Search Documents

By Frederic Benard

This paper constitutes a preliminary report on experimental work done to date on the anode-furnace treatment of blister copper containing relatively high percentages of nickel. The investigation has not been completed but results obtained to date are considered of sufficient interest to warrant publication. Blister copper treated by the Ontario Refining Co. originates from the Copper Cliff smelter of The International Nickel Company of Canada, of which the former is a wholly-owned subsidiary. The blister as received contains from 0.50 to 1.50 per cent Ni, and in this respect differs from that treated by other refineries on this continent. During the anode-furnace treatment, only a portion of the nickel content is removed, and of the amount remaining in the anodes a varying percentage is found in the anode slimes. This necessitates a more extensive treatment than ordinary nickel-free material. Sometimes it is necessary to repeat roasting and leaching operations several times before the nickel content is low enough to permit the slimes to proceed to the dore furnace. This, besides increasing treatment charges, causes greater selenium stack losses and involves a longer tie-up of precious-metal values. Therefore this investigation was undertaken with the aim of determining the factors influencing the diversion of nickel into the anode sludge. From experience it was known that by keeping the nickel content of anodes at 0.55 to 0.60 per cent maximum a relatively satisfactory slime resulted. One avenue of attack therefork was developing improved elimination in the anode-furnace treatment so that the anode content mentioned above was not exceeded. Another angle was to improve the solubility, during electrolysis, of the nickel constituent in the anodes, thus decreasing the percentage diverted to slimes. It may be mentioned here that nickel dissolving electrochemically and entering the electolyte presents no difficulty from a treatment standpoint. Both of these approaches were studied and are discussed separately below.

Jan 1, 1944

By R. A. Wagstaff

Since the time of the early Egyptians, the use of copper: has been a boon to the life of most of the civilized world. Its use has been varied; in many connections, the art by which it attained its greatest usefulness has been lost. In this paper an endeavor will be made to take into consideration some of the many changes that have occurred during the last 40 years of copper-smelting operations. During that period the copper blast furnace has been almost forced out of existence by new methods of ore concen-tration—first, gravity concentration, and then flotation. These new methods of concentration introduced many new factors into the metallurgy of copper. The product was wet, it was fine and it was high grade. To overcome all these factors many changes have been necessary throughout the smelting agenda. New equipment has been necessary for transportation, unloading, sampling, bedding and moisturing in all smelting plants, especially the custom smelter' During the early days of "flotation," little was known about filtering or the use of an alkaline circuit; often the concentrate going to the smelter contained 2' to '5 Per cent moisture. It is easy to imagine the difficulty arising with a product of this nature when it had to be shoveled by hand labor. To meet these conditions, mechanical unload- ing and sampling devices were soon developed. In many instances such good progress was made that actual costs for handling this unfavorable product were lower than those formerly incurred on the old product. Today most sampling of copper ores and concentrates is being done by mechanical sampling, pipe or augur sampling, or shovel sampling. Where concentrates are uniform in metal content, a great many of the plants now use pipe or augur sampling. For spotty ores, the old shovel sample with cone and quartering during the final stages is still in' use. Where big tonnages are handled, mechanical sampling of concentrates is being done, with satisfactory results. To curtail slow and tedious cutting and mixing procedure, mechanical mixing boxes, splitters and cutters are used to cut down labor costs. Milling As crude ores become scarcer, the use of milling operations increases. The few crude ores that are still smelted are used as fluxing types. To use these ores finer grinding has been found advantageous. This has made for better bedding and fluxillg conditions, which in turn makes for faster smelting, lower fuel consumption and better metal recovery. The degree of fineness is more or less determined by the nature of the product. Most plants have set a standard of through 3/8-in. mesh screen for roaster -reverberatory smelting and through 3/4-in. mesh screen for converter fluxes. For some refractory Ores, even finer grinding has been found to be advantageous for good reverberatory smelting-

Jan 1, 1944

By C. R. Kuzell

Although furnaces constructed of refractory brick have been operated for many decades, there has always been an unfulfilled desire by the operators for a less arduous and more satisfactory method of patching weak spots and burned-out zones in order to postpone and to effectively extend the termination of a furnace campaign. Not only has this desire frequently arisen when small spots have approached incipient failure while the furnace structure as a whole was still in good condition, but also when larger areas showed signs of needing replacement. The available methods have been arduous, sometimes dangerous, inefficient with respect to the life of new refractory material used, and nearly always have necessitated interruption of the continuity of the furnace operations and therefore reduction in furnace production during the period of time required for the repair work. Most of these older methods have involved the replacement of masonry by refractory shapes in the weakened areas. Many attempts have been made to develop a method of patching the burned-out areas by spraying granular refractory material upon old surfaces. While some success has been attained in attempts to renew the grooved, corroded side walls and fettling, and the edges of hearths, there have been no known successful applications of the spraying method in patching the underside of arches of operating furnaces. The method developed recently at the Clarkdale smelter of the United Verde Branch of Phelps Dodge Corporation, at Clarkdale, Aria., and described in this paper, has been found to be not only convenient and successful with respect to patching vertical or inclined surfaces, but has met with a marked degree of success in patching the parts of furnace arches that are exposed to flame, the successful result being obtained without interruption or interference with the continuity of furnace operations. It may be described as a method of renewing the interior refractory surfaces of operating furnaces by pneumatic spraying of an aqueous suspension of finely divided. particles of refractory material. It has been used

Jan 1, 1944

By Leonard Larson

Before direct or wet smelting of copper concentrate was adopted at the McGill smelter, in November 1932, actual furnace smelting tests had indicated the possibility of smelting between 400 and 500 dry tons of wet concentrate per furnace day. Under calcine smelting practice, solid charge up to 116; tons per furnace day had been smelted. It was realized that in order to smelt wet tonnage in equivalent amount per furnace day, a higher combustion rate and higher temperatures would be necessary within the furnace. The reverberatory furnace considered in the following history of wet-smelting developments is a unit rz7 ft. long and 28 ft. 8 in. wide, inside dimensions (Fig. I). This furnace was in operation through all the years from 1934 to date. During 1933, the first complete year of direct or wet smelting, the plant was on a curtailed copper-production basis and the available concentrate tonnage to be smelted was limited. The average coal-firing rate in the reverberatory for that year was 124 tons per furnace day. This rate was lower than the average for years immediately preceding under calcine smelting practice. With this coal-firing rate, an average of 391 tons of new metal-bearing material (N.M.B.M.) was smelted per furnace day. The following year, 1934, the coal-firing rate was increased to 139 tons and the smelting rate increased to 448 tons N.M.B.M. per fur- nace day. In succeeding years production required that either more tonnage be smelted in the one furnace or an additional furnace be placed in operation. With this situation confronting the smelting department, it was determined to continue the program of increasing the coal-firing rate and also to intensify the temperature in the smelting zone of the furnace. In order to accoinplish this, various improvements were made to aid combustion within the furnace. The burners were enlarged, increased primary air was supplied to the burners, and changes were made in the furnace to enlarge and streamline the outlet, thus increasing the furnace draft. The coal-firing rate was increased yearly until 1937 when the firing rate averaged 210 tons per furnace day. Thereafter the firing rate fluctuated according to character of charge and production requirements. In general, however, with the same character of charge the smelting rate increased in almost direct proportion to the increases in firing rate. Table I shows the average coal-firing rate, smelting rate, and other pertinent data for the direct-smelting years of 1933 to 1939 inclusive, the first fivc months of 1940 and May 1940. These figures show that the water evaporated in the waste-heat boilers was in almost direct proportion to the coal-firing rate. In making comparisons in the table, it should be borne in mind that the figures, except for May 1940, are yearly averages. Because of changes in character of charge or operating conditions, records for individual days or even months

Jan 1, 1944

It is claimed that the first mining of copper by Americans in Arizona was done at Ajo, near the Mexican border, in 1854,* a year after this region had been added to the United States, under the terms of the Gadsden purchase. A group of adventurers, organized at Los Angeles under the name of the Arizona Mining & Trading Company, came, to the district. This party of prospectors consisted of 20 men, led by Major B. Allen, J. D. Wilson, and William Blanding. † They went first to Yuma, where the party divided, some of them going southward into Mexico while the remainder took the trail to Tinaja Alta, where they heard of the copper deposits of Ajo, which is 85 miles southeast from Yuma.‡ They found attractive outcrops of copper ore, and set to work energetically, but their operations suffered many interruptions, because that part of the country was still controlled by the Mexicans, who were uncertain about the exact position of the new international boundary. The Mexicans tried to expel them, but they held their ground. The first shipment of ore, consisting of native copper and cuprite, came from what is now the western end of of the New Cornelia workings; it was hauled in ox-carts to San Diego, 400 miles across the desert. § Later shipments were hauled to Yuma and floated on barges down the Colorado river to the Gulf of California, whence the ore was loaded on

Jan 1, 1932

By L. L. McDaniel

During the past few years a nuinber of improvements in smelting and power equipment have been made at the Douglas smelter of the Phelps Dodge Corporation at Douglas, Ariz. In the summer of 1935 work was started on the first unit of a new reverberatory waste-heat boiler plant which ultimately was to furnish steam for a modern high-pressure power pIant. A second reverberatory waste-heat boiler unit

Jan 1, 1944

By Frederic Benard

The treatment of Balbach-Thum slimes at Copper Cliff by the Ontario Refining Co, is of interest because it differs considerably from methods usually employed for the recovery of fine gold from parting-plant slimes. Owing to the comparatively high percentages of palladium and platinum, wet methods are used to remover the greater proportion of these metals before electrolytic refining in the Wohlwill cells. The Plant The platinum-metals department, Wohlwill room, and gold-melting room extend along the north side of the silver refinery and are separated from the parting plant by a glass partition (Figs. 1 and 2). The platinum-metals department resembles a laboratory built on a large scale, with the various pieces of equipment laid out on terraces to facilitate handling. Because of the corrosive nature of the solutions, chemical stoneware is used throughout. Elevation of acids is accomplished by means of Mariotte bottles. On the top level are three steam-jacketed stoneware kettles of 50-gal. capacity, also a packed tower. One kettle is used for digesting the slimes while the remaining kettles are used for precipitating the gold. The dissolver is equippcd with a treated concrete hood, which is connected to the flue system. Gases are drawn off through the packed tower by means of a 6-in. stoneware exhaust fan. A 15 per cent solution of sodium carbonate is continually circulated through the tower. On the intermediate level are two suition filters, one for removing silver chloride and the other for the gold sand. The filter medium consists of an asbestos pad covered with a filter paper on which is a layer of twill. On the bottom level are two rectangular cementation tanks, 24 by 48 by 24 in. deep, and a suction filter used for removing the platinum-metal concentrates. The circulating pump for the packed tower is also here. Suction is provided by a vacuum pump having a capacity of 8 cu. ft. at 20-in. vacuum.

Jan 1, 1944

By Frederic Benard

PRIoR to 1936, the Ontario Refining Co. received all incoming blister copper from The International Nickel Company's smelter in the usual form of 460-lb. cakes, or slabs. These were received in open cars in lots of 250 pieces. At the refinery the cakes were unloaded onto narrow- • gauge cars, for convenient handling and pickup by charging cranes. In March 1936, an investigation was started on the feasibility of transferring converter copper in the molten state from the smelter to the refinery anode furnaces, a distance of 1 1/4 miles. Although common in steel practice, the transportation of blister copper in this form had not previoisly been attempted and a preliminary study of the various factors involved was therefore necessary. For this purpose, a series of experiments was made in which molten copper, removed at different stages during the furnace cycle, was held in a small portable holding furnace. This vessel was essentially a refractory-lined steel cylinder provided with ports for filling and emptying, heated by an oil burner. Dimensions of the steel shell were 3 ft. 10 in. diameter by 7 ft. 3 in. long, and thickness of refractories depended on the design and combination used for the particular experiment. The vessel was filled with approximately 2 1/2 tons of copper at a predetermined temperature, the lining having been preheated to a definite point by means of an oil burner. All openings were well luted and temperature readings were made every 15 min. by means of an immersion-type thermocouple. In this manner the rate of heat loss resulting with different refractory design could be developed. The ratio of exterior surface to weight of metal was taken into consideration in comparing the experimental results with what might be expected for a full-sized unit, and observations were also made on type and degree of slag accretion. It was not possible to get as much information on this latter point as desirable, but all evidence showed that the problem would be control of slag build-up rather than attack on refractories. From the information derived, and in consultation with the engineering staff of the supplying company, an order was placed for the first

Jan 1, 1944

By M. G., Corson

1. In addition to the alloys of copper with iron previously found by Hanson and Ford to show an increase in the concentration of the alpha range with increase in temperature the following binary and ternary systems have been found by the author to exhibit: the same feature: (a) Copper-chromium. (b) Copper-cobalt. (c) Copper-silicon. (d) Copper-nickel-silicon and its pseudo-binary section Cu-Ni2Si (formula of compound definitely established). (e) Copper-cobalt-silicon and its pseudo-binary section Cu-Co2Si (formula of compound established with some probability but not as definitely as for Ni2Si). (f) Copper-chromium-silicon; formation of a compound definitely established (compare Figs. 1 and 2). The formula of this compound may be either Cr2Si or Cr3Si. (g) Copper-iron-silicon; presence of a phase containing both iron and silicon established in an indirect way only. No definite microconstituents observed. 2. RANGES OF SOLID SOLUBILITY (percentage at various tempera¬tures). (a) For copper-chromium: 0.8 at 1000° C.; 0.5 at 900° C.; 0.25 at 800° C.; 0.05 to 0.1 at 500° C. (b) For copper-cobalt: 3.4 at 1000'C.; 2.4 at 900'C.; 1.7 at 800'C.; 0.9 at 600°C.; 0.35 at room temperature. (c) For copper-silicon (Fig. 3) : 6.8 ± 0.1. per cent. at 800° C.; 3 per cent. at 400° C. and probably below 2 per cent. at room temperature if sufficiently long annealed. (d) For copper-cobalt silicide: 3.3 at 1000° C.; 2.4 at 900° C.; 1.6 at 800° C.; 0.7 at 600° C.; 0.3 at 300° C. (e) For copper-nickel silicide (Fig. 4): 8.2 at 1000° C.; 5.5 at 900° C.; 3.6 at 800° C.; 1.6 at 600° C.; 0.7 at 300° C.

Jan 1, 1927

By W. H. Bassett

THE modern smelting and refining of copper is distinctly an American development. The present demand for sound and perfect castings for rolling is due to the development of American industry. Prac-tically every article made from sheet metal or, for that matter, from any form of metal is made by mass produc-tion methods. Faulty material disturbs the orderly sequence of operations and in addition requires expen-sive inspection and unbearable rejection of finished product. This demand, therefore, is positive and will have to be met. Some mines produce ore that can be smelted and re-fined without the electrolytic process into satisfactory copper for the wrought metal industry, and some of these ores carry small amounts of impurities that are of use in hardening and toughening the copper. Most of the copper ore mined contains impurities that must be removed and, fortunately, precious metals which are separated at the same time make this re-moval profitable. The modern electrolytic refineries have their methods so well under control that the product meets all re-quirements without question if properly fire refined and cast. The "if" in this case is important, because there is evidence of carelessness or want of control at times in the material from some of the producers. It is nec-essary that the melted cathodes should be saturated with oxide to dispose of sulfur and that the copper should then be poled to the proper pitch or set. Sometimes statements leak out to the effect that it is unnecessary to oxidize completely the molten cathodes and a little less rabbling saves poles, which are expensive. The re-sult is that the rolling mill gets porous castings and produces seamy or slivered sheets or strips. The condition of the molds, their temperature, and the temperature of the metal are all important to good copper cakes, wedges, or wire bars. Cracked, and burned molds leave rough places and porous spots. Molds which are out of shape, or cracked, make shrink-age cracks which gap open when the castings are hot-rolled. Dampness means porosity. Drops of oil mean porous or honey-combed areas. Poor or irregular dress-ing means mold wash in the castings. Cold copper or slow pouring means cold sets.

Jan 4, 1928

By R. I. Jaffee, J. G. Dunleavy, W. Hodge, H. R. Ogden

Early in 1946, at the instigation of the U. S. Army Signal Corps, the authors made an extensive survey of the available literature covering high-strength, high-conductivity alloys. For the purposes of the Signal Corps, the minimum tensile strength to be considered was 150,000 psi, together with at least 50 pct IACS conductivity. No single alloy was found to be in production, or mentioned in the literature, which had this combination of properties. The constitution diagram for the iron-copper alloys is not well established in the respect to the two-liquid region. Fig I, taken from the 1939 Ed. of the ASM Metals Handbook, shows a two-liquid system existing above the liquidus temperature, but disconnected from it. Although Ruer and Fickl and Ruer and Goerensz definitely established the fact that two liquid layers could occur in the system, Ruer3 was not convinced that the line of immiscibility was closed in the center. Maddocks and Claussen4 found no liquid immiscibility at temperatures up to 157oOC in alloys containing almost no carbon. Although the diagram would indicate age hardening of the copper-iron alloys is possible, Hanson and Fords found that little hardening effect could be obtained. However, Gordon and Cohen6 reported that the response is better if the quenching is very efficient. Although iron has a potent effect on the conductivity of copper, Hanson and Fords have shown that, when hot-rolled copper-iron alloys are aged at 1200°F, the electrical conductivity decreases sharply with iron contents up to 0.2 pct Fe, and then levels off at conductivities between 60 to 70 pct IACS with compositions to 2.1 pct Fe. They also found that aging did not increase the tensile strength, although increasing iron content gave a moderate improvement in strength. High-strength copper-iron alloys, usually containing about 60 pct iron, had been reported by Smith and Palmer.7 They found that an alloy containing 56.4 pct copper, 0.21 pct magnesium, balance iron, would give a maximum tensile strength of 170,000 psi and an electrical conductivity of slightly over 35 pct IACS. The alloy could also be processed to have I50,000 psi tensile strength, and about 41 pct I.\CS conductivity. Decreasing the iron content increased the electrical conductivity, but the long mushy range during freezing of the alloy containing 10 to 20 pct iron made the production of good castings difficult. In spite of the difficulty encountered by Smith and Palmer with alloys of around 20 pct iron, it was decided to make a rather complete investigation of this type of alloy. As a result of this investigation, a method of alloying and fabricating copper alloys containing 10 to 15 pct iron was developed

Jan 1, 1949

By R. I. Jaffee, J. G. Dunleavy, W. Hodge, H. R. Ogden

Jan 1, 1949

By J. W. Craig, H. L. Walker

Industry has for many years recognized the dependence of certain mechanical and physical properties. as well as workability, upon grain size variations in brass. Although the dependence of properties upon grain size has been recognized there are not many instances where a planned program of testing has been conducted and where the only intentional variable has been grain size. In particular the literature is limited with respect to the effect of grain size on the endurance limit, though there are many industrial applications where the brasses are subjected to cycles of repetitive stresses. Literature A number of the physical and mechanical properties of a given brass sheet, in the annealed condition, will depend upon: (I) The ready-to-finish grain size before final rolling, (2) the percentage deformation by cold working, and (3) the time and temperature of annealing. The first and second variables affect the recrystallized grain size, and the third variable affects the coalesced grain size. Gibbsl has shown the effect of two different ready-to-finish grain sizes upon the tensile strength and hardness of deep drawing brass strip after a definite amount of cold rolling. The smaller the ready-to-finish grain size, the higher the tensile strength and hardness, but with lowered ductility, Gibbs also demonstrated, for a given time and temperature of anneal, that the tensile strength, hardness, and elastic limit are higher for the smaller ready-to-finish grain size, but the elongation is lowered. Townsend and Greena112 have reported data which show that cold working increases the tensile strength, hardness, and load carrying ability of deep drawing brass sheet under reversed stress. Greenall and Gohn3 have reported data which show that the smaller grain sizes have higher tensile strength and hardness properties, and the data also show that increased cold working increases the tensile strength and hardness properties of deep drawing brass sheet. Kommers4 has reported data for tests of brass of known ready-to-finish grain size, a specific amount of cold reduction, and annealed at different temperatures above and below the recrystallization temperature. Kommers' results indicate that annealing below the recrystallization temperature does not change the tensile strength and hardness properties, but the endurance limit may be increased from 22,000 to 28,000 psi. The increased endurance limit is explained as resulting from the reduction of internal stresses by a stress-relief anneal. Burghoff5 and associates have reported data on creep tests, over

Jan 1, 1949

By J. B. Newkirk, A. H. Geisler

Certain of the permanent magnet alloys provide ideal systems for the study of the kinetics of the precipitation reaction and the correlation of structure with properties. One such system, Cu-Ni-Fe, was found by Bradley1,2 to exhibit a coherent transition state in the precipitation process analogous to that reported for Al-Cu alloys somewhat earlier.3 The attractiveness of some perrnanent magnet alloys for study lies in the fact that vertical sections of the ternary phase diagram in certain regions of composition (Fig I) have as their prototype the binary Ni-Au diagram. Alloys of this type decompose into products that have the same crystal lattice type but only slightly different lattice parameters. The advantages that such alloy systems oA&apos;er for study over the usual in which an intermetallic compound is formed are many: 1. Since the precipitate has the same crystal strutture as the matrix, complex atomic movements are not required to form the new lattice. 2. Similarly, complex orientation relationships are not involved for both the matrix and the precipitate would be ex. petted to have the same orientation. 3, Small &,registry of the decomposition products at equilibrium (in contrast with Cu-Ag alloys) is conducive to extensive coherent growth in the transient state and thus the transition lattice can be detected by the usual X ray diffraction methods. 4. Finally, the relative quantities of precipitate and depleted matrix can be varied from 0 to 100 pet* thus permitting wide freedom for the study of the effect of cornposition on coherent growth and properties. In the Cu-Ni-Fe alloys of appropriate composition, the face-centered cubic precipitate and also the depleted matrix when first formed are coherent with the parent matrix.1&apos;2 The two have the same <Jo parameter as the original matrix but they are both tetragonal; the precipitate has an axial ratio c/a < I while that of the depleted matrix is c/a > I. When coherency is lost they assume the normal face-centered cubic structure with the depleted matrix having a lattice parameter greater than the original matrix and that of the precipitate less. Such a mechanism would also be expected for CU-Ni-CO alloys because of the similarity in constitution but this had not been demonstrated. The Present investigation was conducted on a CU-Ni-CO alloy. The constitution diagram and magnetic properties of these alloys have been fairly well established, however, no previous determinations of mechanism of precipitation and no correlation of structure with properties had been made. Thus, an alloy of this system Was chosen for a comprehensive investiga-

Jan 1, 1949

By R. P. Carreker, J. G. Leschen, J. H. Hollomon

The external appearance of a crystal which has undergone plastic flow suggests that adjacent blocks of the crystal have glided bodily past one another along the slip planes. However, the great discrepancy between the shear stresses calculated to effect such a movement and those actually observed has for many years stimulated searches for more satisfactory interpretations of the slip process. Appreciable success in these efforts has been realized by assuming the presence in the crystal of various sorts of imperfections and by considering the consequences of the presence of these flaws when stress is externally applied. The theory that has received the most attention recently—that of dislocations—postulates the occurrence of small, mobile imperfections or dislocations which move through the crystal along the slip direction, each dislocation being attended by a local shear displacement of one interatomic distance. Thus only that portion of the crystal through which the dislocation has passed may be said to have slipped. Since each dislocation produces a displacement of only one interatomic distance, an avalanche or train of many dislocations would be necessary to account for the observed displacements of several hundred or thousand interatomic distances. The salient point of the theory of dislocations is the concept that a slip band does not suddenly appear in its entirety. On the contrary, it grows progressively across the slip plane as the individual dislocations successively appear and advance in a train through the crystal. However, the concept of growth is not intrinsic only to the theory of dislocations, since atomistic mechanisms other than that postulated by the dislocation theory can quite probably also describe the growth of a slip band. For instance, a slip band might resemble a disc-like interface that appeared and then grew continuously from a small size, producing a displacement that increased with the size of the disc. Whatever the precise mechanism, the concept of growth is in itself consistent both with the observed low values of critical shear stress and with the experimental observation that slip bands frequently do not traverse the entire crystal. The likelihood that slip bands grow implies that they must first be nucleated and that the initiation of slip can therefore be analyzed in terms of the general concepts of nucleation. Indeed, this suggestion was advanced1 several years ago, but the matter does not appear to have been further pursued. Recent advances in the theory of nucleation 2,3,4 have encouraged such a treatment of the slip process. As will presently be demonstrated, this approach to the problem not only accounts successfully for the incubation periods that have been observed 5,6 to precede slip but also predicts certain transient effects in the deformation process. The existence of these transients apparently has not hitherto

Jan 1, 1949

By W. R. Hibbard

The increased strength of a polycrystal-line metallic aggregate compared with that of its individual crystals generally has been associated with complex stress distributions at the grain boundaries resulting from atomic irregularity between grains. This disregistry, together with the necessity for adjacent crystals to deform compatibly regardless of the single crystal crystallo-graphic strain mechanism, complicates this mechanism so as to increase the metal&apos;s resistance to deformation. Although Taylor1 has shown from purely mechanical considerations the requirement for at least five active slip systems to absorb the complexities during the homogeneous deformation of polycrystalline face-centered cubic metals, experimental studies by Barrett and Levenson2 indicate that this is feasible only in a qualitative sense because of deviations from homogeneous deformation. Seitz and Read in their review3 state: " Unfortunately . . . present knowledge of the grain boundary influence is not sufficient to complement the work in single crystals and provide us with a good foundation for treating polycrystals." The purpose of this investigation was to attempt to clarify the gap between single crystals and crystalline aggregates by an analysis of the effects of dcformational stresses. Some of the results were described by C. H. Mathewson in his 1943 Campbell Memorial Lecture.zl Previous Investigations Since the performance of single crystals deformed under pure stresses has been relatively thoroughly investigated, the most significant researches are concerned primarily with deviations from single crystal behavior of large grained specimens. Qualitative grain boundary effects involving the observed amount of slip werc reported by Carpenter and Elam5 in 1921. Individual crystals in a two or three grained aluminum specimen deformed by tension acted in the normal manner described by Taylor and Elarn4 except for marked modifications at the grain boundaries. There the power to withstand deformation was considerably increased, leaving a ridge on the specimen where decreased slip occurred, often noticeable yi in. from the boundary. Adcock,6 Yamaguchi,&apos; Seumel, Hirstg and others have also reported this phenomenon as a qualitative trend. Yet attempts by O&apos;Neill&apos;o and Ljunggrenll to measure a difference in hardness at grain boundaries in undeformed specimens were unsuccessful. Miller12 measured quantitatively the reduction in slip near the boundary between the single crystal and polycrystalline areas of zinc specimens. These effects consisted of two parts: I. Next to the polycrystalline interface thc distribution of stresses was so altered

Jan 1, 1949

By J. E. Burke

Recent investigations have elucidated many of the phenomena of the grain growth process, but have also revealed some conflicting and unexplained results. Beck and his co-workers 1,2,3 have shown that grain growth continues with continued heating at one temperature until the average grain size approaches the thickness of the specimen. Plotting the logarithm of the average grain diameter, D, against the logarithm of the annealing time, t, they find the straight line relationship given by the equation: D = Ktm [,] with possible small deviations at short annealing times. K and n are constants at constant temperature. On the other hand, Wa1ker4 finds that, in alpha brass) the rate of growth falls off more rapidly than indicated by Eq i after long annealing, so that an equilibrium grain size is approached. Burke5 showed, using Walker&apos;s data, that time and temperature for grain growth are apparently related through an activation energy, H, according to the relationship: ] where h and tz are the annealing times necessary for the average grain size to increase from Do to D at temperatures T1 and T2 and R is the gas laws constant. In the case quoted, H had the value 60,000 cal per mol. The apparent validity of this relationship, at least up to 700°C, and the approximate value of H were later confirmed by Beck3 The results obtained by Beck Kremer, Demer and Holzworth2 for aluminum and aluminum-magnesium alloys could not be satisfied by such a relationship, since the slope of the isothermal curves increased with increasing temperature. These data indicate and Beck, Holzworth and HuV ater demonstrated expcrimentally that the only apparent factor controlling the rate of grain growth at constant temperature is the grain size. While many of the phenomena have been elucidated, a reasonably quantitative explanation of the process is lacking. There exists uncertainty even as to the primary driving force for grain growth, although there is a considerable body of indirect evidence that the interfacial energy of the grain boundary is responsible.7,8,io Making implicit use of this assumption, several authors14,16 have pointed out that grain boundaries should migrate toward their center of curvature. Harker and Parkerg have advanced this concept furthest, and show that one can expect grain boundary migration to occur only when the grain faces are curved, or when they meet at non-equilibrium angles. When the face is curved toward a grain A, for example, the atoms in the surface of grain B, on the other side of the boundary7 are on the average more surromded by the atoms of their lattice than the atoms on the surface of grain A. Because of thermal motion, the atoms at the interface will sometimes be attached to grain A and

Jan 1, 1949

By J. N. Hobstetter

It is widely believed that the nucleation of phase transformations in solid metals is accomplished by some type of local atomic fluctuations in the parent phase which arise from spoutaneous difiusion in the metal or alloy system. Under appropriate conditions, these fluctuations are believed to be stable in the sense that the free energy of the entire system is lowered if the fluctuations, or nuclei, grow. Under other conditions they are thought to be unstable in the sense that the free energy is lowered if they redissolve. The criterion for stability is usually taken as the size of the fluctuation. This view has been rationalized by Beckerl in terms of the free energy released by the volume of the nucleus and the free energy associated with the formation of a new surface surrounding the nucleus. The qualitative result of this reasoning is shown in Fig I where the changes in free energy for the surface and volume effects are plotted together with their sum against the radius of the nucleus. There is evidently always a critical size r, beyond which the release of free energy which is proportional to the volume always dominates the increase of free energy which is proportional to the surface area. Borelius2 has also attempted to rationalize the stability criterion for binary alloys, but in terms of the internal concentration of the fluctuations. He has found that, if surface energies are entirely ignored, the fluctuation must still reach a critical concentration if it is to become stable and avoid subsequent dilution. Although the Borelius theory fails to predict many observed nucleation phenomena such as the vanishing of the nucleation rate at the critical temperature, the major premise must certainly be valid. In this paper a reconciliation of the Becker and Borelius views is made by means of a search for nuclei which are at once stable with respect to both size and internal concentration. The increase in free energy per mole accompanying the formation of a nucleus which is just stable, AF., is independent of the path whereby the nucleus is brought into being. It is thus appropriate to seek critical nuclear attributes without reference to the fluctuation mechanism. The rate of nucleation, however, cannot be treated without a detailed understanding of the exact fluctuation process which produces stable nuclei. In this and succeeding papers the purpose will be to find the size, shape and internal concentration of the stable nucleus which is formed with a minimum increase in free energy in a wide variety of systems. The Nature of the Nucleus It is necessary ti have rather detailed knowledge of the nature of a nucleus if AF, is to be computed. Fortunately, enough facts are known or suspected about the nucleus to make possible a detailed analysis. X ray diffraction studies have shown

Jan 1, 1949

By W. P. Saunders, W. R. Hibbard, G. H. Eichelman

Interest in the various products of the austenite eutectoid transformation in iron-carbon alloys, particularly as produced by the isothermal sub-critical techniques introduced by Davenport and Bain,&apos; has resulted in the application of similar heat treatments to other eutectoid transformations. Studies o f the beta copper-aluminum alloy eutectoid transformation revealed microstructures very similar to those found in iron-carbon alloys. Investigations5t6 of the kappa phase in the copper-silicon alloy system have established the eutectoid type transformation of this phase into terminal alpha and intermediate gamma phases. This transformation has been reported5 as forming a eutectoid type structure in a very sluggish manner by a long time isothermal treatment. The purpose of the present investigation is to study the products of the kappa eutectoid transformation as affected by transformation temperatures by means of microscopic and hardness methods. Preparation of Alloys The copper-silicon alloy equilibrium diagram is shown in Fig 1. The kappa field converges to a minimum silicon solubility at the eutectoid composition and all other completely kappa alloys are hypereutec-toid. Thus, kappa decomposition involves first the proeutectoid precipitation of gamma, followed by the eutectoid transformation of a lower silicon content kappa (5.2 pct) Smith6 has indicated that alloys containing between 5.4 and 5.7 pct silicon could be hot rolled, but that the latter composition was desirable in order to permit a wider kappa temperature range. The alloy was prepared by adding Baker&apos;s commercially pure silicon to an ingot of copper-silicon alloy containing 2.34 pct silicon remaining from the investigation of Brick, Martin and Angier.&apos; Following the casting techniques of Krynitsky, Saunders and Stern,s the ingot was melted under a heavy layer of dried charcoal in a clay-graphite crucible in a gas-fired pot furnace. Small lumps of silicon were held under thc melt surface with a graphite rod. After the silicon had dissolved, the alloy was chill-cast and then remelted without a cover, stirred thoroughly, skimmed and poured into a preheated cast iron mold. The cast microstructure is shown in Fig 2. The casting, 10 X 4 X K in., was annealed at 750°C for 21 hr and furnace cooled, About }{& in. was milled from all surfaces. The plate was cut in half longitudinally and the bottom half analyzed by A.S.T.M. and The composition by weight was as follows: copper 94.24 pct, silicon 5.56 pct, iron 0.11 pct.

Jan 1, 1949

By J. D. Lubahn, R. P. Carreker, J. G. Leschen

The formation of slip bands in crystalline solids undergoing plastic deformation has recently been treated as a problem of nu-cleation and growth.&apos; A simplified theory was developed and shown to be qualitatively consistent with certain experimental observations reported in the literature, such as the frequent failure of slip bands to traverse the entire crystal, the effect of specimen diameter and temperature on the yield stress, and the occurrence of incubation periods in the deformation of suddenly loaded crystals. The theory further predicts the appearance of transients in the rate of strain of a crystal when the applied stress is suddenly changed from one constant value to another. A transient consists of a discontinuous change in the strain rate followed by a gradual approach to a steady-state value. Such transient behavior apparently has not been recognized in previous experimental work. The preliminary experiments reported in the present paper were undertaken to confirm the existence of the predicted transients and thus to determine whether a more detailed treatment of the nucleation theory of slip might be theoretically and experimentally fruitful. Experimental Procedure A simple creep apparatus was assembled as shown in Fig I. Very little insulation was used on the furnace in order to facilitate rapid changes of temperature. Temperature was measured with a thermometer placed in the center of the furnace and was controlled by adjusting the Variac setting. The specimens were cold-headed in split-cylinder grips. The strain was measured as total elongation by a dial gauge sensitive to I/I0,000-in movement of the weight pan. Since the specimen gauge lengths were

Jan 1, 1949