Search Documents

By F. V. Lenel, G. S. Ansell, J. A. Dromsky

The growth of aluminum oxide particles in a nickel matrix was studied eve?. the temperature vange of 2140° to 2470°F. The instability of the dispersed alumina was shown to be independent of the crystal structure of the alumina. The activation energy for the growth of the dispersed alumina was found to be 84.7 1 2.0 kcal. The particle radius increased as a function of time. These results indicate that the growth is not diffusion controlled. It is believed that the rate controlling mechanism is the dissolution of the aluminum and oxygen atoms into the nickel lattice. THE strength properties of alloys which consist of a finely dispersed second phase in a metallic matrix depend upon the spacing between the second phase particles. It is therefore desirable to achieve very fine dispersions in these alloys. Furthermore, to retain the properties these very fine dispersions must be stable during fabrication and service of the alloys. The best known of the dispersion strengthened materials, the SAP type alloys, which consist of a dispersion of aluminurn oxide in aluminum, have exceptionally good stability up to the melting point of aluminum. There is evidence, however, that other dispersion strengthened alloys, even those consisting of refractory oxides in a metal matrix, may be less stable. This investigation is concerned with the stability of Ni-Al2O3 alloys in the temperature range in which these alloys are usually fabricated. The mechanical properties of Ni-Al2O3 alloys at elevated temperatures have been previously investigated by Crelnens and rant,' and Gregory and Goetzel.2 The behavior of these alloys in stress rupture tests appears to indicate that at temperatures below 1800°F they are highly stable. There is some doubt, however, as to their stability at the higher temperatures used during the conventional fabrication. Cremens and Grant, in preparing their test alloys, cousolidated, by powder metallurgical techniques, nickel powders as fine as 0.13 µ diam and alumina powders as fine as 0.018 µ. Metallo- graphic examination of the alloys following fabrication revealed that none had interparticle spacings of less than 2 µ. Considering the size of the original component powder particles, it is likely that the dispersions coarsened during fabrication. Gregory and Goetzel, in their studies of extruded alloys of 80 pct Ni—-20 pct Cr matrixes with nonmetallic dispersion, observed a definite coarsening of the alumina dispersions in alloys sintered at 2280°F as cantrasted to those sintered at 2000oF. Similar observations on the spheroidization and growth of thoria particles finely dispersed in a nickel matrix were made by D. K. Worn and S. F. Marton.3 As a result of such coarsening, much of the effort expended in the preparation of very fine powder mixtures would be lost. The mechanical properties of the alloys which had experienced coarsening would be expected to be poorer than if the original dispersions had been retained. EXPERIMENTAL PROCEDURE Ni-Al2O3 alloys were produced from powder prepared by a coprecipitation technique. Aluminum hydroxide and nickel hydroxide were coprecipitated from chloride solutions of the metals. The mixed hydroxides were calcined to form metal oxides and the nickel oxide in the mixture was selectively reduced to nickel by treating it in hydrogen. Specimens were compacted from the resultant powder which consisted of a fine, uniform mixture of aluminum oxide and nickel particles. The compacts were sintered by resistance hot pressing4, a densification technique which requires exposure times of the order of only a second or less at elevated temperatures. A conventional sintering process was not used, since the temperature required for densification would have to be in the region in which the stability of the dispersion was to be studied. A series of hot pressed specimens were treated in hydrogen at temperatures from 2140o to 2470°F (1171o to 1355oC, for times up to 120 hr. Changes in the microstructures were studied by electron microscopy using the two-stage preshadowed carbon replica method.5 In performing a lineal analysis on a series of micrographs from each specimen it was found more convenient to determine the mean free path between alumina particles rather than particle radii as the parameter of growth. Although these quantities are directly proportional for only spherical particles, the alumina particles in these alloys

Jan 1, 1962

By Leslie W. Austin

ROTARY kiln operators will agree that some of the most severe conditions a refractory must stand occur in the hot zone of a kiln burning Portland cement, dead burn dolomite, magnesite, periclase, and similar materials. Frequently the operator is faced with factors beyond his control which drastically shorten the life of refractories. Shutdown due to mechanical failure can be serious if the period is of sufficiently long duration to cause the dropping of coating or the loosening of the lining. A change in slurry can affect the coating and cause ring buildup. A change in type of fuel and its effect upon the flame can cause a shift in location of the hottest zone. Weekend shutdowns or any other interruption can cause the operator trouble and may damage the refractories, since stopping and starting a rotary kiln is certainly more difficult than stopping and starting a motor. Some operators have tried to set an estimate of damage for each shutdown in equivalent days of running time. Conditions affecting the refractory may be roughly grouped in four classes: chemical attack, mechanical stress, thermal shock, and abrasion. Chemical Attack: The drive to obtain maximum production through a kiln demands maximum operating temperatures, temperatures which are limited more by the ringing up or melting of the clinker. This can cause interface temperatures at the junction of coating and refractory which require the use of a basic kiln block to withstand the chemical attack. Chemical changes take place within the refractory itself, especially in chemically bonded or unburned kiln blocks: These changes cause the formation of the ceramic or burned bond. Migrating liquids or fluxes from the kiln charge have an effect within the refractory and lead to mineral or glass formation. The alkalies, sodium and potassium, migrate into the refractory as silicates, chlorides, sulphates or other salts. They may move under capillary action or may be caused to move by volatilization with condensation in the cooler portion. Mechanical Stress: Concentrated stress may be caused by several factors or combinations thereof. 1-The rings of refractories must be kept tight and rigid within the kiln, and this alone demands considerable force to hold the blocks in place. So that the force will not be concentrated, the blocks should fit the circle as perfectly as possible, with the faces in contact overall. 2-As the kiln is heated, thermal expansion takes place at the hot end of the kiln block. Since this disturbs the plane face it too can cause a concentrated stress at the two ends of the block, and shearing stress can be set up within the brick itself because of the difference in expansion between the two ends. 3-If a lining becomes loose and moves in the shell very severe stress can be set up, and as the kiln rotates this load changes and gives the effect of repeated loading. Permanent expansion of the refractory can also cause severe loading. 4-Not least important, flexing of the kiln is frequently the cause of concentrated stresses. Thermal Shock: Thermal shock, the result of heating and cooling too rapidly, occurs on starting and stopping or when a large patch of coating drops, exposing the bricks. Again, its destructive effect is often the result of phase change, liquid to solid or the reverse; dense refractories loaded with glass-forming impurities are particularly susceptible. Thermal shock is a problem with refractories set in the wall or roof of a stationary furnace, and becomes even more serious in a rotary kiln, the tendency to spall being magnified with movement and concentration of stress. Uniform rate of feed and loading insures both better coating and a more uniform stress. Abrasion: If the refractories do not take a coating, abrasion can become a most destructive factor. Movement of the lining in shell or movement of loose blocks causes abrasion, which is also most destructive if the refractories do not take a coating. An analysis of the problem of basic lining for the hot zone reveals, therefore, a number of desirable characteristics: high refractoriness, basic chemical reaction, resistance to spalling, good strength at all stages, ability to take coating, true sizing, volume stability, and abrasion resistance. Increased demand

Jan 1, 1952

By Ben-Zion Weiss, Melvin R. Meyerson

Chromium diffusion coatings on commercial Armco iron lead to carbide precipitation at the grain boundaries in and below the coatings. High compressive stresses are introduced into the coating and, as a result, tensile stresses are introduced into the base material below the coating. In coated samples, fatigue cracks form at the grain boundaries below the coating after only a limited portion of the total lifetime (5 to 10 pct). Residual tensile stresses and a stress concentration caused by precipitated carbides seems to be responsible for early crack nucleation. Stage I of Propagation may be divided into two substages: I-a in which the crack Propagates along grain boundaries, and I-b in which the crack propagates along slip boundaries according to a shear mode. In uncoated samples, the cracks form at slip bands after 40 to 50 pct of the total lifetime. In coated samples the Propagation process takes longer than in uncoated samples because of the moderate rate of crack extension until the crack breaks through the coating. The chromium-diffusion coating causes little if any increase in fatigue life. CHROMIUM diffusion is one of the most popular processes used to apply coatings to iron alloys. This popularity stems from the fact that the diffusion treatment is comparatively easy to carry out, and it improves certain surface properties such as corrosion and wear resistance as well as some other properties. Very little effort has been devoted to the investigation of the effect of chromium-diffusion coatings on mechanical properties and particularly on fatigue properties. Since the chromium coating usually represents a small fraction of the total volume of the base material it is generally assumed that the macro properties are dependent principally on the base material rather than the coating. Insofar as fatigue is concerned, there are indications that the fatigue life of the chromium coating is not less than that of the basic material and the fatigue properties are generally not reduced by it.' But chromium diffusion causes drastic structural and chemical changes in the region of the material surface2-4 which introduce additional residual stresses. Such changes must affect fatigue crack nucleation and may sometimes affect the initial stages of propagation. Furthermore, it is conceivable that the secondary stages of the crack propagation could be influenced by some irreversible structural changes in the core of the material, which stem from the long-time heat treatment at high temperatures required to apply the coating. This paper analyses the structural and compositional changes produced in commercial Armco iron by a chromium-diffusion coating and the effect of the coating on fatigue crack initiation and propagation. EXPERIMENTAL PROCEDURES The investigations were carried out with commercial Armco iron. The chemical composition of the iron, the number of specimens tested, the grain size produced during chromizing and heat treating, and the thickness and microhardness of the chromized layers are shown in Table I. Samples were chromized by a gas diffusion process. The temperature and time of the diffusion coating process were chosen according to a previous grain growth study. It had been found that at a temperature of 970°C and a holding time of between 3 and 9 hr there are practically no changes in grain size. To describe the grain size more accurately the ASTM method5 was supplemented by a statistical analysis of the mean volume diameter as outlined by Fullman6 and The The mean volume diameter was deter- mined from D = p/2m where D is the mean volume diameter, and m is the mean reciprocal value of grain diameters as seen on the micrograph. The shape and the size of the asymmetrical sample, see Fig. 1, were chosen so as to facilitate the study of fatigue crack initiation. Before coating, the samples were machined and the surface of the groove was polished. After chromizing, the two side surfaces were machined and polished so as to retain the coating only on the top surface of the groove. Residual stress in the chromized layer was measured on the top of the grooved section of the specimen after the diffusion process. The inclined incident X-ray beam procedure was used.9'10 The computations were performed with the initial assumption of a zero surface normal stress component. An electron microprobe analyzer was used to obtain the relative amounts of chromium and iron con-

Jan 1, 1970

By Louis Raymond, John E. Dorn

The object of this investigation was to analyze the recovery that arises when the stress on a specimen undertaking creep is reduced. For this purpose annealed specimens of high-purity aluminum were precrept under a stress of 1000 bsi to a strain of 0.08 following which the stress was reduced for various periods of time to 10, 250, 500, or 700 psi. When the original stress was reapplied the subsequent creep curve lay above that for the unre-covered state and below that for the original annealed state. Analyses on the kinetics of this recovery as a function of the temperature gave a stress-sensitive activation energy that decreased as the reduced stress was increased from a value of 64,000 cal per mole at 10 psi to 37,000 cal per mole at 750 psi. Recovery was also detected and measured during creep under the reduced stress. Following a short initial period, the creep rate under the reduced stress increased monotonically until it reached the secondary-creep rate for the reduced stress. The temperature dependence of this phenomenon was also shown to be correlatable in terms of the previously deduced activation energy for recovery. The activation energies for creep of most pure metals at high temperatures have been shown to agree well with those for self-diffusion.'j2 Since the true secondary stage of creep is usually due to the steady-state balance between the rate of strain hardening and the rate of recovery, it is generally thought that the activation energy for recovery of the creep-induced substructure equals that for creep itself. A shoft time ago, however, Ludemann, Shepard, and Dorn~ found that the activation energy for recovery of the creep-induced substructure in high-purity aluminum under zero stress was almost twice that for self-diffusion, namely about 65,000 cal per mole; obviously recovery under reduced stresses differs in some significant way from the recovery that accompanies the secondary stage of creep. The major purpose of this investigation is to study the effect of stress on the re- covery of the creep-induced substructure in order to provide a better understanding of the recovery mechanism itself. EXPERIMENTAL TECHNIQUE High purity aluminum, containing 0.004 pct Cu, 0.002 pct Fe, and 0.001 pct Si, used in this investigation, was in the form of 0.100-in.-thick sheet which has been cold-rolled to the H-18 temper. Creep specimens were milled from the sheet with their tensile axes in the rolling direction. All specimens were then heated at 686°K for 1 hr followed by air cooling in order to produce an annealed structure which exhibited a uniform equiaxed grain size of about 4 grains per mm. Tests were run in creep machines fitted with Andrade-Chalmers type of lever arms so contoured as to maintain the stress constant to within 0.05 pct of the reported values. Constant temperatures to *O.l°K were obtained by complete immersion of each specimen in a temperature-controlled and agitated bath of molten KN02-KNOs mixture. Where changes in temperature were involved, the change was effected in less than 2 min by manually replacing one bath by another controlled at the second temperature. Displacements over the gage section were sensed by linear differential transformers, the output of which was autographically recorded. The calculated strain measurements were sensitive to 5x EXPERIMENTAL PROCEDURE The following analyses are based on extensions of the previously announced effect of the temperature on the creep strain,2 namely for a = constant, where e = the total true tensile creep strain for a given applied true tensile stress, t = the duration of the test, R = the gas constant, T = the absolute temperature, Q, = the activation energy per mole for creep which is independent of the stress, / = a function of 8, = and of the stress, and a = the stress. The validity of this correlation for high-purity aluminum is demonstrated in Fig. 1 for temperatures in the near vicinity of 600°K; the activation energy for creep, Q,, which is approximately that for self-diffusion, is insensitive to the applied stress

Jan 1, 1964

By Leslie W. Austen

ROTARY kiln operators will agree that some of the most severe conditions a refractory must stand occur in the hot zone of a kiln burning Portland cement, dead burn dolomite, magnesite, peri-clase, and similar materials. Frequently the operator is faced with factors beyond his control which drastically shorten the life of refractories. Shutdown due to mechanical failure can be serious if the period is of sufficiently long duration to cause the dropping of coating or the loosening of the lining. A change in slurry can affect the coating and cause ring buildup. A change in type of fuel and its effect upon the flame can cause a shift in location of the hottest zone. Weekend shutdowns or any other interruption can cause the operator trouble and may damage the refractories, since stopping and starting a rotary kiln is certainly more difficult than stopping and starting a motor. Some operators have tried to set an estimate of damage for each shutdown in equivalent days of running time. Conditions affecting the refractory may be roughly grouped in four classes: chemical attack, mechanical stress, thermal shock, and abrasion. Chemical Attack: The drive to obtain maximum production through a kiln demands maximum operating temperatures, temperatures which are limited more by the ringing up or melting of the clinker. This can cause interface temperatures at the junction of coating and refractory which require the use of a basic kiln block to withstand the chemical attack. Chemical changes take place within the refractory itself, especially in chemically bonded or unburned kiln blocks. These changes cause the formation of the ceramic or burned bond. Migrating liquids or fluxes from the kiln charge have an effect within the refractory and lead to mineral or glass formation. The alkalies, sodium and potassium, migrate into the refractory as silicates, chlorides, sulphates or other salts. They may move under capillary action or may be caused to move by volatilization with condensation in the cooler portion. Mechanical Stress: Concentrated stress may be caused by several factors or combinations thereof. I—The rings of refractories must be kept tight and rigid within the kiln, and this alone demands considerable force to hold the blocks in place. So that the force will not be concentrated, the blocks should fit the circle as perfectly as possible, with the faces in contact overall. 2—As the kiln is heated, thermal expansion takes place at the hot end of the kiln block. Since this disturbs the plane face it too can cause a concentrated stress at the two ends of the block, and shearing stress can be set up within the brick itself because of the difference in expansion between the two ends. 3—If a lining becomes loose and moves in the shell very severe stress can be set up, and as the kiln rotates this load changes and gives the effect of repeated loading. Permanent expansion of the refractory can also cause severe loading. 4—Not least important, flexing of the kiln is frequently the cause of concentrated stresses. Thermal Shock: Thermal shock, the result of heating and cooling too rapidly, occurs on starting and stopping or when a large patch of coating drops, exposing the bricks. Again, its destructive effect is often the result of phase change, liquid to solid or the reverse; dense refractories loaded with glass-forming impurities are particularly susceptible. Thermal shock is a. problem with refractories set in the wall or roof of a stationary furnace, and becomes even more serious in a rotary kiln, the tendency to spa11 being magnified with movement and concentration of stress. Uniform rate of feed and loading insures both better coating and a more uniform stress. Abrasion: If the refractories do not take a coating, abrasion can become a most destructive factor. Movement of the lining in shell or movement of loose blocks causes abrasion, which is also most destructive if the refractories do not take a coating. An analysis of the problem of basic lining for the hot zone reveals, therefore, a number of desirable characteristics: high refractoriness, basic chemical reaction, resistance to spalling, good strength at all stages, ability to take coating, true sizing, volume stability, and abrasion resistance. Increased demand

Jan 1, 1953

By Leslie W. Austen

ROTARY kiln operators will agree that some of the most severe conditions a refractory must stand occur in the hot zone of a kiln burning Portland cement, dead burn dolomite, magnesite, peri-clase, and similar materials. Frequently the operator is faced with factors beyond his control which drastically shorten the life of refractories. Shutdown due to mechanical failure can be serious if the period is of sufficiently long duration to cause the dropping of coating or the loosening of the lining. A change in slurry can affect the coating and cause ring buildup. A change in type of fuel and its effect upon the flame can cause a shift in location of the hottest zone. Weekend shutdowns or any other interruption can cause the operator trouble and may damage the refractories, since stopping and starting a rotary kiln is certainly more difficult than stopping and starting a motor. Some operators have tried to set an estimate of damage for each shutdown in equivalent days of running time. Conditions affecting the refractory may be roughly grouped in four classes: chemical attack, mechanical stress, thermal shock, and abrasion. Chemical Attack: The drive to obtain maximum production through a kiln demands maximum operating temperatures, temperatures which are limited more by the ringing up or melting of the clinker. This can cause interface temperatures at the junction of coating and refractory which require the use of a basic kiln block to withstand the chemical attack. Chemical changes take place within the refractory itself, especially in chemically bonded or unburned kiln blocks. These changes cause the formation of the ceramic or burned bond. Migrating liquids or fluxes from the kiln charge have an effect within the refractory and lead to mineral or glass formation. The alkalies, sodium and potassium, migrate into the refractory as silicates, chlorides, sulphates or other salts. They may move under capillary action or may be caused to move by volatilization with condensation in the cooler portion. Mechanical Stress: Concentrated stress may be caused by several factors or combinations thereof. I—The rings of refractories must be kept tight and rigid within the kiln, and this alone demands considerable force to hold the blocks in place. So that the force will not be concentrated, the blocks should fit the circle as perfectly as possible, with the faces in contact overall. 2—As the kiln is heated, thermal expansion takes place at the hot end of the kiln block. Since this disturbs the plane face it too can cause a concentrated stress at the two ends of the block, and shearing stress can be set up within the brick itself because of the difference in expansion between the two ends. 3—If a lining becomes loose and moves in the shell very severe stress can be set up, and as the kiln rotates this load changes and gives the effect of repeated loading. Permanent expansion of the refractory can also cause severe loading. 4—Not least important, flexing of the kiln is frequently the cause of concentrated stresses. Thermal Shock: Thermal shock, the result of heating and cooling too rapidly, occurs on starting and stopping or when a large patch of coating drops, exposing the bricks. Again, its destructive effect is often the result of phase change, liquid to solid or the reverse; dense refractories loaded with glass-forming impurities are particularly susceptible. Thermal shock is a. problem with refractories set in the wall or roof of a stationary furnace, and becomes even more serious in a rotary kiln, the tendency to spa11 being magnified with movement and concentration of stress. Uniform rate of feed and loading insures both better coating and a more uniform stress. Abrasion: If the refractories do not take a coating, abrasion can become a most destructive factor. Movement of the lining in shell or movement of loose blocks causes abrasion, which is also most destructive if the refractories do not take a coating. An analysis of the problem of basic lining for the hot zone reveals, therefore, a number of desirable characteristics: high refractoriness, basic chemical reaction, resistance to spalling, good strength at all stages, ability to take coating, true sizing, volume stability, and abrasion resistance. Increased demand

Jan 1, 1953

By R. C. Buehl, M. B. Royer

High manganese slags of low phosphorus and iron content are produced by air oxidation of high phosphorus spiegeleisen in a basic-lined converter. Control of phosphorus and iron within specification limits for ferromanganese ore feed is obtained by a unique cyclic operating procedure. Various types of slags, or synthetic manganese ore, can be made. AN affiliated paper' describes the production of high phosphorus spiegeleisen containing 14 to 23 pct Mn and up to 4 pct P in an experimental blast furnace from open-hearth slag or manganiferous iron ore. This phosphorus content is too high for use of the spiegel in normal steel-production operations. Consequently, a preferential separation must be effected whereby the manganese is concentrated in a product that is usable in industrial operations. The preferential separation methods being investigated for this phase of the work are confined to pyrometal-lurgical processes for ready incorporation into steel-plant operations. It is fortunate that manganese is more actively oxidizable than iron or phosphorus, and therein lies a preferential separation method for isolating manganese from these two elements. Silicon is more strongly oxidizable than manganese, and its oxidation precedes, or occurs simultaneously, with manganese. Therefore, manganese and silicon are separable into a high manganese oxide slag phase while the phosphorus and iron remain in the molten metallic state. Complete separation of these elements represents an ideal condition which is not generally attainable in actual operations. This pyro-metallurgical method for separating manganese from phosphorus and iron by preferential oxidation was investigated by blowing 500 lb of metal in a basic-lined vessel to obtain preliminary information on the effectiveness of the procedure. Certain conditions of attainment in the separation process are desirable in any method for removing manganese from high phosphorus spiegeleisen, such as: 1—High recovery of manganese in the oxidation product; 2—the product to be of equal or better quality than ferromanganese-grade ore—48 pct Mn minimum, less than 10 pct SiO2, and Mn-P ratio of 300:1; and 3—attainment of the enumerated objectives in a product amenable to industrial handling, such as a slag of high manganese content that will flow from the processing vessel and of sufficient fluidity for ready separation of metallic granules that may have become mixed with the manganese slag phase. Previous Work Recovery of manganese from low grade ores and industrial byproduct slags by smelting to spiegeleisen, with subsequent oxidation to synthetic ferro-manganese-feed ores, has been a subject for periodic investigation during the past four or five decades in the United States and Germany. Joseph' and associates of the Bureau of Mines North Central Experiment Station, Minneapolis, Minn., smelted manganiferous iron ores of the Cuyuna range, Minn., to spiegeleisen and subsequently tried air and ore oxidation procedures for concentrating the manganese into synthetic ferrograde manganese ore. Joseph's results on the operation of a side-blown basic-lined converter for air oxidation of manganese from molten spiegeleisen were so discouraging that the procedure was abandoned after 11 blowing tests. Much slag was thrown from the converter vessel; phosphorus content was four to five times the maximum allowable for ferro ore; and a manganese-iron ratio of only 2 or 3:1 was obtained in the slag instead of the required 8:1. Oxidation of manganese from spiegeleisen with about 0.5 pct P, by adding iron ore to molten spiegel in an experimental open-hearth furnace, proved moderately successful in Joseph's experiments. The main drawback to the operation was the necessity for over-oreing for effective oxidation of the manganese with consequent too high iron oxide residual in the slag. Furthermore, the phosphorus content was proportional to the iron concentration and therefore much too high, so that several hours of reduction by a carbonaceous material were required to adjust the iron and phosphorus to the desired content of high grade ferro feed. A troublesome feature of the slag product of the spiegel oxidation process was its lack of fluidity if the silica content was less than 10 pct, the concentration considered desirable for the synthetic ore. Approximately twice the desired concentration of silica was required to impart enough fluidity for transfer from the vessel by pouring. The acute shortage of manganese in Germany

Jan 1, 1953

By R. J. Charles

SEARCH for a consistent theory to explain the relationship between energy input and size reduction in a comminution process has accumulated, over the years, an enormous amount of plant and laboratory data. Although some correlation of these data has been possible for purposes of engineering design and for the advancement of research in fracture, there is still great need of a means of predicting behaviour of a solid when it is reduced in size by mechanical forces. The best known hypotheses proposed to describe the energy-size reduction relationships in crushing and grinding stem from a common origin. The present article analyzes problems of comminution in the light of the precepts of this origin. Its object is to reconcile points of difference between these well known hypotheses and to present relationships more widely applicable to comminution studies. Theoretical Considerations: Most existing relationships between energy and size reduction of a brittle solid stem from a single, simple, empirical proposition.&apos; Although this proposition can be demonstrated by observation and experiment, no theoretical derivation is yet possible. Mathematically, the proposition may be stated as follows: dE = -Cdx/xn [1] where dE = infinitesimal energy change, C = a constant, dx = infinitesimal size change, x = object size, and n = a constant. Eq. 1 states that the energy required to make a small change in the size of an object is proportional to the size change and inversely proportional to the object size to some power n. No stipulations are placed on the exponent n in either magnitude or sign. In 1867 Rittinger2 postulated that the energy required for size reduction of a solid would be proportional to the new surface area created during the size reduction. As far as can be determined there is as yet no physical basis for Rittinger&apos;s hypothesis. Rittinger&apos;s hypothesis can be stated mathematically as follows: E, = K(oa-a-0 . [2] Er = energy input per unit volume, K = a constant, <ti = initial specific surface, and o2 = final specific surface. In the size reduction of particles of size x, to particles of another smaller size, x2, Eq. 2 becomes the well known relation: ET = K&apos; {l/x2-l/xx) [3] where K&apos; is a constant. Eq. 3 may be arrived at from the proposition given in Eq. 1 by integrating and by assigning a value of 2 to the exponent n. J dE = J - C dx/x2 E = Kt (1x - 1x) where K&apos; = C. In 1885 Kick3 proposed the theory that equivalent amounts of energy should result in equivalent geometrical changes in the sizes of the pieces of a solid. For example, if one unit of energy reduced a number of equal-sized particles to particles of one half the size, then the same amount of energy applied to the particles resulting from the first test should result again in a size reduction of one half or a final size one quarter the original size. The Kick concept may be expressed as follows: Ek = K" log x1/x2 [41 K" = a constant and E, = energy per unit volume. The expression for Kick&apos;s law may be arrived at by again integrating Eq. 1 and in this case assigning a value of 1 to the exponent n. dE= J - C dx/x Eh = - C In {x/x2) = K" log (x,/x,) where K" = 2.3 C. Application of Kick&apos;s and Rittinger&apos;s laws to comminution has met with varied success. Gross and Zimmerley4 and Piret5 have shown that Rittinger&apos;s equation applies under certain conditions of experimentation. Walker and Shaw6 express the belief that in metal turning and shaping and in grinding of both metals and minerals the production of very fine particles (less than lP) follows Kick&apos;s hypothesis, whereas Rittinger&apos;s concept is valid for the size reduction of coarse particles. For the practical case of crushing and grinding, however, neither of the above hypotheses has received general acceptance. Bond&apos; has lately proposed that since neither Kick&apos;s nor Rittinger&apos;s hypotheses seem correct for plant design work, an energy-size reduction relationship somewhere between the two would be more applicable. The fundamental statement of Bond&apos;s work

Jan 1, 1958

By M. H. Shumate

Rope-type haulage has had many applications in the mining and allied industries. Records have indicated favorable results both from a standpoint of efficiency and investment. The Truax-Traer Coal Co. has used some variations of rope haulage at their mines and preparation plants. They have built a system which solves their particular problems, and have an investment which has indicated an annual savings in operating costs. Rope haulage, as applied to the mining industry, goes back many years for both underground and surface mining, or a combination of both. The Truax-Traer Coal Co. has used gravity retarding hoists as well as several variations of rope haulage, including electrical and combinations of both, but each was limited in its use and application by natural conditions and economies of the operation. Several systems of rope haulage equipment are offered by manufacturers today for handling railroad car movement on a limited or continuous basis. The portal and railroad loading facilities of Truax-Traer Coal Co.&apos;s Burning Star Slope mine, located in Jackson County, Ill., were moved in 1960. The old location was abondoned, eliminating 3% miles of underground track haulage. The mine was converted to an all-belt system, and coal is loaded at a new raw coal plant and shipped by railroad cars to a central cleaning plant. The company wanted to operate the surface facilities as efficiently as possible, employing a minimum number of workers, using the latest type railroad car movers. The existing rope haulage facilities at several locations throughout the country were examined and considered for the Slope mine location. The application of each appeared favorable but lacked flexibility, and it was difficult to justify the capital investment. The company decided to investigate the possibility of building a system that would apply to their problem, and have an investment that could be amortized in a relatively short period. SPECIFICATIONS They employed the services of Allen and Garcia Co. of Chicago, Ill. Through combined efforts, a railroad car rope system was designed to specifications as shown in Fig. 1. The Falk Corp. of Milwaukee, Wis. built the hoist to the specifications as shown in Fig. 2. It has an all welded base and pedestals. The all welded drum is not grooved. A single helical gear was used in lieu of herring bone type. A Falk speed reducer, Unit 90Y3-A, is driven by a 25-hp 1800-rpm, 440-V ac type &apos;C&apos; (high starting torque) drip-proof, Frame 324-U motor. The speed reducer shaft is equipped with a solenoid operated ac spring set shoe brake, operating on a 7-in. diam. brake wheel. The dolly car, shown in Fig. 3, was field constructed, using the trucks and frame of a railroad tank car. Truax-Traer did not alter the frame but added plates and anchors for hoist ropes and frames for 8.5 tons of concrete to be used as ballast on the car. The limit switch operating shoe was also installed and the car coupler latch mechanism overhauled. The hoist rope selected was 1% in. in diam., 6x37 improved plow steel extra flexible, right lay, regular lay, with independent wire rope core, and preformed. Wedge sockets were used to connect rope to the dolly car as shown in Fig. 4. The system employs two 30-in. and three 16-in. diam. sheaves. All are bronze bushed and equipped with grease fittings. Larger sheaves are used for directional change of hoist rope and the smaller ones as snub sheaves to correct fleet angle of rope as it approaches the hoist drum. The Trench lay cable, buried between the two tracks, is used for electrical distribution and controls between hoists and operator&apos;s cab. Foundations for the system required 187 cu yards of reinforced concrete using approximately 60 cu yd each at three locations. Four hoists are employed in the system, two for each track. A pair is interconnected electrically and when one operates, moving the cars, the tail hoist idles with brake released. The process is repeated for the reverse direction. A limit switch, mounted in the track near each end of the system, is tripped by the dolly car as it approaches the hoist. The limit switch controls movement in one direction only, protecting the dolly car and establishing positive con-

Jan 1, 1962

By C. Law McCabe, R. G. Hudson

The Knudsen cell has been employed to determine the free energy of formation of Mn7Cs in the temperature range 800" to 950°C. A value of 66,440 cal was found for hH°o for a-manganese. Measurements of the pressure of manganese over a mixed carbide, (Fe,Mn),C, points to a power relationship between aun7cs and N.4,. RECENTLY Kuo and Perssonl have reported that the carbide of manganese which is in equilibrium with graphite at temperatures up to 1100° C is Mn7Ca. There are no published data on the thermo-dynamic properties of this compound. In order to determine the stability of Mn7Ca, it appeared that, by obtaining the pressure of manganese above 8-manganese and also above Mn,C, in equilibrium with graphite, the free energy of formation of Mn7Ca from 8-manganese and graphite could be obtained. In addition, the vapor pressure of manganese, reported by Kelley From data of Bauer and Brunner,&apos; is subject to some uncertainty and further determinations of the vapor pressure of manganese seemed warranted. In this investigation of the pressure of manganese vapor above pure manganese and also above the carbide of manganese in equilibrium with graphite the apparatus used is the Knudsen orifice cell. The same apparatus, experimental procedure, and method of calculating the pressure was used in this investigation as in one previously reported.~ Care was taken to insure that the cells were at constant weight before using them in a run. The manganese charged in the cell was CP grade powder, carbon free, obtained from the Fisher Scientific Co. A spectroscopic analysis of the manganese after appreciable amounts of it had vaporized from the Knudsen cell showed that no element was present in sufficient quantities to contribute to a weighable weight loss or to decrease the vapor pressure of manganese to any appreciable extent. The spectro-graphic analysis was 0.002 pct Cu, 0.05 pct Fe, 0.002 pct Pb, and 0.002 pct Ni. 8-manganesea is the allo-tropic form of manganese which was present in the cell at temperatures used in this investigation. The manganese carbide, Mn,Ca, was made in the following way: In a closed graphite cell manganese powder was added to graphite powder, which was made from graphite rods for spectrographic use. The manganese powder was the same as that described previously; 5 pct excess graphite was added over that required for the formation of Mn7C,. The mixture was heated in a closed graphite cell for approximately 20 hr at 1350°K under vacuum. X-ray analysis revealed that there was no manganese present after this treatment, but that the lines due to Mn,C, were present. In order to prove that there was no volatile carbide of manganese which was effusing out of the cell, the following experiment was performed: A graphite effusion cell containing graphite power, in excess of that to form Mn,C, of a desired amount, was brought to constant weight on heating at 1228°K. The required amount of manganese was accurately weighed and then added to the graphite effusion cell. The cell was placed in a vacuum at 1228°K for one week, which was the time calculated for the manganese to have effused completely, assuming instantaneous formation of Mn,C8. The cell was then weighed again. This experiment was carried out on two different occasions and both times the weight loss of the cell came within 1 pct of the weight of manganese originally charged minus the weight of manganese left in the cell, as determined by chemical analysis. These data are summarized in Table I. This agreement is considered to be within experimental error and is taken as proof that no carbide of manganese is volatile in this temperature range. It was established, by X-ray analysis, that Mn,C, formed before appreciable amounts of manganese vaporized from the metal powder which was charged. The identification of the carbide of manganese which was present in the Knudsen cell in equilibrium with graphite and manganese vapor was carried out by Kehsin Kuo at the University of Uppsala. He established that the authors&apos; sample, which was submitted to him for analysis, contained the phase

Jan 1, 1958

By J. Szekely, G. P. Martins

A formulation is given for describing the rate of expansion of spherical bubbles rising in liquids the freeboard of which is evacuated. The computer solution of the resultant differential equations has shown that, for low freeboard pressures (less than about 5 mm Hg), on approaching the free surface the bubbles expand much less rapidly than predictable from equilibrium considerations. In other words, in this region the pressure inside the bubbles will be significantly larger than the static pressure in the liquid corresponding to the position of the bubble. These theoretical predictions were confirmed by experimental work, using two-dimensional air bubbles rising in mercury. The important consequence of these findings is that the expansion of gas bubbles in vacuum degassing operations will be a great deal less than expected from hydrostatic considerations. This would lead to a significant reduction in the available interfacial area and may explain the apparent poor efficiency of many vacuum degassing units. VACUUM degassing as a treatment for liquid steel has gained widespread popularity in recent years; the number of known installations exceeds several hundred at the present time.&apos; Although much information is available on both the thermodynamics of the system and the overall performance of various industrial units, much less is known about the fundamental aspects of the process kinetics.2-4 The basic physical situation common to virtually all vacuum degassing operations is the interaction between gas bubbles (swarms of bubbles) and the surrounding molten metal, held in a container, the freeborad of which is at a low absolute pressure. Once formed (or introduced from an external source) the bubbles will ascend, due to the buoyancy forces, and, during this ascent, a significant increase in their volume will occur. This progressive increase in the bubble volume is due to two factors: a) the continuous reduction in the static pressure acting on the bubble during its rise; and b) mass transfer into the bubble from the surrounding molten steel. In a recent paper Richardson and Bradshaw developed equations5 for describing mass transfer into gas bubbles from molten metals at reduced pressures. However, in deriving these expressions it was assumed that the pressure inside the bubble was identical to the static pressure in the adjacent liquid. In other words, the volume of the bubble was considered to be in equilibrium with the pressure of the fluid adjacent to it. This assumption, thus their analysis presented, was thought to be reasonably accurate for most of the bubble&apos;s ascent; however, it was unlikely to be valid in the immediate vicinity of the free surface, held at a low pressure. It was pointed out in the discussion6 that the region close to the surface may be of considerable importance as both the driving force and the interfacial area available for mass transfer are at their highest value here. The &apos; &apos;anomalous" behavior of gas bubbles when approaching a free surface at low pressures was recently confirmed in a preliminary investigation by Szekely and Martins. ?1 Here high-speed motion photography was used to study air bubbles rising in a column of n-tetradecane with a freeboard pressure of 0.1 mm Hg. It was found that significant distortion of the bubbles occurred on approaching the free surface; furthermore, the expansion observed was much less than what one could expect from hydrostatic considerations, i.e., factor a previously discussed. It follows from the foregoing that a detailed study of these phenomena would be justified both from fundamental considerations and because of their potential relevance to technology. The purpose of the paper is to present a more realistic formulation for the expansion of a gas bubble approaching a low-pressure region, together with a comparison of the theoretical prediction with experimental measurement. An inert bubble will be considered in the first instance; it is thought that the understanding of the fluid mechanics is an essential first step toward the realistic formulation of the mass transfer process. This latter problem will be the subject of a subsequent publication. FORMULATION The Physical Model. Consider a spherical bubble, of initial radius Ro, rising in a fluid, having a density pL. Initially let the bubble be at a distance H from the free surface, and at a pressure Pgo, as illustrated in Fig. 1. Pgo, the initial pressure in the bubble, is given by the following equation: pgo = Po = Ptp +pLgH [ la] where Po is the pressure in the liquid corresponding to the initial position of the bubble and Ptp is the pressure at the free surface. The fluid pressure at

Jan 1, 1970

By G. F. Bolling

STUDIES of grain boundary migration in zone-refined metals have all shown that the rate of migration is greatly reduced by small added solute concentrations. However, it is apparent that a difference exists between boundary migration during normal grain growth and single boundaries migrating in a bicrystal to consume a substructure. To effect the same reduction in velocity in the two cases, much more solute is required for grain growth than for the single boundary experiments. One case is available for direct comparison; both Bolling and winegardl and Aust and utter&apos; added silver and gold to zone-refined lead to study grain growth and single boundary migration, respectively. For comparable reductions in migration rates, about 500 times more solute was required to retard grain growth than to retard the single boundaries. A reason for this difference is suggested here. The rate of grain boundary migration is dependent on solute concentration and must therefore also depend on the solute distribution; i.e., regions of higher solute concentration encountered by a moving boundary must produce greater retardation and thus could determine any observed rate. A dislocation substructure can be the source of a nonuniform solute distribution since it can attract an excess concentration of certain solutes. In fact, it is probable that the solutes which impede grain boundary migration most would segregate most severely to a substructure for the same reasons. Thus a dislocation substructure present in a crystal being consumed could locally magnify the concentration of solute confronting an advancing grain boundary. In the single boundary experiments a low-angle substructure, within single crystals obtained by growth from the melt, was used to provide the driving force to move a grain boundary; in grain growth, no substructure of this magnitude was present. The increased solute concentration at subboundaries should be given approximately by C, = G e c,/kT, where t, is a binding energy and CO the bulk concentration. To account for the difference between the two experiments in the Pb-Ag and Pb-Au cases, C, must be the concentration impeding the single boundary migration, and a value of t, = 0.25 ev is necessary. This is reasonable, even though calculation on a purely elastic basis gives t, = 0.12 ev. because electronic effects must enter for silver and gold in lead. The compound AuPbz forms3 and the metastable compound AgrPb has been reported to nucleate at dislocations prior to the formation of the stable, silver-rich phase.4 Other observations support the hypothesis that a magnified solute concentration impedes the single boundary migration. For example, some crystals were grown by Aust and Rutter at concentrations of ~ 0.1 wt pct Sn and 2 x X at. pct Ag or Au which exhibited a cellular substructure, and in these crystals no boundary migration was observed. It is therefore evident that the higher concentrations at cell boundaries drastically inhibited migration. Inclusions would not have been responsible for this inhibition since according to recent work on cellular segregation,5 no second phase should have occurred in the segregated regions at the cell boundaries for the conditions of growth used, at least in the Pb-Sn system. In the purest lead, only the "special" boundaries observed by Aust and Rutter gave rise to the same activation energy as that obtained in grain growth. It is reasonable to suppose that the structure of special boundaries does not favor segregation at low concentrations and thus solute, or an inhomogeneity in its distribution, would have no effect. Random boundaries, on the other hand, are affected by solute and the substructure would enhance residual concentrations in the zone-refined lead, leading to a higher activation energy. It is clear, even without a detailed theory, that the apparent activation energies and exact solute dependence in the two experiments must be different as long as the non-uniform solute distribution produced by the substructure is important. Recrystallization experiments should also be susceptible to the same kind of local segregation at subboundaries or disloca tion cell walls; a suggestion similar to this has been made by Leslie et al.&apos; Following the arguments presented here, the effects of a given solute concentration would be like those observed by Aust and Rutter if segregation occurred, and like those of grain growth otherwise. This work was partially supported by the Air Force Office of Scientific Research; Contract AF-49(638)-1029.

Jan 1, 1962

By C. E. Richards, C. N. Reid

The influence of preferred orientation on the incidence of defbrtnation tuinning has been studied. High-purity iron with almost vandonz grain orientation was cotnpared uitll iron of the sa)ne grain size and composilion lza,ing a strong (110) fiber texture. As expected from published work on single crgslfls, /he ))lean stress for the onset of luitzning-, and the l,olu)nt. fraclion of twinned nzaterial obserlled in lension differed fron the 1-a1ue.s it2 co?nPression for tnolerial with a slrong texlure. The llinning stress of "rctndorrl " )zalerial did not 17ary with the sense of the aPPlied unin.via1 stress, but sirprisinglg the incidence of 1c)i)zning- was about three 1i))zes greater ill conzp?&apos;ession Illon in lension. These results (Ire attributed entirely to ovienbation and may be nderslood in ler?ns of the shear slresses acting on the allowed twinning syster)is. J. HE twins most commonly formed in bcc metals may be described as regions of the crystal in which a particular set of (112) planes is homogeneously sheared by 0.707 in the appropriate ( 111) direction. A similar twin-related crystal could be produced by a shear of 1.414 in the reverse (111) direction but twinning by this large displacement has never been reported. Thus, twinning is unidirectional and a shear stress which produces twinning does not do so when its sense is reversed. The sense of a shear Stress is reversed when the loading is changed from tension to compression, or vice versa. Consequently, for a given orientation of a crystal relative to a uniaxial stress, only a fraction of the twelve (112) twinning systems are geometrically capable of operating in tension, and the remaining systems may operate only in compression. Therefore, when twinning is involved, there are expected to be differences in behavior between crystals tested in uniaxial tension and those tested in compression. This has been verified experimentally by Reid et 01.&apos; and Sherwood el al.,&apos; although a critical stress criterion was not encountered. Furthermore, twinning stresses in colmbium," tungten, tantalum,&apos; irn,&apos; i-Fe,\ nd molybdenum7 single crystals have been shown to depend critically on orientation, although again twinning did not occur at a critical value of the macroscopic shear stress. However, when twinning occurs, it generally does so on the most highly stressed systems, 1--4&apos;6&apos;8&apos;9 implying that the stress level does have some relevance to twin formation. In view of the large orientation dependence of twinning in bee single crystals, it might be expected that such an effect would be present in poly crystalline material which possesses a recrystallisation texture. Indeed, riestner" showed that the twinning stress in tension is very orientation-sensitive it1 <&apos;grain-oriented, silicon-iron;" this material possessed a very strong t c m^ii a nnr x_____k . i-_ii__ ri_______j. _x r»i_._:__i preferred orientation obtained by secondary recrystallisation. Reid et a/.&apos; observed a marked difference in the tensile and compressive yield stresses of polycrys-talline columbium which was rationalised in terms of the effect of a preferred orientation on twinning. No other such illformation is known to the authors. Several investigations of twinning in polycrystalline bcc metals have been reported in which the possible existence of a preferred orientation was not even mentioned. It is the purpose of this paper to show that there is a strong effect of texture on twinning in polycrystalline iron, and to poilt out the difficulty in eliminating preferred orientation in recrystallised metals. 1. EXPERIMENTAL METHOD Material and Specimen Preparation. Low-carbon, high-purity iron was obtained from the National Physical Laboratory in the form of $-in. diam rod which had been cold-swaged from a diam of 1 in. The composition of the material is given in Table I. The as-received bar was cold-swaged directly to 0.185 in. diam from which cylindrical tensile and compression specimens were machined. Specimen geometry is illustrated in Fig. 1. The gage length was 0.30 in. long and 0.10 in. diam; it should be noted that, apart from the extra heads which are necessary for tensile loading, the geometry and dimensions of the two types of specimen are identical. The specimens were heat treated either by sequence A or B outlined in Table 11. The essential difference between these two treatments is that in one case the material was repeatedly cycled through the y- to a-phase change in order to produce grains of almost random orientation ("random" iron)

Jan 1, 1969

By A. E. Jenkins, L. A. Baker, N. A. Warner

Rates oj decarburization of levilated Fe-C droplets conlaining 5.5 to 0 pct C have been measured at 1660°C. Gas mixtures of 1, 10, and 100 pct 0, with helium diluenl were used at velocities of 12.5 and 62.5 cm per sec. Rates were independent of carbon concentration in the mell and in good agreement with the calculated rule of oxygen diffusion through the gas boundary layer. The effects of flow rale and total pressure are as predicled and the rates are approxitnalely 2.5 times those with CO2 as oxidant. The mass-transfer correlation used incorporaled the efject of natural convection as well as forced conrection. Graphile spheres are shown to oxidize at the same rate as Fe-C droplets under the same experimental codlions. It is concluded that, for high carbon concentrations in the melt, the rate of- decarburizalion is controlled wholly by the rate of gaseous diffusion. Rate measurements with pure CO, are reported for low carbon concentrations where CO bubbles nucleate within the droplet. Under these circumstances the decarburi-zation decreased with carbon concentration and it is proposed that carbon diffusion is significant in conlrolling the decnvburization rate. In an earlier paper1 decarburization rate measurements were reported for levitated Fe-C alloys at 1660°C but with CO2 as the oxidant. The decarburization rate was found to be independent of carbon concentration in the melt but slightly affected by total pressure. The authors were unable to explain the slight pressure effect but in all other respects the results were consistent with control by diffusion in the gas boundary layer. Subsequent work has been directed at finding the reason for the slight pressure effect and whether the kinetics with oxygen as oxidant parallel those with CO2. Recently Ito and Sano2 have shown that with water vapor-argon atmospheres the decarburization rate is gaseous diffusion controlled until an oxide film appears on the surface. In this work the melts were contained in crucibles. MASS TRANSFER IN THE GAS PHASE In the earlier analysis1 only forced-convection mass transfer was considered. Subsequent recognition of the existence of some free-convection mass transfer explained the observed small effect of total pressure on the decarburization rate. Steinberger and Treybal3 and Kinard, Manning, and Manning4 have developed correlations involving the linear addition of the contribution of radial diffusion, free and forced convection. Steinberger and Treybal&apos;s correlation was chosen as the most applicable to the present work since it correlated most of the data available in the literature and handled the low Reynolds number region exceptionally well. The correlation for (Gr&apos;Sc) < 108 is where Nu&apos; is the Nusselt number for mass transfer based upon the total surface of a sphere in an infinite medium, G&apos; is the mean Grashof number for mass transfer defined by Eq. [2], Sc is the Schmidt number (µ/pDAB)f, Re is the sphere Reynolds number (dpu,pf/µf), p is the viscosity of the gas (poise), p is the density of the gas (g cm-3), Dab is the binary diffusivity for the system A-B (sq cm sec-&apos;), dp is the sphere diameter (cm), u is the approach velocity of the gas (cm sec-I), and subscript f denotes the property value is computed at the film temperature Tf defined by Tf = +1/2(To + Tr) where To is the specimen temperature and T, is the approach gas temperature (oK). Natural convection occurs when inhomogeneities exist in gas density. These may be caused by concentration gradients, temperature gradients, or both. In the present work the temperature gradient between the sphere and the bulk gas was very large and in some cases, for example the runs with pure oxygen, the concentration gradient was also appreciable. The Grashof number defined by Mathers, Madden, and piret5 was used since it took account of both temperature and concentration gradients: where Gr&apos; is the Grashof number for mass transfer (p2fgd3|-yA-yA|/µ2f), Gr is the Grashof number for heat transfer (p2f gd3p|To - T,]/µ2fTf), Pr is the Prandtl number (cpµ/k)f, g is the acceleration due to gravity (cm sec-&apos;f, a is the concentration densification coefficient (1/p)(ap/ayA)T, yA is the mole fraction of component A at the gas-metal interface, yA is the mole fraction of component A in the bulk gas stream, cp is the heat capacity of the gas per unit mass at constant pressure (cal g-I OK-&apos;), and k is the thermal conductivity of the gas (cal cm-&apos; sec-1 OK-1). Mathers et al. tested this combined Grashof number

Jan 1, 1968

By B. L. Landrum, J. Simmons, J. M. Pinson, P. B. Crawford

Patentiometric model data have been obtained to estimate the effect of vertical fractures on the areas swept after breakthrough in water flooding and miscible displacement programs such as gas cycling where the mobility is near one, The data are presented for the case of the fire-spot pattern in which the cemer well is fractured various lengths and orientations, the data indicate that for 10-acre spacing, fractures extetidirrg over 1300 ft in either directior1 from the fractured well may re.srrlt in reductions in sweep efficiencics from 72 to approximately 34 per cent. However. the area swept after break through may be quite largr and only 10 or 12 per cent 1ess than would be obtained if the reservoir were trot fractured. For the specific case when the volume of fluid injected is equivalent to 100 per cent of the pattern vol-unie, the swent area may vary from 80 to 88 per cent, depending on the lenght of the fracture. The former value is that which occurs when the break through or sweep efficiency was orrly 34 per cent and the latter figrrre of 88 per cent is that which is obtained if the reservoir were unfrac-ttm&apos;d. It is pointed out that although the sweep efficiency may he very low in vertically fractured five-spot patterrz.s, the area swept at low water-oil ratios may be only 5 to 10 per cent less than those achieved if the reservoir were unfractured. INTRODUCTION Since the initiation of commercial reservoir fracturing techniques it has been desirable to determine the effect of fractures on the areas swept after breakthrough. Most water flooding or gas cycling projects are continued for substantial periods after the brcakthrough of the injected fluid. Although the sweep efficiency serves as one criterion for rating various flooding patterns. the area swept after breakthrough for various water-oil ratios or percentage wet gas, if cycling. is of perhaps more importance than the sweep efficiency alone. Sweep efficiency data on the vertically fractured five-spot have been presented3. Previous work on the line-drive pattern has shown the effect of vertical fractures on the area swept after breakthrough for the case in which the distance between injection and producing wells divided by the distance between adjacent input wells was equivalent to 1.5 (see lief. 2). The data indicated that for the line-drive pattern it may be desirable to flood or cycle substantially perpendicular to the fractures in order to achieve the greatest recovery for the smallest volume of fluid injected. For this study the center well of a five-spot is assumed as the fractured well. All fractures were assumed to originate at this well and extend into the reservoir for various distances and orientations. All the fractures are straight and are of large permeability compared to the matrix proper. These data are presented to aid the engineer in estimating fractured five-spot pattern performance. ANALOGY The potentiometric model was used in making this study. The model used was 20 20 in. by approximately 1-in. deep. For certain portions of the study one corner of this model was considered to be an injection well and the opposite corner a production well. To simulate vertical fractures a copper sheet was soldered to the wire well and made to conform to the desired length and orientation. In other studies the same model was used except that the four corners of the model might be considered as the corner wells of a five-spot pattern and a fifth well was placed in the center of the model. The well placed in the center of the model was fractured. The total fracture length is L and the well spacing. d. The complimentary fracture angles will be obvious from Figs. 3 and 4. The data obtained on the potentio-metric model assumes the pay to be uniform and homogeneous, the mobility ratio is one, steady-state conditions exist and gravity effects arc neglected. The permeability of the fractures is very great compared to that of the matrix proper. The po-tentiometric model has been used widely both in water flooding and gas cycling projects, and may be used for miscible displacement; how-ever. it is believed that the poten-tiometric model data are more properly applicable to gas cycling than water flooding because the model as-

By P. J. Dean

This Paper contains a comprehensive survey of the known electron-hole radiative recombination mechanisms in the family of III-V compounds. Because of space limitations, the luminescence properties of each III- V compound are not reviewed separately and exhaustively. Instead, the different known types of recombination processes are discussed in turn and exemplified with reference to the III- V compound in which they were first recognized, or are best understood. Electron-hole recombinations usually occur predominantly at impurities or lattice defects either introduced de1iberately or inadvertently present, but radiative intrinsic interband electron-hole recombinations, which occur in perfect crystals, have been observed. Recombination processes which involve the participation of impurities or lattice defects ("extrinsic" recombinations) considered include transitions in which a) free carriers recombine with carriers trapped at impurities ("free to bound" transitions) , b) electrons bound at donor impurities recombine with holes trapped at acceptor impurities ("donor-acceptor pair" recombinations), C) excitons bound to charged or neutral donor or acceptor impurities recombine radiatively (both "resonance" and "two-electron" "bound exci-ton" transitions have been observed), d) excitons bound to neutral donor or acceptor impurities recombine non-radiatively (an example of an "Auger" recombination), and e) excitons bound to impurities with the same number of valence electrons as the host atom which they replace ("isoelectronic " traps) recombine radiatively. In addition, Auger recombination processes involving one or more free carriers have been observed. These extrinsic processes all involve impurities which are present as point defects. Some apparently well-authenticated examples of the recombination of excitons bound to complex impurity-lattice defect centers including nearest-neighbor donor-acceptor pairs are also discussed. Identificalions of the transitions involved in stimulated emission from the direct gap III-V compounds are briefly reviewed. Although the examples of these recombination mechanisms are selected from the III-IV compounds ia this review, these processes have quite general relevance in semiconducting crystalline solids; irrdeed most of them have also been identified in the 11- VI compounds and elernental semzconductors. THE development of crystal growth and purification techniques in recent years and concurrent advances in the understanding of physical processes in solids has accelerated the development of a wide variety of solid-state electronic devices of proven utility. These de- vices are generally used for switching or amplifying operations in electrical circuits. Most solid-state circuit elements are very photosensitive. This photo-sensitivity is generally undesirable and the single-crystal chip forming the active portion of the solid-state device is mounted in an opaque container. The photosensitivity is made use of in phototransis-tors and photodiodes, which are among the most sensitive detectors of electromagnetic radiation particularly in the near infrared.&apos; In these devices, light is converted into electrical power. The solid-state lamp utilizes the inverse effect, namely the conversion of electrical power into light. There is an increasing tendency to use single-crystal diodes rather than the earlier electroluminescent cells in which the active material is present as a powder embedded in a suitable dielectric.&apos; The radiation is emitted at a rate far in excess of the thermal equilibrium rate for the frequencies and temperatures involved; i.e., luminescence occurs. The development of practically efficient solid-state lamps is at an early stage compared with solid- state circuit elements or even photodetectors. Considerable progress has been made in recent years, however.3 The present review is devoted to a survey of the radiative recombination processes in the semiconducting compound crystalline solids formed from elements in groups I11 and V in the periodic table. These materials exhibit the full range of known recombination processes in solids. In fact many of these processes were discovered in 111-V semiconductors. Nonradiative recombination processes, which control the lutninescence efficiency, are also discussed. Luminescence is efficiently excited in semiconductors through processes which produce large excess concentrations of free electrons and holes in the energy bands of the crystal. Transitions induced by lattice defects or impurities usually predominate in the recombination process. By contrast, luminescence in the conventional fluorescent lamp is excited by optical absorption at the luminescent impurity center itself (the activator) and/or at a second type of impurity center (the sensitizer). This latter type of photoluminescence process, occurring in doped ionic crystals with wide band gaps, is outside the scope of this review.4 I) ENERGY BAND DESCRIPTION OF ELECTRON STATES IN CRYSTALS The energy band description of the energy states available to an electron in a crystal forms the basis of our understanding of the empirical division of crystalline solids into metals, semiconductors, and insulators in accordance with their electrical and optical properties.&apos; Nonmetallic crystals have a finite energy gap between the highest energy band which is

Jan 1, 1969

By E. A. Steigerwald, G. L. Hanna

The influence of work-hardening exponent on the variation of fracture toughness with material thickness was studied for high-strength steel, aluminum, and titanium alloys. The results indicate that, when materials are compared at similar fracture toughness to yield strength ratios, the material with the lower work-hardening exponent undergoes the transition from flat to slant fracture at a larger thickness than material with a high work-hardening exponent. In the thickness range where complete slant fracture is obtained the reverse is true and a lower work-hardening exponent results in a lower fracture toughness. The influence of work-hardening exponent on fracture toughness is, therefore, dependent on the particular fracture mode. In the transition region a low work-hardening exponent is beneficial for fracture toughness while in the 100 pct slant region it is detrimental. THROUGH the use of fracture mechanics analyses, the influence of geometric variables on the crack propagation resistance of structures can be interpreted with reasonable consistency. However, in order to gain a more complete understanding of the fracture process, the influence of material parameters on crack propagation must be defined and coupled to the macroscopic fracture mechanics approach. The work-hardening exponent, which characterizes specific material behavior, may serve as an effective parameter to allow some degree of coupling to be accomplished. In the extension of a crack in a specimen of suitable dimensions the propagation process occurs in a stable manner when the crack extension force is balanced by the resistance to crack extension, which exists in the material at the crack tip. As the applied stress, and therefore the crack extension force, on the specimen increases, the resistance also increases primarily because the effective plastic zone at the crack tip, which is the main energy absorption process, becomes larger. Since the work-hardening rate of a material influences the stress-strain relationship, it will also influence the energy absorption process in the plastic enclave at the crack tip and hence should have an effect on crack propagation. A number of studies have been made correlating the strain-hardening exponent or the strain to tensile instability with the ability of a material to resist fracture. Gensamer1 concluded that a low-strain-hardening exponent would result in a steep strain gradient at the base of a notch. He reasoned that a large work-hardening coefficient would result in high-energy ab- sorption due to the increased area under the stress-strain curve. Larson and Nunes2 experimentally observed in Charpy tests on steels heat-treated to below 200,000 psi yield strength that the energy to failure in the fibrous mode, i.e., above the brittle-to-ductile transition temperature, was logarithmically related to the strain-hardening exponent. In order to avoid the complicating effects present in studying materials which undergo a brittle-to-ductile transition, Ripling evaluated the notch sensitivity of a variety of fcc metals with varying work-hardening exponents.3 The results indicated that the relative notch sensitivity, as determined from tests on specimens with a sharp notch, decreased with increasing values of strain hardening. Although the energy required for ductile or fibrous fracture increases with increasing work hardening, high-strength steels often exhibit improved crack propagation resistance when heat-treated to obtain low values of strain hardening.4,5 An analysis of whether low strain hardening is beneficial or detrimental to crack propagation resistance must depend on the particular fracture criterion involved. At temperatures where the material is relatively ductile and the development of a critical strain is required for fracture, high strain hardening increases the energy required to produce failure. In the transition region and below, however, a critical stress law appears to be valid6 and a low rate of work hardening may produce superior resistance to semibrittle crack propagation. The experimental program is aimed at studying these possibilities and determining the specific influence of strain hardening on the crack propagation resistance of several high-strength materials. MATERIALS AND PROCEDURE The alloys, chosen as representative of various classes of high-strength materials, are summarized in Table I. The heat treatments evaluated along with the smooth tensile properties are presented in Table 11. Pin-loaded sheet tensile specimens were employed to determine the smooth tensile properties. A strain gage extensometer (measuring range 0.200 in.) was used at a strain rate of 0.02 in. per in. per min. The work-hardening exponents were determined from the stress-strain curves generated in the smooth tensile tests and the assumption that the portion of the stress-strain curve beyond the yield point can be described by the power relationship: where a is the true stress, P is the true plastic strain,

Jan 1, 1969

By Robert C. Paddock

INTRODUCTION A way of producing 3,000 tpd from the El Mochito Mine was needed. Of this production, 2,000 tpd must come from the San Juan orebody. The original sub-level stoping method did not give satisfactory results due to ground instability, and the highly irregular ore/waste contacts encountered . The experience gained from the initial system helped guide research into the ground instability problem. Results from this work, combined with knowledge gained about the orebcdy configuration, defined constraints that were previously not fully appreciated. These constraints, and others, combined with objectives, were considered together to develop a new mining method. No single technique was found to be suitable, so a hybrid mining system was developed. A combination of ramping, cut and fill, and vertical crater retreat, with an option to use top heading and benching was developed. To complement the mining system, the type of equipment needed was decided upoun. Also, to support the mining system at this expanded rate of product ion, major modifications of existing infrastructure were required. THE EL MOCHITO MINE The El Mochito Mine, of Rosario Resources Corporation, has been in continuous product ion since 198. The mine began operations in April of that yeas at a rate of 100 tpd. The reserves in 198 were 100,000 tons of silver ore assayed at 1,250 grams per tonne. As of the end of 1979, the El Mochito orebodies have produced over 5.6 million tonnes of ore averaging 516 grams per tonne silver, 6.8 lead, and 7.8% zinc. Present ore reserves are about 7.9 million tonnes, averaging 138 grams per tonne silver, 4.6% lead, and 8.7% zinc, with minor quantities of copper, cadmium and gold. An expansion plan to increase mill production two fold to 2,500 tonnes per day is underway. This expansion will require the mine to produce 3,000 tpd. The mine consists of numerous orebodies, all of which have been mined to a certain extent. Of all the orebodies, the San Juan contains 8% of known reserves. This amounts to about 6.7 million tonnes. The significance of the San Juan orebody to the future life of the El Mochito Mine is obvious. If the required mine production of 3,000 tpd is to be sustained, the San Juan must be the source of the majority of that production. Due to the mineability and overall logistics concerned with the other orebodies, the San Juan must be able to reach and maintain a production rate of 2,000 tpd by 1982. GEOLOGY OF THE SAN JUAN OREBODY The El Mochito Mine is a classic example of a chimney replacement deposit in limestone. Similar deposits axe found in Mexico, at the Naica, Providencia, and Santa Eulia Mines. The El Mochito Mine is located at the south- western end of the Sula Valley on the western edge of the Honduras Depression in the Central Cordillera and Central Highlands of Honduras in a setting of Mesozoic sediments. The orebodies occur in a structural basin developed between NNE trending normal faults and apparently hinged on the south end. Topographically, the Mochito Basin lies between the uplifted Santa Barbara mountain in the west and the Palmer Ridge on the east. The San Juan orebody occurs near the intersection of the NE trending San Juan fault and the ENE trending Porvenir fault. The downward continuation of the orebody is controlled by the westward rake of these NW and N dipping structures. The discovery of the San Juan orebody is attributed to analysis of structural evidence of known ore deposits by in-company geologists. The composition of the San Juan orebody is primarily garnet skarn, with local concentrations of hedenbergite and magnetite. The economically important sulfide mineralization consists of (in decreasing abundance), sphalerite , galena, pyrrhotite , and chalcopyrite. There is some indication that a Cu-Ag mineral such as tetrahedrite may also be present. The skarns were formed by replacement of the original limestone by hydrothermal water migrating upward roughly along the intersection between the Porvenir fault system and the San Juan fault system. Textural evidence suggests that the orebody is a composite of several pulses of hydrothermal activity which would explain, in pat, the great irregularity of the contacts and the large horizontal variation in mineralogy. A general pattern of skarn types can be seen in the orebody, partially accounting for the observed lateral variation in grades. This zonation is very generalized, and one or more zones may be missing in any given locality. The orebody is almost invaxiably surrounded by a 2 cm to 25 cm zone of bustamite skaxn with low values. The border skarn is usually

Jan 1, 1981

By David A. Thomas, David N. French

The lrnrdness of WC crystals has been measured with the Knoop indenter at loads of 100 and 500 g on the (0001) and (1070) planes. The hardness as tneasitred on the basal plane is 2400 kg per sq mm and shows little anisotropy. The hardness on the prism plane, however, shows a marked orientation dependence, with a low value of 1000 kg -per sq mm when the long axis of the Knoop indenter is oriented parallel to the c axis and a high value of 2400 kg per sq mm when the indenter is perpendicular to the c axis. Slip lines (Ire observed surrounding the microhardness indentations and they show slip on (1010) planes, probably in [0001] and (1120) directions. This slip behavior can be explained by the crystal structure of TVC, which is simple hexagonal with a c/a ralio of 0.976. The hardness anisotropy call be explained by [0001]{1010} and (1130) {10l0] slii) and the resolved shear-stress analysis of Daniels and Dunn. HARDNESS anisotropy is a well-known phenomenon and has been reported for many metals, with both cubic and hexagonal structure.1-6 For hexagonal tungsten carbide, WC, a wide range of hardness values is reported in the literature. For example, Schwarzkopf and Kieffer7 give a hardness of 2400 kg per sq mm and report a value of 2500 kg per sq mm by Hinnüber. Foster and coworkerss give the average Knoop microhardness as 1307 kg per sq mm with a maximum value of 2105 kg per sq mm. Although these values and the structure of WC suggest the likelihood of hardness anisotropy, no such measurements have been made. We first detected a large apparent hardness anisotropy in WC crystals about 75 p large, in over-sintered cemented tungsten carbide. Prominent slip lines also occurred around many indentations. This report presents further observations and interpretations of hardness anisotropy and slip in WC crystals obtained from Kennametal, Inc. Both Knoop and diamond pyramid indenters were used on a Wilson microhardness tester with loads of 100 and 500 g. EXPERIMENTAL RESULTS The carbide crystals tended to be triangular plates parallel to the (0001) basal plane of the hexagonal structure. The side faces were parallel to the ( 1010) prism planes. Specimens were mounted approximately parallel to these two types of faces and metallographically polished. Laue back-reflection X-ray patterns were used to orient the specimens, which werethen ground to within ±1 deg of the (0001) and (1010) planes. The Knoop hardness values measured on the basal plane are plotted in Fig. 1. There is only a small anisotropy, with a hardness range of 2240 to 2510 kg per sq mm. The additional points at angles from 52.5 to 67.5 deg confirm the sharp minimum hardness at 60-deg intervals, consistent with the sixfold hexagonal symmetry. The average hardness of all values obtained on the basal plane is 2400 kg per sq mm. While the basal plane shows only slight anisotropy, the (1010) plane exhibits marked hardness anisotropy, from 1000 to 2400 kg per sq mm. Fig. 2 shows the hardness as a function of the angle between the long axis of the indenter and the hexagonal c axis, the [0001] direction. The minimum and maximum occur when the indenter is oriented parallel and perpendicular to the [0001] direction, respectively. The anisotropy of the prism plane is contrary to that reported for hexagonal zinc and hard- However, the basal-plane anisotropy is similar to these two metals.1&apos;2 To check the accuracy and reproducibility of the measurements, a series of fifteen impressions was made at 100-g load in the same orientation in the same area of the specimen surface. The average for all was 2040 kg per sq mm, with a range of 1950 to 2130 kg per sq mm, giving an accuracy of about ± 5 pct. Thus the slight anisotropy on the basal plane is almost within experimental error. Fig. 3 shows slip lines around the Knoop indentations on the basal plane. The slip traces are in directions of the type (1130). The presence of slip steps on the basal plane indicates that the slip direction lies out of the (0001) plane. Because WC has a c/a ratio of 0.976,&apos; the shortest slip vector is [0001], which suggests slip systems of the type [0001] (1010). Fig. 4 shows slip lines around the Knoop intentations on the (1010) plane. These slip lines are inconsistent with [0001] slip but can be

Jan 1, 1965

By P. Duwez, C. B. Jordan

A. J. SHALER*—I should like to congratulate the authors for having carried out such a precise set of experiments. It has been found useful, in sintering experimental compacts in vacuo, to make certain that the residual gas is not one which reacts with the metal. Since traces of oxygen can be kept away only with great difficulty, the technique is often adopted of using a "getter " of powder in the vicinity of the compacts, and, in addition, of permitting a small hydrogen leak to flow into the vacuum chamber. Did the authors use similar devices? This paper brings up a question concerning the definition of the word &apos; sintering.&apos; The authors restrict its use to the adhesion between particles. Kuczynski, in a paper presented at this meeting, applies the word to the growth of areas of contact between particles. I have used it to mean both these phenomena and also the dimensional changes which continue to take place after the first two have run their course. May I suggest that we should come to an agreement on the use of these words ? Fig 1 and 2 show an interesting feature: extrapolation of the curves to zero time does not give a densification parameter of zero. The higher the temperature, the higher is the intercept on that axis. These observations agree with the concept of a practically instantaneous densification taking place while the compact is being brought to heat. Such a change may be brought about by plastic deformation and primary creep. The stress pattern causing this first rapid flow is, to my mind, due to the force of attraction between the surfaces of opposite particles in the regions immediately flanking their common areas of contact. The stress is not temperature-sensitive, but at room temperature plastic deformation only proceeds until the metal in the area of contact can support it elastically. As the metal is heated, the elastic limit falls, and further plastic flow occurs. At the higher temperatures, this is followed by primary creep, and finally by the steady-state rate-reaction which the authors are seeking. If they were to recalculate their densification-parameter values, using, not the initial density of the cold compact, but the density after the compacts have been brought to temperature, the systematic deviations from linearity in Fig 3 and 4 might be eliminated. Such initial densities might be obtained by extrapolating the curves of Fig 1 and 2 to zero time. I am naturally pleased to see that such a very well done series of experiments leads to a heat of activation (for the densification process in hydrogen) that is much higher than that for self-diffusion, in confirmation of the less elaborate results reported by Wulff and myself (Ind. and Eng. Chem., (1948) 40, 838). J. T. KEMP*—I would like to comment on Dr. Shaler&apos;s remarks. There are apparently different interpretations of the word "sintering." It seems to me that an accurate definition of our word is essential in all metallurgy. May I point out, in this connection, that in practical metallurgy the word "sintering" has been applied to a bonding process in the preparation of ores and flue dust for fur-nacing. It would be unfortunate if in the area of powdered metallurgy we should establish a definition that is essentially different in meaning. F. N. RHINES*—I think that I can answer the question by saying that I see no essential difference between the use of the term "sintering" in extractive metallurgy and in powder metallurgy; physically the same things are going on. I admit sintering is used for different end purposes in the two cases. When we resort to the sintering of lead ore mixture we are doing so to obtain a chemically reactive, loose texture of some rigidity. This is only a difference in use. After all, in powder metallurgy we sometimes deliberately produce a very porous material which has just a little strength, just as in the case of sinter cake. P. DUWEZ (authors&apos; reply)—We agree that it would be helpful to have well-established definitions of such terms as "sintering." Since the question has now been raised, the time might be appropriate for its consideration by some suitable committee of one or more of the metallurgical societies. In answer to Dr. Shaler&apos;s first question, no getter nor hydrogen leak was used in our vacuum experiments, except insofar as the guard disks (used to reduce friction between specimens and trays) may have acted as getters. Dr. Shaler&apos;s statement that extrapolation of the curves of Fig 1 and 2 does not lead to zero densification at zero time apparently overlooks the logarithmic

Jan 1, 1950