Search Documents

By Klaus Schwerdtfeger

The rate of formation of silicon monoxide from silica spheres has been measured at 1500°C in H2, H2-H2O, H2-He, H2-Ar, and CO-CO2 gas mixtures. Indications are that in the hydrogen-containing gases the reaction rate is mainly determined by convective diffz~sion. In the CO-CO2 mixtures the rate approches a constant value which is independent of sphere diometer (and gas velocity). This suggests that a surface reaction becomes rate controlling in CO-CO2 gas. WHENEVER silica is exposed at elevated temperature to a strongly reducing atmosphere there occurs, inevitably, a volatilization of silica by the formation of gaseous silicon monoxide. Because of the importance of this reaction in metallurgical processes and laboratory experiments, its kinetics were studied in the present work. EXPERIMENTAL The spheres made of vitreous silica were suspended by a molybdenum wire in the hot zone of a gas-tight high-purity recrystallized alumina furnace tube of 3.1 cm internal diam and 100 cm length. Molybdenum wire resistance heating was used. A purified H2, H2-H2O, H2-He, H2-Ar, or CO-CO2 gas mixture was passed through the tube, from top to bottom. The amount of silicon monoxide formed was determined by weighing the silica balls before and after the experiments. The silica balls were made from commercially available translucent fused quartz glass. In experiments with hydrogen-containing gases the silica stayed vitreous during the experiment, whereas in experiments with CO-CO2 mixtures, which were performed over longer times, the balls were always crystallized to cristobalite (as determined by X-ray examination) at the surface. Constant-pressure head capillary flowmeters were used to control the composition of H2-He, H2-Ar, and CO-CO2 gas mixtures. The oxygen and water vapor were removed from the gases containing hydrogen by passage through columns containing platinized asbestos (450°C) and an-hydrone. The CO was purified from O2 and CO2 by passing it through activated copper (200°C) and through ascarite. The H2-H2O gas mixtures were prepared by passing purified Hz through a mixture of oxalic acid dihydrate and anhydrous oxalic acid held at a constant temperature in a water bath. The equilibrium water vapor pressures of oxalic acid dihydrate-anhydrous oxalic acid mixtures were taken from the available data.1 The furnace system was checked for sufficient gas tightness at regular intervals. As an additional check, the H2O and CO2 contents were measured in the exit gas and were found to agree closely with the metered contents when H2-H2O or CO-CO2 mixtures were used. When purified H2 was used, H2O pressures found in the exit gas were always below 0.1 mm Hg. These contents originate either from an incomplete reoxidation of silicon monoxide in the cold part of the furnace or from small leaks which cannot be avoided. If the latter is the case, such small uncontrolled H2O contents decrease the partial pressure of SiO in equilibrium with SiO2 by less than about 7 pct, which is considered small in comparison to other experimental uncertainties. The temperature of the furnace was controlled automatically in the usual manner. The reported temperature was measured with a Pl-Pt/l0pct Rh thermocouple and was accurate to ± 5°C. TRANSPORT-CONTROL LED RATE OF REDUCTION The gaseous convective diffusion from the surface of a sphere can be predicted from the semiempirical relationship obtained by Ranz and Marshall2 in a study of the evaporation of water and benzene drops in air. The following equation applies:

Jan 1, 1967

By K. Leutwyler

The key to realistic casing stress analysis in thermal recovery installations is accurate knowledge of the temperatures involved. Much information leading to prediction of heat losses from tubing string to wellbore has been published. This .same information permits calculation of casing temperature under certain conditions. This paper reviews the basis for most of the published work, the line source solution to the diflusivity equation and examines some of its application criteria in the light of unsteady-state conditions. An alternate finite differences method is presented for correlation, and experimental data pertaining to the heat transfer mechanism in the annulus is reported. In the light of the information presented in the paper, it appears that the assumption of a line source for the tubing-wellbore system is valid even for reasonably short periods of heat flow, .such as "steam soak" operations. The empirically modified equation presented earlier matches computer derived results of the finite differences approach lor periods as short as 50 hours. Previously predicted cas ing temperatures and the associated stress levels are thus verified based on these correlative computations and the reported experimental information. Fall out from the computer output resulting in some studies of the influence of insulating cement properties on casing temperature, and experimentally studied effects of nitrogen placed in the annulus at various pressures., are described. INTRODUCTION Generally speaking, predictions of average casing temperatures have been based on an idealized model involving a centralized tubing string at uniform constant temperature radiating energy towards the casing under steady-state conditions.'. ' Heat is transferred away from the casing to the formation by unsteady-state conduction across thermal barriers consisting of cement and the mud sheath. Heat balances written on the casing permit prediction of instantaneous casing temperatures. During early periods of injection, casing temperature varies appreciably as a function of time; however, for long injection times this variation decays as heat flow from the wellbore to the formation stabilizes, and it becomes virtually negligible as conditions surrounding the wellbore become quasi-steady-state (Fig. 1). Carslaw and Jaeger -escribe a solution to the diffusivity equation for a line source (Fig. 2). A constant supply of heat is introduced into the medium from the line placed into the origin of the x-y coordinate system and parallel to the z-axis. Heat spreads outward in the infinite cylindrical solid and the temperature increase on a line going through point (x, y) parallel to the z-axis at a distance r can be predicted by: The following criterion must be considered: the Carslaw and Jaeger equation is obtained on the basis of a large

Jan 1, 1967

By D. W. Peaceman

In previous papers by this author on numerical calculations of multidimensional miscible displacement, some simplifying assumptions were made in writing the dispersion term of the differential equation. It was assumed that the flow vectors were essentially parallel to the x-axis. However, in most multidimensional miscible displacements, the flow vectors are not so simply oriented with respect to the coordinate axes. This paper derives the correct dispersion term /or the more general case; gives a difference approximation /or the dispersion tern; and derives the stability criterion for the corresponding explicit difference eguation.

Jan 1, 1967

By R. J. Brown

Design of a submarine pipeline system is governed by many factors, one of which is the effect of transverse horizontal currents on the pipeline structure itself Although this feature alone can be of utmost importance to the life span and stability of the underwater pipeline, it is usually the least recognized, with these hydro-dynamic forces being either completely overlooked or grossly underestimated. In an attempt to supply the methods and means for stabilization of sub-marine pipelines, much capital has been spent by pipeline companies. Design criteria incorporated by these organizations extend over a broad spectrum of engineering with bases varying from pure theory to cut and dry methods. A history of numerous failures has substantiated the postulate that many of these submarine pipelines have been underdesigned from the stability standpoint. By the same token, equally as many have been excessively overdesigned with resultant excess expenditures. An attempt has been made in this paper to establish the magnitudes of these hydrodynamic forces by measuring the lateral current-induced differential pressure distribution around the periphery of the submarine pipeline. INTRODUCTION The expenditure of money for research by corporations obviously requires beneficial results to justify the investment. One of the functions of the engineer is first to make the problem known and then to provide a means of solution at the least possible cost. The initial procedure is to investigate all pertinent published technical articles on the subject to determine if this problem has been previously solved. If resolution of this problem is not obtainable, the next course of action is to evaluate through experiments these forces and their magnitudes. On the specific subject of drag and lift, the accumulation of data is provided from three separate and distinct sources: from experience, from theoreticians and through experimental observations. The experience source has proved costly with money wasted either because of underdesign and failure or excessive expenditures for overdesign, and has not been sufficient to form a definite conclusion on these hydrodynamic forces. The pure theoretician is often not completely cognizant of the situation and his works bear little adaptation to the real problem. The experimentalist has his own problems with regard to obtaining results from small-scaled models and successfully adapting them to the larger prototypes. Each method has its own advocates who diligently adhere to their own separate routes as being the only means of resolving this problem. However, a review of the various methods of analysis leads one to the conclusion that if the advantages of all three are combined, a reasonable solution to this problem can be reached. Briefly then, the problem is to obtain the magnitudes of the forces on the submarine pipeline subjected to a lateral current by investigations of previous design considerations, by a study of existing theoretica1 data and by appropriate experimentation. This paper deals mainly with obtaining experimental data and the interweaving of theoretical and experience factors for resolution of these forces. APPARATUS AND PROCEDURE These tests were performed in 1961 at Collins Construction Co.'s main office near Port Lavaca, Tex. Test equipment consisted of a steel tank 3 ft high by 4 ft wide by 40 ft long, with several pumps circulating the water through the tank via suction lines from the downstream portion of the tank and discharging from the pumps directly into the upstream end of the tank. For reduction of turbulence, a wire section was constmcted in the upstream portion and straightening vanes in the form of pipes, screens, etc., were used to reduce the vortex and eddy action in the tank (Fig. 1). A venturi was built at the test specimen for obtaining velocity sufficient to simulate actual conditions. This full-scale test was initiated to eliminate the error introduced in modeling and obtain a truer picture of forces actually encountered. Six- and 10-in. specimens were prepared by

Jan 1, 1967

By George A. Roberts

The development of an alloy system, high-speed steel, is used as an example of the progress of physical metallurgy. Tracing the history of men and their thoughts as they studied and invented and modified these materials, the story of high-speed steel since the 1890's is seen to be a long, slow, but steady progression of achievements. Today's multiplicity of high-speed steels are the result. Data are presented on quantitative metallography of carbides in high-speed steel with high vanadium content using the QTM instrument. Grindability as well as mechanical properties are related to the content of excess MC and M6C carbides in a series of high-speed steels and the newer matrix steels. These emphasize the present state of knowledge and show the current stages of thought that may affect the development in succeeding years. It is a great thrill for me, an ex-research metallurgist and ex-practicing metallurgical engineer, to speak to an audience currently engaged in both pursuits on the occasion of the 43rd Howe Memorial Lecture before the Iron and Steel Division of the AIME. I shall not review Henry Marion Howe's distinguished career, so adequately reviewed in Maxwell Gensamer's contribution in the 1965 AIME History of Metallurgy, except to point out that, while leading the life of an intellectual, he worked as a practicing research metallurgist and a practicing metallurgical engineer for significant periods of time and contributed greatly in both areas. For those of you who believe, as I do, that in the intellectual activity of metallurgy we are blessed with an intimate mixture of science and engineering (see Mehl's 1960 Howe Lecture), we might recall that Howe, while pursuing a life of writing and teaching in later years, served as the first Chairman of the Division of Engineering of the National Academy of Sciences. That metallurgists were called upon then to lead an engineering recognition movement is perhaps related to the strong part played by the AIME and its members in the recent establishment of a National Academy of Engineering and to the fact that a metallurgist is its head, 47 years after Howe served. I stress these engineering aspects for I wish to talk about a metallurgical engineering subject today. Alloy systems, physical metallurgy, and steel blended with an historical perspective and a report on new techniques to improve the old will, I trust, be of sufficient interest to bring these lectures back to the physical-metallurgy banner from which they have departed for 4 years. Those process metallurgists, open-hearth fiends, and physical chemists among you will just have to be patient. The development of an alloy system is the engineering application of scientific thought and fact. There are many interesting alloy systems, developed to a high degree of technology and sophistication over the past 40 to 70 years. These systems are bound together and related by a similar set of characteristics. For example, stainless steels are a system in which the members are related by a high chromium content which gives a corrosion-resistant characteristic. A subfamily in this system would be austenitic steels

Jan 1, 1967

By C. M. Wayman, M. Miyagi

COMPARED to many of the binary alloys of iron, Fe-Ir alloys have received very little tud.'' Virtually nothing is known about the Fe-Ir equilibrium diagram' except for a few details concerning the liquidus-solidus positions for dilute iron-based alloys. It is reported1 that the y to a transformation temperature in Fe-Ir alloys (up to 40 wt pct Ir) is composition-dependent, varying from -900°C for pure iron to 0" to 100°C for Fe-40 pct Ir in an almost linear manner. The y to a transformation appears to be reversible,' the transformation hysteresis width increasing with increasing iridium content. In a qualitative way, the composition dependence of the y to a (and vice versa) transformations appears similar to those in the widely studied Fe-Ni system. In that the y phase (in Fe-40 pct Ir) can transform at -1000C, it

Jan 1, 1967

By F. A. Crossley

The titanium-rich end of the Ti-A1 system has been investigated up to 35 at. pct A1 (23 wt pet). One conzpound Ti3Al was found to occur between primary a and TiAl. It is ordered hcp with DO19 structure, it has virtually no solid-solubility range, and it has a closed maximum at about 875°C. OIL either side of the compound are a +Ti3Al two-phase fields. The limiting a1uminum solubility in primary a at the titanium-rich end is indicated to be 7.5 at. pct A1 (4.4 wt pet) at 550°C and about 6.8 at. pct Al fl wt pct) at 500°C. Quenching alloys from above the a + Ti3Al two-phase field produces the following structures with respect to alloy composition: Up to 13 at. pct A1 (7.8 wt pet), a solid solution; from 15 to 18 at. pct A1 (9 to 11 wt pct), shear transformation product or martensite; from 19 to approximately 30 at. pct (11 to 19 wt pet), submicro-scopic coherent Ti3Al in an a malvix. The twin hcp phase fields reported in the literature are the result of nonequilibrium corzdztions. Ti-A1 alloys, once partitioned by dwelling- in the a + ß phase field during either hot working or heat treatment, are extremely difjicult to homogenize at temperatures below 1000°C. Such partitioned alloys exhibit the characteristics or symptoms of two-phase materials, and may be said to suffer the "twin-phase syndrome". THE earliest investigations of the Ti-A1 system by Ogden et al.1 and Bumps et al.2 reported wide solubility of the primary solid solutions. Aluminum was reported soluble in the low-temperature allomorph to the extent of 37 at. pct (25 wt pct), and the first intermediate phase was reportedly TiA1. Somewhat later Kornilov et al.3 reported a similar diagram with phase boundaries displaced towards lower aluminum contents and higher temperatures. Beginning about this time (1956) reports in the literature made it very clear that one or more intermediate phases occurred at lower aluminum contents than TiAl.4-17 These reports included five major investigations of the titanium-rich end of the Ti-A1 diagram.4,12,14,16,17 Three of these diagrams show two two-phase fields below 37 at. pct Al, while two of them show a single two-phase field. The existence of the phase Ti3A1 is firmly established and is included in each of the diagrams, except one—that of Sato and Huang.12 The new phases are reportedly hcp and differ from primary a only slightly when disordered, and when ordered the "a" parameter is approximately one,4,12,15 two, 6-10,13,14 or four14 times that for primary a. Beyond this, however, the diagrams are remarkable for their lack of agreement. Two tacit assumptions are usually made in phase-diagram determinations of metal systems. These are: 1) equilibrium anneals bring the alloy to equilibrium or to indistinguishable closeness to it, and 2) equilibrium conditions established at elevated temperatures are either "frozen" by rapid quenching for evaluation at room temperature, or quench-transformation products are recognized as such. In the current investigation evidence was obtained that over substantial composition ranges neither of these two conditions was met in any of the more recent major investigations. I) MATERIALS, METHODS, AND TECHNIQUES The alloys of this investigation were prepared by nonc on sum able electrode arc melting. Materials used in the preparation of the alloys are summarized in Table I. The investigative tools employed were: optical and electron microscopy, differential thermal analysis (DTA), disatometry, X-ray diffraction, electron diffraction, and resistometry. Alloys for microscopic and X-ray investigations were prepared as 15-g melts. Alloys containing from 7 through 11 at. pct A1 were hot-rolled out of a furnace at 900°C, from 12 through 15 at. pct out of a furnace at 1000°C, and from 16 through 18 at. pct out of a furnace at 1125°C. Alloys containing more than 18 at. pct A1 could not be hot-rolled. The ingots were covered with Markal coating prior to hot rolling to minimize atmospheric contamination. After hot rolling, alloys containing up to 15 at. pct A1 were ground and pickled to remove 7 mils from each surface; alloys containing 16 and 18 at. pct A1 were skinned to a

Jan 1, 1967

By R. F. Hehemann, Darrell W. Smith

The kinetics of alloying- in W-5 rot pct Re powder compacts were investigated at temperature about the upper limit for the formation of the x phase, such that a single intermediate phase was involved in the process. The rates and directions of motion of the a-o and o-b interface were assessed as functions of tinze at the sintering teri7peraticve by quantitative izetallog-ralhic techtziques. The iiotion 01 these interjuces could be specified approxitnate1.y by u Parabolic )-ate law and deviations from this ,rate law were attribuled to a complex interaction between surface dif'fusion and volume diffusion during the initial stages of sintering. WHEN binary or more complex alloys are produced by sintering compacted blends of elemental powders, a complication is encountered which does not arise in the sintering of pure metals. In addition to concern for grain size and densification, the rate of inter diffusion of the component elements must be considered. The general problem of alloying during sintering has been treated by numerous investigators.'-7 These investigations have been concerned only with systems that form continuous series of solid solutions and a concentric-sphere diffusion model4 provides a useful description of homogenization in these stems.' Some controversy remains, however, over the contribution of surface and grain boundary diffusion to the homogenization process. Heckel% ePorts that homogenization can be described by volume diffusion at all temperatures, whereas other invetiators'' recognize the importance of surface and grain boundary diffusion, particularly at lower temperatures. Analysis of the homogenization process in partially miscible systems is complicated by the motion of the interfaces separating the individual phases. A simple analytical solution does not exist for such finite systems with spherical symmetry. It was the purpose of this work to examine the kinetics of the formation and dissolution of an intermediate phase in a binary system through an assessment of the motion of the interfaces separating this phase from the terminal solid solutions. PROCEDURE The W-Re system is an example of a partially miscible binary system and was selected for study because of its current importance in refractory-metal technology. The equilibrium phase diagram, Fig. 1, shows that the o phase will be formed by interdiffu-sion between tungsten and rhenium. At temperatures between approximately 2125" and 282O°C, the o phase will appear as the only intermediate phase in the diffusion system. In this work, a mixture of elemental powders containing 5 pct Re was employed. In such a mixture, the rhenium particles will be distributed randomly in a tungsten matrix. During sintering, a o-phase layer will form at the W-Re interface. The development and subsequent disappearance of the o phase are controlled by the relative rates and directions of motion of the 0-0 and a-p interfaces. The motion of these interfaces can be inferred from quantitative metallographic measurements of the average radii of the o shells surrounding rhenium particles. Elemental tungsten and rhenium powders of commercial purity were employed in these investigations. The rhenium powder was spheroidized from +200 mesh agglomerates by plasma-arc spraying and was then classified to yield a powder composed of essentially

Jan 1, 1967

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

Use of the Weisseizbevg goniometer for the study of the morphology of whisker crystals and transfer of the results to a stereograplzic projection are described. While in almost all instances a complete and rapid solution is obtained with a purely gaphical treatment, an absolute analytical treatment applicable to difficult cases is given. Examples are given for determination of growth direction and external features of a crystal, for detection of 'twintzittg, and for establislting the orientation relntionships in a kinked whisker crystal. We give here a method for the use of the Weissen-berg goniometerl" in the study of orientation relationships in "whisker" (micro) crystals3-8 and direct transference of the data to the stereographic projection. The operations are straightforward and require no involved calculations: in almost all cases solely graphical treatment is adequate. In a few complex instances, a more exact analytic method may be required. In the method, not only are equatorial reflections of an oscillation photograph used, but also those on higher layer lines, or those not occurring on a recognized layer line. Such reflections off the zero level are often essentialy to detect the presence of twinning. The method is particularly suitable for examination of whisker crystals since the axial or growth direction is almost always one of high symmetry, permitting a more restricted oscillation range to obtain the necessary reflections and an accordingly shorter exposure time. The method is also applicable when there is no lattice-related external symmetry, and is as well able to identify external crystal surfaces, when such exist, with the aid of an optical beam. Thus growth direction, identification of lateral faces, twinning, and orientation relationships in a kinked crystal can be established. A test for the Eshelby twist6-8 can readily be made. The method can be used for orientation, for structure analysis, and as an aid in the identification of an unknown crystal. Moving-film techniques have previously been employed variously for crystal-orientation studies.1,2,9-13 In particular, use of the Weissenberg method for orientation determination has been reported by Wooster and wooster.14 While their method derives the same data as here described, it differs in depending on the tracing of constant p streaks, a less than dependable phenomenon with small specimens, a few microns or less in cross section. The Laue method can and has been used for orientation studies of microcrystals,15 e.g., 2 p in cross section, but requires the extreme intensity and concentration from a microfocus tube to obtain reasonable exposure time. PRINCIPLES OF THE PRESENT METHOD Consider a crystal of random orientation with respect to the oscillation axis in a Weissenberg goniometer. Its diffraction pattern need not necessarily result in recognizable layer lines. In Fig. 1, a stereo-graphic projection of the reference sphere is given with the equatorial plane XZ normal to the oscillation axis Y, and the X-ray beam directed from X towards 2. The great circle ZZ is then normal to the beam, and the axes are right-handed. Thus the reflected X-ray beam lies at the intersection R of small circle ARB of radius 28 from 2, and of small circle RE of radius a from the rotation axis Y. a is the conventional semiapex angle of a cone of diffracted rays from the atoms aligned in the direction

Jan 1, 1967

By P. J. Burkhardt, L. V. Gregor

The oxidation kinetics of single-crystal silicon has been investigated using extremely dry oxygen as the oxidant. Two techniques were used. The first involved a flow system with which incremental thickness measurements were made. This sytem provided an oxygen ambient at atmospheric pressure which had a moisture content of less than 0.1 ppm of water. The second technique involved measuring the change in volume of oxygen with time in a closed system. A temperature range of 900O to 1200 "C and a pressure range of 20 to 760 Tory were covered. Both techniques showed that a simple parabolic rate law holds only at the higher temperatures. At lower temperatures, the data up to 5000Å can be fitted well to a linear-parabolic equation. The two constants in this rate equation are examined in terms of a physical model. The parabolic portion of the low-temperature oxidatLon fits along the high-temperature curve of an Arrhenius plot. The parabolic rate constant for the flow-system technique can be expressed by k = 2 x 10-9 exp (L 31 kcalRT) sq cm per sec. The manostatic technique gave k = 1 x 10-exp (- 23 kcalRT) sq cm per sec. By extending the thickness range to over 104. a rate law of X = At seems to fit all of the data better than the linear-Pavabolic law. Here the exponent n increases from 0.518 at 1150 V to 0.637 at 900°C. PASSIVATION of silicon planar devices through thermal oxidation is of such great importance that considerable effort has been devoted to the study of the process.17 Much of this work involves hydrothermal oxidation, in which H2O is the oxidant: whereby H2 is produced. It has been proposed that the reaction proceeds by the formation of Si-OH groups and hydroxyl diffusion through the film to react with the silicon surface. Thus, the silicon oxide layer probably contains residual hydrogen in some form after oxidation in H2O and it has already been established that hydrogen has a marked effect on the electronic properties of oxide-passivated silicon surfaces. Silicon is oxidized by oxygen at elevated temperatures; but the rate of oxide formation is considerably lower than that for 0." By using dry O2 as the sole oxidant, the presence of residual hydrogen is presumably avoided. At present, the mechanism of thermal oxidation by dry O2 is a matter for conjecture, and the presence of small amounts of H2O was reported to have a marked effect on the rate of oxide formation10 (perhaps on the electronic properties of the surface, as well). Hence, this study of the kinetics of silicon oxide growth in 0, was undertaken: 1) to determine the oxidation rate at atmospheric pressure in a system designed to exclude rigorously all traces of H2O in the ambient 0,; 2) to obtain oxidation-rate data in situ in dry O2 by means of a continuous monitoring of the amount of 0, consumed during oxide growth. These objectives are complementary—the first assesses what effects (if any) are obtained by excluding H2O from the reaction, and the second expands this information while at the same time avoiding the uncertainty inherent in relying upon incremental measurements of oxide growth to furnish kinetic data. EXPERIMENTAL PROCEDURE The conventional silica open tube in a resistance-heated oxidation furnace is subject to sources of trace H2O, even after the input O2 is thoroughly dried and the outlet gas stream is passed through a trap to prevent

Jan 1, 1967

By M. Nicholas, D. M. Poole

The diameters of sessile drops have been found to increase linearly with time in five reacting binary metal systems. The spreading rates of the drops are markedly dependent on temperature and on prior alloying of the solid with the lower melting point metal, hut are independent of the drop volume, wetting atruosphere , solid-surface roughness, and prior alloying of the drop with the substrate metal. A mechanism has been suggested that relates the linear-spreading rate to lateral diffusion of the liquid-metal atoms into the solid at the drop edge. An Arrhenius- type equation has been derived that describes the temperature dependence 0) the spreading rate, and although the agreement between the actual and the predicted pre-exponen-tial terms is poor that between the activation energies is excellent and the variation in the spreading rate of copper on Ni-Cu alloys produced by different extents of alloying can be predicted with considerable accuracy. CHEMICAL interactions frequently change the wetting behavior of solid-liquid systems causing, for example, "secondary spreading1 of sessile drops beyond the size defined by the surface and interfacial tensions of the unreacted components. The kinetics of the contact-angle decreases associated with this spreading are similar for many systems, but few studies have been made with the objective of determining whether the similarities are a reflection of a common mechanism. Some workers2,3 have assumed the secondary spreading is controlled by changes in the liquid surface and liquid-solid interfacial tensions and hence by the composition of the liquid, and contact-angle changes measured by the vertical-plate technique have been used to follow the course of liquid-solid chemical reactions.4 Other processes that have been invoked to explain these time-dependent changes in specific systems include the removal of adsorbed gas from the liquid-solid interface.5 penetration of containment layers on the solid Surface,6 interdiffusion,1,7 reori-entation of the solid surface into a wettable configuration: vapor-phase transport of the liquid onto the solid in advance of the drop,9 and, from vertical-plate studies. capillary flow between oxide layers and the solid surface.10 One of the reasons for the profuseness of these suggestions may be the complexity of the contact-angle change kinetics. However, in an analysis of secondary spreading gold and copper on UC,11 it was found that the diameter of the contact area between the sessile drop and the solid surface showed a simple linear increase with time although contact-angle changes were more complex. To check whether the linearity was merely fortuitous! additional exploratory work was conducted with four reacting metal-metal systems: Au on Ni. Cu on Ni, Cu on Fe, and Ag on Au. Linear spreading was observed in every case even though the kinetics of the contact-angle changes were complex. A further detailed study of the kinetics of linear spreading of five reacting metal-metal systems has been made with the object of determining the mechanism involved. The influence of variables such as temperature, drop volume. and the initial composition of the drop on the linear-spreading rate has been measured and compared with those predicted by a number of possible mechanisms. The systems employed in this study (Cu and Au on Ni and Pt, and Ag on Au) were selected because of the availability of potentially relevant chemical and physical property data. the simplicity of their phase diagrams at the wetting temperatures, and the ease of experimentation. EXPERIMENTAL TECHNIQUES The purities of the metals used in the study were: copper, 99.9 pct; gold. 99.96 pct; nickel, 99.2 pct; platinum 99.99 pct; and silver, 99.999 pct. The wetting tests were performed in a split tantalum tube vacuum resistance furnace of a conventional design. The furnace element was held vertically and was 1 $ in. in diam and 6 in, long. Viewing ports were provided in the water-cooled chamber to enable the specimens to be observed in both the horizontal and vertical planes. The temperature in the hot zone of the furnace could be held at 1500" i 5°C for an indefinite time. The surfaces of the solid-plaque metals were ground flat on Microcut paper and both the sessile drop and substrate metals were ultrasonically cleaned in methyl alcohol prior to their insertion in the furnace. After loading, the furnace was pumped down to a pressure of 2 x 10-5 mm of mercury and degassed for 30 min at 900° to 950°C. The temperature was then increased at more than 100°C per min to that used in the wetting test. The vacuum at the wetting temperature was better than 5 x 10-5 mm of mercury. Dewetting and retraction of the drop on cooling did not occur and the contact-area diameters, therefore, were measured after solidification with a vernier traveling microscope. The diameters quoted later are arithmetic means of ten measurements. The standard error of the mean never exceeded 3 pct and was often less than 1 pct.

Jan 1, 1967

By C. Dirksen

This study investigates whether oxygen consumption during segregated forward combustion may be affected by natural convection. Linear theory indicates that thermal instability occurs in a horizontal porous medium when the modified Rayleigh number N 'Ra exceeds a critical value of about 40. Most experimental results, including those described here, indicate this value to be about an order of magnitude smaller. N ', is derived from "differential similarity" and the proper time scale factor is also obtained. The critical value of N'ra for a horizontal, water-saturated porous medium at atmospheric pressure subjected to a uertical temperature gradient was found to be about 3.0. The ambiguity in the value of N'R, arising when fluid properties cannot be assumed constant is indicated. Natural convection was observed for oblique systems below N 'Ra , while simultaneous horizontal flow did not affect N Racrit. The expected range of values for the various parameters pertaining to segregated forward combustion is indicated and the corresponding values of N 'Ra calculated. Only under very favorable conditions, such as high permeability, high pressure and low top temperature, can the critical value of N 'R, be exceeded. Thickness of the convection layer should not become too large, since the time required to obtain significant mixing is proportional to the square of the thickness. For an oblique combustion frant the conditions for significant mixing may be less strenuous. LNTRODUCTION Laboratory1,2 and field314 experiments have shown that a markedly upgraded oil can be produced by in situ forward combustion. Under certain conditions injected air may move through a high permeability, gas-saturated channel over the oil-saturated layer, while combustion takes place, exclusively, at the horizontal interface (Fig. 1). Such a channel may be a natural gas cap, or it may result from a forward combustion process in which injected air and combustion gases fingered through the top of the formation, causing a premature breakthrough at the production well. In this particular case of segregated forward combustion, the air stream moves parallel to the combustion front. Molecular diffusion and transverse mechanical dispersion will be insufficient to cause all the oxygen in the air stream to reach the combustion front. When the oxygen in the air stream is not completely consumed in the combustion process, it will reach the production well and create a serious fire hazard. In addition, the reduced efficiency may make the process economically unprofitable while also serious corrosion problems may be experienced in the production we11.3 This undesirable situation should, therefore, be prevented. However, in a situation as depicted in Fig. 1, an unstable temperature gradient exists in the gas phase. The combustion is taking place at the lower boundary at a nearly constant temperature that may be anywhere from 600 to 1,200F, depending on the conditions. The upper boundary of the gas phase is overlying, impermeable rock; its temperature might be around 100F, but may increase to 300F or higher due to conduction and convection as the combustion process continues. This represents a steep temperature gradient and it is conceivable that under those conditions free or natural convection will take place in the gas

Jan 1, 1967

By H. M. Otte, A. G. Crocker

SLIP, twinning, stacking faults, and precipitates on well-defined planes in a crystal produce traces that are visible on either a polished or an etched surface. The purpose of this note is to establish the usefulness and limitations of the information obtained from such traces when determining the orientation of single crystals or grains in a polycrystal. Attention will be confined specifically to traces produced by octahedral planes in cubic crystals, though most of the remarks can be taken as more generally valid. The problem of determining the crystal orientation is completely solved in principle' and in particular for the special case under consideration.1'2 Tables are available3 in which the orientations can simply be looked up when given the two or three angles between the traces. Thus a quick and simple method is available for obtaining the orientation of a crystal from surface traces. In this note two examples will be given from recently published papers and will show the use of the method with the aid of the Tables, as well as some of the precautions that should be taken in the procedure. As part of an analysis of X-ray Kossel lines, Peters and Ogilvie 4 determined from them the crystallographic orientation of the specimen. They chose a Cu-5.10 wt pct Si alloy that had been partly transformed from the a (fee) phase to the k (hcp) phase. The latter formed as very straight platelets along the octahedral planes of the a phase, and a photograph (Fig. 7 of Ref. 4) was reproduced in the paper4 of the region for which the orientation had been determined from the Kossel lines. Apart from a very small area (about 0.15 mm diam) near the top center of the photograph, the region shown was clearly all of one orientation. Traces in four different directions were readily identified. Starting with the smallest angle and proceeding counterclockwise, the angles between the traces were measured to be 27° 43', 55° 16', 64 3 43 1, and32"26'. These add up to 1802 08 1, indicating a cumulative error of 8'. The correct values should, of course, be taken from the original glass plates, or at least from prints on distortion-free paper, so the precision of the quoted values is actually greater than their accuracy. The solutions can be looked up in the Tables3 in four ways, depending on how the angles are paired, tion for the minutes of arc is required and the solution obtained will be identical for all the four pairs if the values of the angles are not in error. In a practical situation, such as this one, there are several sources of error, so that the four solutions obtained will all be slightly different. An arithmetic average of the four solutions is mathematically incorrect, but may, in special cases, give an answer close to the correct one. A rigorous procedure for refining the orientation has been described by Mackenzie;5 he gives a least-squares method that can be applied, if necessary, to this type of problem. The orientation thus determined can be compared with the one from the Kosel lines and discrepancies larger than the experimental error assigned to nonor thogonality of the X-ray or electron beam to the specimen. Here the example will only be used to illustrate the general procedure for making a very rapid rough check. From measurements on the stereographic projection, Fig. 9 of the paper,4 we can determine that the surface normal is approximately (0.434, 0.704, 0.562). From the Tables3 we find, using 28°, 55" as a good approximation to the first pair (27°43', 55°16'), that the surface normal is (0.453, 0.678, 0.579).* These two values are within 2 deg of each

Jan 1, 1967

By I. C. Edwards, G. E. Agar

INTRODUCTION Many research efforts over the past years have been devoted to the study of energy consumption in comminution and the characteristics of the broken material. The old "laws" have been superseded by more general methods of analysis1,2,3 but these methods have not received much attention outside the laboratory. The work to be described here was done in order to obtain scale-up information on wet rod milling of sylvinite. EXPERIMENTAL A 30 x 51 in. Denver rod mill with a scoop feeder was used for these tests. Ore was fed from a variable speed screw conveyor into the feed box along with saturated brine. The mill was filled to approximately 45% of its volume with 2400 lbs of a mixed rod charge. The minimum possible feed rate was 2.5 tph and the practical maximum for this mill was 8 tph. Individual tests were run by setting the solid feed rate, then adjusting the brine flow rate to give the desired pulp density. After steady state, as evidenced by constant discharge pulp density, had been reached, samples of the discharge were taken at 15-minute intervals. These samples were filtered, washed with alcohol, dried, then screened on Tyler screens using a Ro-tap shaker. The sylvinite feed was regular plant feed that had been scalped at 20 mesh and was tapped from the coarse ore bins. It had been previously passed through a roll crusher set at 3 inches and after scalping was essentially minus 1 in. plus 20 mesh. RESULTS The screen analysis of samples from each series of tests was plotted on double logarithmic scale paper to determine the reliability of the series. In some cases, a faulty sample or screening procedure dictated that the results of one test be discarded. The criterion is a simple one — if the slope of one plot was noticeably different from the others in the series, it was discarded. The only cases that caused difficulty were those in which the brine was not thoroughly displaced from the solids. When this occurred agglomerates were easily identified on the screens and the whole of the sample was retained on 3 or 4 screens. There were no border line cases. The number N given in Table 1 is the number of size analyses that were pooled to form the data shown in Fig. 1. While examining the reliability of the sizing and sampling by plotting the data, it became apparent that there was no significant difference between grinds made at 50 and 60% solids. For the analysis given here, these tests were regarded

Jan 1, 1967

Jan 1, 1967

By Donald L. Johnson

REVIEWS on the applications of optical ellipsometry to the study of polycrystalline metal surfaces and oxidation-corrosion processes have been presented by Kruger and Hayfield, and white.' In view of the ability of this method to measure differences in film thicknesses to below 10A, ellipsometry was used to study nitrogen chromium surface reactions during the early stages of growth. Chromium specimens in the form of a vapor-deposited film on glass substrate were exposed to purified nitrogen for selected time increments at 700°F. This temperature was selected because it gave the best combination of exposure time and thickness change. Ellipsometer readings were taken immediately after exposure to determine polarizer and analyzer angles using monochromatic light at 5000A.2 After each set of readings was obtained, the specimens were exposed for an additional increment of time in nitrogen and a new set of readings taken. Film thickness was computed from LUCy's3 equation for nonabsorbing thin films. In Lucy's derivation, two equations for computing thickness were obtained. Peterson and Bashara4 have discussed the validity for using one of these equations when an independent value of the refractive index is used. The film was assumed to be Cr2N (Ref. 5) and was determined to be optically isotropic. Optical isotropy was determined by rotating the films about the surface normal and noting that the azimuth and phase difference did not change. Since Cr2N is reported to be hexagonal, it is assumed that optical isotropy results from randomly oriented grains of very small size. The thickness was computed using refractive indexes varying from 1.6 to 2.0. The maximum variation in computed thickness over this range for a given exposure was 20 pct. The data reported are based on a refractive index of 1.8. Reference readings on the metal surface were taken just prior to nitrogen exposure. No attempt was made to deoxidize or denitride the metal surface after vapor deposition, although specimens were observed to gain approximately 2A per day in air at room temperature. Twelve experimental exposures were made on a single sample over a total accumulative time of 132 min. Additional experiments over longer time intervals were made to determine the effect of heat up and cool down during and after exposure. As a result of these experiments, the thickness computed for each exposure was increased by 4A. Time profiles based on the linear law are similar to the oxidation profiles for many metals.' The rate of accumulation of the nitride film is rapid at first and slows down as the film thickness increases. The parabolic law x2 =k1t +k2 [11 gives a straight line in Fig. 1; therefore, it is concluded that the kinetic behavior is parabolic. Sey-bolt and Haman5 reported similar results for CrlN on chromium at 1000°C although the order of the thickness was much greater (10-3 cm). Analysis of thickness data based on other values of refractive index shows that the curves are shifted vertically but retain the same shape. According to Mott,7 it is expected that for thin films k2 is zero in Eq. [I]. Since the plot in Fig. 1 is linear through the origin, k2 is zero and the parabolic nature of the film growth is further indicated. The parabolic rate constant, calculated from the slope in Fig. 1, is 4.4 x 10-16 sq cm per sec. The corresponding work (+) required to bring an electron from the metal into the oxide is estimated to be 1.5 ev.7 The author thanks professor N. M. Bashara and D. W. Peterson, Electrical Materials Laboratory, University of Nebraska, for making the ellipsometer available and providing technical assistance. The Electrical Materials Laboratory is supported by the office of Naval Research, NONR 3718 (01). This work was partially supported by the Engineering Experiment Station and the Mechanical Engineering Department, University of Nebraska.

Jan 1, 1967

By C. N. J. Wagner, E. N. Aqua

The bee alloys of the terminal solid solution of rhenium in niobium were investigated by X-ray diffraclion methods. The analysis of the broadening of the powder pattern peaks from the niobium-rich alloys, cold-worked by filing, showed evidence for faulting on (211) planes. The alloys with larger rhenium concentrations were brittle, and fractured when filed. The analysis of the shapes of the diffraction profiles from powders produced by hammering with steel plates) of these brittle alloys indicated that the broadening was due predominantly to platelets bounded by clibe planes. This result was consistent with the Laue patterns from single-cleazlage facets which showed evidence for clearage fracture on {100} planes. The influence of faulting on the mechanical properties of face-centered cubic (fcc) metals and alloys, i.e., how faulting affects the various dislocation configurations in fcc metals, is well-known.' The important question is whether or not faulting occurs in body-centered cubic (bcc) metals and alloys, and, furthermore, if faulting does occur, to what extent it affects the mechanical properties. Theoretical calculations indicate that the stacking-fault energy is rather high in most bcc metals.2 The results of sev-eral studies using transmission electron microscopy have indicated that faults are present in well-annealed and lightly deformed bcc metals and alloys, such as tungsten,3 vanadium,4 and Mo-Re alloys.5,6 These faults, as have been observed, may be stabilized by the presence of interstitial (C,N,O) atom segregation to the extended dislocations. In addition, there is considerable X-ray diffraction evidence for faulting in bee metals,7-8 using the same techniques that were successful for the study of faulting in fee metals and alloys.7 This X-ray diffraction method is currently the only technique applicable for the investigation of faulting in metals with high stacking fault energies, i.e., when severe amounts of deformation are required to produce sufficient amounts of faulting to be detectable by X-rays. The thinned films of heavily deformed metals will become opaque in the electron microscope due to the high dislocation densities present. In a previous investigation of faulting8 in bee metals of group Vb and VIb, the authors observed that niobium was ductile when filed at room temperature, and showed the highest density of faults on the (112) family of planes. A niobium-base alloy system was therefore chosen to study the effect of alloying on the occurrence of faults in a bcc alloy. There is a large range of terminal solid solubility of rhenium in the alloys of the bee refractory metals. All of the Group VIb-rhenium alloys, i.e., Cr-Re, Mo-Re, and W-Re, exhibit solid solution hardening without loss of ductility.9 In fact, the ductility reaches a maximum for alloys with compositions near to the limits of solubility. This enhanced ductility has been attributed to the lowering of the stacking-fault energy in the alloy; i.e., the increased amounts of twinning provide an additional mode of deformation, resulting in the observed high ductility.' Because of this demonstrated effect of rhenium in bee alloys, it was decided to study a series of alloys of Nb-Re. Using the same X-ray diffraction methods as described in the previous investigation,' one is able to identify the possible causes of broadening of the powder pattern peaks, e.q., particle size, microstrains, stacking faults, and/or twin faults. The separation of these various contributions to the measured peak shape is accomplished by the Warren-Averbach analysis.7 I) EXPERIMENTAL PROCEDURE A series of seven bee solid solution alloys containing from 2.5 to 21.5 at. pct Re (remainder niobium) were prepared at the General Electric Research Laboratory. Buttons, weighing 25 g, were alloyed from electron-beam-melted niobium chips and random rhenium sheet by arc melting in a water-cooled copper crucible, under an argon atmosphere. The buttons were inverted and completely remelted to promote homogenization. The nominal alloy compositions are listed in Table I. Samples of the cold-worked alloys were obtained by filing to produce a powder for the X-ray study. Alloys too brittle to be filed were hammered with steel plates. The resultant filings or brittle impactings were screened through 150-mesh classifiers and compacted to the required shape for the diffractometer holder. The X-ray diffraction peaks were automatically recorded by fixed time counting at uniform 28 intervals, or by continuous registration with a ratemeter. In all cases, the (110)-(220) and (200)-(400) pairs of peaks and the (211), (310), (222), and (321) reflections were measured on a GE-XRD5 diffractometer using CuKa radiation with nickel filter or MoKct radiation and a LiF monochromator in the diffracted beam. The first part of the data reduction, completed with the aid of a series of programmed computations,10 was the analytical resolution of the Kol component of the diffraction profile. Subsequent computations were then made with reference only to the resolved Kol peak shape. These included the measurement of the

Jan 1, 1967

By A. W. Jenike

A theory of gravity flow of bulk solids like ores, concentrates, coal is outlined and an example of design for unobstructed flow is given. Mass-flow and plug-flow patterns are defined and related to the geometry of the bin. The effect of these patterns on segregation, first in-first out flow, and caking is described. This paper is a report on the results from the AIME sponsored project "Flow of Bulk Solids" which was carried out under the author's direction at the Utah Engineering Experiment Station, University of Utah, during the period 1956-1962. The theory has been applied to the design of numerous industrial storage plants. A QUANTITATIVE METHOD OF DESIGN FOR FLOW There is hardly an industry which does not store and handle solid materials in bulk form. When the volume of the solids is large, gravity is invariably relied upon to cause the solids to flow out of storage, through channels and chemical reactors. Such materials as ore, coal, cement, flour, plastics, clay, soil, to which a general term of bulk solids is applied, flow by gravity or are expected to flow by gravity in thousands of installations and by billions of tons annually. The chemical and process industry depends on gravity flow of solids in process and out of storage. Mining relies on gravity flow in block-caving, in ore passes, as well as in storage and loading bins. Agriculture uses gravity flow in elevators, in feed plants, and in farm silos. In practice, satisfactory gravity flow is seldom attained. Such problems as doming, piping (rat-holing), lack of live capacity, segregation, deterioration in storage, harass industry. In spite of the universality of these problems, until recently, there has been no theory of solids flow, no method of predicting how a solid would behave, and how the channel, bin, or hopper should be designed. In the late nineteen forties and early fifties, the theory of plasticity came of age and simultaneously the computer became widely available. These two factors permitted the development of a theory of flow of bulk solids. The work was conducted during the years 1953-1961. In conjunction with the development of the theory a method to measure the flowability of solids was also developed1-7 and the theory was ap-plied to the design of ore pockets, storage plants,16 chemical reactors and adsorption columns. A comprehensive report of this work is given in a previous work." THEORY OF FLOW The most important question which a theory of flow has to answer is: what conditions need to be satisfied for a given solid to flow uniformly and without obstructions through a channel? First of all, the following observations are made:

Jan 1, 1967

By H. U. Schutt, J. M. Toguri

REGARDLESS of the degree of purification effected during the electrorefining of copper, a danger of re-introducing impurities exists in the operation of melting and casting of the refined copper. Logically only a combination of the purification and casting operation would remedy the situation. A well-known method of metal purification is the "Gross"-type process, whereby a volatile halide is produced, purified, and decomposed.' In the present case, this concept was used to remove the impurities rather than to form possible volatile copper halides. The object was to produce volatile halides of the impurities present and distill these away, leaving high-purity copper. It is interesting to note that such techniques are used in spectrographic procedures in order to shorten exposure time and enhance the density of impurity lines.' In the present investigation, copper hydroxy fluoride was used, since the temperature of decomposition of copper hydroxy fluoride and the melting point of copper nearly coincide. The fluorine released by the thermal decomposition of copper hydroxy fluoride purges the molten copper. Copper hydroxy fluoride was prepared by dissolving high-purity copper in nitric acid and adding hydrofluoric acid and ammonium hydroxide in the required proportions. The melt was made in a high-purity graphite crucible heated by an induction furnace. Trial experiments showed that impurities can be introduced from the crucible material. Therefore, preceding each cast, the crucible itself was cleaned by decomposing copper hydroxy fluoride and discarding the residual copper. Best experimental conditions were obtained by adding 30 g of copper hydroxy fluoride to 500 g of nor-

Jan 1, 1967

By Klaus Schwerdtfeger

Activities of lead in Pt-Pb and Pd-Pb melts in the temperature range of 800° to 1200°C have been detev-mined from electromotive-. force measurements 202th the cells Both systems display strong negative deviations from Raoult's law. The data obtained with cell [1] were combined with available data on the activity of PbO in silica-satuvaterl PbO-SbO2 melts to calculate the standard free cnergy of formation 01- liquid PbO. FREE energies of formation of oxides and activities of components in metal and oxide melts can be determined with galvanic cells consisting of a metal phase, an electrolyte containing the oxide of one of the components of the metal phase, and an oxygen electrode. This technique has been used previously to measure the standard free energy of formation of solid sio2, 1 and activities in Fe-si, CO-si,2 Ni-Si,2 and CaO-P2O5 3 melts. The present paper reports activities in Pt-Pb and Pd-Pb melts as measured with the same experimental technique. EXPERIMENTAL METHOD General Considerations. The following cells were used: Pbl (PbO-SiO2)liquid, SiO2 Pt-Pb 1 (PbO-SiOz)liquid, SiO2solidO2 [2] Pd-Pb (PbO-SiOz)liquid, SiO2solid O2 [3] Liquid alloy and' oxide melt were contained together in a silica glass crucible of essentially the same design as described previously.1'2 The PbO-SiO2 melts are ionic conductors.' During the experiment, the silica glass crystalizes in a layer adjacent to the electrolyte. The electrolyte equilibrates with solid silica and hence attains the composition of the Si02-liquidus curve in the system PbO-SiO2 at any particular temperature. Molybdenum wires were used to establish contact with the liquid metal. The oxygen electrode, which reaches the surface of the PbO-SiO, electrolyte from above, consists of a platinum wire inserted into a ceramic tube and provided with a 2-mm-diam platinum bead at its end. Gas of known oxygen partial pressure (pure oxygen of 1 atm: air, or N2 with 1.04 pct 0,) is blown through the ceramic tube onto the surface of the electrolyte. The cell reaction is Pb (metal) + 1/2O2 (gas) = PbO (electrolyte) [4] No oxidation of platinum or palladium ("displacement reaction") occurs. This is to be expected from the noble character of platinum and palladium, but was also shown experimentally, as described in the next section. The galvanic electromotive force of the cell is given by where AG is the change in free energy for Reaction [4], E(=2) is the number of electrical equivalents necessary to form 1 mole of PbO, and F is the Faraday constant. Formula [5] is valid only if no junction potential exists within the electrolyte. Junction potentials occur only when the compositions of the electrolyte are markedly different at the electrodes. Such differences are negligible in the PbO-SiO, electrolyte, as shown in the next section. Since molybdenum contact leads are used, appropriate corrections were made on the measured electromotive forces to account for the thermoelectric effect of the Mo-Pt couple. The thermoelec-tromotive forces were measured separately with a Mo-Pt couple, and were found to vary linearly from 19 mv at 740°C to 40 mv at 1310°C with molybdenum being positive and platinum negative (cold junction at 25°C). Extrapolation of these values to higher temperatures yields values which are in satisfactory agreement with those obtained previously in the temperature range 1440° to 1630°c.1,2 The change in free energy for Reaction [4] is expressed as where chemical potentials of the species. The value of constant for a given temperature because the slag composition is fixed. Inserting [6] into [5] and using appropriate expressions for upb and uO2 the relation

Jan 1, 1967