed ordered structures. It was observed that these ordered structures also may be described by suitable stacking of close-packed ordered layers, Fig. 1, as first shown by Laves and Wallbaum for TiNi3.' It is clear that the units of translation in such a close-packed layer consist of four atoms, one of which (black circle) is atom A and the three others (empty circles) are atoms B. Because of the alternation of A and B atoms in every second row of each layer, the unit-cell dimension in the layer is equal to the sum of the diameters of atoms A and B: a0 = 2(rA + rB). Stacking of such layers in the sequence abab results in a hexagonal close-packed structure of the MgCd, type. The unit-cell dimension in the basal plane is approximately double of that for the disordered hexagonal close-packed structure, but in the direction of the hexagonal axis it is about the same, so that 2co/uo = 1.63 in case of spherical close-packing. Ordered hexagonal close-packed structures of this type were found to occur in MoCo3,2 WCo3,, and UPt, .4 When the ordered layers shown in Fig. 1 are stacked according to the scheme abcabc, the resulting ordered structure is of the AuCu, type (cubic). Structures of this type were found to occur in TiPt3,5 CrPt,,' CrIr,,' and URu, .4 Finally, the stacking scheme abacabac of the same ordered layers results in a hexagonal close-packed structure of the TiNi3 type.' In this case, the unit-cell dimension is doubled not only in the basal plane, but also in the direction of the hexagonal axis, so that co/ao = 1.63. Structures of this type were found to occur also in TiPd3,5 ZrPt,," and UPd,.4

Jan 1, 1960

By C. Chiou, D. P. Seraphim

The effect of clustering on the superconducting properties of Al-Zn alloys has been studied by the ballistic induction techniques. The superconducting critical temperature, T,, changes when zinc-rich clusters are formed. The change in T, depends on the concentration and may be positive or negative, but always correlates with an effective decrease in the solute concentration of the alloy. Reversion after clustering effectively returns a 5 at. pct Zn alloy to the quenched state. These methods might be use&lly extended to the Al-Mg system, where X-ray data are difficult to obtain. In very dilute Al-Zn alloys, as in other tin,' indium, aluminum,2 or tantalums base solid solutions, there is a decrease in the superconducting critical temperature, Tc, roughly proportional to the temperature independent resistivity, i.e., -0.2°K per pS2cm. With increasing solute concentration, the T, generally reaches a minimum and then increases4 to a value well above that for the pure solvent. The change of T, with solute concentration follows an empirical formulation involving two constants which correlate with the scattering cross section of the impurities. As will be shown below, the Al-Zn alloys more concentrated than 1or2at. pct do not follow the empiricism, probably because they are unstable even when rapidly quenched.5 However, it seems possible to study at least the later stages of solute clustering by precise measurements of the 1', following various thermal treatments. A few experiments on quenched alloys, aged alloys, and tests on reversion are reported here. The heat treatments selected for obtaining the quenched and aged states were mainly deduced from the work of Dash and Fine.6 For the quenched state, specimens were homogenized at 480°C, quenched to -160°C, and immediately transferred and mounted in liquid nitrogen. Measurements on the superconducting properties and residual resistivities were then made at helium temperatures, after which the specimens were aged at room temperature for 3 to 5 hr, i.e., much longer than the time required for clustering to be essentially completed. The superconducting properties and residual resistivities were once again examined. To reach the reverted state of the A1-5at. pct Zn alloy, the aged specimen was heated for 5 min at 200°C and then tested as above.

Jan 1, 1963

By B. H. Kear

Clustering of slip bands in different systems has been observed in both ordered and disordered Cu3Au crystals compressed in the <110> orientation. Certain features of the slip band patterns are explained in terms of reactions between dislocations with Burgers vectors at 120 deg; in a few rare cases, the slip band evidence suggests that the Products of such reactions form diffuse tilt boundaries between the differently slipped regions of the crystal, similar to the known behavior of ionic crystals, but not nearly so well defined. WHEN crystals with orientations favorable for multiple slip undergo plastic deformation, slip may occur in different systems in localized regions, so that clusters of slip bands are developed. This phenomenon has been observed in fee1-4 metals and also in ionic crystals having the sodium chloride structure.5-8 In the ionic crystals, the boundaries formed between the differently slipped regions of the crystal assume characteristic orientations and are associated with macroscopic tilts that are plainly visible on the sides of the crystal. Recently, this interesting feature was explained in terms of the formation of diffuse low-angle boundaries as a result of interactions between families of glide dislocations in the different systems meeting in the boundaries.9 This paper describes an attempt to obtain similar evidence for boundary formation in an fee crystal. The alloy Cu3Au was chosen for the present study simply because this material is currently being used in a general study of differences in deformation behavior of the ordered and disordered phases of this alloy, which will be described elsewhere. EXPERIMENTAL PROCEDURE The two specially oriented compression specimens used in the present study were prepared from a melt-grown single crystal by careful cutting, grinding and polishing. The finished specimens were about 6 mm by 2 mm in cross section by 6 mm tall, with their ends parallel to (101) and sides parallel to (111) and (121); the broad faces of the specimens were made parallel to (711). Both specimens were annealed in an evacuated quartz tube at 850oC for several hours, slowly cooled to 500°c, and then quenched into water in order to obtain the disordered state for the alloy. One of the specimens was partially ordered (with S- 0.8) by reannealing at 350oC for 100 hr. After these annealing treatments, both of the specimens were electropolished. 10 Finally, the specimens were compressed in the [loll orientation and their stress-strain curves were recorded. The deformation was carried out in increments of a few percent strain. At each stage in the deformation the specimens were unloaded, removed from the compression jig, and examined optically for slip bands. Before reloading, the specimens were repolished, so that the observed slip band patterns were characteristic of a particular stage in the deformation. EXPERIMENTAL RESULTS Fig. 1 show!; the stress-strain curves for the ordered and disordered specimens and also indicates the stages 1, 2, and 3 where the deformation was interrupted for examination of slip bands. The curves show marked differences in flow stress and in work hardening rate in the two specimens, as observed by previous workers.11,12 Also, it was found, in agreemement with the observations of Taoka and Sakata,&apos;3 that the plastically deformed disordered alloy showed the characteristic coarse slip bands of alloys, whereas the ordered alloy exhibited a fine slip band structure similar to that of pure metals. Slip Band Observations. In what follows, the description of the slip band patterns, with accom-

Jan 1, 1962

By M. G. Nicholas

The influence of deformation, relative movement and surface cleanliness on the strength of room -temperature "interparticle" bonds between copper surfaces has been studied. No bonding resulted when clean surfaces were merely pressed together, even under pressures of 50,000 psi. Bonding resulted, however, when clean surfaces were simultaneously pressed and moved relative to each other, some bonding occurring no matter how small the pressure or relative movement. Exposure of the clean surfaces to various chemicals inhibited bonding, none at all being produced by small relative movements. The size of the movements which first produce bonding between exposed surfaces depends on the nature of the chemical, being less than 0.002 in. for cleansing reagents and greater than 0.004 in. for lubricants, water and hydrogen sulfide. It is concluded from the experimental data that the prime requirement for bonding is the dispersal of surface films such as oxides and adsorbed contaminants. THE formation of interparticle bonds during the room-temperature compaction of powder metallurgy components greatly increases the ease of handling in subsequent production operations prior to sintering. The strength of unsintered compacts follows the same general trends as their density and hardness, but quantitative data are extremely Scarce. Quantitative correlation of the data on the overall strength of a compact with that of the individual interparticle bonds is virtually impossible due to the extremely complex geometry of powder compacts. Qualitatively, however, the strengths of unsintered compacts can be correlated with variations in parameters known to influence the extent of cold bonding, and hence interparticle bonding, such as surface cleanliness, relative movement, and deformation of the mating surfaces. Kuzmic and seelig4 found that contamination of powder particle surfaces with lubricants reduced the strength of unsintered iron and copper compacts. Interparticle shear has been shown to promote bonding by Train and Hersey7 while Gregory has shown that die rotation is beneficial. Eilender and schwalbe6 concluded that most bonding occurred during the early stages of pressing iron powder compacts when interparticle shear is one of the principal processes occurring. mile there can be little doubt that these and other factors, such as particle size and shape, affect the extent of interparticle bonding produced by compaction, evaluation of the influence of individual factors is not easy. Particle deformation and relative movement, for example, occur simultaneously during compaction, making any precise measurement of their relative influences difficult. Because of these and other difficulties encountered in studying interparticle bonding in actual powder compacts, the approach adopted in this study was to

Jan 1, 1963

By C. H. Li, R. J. Stokes, T. S. Liu

Poly crystalline silver chloride specimens having different micro structures were prepared by extruding monocrystals or precompressed powder at varying temperatures. Extrusion at high temperature (370°C) produced a Fully recrystallized equi-axed material which was brittle and fractured by simple cleavage below its transition temperature (60°C for notched specimens). Extrusion at low temperatures (down to -196°C) produced an extremely tough cold-worked material with grains elongated in the extrusion direction. This material fractured below its transition temperature (which varied from —20" to —200°C) by an intergranular mode. It is concluded that cold work increases the resistance to transgranular cleavage. Fracture is then de-flected to propagate intergranularly ouer surfaces making an angle with the tensile axis with a corresponding improvement in notch toughness. This deflection is greater for extruded powder than for extruded monocrystals because of the extremely elongated grain shape. PREVIOUS work1,&apos; on the fracture properties of fully recrystallized silver chloride under notch impact conditions has indicated that the fracture behavior is rather invariant. Thus polycrystalline materials originating from a number of sources ranging from relatively pure Harshaw single crystals to commercial-purity powder and prepared by hot extrusion at 370°C all fracture by simple cleavage with a ductile-brittle transition at 60°C, see Fig. 1. This behavior is also typical for fully annealed single crystals and appears to be fundamental to silver chloride for the particular loading conditions and notch configuration adopted. One way in which this characteristic fracture behavior can be modified is by the dispersion of alumina particles in the silver chloride matrix.&apos; Then, particularly when the alumina particles are in the shape of spheres, the duc tile-brittle transition is much less sharply defined and occurs over a range of temperature, the transition temperature is lowered considerably and the fracture mode changes from simple cleavage to a fibrous or shear mode. The reason for this change is believed to be partly due to the nucleation of cavities in the region of the dispersed alumina particles which influences the state of stress locally and the distribution of plastic constraint at the root of the notch. It is emphasized however that this is not the sole factor responsible for the increase in toughness. In extruded materials it is reasonable to expect that the resultant grain size, texture, and the residual cold work will be influenced by the presence of a dispersed phase. The present experiments were initially designed to determine the extent to which cold work influenced the fundamental ductile-brittle transition temperature and fracture mode of silver chloride. They were stimulated by the discovery that silver chloride could be extruded at temperatures as low as -25°C when commercial-purity powder was used and as low as -196°C when single crystals were used. This working temperature was far below that previously used (370°C) and it was found that such extruded material was exceedingly tough. The reasons for the enhanced toughness and its relation to the previous work on silver chloride-alumina compacts will be discussed. EXPERIMENTAL PROCEDURE Materials Used. Two forms of silver chloride were used in this study: i) silver chloride powder

Jan 1, 1964

By P. K. Koh, G. C. Dunn

Si-Fe single crystals in a number of selected orientations were cold rolled 70 pct and analyzed to obtain quantitative (110) pole figures. Stable end orientations were ddetermined, and the effect of orientation on deformation behavior was investigated. WHEN single crystals are cold rolled sufficiently, they rotate into end orientations that are stable or unchanged in position with additional deformation. The end orientations of heavily cold-rolled iron crystals, positioned in holes in copper to limit lateral flow and to simulate polycrystalline conditions during deformation, were determined by Barrett and Levenson.&apos; They found that the sum of the end orientations, a pole figure for all the crystals studied, reproduced the essential features of the cold-rolled texture of polycrystalline iron. In. an investigation of Fe-Si alloys, Barrett, Ansel, and Mehl&apos; observed little effect of silicon content on the textures of heavily cold-rolled samples. Since alloys with less than 4 pct Si are known to have the same slip systems as iron," it seems logical to expect that they would also have the same end orientations as those for iron. Whereas Barrett and Levenson in their study of single crystals were interested in the problem of the origin of the cold-rolled texture of polycrystalline iron, the present authors are more interested in the final cold-rolled textures of single crystals of Si-Fe from the point of view of their relationship to subsequent recrystallization phenomena which will be reported in another paper. Nevertheless, such points as the tendency of an end orientation not to rotate (stability), the tendency of a pole concentration to spread out, and the tendency of the crystal to widen during cold rolling become interesting separately when the initial orientations are selected already in end orientations. Also the initial orientations having a [I101 axis parallel with the transverse direction, which include the two end orientations (111) [Ti21 and (001) [110], comprise a series that needs further clarification regarding the way such crystals rotate during deformation. Finally the present use of quantitative pole figure methods* represents an advance over the X-ray photographic method employed by Barrett and Levenson in their work on end orientations. The present investigation provides information on the cold-rolling behavior of two groups of Si-Fe crystals: 1-—crystals initially oriented in recognized end orientations for iron crystals with a [I101 direction parallel with the rolling direction, and 2— crystals initially oriented with a [I101 direction 90" to the rolling direction and in the rolling plane. For the second group, there is a 90" range between the (001) [110] orientation and the (110) [001] orientation. Since the 35" portion of this range from (110) [OOl] to (111) Di2] was included in a previous investigation," only the remaining 55" portion from (111) fii2] to (001) [I101 needs further consideration. Also, since both (111) [xi21 and (001) [I101 are end orientations for iron crystals,&apos; there might be a critical orientation* or orientation range be- • W. W. Martin and C. G. Dunno noted that (112) [Till was such a critical orientation. tween them such that early in the deformation an orientation within this critical range splits into two orientations. Upon further deformation, these two orientations rotate to produce a final texture consisting of (001) [1101 and (111) [?i2] components. The end orientations obtained are discussed in connection with the work of Barrett and Levenson, while the observed tendency for lateral flow of crystals in these orientations is treated in terms of the theory of Hibbard and Yen.&apos; Lateral flow according to the theory increases with deviation of the plane containing the rolling direction and the slip direction from a position perpendicular to the plane of rolling. Experimental Procedure Single crystals in sheet form with preselected orientations were prepared by methods that have been described elsewhere." Three lots designated A, B,

Jan 1, 1956

By John K. Kennedy

ThE basic equations describing continuous zone-melting via the zone-void method were reported by Pfann. Using these equations, a theoretical study was undertaken in this laboratory on the effect of the variation of the significant parameters of the zone-void system. These parameters are the zone length, 1, the heater length, h, and the effective distribution coefficient, k. The object of this study was to determine the effect of these parameters on and (where a = Cp/Cf and ß = Cw/Cf, Cp, C,, and Cf being, respectively, the solute concentration in product, waste, and feed) so that, in future application of this technique to the ultrapurification of materials, the effect of 1 and h would be predictable regardless of the value of k. The basic assumptions made by pfannl were, for the purposes of this study, assumed to hold. For k < 1, the steady-state solute distribution in the stripping section and the shift in this distribution brought about by passage of a void is shown in Fig. 7 of Ref. 1. In this section the definition of k and ß precludes the possibility of ß assuming values less than unity when k < 1. In the course of this investigation it was found that computations based on Pfann&apos;s equations gave values of ß less than unity in the region 1 > k > l/h for almost all values of Ls, where Further computations showed that 6 assumes negative values when k > 1 for many values of L,, 1, and h. A check of the derivation of the stripping-section equations showed that an error had been

Jan 1, 1964

By B. Ramaswami, Y. Nakada, U. F. Kocks

Compression tests were performed on single crystals of aluminum, silver, gold, and a Ag-10 pet Au alloy, with height-to-width ratios from 8:1 to 1:1. As this ratio decreased, so did the amount of easy glide until, for cubic specimens of all the pure materials, it was entirely absent. Heterogeneity of deformation is proposed to he a major contribution to the explanation of easy glide in slender specimens. For comparison, one single-crystal cube of zinc was tested and showed a behavior very similar to that in tension. Beyond easy glide, tension and compression curves are very similar. It is concluded that, in this range, neither grip effects nor orientation change influence the work-hardening characteristics to any significant extent. (121) and (100) crystals of all materials had similar stress-strain curves in tension and compression, except that the stress level of the flat portion of aluminum (109 crystals at high strains mas somewhat higher in compression. These results are at variance with previously proposed mechanisms of orientation dependence by both Hosford and Kocks. The compression tests themselves are described in detail and it is shown that they are essentially frictionless and produce single slip when the crystal is suitably oriented. DEFORMATION in uniaxial compression does not necessarily give the same result as deformation in uniaxial tension, essentially because the boundary conditions chosen are practically always different. 1) The shape and size of the specimens are usually different, and any genuine effect of size on work hardening should thus influence the result. 2) The grip constraints are usually different and can vary substantially for different kinds of tests. If the compression specimen is a thin plate and the friction on the platen is large, the grip contraints are more serious than in tension and lead, in the extreme, to uniaxial strain rather than uniaxial stress. If the specimen is roughly cubical and the friction is low, the specimen is less constrained than in tension inasmuch as displacements in the contact face are allowed. 3) The orientation change of single crystals is of opposite sign in tension and in compression; it is furthermore different in detail, because the relevant orientation in a squat specimen is the normal to the contact face, whereas that in a slender specimen is the generator of the side faces.&apos; The considerable dependence of work hardening on the initial orientation in tension would suggest that the orientation change should also be very important. 4) Finally, localized deformation instabilities found in tension tests, such as necking and Lüders band propagation, are not possible in squat specimens. This is, for some investigations, a favorable consequence of the large grip influence. Even if one is not interested in any of the effects mentioned above, compression tests are often desirable just because the specimens needed are so much smaller and sturdier. For example, they have been very helpful in the investigation of latent hardening by means of alternate longitudinal and lateral compression of cubes,* and for testing

Jan 1, 1964