Search Documents

By K. E. Gray, G. M. Myers

In previous publications,5 results of single-blow bit tooth impacts on saturated rocks at various stress states were reported. This paper extends these earlier works to include study of bit impact tests on salt water-saturated Berea and Bandera sandstone samples under conditions of elevated confining and pore pressure. During the tests dynamic filtration and deposition of a mud cake were occurring due to the presence of drilling mud in the borehole and a borebole-to-formation pressure differential. Results indicate that saturation of these sandstones with salt water tends to make them weaker than when saturated with nonreactive fluids. Plastic failure often occurs even when extremely high fluid-loss muds are present in the borehole. The failure mode tends toward plasticity with decreasing fluid loss. Brittle failure of sandstones in mud-filled holes is apparently relatively rare. Depending on the stress level, and the associated failure mode, withdrawal of the bit tooth induces a tensile force that seems rather important relative to the quantity of rock removed by vertical tooth impact. INTRODUCTION A vast majority of the oil wells drilled today involve the use of colloidal muds having a measurable mud filtrate loss. It is known from field experience that reduction of the water loss of a mud generally results in a reduction of the penetration rate. This paper describes an investigation of crater formation at simulated bottom-hole pressure conditions for drilling fluids having different water losses. The literature on single bit tooth impact (cratering) at atmospheric conditions is extensive, but only a limited amount of work has been performed at the stress conditions similar to that found in oil wells. Spherical penetrator cratering tests on rocks at hydrostatic pressure were performed by Payne and Chippenda1e.1 Chisel impact studies on limestone with independently varying overburden and borehole pressure and atmospheric pore pressure were reported by Garner, et al.2 Gnirk and Cheatham3 performed "static" penetrator tests on dry rocks at equal overburden and borehole pressures. Podio and Gray4 studied the effect of pore fluid viscosity at atmospheric pore and borehole pressures and varying overburden pressures. Their results illustrated the importance of fluid saturation of the rock. Yang and Gray5 reported single tooth impact tests on saturated rocks at elevated borehole, pore and confining pressures, but with equal pore and borehole pressures. For the work reported in Refs. 4 and 5, filtration across the hole bottom purposely was avoided. Maurer6 has investigated the effect of independently varying pore and borehole pressures at elevated overburden pressures by using a zero water-loss mud in the borehole. His tests were carried out under conditions that allowed independent control of overburden, pore and borehole pressures, but without control of filtration at the hole bottom. EXPERIMENTAL APPARATUS AND PROCEDURE APPARATUS The single blow chisel impact apparatus used by Garner,2 as modified for this study, is shown in Fig. 1. A cross section of the pressure vessel is shown in Fig. 2. Several pieces of auxiliary equipment were added to the original apparatus in order that the borehole and pore pressure could be applied in a manner analogous to the actual bottom-hole situation. The pore-pressure system included a filtrate volume-measuring transducer and had sufficient surge capacity to allow filtrate to be injected into the sample without significant changes in pore pressure at points remote to the borehole. The borehole-pressure system contained a cross connection to the pore-pressure system so that both systems could be pressured simultaneously. The borehole-pore pressure differential was applied rapidly by the use of a quick and full-opening valve. The operation of this valve triggered two industrial

Jan 1, 1969

By D. W. Levinson, L. F. Mondolfo, N. A. Gjostein

MANY investigations are reported in the literature on the age hardening of Al-Mg-Zn alloys but most of them are concerned mainly with mechanical property changes. The present investigation was started to determine the structural changes in an AlMgZn alloy. The alloy composition was chosen in the existing phase diagram" so as to lie on the quasi-binary section Al-(AlZn)Mg Actually, it was found that the diagram as published was inexact at the lower temperatures and that the phase in equilibrium with aluminum at the lower temperatures is not the ternary compound. Technique The alloy used for the investigation was prepared from high purity materials and the resulting composition as determined by chemical and spectro-graphic analysis was: Zn, 6.17 pct; Mg, 2.13 pct; CU, 0.03 pct; Si, 0.009 pct; Fe, 0.005 pct; and the remainder, aluminum. An ingot -in. thick was cast, homogenized, hot and cold-rolled to strip 0.064-in. thick for dilato-metric and powder work, and to foil 0.005-in. thick for single crystal work. Solution treatment temperature for all specimens was 475°±5°C for periods of at least 1 hr, either in air furnaces or lead baths. Aging was done in oil or Pb-Bi baths controlled to within ±3°C. The aging was investigated at the following temperatures: room (30°), 130°, ZOO0, 260°, and 300°C. For the dilatometric study, the length changes were measured continuously, at temperature, using strips of alloy 10 in. long, shaped as channels to eliminate undesirable deflection due to bending. To compensate for thermal expansion and contraction due to the variations in the temperature of the aging baths, the dilatometers were built of the same alloy as the one being tested, with the same shape and size, but with material fully annealed. Gages on which 0.0005 in. could be read and 0.0001 in. estimated were used to measure the dilation. Length changes were also measured at room temperature. Strips approximately 10 in. long were solution-treated, quenched, and measured in a supermicrome-ter in which 0.0001 in. could be read and 0.00001 in. estimated. The specimens were immediately aged for a given time, remeasured, and then the cycle was repeated, including the solution treatment. The error of measurement with this technique was somewhat larger, but the results confirmed those obtained by continuous measurements. The sheet used for the dilatometric work was checked to determine grain orientation. Both X-rays and etching methods revealed a rather coarse grain size but only a negligible amount of preferred orientation. Thus, volume changes calculated from length changes are not invalidated by preferential orientation effects. Pieces of the sheet used for the dilatometric work were also used for the metal -lographic work. Filings for the powder X-ray study were produced from a homogenized piece of sheet. These filings were magnetically purged of iron, screened to 200 mesh, sealed in a Pyrex tube containing an argon atmosphere, and solution-treated. When quenching, the glass tube was broken, so that the powder dropped directly into cold water. After drying with alcohol, the powder was sealed into capillary tubes containing an argon atmosphere and then aged for the required times. In spite of all these precautions the three strongest lines of a( Al2o) appeared in all patterns, and were used as internal standards. By the use of these standards, accurate corrections for film shrinkage were obtained. For the back reflection patterns, annealed nickel powder was mixed with the specimen to act as internal standard. A preliminary investigation was made in small (57.3 mm diam) powder cameras. From these patterns the changes which took place in the solid-

Jan 1, 1957

By R. F. Hehemann, R. H. Goodenow

Thermal arrest, hot-stage microscopy, and transtnission electron microscopy techniques have been employed to study the transformations in low-carbon iron and Fe-9 pct Ni alloys. In continuous cooling experiments, each alloy transforms at an essentially constant temperature for cooling rates below the critical rate required for martensite formation. However, high-tenzperature transformation in pure iron takes place by a different mechanism than that in the 9 pct Ni alloys. Pure iron exhibits an equi-a a structure with a low and random dislocation density while the 9 pct Ni alloy exhibits a cell or lathlike substructure analogous to that of low-carbon martensite. This same substructure characterizes upper bainite in higher-carbon inaterials. UNTIL relatively recently, diffusionless transformations have been considered to take place by the same mechanism and have been termed mar-tensitic transformations.l-3 The most characteristic feature2 of this mode of transformation is the shape change produced by a shear deformation and growth of the product phase frequently occurs at an extremely high velocity approaching that of an elastic wave.4'5 A different mechanism of diffusionless transformation was recognized first in Cu-Ga and other copper-base alloys.6,7 From the absence of surface tilting in these alloys, it has been concluded that transformation is not accomplished by the cooperative shear displacements that occur in mar-tensitic reactions. Transformation presumably takes place by the rapid movement of an incoherent interface which is capable of propagating across parent grain boundaries. Although the transformation is diffusionless in the macroscopic sense, advancement of these interfaces is accomplished by an atom by atom rearrangement. Transformations of this type therefore require thermal activation and are generally operative only at high temperature. Although requiring short-range diffusion at the interface, propagation still occurs so rapidly that the reaction cannot be suppressed with normal cooling rates.7 The resulting microstructures exhibit large, irregular-shaped regions of the product phase, which are essentially free of crystallographic features. Consequently, these reactions have been termed massive transformations. Transformation by the massive mechanism has been reported to occur in ferrous systems involving pure iron and binary alloys of iron with nickel, chromium, and silicon by Gilbert and owen.' The Fe-Ni system is of special interest in that the massive transformation was observed in alloys with less than 15 pct Ni while alloys having more than about 28 pct Ni transformed by the conventional mar-tensitic mechanism. By employing cooling rates substantially higher than those used by Gilbert and Owen, Swanson and Parr have been able to suppress the massive transformation in alloys with 0 to 10 pct Ni9 and produce martensite as revealed by surface relief. The massive transformation in the alloys having less than 15 pct Ni was identified by the lack of surface relief and an irregular rather than clearly acicular microstructure.' Speich and Swann, using thin-foil electron microscopy, identified three distinctly different structures10 in quenched binary Fe-Ni alloys. From 0 to 4 pct Ni the structure consisted of blocky grains of a with a low and random dislocation density. Alloys with 4 to 25 pct Ni exhibited a cell structure with a high dislocation density; and at greater than 25 pct Ni the structure consisted of internally twinned plates. The structural change at approximately 4 pct Ni suggests that alloys with low nickel content transform by a different mechanism than those with intermediate nickel contents and the transition from the cellular to the internally twinned structures at approximately 25 pct Ni is analogous to the transition from needles to internally twinned plates that occurs over a narrow range of carbon contents in Fe-C martensites.11,12 Conventional bainitic and martensitic modes of austenite decomposition operate in ternary Fe-C-Ni alloys having less than 10 pct Ni. This investigation was conducted in order to explore the conditions under which the massive, bainitic, and martensitic transformations occur and the relationships between these modes of decomposition.

Jan 1, 1965

By R. E. Goodman, M. A. Mahtab

The, state of stress around a vertical wellbore in rock following nonlinear stress-strain laws is examined by means of finite element analysis. The wellbore, is considered an axisymmetrier body with axisymmetric loading. The initial vertcal and borizontal stresses are "locked" in the rock elements around the wellbore and a new state of stress is generated by the displacements which occur around the borehole. A point-uise variation of the elastic moduli is made on the basis of the new stress state and the triaxial data. The initial stresses tire now reintroduced along with tile changed moduli and original boundary constraints. This procedure is repeatrd until convergent stresses are reached. The effect of nonlinearity on streases is examined for a 6,000-ft wellbore in a scbistose gnezss and Berea sandstone using results of laborratory triaxial compression tests. The results sbou that the effect is restrited to one well radius from the bottom periphery of the bole. Beyond a distance of one-quarter radius, the elfect of nonlinearity on stresscs IS almost always less than percent for the cases conszdered, The consideration of a static pressure inside the well does not magnify the effect of monlinearity on borehole stresscs. INTRODUCTION The terms " wellbore" and "borehole" here designate cylindrical openings in the ground with vertical axis and a circular cross-section. A knowledge of the stress redistribution that occurs on excavating a wellbore is important in understanding (1) the behavior of the lined or unlined hole, (2) hydraulic fracture response, and (3) the effect of stress redistribution on drillability; also it is important in predicting initial stresses in the virgin ground, and in analyzing the response of measuring instruments placed in the borehole. Our knowledge of the state of stress around a wellbore has been restricted to homogeneous, isotropic, elastic material and derives chiefly from the analysis by Miles and Toppingl and the photoelastic work of Galle and Wilhoit2 and Word and Wilhoit.3 In this investigation the state of stress is examined for a nonlinear elastic material by means of finite element analysis. Many rocks possess stress-strain curves that depart notably from straight lines in their initial or final portions. While the literature contains abundant stress-strain data from triaxial tests (axisymmetric loading) on cylindrical rock specimens, there is little information on rock deformability under nonaxisymmetric loading conditions such as occur at each point around the bottom of a wellbore. Although there is some knowledge of the effect of intermediate principal stress on rock strength, there is virtually nothing known about its effect on rock deformability; therefore, we have assumed here that the effect of intermediate principal stress can be ignored. A schistose gneiss4 and Berea sandstone5 were selected as representative rocks for this analysis. The traditional graphs of deviator stress (ol -o3) vs axial strain were reworked to give the tangent modulus as a function of the deviator stress for varying values of the minor principal stress. The result is a nesting family of skewed, bell-shaped curves for tile gneiss (Fig. 1A) and the sandstone (Fig. 2A). A similar replotting of the lateral strain data defines the variation of Poisson's ratio (v) with the deviator stress and confining pressure. These curves, shown in Fig. 1B for the gneiss and in Fig. 2B for the sandstone, are not so well ordered as the tangent modulus curves. However, all of these display an increase of v with deviator stress application, but the rate of increase diminishes with confinement. The ET and v curves for the two rock types are tabulated in Tables 1A and 1A for use in a digital computer so that material properties corresponding to a given state of stress can be assigned by interpolation. STATEMENT OF THE PROBLEM Given a state of stress in the ground prior to drilling a hole (initial stresses), we assume that

Jan 1, 1969

By J. S. Kirkaldy, D. H. Weichert, G. R. Purdy

Two-phase diffusion couples have been used to simulate the growth of proeutectoid ferrite in ternary Fe-C-Mn austenites. It has been shown, theoretically and expermentally, that the results fall into two classes: a high-super saturation class in which the major role of manganese additions is to influence the boundary conditions for carbon diffusion; and a low-supersaturation class in which the partition of manganese is required. In the latter case, a drastic inhibition of the reaction rate is to be expected. As an aid to the kinetic analysis, a portion of the iron-rich comer of the Fe-C-Mn constitution diagvam has been determined. The conclusions drawn from this work have a direct bearing on the kinetics of the grain boundary ferrite precipitation reaction and thus yield insight into the manner in which alloying elements can influence the rate of austenite-decomposition reactions. THE role of alloying elements in the austenite-decomposition reactions has provoked considerable interest, since it is apparent that an understanding of this phenomenon will yield insight into the harden-ability problem. In the present investigation, attention is focused upon the proeutectoid-ferrite transformation in the system Fe-C-Mn, and upon the manner in which manganese additions affect the growth rate of ferrite. When hypoeutectoid austenite is supersaturated, the ferrite first formed usually exhibits two distinct types of morphology. At low super saturations, grain boundary layers grow into the austenite grains with an approximately planar interface, and usually possess at least one incoherent phase boundary. At high supersaturations, Widmanstiitten plates appear, which bear an orientation relationship' to the parent austenite. In this case, there is a chance of partial lattice matching along the sides of the plates and an attendant semicoherent interface structure. Purdy and Kirkaldy2 have used a two-phase binary Fe-C diffusion couple to demonstrate that carbon volume diffusion controls the incoherent reaction, while Rouze and rube,' using thermionic-emission microscopy, have shown that the thickening of Widmanstatten ferrite plates is appreciably slower than expected for carbon-diffusion control. This suggests that the plates are bounded by partially coherent, low-mobility interfaces. The redistribution of alloying elements during transformation has been the object of studies by Aaronson4 and owmman,5 who have shown that no alloy partition occurs during ferrite precipitation in the systems Fe-C-Cr and Fe-C-Mo, respectively. More recently, Aaronson et al.6 have shown that some manganese partition occurs at low super-saturations in the system Fe-C-Mn, but that no partition occurs at higher super saturations. Pickle-simer et al.7 observed that no manganese partition occurred during the early stages of the pearlite transformation in a manganese steel, and concluded that volume diffusion of the alloying element is not the rate-deter mining factor. The diffusion coefficients pertinent to this investigation have been measured by Wells, Batz, and Mehl8 (carbon in austenite), R. P. smith9 (carbon in ferrite), Wells and Mehl 10 (manganese in austenite), and Kirkaldy and Purdy11 (diffusion of carbon on a manganese gradient in austenite). Kurdjumov 12 has shown that the diffusivity of substitutional elements in ternary austenites up-quenched from martensite may be greatly enhanced, due, apparently, to the presence of a persistent defect structure. The recent work of Krauss 19 showing high densities of tangled dislocations in reversed Fe-Ni martensite is in agreement with this conception. THEORETICAL Diffusion in ternary or higher-order systems may be described with the aid of Onsager's extenbion of Fick's first law,14 in which the flux of each of the n components is assumed to be a linear function of all concentration gradients:

Jan 1, 1964

By C. Laird, C. E. Feltner

An investigation into the effect of a material's slip character on its high strain fatigue properties has been carried out using copper and Cu-7.5 wt pct A1 as representative wavy and planar slip mode materials, respectively. In copper, the constant term and exponent in the Coffin-Manson equation, ?EpNf1/2 = const, are unaffected by the temperature of testing or the history of the metal. On the other hand, in Cu-7.5 wl pcl Al, while the exponent is invariant, the constant is increased with respect to copper and increased further with decrease in temperature. The life in Cu-7.5 wt pct A1 is double that of copper at room temperature and about four times at 78°K. Fractoflaphic studies indicate that these diffevences in life are associated with the early stages of cracking. In copper, severe surface rumpling provides crack nuclei at pain boundaries independent of' testing temperature or the history of the metal. In Cu-7.5 wt pct Al, the planar slip mode of deformation prevents grain boundary folding; crack nucleation is therefore mainly transgranular and the increase in life with respect to copper is associated with a prolonged linking-up stage of such transgranulay cracks. At the lower temperature the increased difficulty of cross slip inhibits multiplication of trans-granular cracks and retards the linking-up process, giving a further enhaancement in life. It is expected that such behavior will be found general in materials of wavy and planar slip mode, respectively. STAGE II crack propagation in high strain fatigue takes place by a repetitive process of crack tip blunting in the tension part of a fatigue cycle followed by resharpening of the crack in the compression part.1-3 This mechanism has come to be called the "plastic relaxation process".3 At very high strains (lives to failure less than 103 cycles), most of the life of a specimen is spent in crack propagation by the plastic relaxation process.4,5 Since the differences in the cyclic strain hardening behavior of metals and alloys are comparatively small in this strain range,6,7 the amount of blunting at a crack tip depends primarily on the applied strain amplitude. The rate of crack propagation in this life range is therefore nearly constant with respect to material and temperature of testing. In the longer life portion of the high strain fatigue range (l03 to 103 cycles to failure), a considerable fraction of the lives is spent in crack initiation and Stage I growth. Grain boundaries have frequently4,5,8 been found to be the sites for this growth which is also associated with a surface folding process.4'5 Such folding behavior may also be general in any material having a grain structure and capable of plastic deformation. It has therefore been suggested475 that the universality of the Coff in-Manson law,9'10 constant,* where ?Ep is the plastic strain range and Nf the number of cycles to failure, is reflected in the approximate invariance of both the crack initiation and propagation behavior. It happens that, in most investigations of the phenomena associated with failure,4,5,8 materials with a wavy slip mode have been used as vehicles for the studies, or else "normally" planar slip mode materials have been tested under conditions giving rise to wavy glide deformation, e.g., stainless steels cycled at high temperature.'' The purpose of the present investigation was to determine whether or not different cracking phenomena would occur when the slip mode of a material is planar and if the associated life to failure would be different than that of a wavy slip

Jan 1, 1968

By G. Townsend Harley

STOPING methods in which shoveling plays an important part are gradually being replaced by other and cheaper methods. But there will always be considerable shoveling done underground in stopes as well as in drifts, tunnels, urinzes, and shafts. At the mines of the PhelpsDodge Corporation at Tyrone, N. M., the cost of shoveling in all stopes in 1917 amounted to 24c. per T. In the top-slice stopes for the same period, it cost 27 c. per T. or 16 per cent, of the total cost of these stopes. The tonnage for shovelers from all stoping was 9.3 T. per man, and for topslicing 8.2 T. per man. These stopes were not unduly hot, and there was not more than the usual amount of timber to interfere with the work of the men. The tonnages obtained per shoveler were considered low; first, because of a poor grade of Mexican labor, many of the men having come in from railroad grading camps; and second, because of a poor spacing of raises, especially in the top-slice stopes, where, in general, they were spaced 25 ft. by 66 ft. (7.6 by 20.1 m.) centers. The average wage per laborer shift was $2.67 during the year. It was thought, however, that even under these conditions the men were not producing the tonnage that they should, so, with the consent of the management, the writer undertook to determine how the general efficiency of the underground shoveling could be improved. No predetermined plan for conducting these experiments was arranged because we had no definite ideas as to the scope of the work or the number of elements into which the investigation would resolve itself, before all of its phases could be determined. We were sure, however, that any work that would thoroughly cover the ground would have to be in the nature of a systematic time study, combined with a course of instruction in correct shoveling methods and adequate and intelligent supervision of the work. Two or three companies in the Southwest have done some work to determine the proper shovel to be used undcrground, but so far as is known the work has been limited to equipping certain parts of their mines with a particular type and size of shovel and thereafter watching the cost and efficiency records. In each case it seems to have been the shovel that held the 21-lb. (9.5-kg.) load that gave the bcst results. Excepting personal communications from these companies, the only data available on scientific shoveling are contained in F. W. Taylor's book, "The Principles of Scientific Management," and D. J. Haucr's article, in The Contractor for March 29, 1918, "A Hundred Hints for Shovelers." This paper discusses and draws conclusions from several thousand time-study readings, taken both on the surface and in the mines for nearly a year. A sufficient number of readings were taken on each factor in the problem for the plotting of curves of the performances and to obtain accurate indices of the work to be expected from this class of labor. The work is not as complete in all of its details as we would like to havc it, because we were forced to stop the work temporarily, owing to the numbcr of men going into the National service and our inability to get others who could make time studies. The results obtained so far, however, have been of such a startling nature that we have decided to submit them at this time, subject to future modification. It is hoped, also, that a free discussion of this paper will lead to a disclosure of any errors that may have becn embodied and offer some valuable suggestions for the conduct of future work.

Jan 1, 1920

By P. E. Baker

An experimental study of heat flow in steam flooding was carried out with steam displacing water in a plane-radial fluid-flow model. Temperature distributions in the model reservoir, overburden, and substratum were measured at frequent intervals during flooding by means of a fixed array of thermocouples. Heat losses to overburden and substratum and the division of heat between the steam and hot-water zones in the reservoir were calculated from the temperature data by performing numerical integrations over the appropriate volumes. Temperature profiles and heat distribution results are presented. Some results were (1) the fraction of injected heat that is lost to overburden and substratum, when expressed as a function of time, did not depend on injection rate; (2) a significant portion of the injected heat was contained in the hot-water zone ahead of and underlying the steam front in the reservoir. The division of heat between the steam and hot-water zones did depend on injection rate; and (3) considerable gravity override was observed. The experimental results are compared with those of several published theoretical studies of heat transfer in hot-fluid injection; approximate agreement on heat losses is found with some of the most idealized theories. A valid comparison cannot be made in regard to steam us hot-water zone; basic interpretation of the hot-water zone, in terms of a vertical steam front, is clouded by gravity override in the experiments. INTRODUCTION The experimental study reported here primarily concerns two problems of heat transfer in steam flooding that have economic and engineering importance. These are (1) the fraction of injected heat lost to the overburden and substratum, and (2) the division of the remaining heat between a zone containing steam and a zone of hot water at lower-than-boiling temperature. The volume of the steamed region, which has a direct effect on oil recovery, can be calculated readily from these results. Several valuable theoretical studies have been published on this problem, but no experimental tests of the theories have previously appeared. Tests were needed because the effects of the necessary simplifications in the theories could not be determined otherwise. The basic, untested theoretical problem is the effect of "decoupling" heat and mass transfer, and the resulting inability to include the effects of heat transfer and condensation on fluid flow. The theoretical problem solved so far has been a heat transfer problem in which fluid is present as a heat carrier and flows in a stipulated manner, either linear or plane-radial. The simplifications that have sometimes been made in regard to thermal conductivity by, for example, Marx and Langenheim,l Lauwerier2 and Thomas3 apparently do not have an important effect on heat loss calculations, although the results of the different methods disagree to a certain extent, as shown by Spillette4 and by Ramey.5 A recent paper by the author6 compared the results of experiments in hot-water flooding with several theoretical results and showed a rather poor agreement. It will be seen that the experimental heat loss results for steam flooding, however, show acceptable agreement. These comparisons are discussed in a later section. EXPERIMENTAL APPARATUS AND PROCEDURE The system studied was one in which a sand pack, initially water saturated and at a uniform temperature, was flooded with steam at a constant rate and temperature. The temperature distribution in the "reservoir" and adjacent "strata" was measured with an array of thermocouples, and heat losses were determined from these measurements. The sand pack contained no oil because the only concern of the experiments was heat flow, and it was important to have the flow pattern as simple

Jan 1, 1970

By P. H. Turnock, R. D. Pehlke

The effects of mutual interactions between elements on the solubility of nitrogen in liquid iron alloys containing chrormium, columbium, molybdenurn, nickel, or silicon have been determined. The equilibrium nitrogen solubilities in the liquid iron alloys were measured by the Sieverts method. The second-order effects caused by the presence of chrormium, columbium , or silicon in the melt were found to be significant. The solubility of nitrogen in liquid iron alloys containing several alloying elements has also been measured as a function of melt temperature and nitrogen pressure. The heat of solution of nitrogen in microcomponent iron alloys has been found to be a function of the logarithnz of the activity coefficient of nitrogen, irrvespectilje of the composition of the melt. The heat of solution of nitrogen in pure liquid iron was determined to be 1265 ± 450 cal per g-atom of N over the temperatxue range 1600° to 1800°. Sieverts ' law was obeyed for all melt compositions studied in the pressure interval 0.4 to 1.0 atm. The solubility of nitrogen in liquid pure iron and in many dilute binary liquid-iron alloys has been the subject of many investigations. Pehlke and Elliott' made a comprehensive investigation on this subject in which they summarized and compared previous researches. However, the practical application of these solubility studies requires prediction of the nitrogen solubility in the multicomponent alloys encountered in operating practice. The interaction parameter as defined by wagner2 has been used to predict the behavior of a solute in a complex alloy, and a modification of this approach has met with reasonable success in several instances.'1"5 The necessity for extending the nitrogen solubility data on binary alloys is caused by the lack of direct measurements of the solubility of nitrogen in molten alloy steels. A direct comparison between measured solubilities and those predicted for multicomponent alloys based on data in binary systems has been made in only a few instances, and most of these were reviewed by Langenberg,6 who developed a graphical technique for predicting nitrogen solubilities in multicomponent iron alloys. The solubility of nitrogen has also been studied over the entire Fe-Cr-Ni ternary system by Humbert and Elliott.? Both studies compared the solubilities computed from binary iron alloy data with those measured in ternary alloys in the liquid state at 1600°C. Most of the nitrogen solubility measurements referred to above were for iron alloys at 1600°C; thus, the ability to predict nitrogen solubilities in molten iron alloys has been restricted primarily to alloys at 1600°C. Recently, Nelsson8 devised an approximate method for calculating the change in nitrogen solubility in molten iron alloys as a function of temperature, but an assumption of regular solution behavior limits the applicability of the method. Chipman and Corrigan9,10 have developed an empirical correlation. based on the activity coefficient of nitrogen in the iron alloy at 1600°C, which facilitates the estimation of nitrogen solubilities at higher temperatures. This method also has limitations, due to the necessity of calculating the activity coefficient for nitrogen in a multicomponent iron alloy at 1600°C from the nitrogen solubility data for binary iron alloys. The shortcoming of such a straightforward approach to the problem of predicting nitrogen solubilities is the disregard of mutual interactions between elements in a complex iron melt. This investigation was conducted in an attempt to provide a more reliable method of predicting nitrogen solubilities in multicomponent iron alloys. The effects of the mutual interactions between chromium, nickel, silicon, molybdenum, or columbium in iron were studied, using the Sieverts' technique1,7 to measure the effects of alloy composition, temperature, and pressure on the solubility of nitrogen in ternary liquid iron alloys.

Jan 1, 1967

By G. H. Gronhovd, W. H. Oppelt

Complete gasification to produce synthesis gas for manufacture of synthetic chemicals and fuels offers possibilities for utilization of the vast lignite fuel deposits in the U.S. The design of a slugging, fixed-bed, pressure gasification pilot Plant is described. Results are presented for initial operation of the plant at moderate pressure and using low-temperature char as fuel. Gasification of solid fuels is becoming increasingly important in the future overall energy picture, both in this country and abroad. Snythesis gas from gasification of coal consists chiefly of carbon monoxide and hydrogen, and after suitable purification and modification this gas can be used in the production of synthetic ammonia, methanol, higher alcohols, and synthetic liquid fuels, or as a reducing gas in the metallurgical field.' Most synthesis gas used in this country is made from natural gas, petroleum oils, and refining tail-gases, but the direct gasification of coal will become more attractive as costs of petroleum oils and natural gas increase.' A gasification process designed for use with brown coal and lignite was developed by the German Lurgi Co. prior to World War 11. In this process a fixed bed of lump-sized fuel is gasified with steam and oxygen at pressures ranging from 370 to 440 psi. Effective heat exchange between the ascending high-temperature gases and the descending solid fuel results in low gas outlet temperatures, high thermal efficiency, and favorable oxygen consumption. Lurgi gasifiers are used commercially for the production of synthesis gas or as a substitute for coal gas in many countries, including Germany, Czechslovakia, Great Britain, South Africa, Australia and Pakistan.3-6 The chief disadvantage of the conventional Lurgi gasifier which operates under nonclinkering conditions is that large quantities of excess steam must be supplied with the oxygen-steam mixture to reduce the temperature in the fuel bed below fusion temperature of the ash. Most of the steam leaves the gasifier undecomposed with the product gas and, after cooling, produces large amounts of dilute liquor. Evaluation of fixed-bed gasification experiments using dry-ash removal indicated that by increasing the oxygen concentration in the oxygen-steam blast the temperature in the combustion zone could be raised high enough to operate under slagging conditions. 7-8 Operation of a fixed-bed pressure gasifier under slagging conditions would reduce the steam requirements and the volume of waste liquor and would increase the steam decomposition, the overall thermal efficiency of the process, and the rate of gas production per unit cross sectional area of the gasifier. Because the Lurgi gasification process, modified for slagging operation, offered possibilities for efficient gasification of American lignite, the Bureau of Mines decided to investigate the process on a pilot-plant scale at the Grand Forks, N. D., Lignite Research Laboratory. DESIGN Description of Pilot Plant: A flowsheet of the Grand Forks slagging gasification pilot plant is presented in Fig. 1. The gasifier unit consits of the coal lock, the gas generator, and the slag lock. The fuel, which is periodically charged to the coal lock, moves by gravity flow into the generator and is continuously gasified by an oxygen-steam mixture introduced at the bottom of the gasifier. Molten slag is formed at the hearth and flows through a central taphole into a water quench bath. The slag-water slurry is periodically discharged from the slag lock. Process steam is generated and superheated in separately-fired, oil-burning units. Oxygen is supplied to the gasifier from a gas storage bank located some distance from the pilot plant. The capacity of the oxygen storage bank is about 130,000 standard cu ft at 2200 psig. Liquid oxygen is delivered by truck and is converted to gas during the filling operation. The superheated steam and oxygen are metered separately, then mixed and fed to the gasifier tuyeres. The gas generator is surrounded by a water jacket. The steam formed is separated in the steam drum and then discharged to the gas offtake, thus equaliz-

Jan 1, 1961

By J. Handin

Triaxial compression tests with independently applied external confining pressures and internal pore pressures show that the ultimate compressive strengths of representative oil well cements are nearly linear functions of effective pressure — the difference between external and internal pressures on the jacketed cylindrical specimens (to 15,000 psi). The strengths are little affected by the test temperature to 350F (not to be confused with the curing temperature). At an effective pressure of 15,000 psi, strengths are in the rage of 30,000 to 50,000 psi, comparable to those of sedimentary rocks under similar conditions. The cements become very ductile even under low effective pressures; permanent shortenings of 30 per cent or more are attainable without rupture. INTRODUCTION Since the pioneering work of Richart, Brandtzaeg and Brown1 on the failure of cement under combined compressive stresses, it has been recognized that ultimate compressive strength is greatly enhanced by the application of confining pressure. More recently, Mc Henry2 showed that the strength of concrete was a linear function of the effective pressure (the difference between the external confining pressure on a jacketed specimen and the internal fluid pore pressure), at least for a range of 0 to 1,500 psi. The effect of temperature had not been investigated, and no previous systematic triaxial compression testing of naterials used for oilwell cementing seems to have been done. The present work was suggested by the late J. hi. Bugbee3 who stated that "consideration of the common application of high-pressure hydraulic fracturing to the initial completion or recompletion of wells, and the large pressure drawdowns in some producing wells, particularly those in abnormally high-pressure gas-condensate reservoirs, raises the question of what is a suitable cement strength for various completions. The intuitive belief exists that cement strength need be no greater than formation strength. Tests should, however, be conducted at downhole conditions." The ultimate compressive strengths of rocks penetrated by the borehole must rise several fold with increasing depth.4 This marked enhancement of strength is due to the influence of the effective pressure, the total weight per unit area of the overburden less the hydrostatic pore pressure. (The effect of temperature due to the geothermal gradients is relatively small for depths to 30,000 ft.) .A significant comparison of the strengths of rocks and cements at downhole conditions requires knowledge of the confined compressive strengths of cements as well. EXPERIMENTAL PROCEDURES The theory and technique of triaxial compression testing are fully discussed in earlier reports.4-6 Briefly, cylindrical specimens 1-in. long and 0.5-in. in diameter are jacketed in thin copper tubes of negligible strength, placed in the test chamber and subjected to an external confining pressure of kerosene and loaded axially by the piston at a strain rate of 1 per cent per minute. Pore pressures of water (or kerosene) are applied independently through the hollow piston and are maintained constant during the shortening of the specimen, Tests at sensibly 0 pore pressure are arranged so that any free water in the cement can escape to the atmosphere. (If egress of water were denied, pore pressure would rapidly attain the value of the external confining pressure because of reduction of pore space.) The test chamber can be heated for high-temperature experiments. Unless otherwise noted, the cement samples were air dried for about a week. Recorded during a test are pore and confining pressures, shortening and axial differential force (total force less the product of the confining pressure and the area of the piston). True stress-strain curves are constructed by dividing the measured differential force by the actual cross-sectional area of the specimen at a given

Jan 1, 1966

By H. I. Aaronson, S. Toney

When the component grains of .ferritic hicrystals of an Fe-1.55 pct Si alloy are disoriented through an angle "6 " about a conzmon [ll0] axis, the tendency for preferential growth of austenite crystals along the grain boundary during transformation at elevated temperatures is small when 11 deg, but increases rapidly at larger angles. This type of orientation-dependence indicates that grain boundary diffdsion promotes preferential growth along large-angle boundaries. Morphological differences between austenite crystals formed at small-angle [1101 and [loo] boundaries suggest that precipitate morphology can be dependent on the dislocation structure of the boundary. ThE morphology of precipitate crystals nucleated at a grain boundary can be significantly affected by the structure of the boundary.' The limited amount of experimental evidence available in the literature indicates that the morphological effects of boundaries made up of arrays of dislocations, such as subbound-aries and small-angle grain boundaries, are different from those of boundaries having essentially disordered structures, i.e., large-angle grain boundaries. On the basis of indirect evidence, it has been concluded that large-angle grain boundaries give rise to the formation of grain boundary allotriomorphs (crystals which nucleate at grain boundaries, and grow preferentially and more or less smoothly along them)2 in the proeutectoid ferrite and the proeutec-toid cementite reactions in plain-carbon steels, and apparently also in many non-ferrous alloys.' At small-angle grain boundaries in a plain carbon steel, on the other hand, ferrite crystals were found to take the form of primary side plates. Similarly, Guinie, loys. Primary sideplates formed at a subboundary with a constant orientation tend to be parallel to only one, or occasionally two matrix habit planes, and a marked change in the orientation of the boundary is accompanied by a change in the habit plane. Previous studies on the morphological effects of grain boundary structure were performed on poly-crystalline aggregates. Information on the disorien-tation of the pairs of grains forming the boundaries at which the various morphologies appeared in these specimens was largely either qualitative or semi-quantitative. Precipitate morphologies accordingly could not be accurately and systematically correlated with grain boundary structure, and thus theories which have been proposed for the various morphological effects could not be satisfactorily tested. This investigation was undertaken in an attempt to remedy these deficiencies by studying the morphological effects of grain boundary structure with a method in which the boundaries are formed by matrix grains whose disorientations are known and controlled with reasonable accuracy. EXPERIMENTAL PROCEDURE The crystallographic requirements of this study were fulfilled by means of oriented bicrystals of silicon-iron. Disorientation of the component ferrite crystals was carried out about common, major crystallographic axes through angles ranging from 1/2 to 44 deg. The silicon content was low enough SO that the bicrystals could be partially transformed to austenite by heating to elevated temperatures. The silicon-iron used had the following initial composition: 1.55 pct Si, 0.04 pct C, 0.0031 pct N, 0.17 pct Mn, 0.020 pct S, and 0.002 pct P. The alloy was obtained in the form of 0.036-in. sheet. The procedures employed to prepare seed crystals in strips of this sheet, to reorient the seeds, and to grow them into bicrystals are essentially those described by Dunn and Nonken and aynes." The characteristics of the bicrystals are given in Table I. The "bicrystal type" indicates the crystallographic plane parallel to the broad faces of the strip in both grains and the crystallographic direction which was parallel to the long edges of the strip in both grains prior to disorientation. The angular disorientation of the grains, 8, which was performed about the direction normal to the plane of the broad faces of the strip, was measured between the [ 00l] directions. Orientation of the grain boundary, , was taken as the angle between the plane of the grain boundary and a plane containing the axis of disorienta-

Jan 1, 1962

By R. E. Collins

The existence of fluid migration across fixed boundaries in oil and gas reservoirs has been known for many years. Several techniques have been developed in the past for estimating The rate of migration across fixed boundaries as an aid in planninng field development and in the valuation of oilfied properties. The principal de/errc7nt to the use of these techniques lies in the rather extensive reservoir and field data required for ca1cula1ion.s of fluid migration. For this reason, a new, simplified procedure has been developed which makes possible the calculation of fluid migration with a minimum of field and resrvoir data. This new technique is based on certain. solutions of the differential equations describing flour in the reservoir which assume, for the portion of the reservoir of interest, that the formation can be approximated by a homogeneous rock of uniforrn thickness, that only a Single mobile fluid phase exists, and {hut fluid production at the well is solely a result of expun.rioti of rile resersoir fluds. The results of the, present work are compiled in a set of curves. These curves can he used to calculate both rule and cummulative fluid migration when the abovc assumptions ore ju.r/ifiril. The only data required for such calcu1ations are the production hisrories of all wells in the field, the permeability and porosity of the reservoir., the, compressibility and viscosity of the fluid. and the shape of the reservoir, INTRODUCTION The migration of fluids across fixed boundaries in oil and gas reservoirs has long been recognized as an engineering and economic problem, but no methods for estimating the extent of migration during the development stage of a reservoir have been published. The investigation reported here has been directed toward providing a simple technique for estimating the extent of such migration and has therefore been restricted to the simple case of a homogeneous reservoir of uniform thickncss containing compressible liquids. It is assumed that only a single mobile fluid phase exists and that fluid production at the wells is solely by expansion of the reservoir fluids. Obviously, few, if any. reservoirs conform to these assumptions during their entire productive life. On the other hand, most reservoirs approximate fluid expansion reservoirs during their initial stages of primary production. Consequently, the present work should be applicable during the field development of most reservoirs, regardless of whether or not they ultimately are gas cap, dissolved-gas, or water-drive reservoirs. The mathematical analysis of this problem has been outlined in the Appendix, while the results of the investigation are presented in the following discussion. DISCUSSION OF CALCULATION PROCEDURE SLNGLE Well, Ini-iimite Reservoir The procedure for calculating fluid migrations is based on solutions of the partial differentia1 equations describing the flow of a homogeneous compressible liquid in a homogeneous reservoir of uniform thickness. This differential equation has the form of the well-known diffusion equation. The equation, as applicable to our problem, is where p is fluid density, is porosity, EL is viscosity, c is compressibility,* and k is permeability; .r, y and t are the space coordinates and time. Derivation of this equation is given in the Appendix. The analytical solution of this equation for a single well of negligible radius producing at a constant rate. q, from a reservoir of infinite areal extent for a time, 1. can he readily obtained. From this solution the rate of flow and also the cumulative flow across a straight line segment of length, 1, at a perpendicular distance. 6. from the well can be computed. One end of the line must be on the perpendicular joining the well to the

By R. R. Boucher, O. J. C. Runnalls

A pronounced thermal effect has been observed on heating or cooling a1wninum-rich Al-U and Al-Pu alloys. From microscopic and X-ray diffractionstudies, the effectl has been attributed to trnsfor)nations in the inlemetallic phases UAl, and PuAl , involuing rearrangement or cluslering of vacancies The transformation obsrved in Al-20 wi oct U alloys occurred near the a Al-UAl4 eutectic tem-perature of 646°C making it difjicult to detect from heating curves. When cooling, howeoer, thermal mad X-ray analyses indicnted that the high-temperati~re phase labeled p UAL, precipitated initially during euitectic solidification. As solidification continued and after an induction period o-f 5 lo 20 tnin the nu-cleatiorz and growth of the low- twmperature phase, a UAl4 produced an euolution of energy which raised the observed alloy temperature by 1° to 2°C. in Al-Pu alloys the transformation occurred at 645°C, 5 deg below the eutectic solidification temperccture. The eslimated latent heat of transformation was 25 +5 cnl per g PuAl,. The high-tetnpevcl-ture forms of UAl, and PuAl, were easily 9-etccined by chill casting Al-U and Al-Pu alloys and on reheating the alloys the trcltzsforrnation to the a phase was sluggish even above 600°C. Separated single cryslals o-f UAl4, joy example, showed no evidence Of transformalion after heating for 16 hr at 620°C. The transformation rates increased yapidly enough with increasing temperature, however, that a a spontaneous liberation of energy lens observed on occasion when bealing cast Al-20 wt pet U to neal- the RECENT phase studies on A1-Pu alloys have indicated that earlier reported equilibrium diagrams should be modified to take account of several poly- morphic transformations in the intermetallic compound PuAl,.' During the course of the latter work an unexpected heat evolution was observed while cooling a solid alloy consisting of a mixture of a aluminum and PuA1,. The evolution occurred at 645°C, y below the a A1-Pu A1, eutectic solidification temperature. Subsequent thermal-analysis experiments with A1-U alloys showed that heat was evolved there also at a temperature nearly coincident with the a A1-UA1, eutectic solidification temperature of 646°. A search of the literature revealed that similar thermal effects had been observed before for both A1-U and A1-Pu but had never been fully explained.2,3 The purpose of the present paper is to describe thermal-analysis and X-ray diffraction experiments which have led to the conclusion that the thermal effects are due to transformations which involve the rearrangement of vacancies in the intermetallic compounds UA1, and PuA1,. EXPERIMENTAL PROCEDURE The A1-U alloys were prepared by heating super-pure aluminum (99.99 pct +) obtained from the Aluminum Co. of Canada in a graphite crucible surrounded by an induction coil. At 1000°C high-purity uranium metal produced by the Eldorado Mining and Refining Co., Ltd. was dropped into the liquid aluminum and was dissolved completely within 15 min. The impurities in the uranium were A1—4 ppm, Ca-4, C-70, Cr-3, Cu-6, Fe-30, Mg-2, Mn-10, Ni-25, Si-25, and Zn-2. The A1-Pu alloys were prepared by reacting plutonium dioxide with liquid aluminum under cryolite (3 NaF . AlF,) for 10 min at 1100°C in graphite crucibles exposed to the air. The plutonium was 99.8 pct pure, containing trace impurities of Ca, Cu, Fe, Pb, Si, and Zn. The liquid alloys were usually poured into water-cooled molds to provide castings of uniform composition from which samples could be cut for heat treatment and thermal-analysis experiments.

Jan 1, 1965

By D. H. Polonis, Gary A. Dreyer

This investigation involved a study of the reactions occurring during aging of supersaturated a phase in a CIL-Si alloy. The aging processes at temperatures below 552°C were studied by means of metallography, microhardness, and electrical resistivity measurements. During the initial stage of aging in the temperature range 275o to 525oC an anisotropic phase, which is visible in the form of striations in the micro-structure formed prior to the appearance of the -phase. This transition phase has been designated k' due to its structural similarity to the k phase. The rate of K' formation during aging increased with aging temperature. At 450o and 525°C the reaction appeared to reach completion before significant amounts of precipitation were observed. The overall precipitation reaction in this system involved four stages: formation of formation of at a grain boundaries, growth of particles in 7-egions, and coagulation of into large particles. The relative role of each of these processes is dependent on the temperature of aging. The formation of phase from supersaturated a during aging increases the electrical resistivity and slightly increases the hardness of a 4.9 pct silicon alloy. The formation of during aging decreases the electrical resistivity and increases the hardness until coalescence of particles begins during the final stages of aging. The coalescence of results in a decrease of both electvical resistivity and hardness. THE available literature does not record a detailed analysis of the aging processes in Cu-Si alloys. Such an analysis has been difficult due to both the limited data available and misleading microstructural observations which have been reported. Previous work suggests that a rather complicated series of processes may be involved in the aging of supersaturated a. The present investigation is concerned with an analysis of the sequence of these aging processes in a 4.9 pct Si alloy. The currently accepted version of the Cu-Si equilibrium diagram, Fig. 1, was published by C. S. Smith in 1940.' This diagram shows a eutectoid at 5.2 pct Si with the high temperature phase, having its composition range entirely to the right of this value. By quenching alloys in the composition range 4.6 to 5.2 pct Si from the region, Smith was able to retain the phase. However, this phase exhibited a striated microstructure which was anisotropic to polarized light. These striations were suggested by Smith to be duplex twins. Using an oscillating crystal X-ray method Barrett2 studied alloys containing 4.0 to 5.4 pct Si quenched from the and regions. The striations observed in quenched specimens were attributed by Barrett to quenching stresses. He found that striations could also be formed by deformation after quenching and identified them as stacking faults in alloys containing 4.0 to 5.0 pct Si. Barrett related the striations in alloys containing 5.0 to 5.4 pct Si both to stacking faults and to a metastable phase which he called The striated structure reported by smith' was also obtained by Hoffmann, et al. 3 By deeply etching their specimens the anisotropic phase on the surface was removed, leaving a structure which was microscopically similar to pure copper. By means of X-ray analysis before and after deformation at room temperature it was concluded that the anisotropic phase was metastable This conclusion has been questioned by Barrett who attributes the discrepancy to superposition of k and lines on the powder patterns of Hoffmann, et alm3 The decomposition of supersaturated according to smith' and Anderson,4 is very sluggish. However, Nony and Dreyer,5 and Hoffmann, et al.' found that deformation increases the rate of decomposition. X-ray studies by Hoffmann, et al on

Jan 1, 1962

By G. Krauss, J. M. Thompson, H. Y. Kumagai

Boron-doped p-type and arsenic-doped n-type source materials were used to deposit thin silicon films on amorphous fused quartz substrates by cathodic sputtering in argon atmospheres. All as-sputtered films were found to have high electrical resistivities of about 104 ohm-cm, despite controlled variation of the substrate temperature from 50O to 400°C. The high resistivity of the n-type films persisted even after postdeposition heat treatments as high as 1000°C, while p-type films showed a sharp decrease in re-sistivity to values on the order of 1 ohm-cm or lower after heat treatments above 650°C. X-ray and electron diffraction, together with transmission electron microscopy of the p-type films, revealed that the improved electrical properties resulted from two distinct processes. A primary crystallization or recovery within the pains, which were approximately 200Å diam in the as-sputtered films, accompanied the initial sharp drop in resistivity at 650°C. Following this process, a secondary recrystallization or discontinuous grain growth was nucleated in films heated at 1000°C. The resulting significantly increased size of certain grains and the corresponding decrease in grain boundary area accompanied an additional decrease in resistivity after heat treatment at 1000°C. The properties of thin silicon films are of interest because of their possible application in device technology. Although evaporated silicon films on amorphous substrates1-3 and both sputtered and evaporated germanium films4-' have been studied in some detail, sputtered silicon films have not been widely investigated. It has been observed that evaporated silicon films deposited at substrate temperatures below approximately 600°C are amorphous.1,3,9,10 At substrate temperatures higher than this, the films are polycrys-talline, sometimes exhibiting a fiber texture with the fiber axis normal to the substrate surface.' Similar results have been obtained with germanium films, although in this case the substrate temperature for the amorphous to crystalline transformation is reported to be as low as 150OC.5 The mechanism of film deposition by cathodic sputtering differs markedly from that of thermal evaporation. It might be expected, therefore, that the properties of sputtered and evaporated silicon films would also differ. The purpose of our study was to determine the effects of deposition conditions and heat treatments on the electrical resistivity, structure, and composition of thin silicon films sputtered on fused quartz substrates. EXPERIMENTAL DETAILS Apparatus. A schematic diagram of the sputtering apparatus is shown in Fig. 1. The vacuum chamber was enclosed by an 18-in.-diam by 12-in.-high Pyrex bell jar mounted on a stainless-steel base plate. Access holes in the base plate were used to supply argon gas to the chamber, to allow for various electrical feedthroughs, and to provide mechanical motion to the shutter between the source material and substrate. The chamber was evacuated by a pumping system consisting of a 720 liter per sec oil diffusion pump backed by a 13 cu ft per min mechanical pump. Ultimate pressures of less than 1 x 10-6 torr could be achieved. The substrates were securely mounted on the stainless-steel substrate holder by quartz tabs held in place by stainless-steel screws. Temperatures of the substrate holder were measured by a thermocouple inserted within the holder. A quartz lamp was used to heat the holder and temperatures as high as 420°C could be maintained. Deposition Conditions and Procedure. The substrates used were polished fused quartz slides 116 by 1-12 by 3 in. and were outgassed at 200°C for 1 hr at a pressure less than 1 x 10-5 torr before film deposition. This was followed by a presputtering period of 2 hr during which the movable stainless-steel shutter was so positioned that deposition on the substrate was prevented. At a sputtering voltage of 5000 v dc and an argon pressure of 0.03 torr, the cathode cur-

Jan 1, 1967

By S. H. McFarlane, K. H. Zaininger, G. W. Cullen, C. C. Wang, G. E. Gottlieb

Heteroepitaxy of silicon on stoichiometric spinel has been studied. Both boron-doped (p-type) and arsenic-doped (n-type) single-crystal silicon films have been grown by the pyrolysis of silane on sioichiometric spinel (MgAl2O4) substrates prepared by a flux technique. The spinel crystals are free from strain and subgrains, and exhibit no thermal instability at temperatures employed for silicon deposition. The growth parameters of epitaxial silicon have been studied. Effects of substrate orientation, surface preparation, and growth conditions on the silicon Properties have been determined. The silicon-spinel composite was characterized by X-ray diffraction, electron diffraction, electron microscopy, and optical techniques. Information on epitaxial orientation relationship, defect structures and film-substrate interface has been obtained. A parallel orientation relationship was found between silicon and spinel. The silicon films exhibit no subgrain structures. However, growth pyramids, microtwins, stacking faults, and dislocations are the prevalent defects. The deformation of epitaxial silicon was measured by an X-ray technique. It was found that the epitaxial silicon is strained under compres-sive stress due to the thermal expansion difference between the film and substrate. The stress was estimated using the classical plate theory and beam equation. Electrical properties of the epitaxial silicon have been measured, including resistivity, carrier concentration, and Hall mobility. MIS and vertical through diffused junction structures have been fabricated wing silicon-spinel composite. Information on MIS characteristics, minority carrier life time, and junction properties has been obtained. THE achievement of single-crystal growth of large area silicon semiconductor films on oxide insulating substrates is of technical importance to many solid state electronic devices. The semiconductor-insulator composite structure is also of scientific interest because the heteroepitaxy is determined by the spatial relationship between the atomic arrangement in the substrate and that of the atoms in the appropriate crystallographic plane of silicon. The degree of crystalline perfection of the silicon deposits depends also on the physical condition of substrate surface. Epitaxial growth of silicon on commercial flame fusion Al-rich spinel (MgO:3.3A12O3) has been reported by several investigators.'-4 The lattice of spinel of this composition is distorted. The material cracks and exsolves the excess alumina in the lattice at temperatures (-1200°C) used in fabricating silicon devices. Stoichiometric MgA12O4 spinel exhibits neither lattice distortion nor thermal instability. Results are given on the epitaxial growth and characterization of silicon on stoichiometric spinel. EXPERIMENTAL Fabrication of Silicon-Spinel Composite Structures. The MgA12O4 spinel substrate wafers were prepared from bulk single crystals grown at about 1200°C by a flux technique5 using a PbF2 flux. The crystals are free from strain and subgrains, and exhibit a typical dislocation density of less than 50 lines per sq cm. The crystals are thermally stable without cracking and exsolution of alumina to at least 1500°C. The spinel crystals were oriented for cutting by the Laue X-ray back-reflection method.6 Substrate wafers (about 25 mil thick) of (Ill), (loo), and (110) orientations have been obtained bu cutting the X-ray oriented crystals using a standard type diamond wheel. An accuracy of better than 1/2 deg is maintained throughout the operation. The spinel substrate wafers were mechanically lapped and polished to produce the flat and smooth surface which is required for silicon epitaxy. Lapping was carried out with about 30-µ boron carbide abrasives to obtain a flat coplanar surface. The lapped surface was further polished using successively finer grades of alumina, ending with the 0.3-µ grade. After polishing the wafers generally have a flatness of ±0.4 µ per cm, as revealed by interferometry. Fig. 1 shows (111) oriented MgA12O4 spinel wafers of various shapes.

Jan 1, 1970

By Lawrence A. Shepard, Richard H. Krock

A series of ductile, two phase W-Ni-Fe composites, sintered in the presence of a liquid phase, were tested in tension. Identical room temperature stress-strain curves were obtained for specimens containing from 80 to 92 wt pct W (58 to 75 vol pct W particles). The composites exhibited a maximum elongation of 29pct at room temperature, and 10.7 pct at 77 °K. The tungsten particles in the composite elongated by the same amount at these temperatures. Single-phase alloy specimens matching the composition of the composite matrix showed about one half the flow stress of the composites. The test results demonstrated that the mechanical properties of W-Ni-Fe composites are determined by the tungsten particles alone and are independent of matrix volume fraction or mean free path over the composition range studied. A general prediction of the plastic behavior of two-phase alloys or composites from a knowledge of the properties, size, shape, and dispersion of the individual components is not, at present, possible. The excellent theoretical and experimental exposition of the dispersion-strengthening problem applies only to cases where the stronger phase is present as fine particles and in volume fractions of a few percent. Materials in which the stronger phase constitutes a significant fraction of the volume, although of broad engineering interest, have received comparatively little analytical attention. This important group of materials is the concern of the present study. The degree of deformation of each phase during the plastic working of a ductile two-phase alloy has been established by Boas and Honeycombe,1 Clare-brough,2 and Clarebrough and berger.3 These workers empirically correlated the recrystallization temperature of the individual phases with the degree of cold work sustained by each. It was shown that the deformations of both phases are equal and therefore the same as that sustained by the composite whole when the volume fraction of the stronger phase exceeds 35 pct. At yielding, however, deformation begins in the weaker phase. A direct correlation of composite strength, as a function of deformation, with the strength of the individual components cannot be made. In the presence of stronger phase particles, the flow stress of the weaker matrix phase is enhanced by an additional hydrostatic component. The development and potential magnitude of matrix strength promotion upon straining has not even been experimentally established. Quantitative work in this field has been confined to ceramic-metal composites. These materials fracture at strains considerably below 1 pct by cracking of the ceramic particles.4-6 An indication of the potential constrained matrix strength capacity is given by the observation by Nishimatsu and Gur-land4 that the maximum tensile strength of a WC-Co composite is almost double the bulk strength of the cobalt matrix. The relationship between composite strength and geometry is ordinarily expressed in terms of the mean-free matrix path (referred to subsequently as MFP) between the stronger particles.* This factor takes into account both the volume fraction of matrix phase and the average size of the particles. unke17 concluded that composite strength varied directly with the decreasing log MFP, following the Gen-samer8 relationship. However, Gurland and ~radzil~ reported a strength maximum at an MFP of the order of 1 p in WC-Co composites. A ductile two phase metallic composite was chosen for the present investigation of composite mechanical properties with the view of avoiding the difficulties and ambiguities noted above. The W-Ni-Fe heavy alloy of Green, Jones, and pitkin9 contains 80 to 94 wt pct W, and nickel and iron in a weight ratio

Jan 1, 1963

By Stuart S. Forbes

Changes and additions made to the Canadian Copper Refiners Ltd. electrolytic refinery between 1949 and 1955 are reviewed. The effect of high current density on current efficiency and section work is discussed. OPERATING and physical changes which have been made to the electrolytic tank house of Canadian Copper Refiners Ltd. suggest a review of present conditions. Previous papers1 a have described the general arrangement. Since then, expansion, closer anode spacing, and higher current densities have increased annual cathode capacity from 112,000 to 182,000 tons. A recent extension to the tank house, begun in the spring of 1954 and completed in August 1955, was undertaken to handle the monthly production of about 3,000 tons of anodes from Gasp6 Copper Mines. The anticipation of these additional receipts and large stocks of unrefined copper caused by the increase in custom smelting at Noranda made it necessary to increase production without immediately available additional cell capacity. Accordingly, from December 1954 through March 1955, the tank house was operated successfully at a current density of 24.0 amp per sq ft. Again, as of September 1955, one of the two tank house circuits is being operated at this high density. Although other refineries may operate at current densities higher than 24 amp per sq ft, none is doing so with anodes which contain silver and gold content as high as those being refined at Canadian Copper Refiners Ltd. General Arrangement The original building consisted of two parallel bays, each 660 ft long and 60 ft wide, with a 24 ft pump bay extending along the west side of the building. A previous expansion, completed in 1939, was to the west of the original building and consisted of two 60 ft parallel bays 240 ft long. The length of these two bays was extended in 1954 to 540 ft, adding one stripper and 12 commercial sections to No. 2 electrical circuit. Space is still available for an additional six commercial sections and one stripper section. Construction on these seven sections started in September 1955, and will be completed early in 1956. The tank house capacity will then be 206,000 tons of cathodes per year. In the two tank houses there is a total of 900 cells distributed as follows—commercial, 468; stripper, 36; total, 504 in the No. 1 tank house (No. 1 circuit): commercial, 360; stripper, 36; total, 396 in the No. 2 tank house (No. 2 circuit). Section arrangement consists of nine cells to a tier and two tiers or 18 cells to a section. Anodes, weighing 620 lb each, are cast with a Baltimore groove and are refined by the multiple system. Solution enters each cell through a bottom inlet at one end and overflows at the top of the opposite end. Rate of solution flow through the cells is controlled by valves at each section to 4.5 gpm. Electrical power to the tank house is now provided by nine 675 kw motor generator sets. Each set consists of a 135-12 v 5000 amp dc generator, driven by a 980 hp 2300 v synchronous motor. When No. 1 circuit is operated at 22,000 amp, five sets, at 4400 amp each, are connected in parallel. As the maximum allowable current on No. 2 electrical circuit has recently been set at 18,000 amp, only four sets are connected in parallel on No. 2 circuit. A tenth set, now being installed, will be reserved as a spare, for use in either circuit as required. Recent Changes In 1949, the annual capacity of the tank house was increased from 112,000 to 132,000 tons by closing the

Jan 1, 1957

By Bernard G. Carbajal

The dielectric properties and thermal stability of glow-discharge polytnerized films of styrene, chloro-benzene, and other organics and photoresist filrns are presented. Variations in the glow-discharge polymerization procedure are related to thermal stability. Thermal treatwzents to improve the dielectric properties of both positive (A2 340) and negative (KMER) photoresists are given. THIS study of the dielectric properties of thin organic polymer films is part of a program to investigate the electronic applications of organic materials. Thin organic polymer films have been produced in an ac glow discharge,13 with a low-energy electron-beam process,4 and in an rf dicharge. A process has been developed during the course of this program for forming thin organic polymer films in a dc glow discharge. These glow-discharge polymerized (GDP) films exhibit some useful dielectric and insulative properties. Variations in the properties of the GDP films can be induced by changes in the source material (monomer) andor in the conditions of the glow discharge. Thin organic polymer films can also be produced from typical positive and negative photoresists. Some variations in the properties of the photoresist films can be induced by suitable post-application treatments such as thermal crosslinking. The properties of interest in this study are dielectric constant, dielectric strength, resistivity, and thermal stability. POLYMER FILM PREPARATION I) GDP Films. The GDP films were prepared in a dc glow discharge in an atmosphere of organic monomer vapor. The details of the glow-discharge procedure have been presented previously.6 Briefly, under the influence of a dc glow discharge organic vapors can be polymerized in the vapor phase and the resultant polymer (GDP film) deposits on and in the vicinity of the cathode. Substrates placed behind a screen cathode can be uniformly coated over all exposed areas. Pattern definition can be achieved using normal mask stencil techniques. Table I presents the pertinent parameters in the preparation of GDP styrene. Conditions of Forrrzation of GDP Styrene. The rate of deposition for GDP films was monitored using a current-time integrator. The thickness of the deposited film for a given set of experimental conditions was a function of the current-time integral. The fifth row in Table I is a compilation of the deposition rate of GDP styrene films. Other organic monomers behave in a similar fashion. With the exception of the highly halogenated saturated hydrocarbons, all organics subjected to the conditions of the glow discharge form GDP films. 11) Photoresist Films. Photoresists are normally recognized to exist in two forms, a low molecular weight, soluble polymer form (A) and a high molecular weight, crosslinked insoluble form (B). Typical negative photoresists are applied in the low molecular weight form A and converted to the high molecular weight form B by the action of light. (Positive photoresists are applied in form B and converted to form A by the action of light.) Eastman Kodak Metal Etch Resist (KMER) has been chosen as a typical negative photoresist. Polymer film samples were prepared in the same configuration used for GDP films. The KMER was applied in the usual fashion using a 120°C drying step (infrared-lamp heating) and a 180°C post-bake. Care must be taken to completely remove the solvent in the drying step and to properly develop and postbake to exposed photoresist. If the photoresist film is not properly dried and postbaked, severe blistering and peeling will occur during subsequent elevated-temperature exposure. Thickness can be controlled by varying the viscosity of the form A photoresist and the spin rate at application. AZ 340 was chosen as a typical positive photore-

Jan 1, 1967