Search Documents

By D. G. Westlake, D. T. Peterson

The saturation solubility of thorium dihydride in thorium was studied by saturation of samples and subsequent analysis. The solubility increased from about 1 at. pct at 300°C to above 20 at. pct at 800 °C. Over this temperature range the log of the solubility was a linear function of the reciprocal absolute temperature for thorium of good purity. The heat of solution of thorium dihydride was found to be 7.9 kcal per atom of hydrogen in crystal bar thorium and 6.6 kcal per atom of hydrogen in Ames thorium, The important effect of hydrogen on the physical and mechanical properties of many metals is receiving increasing recognition. The basic physical and chemical data necessary to help explain the observed effects of hydrogen on the mechanical proper ties of metals are all too often not available. The solubility of hydrogen in thorium metal was determined to provide a portion of this basic data for the thorium-hydrogen system. The absorption of hydrogen by thorium metal and the formation of thorium hydrides was studied by several investigators, beginning in 1891, and their work is summarized in Gmelin's Handbuch.1 No effort was made to determine the extent of solid solubility in the metal in these early investigations and the impure metal available made interpretation of the results uncertain. Nottorf2 reported the results of a careful study of pressure-composition isotherms for the thorium-hydrogen system. The principal interest in this work was in the composition and dissociation pressures of the thorium hydrides. Mallett and campbel13 reported the pressure-composition isotherms for the thorium-hydrogen system up to the composition of thorium dihydride. The saturation solubility of thorium dihydride in thorium was reported to be 13 at. pct at 650°C and 23 at. pct at about 900° C. Mallett and Campbell used thorium metal which had been prepared by the calcium reduction of thorium tetra-fluoride or thorium oxide. The results obtained with thorium of these two types agreed within the experimental error. PROCEDURE AND MATERIALS Procedure—The method used to determine the solubility was to saturate a sample of thorium with hydrogen and analyze a portion of the saturated sample. This method was chosen rather than a determination of pressure-composition isotherms since it was much less time consuming and could be applied at temperatures at which the equilibrium hydrogen pressure was too low to measure. The

Jan 1, 1960

By L. M. Pidgeon, K. T. Aust

There has been considerable interest in the possible use of titanium in magnesium alloys.' Zirconium has shown some promise in this connection2 and its general similarity with titanium suggests that the latter might act in a similar manner. A literature survey revealed that quantitative data on the Mg-Ti system was unavailable. Several patents3 have claimed that titanium additions from 0.2 to 4 pct to magnesium alloys were possible, but no mention was made as to the form in which the titanium existed in the alloy. Kro114 succeeded in introducing only traces of titanium into magnesium by bubbling TiCl4 through the metal under argon or by reacting it with sodium titanium fluoride. The application of theoretical data given by Carapella5 based on Hume-Rothery's principles, involving atomic size factor, crystal structure, valency and the electro-chemical factor, suggests that a Mg-Ti alloy is a favorable case, and the system appeared to warrant experimental examination. Experimental Procedure and Results THERMAL ANALYSIS If titanium is appreciably soluble in magnesium, a change in the melting point of the magnesium might be detectable using standard cooling curve methods. Magnesium was melted in graphite crucibles under an argon atmosphere, the assembly being enclosed in a silica tube. Graphite thermocouple protection tubes served also to stir the melts. The apparatus was very similar to Fig 1, with the addition of a refractory and baffle system to prevent undue heat losses from the top of the crucible. Chromel-alumel thermocouples were calibrated using Al of 99.97 pct purity. Dominion Magnesium Limited sup- plied redistilled high purity magnesium of the analysis given above. Titanium was added in three different forms: 1. Titanium powder —100 mesh, from the Titanium Alloy Manufacturing Co., Niagara Falls, N. Y. 2. Sheet titanium from the U.S. Bureau of Mines, produced by Mg reduction of TiCl4. 3. Magnesium —50 pct titanium master alloy from Metal Hydrides Inc., Beverly, Mass. The melting point of the high purity magnesium used was measured experimentally as 651.0°C. More than a dozen tests were conducted using titanium from the three sources referred to above, in calculated additions up to 20 pct titanium, at temperatures between the melting point and 1000°C and holding periods up to 6 hr. In no case was evidence obtained of solubility of titanium in magnesium, using inverse-rate and time-temperature curves. The melting point of the magnesium was unchanged within the accuracy of measurement, namely -+0.5°C; and no other thermal arrests were detected. Metallographic investigation of the thermal analysis billets indicated that the titanium additions were apparently mechanically entrapped in the magnesium in segregated areas. Consequently, these samples were not analyzed for titanium. The master alloy proved to be a mechanical mixture of titanium particles in a magne- sium matrix. These results indicated that the titanium solubility, if such existed, could not be obtained by the usual thermal methods. X RAY DIFFRACTION INVESTIGATION In an effort to detect solubility of titanium in magnesium, samples were investigated using both the Debye-Scherrer and the Focusing Back-Reflection methods. Filings from samples of the thermal analysis billets and from pure magnesium were annealed in argon one hour at 350°C to relieve mechanical strain. Measurements made of the interplanar spacings showed no difference between the Mg-Ti samples and pure magnesium. The interplanar spacings could be measured to within 0.0002A, and the greatest variation found was 0.0004A, in the back-reflection method. The diffraction lines for magnesium were not shifted by the titanium additions indicating that the solid solubility of titanium in magnesium is of a very low order—less than 0.5 pct. From both diffraction methods, a d or interplanar spacing of 0.817A was obtained for the redistilled high purity magnesium. This latter value is not given in the standard X ray diffraction cards for magnesium metal or vacuum distilled magnesium. Theoretical calculations for a close-packed hexagonal space lattice for magnesium indicate that the planes {2134) should give a line which was found. The relative intensity for this reflection at 0.817A is slightly less than that at 0.870k for magnesium. SOLUBILITY OF TITANIUM IN LIQUID MAGNESIUM The Mg-Mn system was examined by Grogan and Haughton6 who were

Jan 1, 1950

By A. L. Mowry, John T. Norton

Isothermal sections of the pseudo-ternary carbide systems TiC-VC-ZrC, TaC-VC-ZrC, and NbC-VC-ZrC have been examined. The sections contain an inverted U shaped two-phase field whose maximum extent is 77 molecular pct for Tic, 64 pct for TaC, and 51 pct for NbC. lsoparametric lines were determined which show the curvature of the single-phase field and the direction of tie lines in the two-phase field. It was not possible to show any relationship between atom size and the extent of the two-phase field. REVIOUS work' on the solubility relationship of several pairs of refractory cubic monocarbides from the group Tic, ZrC, VC, NbC, and TaC has shown that, with the exception of the system VC-ZrC, they all form continuous series of solid solutions. It was demonstrated that the limited solubility in the case of VC-ZrC system was primarily a question of the size factor, the size difference between vanadium and zirconium being greater than the limiting value of about 15 pet. It is interesting to consider the results of combining these isomorphous carbides in groups of three and to determine the isothermal section of the pseudo-ternary diagram at, for instance, 2000°C, a temperature that will assure an equilibrium structure in a reasonable length of time. In the ternary systems where the three pairs form complete solid solution series, it is probable that the isothermal ternary section would be a single- phase field, although this is not necessarily the case. However, if one pair of carbides has a very limited solubility, then there must be a two-phase field of some extent, and it is the purpose of the present work to examine the extension of this field and to determine if it can be related to the nature of the third carbide. This condition is fulfilled if the VC-ZrC pair, which has only limited solubility, is combined with Tic, TaC, or NbC. The relative sizes of the metal atoms undoubtedly play a role in the extent of the single-phase field, and the values obtained from the pure carbides, assuming close-packing between metal and carbon atoms and a carbon atom diameter of 1.54A, are

Jan 1, 1952

By J. T. Norton, A. L. Mowry

The monocarbides of the A subgroup elements in the fourth and fifth group of the periodic table in addition to being hard and refractory are of special interest in that they are isomorphous in crystalline structure. They are cubic with a sodium chloride type structure in which the metal atoms are essentially close packed in a face-centered cubic arrangement with the carbon atoms placed in the interstices between. Interstitial structures of this close packed type were first investigated systematically by Egg1 and he gave the rule for their formation, stating that the radius ratio of the nonmetal to the metal atom should not exceed the value of 0.59. The carbides of interest are those of titanium and zirconium of the fourth group and vanadium, columbium and tantalum of the fifth group. Table 1 shows the radius ratio using the Goldschmidt radii for 12 coordination for the metal atoms and the diamond radius for the carbon atom. It will be noted that while there is considerable variation in the size of the metal atom, in all cases the ratio is smaller than the limit of 0.59 placed by Hägg. It has been known for some time that these cubic carbides are soluble in one another, at least to some extent or, in other words, the metal atoms can be replaced, one by another without destroying the stability of the structure. Since the stability of these close packed interstitial substances appears to depend more upon geometry than upon the exact chemical nature of the atoms involved, it is of interest to examine the possibilities of replacement in these carbides in some detail. Hume-Rothery2 has pointed out the importance of the difference in size of solute and solvent atom as a factor in limiting the solubility in simple binary solid solutions. Largely on an empirical basis, he states that if the difference in size between solvent and solute atom is more than 14-15 pct of the solvent atom, the range of solubility is very restricted. The atom size was based on the distance of closest approach in the elements involved. While there is some question as to how one should calculate the size of the metal atom in the carbide structures, reference to Table 1 will show that zirconium is the largest and vanadium the smallest of the group and that the difference is about 15 pct. The Ti-Zr difference is about 9 pct and the others are smaller. Thus one would predict that if the size factor controls the solubility, all of the pairs except VC-ZrC would have wide or complete solubility whereas this latter pair is on the border line and might have restricted solubility. The purpose of the present investi- gation was to examine the solubility of the several pairs of carbides by heating them together until equilibrium was established and then examining the product by X rays. Previous Work Agte3 and his associates prepared various transition metal carbides and determined the melting points of binary mixtures. He concluded from the shapes of the melting point curves that there was extensive solubility in the case of the cubic carbides. Umanskii and his colleagues made an investigation of a number of pairs of the cubic carbides, using X rays and plotted lattice parameter vs. composition curves for the systems TaC-Tic, CbC-Tic, TaC-ZrC and CbC-ZrC. All pairs showed a continuous series of solid solutions. The first two pairs gave a linear relation while the latter two showed a negative deviation from Vegard's law. Kiefer and Nowotny, in a paper which became available after the present work was well advanced, investigated the binary pairs of the five cubic carbides by means of X rays. Relatively few points were obtained and results indicated that in some cases, at least, equilibrium was not reached at the temperatures used. The results indicated that solubility in the VC-ZrC system was not complete. All of the results of previous investigations indicated the desirability of a more detailed study. Materials The raw materials used were mono-carbides of titanium, zirconium, vanadium, columbium and tantalum and were the purest which could readily be obtained commercially. Spectrographic qualitative analysis showed that the CbC and TaC contained less than 1 pct

Jan 1, 1950

By J. T. Norton, A. L. Mowry

S. J. SINDEBAND*—(1) Discussing the properties of the powders used, Mr. Rostoker mentioned a silicon powder as being between 150 and 325 mesh. We always had much difficulty in measuring particle size of silicon powder by screen analysis, because of its tendency to agglomerate. I would be interested in the method used to measure these particle sizes. (2) Mr. Rostoker maintains that the measurement of magnetic properties is more accurate in determining complete homogenization of an alloy, than is the X ray diffraction technique. I wonder whether this is really true. Even when diffusion is complete, magnetic properties will change during continuation of the heat treatment, due to change in density and pore shapes, as has been demonstrated by Mr. Rostoker himself. Which effect is faster, the diffusion of the alloy components or the attainment of final density, will, in my opinion, depend very much on initial material and sintering conditions used. For magnetic applications, of course, the magnetic measurements are most suited for determining the point at which equilibrium conditions are reached. (3) I was glad to see that Mr. Rostoker had used sample sizes which are similar to those I had used previously, and I believe that ring samples of about 2 in. od will become more or less standard samples for the determination of magnetic properties of powder metallurgy products. I should like to ask how the density was changed for the samples on which the effect of porosity was investi- gated, and, further, why the sintering temperature used for this series of experiments was so low. It is somewhat surprising that a 24-hr treatment at 850°C acts so much more to spheroidize the pores than a treatment for 1 hr at 1100°C. But the effect seems to be indisputable. I have made an investigation of the influence of sintering temperature for longer times on the maximum permeabilities of iron powder compacts which was presented at the International Powder Metallurgy Conference in Graz, in 1948, and I hope that these results will be published here within a short time. The results of this investigation are confirmed by Mr. Rostoker's findings that the change of time from 1 to 24 hr changes the permeability behavior considerably in the direction as indicated by Polder and Van Santen for spheroidiza-tion of pores. I do not quite understand Mr. Rostoker's explanation of the influ-

Jan 1, 1950

By Allan Martin, R. A. Swalin

Diffusion rates of manganese, aluminum, titanium, and tungsten in nickel were measured at temperatures between 1100° and 1300°C. Activation energies, Q, and values of the frequency factor, Do, were calculated for each system. Values of Q were found to be within 15 pet of that for self-diffusion of nickel in agreement with present theories of diffusion. ONE of the most interesting problems in diffusion concerns factors governing the order of diffusion coefficients of various solutes in a common solvent metal at low solute concentrations. An early suggestion by Rhines and Mehl' was that diffusion coefficients might each approach that of the self-diffusion coefficient of the solvent itself. Later Thomas and Birchenall' pointed out correlations between the order of the diffusion coefficients and the freezing point depressions of the solvent by the various solutes. Other correlations have been sought. To settle this question it seems likely that much more data will be needed and, in particular, data collected at low solute concentrations. This type can be obtained by special analytical techniques as well as by use of radioactive tracers. The following study was made of diffusion of manganese, aluminum, titanium, and tungsten in a common solvent, nickel, at relatively low solute concentrations: 1—Production of alloys of nickel with manganese, aluminum, titanium. and tungsten at concentrations of the order of 1 atomic pet. 2—Use of pressure-welded diffusion couples in cylindrical form. 3—Diffusion anneals in the temperature range of 1100" to 1300°C. 4—Lathe sectioning of the diffused couples. 5—Spectrophotometric analyses of lathe turnings for the solute being studied. 6—Calculation of the diffusion coefficient for each of the solutes as a function of temperature. 7—Interpretation of results with current theories of diffusion. Experimental Work Materials—Two types of nickel were used in this research. The grade designated as Vacuum melted Mond nickel was used for all diffusion systems except the Mn-Ni system for which the Vacuum melted electrolytic nickel was used. Analyses of the two grades are given in Table I. The manganese used in making alloys was 99.9 + pet pure, aluminum 99.99 pet pure, titanium 99.9 pet pure, and tungsten of C.P. grade powder. Alloys used in this investigation were prepared by melting in vacuo or under a low pressure of argon. After melting, all heats except those of nickel-manganese were poured into a cylindrical copper mold 3 in. long and 1 in. ID. In the case of nickel-manganese, the alloys were solidified in the melting crucible. For nickel-aluminum heats alumina crucibles were used and for all other heats, zirconia crucibles.

Jan 1, 1957

By Roger A. Gregg, Barry J. Piearcey

MANY of the nickel-base superalloys developed recently for use in the as-cast condition exhibit a massive "white-etching" constituent1 in the inter-dendritic regions. Commercial alloys in this category are SM 200,* IN 100, INCO 713, and Nicrotung. The appearance of a similar microconstituent in nickel-rich Ni-A1-B alloys has been reported by Schijbel and Stadelmaier. Their investigation revealed that this constituent was the binary eutectic of the terminal nickel-base solid solution and the inter metallic based on Ni3A1. This binary eutectic is degenerate and NhA1 is the major phase. At lower temperatures the eutectic composition lies within the NkA1 phase field and under equilibrium conditions is single-phase NisAl. Taylor and F&apos;loyd3 have shown that in Ni-Al-Cr alloys containing at least 8 at. pct Cr, liquid compositions which solidify by the eutectic reaction in the nickel-rich portion, precipitate nickel-base solid solution and NiAl rather than nickel-base solid solution and NkA1 in the binary Ni-A1 system. Since the nickel-base alloys contain chromium in excess of 8 at. pct and are further alloyed with one or more of the elements titanium, cobalt, molybdenum, or tungsten, it is possible that the eutectic could contain NiAl rather than NisAl. In addition, Benson4 has reported that in low-nickel austenitic high-temperature alloys, which possess Ni/(Al + Ti) atomic-weight ratios <18.0, the Ni (Al, Ti) phase may precipitate from the melt during solidification. Since SM 200 and similar high-temperature materials were developed from alloys which were designed for solution treatment, the appearance of a eutectic would indicate that the limit of solid solubility has been exceeded. The purpose of this note is to confirm that the massive white constituent in the complex nickel-base alloys investigated is based on the binary eutectic Ni + NbAl and that the limit of solid solubility need not be exceeded to produce it, in common with other observations of cast alloys. Successive polishing of a sample of SM 200 showed that, in agreement with Schobel and Stadelmaier, particles of the constituent were spherulitic and in particular were a) situated between cored dendrite arms, Fig. 1, and b) nucleated at the surface of dendrites, Fig. 2. The Ni/(Al + Ti) atomic-weight ratios for the nickel-base alloys which were observed to contain the massive eutectic constituent in the interdendritic regions are 12.1 (Nicrotung), 9.2 (IN loo), 19.5 (INCO 713), and 14.5 (SM 200). However, the results of X-ray analysis of these alloys in the as-cast condition failed to reveal the presence of the Ni (Al, Ti) phase. An electron microanalyzer was therefore employed to analyze the composition of this constituent in SM 200. The results of analyses of representative particles are as follows5: 8.4 at. pct Ti, 8.8 at. pct Co, 5.5 at. pct Cr, 0.9 at. pct W, 12.8 at. pct Al, and 63.6 at. pct Ni. Assuming that

Jan 1, 1964

By F. Weinberg

The distribution of solute during the progressive solidification of dilute Sn-Ag alloys was determined in both solid and liquid as a jbnction of growth rate, rod diameter, temperature gradient, and solute concentration, using radioactive tracer techniques. The results indicated that partial mixing occurs in the liquid during freezing for most of the conditions investigated, approaching complete mixing for larger rod diameters (>2mm) and slow growth rates. Measured solute distributions were fitted to distribution functions based on diffusion controlled solute transport in the liquid state, by treating the normal solute diffusion constant and distribution coefficient as variable solute transport and segregation parameters. Values of these parameters were determined for the growth conditions investigated. Observations were made to determine whether mixing in the liquid was primarily due to volume changes in freezing or to differences in density between solute and solvent. The former effect was investigated using a selected Sn-Bi alloy as a solvent with zero volume change on fusion, and the density effect was studied using vertical solidifications of Sn-In and Sn-Ag alloys, systems having, respectively, very small and large differences in density between solute and solvent. The result indicate that mixing in the liquid cannot be primarily attributed to either volume changes or density differences. THE solute distribution in progressively solidified rods of dilute alloys, in which there is complete mixing in the liquid, is given by the expression1 where C, is the solute concentration in the solid at the point where the fraction g of the material has solidified, C, is the initial solute concentration and k is the equilibrium distribution coefficient. If it is assumed that there is no significant con-vective mixing in the liquid, then the solute distribution in the initial portion of a semi-infinite rod has been calculated to be2 where t is the time from the start of freezing and x1 is the distance in the liquid ahead of the solid-liquid interface at time t. Expressions [1] and [2] describe the limiting cases of solute distribution in the liquid, i.e., complete mixing, or no convective mixing; in practice it might be expected that the solute distribution would be somewhere between these two cases. pfann3 has suggested that Eq. [I] is applicable for partial mixing, as well as complete mixing, if an effective distribution coefficient k1 is used instead satisfactorily describes solute distributions in rods of magnesium, n-g, and b-sb6 alloys that were progressively solidified from the melt without mechanical mixing. The values of kl were observed to depend on the growth rate4-6 and the temperature gradient.6 The purpose of the present investigation was to measure solute distributions in both the solidified rods and the melt as a function of the solidification variables, and to determine to what extent the observed solute distributions could be described by the above theoretical expressions, keeping in mind that expressions [2] and [3] have been derived solely for the case of diffusion controlled solute transport in the liquid and are not theoretically applicable to

Jan 1, 1963

By K. G. Davis, P. Fryzuk

WEINBERG,&apos; in an investigation of solute distributions along unidirectionally solidified rods of dilute silver in tin alloys, concluded that, for rods of 2 mm diameter or larger, a very high degree of convective mixing takes place in the melt ahead of the advancing interface. Suggested causes for this mixing were (1) thermal convection, (2) convective currents set up by changes in fluid density associated with concentration gradients, and (3) currents set up by changes in volume during freezing. Evidence was presented that causes (2) and (3) were not primarily responsible for solute mixing. Further, convective mixing took place in horizontal rods with temperature gradients as low as 6°C per cm, putting doubt on the importance of thermal convection. In the present work, convective mixing is examined in melts with known initial solute distributions. The absence of a moving solid-liquid interface eliminates convection caused by volume changes on solidification. In the first experiment, three silica tubes 2 mm in diameter and approximately 20 cm long were filled with an alloy of 100 ppm Ag in tin (99.999 pct purity), placed adjacent to one another in a horizontal furnace, and solidified from one end at 1 cm per hr. A110, a Y emitter, was used as tracer. A few centimeters at the start of the rod were left un-melted, to determine the initial concentration Go. After solidification was complete, one sample was removed. The hot furnace was then brought back to

Jan 1, 1965

By J. J. Kramer, W. A. Tiller, G. F. Bolling

The temperatures of solid-liquid interfaces have been measured during the growth of Sn-Pb alloy crystals which exhibited a cellulur substructure. A simple descriptive theory was used in combination with this data and the other measurable growth variables to predict some features of the solute segregation such as the intercell excess concentration and the depth of cell grooves during freezing. The effective distribution coefficient at the tip of a cell, a, was predicted and experimentally found to vary in the range, 1 >a > k0. At the lower limit, the severe segregation that must arise between dendritic cells was substantiated by various observations such as the existence of a second Phase even for alloys of only 0.2 wt pct Pb. ThE cellular morphology of a solid-liquid interface is known to coincide with the formation of a zone of constitutional supercooling ahead of the interface during growth.&apos; The occurrence of various cell structures, (two-dimensional cells, regular cells, dendritic cells), has been quantitatively investigated,&apos;2,4 but the segregation of solute which arises as a consequence of this mode of growth has only been examined qualitatively. An indirect method of obtaining the solute concentration at the tip of a cell cap during solidification has been de-

Jan 1, 1963

By F. Weinberg

Measurements have been made of solute segregation during dendrilic growth by using radioactive solute elements and ,measuring the activity of den(12-ites cut from decanted specimens. This has been done for both lead awl tin based binary alloys contaitzing the following solute additions: Ag, T1204, was dependet on ko, the equilibrium distribution coefficient in the following way Fay k &apos;c 0.1, C/C 0.6; for k0 >0.1. 0.6 <c,/c,< I. Qualitative obse?-vations were madc of dendritic segregation, by using autoradiographic techniques, for the Sn + Ag110 and Sn + Tlo4 systems. The observation were found to he in general agreement with the measurements ofCA/Co. Autoradiographic were also obtained of scctiolccl delzr11-iie stalks. These indicated that the stalks had a substructure, dclileated by solute corzetlt?atio?zs nlolg the substructure walls. A new dendrite growth direction <JI2> is reported for tila. SOLUTE segregation in dilute binary alloys has been investigated by Pfann,&apos; Smith, Tiller, and Rut-ter,&apos; and others. They considered the case of a slowly advancing plane solid interface, and derived expressions for the distribution of solute in both solid and liquid during solidification. To determine these expressions, they assumed no diffusion in the solid and either complete mixing in the liquid:&apos; or diffusion controlled solute movements in the liquid without any convective mixing.&apos; The present investigation considers solute segregation during dendritic growth, in which case the solid-liquid interface is not plane, and the growth rates are rapid. Segregation under these growth conditions has not been treated mathematically, because of the relative complexity of the system. It has been suggested by Chalmer, on the basis of preliminary results, that an alternative to the diffusion and heat flow controlled conditions during growth is &apos;diffusionless" dendritic growth in which solid is formed with the same composition as the liquid. He suggests this type of growth may depend upon a solvent-solute relationship that permits some solid solubility without excessive increase in internal energy, as is the case for solutions of tin in lead. On the other hand, Montariol,4 and others, have shown experimentally that some segregation does occur during dendritic growth in metals using etching and radioactive tracer techniques to indicate the concentrations of the solute. The present investigation was undertaken to determine, both qualitatively and quantitatively, the extent of solute segregation associated with dendritic growth in a series of binary alloys, as a function of solute concentration. PROCEDURE The solvent materials used were Vulcan Electrolytic tin (99.997 pct purity) and Tadanac lead (99.998 pct purity). The solute materials were Zn, Sn, and Sb (better than 99.998 pct purity), Ag and Co (99.5 pct purity), and T1 (Fisher "purified" metal sticks). Activation of the solute metals was carried out in the reactor at Chalk River, Canada. Master alloys were prepared by induction heating from the radioactive solute metal and the pure solvent, under argon, in graphite crucibles. Pieces of these alloys were then added to the solvent to give the required solute concentration. Dendrites were grown in essentially the same manner as that described by Weinberg and Chalmer, , in which controlled orientation single crystals were grown dendritically in horizontal graphite boats, and the liquid decanted. The crystals were grown and decanted in an atmosphere of tank argon. Before decanting, a sample of the liquid was drawn up in a glass tube and allowed to solidify rapidly. The orientations of the single crystals were such that <loo> was parallel to the growth direction, and (100) in the horizontal plane for lead, and [1101 and (110) respectively for tin. With these orientations long dendrite stalks formed along the bottom of the boat in the dendrite direction (<100> for lead and [I101 for tin) from which secondary branches grew. Only these secondary branches, which grew freely in the liquid from the dendrite stalk to the liquid surface, were used in the measurements. Accordingly, effects due to substrates and oxides on the surface of the liquid need not be considered. In order to measure the solute concentration C, of the dendrites, individual dendrite stalks were cut from the decanted specimens, remelted, and formed

Jan 1, 1962

By E. W. Cawthorne, R. I. Jaffee, C. M. Craighead

SOLUTION of a solid metal or alloy in a molten metal bath is used daily in melting operations, extractive metallurgical processes, and in brazing. It is generally recognized that temperature, time, agitation, and the composition of the solute metal and the solvent bath all affect solution rate. However, there is little information available in the literature on the kinetics of the process or the relative effect that these variables have on solution rate. In connection with an aluminum brazing study, data on the solution of aluminum in an A1-Si* bath to promote rapid solidification. The chemical analyses were made on drillings taken from the cast plates. The results of this work are listed in Table I. The thermal-analysis data were obtained on 200 gram samples cooled at approximately 1.5" per min. A calibrated chromel-alumel couple was used. The thermal arrests were recorded autographically on a General Electric photoelectric potentiometer recorder. The instrument was calibrated to 22.5 pct. The thermal emf was bucked with a Leeds and Northrup semiprecision potentiometer so that 1 mv represented the full scale of the GE recorder. Referring to Table I, it will be noted that the analytical results tend to be lower than the intended compositions by about 0.3 pct Si in the higher ranges. In view of the known difficulties in obtaining accurate silicon determinations by wet analysis when the silicon content is high and in view of the care that was employed in weighing and preparing the alloys, the intended compositions are believed to be more accurate than the analyses. Furthermore, when the thermal-arrest data are plotted against the intended compositions, a smooth curve is obtained, as shown in Fig. 1. This is not the case when the data are plotted against the analytical results. It will be noted from Table I that the 12.5 pct Si alloy (melt 4) showed a primary arrest at 577.6"C and the eutectic at 576.l°C. The metallographic examination of this chill-cast plate sample showed a small amount of primary silicon. These results, Fig. 1, place the eutectic at slightly less than 12.5 pct Si and are in rather good agreement with the unpublished Hanford data. In all of the subsequent work, the composition of the A1-Si dipping baths was determined by thermal analysis (spread between the primary arrest and the eutectic temperature) employing 12.5 pct Si and 577.2 °C as the composition and temperature of the eutectic. Material Preparation and Selection of the Techniques: At the beginning of this work, it was anticipated that, owing to oxide film present on the surface of the aluminum, difficulty would be encountered in obtaining rapid uniform wetting of aluminum specimens dipped in the A1-Si bath. The following etchants were investigated as possible methods of providing a thin uniform oxide film on the specimen surface: 1—phosphoric acid (H³O4), 2—fluoboric acid (HBFI), 3—sodium hydroxide (NaOH), and 4—hydrofluoric acid (HF). Tests with sheet .specimens (2x½x0.06 in.) of 2s aluminum prepared with these etchants indicated that the fast uniform wetting required for this work could not be obtained if any oxide film existed on the specimen surface. Tests were then conducted by dipping similar specimens etched in 5 pct NaOH and given a thin

Jan 1, 1956

By W. A. Jr. Slippy, E. P. Dahlberg, R. B. Reed-Hill

A large room-temperature mechanical-hysteresis effect under cyclic tensile loading was observed in zivconium specimens prestrained at 77°K so as to form large numbers of (1121) twins. The observed hysteresis loss was a maximum when the prestrain was just under 1 pct It also depended strongly on the maximum applied cyclic stress, but only moderately on loading rate and temperature (-72°to 100°C). The phenotnena have also been studied in terms of the elastic aftereffect where the data has been found to cotlforln closely to the functional relationship between strain and time, c = -RT/a In tank ft. t to)/Zr, for strains cts large as about 10-4 This eyication can be derived directly from the strain-rate equation There may be important commercial implications to the present discovery since it appears that within rather wide limits it may be possible to obtain any desired magnitude of damping capacity in zirconium melal. The damping propevties may also be given directional characteristics. It is shown that the observed anelaslic phenomena can he explained on the assumiplion that they are the resuilt of stress-induced twin-boundary movements in which the average twin increases its thickness by only a few percent. In polycrystalline zirconium, a grain is least favorably oriented for slip when its basal plane, containing the (1130) slip directions, lies nearly perpendicular to the principle normal stress. Mechanical twinning is favored by this orientation and under simple tensile deformation at room temperature the primary twinning mode is {1012).&apos; The (1012) twinning shear is small (0.17) so that, during deformation, twin growth does not greatly enhance ductility. At 77°K, on the other hand, the predominant twinning mode shifts to {1121), and many thin (1121) twins can form at small strains.&apos; Twins thus nucleated can grow readily during subsequent room-temperature deformation and permit easy deformation in unfavorably oriented grains. In addition, the (1131) twinning shear is large: 0.63. These facts have lead to a process&apos; for greatly improving the room-temperature ductility of zirconium when it contains a sizable fraction of grains with basal planes nearly normal to the stress axis. Optimum results are obtained by slightly less than 1 pct prestrain. The effect is large so that a 50 pct elongation increase is readily attained. A large room-temperature mechanical-hysteresis effect has also been observed in zirconium specimens prestrained at 77°K. The present paper is concerned with some aspects of this effect which the experimental observations strongly indicate is primarily the result of strain-induced {1121) twin-boundary movements. This clearly suggests that mechanical twins formed by plastic deformation can cause anelastic effects similar to those previously observed2 for annealing twins and twins associated with phase transformations. EXPERIMENTAL PROCEDURES All specimens were prepared from arc-melted sponge zirconium plate stock as previously described,3 which has a preferred orientation with basal planes aligned parallel to and uniformly distributed around the rolling direction. Both lonti-tudinal and transverse tensile specimens, with axes parallel and transverse, respectively, to the plate rolling direction, were cut from this plate. In a longitudinal specimen most grains are favorably oriented for slip, while in a transverse specimen a large fraction are unfavorably oriented. Prestraining at 77°K was performed in two ways. In one, annealed zirconium plate stock was cooled in liquid nitrogen and then rolled in the transverse plate direction to strains varying from 0.5 to 12 pct. Cylindrical 1/8-in.-diameter by 1-1/4-in.-gage-length tensile specimens were machined from this material. In the other, previously machined tensile specimens were prestrained in tension at 77°K between 0.2 and 3.8 pct. Type FA 25-12 SR4 strain gages were cemented to all specimen gage sections. All tests were performed on an Instron testing machine of 10,000-lb capacity. EXPERIMENTAL RESULTS Fig. 1 shows a typical two-cycle room-temperature stress-strain diagram for a specimen prestrained 0.65 pct by rolling at 77°K. The maximum stress in both cycles was 14,500 psi. Note that in the first cycle the loading and unloading curves are unsymmetrical so that a residual strain, er, remains after unloading. In the second cycle the stress-strain curves form a nearly symmetrical hystere-

Jan 1, 1965

By E. S. Machlin

An extension of the thermodynamic theory has been made for the case of irreversible growth processes occurring by the motion of an interface. The theory is applicable to such diverse phenomena as diffusion, growth in recrystallization, continuous grain growth, growth of carbides, etc., growth of eutectoid products, growth in solidification, recovery, "slipless" flow, etc. THE publication of a recent book1 has served to focus attention on a very powerful means of treating certain irreversible processes. This method which has been formally described as the thermodynamic theory of irreversible processes is applicable to processes which involve an approach to equilibrium and for which the deviations from equilibrium are small. The irreversible processes can be classified into three groups: chemical reactions, transport processes (diffusion, heat and electrical flows), and relaxation phenomena involving a degradation of internal energy to more stable states. The limits of applicability of the theory can be precisely defined1,2 for each type of irreversible process. For example, in a chemical reaction at constant temperature and pressure, the Gibbs free energy released per mol in the process should be less than the thermal energy, RT. Another type of irreversible process, which sometimes involves a combination of the first two of the above-mentioned groups, is treated in this paper. This process is one characterized by the motion of an interface separating two regions having different values of free energy and may be briefly described as a growth process. Prigogine2 and Herring3 have treated special cases of this type of process. Previous conscious applications of the theory in metallurgy have been limited to the field of diffusion, one of the transport processes. Darken,4 Bardeen,5 Prigogine2 and others have made significant contributions in this respect. Some other applications of the theory are described in this paper. The Theory The thermodynamic theory of irreversible processes is based on the work of Onsager,6 DeDonder,7 Prigogine2 and others. A resume of the theory has been given by Prigogine2 and DeGroot.1 A brief description of the theory follows, although for a complete understanding the reader is urged to read the references. It is first assumed that the change in entropy, even for a system removed from equilibrium, is given by dS = [dU + pdV - Si µi, dn, - Sk Pk dxk]T-1 Using this expression, the irreversible entropy production in the system is calculated by subtracting the contribution to the change in entropy of the system by transfer of heat, work, or matter from the surroundings. Thus, for example, the irreversible entropy production of a system at constant temperature and pressure is given by dtS = — — dG (DeDonder7) Now it has been shown by Prigogine,2 DeDonder,7 and DeGroot1 that the rate of irreversible entropy production diS/dt, when calculated in the manner suggested, can be written as a sum of products of conjugate forces XK and fluxes Jk, i.e., d,s/ = Sk Jk Xk [I-1] dt The assumption is next made that a linear relation exists between a given flux and the forces, obtained from Eq. I-1, i.e., Ji = Sk Lik Xk (i = 1, 2, . . . n) [I-2] Generally, there is a degree of freedom in choice of the fluxes and forces. However, for all the choices consistent with Eq. I-1 the Onsager8 relations, based on the principle of microscopic reversibility, namely, Lik = Lki, are valid. The coefficients Lik, depending upon the choice of the conjugate fluxes and forces, may be more or less dependent upon the parameters defining the state of the system. In general, the proper choice of fluxes and forces utilizing Eq. I-1, makes the coefficients Lik independent of time. Also, because many of the three classes of irreversible phenomena have already been treated,1,2 a proper choice can be made by analogy. In any case, the validity of Eq. 1-2, from whatever choice made using Eq. I-1, must be tested by experimentation. The theory, thus, comprises the following steps: 1—The irreversible entropy production is calculated to yield Eq. I-1. 2—The terms in Eq. I-1 are grouped

Jan 1, 1954

By W. C. Hittinger, J. McGlassan, J. W. Peterson

THIS paper describes the result of an investigation of the production of thin alloyed layers on a thicker substrate of pure germanium as one step in the manufacture of transistors.&apos; The technique of alloying materials with semiconductors is a means of introducing small amounts of impurity elements in the alloyed regions of the crystal lattice of the semiconductor, the presence of which make possible the junction type transistor.2 Impurity elements are added which will produce regions having differing electrical resistivities and means of carrying current. Impurities are of two types: those which produce an excess of electrons (negative or n-type), and those which produce a deficiency of electrons (positive or p-type) in comparison to the pure semiconductor. The boundary between two such regions is designated a p-n junction. Elements in Group V of the periodic table, having five valence electrons, give rise to an excess of electrons when present substitutionally in the semiconductor lattice (usually germanium or silicon from Group IV). Similarly, elements in Group III create a deficiency of electrons. Germanium and silicon which are substantially free of impurities of either p or n-type are designated intrinsic (i-type) semiconductors. The usual junction ----trarnsistor has

Jan 1, 1958

By V. Balasubarmanian, Rajendra Kumar

ISOTHERMAL formation of martensite in a copper-aluminum-nickel alloy was previously reported by Hull and Garwood.&apos; In the present work an attempt has been made to investigate some of the character- istics of the formation of isothermal martensite in binary copper-aluminum alloys containing 14.00 and 14.50 pct Al, respectively. Chill-cast alloys were prepared from OFHC copper and aluminum of greater than 99.5 pct purity. They were homogenized for 24 hr at 950° C. Polished samples were soaked at 1000°C and quenched in water to retain the p phase untransformed; martensite did not form during the quench. The surface was then repolished for isothermal microscopic studies at room temperature. The salient features of the martensitic transformation in copper-aluminum alloys observed in this investigation are: a) The transformation progresses (i) by nuclea-tion at new sites and (ii) by simultaneous slow isothermal growth of existing martensite plates. Slow isothermal growth of martensite has previously been observed by Holden2 in an uranium-chromium alloy and by Hull and Garwood in copper-aluminum-nickel alloy. b) Martensite plates nucleate preferentially at the pin holes formed during the casting of the specimen, Fig. 1. c) Some of the martensite plates pass right through other plates lying in their path, Fig. 1. d) Martensite plates cross grain boundaries without any change in the direction of growth. e) An irregular network of markings, Fig. 2, appears simultaneously with the formation of martensite. It was less prominent in the case of 14.5 pct alloy. The markings may arise due to spontaneous deformation under the lattice strains of the martensitic transformation. Crossing of martensite plates was first observed by Kulin and cohen3 when a high-nickel steel was cooled to 77°K. They suggested that the rapidly advancing plates atquire considerable kinetic energy and strike the obstructing plate with a momentum so as to nucleate the process on the other side of the plate. This, however, fails to explain the present observation where the slow isothermally growing martensite plates cross each other. Such plates would not possess large momentum to bodily displace the obstructing plate. Fig. 1 shows the martensite plates crossing each other without body displacements. This should be contrasted with Fig. 3 of Kulin and Cohen&apos;s paper where crossing of plates is accompanied by shearing or body displacements of the plates. The present observations can, how-

Jan 1, 1961

By R. G. Treuting

Germanium single crystals strained in tension at 600°C slip on the {Ill} plane and, macroscopically at least, in the <110> direction. Deformation is in homogeneous: various localized rotations are observed, as is a banding consistent with secondary slip banding. The structure after deformation is polygonized with a domain size of about 2x1O-3 cm. In relaxation tests, an incubation period prior to flow is observed, of duration inversely related to temperature and applied stress. Under continuous loading, there is a sharp first yield point. A critical resolved shear stress therefore must be cited with respect both to temperature and to rate of loading. At 600°C, when loading proceeds at 2900 psi per min, it is 1310 psi. The yield point phenomenon is suggestive of Cottrell&apos;s solute atom atmosphere theory and five points of qualitative agreement with this theory are found. PLASTICITY of germanium at elevated temperatures has been reported by Gallagher,&apos; who ascribed it to slip on {Ill) planes and described some associated effects. Preliminary experiments with single crystals of silicon and germanium loaded as simple beams confirmed {ill) as the slip plane of both materials.V his was demonstrated by the standard technique employing traces on two surfaces of known angle on a deformed crystal of known orientation.3 These experiments did not fix the slip direction, since more than one glide system operated. Uniaxial tensile straining is preferred for this purpose in order to restrict slip to one predominant system and to provide an unequivocal determination of the orientation change with respect to the stress axis. Experimental Method Specimens were diamond-sawed from germanium crystals containing about 5x10-5 wt pct Sb prepared by the zone-leveling process described by Pfann and Olsen;4 these specimens measured about 81/2 in. long, 0.20 in. wide, and were of various thicknesses from 0.035 to 0.060 in. Prior to testing, a specimen was coated over 1 in. or more of length at each end with Apiezon wax, chemically polished (in a solution of 15 ml HF, 25 ml HNO,, 15 ml CH3COOH, and 3 to 4 drops Br2), and dewaxed in warm xylene. A brief re-etching after scribing through a complete wax coating with vernier dividers provided gage marks at 0.200 in. intervals on the polished surface. A specimen with grips soft soldered to the ends was passed through the furnace tube and the grips pinned to the tensile machine in series with a stress gage. The Schopper testing machine is best described by reference to the illustration on p. 107 of Schmid and Boas5 and is operated with the pendulum arm locked, fixing the upper head. Load is applied through the screw-driven lower head, manually or by reduced speed motor. The furnace is 21/2 in. long surrounding a 4x7/16 in. ID quartz tube and was positioned about the specimen to permit free motion. The inert or reducing atmosphere required was sufficiently provided by a flow of helium introduced at the center of the furnace tube, with the tube ends loosely plugged with asbestos paper. Surface corrosion was slight. The stress gage employs a Be-Cu strip with SR-4 units incorporated in a simple bridge network with provision for balancing and calibrating. The bridge output on loading is fed through a Leeds and North-rup dc amplifier to a Speedomax 10 mv recorder, and calibration is obtained by dead loading. Experiments of two types have been conducted. Most have been in relaxation at constant elongation. One stress-strain curve has been taken with the machine continuously driven. Following extension of a specimen, Laue X-ray photographs were taken to obtain the orientation change between deformed and undeformed regions, recorded on a single film by translating the specimen in a plane normal to the beam between exposures. Sequences of such multiple exposures were made on some specimens to include the entire orientation range on one film. Slip Direction To obtain a maximum orientation change with a single slip system, crystals of very closely <110> axial orientation were used. The classic determination of slip direction rests on establishing that direc-

Jan 1, 1956

By N. E. Brown, F. L. Versnyder

THE apparent dependency of the electrical characteristics of hexagonal crystalline selenium on microstructure has aroused much interest in microscopical studies of selenium. Microscopic observations on the crystallization of selenium have been made by Escoffery and Halperin,&apos; P. H. Keck,&apos; and other investigators. It is the purpose of this paper to discuss the microstructural changes observed on polished cross-sections of single layers of selenium after various heat treatments. Observations were also made on crystallization of the free-surface layer of these deposits. In general, all of the transformations studied were either transformations of the vitreous selenium to hexagonal selenium or micro-structural transformation of the hexagonal selenium itself. Procedure The selenium used in this work was obtained from the American Smelting and Refining Co. and was approximately 99.96 pct pure. An intentional impurity of 1 part per 2,000 of bromine was added to the material prior to evaporation. A thickness of approximately 0.002 in. of this selenium was vapor deposited on an aluminum base plate. The maximum plate temperature during the vacuum vapor deposition was 140°C. Mounting of the cross-sectional specimens for metallographic study could not be done in plastic mounting media, as is customary, since temperatures in excess of 50°C would cause unwanted transformations. Consequently, a simple clamp-type device was used to mount the specimens for preparation. All grinding operations were then done carefully by hand in order that the specimen not become heated during this operation. Wet polishing was done on the conventional metallographic polishing laps, using successively finer grinding powders. An extremely careful polish is necessary, since observation and micrography of the specimens are done in the unetched condition under polarized light. The two observations of crystallization made on the free surface of vitreous selenium deposits (Figs. 4 and 5) were made on surfaces which were perpendicular to the cross-sections studied. These free-surface layers were examined directly, i.e., no pre- vious metallographic preparation, as obtained from the vacuum vapor deposition. Microscopic Observations A study was made of polished cross-sections of the vitreous selenium as-deposited. It was noted that in all cases there was columnar crystallization adjacent to the base plate, which appeared to occur during the vacuum deposition process. This observation has also been made by Keck? It also was observed that vagrant spherulitic crystallization occurred in the vitreous selenium. The term "vagrant" is used, since these spherulitic grains appear to crystallize at random throughout the vitreous selenium during the vacuum deposition process. Columnar crystallization at the A1-Se interface and a typical spherulite observed in a polished cross-section of "as-deposited" vitreous selenium may be seen in Fig. 1. Cross-sectional samples of vitreous selenium studied after heat treating individual samples for 20 min in 10" steps from 80" to 220°C revealed that crystallization—in this case, columnar crystal growth —proceeds from the aluminum base plate to the surface of the specimen (Fig. 2). Crystallization was microscopically observed to be complete after the 130°C heat treatment. Visual examination of the free surface of the specimen after the 130 °C heat treatment revealed the readily recognizable grey appearance of the completely crystallized selenium, in corroboration of the microstructural observations. No microstructural transformations then appeared to take place between 130" and 190°C. At 190°C the beginning of recrystallization appeared and proceeded until the columnar grain structure had been completely transformed to equiaxed grains between 210" and 220°C (Fig. 3). Naturally, the grain size of the recrystallized grains at the lower temperatures (190" to 210°C) was smaller than is illustrated in Fig. 3. In addition, polished cross-sections of deposits heat treated at 140°C for 10 min to cause complete crystallization and, subsequently, heat treated in 10" steps from 80" to 220°C for 20 min were studied. As expected, no microstructural transformations took place until the beginning of recrystallization was observed at 190°C. A comparison with the previously studied specimens revealed that recrystallization proceeded almost identically in the two experiments although in the first case the deposits were vitreous prior to the series of heat treatments and in the second case they had been crystallized by a previous heat treatment. By heat treating for longer times (180 min) at lower temperatures, the

Jan 1, 1956

By B. L. Averbach, M. Cohen, C. H. Shih

The isothermal formation of martensite is studied in Fe-Ni-Mn and Fe-Mn-C alloys under conditions where the athermal transformation is completely avoided, there being no martensite present at the beginning of the isothermal reaction. The initial rate of martensite formation is extremely low under these conditions, but it increases greatly as soon as some martensite is formed. It appears that true initial characteristics of the isothermal reaction can be studied only in samples wherein the formation of martensite on cooling is completely eliminated. THE phenomenon of isothermal martensite formation is well established1 -" but the observations have been made in the presence of varying quantities of athermal martensite. Complete avoidance of athermal martensite and the formation of martensite under pure isothermal conditions have been reported by Kurdjumov and Maksimova2 in an Fe-Ni-Mn alloy (23 pct Ni, 3.4 pct Mn, and the balance Fe) and have been confirmed by recent experiments of Cech and Hollomon in a similar alloy. In each case, the isothermal transformation rate was a maximum at the beginning of the transformation and decreased as the reaction proceeded. These initial transformation rates were used by Fisher7 for the confirmation of calculated nucleation rates based on classical nucleation theory extended to include the effect of elastic strain energy. Cech and Hollomon have admitted, however, that up to 3 pct martensite formed on cooling in their samples and thus was present at the initiation of their isothermal runs. In this paper, experiments on similar alloys are described, except that care was taken to remove the incidental martensite developed at the surface prior to the isothermal reaction. It is shown here that the reaction rate at the beginning of the strictly isothermal reaction is very low. As soon as some detectable amount of martensite is produced, the transformation rate increases rapidly but ultimately decreases as the reaction proceeds. When martensite is allowed to form isothermally in the presence of about 2 pct of prior martensite, the apparent initial reaction rate is a maximum and corresponds to the observations of Kurdjumov and Maksimova and Cech and Hollomon. It is evident that the cooperative nature of the martensitic reaction results in a significant stimulating or autocatalytic effect on the isothermal reaction rate. Experimental Procedure The chemical compositions of the alloys used in this investigation are listed in Table I. The Fe-Ni-Mn alloys A and B were melted in an induction furnace, cast into ingots 2 in. diam by 4 in. long, and forged into % in. diam rods. The manganese steel (alloy C) was melted in an arc furnace and cast into ceramic molds 1/4 in. diam by 5 in. long. The rods were sealed in evacuated Vycor tubes and homogenized at 1175" to 1200°C for 50 hr. The long homogenization resulted in some manganese loss from the surface, producing a thin layer of transformed material. This surface layer was removed by grinding and the specimens were then swaged to 0.050 in. diam wire. Measurements of electrical resistance were made with a Kelvin double bridge to follow the course of the martensitic transformation. A small permanent magnet was also useful in scanning an entire specimen for small localized quantities of martensite. This test is almost as sensitive as the electrical resistance method in detecting the presence of martensite (which is ferromagnetic in these alloys) and it was especially useful in deciding whether a specimen was completely free of martensite. The wire specimens (0.050 in. diam by 2.5 in. long) were sealed in evacuated tubes and austenitized for 5 min at 1150°C. The specimens were quenched into water and, unless otherwise stated, stress-relieved for 1 hr at 650 °C in order to induce greater uniformity in the subsequent martensitic transformation. As dem-

Jan 1, 1956

By R. D. Carnahan, J. E. White

ThE use of strain gages in the measurement of microplastic behavior of materials is well-known.&apos;-3 Recently it has been suggested that similar techniques might be useful for determining stress relaation. It is the purpose of this report to comment on the use of strain gages in the measurement of strains in the vicinity of 10"6. We will also discuss observations of an anomalous anelastic behavior apparently associated with small temperature changes, which accompanies the unload portion of a load-unload microstrain test. The key to successful strain determinations at levels below 1 pin. per in. is a good gage-mounting technique and either precise ambient temperature control or a set of matched temperature-compensated gages. The importance of good mounting cannot be overemphasized, for example, as in the use of Eastman 910 contact cement. If the cement layer is not sufficiently thin and properly hardened, it is possible to observe anomalous negative stresses when small loads are removed, which relax back to zero in an orderly way with time. Such anomalous behavior, thought to be due to gage bonding, has been reported in microstrain experiments by Thomas and verbach&apos; who observed a negative strain of -3 X lo-&apos; upon unloading. This kind of behavior severely limits the over-all strain sensitivity which can be achieved in such tests. The relaxation of a small positive strain subsequent to unloading in a load-unload experiment is quite normal, however, and is to be expected. The occurrence of an apparent negative strain is also encountered in those instances where the elasticity of the bonding cement is exceeded—for example 0.5 pct or 5000 pin. for cured Eastman 910. An elastic strain of this magnitude can be encountered during load-unload type testing particularly in higher-strength materials (-150,000 yield strength). During the course of microplasticity studies on pure nickel and nickel alloys,&apos; the authors have made the following observation. In a four-point bending test, two active gages are mounted one on each side of the beam as arms R1 and R z of a conventional strain-gage bridge circuit shown in Fig. l(a). This technique provides two advantages: the sensitivity is doubled because R1 and Rz experience opposite strains and temperature compensation is provided automatically because any ambient temperature change would cause both gages to see a strain of like sign whose signals would cancel in the circuit.

Jan 1, 1964