Search Documents

By K. M. Busen, J. Lindmayer

A heterojunction model is suggested for descrihzng the electronic conditions at oxide -semiconductor interfaces. Detailed study of the silicon oxide-silicon interface shows that the heterojunction model explains many experimental facts; however, it is necessary to assume that a transition layer exists at the interface. It will he demonstrated that ionic contrbution way also he present. Extended low-temperature heating stabilizes the oxide to a charge carrier density in the order of 10' cln-: determined by the heterojuncton model. OXIDIZED semiconductor surfaces are widely used, since they offer technological advantages and "passivate" the surfaces of semiconductor devices. However, it has been recognized that there are always n-type inversion layers formed in the silicon oxide-silicon system, which must be regarded as a definite disadvantage. An added complication is the "instability" of these inversion layers;' the inversion layer appears to vary with the processing and they change at temperatures as low as 100" to 200°C This paper will point out that the inversion layer is a result of the dissimilarity of the contacting materials and therefore must always be present. The dissimilarity is the essence of the heterojunction or electronic model and involves directly the structure of the materials involved. In addition, one finds the contribution from ions; this disturbance, however. can be eliminated for the case of silicon oxide-silicon interface. The strength of the inversion layer or channel will be expressed in terms of the density of elementary charges per square centimeter, N*, maintaining the channel. The next section contains the formulation of the heterojunction model by assuming a contact between two homogeneous materials. This calculation should not imply that silicon oxide is homogeneous; it serves only to point out the general properties of the electronic model. Following the electronic model, the practical case of the silicon oxide-silicon interface will be discussed. I) THE HETEROJUNCTION MODEL Contacting an oxide with a semiconductor, a work-function difference must be accommodated. Fig. 1 represents the oxide and the semiconductor before and after contact. When the contact has been made, one finds that qVi + qV, = A@; i.e., the work-function difference is distributed between the oxide and the semiconductor. Normally, the work-function difference appears on the wide-bandgap material, but in the present case the oxide is thin and far from being a single crystal. Accordingly, a significant fraction of the work-function difference must appear on the semiconductor. The work function of silicon2 is about 4.8 ev, but we shall not assign any particular work function for the oxide as it may vary. However, oxides usually have lower work functions and therefore the potentials must build up in the manner shown in Fig. 1. Such an argument predicts that the inversion layer must be n type, as opposed to the clean surface which is p type. Another important factor to be considered is that the oxide is disorderly and therefore contains many trapping levels. As a result, the charges in the oxide arise primarily from the change in trap occupation, while on the semiconductor side the space charge is composed of mobile free electrons. The consequence of this fact is that a change to another thermal equilibrium will require some time, dictated by the traps in the oxide. The conditions can easily be calculated on the semiconductor side. Assuming an intrinsic semiconductor, Poisson's equation is

Jan 1, 1965

By Hermann Schmalzried, Carl Wagner

UPON heating a metal oxide or sulfide in H2, first only oxygen or sulfur is removed from the surface. Thus the metal/nonmetal ratio in the oxide or sulfide increases and the thermodynamic activity of the metal increases accordingly. For the formation of metallic nuclei, a certain super saturation is necessary, i.e., the activity of the metal must become greater than unity. Once nuclei of the metal have been formed, the supersaturation is supposed to decrease. Even under steady-state conditions of a reduction process, however, the activity of the metal inside an oxide or sulfide phase must be greater than unity in order to have a driving force for the individual steps of the reduction process. As an example, consider the reduction of Ag2S with a high mobility of cations and electrons. As shown schematically in Fig. 1, the reduction process involves the following steps:' 1) sulfur removal along the entire Ag2S surface in form of H2S, 2) migration of Ag' ions and electrons from the Ag2S surface to nuclei of metallic silver, 3) transfer of Ag+ ions and electrons from the Ag2S phase to the metal. For a finite rate of step 3, the activity of silver in Ag2S next to the Ag2S-Ag interface must be greater than unity. Moreover, for a finite rate of step 2, the activity of silver at the gas-Ag2S interface must be still somewhat greater. Theoretical considerations lead to the tentative conclusion that at least under steady-state conditions the supersaturation will be rather low.' To test this conclusion, the average activity aAg of silver in a Ag2S tablet about 2 mm thick with a diameter of 5 mm was followed as a function of time by measuring the electromotive force E of the cell Pt I Ag | Agl I Ag2S I Pt where AgI is used as a solid electrolyte in which only Ag+ ions are mobile. The activity aAg is calculated as aAg=exp(-EF/RT) Since the diffusion coefficient for transport of silver in Ag2S is very high (- 3 . l0-2 sq cm per sec),3 the average activity aAg is representative for both the bulk and the surface of the Ag2S tablet. Experiments were conducted at about 400°C. The cell was surrounded by an aluminum block in order to secure a uniform temperature and to eliminate thermoelectric effects. With the help of a potentiometer, first a potential of 0.16 v was applied to the above cell in argon in order to obtain a well defined initial state with a Ag/S ratio close to 2.000.~ Then the cell was disconnected from the potentiometer and a gentle stream of H2 was passed along the Ag2S surface. In view of the removal of sulfur in form of H2S, the electromotive force E decreased within 20 to 30 min to zero corresponding to a Ag/S ratio of 2.003 and became negative, indicating a value of aAg> 1. A representative plot electromotive force E vs time is shown in Fig. 2. The minimum of the electromotive force characteristic of the onset of

Jan 1, 1963

By L. W. Sink, B. H. Kear, S. M. Copley, M. F. Hornbecker

IT is well known that under appropriate conditions of chill casting, many metals and alloys develop a columnar grain structure having a high degree of preferred orientation.' In cubic materials, this principle has been employed to produce shaped castings with a unidirectional cube oriented columnar grain structure as well as shaped single crystals with a cube axis parallel to the growth ais.' "he purpose of this note is to describe how this same principle can be enlployed to produce wires, rods, or more complex crystal Shapes with a variety of conlpositions in cube or random orientations by a single casting operation. The procedure for directional solidification of in Fig. 2. Examination under crossed polarizers indicated that the striations were a second phase (presumably the B19 phase) which frequently forms initially at primary grain boundaries and then grows toward the interior as thin ribbons. X-ray studies on filings of this alloy reannealed at 1300°C for 15 min indicated a strong pattern of the fcc phase together with a substantial pattern of the B19 phase. The un-annealed filings, however, showed only the pattern of a cold-worked, fcc structure. In these two-phase alloys it appears, therefore, that the B19 phase is transformed into a disordered fcc structure by cold-working. The mechanical properties of these alloys would be of some interest since a fcc structure might be expected to possess greater ductility than the or-thorhombic B19 structure from which it forms. The author wishes to thank Mr. Harvey Yakowitz and Mr. Richard Manuszewski for obtaining the photomicrograph in Fig. 2. This research was partially supported by Research Grant DE-02455 from the National Institute of Dental Research. columnar grained ingots involves pouring a melt into a preheated mold having a water-cooled copper chill at its base? and controlling the temperature gradient in the mold so as to maintain columnar grain growth from end to end of the ingot.4 This is illustrated schematically in Fig. l(a). It should be noted that the longitudinal temperature gradient promoting unidirectional columnar grain growth is maintained by heat loss primarily through the solidifying ingot to the chill, so that both the temperature gradient and the growth rate diminish with increasing distance from the chill. As the distance from the chill increases? the columnar structure coarsens, and the perfection in preferred orientation tends to increase progressively. Using this same procedure of gradient solidification, Fig. l(b) shows a simple scheme for producing a batch of crystals having the same cube orientation rather than an assemblage of columnar grains. Wires and rods are produced by cementing ceramic tubes to the wall of the mold. The steep temperature gradient produced in the central chamber is controlled by the mold heater to produce favorable conditions for crystal growth. Clearly the success of the operation lies in the selection of individual growing columnar grains by some kind of orifice at the base of each channel in the mold. Using simple constrictions, experience with nickel-base alloys has shown that 4 -in.-diam holes at a distance of -1 in. from the chill yield crystals within 7 deg of (100); closer control of orientation is possible by placing the crystal selector farther away from the chill.

Jan 1, 1970

By Charles W. Chesterman

Perlite is a glassy volcanic rock which will, upon rapid controlled heating, expand or "pop" into a frothy material of low bulk density, valued as a lightweight aggregate. The term perlite also is applied to the expanded product. The chemical composition of perlite ranges from that of rhyolite to that of andesite; the water content ranges from 2 to 5%. Most perlite is pale gray, but some is black, brown, or even green. Inclusions in perlite are common and include phenocrysts of quartz, feldspar, biotite, and hornblende, and miscellaneous rock fragments. Occurrences of perlite are restricted to areas of Tertiary and Quaternary volcanic rocks. Perlite occurs as flows associated with thick accumulations of pyroclastic debris and other flow rocks, as dikes, sills, domes, and as selvages of intrusive bodies of volcanic rock. Individual flows of perlite range in thickness from a few feet to as much as 100 ft and commonly cover several square miles. Dikes and sills of perlite generally are small, but domes generally are substantial in size and may be as much as several miles in diameter. Most perlite deposits are large enough and topographically situated so as to lend themselves to low-cost open pit mining methods. Conventional equipment is employed in the mining of perlite. Although milling procedures may appear to be simple, care is exercised in order to produce a carefully sized product and to minimize the production of excessive quantities of fine materials, which have limited use and are considered as waste. Perlite is processed, or "popped," in one of two basic types of furnaces: horizontal rotary or vertical stationary. Both types of furnaces include product collection and dust control systems. Many commercial deposits of perlite are known to occur throughout the world, and although no exact estimate of reserves is available, there appears to be an adequate supply to meet current and future worldwide requirements. End Uses The uses of expanded perlite are many and varied and are based primarily upon its physical and chemical properties. Because of its (1) low bulk density, which generally ranges between 2 and 20 lb per cu ft; (2) low thermal conductivity "K" factor (Btu per hr per sq ft per OF per in, of thickness), which ranges between 0.267 for loosely packed perlite (bulk density of 4.48 lb per cu ft) to 1.25 for perlite gypsum plaster; (3) high resistance to fire; and (4) low sound transmission, expanded perlite is used to a great advantage as aggregate and insulating material in the construction industry. Early uses of perlite were related exclusively to the construction industry, which now consumes about 60% of the domestic production. The primary use of perlite now is as aggregate in insulation boards. Other uses of perlite in 1971 were as aggregate for plaster, horticultural aggregate, low-temperature insulation (cryogenics), and concrete. Because of its low thermal conductivity ("K" factor), high adsorption of sound, low bulk density, and fire resistance, perlite aggregate plasters hold many advantages over conventional plaster. Perlite aggregate plaster lends itself to machine applications, dries quickly, and has relatively high elasticity which reduces cracking after drying. These same features make perlite useful as aggregate in portland cement concrete and asphalt concrete for construction wherever weight reduction, insulation, and acoustics are of prime importance. Perlite-portland cement concrete offers up to 20 times more thermal insulation than conventional concrete, and because of this it finds wide application in roof decks and floor fill in multistory buildings, in

Jan 1, 1975

By C. H. Samans, G. F. Tisinai

HARDENING and embrittlement of the ferritic chromium stainless steels at temperatures near 885 °F have been known for a long time.' However, no satisfactory explanation has been given. Both ordering and precipitation hardening have been suggested. Fisher, Dulis, and Carroll2 discovered a chromium-rich body-centered-cubic precipitate in embrittled alloys. Subsequently, Williams' and Wright4 also have found evidence of this body-centered-cubic constituent. William's hardness, electrical resistance, and magnetic studies led to his proposal of a new constitutional diagram for the Fe-Cr alloys. The new feature of this diagram is the eutec-toidal decomposition, below about 970°F, of the s phase into two stable body-centered-cubic solid solutions, one rich in iron and the other rich in chromium. An observation of Heger5 that the amount of s phase has increased and the amount of chromium-rich body-centered-cubic phase decreased in the interval 5000 to 32,600 hr at 900°F, even though the alloy increased in hardness, casts some doubt on this diagram. If precipitation hardening causes the embrittlement, the precipitate probably is always coherent with the matrix, since no evidence of overaging has been reported. It cannot be stated positively, however, that Fisher's body-centered-cubic constituent is the embrittling phase. According to Preston's data,6 such a chromium-rich phase should have a parameter less than about 0.5 pct larger than that of the iron-rich matrix lattice. Moreover, the constitutional relationships whereby Fisher's constituent forms, and whether it is stable, metastable, or a transition phase, are not yet completely clear. The recent studies of Wright4 and of Lommel,7 as well as the earlier studies of Becket8 and of Hochmann,9 have shown that 885°F hardening occurs in relatively pure Fe-Cr alloys. No evidence of its occurrence or absence in chromium itself has been reported. The rate of hardening in high purity alloys seems appreciably slower than that in alloys containing larger amounts of impurities (chiefly carbon and nitrogen). This may be due merely to the absence of hardening and straining effects from precipitated carbides and/or nitrides. These observations on high purity alloys make it seem unlikely that 885 °F hardening is caused by a constituent containing other than iron and chromium atoms, e.g., carbides and/or nitrides. Wright,4 Lommel,7 and Lena and Hawkes10 emonstrated the accelerating effects of a finely dispersed precipitate of (probably) chromium nitride on the hardening reaction. This appears to be evidence that lattice straining increases the rate of nucleation of the hardening constituent. Some new observations on 885°F hardening are reported here. They show an effect of preheat treatment that should contribute to a better understanding of the 885°F hardening phenomenon. They also clarify a discrepancy which exists in the literature. Some investigators have reported most rapid hardening at 885", whereas others have reported it at 930°F. The present work also shows that 885°F hardening does not occur in electrolytic chromium, and that any long range ordering type of reaction affecting any sizeable part of the volume is unlikely. Materials Studied Table I shows the currlposition of the alloys used in this investigation. In these alloys carbon and nitrogen are present in roughly similar amounts. Presumably both elements precipitate, as carbides and/or nitrides, during the heat treatments given.

Jan 1, 1958

By Myron A. Pugacz, Frank S. McKenna, Clarence E. Jackson

The notched-bar impact test has proved worth while in certain applications as a test for control of the quality or the heat-treatment of steel. In view of the serious thought that even so simple a test as the tensile test has evoked during the past years, it is not surprising that the notched-bar impact test should have inspired much experimental work. The correlation of the great number of data has not been successful, although steady progress has been made toward a better understanding of this test. Progress has been made, especially, in the isolation of the effects of the many variables that control the results. The stress system imposed on a specimen with a notch is so complicated and unpredictable that no simple evaluation of the test has been possible. Because of this uncertainty, many engineers hesitate to use the notched-bar impact test, and this attitude may well continue until a better understanding of the effects of the variables in the test are available. Without doubt, any additional data that may be presented should be accompanied by as complete a description of the - details of test as possible, so that it may be added to the general fund of information that is to serve as the basis for the theory of the notched-bar impact test. The many factors that may control the test results have been pointed out repeatedly. The importance of the effect of the type and dimensions of the specimen has been recognized. Most of the work reported by American investigators has been limited to small-size (Charpy) specimens tested as a simple beam; two types of notches (Fig. I) have been commonly used. Considerable data have also been presented by many investigators using a small specimen tested as a cantilever beam; this commonly being called the Izod test. Here, however, a shear component of stress makes the stress system even more complex than that present in the simple beam type of test, hence a critical analysis is difficult. Much work has also been reported by investigators using the tension type of specimen. A second variable that must be controlled is the testing temperature. An indication that the tests were performed at room temperature is not sufficient, since in many steels a variation of a few degrees in the actual testing temperature may be sufficient to give an entirely different range of results. Other factors, such as the composition and condition of the material being tested and the speed of test, should also be considered. Many investigators have contributed data and theoretical discussions on the problems encountered in notched-bar test-

Jan 1, 1944

By L. F. Mondolfo

An aluminum-copper 4 pct Cu alloy aged at room temperature for times increasing up to 78,000 hr was annealed at 170°C and the hardness and electrolytic potential determined during retrogression and subsequent aging. It was found that there is a change of behavior with increasing time of aging at room temperature which indicates that, given sufficient time, retrogression disappears and the room-temperature aging tends to progress directly into the high-temperature aging. WHEN an age-hardenable alloy is aged at low temperature after proper quench, changes of properties and structure take place. If this material is then heated to a temperature higher than the previous one, but still well below the solution-treatment temperature, the properties and structure changes produced by the low-temperature aging disappear and the material apparently reverts to the as-quenched condition, ready to undergo aging from the start. This phenomenon is termed reversion or retrogression. First discovered by Cayler3l it has been extensively investigated, as shown by the number of publications listed in the bibliography. From all the information available the following facts emerge: 1) All aeg hardenable alloys can be retrogressed.9 13, 19) 24, 33, 41, 42, 44, 47, 53, 54, 61, 63, 64, 66, 73, 78 2) All the properties changes that take place during aging at low temperature can be retrogressed to a smaller or larger extent.4,5,7,12,19,32,49,50,57,58,60,67,75,76 67, 75, 76 3) Retrogression requires that the temperature be well above the preliminary aging temperature. If the difference between the two temperatures is small, Little or no retrogression takes place.7,20,32,33,41,49 57. 65 4) The larger the difference between the preliminary aging and the retrogression temperature, the faster and more complete is the retrogression.20, 33, 41, 48, 65, 67 5) Retrogression can be produced several times in the same alloy.23, 35,41,57,59 6) The amount of retrogression that can be obtained depends on the aging temperature. Alloys aged at the lower temperatures can be retrogressed the most. The higher the preliminary aging temperature the less retrogression can be produced.20,25,33,37,44,49,54,58 7) Appreciable retrogression after aging at elevated temperatures is possible only if the preliminary aging has lasted a short time. After a certain time of aging at high temerature retrogression does not take place any more,:5,6,9,10,11,21,40,58 8) During retrogression the Guinier-Preston zones redissolve and disappear. The other precipitate

Jan 1, 1960

By Walter R. Hibbard, George H. Eichelman, William P. Saunders

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,1 has resulted in the application of similar heat treatments to other eutectoid transformations. Studies2.3.4 of the beta copper-aluminum alloy eutectoid transformation revealed microstructures very similar to those found in iron-carbon alloys. Investigations5,6 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 reported' 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 I. The kappa field converges to a minimum silicon solubility at the eutectoid composition and all other completely kappa alloys are hypereutectoid. Thus, kappa decomposition involves first the proeutectoid precipitation of gamma, followed by the eutectoid transformation of a lower silicon content kappa (5.2 pct). Smith5 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'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.7 Following the casting techniques of Krynitsky, Saunders and Stern,8 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 the 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 ¾ in., was annealed at 750°C for 21 brand furnace cooled. About 1/16 in. was milled from all surfaces. The plate was cut in half longitudinally and the bottom half analyzed by A.S.T.M. methods.9 The composition by weight was as follows: copper 94.24 pct, silicon 5.56 pct, iron 0.11 pct.

Jan 1, 1948

By John D. Sullivan, Edwin J. Rogers, Chester R. Austin

THE problem of making a building unit combining the necessary physical and mechanical properties and good thermal insulation has been foremost in the minds of architects and ceramic and construction engineers for many years. This paper presents the results of studies on thermally expanding shales or clays to produce light-weight cellular products suitable for making building units. An expanded clay block suitable for building purposes should be large enough so that substantial savings may be effected in labor costs in laying the wall. The permeability should be low and it should have sufficiently good "dry-wall" characteristics to permit exposing the exterior surface to the elements, and to permit plastering directly the inside wall surface. The thermal conductivity should be low and the mechanical properties should be adequate. Obviously, a unit possessing satisfactory qualities for constructing an outside wall would be satisfactory for interior walls and for back-up purposes. The results of studies on bloating various shales or clays† and the properties of the expanded products are reported. In another paper1 basic factors involved in bloating of clays are discussed. Manufacturers of ceramic products, particularly structural clay, may encounter bloating of the ware during burning, and precautions must be followed to avoid this phenomenon. About 1927 the C. F. Burgess Laboratories, Madison, Wis., while carrying on an investigation for the American Face Brick Research Corporation, conceived the idea of intentionally bloating clay while holding it in a restrained condition, with the hope of producing a unitary cellular mass. Work was carried on until about 1931, when studies were terminated because of business conditions. These studies were continued later by The Western Brick Co., Danville, Ill., which developed a process of "inwardly expansive bloating." The major purpose of the latter investigation was to produce back-up tile to compete with concrete blocks made from Haydite or cinder aggregate. Haydite is a light-weight aggregate formed by heating certain clays or shales rapidly in a rotary kiln. While the investigations at the C. F. Burgess Laboratories or at Western Brick Co. were not carried far enough to lead to the commercial production of material for the construction of walls, particularly outside walls, they showed that certain clays could be thermally expanded and that it was possible to produce a material of low permeability with a large percentage of the total voids or cells closed. Commencing in 1939, the Ozark Chemical Co. sponsored a research program at Battelle Memorial Institute for the purpose of determining whether it is technically feasible to make thermally expanded building units, and specifically to get answers to the following questions:

Jan 1, 1942

By Frank S. McKenna, Myron A. Pugacz, Clarence E. Jackson

The notched-bar impact test has proved worth while in certain applications as a test for control of the quality or the heat-treatment of steel. In view of the serious thought that even so simple a test as the tensile test has evoked during the past years, it is not surprising that the notched-bar impact test should have inspired much experimental work. The correlation of the great number of data has not been successful, although steady progress has been made toward a better understanding of this test. Progress has been made, especially, in the isolation of the effects of the many variables that control the results. The stress system imposed on a specimen with a notch is so complicated and unpredictable that no simple evaluation of the test has been possible. Because of this uncertainty, many engineers hesitate to use the notched-bar impact test, and this attitude may well continue until a better understanding of the effects of the variables in the test are available. Without doubt, any additional data that may be presented should be accompanied by as complete a description of the - details of test as possible, so that it may be added to the general fund of information that is to serve as the basis for the theory of the notched-bar impact test. The many factors that may control the test results have been pointed out repeatedly. The importance of the effect of the type and dimensions of the specimen has been recognized. Most of the work reported by American investigators has been limited to small-size (Charpy) specimens tested as a simple beam; two types of notches (Fig. I) have been commonly used. Considerable data have also been presented by many investigators using a small specimen tested as a cantilever beam; this commonly being called the Izod test. Here, however, a shear component of stress makes the stress system even more complex than that present in the simple beam type of test, hence a critical analysis is difficult. Much work has also been reported by investigators using the tension type of specimen. A second variable that must be controlled is the testing temperature. An indication that the tests were performed at room temperature is not sufficient, since in many steels a variation of a few degrees in the actual testing temperature may be sufficient to give an entirely different range of results. Other factors, such as the composition and condition of the material being tested and the speed of test, should also be considered. Many investigators have contributed data and theoretical discussions on the problems encountered in notched-bar test-

Jan 1, 1944

By O. George Specht, Carl A. Zapffe

In recent print, some remarkably contradictory statements have appeared regarding the importance to be attached to sponge Iron,1-6 a metallurgical commodity whose history goes back at least to the time of Diodorus Siculus in the fourth century B.C. This disagreement, however, expresses the lack of a satisfactory method for producing sponge iron rather than defection in the commodity. The blast furnace is admitted to have certain inherent disadvantages, such as a general unsuitability for magnetite and the heating and blowing of vast quantities of undesirable nitrogen, so that a workable sponge-iron process remains attractive. In this respect, low-temperature gaseous reduction shows especial promise; but confusion persists regarding the factors involved. The present paper, therefore, has three principal divisions: (I) an exhaustive examination of all previous work on low-temperature gaseous reduction, (2) experi- mentation on the reduction of a magnetite with hydrogen, and (3) development of certain thermodynamic relationships in reduction, with a summarized depiction of the factors in the general process. PREVIOUS INVESTIGATIONS Gaseous media used in the reduction processes under discussion can be classified in three types: (I) H2, (2) CO, and (3) other gases. In the second, the CO is often admixed with H2. The third principally comprises hydrogeniferous gases, such as natural gas and producer gas, and in addition includes chlorine. Because hydrogen therefore features in all three, a study of hydrogen reduction based on investigations using pure hydrogen alone may be incomplete. The present paper reviews the three groups of investigations in the listed order, paying particular attention to hydrogen reduction. For the benefit of future workers, an exhaustive bibliography has been prepared and similarly classified. Group I: Reduction with Hydrogen Equilibria Although the present problem is specifically one of kinetics, a first consideration is the equilibrium of the system, since the reactions to be studied comprise that equilibrium, and their direction and end points are defined by it. The historic development of these researches is interesting and informative.

Jan 1, 1947

By O. George Specht, Carl A. Zapffe

In recent print, some remarkably contradictory statements have appeared regarding the importance to be attached to sponge Iron,1-6 a metallurgical commodity whose history goes back at least to the time of Diodorus Siculus in the fourth century B.C. This disagreement, however, expresses the lack of a satisfactory method for producing sponge iron rather than defection in the commodity. The blast furnace is admitted to have certain inherent disadvantages, such as a general unsuitability for magnetite and the heating and blowing of vast quantities of undesirable nitrogen, so that a workable sponge-iron process remains attractive. In this respect, low-temperature gaseous reduction shows especial promise; but confusion persists regarding the factors involved. The present paper, therefore, has three principal divisions: (I) an exhaustive examination of all previous work on low-temperature gaseous reduction, (2) experi- mentation on the reduction of a magnetite with hydrogen, and (3) development of certain thermodynamic relationships in reduction, with a summarized depiction of the factors in the general process. PREVIOUS INVESTIGATIONS Gaseous media used in the reduction processes under discussion can be classified in three types: (I) H2, (2) CO, and (3) other gases. In the second, the CO is often admixed with H2. The third principally comprises hydrogeniferous gases, such as natural gas and producer gas, and in addition includes chlorine. Because hydrogen therefore features in all three, a study of hydrogen reduction based on investigations using pure hydrogen alone may be incomplete. The present paper reviews the three groups of investigations in the listed order, paying particular attention to hydrogen reduction. For the benefit of future workers, an exhaustive bibliography has been prepared and similarly classified. Group I: Reduction with Hydrogen Equilibria Although the present problem is specifically one of kinetics, a first consideration is the equilibrium of the system, since the reactions to be studied comprise that equilibrium, and their direction and end points are defined by it. The historic development of these researches is interesting and informative.

Jan 1, 1947

By Thomas A. Read

The internal friction of single crystals of metals is affected markedly by a variety of factors, which, according to the literature, are without influence on the internal friction of polycrystalline metals at strain amplitudes less than 10-5. For example, the internal friction of a suitably oriented zinc single crystal is multiplied by a factor of ioo when the vibration amplitude is slightly increased, and the internal friction of a copper crystal is markedly increased by the application of a stress of 30 lb. per sq. in. for one minute. In this paper it will be shown that these, together with the other phenomena associated with the damping of mechanical vibrations in metal single crystals, can be satisfactorily explained, at least qualitatively, in terms of an atomic mechanism that has been proposed for quite another purpose—that of accounting for slip in metals. The extraordinary agreement between the observed behavior of the metal single crystals and the behavior that could be predicted on the basis of this atomic mechanism leads to the belief that the latter is probably essentially correct. If this is so, it follows that valuable information about the slip process in metals may be gained from measurements of internal friction on single crystals. In this paper the results of the only study to date on the damping of mechanical vibrations in single crystals of metals mill be presented. Experimental Procedure The internal friction of a material is a measure of the rate at which the energy of mechanical vibrations is converted into heat. The measure of internal friction used in the present paper is the decrcment A, which is defined as the fraction of the energy of vibration lost per half cycle. The measurements described here were made for longitudinal oscillations at a frequency of 33.5 kc. for the copper crystals, and at a frequency of 39 kc. for the zinc crystals. The method used is that of the composite piezoelectric oscillator. The single-crystal specimens are in the form of circular cylinders M in. in diameter and I 1/2 to 21/2 in. long. The crystal rods are waxed in a V-block and sawed to the proper length with a fine jeweler's saw. The ends of the specimen are ground flat, and one end is cemented with a very thin film of beeswax and rosin to the end of an x-cut quartz rod which is 1/4 in. in diameter and 3 in. long. The length of the specimen rod is determined by the condition that its natural frequency for longitudinal vibration be the same as that of the quartz rod, within 0.5 per cent. The lengths of the specimens are thus different for different orientations of the cylinder axis with respect to the crystallographic axes. When the specimen and quartz rods are thus matched in frequency the adhesive is at a node of stress, and it has been shown experimentally that under these circumstances

Jan 1, 1941

By Gust Bitsianes, Joseph H. M. Beaty

Successful blast-furnace operation depends upon securing an optimum balance between a number of important variables. This balance will vary somewhat from furnace to furnace in the same plant and with raw materials available in different plants. It is difficult to isolate the effect of any one variable unless it has a rather pronounced effect upon the efficiency of operation. The physical properties of iron ore and coke are known to affect the process, yet it is difficult to set up any definite yardstick by which these differences can be appraised. In recent years, whenever practicable, operators have taken steps to obtain better control over the absolute size and uniformity of size of the ore and coke. The size and physical character of the stone, however, have received comparatively little attention, emphasis being placed upon chemical composition, fluxing efficiency, and action as a desulphurizing agent. Since the stone occupies only about 10 per cent of the volume of the stock column, its influence upon permeability is less than that of the ore and the coke. The large pieces now used have a tendency to roll to the center and thus increase the permeability of this part of the stock column. Smaller pieces would be more effective in promoting the permeability of denser portions of the stock. The size and physical properties of blastfurnace stone are important also for reasons best understood by considering the results obtained by the crushing and the sizing of iron ore. Crushing ore to about 2 in. at Provo, Utah, and to about I in. in Alabama produced material somewhat less permeable to gases, even though sized. In spite of some sacrifice in permeability, the efficiency of the process was improved by crushing and sizing the ore, as indicated by a substantial saving in fuel and increased output. This saving in fuel is due in large measure to more thorough reduction of ore in the upper part of the furnace and to a corresponding decrease in the amount of carbon consumed prematurely above the tuyeres by the reaction CO2 + C = 2CO - 70,080 B.t.u. [I] The physical properties of stone, its size and rate of calcination are important because they influence the rate at which CO2 is released and the completeness of calcination before reaching a level in the furnace where the temperature does not exceed 1050°C. (1920°F.). Briefly, carbon dioxide from the limestone or from the reduction of ore may cause premature corn bustion of carbon above the tuyeres. Assuming that 1000 Ib. of limestone is used per ton of pig iron, about 435 Ib. of CO2 will be liberated during calcination of a good grade of stone. If all this CO2 from the stone were to oxidize carbon prematurely, about 120 lb, would be consumed above the tuyeres in a heat-absorbing reaction. The amount of carbon thus consumed when a

Jan 1, 1943

By Marie L. V. Gayler

Without diffusion taking place in liquid metals and alloys, no castings could be made; it is therefore the most important factor affecting the structure of metals. Diffusion involves the interchange of atoms, a process which takes place during the life history of an ingot. In the liquid, atoms are in a state of high thermal agitation, but this thermal vibration is reduced in amplitude with falling temperature. When solidification sets in, the atoms arrange themselves on an ordered space lattice and the changes in structure which take place in the solid are due to the interchange of atoms on this space lattice. The rate of atomic interchange—e.g., of diffusion—is most rapid at temperatures close to the melting point and decelerates as the temperature is lowered. At room temperature, the rate of diffusion is practically negligible in most alloys, and any change in structure can only be brought about by using methods which will accelerate the rate of diffusion. The rate of diffusion depends on three factors (I) temperature; (2) concentration gradient and (3) activation energy. The growth of grains and particles in the solid alloy also depend on these factors and are therefore intimately associated with diffusion. The rate of homogenization in a cast alloy depends on several factors, such as: (I) rate of solidification, which involves grain and particle size which, in their turn, depend on diffusion: (2) the nature of the constituents; (3) the temperature interval between the melting point of the metal or alloy and the temperature at which it is being heat-treated. It is well known that coarse, cast structures are extremely difficult—if possible— to homogenize, and the finer the grain size and particle size, the more readily solid solution will take place. The rate of diffusion depends, also, largely on the nature of the constituents; for instance, copper and magnesium form solid solutions with aluminum quite readily but nickel and iron do not; neither do certain intermetallic compounds, such as alpha and beta FeSi. Furthermore, the actual formation of intermetallic compounds may be a very slow process, particularly when they are formed in the solid state as equilibrium products. For example, BFeSi is formed in the solid as the result of the reaction aFeSi + A1 —+ PFeSi. The particles of aFeSi become surrounded with a reaction ring of PFeSi. This layer is formed as the result of diffusion of atoms taking place between the aluminum matrix and aFeSi. As the amount of aFeSi surrounding the core of aFeSi increases the concentration gradient becomes low and so the rate of diffusion of atoms also decreases until it practically ceases. This occurs after a comparatively thin layer of ßFeSi has been formed. It follows that when intermetallic compounds of this type are formed in the solid, as the result of diffusion, it is generally impossible ever to attain equilibrium, since the forma-

Jan 1, 1944

By A. V. Deforest, C. W. Macgregor, A. R. Anderson

One of the important problems in the design of structures and machine parts subjected to rapidly applied loads is the determination of the strength and ductility of the material itself under such conditions. Realizing the need for such information, many investigations have been carried out in the past. Most of these experiments have been made using notched bars in bending either with the Charpy or Izod impact machine. While yielding valuable data on the effect of notches in the energy absorption of the material and on the condition of heat-treatment, this form of test does not lend itself readily to a determination of fundamental stress-strain relations. For this and other reasons, more attention has been paid recently to the tension impact test. An extensive literature already exists on the subject and only those investigations somewhat closely related to the present paper will be mentioned. For a more comprehensive list, the reader is referred to the Symposium on Impact Testing or the A.S.T.M., 1938. The work of Mann.' Clark and Dätwyler,2 Nádai and Manjoine,3 Winlock and Leiter,4 Deutler,5 and Brinkmann,6 however, deserves especial mention. In most of the previous investigations on the subject, when stress-strain relations were obtained under high rates of loading, it was the common procedure to determine stresses based on the original area and strains based on the original length. Stress-strain curves were then constructed by P ALo plotting So = 1/A0 as a function of eo = ALr,o where A0 and Lo refer to the original area and gauge length, respectively. At present there exists a paucity of true stress-strain data in the impact problem. The advantages of constructing so-called true stress-strain curves in order to reveal the fundamental physical properties of the material for the tension test were discussed recently by one of the authors.7'8 It was shown that the true physical behavior of the material in the tension test could best be represented P by plotting the average true stress, S = 2P as a function of 6 = q' = log —r = log ~9L where P, A, L, S, E, and q' refer to the load, the cross-sectioned area while the load P is applied, the actual gauge length, the average true stress, the true strain and true reduction in area, respectively. Besides giving a better physical representation, the method has certain additional advantages in that the stress-strain curve so plotted becomes a straight line from the maximin load point to fracture. The construction of such curves, however, while comparatively simple for the slow rates of loading customarily used in the tension test, presents certain dificulties for tensile impact in that both loads and diameters are required throughout the test. The continuous measurement of test-piece diameters to fracture under impact conditions offers considerable experimental difficulty. In order to make it feasible to

Jan 1, 1942

By Thomas A. Read

THE internal friction of single crystals of metals is affected markedly by a variety of factors, which, according to the literature, are without influence on the internal friction of polycrystalline metals at strain amplitudes less than 10-5. For example, the internal friction of a suitably oriented zinc single crystal is multiplied by a factor of l00 when the vibration amplitude is slightly increased, and the internal friction of a copper crystal is markedly increased by the application of a stress of 30 lb. per sq. in. for one minute. In this paper it will be shown that these, together with the other phenomena associated with the damping of mechanical vibrations in metal single crystals, can be satisfactorily explained, at least qualitatively, in terms of an atomic mechanism that has been proposed for quite another purpose-that of accounting for slip in metals. The extraordinary agreement between the observed behavior of the metal single crystals and the behavior that could be predicted on the basis of this atomic mechanism leads to the belief that the latter is probably essentially correct. If this is so, it follows that valuable information about the slip process in metals may be gained from measurements of internal friction on single crystals. In this paper the results of the only study to date on the damping of mechanical vibrations in single crystals of metals will be presented. EXPERIMENTAL PROCEDURE The internal friction of a material is a measure of the rate at which the energy of mechanical vibrations is converted into heat. The measure of internal friction used in the present paper is the decrement A, which is defined as the fraction of the energy of vibration lost per half cycle. The measurements described here were .made for longitudinal oscillations at a frequency of 33.5 kc. for the copper crystals, and at a frequency of 39 kc. for the zinc crystals. The method used is that of the composite piezoelectric oscillator. The single-crystal specimens are in the form of circular cylinders ¼ in. in diameter and 1 ½ to 2 ½ in. long. The crystal rods are waxed in a V-block and sawed to the proper length with a fine jeweler's saw. The ends of the specimen are ground flat, and one end is cemented with a very thin film of beeswax and rosin to the end of an x-cut quartz rod which is ¼ in. in diameter and 3 in. long. The length of the specimen rod is determined by the condition that its natural frequency for longitudinal vibration be the same as that of the quartz rod, within 0.5 per cent. The lengths of the specimens are thus different for different orientations of the cylinder axis with respect to the crystallographic axes. When the specimen and quartz rods are thus matched in frequency the adhesive is at a node of stress, and it has been shown experimentally that under these circumstances

Jan 1, 1941

By Marie L. V. Gayler

WITHOUT diffusion taking place in liquid metals and alloys, no castings could be made; it is therefore the most important factor affecting the structure of metals. Diffusion involves the interchange of atoms, a process which takes place during the life history of an ingot. In the liquid, atoms are in a state of high thermal agitation, but this thermal vibration is reduced in amplitude with falling temperature. When solidification sets in, the atoms arrange themselves on an ordered space lattice and the changes in structure which take place in the solid are due to the interchange of atoms on this space lattice. The rate of atomic interchange-e.g., of diffusion-is most rapid at temperatures close to the melting point and decelerates as the temperature is lowered. At room temperature, the rate of diffusion is practically negligible in most alloys, and any change in structure can only be brought about by using methods which will accelerate the rate of diffusion. The rate of diffusion depends on three factors (I) temperature; (2) concentration gradient and (3) activation energy. The growth of grains and particles in the solid alloy also depend on these factors and are therefore intimately associated with diffusion. The rate of homogenization in a cast alloy depends on several factors, such as: (I) rate of solidification, which involves grain and particle size which, in their turn, depend on diffusion: (2) the nature of the constituents; (3) the temperature interval between the melting point of the metal or alloy and the temperature at which it is being heat-treated. It is well known that coarse, cast structures are extremely difficult-if possible to homogenize, and the finer the grain size and particle size, the more readily solid solution will take place. The rate of diffusion depends, also, largely on the nature of the constituents; for instance, copper and magnesium form solid solutions with aluminum quite readily but nickel and iron do not; neither do certain intermetallic compounds, such as alpha and beta FeSi. Furthermore, the actual formation of intermetallic compounds may be a very slow process, particularly when they are formed in the solid state as equilibrium products. For example, ßFeSi is formed in the solid as the result of the reaction aFeSi + Al?ß FeSi. The particles of aFeSi become sur¬rounded with a reaction ring of ßßFeSi. This layer is formed as the result of diffusion of atoms taking place between the aluminum matrix and aFeSi. As the amount of ßFeSi surrounding the core of aFeSi increases the concentration gradient becomes low and so the rate of diffusion of atoms also decreases until it practically ceases. This occurs after a comparatively thin layer of ßFeSi has been formed. It follows that when intermetallic compounds of this type are formed in the solid, as the result of diffusion, it is generally impossible ever to attain equilibrium, since the forma-

Jan 1, 1944

By A. V. Deforest, A. R. Anderson, C. W. MacGregor

One of the important problems in the design of structures and machine parts subjected to rapidly applied loads is the determination of the strength and ductility of the material itself under such conditions. Realizing the need for such information, many investigations have been carried out in the past. Most of these experiments have been made using notched bars in bending either with the Charpy or Izod impact machine. While yielding valuable data on the effect of notches in the energy absorption of the material and on the condition of heat-treatment, this form of test does not lend itself readily to a determination of fundamental stress-strain relations. For this and other reasons, more attention has been paid recently to the tension impact test. An extensive literature already exists on the subject and only those investigations somewhat closely related to the present paper will be mentioned. For a more comprehensive list, the reader is referred to the Symposium on Impact Testing or the A.S.T.M., 1938. The work of Mann.' Clark and Dätwyler,2 Nádai and Manjoine,3 Winlock and Leiter,4 Deutler,5 and Brinkmann,6 however, deserves especial mention. In most of the previous investigations on the subject, when stress-strain relations were obtained under high rates of loading, it was the common procedure to determine stresses based on the original area and strains based on the original length. Stress-strain curves were then constructed by P ALo plotting So = 1/A0 as a function of eo = ALr,o where A0 and Lo refer to the original area and gauge length, respectively. At present there exists a paucity of true stress-strain data in the impact problem. The advantages of constructing so-called true stress-strain curves in order to reveal the fundamental physical properties of the material for the tension test were discussed recently by one of the authors.7'8 It was shown that the true physical behavior of the material in the tension test could best be represented P by plotting the average true stress, S = 2P as a function of 6 = q' = log —r = log ~9L where P, A, L, S, E, and q' refer to the load, the cross-sectioned area while the load P is applied, the actual gauge length, the average true stress, the true strain and true reduction in area, respectively. Besides giving a better physical representation, the method has certain additional advantages in that the stress-strain curve so plotted becomes a straight line from the maximin load point to fracture. The construction of such curves, however, while comparatively simple for the slow rates of loading customarily used in the tension test, presents certain dificulties for tensile impact in that both loads and diameters are required throughout the test. The continuous measurement of test-piece diameters to fracture under impact conditions offers considerable experimental difficulty. In order to make it feasible to

Jan 1, 1942

By Marie L. V. Gayler

Without diffusion taking place in liquid metals and alloys, no castings could be made; it is therefore the most important factor affecting the structure of metals. Diffusion involves the interchange of atoms, a process which takes place during the life history of an ingot. In the liquid, atoms are in a state of high thermal agitation, but this thermal vibration is reduced in amplitude with falling temperature. When solidification sets in, the atoms arrange themselves on an ordered space lattice and the changes in structure which take place in the solid are due to the interchange of atoms on this space lattice. The rate of atomic interchange—e.g., of diffusion—is most rapid at temperatures close to the melting point and decelerates as the temperature is lowered. At room temperature, the rate of diffusion is practically negligible in most alloys, and any change in structure can only be brought about by using methods which will accelerate the rate of diffusion. The rate of diffusion depends on three factors (I) temperature; (2) concentration gradient and (3) activation energy. The growth of grains and particles in the solid alloy also depend on these factors and are therefore intimately associated with diffusion. The rate of homogenization in a cast alloy depends on several factors, such as: (I) rate of solidification, which involves grain and particle size which, in their turn, depend on diffusion: (2) the nature of the constituents; (3) the temperature interval between the melting point of the metal or alloy and the temperature at which it is being heat-treated. It is well known that coarse, cast structures are extremely difficult—if possible— to homogenize, and the finer the grain size and particle size, the more readily solid solution will take place. The rate of diffusion depends, also, largely on the nature of the constituents; for instance, copper and magnesium form solid solutions with aluminum quite readily but nickel and iron do not; neither do certain intermetallic compounds, such as alpha and beta FeSi. Furthermore, the actual formation of intermetallic compounds may be a very slow process, particularly when they are formed in the solid state as equilibrium products. For example, BFeSi is formed in the solid as the result of the reaction aFeSi + A1 —+ PFeSi. The particles of aFeSi become surrounded with a reaction ring of PFeSi. This layer is formed as the result of diffusion of atoms taking place between the aluminum matrix and aFeSi. As the amount of aFeSi surrounding the core of aFeSi increases the concentration gradient becomes low and so the rate of diffusion of atoms also decreases until it practically ceases. This occurs after a comparatively thin layer of ßFeSi has been formed. It follows that when intermetallic compounds of this type are formed in the solid, as the result of diffusion, it is generally impossible ever to attain equilibrium, since the forma-

Jan 1, 1944