Search Documents

By T. A. Pollard, C. R. Sandberg, E. B. Elfrink

This paper presents an evaluation of compressibility factor data and a discussion of their application to the estimation of gas reserves. A correlation is presented which provides compressibility factors for use in both two-phase and single-phase hydrocarbon systems. Accuracies comparable to those obtained previously for single-phase systems only can be expected. A simple means of predicting the presence or absence of a liquid phase in a condensate system of known composition is illustrated. The correlation is based on 1,030 compressibility determinations from 21 hydrocarbon samples taken from eight oil fields. Of the data used, 75 per cent were from California, 15 per cent were from the Mid-Continent area, and 10 per cent were from South America. The average numerical deviation of the experimental data from this compressibility chart is 1.22 per cent. Charts and tables are included and discussed which illustrate the errors involved through the misuse or nonuse of compressibility factors in estimates of gas-in-place. INTRODUCTION The compressibility factor is a coefficient which expresses the deviation of a gas of given composition from the Perfect Gas Laws. A factor such as this is necessary in any calculation involving volumes of gaseous mixtures, and finds extensive application in the estimation of natural gas and condensate reserves. It is used in the decline curve, the volumetric, and the material balance methods of gas reserves estimation. The behavior of gases at high pressures has been investigated extensively in recent years and considerable data have been assembled on the compressibility characteristics of a number of gases. Notable among the investigators of high pressure gas relationships are Kvalnes and Caddya, Kay" Sage and Laceys, and Brown, Standing" and Katz4. Other contributors are Smith and Watson, Roland and Kaveler', and Stevens and Vance". The data on compressibility factors assembled by these investigators have been presented and correlated in numerous ways; particularly noteworthy is the method first presented by Kay5 which relates compressibility factor "Z" to pseudo-reduced temperature and Pseudo-reduced pressure. Other investigators have used this empirical relationship successfully and numerous charts covering a large number of gases have been constructed on this basis. It is the purpose of this paper to evaluate compressibility factor data and to illustrate the importance of their proper use in the estimation of gas-in-place for both gas and condensate reservoirs. EVALUATION OF COMPRESSIBILITY FACTOR DATA The compressibility chart presented by Standing, Katz, Brown and Holcomb in 1941 is probably the most recent and reliable chart for natural gases. The reliability of the Katz chart has been investigated by the writers for a fairly large number of gases (single-phase) in the pressure range from 500 to 10,000 psia and temperature range from 80 to 400°F. A summary of this comparison is presented in Table I. The numerical average deviation of the chart compressibility factors from the experimental compressibility factors is 1.05 per cent, and the algebraic deviation is -0.28 per cent. The Katz correlation appears to be quite accurate for single-phase hydrocarbon systems. It was based on single-phase systems and does not necessarily provide a means of accurately predicting the behavior of two-phase systems which might be encountered in condensate work. Data published recently by Sage and Oldsl' have enlarged the knowledge of phase behavior in condensate systems by covering extensively the Compressibilities of systems in the two-phase as well as single-phase regions for a wide range of temperatures, pressures and gas-oil ratios. An evaluation of these data indicates that conditions can arise, especially in condensate systems having fairly low gas-oil ratios, in which two phases occur and the usual gaseous compressibility factors do not accurately apply. The measured PVT data of Sage and Olds covering two-phase systems have been correlated with additional such data from the Mid-Continent area and South America; these are presented in Fig. 1. This chart is the result of 1,030 compressibility* determinations from 21 hydrocarbon samples taken from eight oil and condensate fields. Of the data used, 74.75 per cent were from California, 15.45 per cent were from the Mid-Continent area, and 9.80 per cent were from South America. The chart construction is based on the original

Jan 1, 1949

By J. C. Shyne, T. D. Gulden

The steady-state creep behavior, in compression, of indium containing a dispersion of atomized glass particles was studzed over a range of temperature, stress, and composition. The observed behavior was not consistent with a simple model of dislocation climb over dispersed particles. From room temperature to about 100°C the temperature dependence could be described by an activation energy about one and a half times that of self-dijfusion in indium while at higher temperatures a much higher ternperature dependence was observed. THE value of dispersion-hardened and precipitation-hardened alloys has long been recognized. The mechanism of hardening, however, is not completely understood, especially at elevated temperatures where dispersed particles increase creep resistance. Following the suggestion of Schoeck1 that the rate of creep in dispersion-hardened materials is controlled by dislocation climb over second-phase particles, Ansel and weertman2 derived steady-state creep equations for dispersion-hardened materials. Their derivation was based on dislocation climb over non-deformable spherical particles in a metallic matrix. For low stresses they predicted a linear relation between applied stress and steady-state creep rate. For stresses large enough to cause dislocations to pinch off around the precipitate particles, the theoretical creep rate was shown to be proportional to the fourth power of stress. For very high stresses an exponential stress dependence was predicted. In all cases the theoretical creep rate is linearly proportional to the diffusion coefficient. This provided the primary temperature-dependent factor. Subsequently, Ansel and Lenel3 reported some experimental data on SAP, a composite of aluminum and aluminum oxide, that was in good qualitative agreement with the above theories. Their experimental creep rates, however, were four orders of magnitude less than the theory predicts. This was attributed to a low density of active dislocation sources in the material; somehow the dispersed particles inactivated the dislocation sources instead of merely interferring with dislocation motion. Meyers and sherby4 have recently reported on the creep behavior of SAP. Their results were inconsistent with a simple dispersion-strengthening model and led them to conclude that a continuous network of oxide controls the high-temperature mechanical characteristics of SAP. Because of the rather complex mechanical behavior of SAP and the lack of agreement about the morphology of its oxide phase, it appeared desirable to study the creep of a dispersion-strengthened material of controlled morphology. For the present work a dispersion of spherical glass particles imbedded in a matrix of indium was chosen. The microstructure of this material was similar to the idealized uniform dispersion of hard spheres in a deforming matrix assumed by Ansel and Weertman in their theoretical treatment. SPECIMEN PREPARATION For experimental convenience, it was decided to use a low melting-point metal in this investigation. It was necessary that the metal not be oxidized readily up to its melting temperature and that it wet the second-phase particles. It was desired that the material chosen for the dispersed phase be available as a fine powder with closely controlled size and shape. The material chosen to satisfy these requirements consisted of a matrix of indium metal containing a dispersion of atomized particles of a soft soda-lime-silica glass ranging from 5 to 30 µ in diam. The glass particles had a highly regular spherical shape. The composite was made by stirring the glass powder into liquid indium maintained at a temperature just above the melting point of indium, 156°C. The creep specimens were prepared by a hot pressing operation. Five to ten g of the mixture were placed in a steel die and pressed at 125°C for 5 min under a pressure of 30,000 psi. The resulting specimens were cylinders about 3/8 in. in diam and ranged from 0.4 to 0.7 in. high. By following this procedure a uniform dispersion was obtained in the creep specimens. Typical microstructures of specimens containing twenty and forty volume percent glass powder are shown in Figs. 1 and 2. Unfortunately, grain size could not be determined by the metallo-graphic techniques used. CREEP TESTING PROCEDURE Creep tests were performed in compression under constant stress. The specimens were placed in

Jan 1, 1963

By C. P. Walker

Determining the distance to remote objects by observing the time required for sound to traverse the intervening space is an old practice. Attempts have been made to use this method for determining the fluid level in oil wells but as far as the writer is aware it was not found to be practicable, because vibrations in gas in the audible frequency range do not travel as well and as far as low-frequency pressure waves, which are capable of transmitting a greater amount of energy than can be transmitted by audible waves. Wood, in his book "Sound," states that in a given medium the range of sound transmission will vary inversely as the square of the frequency. The fact that the sound from a whistle having a low tone has a much lower attenuation rate than that from a whistle of high pitch is well known. Also, the pressure waves caused by cannon fire often rattle windows at points so distant that no sound of the cannon is heard. In 1932 P. E. Lehr and H. T. Wyatt, of the Shell Oil Co., experimented with the principle of the concussion, or pressure-wave echo, method of determining fluid levels in oil wells. They released compressed gas into the top of the casing, thereby creating a pressure wave, which traveled down the well and back again. Their experiments proved conclusively that, with the proper apparatus for recording the echoes and determining the velocities at which the waves travel, the method could be used successfully for determining fluid levels in oil wells. Wells are often equipped with from one to three tubing catchers and these together with the top of the liner are frequently above the fluid. Each of these obstructions reflects a part of the pressure wave. In such wells there may be, therefore, as many as five pressure-wave reflections. Thus it is obvious that in order to identify the various reflections and to determine therefrom the fluid level in a well it is necessary to obtain a graphic record of these reflections.

Jan 1, 1937

By C. R. Sandberg, T. A. Pollard, E. B. Elfrink

This paper presents an evaluation of compressibility factor data and a discussion of their application to the estimation of gas reserves. A correlation is presented which provides compressibility factors for use in both two-phase and single-phase hydrocarbon systems. Accuracies comparable to those obtained previously for single-phase systems only can be expected. A simple means of predicting the presence or absence of a liquid phase in a condensate system of known composition is illustrated. The correlation is based on 1,030 compressibility determinations from 21 hydrocarbon samples taken from eight oil fields. Of the data used, 75 per cent were from California, 15 per cent were from the Mid-Continent area, and 10 per cent were from South America. The average numerical deviation of the experimental data from this compressibility chart is 1.22 per cent. Charts and tables are included and discussed which illustrate the errors involved through the misuse or nonuse of compressibility factors in estimates of gas-in-place. INTRODUCTION The compressibility factor is a coefficient which expresses the deviation of a gas of given composition from the Perfect Gas Laws. A factor such as this is necessary in any calculation involving volumes of gaseous mixtures, and finds extensive application in the estimation of natural gas and condensate reserves. It is used in the decline curve, the volumetric, and the material balance methods of gas reserves estimation. The behavior of gases at high pressures has been investigated extensively in recent years and considerable data have been assembled on the compressibility characteristics of a number of gases. Notable among the investigators of high pressure gas relationships are Kvalnes and Caddya, Kay" Sage and Laceys, and Brown, Standing" and Katz4. Other contributors are Smith and Watson, Roland and Kaveler', and Stevens and Vance". The data on compressibility factors assembled by these investigators have been presented and correlated in numerous ways; particularly noteworthy is the method first presented by Kay5 which relates compressibility factor "Z" to pseudo-reduced temperature and Pseudo-reduced pressure. Other investigators have used this empirical relationship successfully and numerous charts covering a large number of gases have been constructed on this basis. It is the purpose of this paper to evaluate compressibility factor data and to illustrate the importance of their proper use in the estimation of gas-in-place for both gas and condensate reservoirs. EVALUATION OF COMPRESSIBILITY FACTOR DATA The compressibility chart presented by Standing, Katz, Brown and Holcomb in 1941 is probably the most recent and reliable chart for natural gases. The reliability of the Katz chart has been investigated by the writers for a fairly large number of gases (single-phase) in the pressure range from 500 to 10,000 psia and temperature range from 80 to 400°F. A summary of this comparison is presented in Table I. The numerical average deviation of the chart compressibility factors from the experimental compressibility factors is 1.05 per cent, and the algebraic deviation is -0.28 per cent. The Katz correlation appears to be quite accurate for single-phase hydrocarbon systems. It was based on single-phase systems and does not necessarily provide a means of accurately predicting the behavior of two-phase systems which might be encountered in condensate work. Data published recently by Sage and Oldsl' have enlarged the knowledge of phase behavior in condensate systems by covering extensively the Compressibilities of systems in the two-phase as well as single-phase regions for a wide range of temperatures, pressures and gas-oil ratios. An evaluation of these data indicates that conditions can arise, especially in condensate systems having fairly low gas-oil ratios, in which two phases occur and the usual gaseous compressibility factors do not accurately apply. The measured PVT data of Sage and Olds covering two-phase systems have been correlated with additional such data from the Mid-Continent area and South America; these are presented in Fig. 1. This chart is the result of 1,030 compressibility* determinations from 21 hydrocarbon samples taken from eight oil and condensate fields. Of the data used, 74.75 per cent were from California, 15.45 per cent were from the Mid-Continent area, and 9.80 per cent were from South America. The chart construction is based on the original

Jan 1, 1949

By G. Y. Chin, M. T. Dolan, W. L. Mammel

We have obtained by computer methods the solutions of the Taylor analysis1 for axisymmetric flow in bcc metals. Four modes of slip have been treated in detail:2-4 (111), {112}(111), {123}( 111), and a mixture of all three. In addition to obtaining anisotropy of strength data for randomly oriented or textured polycrystalline material,4 we have also obtained the lattice rotation resulting from slip on the predicted systems. The patterns of lattice rotation are essential to an improved understanding of texture development during deformation. The present note shows the results of lattice rotation for 5 pct axisymmetric compression, for fifty-two axial orientations distributed at 5-deg intervals throughout the standard stereographic triangle. The data are presented in Figs. 1 to 4 for the four modes of slip. The lattice rotation is indicated by a vector, the ends of which correspond to the positions before and after the deformation. The different vectors emanated from a given orientation represent different combinations of five slip systems that satisfy Taylor&apos;s minimum shear criterion.&apos; Fig. 1 shows the results for {110}< 111> slip. They are similar to those obtained by Bishop5 for the case of (111 )(110) slip in axisymrnetric tension, since in axisymmetric flow the sense of lattice rotation for {111)( 110) in tension is the same as that for (110) (111) in compression. The pattern of lattice rotation may be divided into three regions: region A in which the rotation is toward [100]; region B, toward [lll]; and region C, toward either [ loo] or L lll]. Although some slip combinations in region C lead to [1101, the latter is not stable as continued activity of the same combinations will eventually move the axis toward [loo] or Llll]. It may also be noted that [loo] is itself not stable. However, close inspection of the vectors in region A shows that it is impossible for an axial orientation within A to rotate outside its boundary. Hence after heavy compression, all orientations from A and some from C are expected to form a diffuse band in the [100] vicinity. By far the largest region is B. Here all orientations rotate to [lll], which is a truly stable position. Thus the [111] texture component is expected to be rather sharp. For the case of (112 }( 111) slip, Fig. 2, region C is enlarged at the expense of A, with B about the same as for {110}( 111) slip. A significant difference, however, is that [loo], as well as [Ill], is now truly stable. Fig. 3 shows the results of slip on {123}( 111) systems. Region B is substantially enlarged at the expense of C, with A being intermediate in size between the {110}(111) and {112}(111) cases. Like the case of {110)( 111) slip, [100] is itself unstable. However, the lattice rotations here are confined to less than 5 deg; hence it may be considered "stable" for practical purposes. Finally, Fig. 4 shows the results of mixed slip. Here region B is the largest, and C the smallest, of all the cases considered. Both [loo] and [ill] are stable. The behavior of the three corner positions is, in fact, identical to the case of {112)( 111) slip, Fig. 2; it turns out that in these positions for the mixed slip case, all the active slip systems (based on Taylor&apos;s criterion) belong to pure {112}( 111) .4 In summing up the results for the four modes of

Jan 1, 1968

By W. D. Robertson, J. -P. Briant, M. Gell

The distribulion of dislocatiorls genevnted during the propagalion of hydrogen-induced cracks in Fe-Si cryslcrls was studied as a funclion of temperature f-110o lo 243°C) and silicorz conlenl (3.1 and 4.2 pct). It zoas found that the clislocntiotl densily and the size of the plctslic zone increased with increasing tempevalnre and decreasing silicon conlent. Furtheruvzove, there was a correlcrtion in the dislocalion patterns formed around cracks that propagated at different tempevattrt-es in cvyslals of diffevent silicon contents such that a 1 pcl itlcverise in silicon content was equivalent to a lemperaluore redziclion of approximately 120°C. The experimenlal YeSnllS were compaved with the results of an elastic-plastic wodel that relates the size of the pluslic .zone to the cvack length, internal pressure, and yield stress. it was found that the clzunges in the dislocctlion cryrays with lemperature and combosition can be e-kplaincd by the lemperatrnre and composilion dependence of the yield stress. An wmtsual feature of this study was the observation of low -temperature noncrystrrllogruphic cracks. PLASTIC relaxation can occur at the edge of a moving crack by the operation of pre-existing dislocation sources or by the creation of new sources in a region of perfect 1attice.l&apos;2 The amount of relaxation in a given crystal depends on the crack velocity and the temperature, and can determine whether a crack propagates in a brittle or a ductile manner.1-3 Tetelman and Robertson4 observed microcracks m Fe-3 pct Si single crystals that formed on (100) planes by the precipitation of hydrogen gas during electrolytic charging at 23°C. The plastic deformation accompanying crack propagation was observed in the form of dislocation etch pits and an extensive analysis has been made of the size, shape, and crystallography of the dislocation arrays5-7 This paper describes the effect of temperature on the plastic deformation generated around hydrogen-induced cracks in Fe-3.1 pct Si and Fe-4.2 pct Si single crystals. The observations are compared with the results of an elastic-plastic model for yielding around a crack containing an internal pressure. 1) EXPERIMENTAL PROCEDURES The composition of the materials used in this study are given in Table I. Single crystals of alloy A and the 4.2 pct Si alloy were grown by the strain-anneal method; crystals of alloy B and the 5.1 pct Si alloy were cut from an ingot grown from the liquid. he ~ingle-cr~stal metallographic specimens had a Cross section of 0.100 by 0.050 in. with a length of at least 3/8 in. Specimens were annealed for 3 hr at

Jan 1, 1968

By A. L. Foell

The development of the modern blast furnace began more than one hundred years ago, with the abandonment of the small hillside furnaces. Its development, especially during the past 50 years, has been accomplished through a rapid succession of enlargements and improvements. The remarkable growth in furnace output during this period is a tribute to the engineers and operators who have constantly endeavored to build larger and more efficient blast furnaces. The blast-furnace operators and their mechanical and electrical maintenance staffs have originated most of the furnace improvements as a result of their practical operating experience. Furnace Lines About 50 years ago furnaces consisted of hearths about 11 ft. in diameter, which merged into rather high flat boshes, with no cylindrical section between the bosh and inwall and no cylindrical section at the stock line. Since then the lines of the furnace invariably have included a cylindrical section above the bosh and a cylindrical section at the stock line. Notable increases have been made in both hearth diameters and furnace heights. During the past 10 years hearth diameters in new and replaced furnace stacks have rather consistently fallen within the range of 25 to 28 ft. while their heights have varied be- tween 100 and 106 ft. The lines of most large furnaces using Mesabi ores have followed a fairly uniform pattern (Fig. I) in which the principal dimensions for 25-ft. hearth furnaces are as follows: Hearth diameter............... 25 it. o in. Height of bosh................. 10 ft. o in. Bosh angle.................... 81 1/2º Diameter top of bosh........... 28 ft. 0 in. Inwall batter per foot........... 15/16 Stock-line diameter............. 19 ft. 6 in. Large-bell diameter............. 14ft. 6 in. Height from iron notch to top ring casting...................... 100 ft. o in. In determining the size of a new furnace for the replacement of an existing stack or for an addition to productive capacity, it is not possible to formulate a procedure that can always be followed. Local conditions, requirements as to capacity, available raw materials and prospects for future expansion dictate the size of such proposed improvements or additions. However, where the immediate requirements indicate a stack or furnace of a certain productive capacity, and the future prospects indicate that more capacity can be utilized, it is well to consider building the furnace ior the ultimate capacity rather than to install a smaller hearth with provision for future enlargement. The decision to build the larger hearth and stack for the ultimate capacity probably will require no greater initial investment, since the saving in the cost of brickwork will about offset the greater cost of the hearth jacket, tuyere breast and bosh bands. In the meantime. the furnace of larger hearth dimensions, while operating below its ultimate output, will yield a better coke rate and

Jan 1, 1943

By John Chipman, Carl F. Floe

A year ago the authors published a paper on the solubility of sulphur dioxide in molten copper.l The data in that paper agreed closely with that obtained by previous investigators, which, however, did not agree with the theoretical predictions for solubility of triatomic gases of this type. It has generally been assumed that the reaction of solution is: 6Cu(l) + SO2(g) = Cu2S (in Cu) + 2 Cu2O (in Cu) [I] _ (CU2S)(CU2O)2 K = (Cu)6(Pso2) [2] Neglecting the slight change in copper concentration and considering that the concentrations of sulphur and oxygen are both proportional to the volume of gas absorbed, it follows that this volume should be proportional to the cube root of the pressure. It was found that a plot of the volume absorbed versus the cube root of the pressure gave a straight line, but that this line did not pass through the origin. Calculations of the equilibrium constant for the reaction showed that it was fairly constant for pressures above 100 mm. but that actually it approached zero at lower pressures. The conclusion was made therefore that the solubility as determined could not be expressed by Eq. I or by any other simple equation, particularly at low pressures. Other investigators have concluded that the solubility of sulphur dioxide in copper as well as in several copper alloys is more nearly proportional to the square root of the pressure.* Such a conclusion, however, does not satisfactorily account for the results, even when rather complicated mechanisms for the solution are assumed. The highly improbable speculations of Lepp2 and of Schneider and Esch3 based upon hypothetical dissociation pressures of Cu2O and CU2S dissolved in copper illustrate the type of reasoning that has been used to account for the discrepancy. In the discussion of the previous paper, the authors pointed out that the discrepancy might possibly be due to impurities in the copper, although the purest form available was used. Further consideration of the problem suggested that the impurities likely to have the greatest effect on the solubility of so2 in copper were oxygen, sulphur and carbon. Accordingly, further experiments have been made to show the effects of these elements. Experimental Methods The apparatus used for measuring solubility was essentially the same as that shown in the previous paper. However, since it was desired to make measurements at low pressures, a McLeod gauge was added and 6-mm. glass tubing was substituted for the capillary. This increased the "hot volume" considerably but was necessary in order to obtain good vacuums in a reasonable length of time. A mercury diffusion pump was also installed.

Jan 1, 1942

By A. L. Foell

The development of the modern blast furnace began more than one hundred years ago, with the abandonment of the small hillside furnaces. Its development, especially during the past 50 years, has been accomplished through a rapid succession of enlargements and improvements. The remarkable growth in furnace output during this period is a tribute to the engineers and operators who have constantly endeavored to build larger and more efficient blast furnaces. The blast-furnace operators and their mechanical and electrical maintenance staffs have originated most of the furnace improvements as a result of their practical operating experience. Furnace Lines About 50 years ago furnaces consisted of hearths about 11 ft. in diameter, which merged into rather high flat boshes, with no cylindrical section between the bosh and inwall and no cylindrical section at the stock line. Since then the lines of the furnace invariably have included a cylindrical section above the bosh and a cylindrical section at the stock line. Notable increases have been made in both hearth diameters and furnace heights. During the past 10 years hearth diameters in new and replaced furnace stacks have rather consistently fallen within the range of 25 to 28 ft. while their heights have varied be- tween 100 and 106 ft. The lines of most large furnaces using Mesabi ores have followed a fairly uniform pattern (Fig. I) in which the principal dimensions for 25-ft. hearth furnaces are as follows: Hearth diameter............... 25 it. o in. Height of bosh................. 10 ft. o in. Bosh angle.................... 81 1/2º Diameter top of bosh........... 28 ft. 0 in. Inwall batter per foot........... 15/16 Stock-line diameter............. 19 ft. 6 in. Large-bell diameter............. 14ft. 6 in. Height from iron notch to top ring casting...................... 100 ft. o in. In determining the size of a new furnace for the replacement of an existing stack or for an addition to productive capacity, it is not possible to formulate a procedure that can always be followed. Local conditions, requirements as to capacity, available raw materials and prospects for future expansion dictate the size of such proposed improvements or additions. However, where the immediate requirements indicate a stack or furnace of a certain productive capacity, and the future prospects indicate that more capacity can be utilized, it is well to consider building the furnace ior the ultimate capacity rather than to install a smaller hearth with provision for future enlargement. The decision to build the larger hearth and stack for the ultimate capacity probably will require no greater initial investment, since the saving in the cost of brickwork will about offset the greater cost of the hearth jacket, tuyere breast and bosh bands. In the meantime. the furnace of larger hearth dimensions, while operating below its ultimate output, will yield a better coke rate and

Jan 1, 1943

By Carl F. Floe, John Chipman

A year ago the authors published a paper on the solubility of sulphur dioxide in molten copper.l The data in that paper agreed closely with that obtained by previous investigators, which, however, did not agree with the theoretical predictions for solubility of triatomic gases of this type. It has generally been assumed that the reaction of solution is: 6Cu(l) + SO2(g) = Cu2S (in Cu) + 2 Cu2O (in Cu) [I] _ (CU2S)(CU2O)2 K = (Cu)6(Pso2) [2] Neglecting the slight change in copper concentration and considering that the concentrations of sulphur and oxygen are both proportional to the volume of gas absorbed, it follows that this volume should be proportional to the cube root of the pressure. It was found that a plot of the volume absorbed versus the cube root of the pressure gave a straight line, but that this line did not pass through the origin. Calculations of the equilibrium constant for the reaction showed that it was fairly constant for pressures above 100 mm. but that actually it approached zero at lower pressures. The conclusion was made therefore that the solubility as determined could not be expressed by Eq. I or by any other simple equation, particularly at low pressures. Other investigators have concluded that the solubility of sulphur dioxide in copper as well as in several copper alloys is more nearly proportional to the square root of the pressure.* Such a conclusion, however, does not satisfactorily account for the results, even when rather complicated mechanisms for the solution are assumed. The highly improbable speculations of Lepp2 and of Schneider and Esch3 based upon hypothetical dissociation pressures of Cu2O and CU2S dissolved in copper illustrate the type of reasoning that has been used to account for the discrepancy. In the discussion of the previous paper, the authors pointed out that the discrepancy might possibly be due to impurities in the copper, although the purest form available was used. Further consideration of the problem suggested that the impurities likely to have the greatest effect on the solubility of so2 in copper were oxygen, sulphur and carbon. Accordingly, further experiments have been made to show the effects of these elements. Experimental Methods The apparatus used for measuring solubility was essentially the same as that shown in the previous paper. However, since it was desired to make measurements at low pressures, a McLeod gauge was added and 6-mm. glass tubing was substituted for the capillary. This increased the "hot volume" considerably but was necessary in order to obtain good vacuums in a reasonable length of time. A mercury diffusion pump was also installed.

Jan 1, 1942

By L. H. Schwartz, S. K. Lahiri, M. E. Fine, D. Chandra

WHILE the yield stress of solution treated Fe-Cu alloys increases rapidly with aging, a precipitate has only been directly observed in overaged samples.&apos;-" This precipitate is essentially pure fcc copper, the E phase. Hornbogen2 suggested that the initial hardening is due to clustering of copper atoms in a bcc iron matrix. Hornbogen was not able to obtain direct confirmation for such bcc clusters by either X-ray diffraction or electron microscopy; however, he did observe analogous clustering in Fe-Au alloys7 where the scattering factor difference between iron and gold is more favorable than that for iron and copper. Fujii, Nemoto, Suto, and Monma4 recently observed diffuse electron diffraction scattering in a 2.5 pct Cu alloy aged at 500°C: for 30 hr. They suggested that part of the diffuse SCattering was due to a modulated structure of 26 to 42 wavelength parallel to the (110) matrix plane. This disappeared with further aging. However, 30 hr at 500°C is well beyond the peak strengthening in this alloy which occurred at less than I hr at 500oc.4 By transmission electron microscopy they could not distinguish between quenched samples and samples aged to maximum strength. The purpose of the present investigation was to try to obtain further confirming evidence for a precipitation stage preceding the formation of the fcc E phase. Change in Young&apos;s modulus was selected for study be- EXPERIMENTAL PROCEDURE The Fe-1.67 at. pct Cu alloy was supplied by the U.S. Steel Corporation in the form of a hot forged rectangular bar. The carbon and nitrogen contents of this alloy were 70 ppm and less than 10 ppm, respectively. The samples for Young&apos;s modulus measurements were made by cold swaging to & in. diam and cutting to 2 in. lengths. Tensile test samples 1 in. long were prepared by cold rolling to a thickness of 0.012 in. and then machining to gage dimensions of 0.5 by 0.012 in. For Mijssbauer studies, foils 1 mil thick were prepared by cold rolling. During fabrication the alloy was given intermediate anneals at 845OC followed by quenching. The specimens after preparation were solution treated at 1000°C (y field) or 845°C (a field) for 1 to 2 and 5 to 6 hr, respectively, in a dynamic vacuum of about 2 x lom6 torr. The tensile and modulus specimens were drop-quenched into iced brine through an aluminum foil which vacuum-sealed the bottom of the furnace tube. The MOssbauer foils were drop-quenched into helium gas cooled by passing through N2 (liq) in a copper coil. The grain diameter of the modulus and tensile specimens was between 0.025 and 0.035 mm. The samples were aged from 400" to 500°C in a dynamic vacuum of 2 x 10-6 torr and cooled by pulling into a water-cooled portion of the vacuum system. The dynamic Young&apos;s modulus was measured at room temperature using electrostatic excitation and detection of longitudinal vibrations.&apos; Neglecting any length change during the aging, (this is expected to be comparatively small), the fractional change of Young&apos;s modulus (?Et)/Eav, is equal to twice the fractional change of the resonant frequency (?ft)/fav . ?ft =ft —fo where fo and ft are the resonant frequencies of a specimen before aging and after aging to time t. The measured f&apos;s, approximately 51,000 cps, were corrected to 25°C for any variation in the ambient temperature. The tensile tests reported were carried out at room temperature in an Instron with a 1000 lb full scale load cell. The flow stress reported herein corresponds to a strain of 0.002. The Massbauer spectra were taken in a constant acceleration model AM-1 NSEC spectrometer. The

Jan 1, 1970

By W. A. Tiller, S. O’Hara

Grain refinement in stirred melts has previously been shown to arise from dendrite segmentation. The present work discusses experiments capable of distinguishing between remelting and mechanical effects which result from fluid motion. Nucleation of still un-dercooled Sn-1 pct Pb melts with a single-crystal seed shows that the grain size of the resulting ingots decreases as the undercooling at nucleation increases up to about 1Z0C, then increases again for larger undercoolings. Pure melts exhibit little refinement as a result of nucleation at any undercooling up to 12°C. The presence of fluid motion gives rise to grain multiplication in pure melts, but the magnitude of this effect is partly obscured by solid-state pain growth. The results indicate that two grain-multiplication pvocesses can operate, one which involves dendrite remelt ing, and another associated with fluid motion. Small ingot experiments indicated that an equiaxed transition can be caused by fluid motion. THE presence of forced convection during recales-cence of undercooled Sn-1.3 pct Pb melts has been shown to give rise to extensive grain refinement during solidification of small ingots of this alloy.&apos;&apos;2 In Ref. 2, this refining was shown to result entirely from dendrite fragmentation and not from any nucleation mechanism. The purpose of the present work is to establish what processes lead to dendrite segmentation. Experiments were therefore carried out to determine the resulting structure of both pure tin and Sn-1 pct Pb melts, nucleated at various undercoolings by single-crystal seeds in the absence of forced or natural convection. By comparing the information obtained from these experiments with the results of Refs. 1 and 2, the effects of fluid motion on the ingot structure, and the details of the segmentation mechanisms, should become clear. The results of these experiments should also be helpful in understanding the solidification structures of larger ingots since natural convection, which is known to play an important role in determining ingot structure13 differs only in degree from the fluid-motion effects studied here. I) EXPERIMENTAL PROCEDURE The apparatus shown in Fig. 1 was constructed to permit controlled nucleation of stirred or unstirred supercooled melts using a seed oriented in the tin dendrite growth direction. Because the melts, in all cases, were capable of larger supercoolings than the one used in each experiment, the structure observed always resulted from a single nucleation event at the surface. The sequential procedure was as follows: Tin was melted under purified hydrogen and the oxide removed at a temperature of 450°C . Electromagnetic stirring was then used to clear the surface completely and during this time the guide tube was lowered underneath the melt surface. A pointed Czochralski-like crystal was then propagated from a specially prepared tin dendrite seed4 inside the guide tube while the temperature of the bulk melt was held just above the melting point, ensuring a perfectly clean seed which was epitaxial with the original dendrite. The dendrite and epitaxial seed, still in the guide tube, were withdrawn from the melt which, in turn, was undercooled to 2C° below the temperature at which the nucleation experiment was to be performed. The melt was held there for several minutes, both in stirred and in still-melt experiments. The temperature was next raised to the required value and the melt nucleated by the carrot seed inside the guide tube, the heater power being shut off just after nucleation. In experiments where quenching after nucleation was required, a high-density graphite crucible was used, but otherwise the procedure was as described above. The temperature gradient in the melt before nucleation was kept below 1C° per cm by carrying the experiment out just below the center of the heater shown in Fig. 1.

Jan 1, 1968

By H. [Eyring], B. J. Zwolinski

[ ] The authors are to be commended for their excellent work on the activities of components in the solid state. It is, however, felt that certain points require clarification with reference to diffusion in the solid state. It is a well established point that any rate process in a non-ideal system can be expressed by the general equation [ ] where ko is the specific rate in an ideal system, and ?A, ?B ... , the activity coefficients of the reactants and ?± and ?M,?N the activity coefficient of the activated complex and the intermediates, respectively. With reference to diffusion, the following simplified treatment in correcting for non-ideality is proposed. The same free energy change is involved in moving a particle from a solution of one activity to a solution of another activity,

Jan 1, 1947

By John H. Hollomon

A REVIEW of the steps which have been made to increase knowledge in the field of ferrous physical metallurgy during the closing period of World War II brings both pleasure and disappointment. Contributions of great value have been made, to be sure. However, due partly to the stress of the times, researchers in this field have been in the main directing their attention toward the short-range problems involved in the development and production of materiel for war. Furthermore, publication of the results of a great number of studies has had to be delayed. Much of the German work may, indeed, be lost entirely, owing to the efficiency of our bombers, which managed to account for most of the famous Kaiser Wilhelm Institut of Dusseldorf, not to mention the other large laboratories and manufacturing plants throughout that country. It would seem safe to assume, however, that fundamental research in Germany had been forced to give way to the shallow, but more pressing, engineering problems of a country trying to avoid defeat. In Russia the picture has been clouded by the limitations imposed upon comunication; but this much does seem clear: Russia alone among the great nations of the world has maintained her long-range view and has continued to foster research of the most fundamental type. English workers; on the other hand, appear to have been able only to meet the demands of the pressing problems of war, and their studies have been of the short-range variety. That metallurgical research of any sort should have continued in France during this chaotic period in her history would seem impossible, but, amazingly, a small handful of scientists did manage to carry on their work, and it is to be hoped that more will be heard from them in the future.

Jan 1, 1946

By Larry Kaufman, Martin Schatz

The effect of pressure on the extension of the ? loop in the FeSi system has been determined by means of metallogvaphic studies and hardness measurements performed on a series of high-purity Fe-Si alloys containing 7.5, 11.0, and 13.9 at. pct Si, respectively. These mensurements, performed at 42 kbar and temperatures up to 1200oC, indicate that the ? loop is expanded to about 10 at. pct Si at 42 kbar as opposed to a maximum extension of 4 at. pct Si at 1 atm. Comparison of the experimental results with thermodynamic predictions of the pressure shifts yields satisfnctory results. DURING the past few years, several studies have been performed in our laboratory1-&apos; in order to determine the effect of high pressure on phase equilibrium in pure iron and iron-base alloys. The purpose of these studies has been to elucidate the effects of high pressure experimentally and to compare the observed results with predicted pressure effects derived on the basis of known thermody-namic and volumetric data at 1 atm. These studies have included work on pure iron2,5,7 as well as Fe-Ni,1,5 Fe-cr,l,5 and Fe-c4-6 alloys. In addition, Tanner and Kulin3 have reported results of pressure studies on two Fe-Si alloys containing 2.0 and 6.25 at. pct Si. At the time of this latter study, no detailed information was available concerning the difference in volume between the a (bcc) and ? (fcc) phases in the Fe-Si system as a function of silicon content. In order to compare their observations with calculated pressure shifts, Tanner and Kulin were forced to assume that silicon had no effect on the difference in volume between a and ? iron. The resulting discrepancy between their calculation of the a/? phase boundary at 42 kbar and the observed results led them to the conclusion that silicon additions probably decrease the difference in volume between a and ? iron. Recently: Cockett and Davis8,9 have reported de- tailed studies of the lattice parameters of a series of Fe-Si alloys at temperatures ranging from 20" to 1150°C. These measurements, performed on alloys in the bcc and fcc range, show that silicon does indeed decrease the difference in volume between a and ? iron. By correcting the calculations of Tanner and Kulin in line with the observed effect of silicon they were able to show improved agreement between computed and observed pressure shifts.&apos; The present measurements were undertaken to provide additional corroboration of this effect, by extending the range of composition, in addition to exploring a situation where large extensions of a ? loop could result in impingement of the ? field with an ordered bcc phase (based on Feo.75Sio.25). I) EXPERIMENTAL PROCEDURES AND RESULTS The alloys investigated were obtained from Dr. F. Kayser of M.I.T. They were prepared at the Ford Scientific Laboratory by vacuum melting electrolytic iron and high-purity silicon. The melts were poured under an argon atmosphere into hot-topped steel molds. Subsequently the ingots were hot-worked down to 1/2-in.-diam rods. Three alloys containing 7.5, 11.0, and 13.9 pct Si were studied. Carbon, regarded as the principal impurity, analyzed at, or below, 0.001 wt pct for all of the alloys. Prior to pressure-temperature treatment, the rod was annealed for 24 hr in vacuum at 1000°C, water-quenched, and subsequently machined into 0.100-in.-diam by 0.100-in.-long specimens. Subsequent to machining, the specimens were again annealed and then examined metallographically. They were found to exhibit a clear coarse-grained ferrite similar to Figs. 10 and 110 of Ref. 1 and Fig. 2 of Ref. 3. Subsequently, specimens of each alloy were equilibrated at 42 kbar at various temperatures in supported piston apparatus.1,3,4,6 Three specimens, one of each alloy, were wrapped in platinum and exposed simultaneously. The pressure-temperature cycle consisted of increasing the pressure from ambient to 42 kbar at 25oC, heating rapidly to the desired temperature, holding for 15 min, and quenching to 100°C, followed by slower cooling to 25°C and pressure release. The temperature was measured with a Pt/Pt-13 pct Rh thermocouple which was not corrected for pressure effects. Subsequently, specimens were examined metallographically and by

Jan 1, 1964

By A. J. Opinsky

ThE superior performance of nonsag tungsten wire in lamps has been ascribed to the large-grained micro-structure that can be developed in it. The orientations of these large grains have been studied several times; most published data1-&apos; appear to be similar to those plotted in Fig. 1. These new data were obtained by X-ray back-reflection Laue technique from all of the large grains occurring in thirteen different wires that had been heated at the lowest possible temperature for the formation of large grains. Here the wire axis is plotted in terms of the principal directions of the cube. Rosi decided that the preferred orientation of the large crystals could be described by a single ideal orientation, (531), being parallel to the wire axis.&apos; Rieck2,3 and Mannerkoski&apos; have essentially confirmed this finding. Opinsky, Orehotsky, and Seigle argued that the data suggest that (211) is parallel to the wire axis instead of a single ideal orientation.5 This {211} contains the (531) ideal orientation and would appear to be a more general case. Ahlborn and Wassermann report that their orientation data stretch from ( 110) through (531) to (311) 8 This is part of the locus of (211). All these investigators have determined the orientations of the large grains by back-reflection Laue techniques. They have all obtained essentially similar results. On the other hand, Swalin and Geisler, who obtained their data by neutron diffraction, indicated that the preferred orientation of the large grains in doped wire is (110) .7 These findings cannot presently be reconciled with the pole density data described in the previous paragraph. Rosi1 and Rieck2 have advanced arguments about how their orientations could occur. Nevertheless, the crystallographic link between the (110) (cold-worked) fiber texture and the preferred orientation of the large grains has remained missing. This link can be provided through the concept of coincidence-type boundaries. Such boundaries separate two grains that have a certain number of common lattice sites. Coincidence relationships that result from rotations about low index directions are given by Brandon et a1 .&apos; What is desired is a calculation of those orientations that bear a coincidence relationship with the (110) fibers of the cold-worked matrix. This situation arises in the following fashion. Heating the tungsten wire for a combination of time and temperature falling short of the formation of the nonsag structure yields a structure that is still fibrous in nature and the preferred orientation remains (110). During grain growth, the growing grain is in contact with these fibers. It is hence possible that the entire process may begin by the passage of a coincidence boundary into a (110) fiber. The calculations were carried out analytically and checked graphically. Following Brandon et a1.,8 coincidence relationships from 1 in 5 to 1 in 19 have been considered. The results of the calculations are given in Fig. 2. Of the coincidences examined, some of those occurring upon rotation about ( 110) come closest to the experimental points in Fig. 1. Nineteen of the twenty-three data points in Fig. 1 can be attributed to the ideal orientations A, B, C, and D in Fig. 2. As determined analytically, three crystals lie within 4 deg of point A, four lie within 21/2 deg of point B, eleven lie within 5 deg of point C, and one lies within 1 deg of point D-a total of nineteen. The off-lying points are near the |110| point of the triangle.

Jan 1, 1968

By Ashok K. Sinha

The quasi-binary systems "VFe3"—VCo,—VNi,— "VCu3"and "TiFe3"—TiCo3,—TiNi3-"TiCu3"have been studied by a combination of microscopic and X-ray methods. Of the Phases encountered, eleven had close-packed ordered layer structures. Within each system, there is an approximately linear variation of the degree of hexagonality of the stacking arrangement with electron concentration, e/a. A study of eighty -four known examples of AB3 structures in transition metal alloys indicates that at e/a = 8.65 the ordered structure of the close -packed layers changes from "triangular" (AuCu3 type) to "rectangular" (SbCu3 type). At e/a < 8.65, the slope of hexagonality us e/a line increases with the group number of the A component. An interesting series of close-packed ordered structures is known to exist at the composition AB3; however, their alloy chemistry has not yet been fully understood. Dwight and Beck1 observed a correlation between position of the components in the periodic table and the type of crystal structure of the binary AB2 structures formed by transition metals. No consistent effect due to atomic size was noted and they proposed that electron concentration* predominates in In the present paper, the AB3 structures are characterized by a symbol describing the stacking and a symbol specifying the ordered layer. The stacking symbol used was introduced by Zhdanov,7 and later used in the International Tables,8 and by Beck9 in his survey of possible layer stacking structures. Alternating positive and negative integers are used to represent the number of positive (e.g., in abcabc...) and negative (e.g., in cbacba...) stacking sequences. T denotes the triangular ordering found in close-packed planes of the AuCu3 structure and R the rectangular ordering found in close-packed planes of the SbCu3 structure, Figs. 3(a) and (b). The present structure symbols are listed in Table I alongside the prototypes, where known, and other structural characteristics of AB3 phases found in the present investigation. EXPERIMENTAL The starting materials were metals of 99.9+ purity. The alloys were prepared by arc melting under argon atmosphere using a nonconsumable tungsten electrode and water-cooled copper hearth. The total weight loss upon melting and subsequent annealing was always less than 0.5 pct and hence the alloys will be referred to by their intended (unanalyzed) compositions. The alloys were homogenized and annealed at suitable temperatures in the range 750" to 1200°C, the annealing periods varying from 3 days at 1200°C to 10 days at 750°C. The heat treatments were carried out in sealed silica capsules under argon atmosphere and with a molybdenum foil cover between the alloy and the walls of the capsule. The quenched alloys were microscopically examined and reduced to powder by crushing or filing. The powders were, whenever necessary, given a stress-relieving anneal at the original annealing temperature. X-ray powder patterns were obtained with an asymmetrical focusing camera, a Guinier-deWolff focusing camera, or a Debye-Scherrer camera. Cr, Co, or Cu radiations were used. In some cases, to ensure the minimum degree of contamination, the deeply etched specimens were mounted in the asymmetrical focusing camera to obtain an X-ray pattern. Internal silicon standards were employed for lattice parameter determinations. The intensity calculations were made by using a Fortran IV program

Jan 1, 1970

By F. G. Miller, R. C. McFarlane, T. D. Mueller

During the process of gas storage in pressure-depleted oil reservoirs, it has been observed that in some instances additional liquid oil is recovered and that the composition of the storage gas is materially altered. A mathematical study was made of the dynamic behavior of such a depleted oil reservoir undergoing gas injection. The important variable considered in this study, not included in previously published work, was that of compositional effects on the phase behavior of two-phase flow. Pressure, saturation and component composition profiles were developed for a linear, horizontal and homogeneous porous medium containing oil and gas but undergoing dry gas injection. Special new techniques were developed to overcome the problems of numerical smoothing which arise in the solution of the equations representing such systems. The method of solution includes the development of partial differential equations describing the behavior of the system, representing these equations by finite difference approximations, making certain simplifying assumptions and, finally, applying methods of numerical analysis with the aid of a high-speed digital computer. In an example calculation, results using the mathematical model are compared with field observations made on a gas storage project in Clay County, Tex. This field project involved a depleted oil reservoir used &apos;for gas storage and gas cycling purposes. As a result of these processes, the reservoir yielded substantial amounts of secondary oil, both in the form of stock tank oil and as vaporized products in the produced gas. The methods derived in this study may be applied to a variety of oil reservoir problems which are dependent on compositional effects. In recent years the number of oil reservoirs being used for gas storage purposes has increased greatly, and there has been at least one published account of additional oil recovery resulting from gas cycling a depleted oil reservoir after repressur-ing with dry gas for storage purposes. Additional oil recovery from oil reservoirs resulting from gas storage operations could become an important secondary recovery process. This is especially true since the use of natural gas in large metropolitan areas continues to increase and more gas storage volume near these areas is needed. These facts provided the motivation for the work reported here. This paper reports on a study of the inter-relations of composition, saturation and pressure changes which occur when hydrocarbon gas is injected into an oil reservoir system. From an understanding of the process, prediction methods may be developed for use in forecasting the secondary recovery products from gas storage operations in oil reservoirs and, consequently, .the economics of such projects can be developed. The process of storing hydrocarbon gas in an oil reservoir and the cycling process both involve the transient multi-phase flow of oil reservoir fluids with changing fluid compositions. There is a scarcity of work published on the subject. One of the first laboratory models developed to study a problem similar to the one posed here was published in 1948 by Standing, Lindblad and Parsons. l They developed a model to study gas condensate recovery by gas cycling, including the effect of heterogeneous reservoir permeability. Their results showed the compositional effects of injecting a gas of known composition into a two-phase condensate reservoir of known composition. A major finding of their study indicated that maximum oil recoveries from gas cycling projects result when low cycling pressures are used. Previous to their work it was generally considered

By William S. Hannan, H. Rush Spedden

Flotation may be defined as a process whereby mineral particles are concentrated by selective adhesion to air-liquid interfaces. The process involves attachment of desired mineral particles to air bubbles, and removal of the mineralized air bubbles from the surface of the pulp as a mineral concentrate. Problem Although great strides have been made in understanding the principles of flotation, one part of the process which has not been adequately explained is the mechanism by which selected mineral particles become attached to air bubbles. A glance at the froth of an operating flotation cell demonstrates adequately that use of proper reagents and conditions mill result in such attachments; but how such attachment takes place is not definitely known. A better knowledge of this factor might lead to improved flotation practice either through altered machine design or the application of scientific innovations. Direct observation of the behavior of mineral particles and bubbles in an operating cell does not appear promising because of the complexity of the system. For this reason, conclusions concerning the problem must be deduced from theoretical considerations or from observation of simplified systems designed to reproduce conditions thought to exist in an operating cell. Present Theories Two mechanisms have been recognized as fundamental in particle-bubble attachment: gas precipitation and direct encounter. It is a simple matter to demonstrate in the laboratory that attachment can be brought about by either mechanism. As to the relative importance of the two in a particular flotation system, however, there is a divergence of views. Taggart1 states, Under exceptional circumstances sufficient force and time to effect attachment may conjoin in an ore pulp undergoing agitation, and a preformed bubble adhere to a collector-coated mineral particle, but the occurrence is rare. . . . Pulp-body froth flotation effects selection in the body of a conditioned pulp by causing gas bubbles to precipitate from solution in the liquid phase preferentially at the surfaces of collector-coated particles, and to cling to these particles, Gaudin2 states that, within the pulp body of an agitation machine, attachment of mineral particles to air bubbles occurs as a consequence of contact between the two. This theory has been supported by theoretical reasoning and much circumstantial evidence. Perhaps the most interesting bit of evidence in favor Of this theory js the difficulty in the flotation of fines. If attach-

Jan 1, 1949

By Hiroshi Oikawa, A. G. Guy

Recent experimental evidence has shown that it is often incorrect to neglect the reactions occurring at an interface with respect to the diffusion reactions. By introducing reaction-rate constants and k for the transfer of components 1 and 2 between the a and b phases, a phenomenological description has been devised for the coupled processes of diffusion and in-terfacial reaction. An illustrative example has been analyzed with the aid of a suitable computer program. The results indicate that for kab values on the order of 10&apos;1 cm per sec and Ð values of about 10¯10 sq cm per sec, the interfacial reactions are significant during the first 10 2 to l0 6 sec of a diffusion experiment. Among the observable effects of the interfacial reactions in this case appear to be the burying Kirkendall markers and an anomaly in the initial shifting of these markers. THE study of diffusion between two phases is especially important because of its direct connection to phase transformations. For example, the transformation of austenite to ferrite involves diffusion in the a and ý phases and also an interfacial reaction (trans formation) at the a/? interface. If ferrite and aus-tenite are the two members in a diffusion couple, the same two reactions occur during the diffusion process. The numerical values obtained for the diffusion coefficients and interfacial-reaction constants from diffusion experiments can then be employed to predict phase-transformation behavior. Refinement of the calculations of phase transformation is also suggested by the present method for treating two-phase diffusion. In the recent paper by Nolfi et al.&apos; only the solute com ponent is considered in the interfacial reaction accompanying a phase transformation. The present analysis demonstrates that both components must be considered. It is generally recognized that reactions of some kind must occur at the interface in a two-phase diffusion couple, but usually it is assumed that the interface reaction occurs so rapidly that its effect on the overall process is unobservable because the slower diffusion process determines the kinetics of the process. However, Seith and Heumann 2 pointed out that effects of the interfacial reaction should always occur in the early stages of diffusion, and recently Eifert et al. 3 have shown that interfacial reactions influence the composition and motion of the interface in a/ß Al-Cu diffusion couples during the entire course of prolonged, high-temperature treatments. A general procedure for handling the combined processes of interfacial reaction and diffusion in a binary, two-phase system is to introduce kinetic coefficients for the interface reactions and to develop the system of equations governing the interlinked processes of diffusion and interface reaction. The resulting equations, although not amenable to solution in closed form, can be usefully manipulated by a digital computer. INTERFACIAL REACTIONS IN TWO-PHASE DIFFUSION Consider the general case of diffusion between two solid solutions, a and ß, which are adjacent phases in a binary system, Fig. 1. Two semiinfinite bars having compositions c1 and c0ß are jointed at x = 0. If concentrations are expressed in volume fraction,4

Jan 1, 1970