Search Documents

By Francis Donaldson

The direct injection of cement grout into water-bearing fissures as a means of checking or stopping the flow of water into shafts and tunnels has been experimented with for a decade or longer and seems to have been first attempted in Europe. The earliest application of which the writer has found a record was at a shaft sunk by the Mining Society of Lens; this is described by C. Dinoire in the Transactions of the Institution of Mining Engineers.1 It is only lately, however, that the process has been entirely successful. This success has been accomplished on the Catskill Aqueduct, now under construction by the city of New York. This aqueduct includes a number of deep pressure tunnels, reached by shafts for both waterway and construction purposes. In portions of these shafts and tunnels considerable underground water was met. The contract for the first of these deep pressure tunnels was let to the T. A. Gillespie Co.; this section is known as the Rondout Siphon. It leads under the Rondout valley at a depth of from 400 to 800 ft. and is about 5 miles long. The engineers expected that considerable water would be encountered, but fortunately this was not the case except at one shaft, known as No. 4, which penetrated the rock at a junction between limestone and conglomerate. Water-bearing fissures were encountered almost immediately. At a depth of about 200 ft. a flow of 1,500 gal. per minute was struck. The. shaft then contained as many pumps as could be used, and it seemed impossible to sink it further. After vain efforts had been made to proceed, John P. Hogan, a division engineer of the Board of Water Supply, suggested that cementation of the fissures be tried. The process was attempted for the first time in this country. Since the shaft was partly full of water it was necessary to drill grout holes with a diamond drill. Platforms were placed on the timbers at the water level, the diamond drill was installed and six 90-ft. holes were drilled in the bottom of the shaft. Several car loads of cement were pumped through these holes into the fissures, the drill casings being used for grout pipes. This largely cut off the flow of water from, the bottom and sinking could then

Jan 1, 1915

By L. G. Graper, E. L. Ramsey

The present program of increased production of armament and lend-lease material for mechanized war has created a problem for the shops that must do the machining. They have naturally turned to the steel producers to provide them with a steel that can be accurately machined with the greatest possible speed. The adding of sulphur to steel as an aid to machinability has long been practiced; however, it may cause red shortness in rolling and may result in billets with seams and cracks, which must be removed before rolling can be continued. This conditioning of billets slows down production and adds to the cost of making steel. Experience with various grades of sulphur steel made to different specifications has shown that Bessemer sulphur steels roll better than similar grades made in the open hearth, and also that low-carbon steels that are open or semikilled as tapped roll better than completely deoxidized or killed steels. This has led to the belief that oxygen plays an important part in the rolling qualities of sulphur-bearing steels. It is generally held that the red shortness of sulphur steels is due to the low melting point of the sulphide compounds that separate out at the grain boundaries when the steel solidifies. At the rolling temperature of steel these sulphide compounds melt and cause cracking. Adding Sulphur as a Sulphite These difficulties can be avoided if sulphur is introduced into the stecl by means of an anhydrous sulphite, such as sodium sulphite or sodium bisulphite, which—upon decomposition—forms sulphur dioxide. It is assumed that, in this case, the oxygen of the sulphur dioxide forms part of the sulphide inclusions in the steel, and raises their solidification temperature enough so that the formation in the steel of a low-melting eutectic is prevented. The result is the prevention of hot shortness and a more uniform distribution of inclusions. Fig. I shows photomicrographs of steels made with sulphur, while Fig. 2 illustrates the structure of steels made with sodium bisulphite. These samples were cut from the steels as cast—that is, before rolling— and were polished and lightly etched to bring out the grain boundaries. Fig. I shows that sulphide inclusions segregate more or less at the grain boundaries, and it can be seen how the continuity of these low-melting inclusions would cause red shortness. Fig. 2 shows that the sulphide inclusions arc more or less uniformly distributed throughout the matrix. This arrangement of inclusions has no apparent effect on the rolling qualities of a steel. This statement has been substantiated by the observation of the rolling of over 6000 tons of steel made with a sulphite addition. Results As stated before, the sulphur dioxide is added to steel by means of the decomposi- 27

Jan 1, 1942

By R. W. Gurry

IN the course of a series of measurements of the rate of diffusion of carbon in austenite at about 960°C. (1760°F.) and 1110°C. (2030°F.), it became necessary to determine carbon concentration when austenite is saturated with graphite, because of indications that the hitherto accepted value of the solubility of graphite in this temperature range is too low. Specimens of carbonyl iron were saturated at temperature in a hydrogen-toluene atmosphere which was in equilibrium with graphite. The carbon content of the steel was measured either by combustion or by the loss of weight after substantially complete decarburization in an atmosphere of hydrogen saturated at room temperature with water vapor. The absence of carbide or graphite as a phase at temperature was demonstrated by microscopic examination of saturated specimens that had been drastically quenched. The mean result of concordant measurements is that gamma iron saturated with respect to graphite contains 1.39 per cent carbon at 957°C. (1755°F.) and 1.89 per cent at 1110°C. (2030°F.). These two values, plotted along with what appears from the literature to be the best value (0.69 per cent) at the iron-graphite eutectoid temperature (738°C. or 1360°F.), yield a curve that is almost linear, and which, continued upward through a short interval, leads to a limiting solubility of 1.98 per cent at the iron-graphite eutectic temperature 1135°C. (2075°F.) instead of the commonly accepted value of about 1.7 per cent. EXPERIMENTAL PROCEDURE The samples of carbonyl iron, with the exception of those for microscopic examination, were cylinders 3.8 cm. long, 0.475 cm. in diameter. They were placed in a graphite holder and suspended in a vertical furnace, the temperature of which was controlled precisely by means of the resistance of the platinum winding.1 In the empty furnace the range of temperature over a distance of 4 cm. at the heat center was 9°C. (16°F.) but at a given location the temperature remained within ±0.5°C. (0.9°F.); it was measured by inserting a platinum-rhodium couple calibrated at the melting point of palladium, 1555°C. Since the presence of the sample probably improved the uniformity of temperature, it was decided to consider the average temperature of the sample to be 3°C. lower than the maximum measured in the empty furnace with the same setting of the controller. Data were obtained in the vicinity of two temperatures, 960°C. (1760°F.) and 1110°C. (2030°F.). The samples for microscopic examination, 2.2 cm. long, 0.475 cm. in diameter, were suspended individually on pure iron wire; for these no temperature correction was applied. The carburizing gas passed into the furnace was dry, oxygen-free hydrogen saturated at room temperature with

Jan 1, 1942

By W. W. Smeltzer

The linear formation rates of wustite films have been determined over the temperature range 590° to 1030°C using a vacuum microbalance technique. These rates are dependent directly on the partial pressure of carbon dioxide, Activation energies for film and scale formation in carbon dioxide and oxygen atmospheres are equivalent to either the dissociation energies of carbon dioxide and oxygen, or the activation energy for the diffusion of cation vacancies in wustite. Of the metals, iron is the most widely used basic constituent of durable alloys despite its intrinsic property of converting at a relatively rapid rate to the oxides from which it is derived. This property has accounted partially for many investigations on the oxidation of iron at low and high temperatures. Interpretation of the results have been complicated by formation of multilayers containing the various oxide phases. To aid in elucidation of the oxidation kinetics, measurements have been carried out in this investigation on the growth of the wustite surface layer. There are several studies directly related to the objective of this investigation. The growth of multilayer scales has been demonstrated to be diffusion controlled in oxygen at atmospheric pressure over the temperature range 450" to 1200°C by Davies, Simnad, and Birchenall.' To gain an insight into the diffusion mechanism, Himmel, Mehl, and ~irchenall' determined the self-diffusion coefficients of iron in wustite, magnetite, and hematite. These results were used to evaluate the oxidation rates of iron and its oxides, taking as a basis the theoretical parabolic rate equation developed by Wagner., The evaluations were compared to experimental data and were shown to provide essential confirmation of this theory. Fischbeck, Neundeubel, and salzer4 have demonstrated that the oxidation of iron in carbon dioxide and other atmospheres at high temperature obeys a linear relationship. Because the Arrhenius temperature coefficients of the reaction rate constants changed at the A, point they concluded that the passage of iron into the oxide controlled the reaction. This viewpoint has been accepted by Benard and ~albot' to explain the occurrence of linear rates in oxygen before the onset of diffusion-controlled parabolic oxidation. This basic premise of a rate-controlling reaction at the iron/oxide interface has been questioned recently by Hauffe and Pfeiffer.' Since the formation of wustite may follow a linear or parabolic relationship depending on the pressure of the gaseous re-actant, they propose that linear oxidation is determined by a chemisorption reaction while parabolic oxidation is controlled by diffusion. As confirmation of this viewpoint, Hauffe and pfeiffer6 and Pfeiffer and ~aubmeyer' have established that the linear scaling constants of wustite formation at 1000°C in carbon-dioxide or low-pressure oxygen atmospheres

Jan 1, 1961

By L. G. Graper, E. L. Ramsey

The present program of increased production of armament and lend-lease material for mechanized war has created a problem for the shops that must do the machining. They have naturally turned to the steel producers to provide them with a steel that can be accurately machined with the greatest possible speed. The adding of sulphur to steel as an aid to machinability has long been practiced; however, it may cause red shortness in rolling and may result in billets with seams and cracks, which must be removed before rolling can be continued. This conditioning of billets slows down production and adds to the cost of making steel. Experience with various grades of sulphur steel made to different specifications has shown that Bessemer sulphur steels roll better than similar grades made in the open hearth, and also that low-carbon steels that are open or semikilled as tapped roll better than completely deoxidized or killed steels. This has led to the belief that oxygen plays an important part in the rolling qualities of sulphur-bearing steels. It is generally held that the red shortness of sulphur steels is due to the low melting point of the sulphide compounds that separate out at the grain boundaries when the steel solidifies. At the rolling temperature of steel these sulphide compounds melt and cause cracking. Adding Sulphur as a Sulphite These difficulties can be avoided if sulphur is introduced into the stecl by means of an anhydrous sulphite, such as sodium sulphite or sodium bisulphite, which—upon decomposition—forms sulphur dioxide. It is assumed that, in this case, the oxygen of the sulphur dioxide forms part of the sulphide inclusions in the steel, and raises their solidification temperature enough so that the formation in the steel of a low-melting eutectic is prevented. The result is the prevention of hot shortness and a more uniform distribution of inclusions. Fig. I shows photomicrographs of steels made with sulphur, while Fig. 2 illustrates the structure of steels made with sodium bisulphite. These samples were cut from the steels as cast—that is, before rolling— and were polished and lightly etched to bring out the grain boundaries. Fig. I shows that sulphide inclusions segregate more or less at the grain boundaries, and it can be seen how the continuity of these low-melting inclusions would cause red shortness. Fig. 2 shows that the sulphide inclusions arc more or less uniformly distributed throughout the matrix. This arrangement of inclusions has no apparent effect on the rolling qualities of a steel. This statement has been substantiated by the observation of the rolling of over 6000 tons of steel made with a sulphite addition. Results As stated before, the sulphur dioxide is added to steel by means of the decomposi- 27

Jan 1, 1942

By D. B. Masson

THE martensitic transformations occurring in ,B AgCd as a result of cold working at room temp.. a-ture and cooling below room ten1erature have been reported by Masson and Barrett.1 These authors found that simple cooling to 127°K caused body-cen-tered-cubic AgCd containing 53.0 at. pct Ag to transform to an orthorhombic structure, while mechanical deformation at room temperature was found to induce a hexagonal structure. Furthermore, they observed that the temperature at which martensite first appeared on cooling, Ms, and that at which the transformation was complete, MI, as well as the corresponding temperatures for the reversion on heating, As and Af, could be ascertained to ± 1°K, although they report these temperatures only for the alloy containing 53.0 at. pct Ag (51.96 wt pct). In the work reported here the temperature range of the transformation on cooling was studied as a function of composition, and further consideration was given to the relationship between the original bcc structure and the transformation-induced hexagonal and orthorhomic structures. EXPERIMENTAL PROCEDURE All samples studied were in the form of annealed 300-mesh powder and were prepared from the constituent metals in the manner described previously.1 The compositions in weight and atomic pct are listed in Columns 1 and 2 of Table I. Method of determination of transformation temperature was the same as utilized previously1—measurement of the intensity of X-rays diffracted by the transformation product. A diagram of the low-temperature sample mount is shown in Fig. 1. This mount was made to fit a GE model XRD-3 X-ray diffraction unit. The powder sample was varnished to a flat surface on a copper sample holder at A. This sample holder was soldered to a brass rod, B, which was in turn soldered to a second copper rod, C. Rod C could be cooled to 78°K by inserting it into a reservoir of liquid nitrogen, D. A heater coil of nichrome wire was wound at E on the sample holder, and by varying the current through the heater a thermal gradient could be maintained along brass rod B. Sample temperatures from 78°K up could be attained and held as long as needed. The sample holder and nitrogen reservoir were insulated by Styrofoam, F, and the sample itself was protected by two radiation shields, G. The X-ray beam entered and left the sample chamber through windoys in the supporting container at H and I. Sample temperatures were measured by a calibrated copper-constantan thermocouple varnished to the sample holder just out of the beam at A. The low-temperature structure had been found to possess orthorhombic symmetry1 and the expected positions of the diffraction maxima were known. Therefore the sample was aligned in the X-ray beam and the Geiger counter of the spectrometer set to receive the most intense orthorhombic reflection without an adjacent interfering cubic reflection. This

Jan 1, 1961

By G. G. Gunther, C. L. Boeke

(Any discussion resulting from presentation at the Los Angeles Extractive Metallurgy Division Meeting will appear in the November 1953 issue.) I. M. Symonds (Cia Minera de Penoles, Monterrey, N. L., Mexico)—This is a most interesting paper describing zinc deleading operations. Mr. Gunther was kind enough to show me his operation last fall. Outside of the excellent results, I was most impressed by the ease of control. I had expected a difficult operation requiring careful control and supervision, but found one requiring little attention. The only disappointing fact of the separation is the poor copper recovery in the concentrate of the deleading float. The final zinc concentrate still has about 3 pct chalcopyrite (assuming the copper is chalcopy-rite) which adversely affects the zinc grade. If the copper mineral is free, the poor recovery is probably due to the high cyanide concentration. Fortunately, about 75 pct of the copper is recovered before the pulp goes to the zinc circuit. Mr. Gunther's remarks about proper cyanide to zinc sulphate ratios cannot be overemphasized. Although our experience has been mostly confined to the conventional Pb-Zn separation, we have found these ratios most important, and more important in the mill than in the laboratory. The laboratory appears to give optimum results over a wider range of reagent control than the mill, especially in regard to the effect of cyanide. In the mill in handling relatively high grade ores, we usually find that ratios of cyanide to zinc sulphate of 1 to 10 or 20 or higher work best. Where there is a substantial amount of chalcopyrite present, it has been noted that cyanide in the range of 0.02 to 0.10 lb per ton of ore is often beneficial in zinc rejection and possibly in copper recovery in the lead concentrate, while larger amounts, say 0.20 to 0.25 lb per ton, will float more zinc and cause a serious rejection of chalcopyrite. On an ore with 15 to 20 pct Zn content, we have had to use 0.8 to 1.2 lb of zinc sulphate per ton of ore with the above cyanide to obtain optimum results. Sometime ago we made a laboratory study following Parral and Santa Barbara practice on an ore which never had a very satisfactory rejection of zinc in the lead concentrate. The results were not encouraging and a mill test was not run. Since then the study has been reopened using fresh zinc concentrate from the mill in the laboratory instead of one made in the laboratory. A reasonably clean lead float from the zinc concentrate was obtained. It has been checked several times and each time the zinc concentrate made in the mill responds and the one made in the laboratory does not. Although undoubtedly some reasonable explanations can be given, it is presented here only

Jan 1, 1953

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

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

Jan 1, 1951

By E. T. Stephenson

Wert and Marx1 pointed out that a straight-line relationship exists between the activation energy of a relaxation process and the temperature at which the maximum relaxation occurs. The data available at the time indicated that this relationship held for a number of Snoek peaks, several grain boundary peaks, and pair relaxation in a brass. Theoretical justification for the observed relationship was found in the following equation, which Wert and Marx derived from the theory of interstitial atomic diffusion coefficients:' where AH is activation energy, R is the gas constant, T is the internal friction peak temperature, v is the interstitial atomic frequency, f is the frequency of the applied stress, S is entropy of activation. More accurate internal-friction data on interstitial diffusion in bcc metals have become available since Wert and Marx's paper. Table I gives activation energies calculated by several authors from a combination of elastic aftereffect data and the peak temperatures at several frequencies of measurement. (In the case of Fe-C and Fe-N, the referenced authors5'7'8'10 also used high-temperature diffusion data.) However, diffusion experiments at 150°C and above1'-13—rather than anelastic techniques—were used to determine the activation energy of hydrogen diffusion in Fe-H. In order to plot this recent data as a AH vs T graph we calculated the peak temperatures of the systems listed in Table I (with the exception of Fe-C, Fe-N, and Fe-H) by using published values3'' of AH and Do. These calculated peak temperatures varied by only a few degrees from experimentally determined peak temperatures. The calculated values were felt to represent the better average. A number of investigators have found the peak temperatures in Fe-C and Fe-N to be 38 to 40°C and 20° to 24°C, respectively. The values of 39" and 22°C (312° and 295°K in Table I) were determined by carefully executed experiments.6'8 The hydrogen peak temperature has been measured at 1 cps,12 20 cps,14 and 35,000 cps.15 All hydrogen peak temperatures were adjusted to 1 cps for Table I using AH = 3000 cal per mole. As Fig. 1 shows, the data of Table I fit a straight line at least as well as the earlier data used by Wert and Marx. Linear-regression analysis was applied to the data in Table I. The Fe-H data was omitted because the activation energy has not been experimentally determined in the temperature range of the peak. The analysis yields the equation: with a standard error of 860 cal per mole. The agreement of the data in Table I with the Wert-Marx equation is excellent. Strict application of Eq. [I] forces the linear regression analysis through the origin. In this case, the analysis gives a slope of 62.1 cal per mole per deg, and a standard error of *890 cal per mole. With this value, it follows from internal-friction theory that Hence, Do varies by only 30 pct among these bcc-interstitial systems. Thus, a concept suggested originally by Wert and Marx1 is confirmed by recently published data. The

Jan 1, 1965

By A. C. Losekamp, R. M. Fincel, J. B. Conway

TANTALUM like several other metals exhibits a great affinity for or reactivity with certain gases. Tested in atmospheres which are not completely pure this metal becomes contaminated by certain impurities (notably nitrogen, oxygen, and hydrogen) with a resultant change in various properties. Well-known are the drastic changes in hardness, ductility, and electrical resistivity. Also previously recorded are the distortions of the lattice1,2 accompanying the increased content of oxygen and nitrogen in tantalum. This pronounced affinity for certain gases along with the significant property variations thereby produced makes it necessary to observe additional experimental precautions in making various property determinations with tantalum. Some recent measurements of the linear thermal-expansion characteristics of tantalum have made this point quite emphatically. Failure to observe the greatest precautions in reducing the contaminating influence has led to some extremely distorted results. Not only does such operation lead to different heating and cooling curves during the thermal-expansion measurements, but upon cooling to room temperature the sample is found to have suffered a permanent (actually semipermanent since the specimen can be returned to its original length by a 3-hr degassing treatment in vacuum at 2300°C) elongation. In addition, subsequent heatings will not follow the initial heating curve unless the expansion measurements are based on the actual specimen length at the beginning of the heating cycle. These thermal-expansion measurements were made by positioning the specimen horizontally at the longitudinal midpoint of a 20-in.-long, 4-in.-diam resistively heated tungsten tube furnace. The specimens were in the form of 20-mil sheet formed into a sharp U-shape about 0.3 in. in width and 0.3 in. in depth. Fiducial marks in the form of 0.010-in. holes drilled at either end of the 2.5-in.-long sheet specimens were viewed by a paired filar micrometer telescope system mounted on an Invar bar on the top of the furnace. Tests were made in helium with temperatures measured by a Pt/Pt-lORh thermocouple to 1000°C; above 1000°C temperature was measured by means of an optical pyrometer sighted into a black-body hole drilled into the specimen-holding fixture in the vicinity of the specimen. Previous tests with black-body holes drilled in a 0.25-in.-diam rod specimen had confirmed the temperature uniformity between specimens and specimen-holding fixture in this furnace. A typical thermal-expansion curve for arc-cast tantalum in helium is shown in Fig. 1 for the case where no attempt is made to remove all the impurities from the cover-gas atmosphere. The initial heating curve appears smooth and has a shape characteristic of all metals. However, it is known that the sample begins absorbing impurities at temperatures above 1000°C which are sufficient to distort the lattice and lead to specimen lengths which are somewhat larger than would be obtained in the absence of contamination. During the cooling cycle the expansion curve is seen to be significantly different from the heating curve. This is due to the impurity content causing a semipermanent (i.e., permanent until the impurities are removed by vacuum degassing) deformation which will not allow the speci-

Jan 1, 1965

By R. S. Silo

A new process for leaching oxidized copper ore (CuO) is presented. Its principle advantage is that it requires low cost raw materials. For large operations the only materials used are sodium chloride (NaCl) and calcium carbonate (CaCO3). In those operations of medium size, sulfur, which is abundant in the North of Chile, is used. However the process may also be applied to small copper plants provided low cost sulfuric acid is available as a substitute for sulfur. That is, it may be applied to those small capacity plants where large additional installations for the regeneration of the leaching solution are not needed. The method is comprised of six stages, namely: Crushing, Leaching, Washing of the Solution, Precipitation, Regeneration of the Leaching Solution, and Fusion. A flow diagram for this process in a large or medium-sized plant is given. LEACHING This process requires hydrochloric acid as the leaching solution for treating copper ores. This is easily produced from sulfuric acid and sodium chloride as follows: A small excess of NaCl is advisable to avoid the presence of free sulfuric acid. The experiments reported here were performed with 36 grams of HC1 per liter of solution to obtain a free acidity equivalent to 50 gpl of H2SO4 (the usual leaching solution concentration). Hydrochloric acid leaching may be done by either of the usual methods; i.e. percolation or agitation. The method selected depends on the characteristics of the ore. No matter which method is chosen, low pressure air can be used either as an air lift for circulation of the solution, or for agitation of the pulp with "Pachuca" tanks. In either case, there are no problems with respect to oxidation. Leaching by agitation produces good recovery; but at times, the problem of high consumption of leaching solution or the overproduction of grinding slimes which interfere with contact between phases negates this advantage. The percolation method was used in this study. Percolation took place in four 15-liter containers, 11 1/2 in. diam. by 12 in. height, having hemispheric bottoms. Each container was fitted with a wooden bottom having 1/4 in- holes. The air lift was a 3/4 in. plastic tube 10 in. high. Air at 6 psi was supplied from a rotating laboratory compressor. 10 kg of ore were charged to the container and counter-current leaching was used. The leaching solutions were fed through the center pipe in order to use the ascending solutions to reject the air in the charge. The solution circulated through the air lift in order to facilitate the movement according to the increase in density produced by the reaction between the phases present. Specific observations made in this report cannot be generalized because the leaching characteristics are dependent on the nature of the ore used, the method of leaching, and the production of primary or secondary slimes while crushing or leaching.

Jan 1, 1967

By R. W. Fenn, H. A. Lepper, W. R. Hibbard

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

Jan 1, 1951

By H. A. Shaw, H. G. G. Whitton

This paper describes the use of the electric-arc furnace for the production of tough-pitch, horizontal cast copper shapes and the production of copper anodes from tank house anode scrap. This installation is the only one of its kind in the United States. SINCE the initial use of electric-arc furnaces for refined copper melting on a commercial scale at Copper Cliff in 1936, this type of melting has been of extreme interest to everyone in any way connected with copper casting and fabricating. Early in 1948, Kennecott Copper Corp. decided to build an electrolytic copper refinery at Garfield, Utah. One of the major decisions to be made at that time was the type of furnace to be used for melting cathodes for casting into refined copper shapes and for melting anode scrap for the production of anodes. Based on the results obtained from arc furnaces by the Tnternational Nickel Co. of Canada, at Copper Cliff, and by the American Smelting and Refining Co., at Baltimore, coupled with certain important economic factors such as having an adequate amount of relatively cheap power available at the Magna Central Power Plant, high cost of poles and oil in the Salt Lake District, and the lack of natural gas at that time, the final decision was to install electric-arc furnaces rather than the standard type of rever-beratory furnaces. Experience has shown certain parts of the original installation to have been inadequate or in need of modification. Although the present status of arc-furnace melting at the Utah Refinery is far from perfect, nevertheless substantial progress has been made. At the Utah refinery, copper is melted in three 15-ft diam, size "M" Lectromelt furnaces, two of which are used for melting cathodes and one for remelting anode scrap. Each furnace is energized from a 13,800-225 v, 3-phase, 60-cycle, 6000-kva (at 40°C rise) oil-immersed water-cooled transformer. Auxiliary equipment for each refined copper furnace includes a charging machine, launders, ladle, 40 ft diam Walker-type casting wheel, bosh, and inspection conveyor. The anode furnace differs in that it is charged by hand from an elevated platform and the anodes are picked off the 24-ft diam casting wheel by a takeoff hoist and loaded directly into specially designed trailers, Figs. 1 and 2. Electrical Equipment Electric power to operate the arc furnaces is generated at the Kennecott-owned Magna Central Power Station two miles distant and transmitted to the Refinery outdoor substation at 44,000 v. At this point it is stepped down to 13,800 v and carried through concrete-lined underground tunnels to the switchgear and then to the primary side of the furnace transformers. The power is conducted from the Y-connected secondary bus bars of the transformer to the electrode arms by means of 22 flexible copper cable conductors per phase. Each cable has a cross-sectional area of 1,750,000 circular mils, giving a total area of 38,500,000 circular mils per phase. The power is carried from these cables to the three individual electrode holders by means of water-cooled copper bus tubing. The electrode clamps are incorporated in the electrode holders and are released by air pressure working against a spring in the clamp control cylinder. The movement of the electrode arms is either automatic or manual. The automatic control is used during melting and is regulated by three Westing-

Jan 1, 1954

By R. H. Doremus, E. F. Koch

The first carbide phase that precipitates at 120°C from a iron containing about 0.02 wt pct, C was studied with the electron microscope. In both strained and unstrained material the carbide particles appeared to form on the dislocations in the iron, in agreement with the results of the kinetic study in part II. The structure of the carbide was tentatively identified as an expanded a-iron lattice (body-centered tetragonal) with about 0.7 wt pct C. This phase appeared to be metastable and apparently transformed into cementite or epsilon carbide when the iron was over-aged. WhEN a iron containing less than about 0.025 wt pct C is quenched from 700°C to room temperature the carbon is retained in solid solution and precipitates out slowly with time. The morphology and structure of the precipitating phase are still in doubt, two different carbide phases, epsilon carbide and cementite, have been identified, but there is considerable disagreement about the time and temperature at which they appear.1-4 Epsilon carbide seems to form at lower temperatures than cementite, although one group of investigators4 found only the latter phase over a wide range of aging temperatures. Usually these two phases have been identified after aging times much longer than needed to drain the carbon from supersaturated solution, at least for aging temperatures below 200oC. The precipitated carbides have been observed with the electron microscope,'-5 but again most results have come from samples aged much longer than necessary to reduce the carbon concentration in solution to a low value. In this study the carbide phases were observed in earlier stages of precipitation with the electron microscope, in order to learn more about the shape, size, and distribution of the carbide particles. Samples strained small amounts before aging, as well as unstrained samples, were studied, and the structure of the carbide phase was tentatively identified. EXPERIMENTAL "Ferrovac E" vacuum-melted iron, from which the wires for kinetic studies were also made (Refs. 6 and 7 and the following paper) was used for this work. An analysis of this material is given in Table I; it agrees well with the manufacturer's values. The number of inclusions in the iron was very small, as can be seen in the optical micrograph in the following paper. Pieces of iron about 1 mm thick and 5 to 10 mm in diam were used for all the studies except for those in which the sample was strained before aging; then the specimen was a rod 3 to 6 mm in diam and 20 cm long. The samples were held at about 700°C in wet hydrogen for at least an hour to remove residual carbon and nitrogen, after which they were heated to 950" (in the y region) and cooled to room temperature in the furnace. This treatment gave an average grain diam of more than 1 mm, so that the effect of grain boundaries on the precipitation process was negligible. The samples were carburized at 710°C for about 2 hr in hydrogen that had passed over toluene held at -30°C; then they were rapidly quenched into water. From chemical analysis this treatment gave a carbon concentration of from 0.012 to 0.016 wt pct, which is less than the solubility of carbon at this temperature. After quenching, the samples were aged at 120 "C in an oil bath for vari-

Jan 1, 1961

By K. D. Sheffler, R. W. Hertzberg, R. W. Kraft

The Ni-Ni,Ti lamellar eutectic alloy responds to unidirectional solidification by alig-nment of the platelets of the two phases roilghly perpendicular to the moving solid-liquid interface. X-ray diffraction and zetallographic experimenls show that the following cryslallographic relationships exist in the directionally solidified ingots: lamellar interface w [0001] Ni3Ti 11 (111 }Ni local growth direction (1120) Ni3Ti n (110) Ni These relationships are rationalized on the basis of the crystallography of the constituent phases. In recent years researchers in several laboratories have shown that an aligned or "controlled" micro-structure can be produced in many binary eutectic alloys by forcing the alloy to solidify unidirectionally in an elongated crucible.'-3 It also has been found that a unique crystallographic relationship often develops between the two phases in such a structure.4, 5 Since a parallel arrangement of phase particles in an alloy should, and, in many cases, does, produce unique and highly anisotropic properties, the alloys so produced are of potential interest in a variety of applications. Studies of these alloys are also of interest at the academic level because they help to improve our basic understanding of the way in which metals solidify. This paper reports the metallographic and crystallographic structure of another of these alloys—that between nickel and Ni3Ti. In a separate paper, hot tensile and stress rupture properties as well as deformation and fracture behavior of the same alloy are discussed.19 EXPERIMENTAL DETAILS AND RESULTS General Characteristics of the Eutectic Reaction. Several small exploratory Ni-Ti melts were made to resolve a conflict in Hansen6 over the eutectic composition. The alloy containing 16.4 at. pct Ti, Fig. 1, appeared metallographically to be closest to the eutectic composition, although small dendritic patches of proeutectic nickel are present. The eutectic morphology is lamellar and appears quite sharp and angular, indicating a strong influence of crystallographic factors upon the microstructure. Some of the eutectic nickel platelets are continuous with the primary phase, whereas in hyper eutectic melts a "halo" of nickel generally surrounded the proeutectic inter-metallic platelets, indicating that nickel is the nucleating phase in this eutectic reaction. Extensive precipitation of Ni3Ti has occurred in the primary nickel phase in a Widmanstatten pattern which makes the dendrites difficult to differentiate from the eutectic mixture. This phenomenon may account for discrepancies among the eutectic compositions reported in the literature. The sloping solvus line bounding the nickel primary solid solution explains the presence of the intermetallic precipitate. Poole and Hume-Rothery7 report that the solubility of titanium in nickel drops from 13.8 at. pct at the eutectic temperature (1304°C) to approximately 9 at. pct at 700°C (Ni3Ti, on the other hand, melts con-gruently and exhibits essentially no solubility for nickel or titanium). Lever rule calculations based upon Poole and Hume-Rothery's phase diagram indicate an increase from 21 vol pct Ni3Ti in the eutectic mixture at the eutectic temperature to 61 pct at room temperature. Selective point count measurements of the volume fraction of eutectic Ni3Ti platelets in the eutectic microstructure yielded a value of 46 i 2 vol pct, which required extensive precipitation of Ni3Ti from the nickel solid solution onto the eutectic intermetallic platelets during cooling from the solidification temperature. Influence of Directional Solidification. Nickel bar and titanium sponge of 99.9 pct purity were induction-melted in zirconia crucibles under argon in the proportion of 13.8 wt pct (16.4 at. pct) Ti to produce 14-lb master heats which were cast into investment molds yielding 5/8-in.-diam pins approximately 10 in. long. These pins were remelted under argon in horizontal zirconia boats and unidirectionally solidified to produce 1/2-in.-sq ingots approximately 44 in. long. Details of the experimental techniques and apparatus are described elsewhere.'

Jan 1, 1970

By G. C. Wallick, W. R. Foster, J. G. Savins

A comprehensive review of the salient features of the differentiation method of rheological analysis in Poiseuille flow from its inception circa 1928 is presented. Here no initial assumptions regarding the nature of the function relating rheological parameters to observed kinematical and dynamical parameters are required in the data-analysis process. In contrast, the integration method involves interpreting flow properties in terms of a particular ideal model. It is shown that, although both methods represent modes of solution of the same integral equation, being relatively bias-free, the differentiation method offers a more discriminating procedure for rheological analysis. The application to problems involving plane Poiseuille flow is also described. INTRODUCTION In most instances, the approach to the problem of interpreting the rheological properties of various compositions as they are affected by changes in chemical or physical environment, assaying the characteristics of a particular constituent of a suspension, analyzing flow behavior in terms of interactions between components in a system, to cite but a few examples, has been in terms of what Hersey1 terms the integration method. Briefly, it consists of interpreting flow properties in terms of a particular ideal model. The usual practice of the integration method is to choose a model with a minimum number of parameters because, other things being equal, it is desirable to use the simplest model which will describe the behavior of a real material and yet be mathematically tractable for the requirements of data analysis. This expression is then substituted into an equation which relates observed kinematical and dynamical quantities, such as volume flux Q and pressure gradient ], and angular velocity and torque T, in a capillary and concentric cylinder apparatus, respectively. The rheological parameters appear on integrating, in an expression relating the pairs of observable quantities such as those just given. In many instances a particular model provides a good representation of rheological behavior over a reasonable range of compositional and environmental changes. Just as often, however, it is obvious that the interpretation of rheological changes by the integration method is not providing realistic information about changes in flow behavior. A more general method of interpreting rheological data for a given material is to make no initial assumptions regarding the nature of the function relating rheological parameters to observed kinematical and dynamical quantities, e.g., flow rate and pressure drop in capillary flow or angular velocity and torque in a rotational viscometer. This general method Hersey1 terms the differentia-ation method. Instead of integrating, one differentiates the integral equation with respect to one of the limits, i.e., one of the boundary conditions; the resulting expression contains the same observable quantities just given, their derivatives, and the rheological function evaluated at that boundary. By obtaining these derivatives from experimental data, graphically or by a computer routine, they can be substituted into the differential equation and a graphical form of the function derived. THEORY OF THE DIFFERENTIATION METHOD FOR POISEUILLE-TYPE FLOWS In this introductory paper, two flow cases which are important in viscometry are considered (one for the first time) from the differentiation method of analysis, flow in a cylindrical tube and flow between fixed parallel surfaces of infinite extent, the basic integral equations being formulated in a manner analogous to the way they originally appeared in the literature. In addition, the following ideal conditions will be assumed: (1) an absence of anomalous wall effects, (2) isotropic behavior everywhere, and (3) steady laminar flow conditions. Before proceeding it would seem appropriate to preface the analysis of these flow geometries with the comment that the basic philosophy of the differentiation method could be said to stem directly from the efforts of Herzog and weissenberg2 to indicate systematically the various ways of using

By M. J. Marcinkowski, E. N. Hopkins

An investigation has been made of the ß(fcc) — a(hexagonal) transformation which occurs in lanthanum using both electrical resistivity and transmission electron microscopy techniques. It has been shown that the ß phase can be retained below the transformation temperature by rapid quenching but that the sample immediately begins to transform to the a phase. The transformation is observed to nucleate in the vicinity of inclusions. Based on the above observations, a detailed model of the transformation has been advanced which involves the nucleation of an extrinsic stacking fault bounded by a pair of Shockley partial dislocations in the vicinity of some heterogeneity, i.e., an inclusion. The stress field of the resultant dislocation pair acts to nucleate extrinsic faults in adjacent planes and leads quite naturally to the B-a conversion with a minimum of strain energy induced in the crystal. LANTHANUM possesses an fcc structure ß) (8) upon cooling transforms to a hexagonal modification (a) in much the same way as cobalt. The one exception, however, is that the stacking sequence of the closest packed planes in a La is ABAC ABAC, and so forth,1 whereas in cobalt it is ABAB, and so forth, i.e., hep. Mainly on the basis of transmission electron microscopy techniques, there seems to be little doubt that the 0 — a transformation in cobalt involves a dislocation mechanism2 although its exact nature still remains obscure. Although the ß - a transformation in lanthanum is somewhat more complex than that occurring in cobalt, it was thought that its very uniqueness would be helpful in understanding fcc — hexagonal transformations in general. Such a general understanding of these transformations is important since they represent what are perhaps the simplest of the martensitic class of transformations. The experimental techniques used were those of electrical resistivity and transmission electron microscopy. EXPERIMENTAL PROCEDURE The lanthanum used in this investigation was prepared by the calcium reduction of lanthanum fluoride in a tantalum crucible under an argon atmosphere as described by Spedding et al. The residual calcium was removed by vacuum remelting in a tantalum container. Portions of the metal were then analyzed by emission spectroscopy as well as vacuum fusion. The amounts of the various impurities that were found are listed in Table I. A portion of the lanthanum ingot was swaged at room temperature into 0.030-in.-diam rod for the resistivity samples while the remainder was rolled into 0.010-in.-thick sheet for the transmission electron microscopy phase of this investigation. In order to eliminate the plastic deformation induced in the samples during fabrication, they were sealed in evacuated tantalum lined quartz capsules and annealed for 1 hr at 700° C. Specimen resistances were measured using the conventional "four-wire" technique described by MacDonald, 4 employing a type K-3 Universal Potentiometer and a standard 0.001-ohm resistor, both manufactured by Leeds and Northrup Co. By taking into account all of the possible errors in the apparatus, it was felt that the absolute resistivities of the approximately 0.87-in.-long specimens measured are reliable to 3 pct. Resistances from room temperature to 700°C were measured using a vacuum furnace. In order to avoid sample contamination by the thermocouple, chromel-alumel leads were spot-welded into a tantalum shield which in turn was welded to the lanthanum specimen. Resistances below room temperature were obtained by transferring the sample to a helium gas-filled quench tube and slowly dripping liquid nitrogen into a surrounding dewar flask. A steady reduction of temperature to about —190°C was completed in about 23 hr, and the resistances were measured at various temperature intervals. As will be shown shortly, rapid quenching from above about 350°C was sufficient to suppress the B -a transformation initially but subsequent annealing at lower temperatures leads to a partial ß --a conversion. To investigate this aspect of the transformation, the samples were suspended in the hot zone of a helium-filled furnace from which they could be rapidly dropped into the quench tube mentioned previously which was now

Jan 1, 1969

By Robert Baschwitz, John Chipman

The distribution of silicon between liquid silver and Fe-Si-C alloys has been studied at 1420oand 1530°C. The data are consistent with earlier studies. New data of Hager on the liquidus lines of the system Ag-Si and the distribution data are used to obtain the activity coefficient of silicon in both liquid phases. Data on the heat of mixing in iron permit accurate extension to 1600°C. Equilibrium data involving SiO2 and silicon in liquid iron together with revised data on the free energy of SiO2 are used to fix the activity of silicon in the infinitely dilute solution. The binary system exhibits strong negative deviation from ideality. At infinite dilution ? Si at 1600" is 1.25 x 10'3, and at concentrations up to NSi = 0.4 the slope d InySi/dNSi has a constant value of r; = 13. It is found that logysi in the ternary solutzon is approximately but not exactly the same function of Nsi + NC as of NSi in the binary. The results are consistent with currently available data on the free energy of Sic and its solubility in molten iron. LIQUID solutions of the system Fe-Si-C have acquired considerable importance as the laboratory prototypes of blast furnace hot metal. Equilibrium studies involving such solutions and slags approximating those of the blast furnace have yielded useful information concerning the thermodynamic properties of blast furnace slags. In studies of this kind great importance attaches to a knowledge of the thermodynamic activity of silicon in the solution as a function of temperature and composition. An attempt was made by Chipman, Fulton, Gokcen, and askey' to evaluate all of the pertinent data on this system and to deduce the desired relation between activity, composition, and temperature. These authors published data on the solubility of graphite and Sic in molten Fe-C-Si solutions and on the distribution of silicon between liquid iron and liquid silver. They showed further how the activity of silicon in very dilute solutions in liquid iron could be calculated from equilibrium data involving the molten alloy and solid SiO,. These calculations rested on the published thermodynamic properties of SiO, in- cluding its heat of formation which at that time was recorded as -209.8 kcal. This value has been under suspicion for some time and has recently been replaced by the concordant results from two independent laboratories2,3 which place the heat of formation of a-quartz at -217.6 kcal. This revision necessitates a re-evaluation not only of the activity of SiO2 in slag but also of silicon in molten iron. It is the purpose of this paper, therefore, to recalculate the activity of silicon, and in furtherance of this objective to present new data on its distribution between liquid Fe-Si-C alloys and liquid silver. HEAT OF SOLUTION OF SILICON IN IRON In order to determine the effect of temperature upon the activity coefficient it is necessary to know the heat of solution of silicon in iron as a function of composition. This is found in the data of Korber and Oelsen4 shown in Fig. 1. The curve corresponds to the following equation, which is of a form suggested by Wagner:5 Here AH is the heat absorbed in kilocalories in forming one gram atom of molten alloy from its molten elements and the N's are atom fractions. The relative partial molal enthalpies of the components, each referred to its pureliquid state and defined as zFe = aFe - PFe and zsi = HSi — -psi, are shown graphically. At low concentrations zSi = -28.5 kcal, in agreement with Kijrber and Oelsen's computation. This is in good agreement with the value of -29.3 kcal obtained by Chipman and Grant6 using an entirely different method. ACTIVITY AT INFINITE DILUTION From the known free energy of SiO, it is possible to obtain the activity of silicon in dilute solution in liquid iron from equilibrium studies. The heat of formation of a-quartz is —217.6 kcal and the heat capacity and entropy data are given by Kelley and ~ing.' The free energy of formation of ß-cristo-balite at temperatures above the melting point of silicon is expressed by the following equation: Si(Z) + O2(g) = SiO2 (crist); AF" =-226,500 + 47.50T [I] The value of the deoxidation product for silicon [%Si] x [%O]2 at 1600°C according to Gokcen and chipmans is 2.8 x 10"5, in agreement with results of Hilty and Crafts.9 More recent works of Matoba,

Jan 1, 1963

By Clarence O. Babcock

Model pillars of limestone, marble, sandstone, and granite, with length-to-diameter (LID) ratios of 3, 2, 1, 0.5, and 0.25 (0.286 for granite), were broken in axial compression to determine to what extent an increase in end constraint increased compressive strength. Radial end constraints of 13 to 23% of the average axial stress in the pillar, produced by solid steel rings bonded with epoxy to the ends of dogbone-shaped specimens, increased compressive strength somewhat above that of cylindrical pillars without ring constraint. However, when the results were compared with those obtained by other investigators for straight specimens of several rock types taken collectively, with LID ratios greater than 0.5, the resulting strengths were not significantly different. Thus, the amount of end constraint produced by the solid steel rings was about the same as that produced by the friction from the steel end plates. In other tests, a radial prestress of 3000 or 5000 psi was applied prior to axial loading by adjustable hardened steel rings to increase the constraint above that obtained for the solid rings. The average radial constraint stress, expressed as a percentage of the average axial pillar stress at failure for the 3000 psi prestress, was 54.3% for limestone, 40.3% for marble, 44.7% for sandstone, and 23.4% for granite. The average radial constraint stress, expressed as a percentage of the average axial pillar stress at failure for the 5000 psi prestress, was 74.2% for limestone, 51.2% for marble, 61.6% for sandstone, and 29.7% for granite. These constraints increased the compressive strength significantly above the strength of straight specimens and solid-ring constrained specimens. These results suggest that large horizontal stresses in orebodies mined by the room-and-pillar method should increase the strength of the pillars and allow an increase in ore recovery by a reduction of pillar size when major structural defects are absent. One important objective of the U.S. Bureau of Mines (USBM) mining research program is the rational design of mining systems. In the design of room-and-pillar mining operations, pillar strength is a fundamental variable. It is customary to estimate this strength from uniaxial compression tests of rock samples and to correct this value for the length-to-diameter (LID) ratio of the in-situ pillar. This method of estimating pillar strength corrects for pillar shape but does not consider the effect of a large horizontal in-situ stress field that sometimes exists in underground formations. The purpose of the work covered in this report was to determine to what extent the compressive strengths of model pillars of relatively brittle rock loaded in axial compression were affected by lateral end constraint. In previous work, Obert l used solid steel rings bonded to the ends of dogbone-shaped specimens to study the creep behavior of three quasi-plastic rocks -salt, trona, and potash ore - during a test period of 1000 hr. These rings provided radial constraint during the loading cycle of 20 to 50% of the axial stress for specimens with LID ratios of 2, 0.5, and 0.25. He concluded that (1) "for a quasi-plastic material the end constraint strongly affects the specimen strength, and (2) as D/L increases (length-to-diameter decreases), the specimens lose their brittle characteristics and tend to flow rather than fracture." He also concluded that model pillars constrained by rings were better for use in predicting the strength of mine pillars than either cylindrical or prismatic specimens. This conclusion appears to be valid where mine pillars, roof, and floor are a single structural element. In the present study, 460 specimens of four relatively brittle rocks — limestone, marble, sandstone, and granite - were tested to failure. The study consisted of two parts: (1) the effect on the compressive strength of end constraint produced during the axial loading cycle by solid steel rings bonded with epoxy to the ends of the specimens, and (2) the effect on the compressive strength of increased end constraint produced in part by a prestress applied prior to axial loading and in part by lateral expansion of the specimen during the loading cycle. The first part of this study was reported in some detail earlier.2 EXPERIMENTAL PROCEDURE AND EQUIPMENT Model rock pillars of the sizes and shapes shown in Fig. 1 were broken in axial compression when the ends were constrained as shown in Fig. 2. he straight specimens were broken without ring constraint. The specimens of dogbone shape were broken with (1) solid

Jan 1, 1970

By H. I. Aaronson, H. A. Domian, G. M. Pound

Zener's two-parameter theory of the y a reaction in Fe-X alloys is extended to encornpass austenite-stabilizing as well as fewite-stabilizing elements, and is then cottzbitzed with statistical thermodynamic theories of Fe-C alloys to permit development of a ther-modynamic description of the proeutectoid fewite reaction in Fe-C-X alloys. Equilibrium tielines are calculated for tile a + y rep'on from experimental data and frort extrapolations of the y/y +a equilibrium surface. Sectzovs of the y/y + 0 surface are calculaled for the tzetastable sstuation LYL u-kich alloyrtg eleitzents do not partition betzseen azrstenite andfer-rite. The finding that the metastable y/a + y curves Lie close to their equilibriuwz counterparts when X = Si, .Vo, Co, Al, and Cu but well below them when X = A/ln or Ari pr-elides a thermodunamic explanation for the expcri1?7e?ztally obserted absence of partition during the proeutec toid ferrite reartlon in the fortner alloys and for the occurrence of partition at hzgher ternperatures in the latter alloys. Cotarison ofno- patitzoa free-energy changes and y/y i a curr,es in Fe-C-X alloys with the equilibrium 11al1es of these quantities in Fe-C alloys furnishes additional qualitative insight into the irqluence of allojling elements upon the kinetics of the proeutectoicl ferrite reaction. THE introduction of a substitutional alloying element, X, appreciably complicates calculation of the thermodynamics of the formation of proeutectoid ferrite from austenite. As noted in the preceding paper,' the positional entropy of the interstitial species is the principal component of the free energy of an interstitial solid solution with which theory has so far been able to deal. One would expect, however, that other components of the free energy of this type of solid solution may be significantly altered by the addition of a substitutional alloying element. Even if the assumptions are made, by analogy to the case of Fe-C alloys,' that changes in the positional entropy of carbon represent a significant part of the thermodynamic effects of an alloying element, and that the remaining effects can be taken into account by fitting the equations developed on this basis to experimental data on Fe-C-X phase diagrams or on the activities of carbon in alloyed austenite and ferrite, the experimental information available on either of these quantities is not yet sufficiently accurate or extensive, respectively, to make such an approach useful. An attempt made by Zener on the former basis to explain the effects of alloying elements on the thermodynamics of Fe-C alloys in terms of a temperature-independent "free-energy change" (actually enthalpy hane') required to transfer 1 mole of an alloying element from austenite to ferrite, in which the "free-energy change" was determined by fitting the theoretical relationships developed to Fe-C-X phase diagrams, thus proved inadequate in part because of deficiencies in the available ternary-phase-diagram data.= Other difficulties of a more fundamental nature, however, also indicated the desirability of a different approach to the problem.3y4y6 zener6 subsequently proposed that the free-energy change associated with y — a transformation in pure iron can be decomposed into magnetic and nonmagnetic components. Alloying elements were assumed to affect those components separately. One parameter was used to describe the quantitative effect exerted on each component. These parameters were evaluated from Ms (martensite-start) temperature data and from other experimental information usually either readily available or measurable with acceptable accuracy. Zener applied this treatment only to the calculation of Fe-X phase diagrams in which X is a ferrite-stabilizing element. In the present study, this treatment is extended to include austenite-stabilizing elements, and then combined with treatments previously considered for Fe-C alloys1 to permit calculation of the thermodynamic quantities of interest in the austenite - proeutectoid ferrite transformation in Fe-C-X alloys, where a number of representative, and commonly used alloying elements are chosen for X, including Si, Mn, Co, Mo, Al, Cr, and Cu. The results are used to explain important features of the partition of alloying elements between austenite and proeutectoid ferrite, as reported in a companion paper7 on the basis of electron-probe analysis, and to provide some additional understanding of the influence of alloying elements upon the kinetics of nucleation and growth of proeutectoid ferrite. Portions of this treatment6 have been clarified in the course of a review by Kaufman and Cohen.3 The consolidated summary of these developments with which this section is begun provides a basis for further clarification of the treatment, from which extension to include austenite- as well as ferrite-stabilizing elements is a natural consequence. The division of the free-energy change associated with the 7 - a transformation in pure iron, into two independent components is formally stated as:

Jan 1, 1967