Search Documents

Jan 1, 1916

Jan 1, 1913

Jan 1, 1944

By B. Alexander, G. Derge W. Peifer

This is part of a research project in the Metals Research Laboratory of Carnegie Institute of Technology, sponsored by the Carnegie-Illinois Steel Corporation. In organizing our program it became apparent that the problem of analyzing molten steel for hydrogen must be divided into two parts: (I) the correct analysis of a given sample, and (2) the correct sampling of the bath. Emphasis has been placed upon these divisions in the order named because methods of sampling cannot be evaluated until a satisfactory method of analysis is available. Vacuum fusion was selected from the list of possible analytical methods because it seemed to us to offer the best possibilities for giving rapid, accurate results on solid steel samples. A number of objections have been raised against this method as usually practiced. We have sought to meet some of these by proper design of the apparatus. Experiments will be described which show that others are not valid. In the conventional apparatus the sample is held under high vacuum for a period of several hours before it is actually analyzed and one might expect considerable hydrogen to escape during this period. We have provided a means of introducing the sample to the vacuum just before analysis. Another difficulty has been that the entire apparatus has been designed primarily for the purpose of analyzing for oxygen, which constitutes upward of go per cent of the total gases evolved from the sample The volume and pressure measurements have therefore been insensitive to the relatively small amounts of hydrogen present. This difficulty obviously can be taken into account in any apparatus designed especially for the amounts of hydrogen expected. Analysis in Vacuum-fusion Apparatus A number of experiments were carried out to determine the importance of several factors that have been stated as possible sources of error or uncertainty in vacuum-fusion analysis for hydrogen; namely, that hot graphite will absorb some of the gases, that metal vapor will cause discrepancies, and that low results may be obtained as a result of the water vapor (produced by the oxidation of the hydrogen) dissolving in the butyl phthalate of the manometer, or that the blanks obtained from the apparatus are so great as to make the method inaccurate. The experiments also checked on the accuracy and feasibility of using differential freezing as a method of gas analysis in a vacuum-fusion apparatus.

Jan 1, 1945

By Richard Franchot

From an efficiency viewpoint, the greatest loss of energy to the blast furnace is in its failure to convert more than about a third of the coke carbon from carbon monoxide to carbon dioxide. This result has usually been ascribed to a necessity of a 2:1 excess of CO in order to reduce iron. It has, however, been proved that iron is completely reduced by equal volumes of carbon monoxide and carbon dioxide. Hence there is room for the hypothesis that vaporization, as cyanid, of accumulated alkalis is a serious primary factor limiting the ratio of ore to coke. Observed cyanid vapor concentrations and those measured by the variations of the nitrogen-oxygen ratio in the hearth gases form the basis for a quantitative explanation of the furnace action. As EVERYBODY knows, the blast furnace is a gas producer. As to the reason for this, people are not so well agreed. The significance of the fact is obvious, for the extent to which the fuel leaves the furnace as combustible gas measures poor economy in iron smelting; while good efficiency in this respect means poor gas. The furnace thus works at cross purposes. The present trend of the steel industry is toward. a primary use of the blast furnace as a gas producer with pig iron as a byproduct. Needless to say, this trend is away from improvement in iron making. The production of more and better iron in a given furnace consuming a given amount of coke would seem, at least, to involve a decrease in the relative proportion of the combustible gas. The possibility at the same time, however, of making more and better gas need not necessarily be excluded. (The more the furnace acts as gas producer, the bigger is the sulfur problem.) Whatever the object, it would seem desirable to have a, clear, comprehensive, mathematically exact estimate of why and how the blast furnace functions, to the large degree that it now does, as a gas producer. The literature does not give a definite answer to the question. That for which explanation is sought is the fact that only a minor part, usually not more than a third, of the coke carbon burned is oxidized to carbon dioxide; that the development in the furnace of the fuel energy is thus limited usually to little more than 50 per cent.; that it requires about twice as much fuel coke to make a ton of iron as is indicated by the rela-

Jan 7, 1925

By Ernest Klepetko

A NOTABLE amount of fossil resin exists in many of the bituminous coal beds of Utah. The upper part of these show a marked concentration of resin, which occurs primarily in the fracture seams. In general, these seams are less than 1/8 in. thick, but it is not uncommon to find coal with areas as thick as 3/8inches. The resin is of the hard kauri variety, with coloration varying from deep blood red to lemon yellow and almost colorless. There is also an almost black variety, which is not readily soluble in the same solvents as the lighter colored resins. These resins are valuable for many industrial purposes. When refined, they are suitable for uses similar to those for which lac resins are usually employed. They are particularly desirable for preparing dielectric varnishes. They are also desirable in preparing printing inks, waterproofing agents, special synthetic rubbers and rubber cements. They are especially suitable as an admixture in synthetic rubbers and cements, as they supply the essential tackiness the synthetics lack. The following chemical trade characteristics have been found by Inter-Chemical Co. research chemists in the refined resins: [Specific gravity (melted) 1.03 -1. o6 Acid number6-8 Softening point, mercury method 160 -165°C Iodine value (Wijs)140-1.50 Refractive index1.544 Color (Hellige, 25 pct toluol) ...14 Molecular weight, average732 Purity, 99 + pct pure hydrocar Bon] "Unaffected by alkalies and resistant to one hour fusion with solid potassium hydroxide. Completely soluble in aliphatic and aromatic hydrocarbons. Insoluble in alcohols, ethyl acetate and other commonly used alcohols and esters. Compatible with natural and synthetic rubbers, chlorinated and cyclized rubbers, waxes, vegetable oils, mineral oils, rosin, ester gum, phenolic and modified phenolic resins and other natural and synthetic resins of like solvent solubility. The recovery and refining of these fossil resins has been the subject of much research work and has intrigued many investigators. Broadly speaking, the problem involves separation of the resin from the coal into a very high-grade concentrate, and subsequent treatment of this concentrate to produce an acceptable refined resin for its various usages. Obviously it is also desirable to recover the residual coal without excessive size degradation. The late W. D. Green, concentration engineer for Combined Metals Reduction Co., obtained U. S. Patent No. 1773997 for the recovery of resin from coal through

Jan 1, 1947

[Admissions: Aurel E. Smith, Chairman Edmond F. Egan Lawrence S. Fennell John S. Fuller Howard J. Ritts, Jr. Gene E. Roark Kenneth W. Robbins Ben M. Gottlieb Advertising: Bernard F. Klein, Chairman Henry N. Lyle R. E. Howard H. M. Krause, Jr. Mrs. Evelyn R. Robertson]

Jan 1, 1961

By William E. Ford, Edward Salisbury Dana

WheweUite. Calcium oxalste CaCz04.Hz0. In small colorless monoclinic crystals. Optically +. j3 = 1.555. From haxony, with coal; also from Bohemia, and Akace. Oxammite. Ammonium oxelate, (N&)rC20r.2HzO. From the guano of the Gudape Isbnds, Peru.

Jan 1, 1922

By T. S. Lundy, J. I. Federer

A graphical method has been developed and tested for separating the effects of grain boundary and lattice diffusion in polycrystalline materials. The method is based on the assumptions that for unidirectional diffusion from a plane source of radioactive material, the penetration of activity far into the specimen from the source is due solely to grain-boundary effects, and 2) pure grain-boundary penetration is described by a linear plot of log activity vs the perpendicular distance x from the source. The former is based on the fact that at any temperature the grain-boundary diffusion coefficient is always very large relative to the lattice diffusion coefficient; and the latter is based on the derivation of Fisher.1 Using these criteria along with appropriate penetration data (in this case, data for the diffusion of Cb95 in pressed and sintered UO2) one first plots the data as log activity vs x2 and as log activity vs x, as shown by curves 1 and 2, respectively, in Fig. 1. A tangent line to curve 2 for large x (curve 3) is then transferred to the x2 plot as curve 4. The subtraction of curve 4 from curve 1 yields curve 5 which represents only lattice diffusion and may therefore be related to the diffusion coefficient by using the appropriate solution to Fick's second law. It should be noted that the method has inherent inaccuracies for small and large values of x2 and only the subtraction at intermediate values can be used. For the illustrated specimen the calculated diffusion coefficient is 3.4 X 10-11 sq cm per sec at1445°C while data using a single crystal of UO2 at the same temperature yielded a value of 2.9 x 10-11 sq cm per sec.2 The same method of subtracting grain-boundary diffusion effects to obtain lattice diffusion coefficients has been successfully used to determine self-diffusion coefficients of Th230 in polycrystalline thoria.

Jan 1, 1962

By Haim H. Chiswik, Ralph Hultgren

In recent years considerable attention has been devoted to the development of new commercial uses for silver, chiefly because of lowered prices and large stocks of the metal. This has led to increased interest in the phase diagrams of silver alloys. The silver-lead diagram is reported' to be of the simple eutectie type, but no measurements of the solubility of lead in silver have been made. The solubility of silver in lead has been estimated2 from diffusion experiments and resistance measurements to vary with temperature from 0.09 to 0.18 atomic per cent (0.05 to 0.10 weight per cent). Silver-rich silver-lead alloys have recently been recommended as high-strength bearing metals3 The silver-bismuth diagram is similar.' The solubility of bismuth in silver. has been estimated to be about 5.5 weight per cent on the basis of X-ray work.4 Matthiessen6 found a solubility of silver in bismuth from 0.25 to 0.50 per cent by conductivity measurements. Silver and thallium form the same type of diagram,' with no data on their mutual miscibilities. We have studied the first two of these diagrams by X-ray methods, verifying their main features and determining the solubilities of lead and bismuth in silver at several temperatures. We also determined the solubility of thallium in silver at one temperature. The solubilities of silver in lead and in bismuth were found to be too low to be detectable. Materials and Methods Mallinckrodt precipitated silver, analytical reagent, was used. Maximum impurities listed on the label were: chloride, 0.005 per cent; copper, 0.001; iron, 0.002; other heavy metals (as Pb), 0.002; and sulphate (SO³), 0.05. The lattice constant was found to be a0 = 4.0782 $., in good agreement with the value of Jette and Foote,6 a0 = 4.0779 A., for

Jan 1, 1940

By Ralph Hultgren, Haim H. Chiswik

In recent years considerable attention has been devoted to the development of new commercial uses for silver, chiefly because of lowered prices and large stocks of the metal. This has led to increased interest in the phase diagrams of silver alloys. The silver-lead diagram is reported' to be of the simple eutectie type, but no measurements of the solubility of lead in silver have been made. The solubility of silver in lead has been estimated2 from diffusion experiments and resistance measurements to vary with temperature from 0.09 to 0.18 atomic per cent (0.05 to 0.10 weight per cent). Silver-rich silver-lead alloys have recently been recommended as high-strength bearing metals3 The silver-bismuth diagram is similar.' The solubility of bismuth in silver. has been estimated to be about 5.5 weight per cent on the basis of X-ray work.4 Matthiessen6 found a solubility of silver in bismuth from 0.25 to 0.50 per cent by conductivity measurements. Silver and thallium form the same type of diagram,' with no data on their mutual miscibilities. We have studied the first two of these diagrams by X-ray methods, verifying their main features and determining the solubilities of lead and bismuth in silver at several temperatures. We also determined the solubility of thallium in silver at one temperature. The solubilities of silver in lead and in bismuth were found to be too low to be detectable. Materials and Methods Mallinckrodt precipitated silver, analytical reagent, was used. Maximum impurities listed on the label were: chloride, 0.005 per cent; copper, 0.001; iron, 0.002; other heavy metals (as Pb), 0.002; and sulphate (SO³), 0.05. The lattice constant was found to be a0 = 4.0782 $., in good agreement with the value of Jette and Foote,6 a0 = 4.0779 A., for

Jan 1, 1940

In the mile-high city of Denver. Colo., the mining industry's political adjunct, The American Mining Congress, held its annual meeting, September 27-30. The Honorable Maurice H. Stans, in what became the keynote address of the meeting, spoke of a "common interest in a basic reality of 1970, a reality involving ecology, energy, raw materials and natural resources.”

Jan 1, 1970

By John A. Putnam, Morrough P. O’Brien

DATA on the physical and thermodynamic properties of hydrocarbons have been made available in recent years but the formal method of applying these data to flow in porous media appears not to have been fully developed. By making usual thermodynamic assumptions and employing existing data on permeability to two-phase flow, Darcy's law may be applied to obtain the flow characteristics for representative combinations of flowing medium and geometry. The numerical examples presented herein are based on hydrocarbons for which complete physical and thermodynamic data are available under equilibrium conditions. INTRODUCTION Thermodynamic data are obtained in the laboratory under equilibrium conditions and their application requires the assumption that equilibrium also exists during flow. This assumption is at least usual in engineering analyses, though not easily justified in many instances, and will be followed here. In regard to hydrocarbons, laboratory experience in determining the pressure-volume-temperature relationships indicates that gas-liquid mixtures must often be stirred vigorously for periods of 15 to 20 min. before reaching equilibrium. On the other hand, boiling and condensation are known to occur on solid surfaces and the large surface exposed to the flowing substance by a porous medium may bring about rapid approach to equilibrium. For the purpose of providing at least a guide to experimentation and a standard for later comparison with experimental results, it will be assumed that the phases present in cross sections perpendicular to the direction of flow are in thermodynamic equilibrium. Another assumption usually made in the solution of problems of flow in porous media is that the fluid remains at a constant temperature; i.e., isothermal flow. This assumption is arrived at by considering the porous medium and surroundings to be an infinite heat source and the rate of heat flow to be infinite at a boiling front. Its validity depends in large measure upon the character of flow, the number of components, and the physical and thermodynamic properties of the fluid. However, the assumption of isothermal flow represents a limiting case. The other limit can be represented by the assumption that the fluid is thermally isolated; i.e., irreversible adiabatic. The thermodynamic process characterizing the actual flow of a given fluid system probably lies somewhere between these two limiting cases and might be thought of as a flow system in which the confining walls are maintained at a constant temperature. For isothermal flow the solid material of the porous media is considered as the infinite heat source capable, without loss in temperature, of giving up heat to the fluid as the latter tends to cool by expansion. For the thermally isolated system, the solid material is considered to give up no heat to the fluid. In the intermediate case the impervious walls confining the porous media are at constant temperature from which heat is transferred along various paths, solid and fluid, throughout the porous media. It would appear that an actual petroleum-

Jan 1, 1941

By C. K. Leith

The question as to what grades of ore it pays to conserve for the future, and the valuation of low-grade reserves, are becoming immediate and definite as applied to the individual ore deposits, and seems to call for some sort of quantitative answer.

Jan 1, 1915

By Leland L. Mink, Roy E. Williams, C. Daniel Kealy

The passage by Congress of Public Law 92-500 (the 1972 Water Pollution Control Act amendment) has placed a new emphasis on the need for methods to minimize the escape of water from tailings ponds. Presumably tailings pond seepage will fall under the jurisdiction of the act. This paper is designed to explain some of the problems caused by leakage from tailings ponds and to present a method by which leakage can be minimized if not eliminated. Ground-water pollution in one case studied by the authors results from leaching of old jig tailings that are intermixed with the upper part of a sand and gravel aquifer. Anomalous concentrations of zinc, lead, and cadmium occur in ground water and soil samples taken from the portion of the sand and gravel aquifer containing the old jig tailings. Analysis of water samples from a tailings pond located in the area indicates that the pond water is not the source of the heavy metal concentrations found in the down-gradient ground water. However, leakage from the pond and water from the pond's decant system combine with natural ground water to compound the problem by raising the water table into old jig tailings which cover the upper portion of the aquifer. The ground water which discharges from this aquifer causes a statistically significant decrease in the quality of the receiving stream water. The shallow ground water contains oxygen; therefore, corrective measures include lowering the water table to depths below the old jig tailings which cover the aquifer at ground level. This, in turn, implies elimination of ground-water recharge by subsurface leakage from the pond. A laboratory study and a mathematical model study have revealed that seepage from such ponds can be minimized by a carefully designed and managed peripheral tailings discharge system. The process involves maintaining the slime zone and the free-water surface near the center of the pond so that the slimes retard downward and semilateral movement of pond water. The free-water surface must not be allowed to extend out over the nonslime portion of the tailings during or after pond operation. Mathematical model output reveals that this method can hold the subsurface discharge rate to as little as 110 gpd per acre of pond water area. The technique should be applicable to any tailings pond system in which the loss via seepage of poor quality water must be minimized or where recharge of ground water by losses from a tailings pond must be minimized, regardless of the reason.

Jan 1, 1974

By H. M. Bannerman, E. N. Cameron

INTRODUCTION DURING the years 1942-1944, about 125 New England deposits were mined for sheet and punch mica, and many others were briefly prospected. During this period the Geological Survey, United States Department of the Interior, carried on detailed studies of the deposits and the pegmatite bodies that contain them. The primary purpose of the studies was to provide geological information that would contribute toward efficient mine development and operation. The studies were aimed directly at practical production problems. Detailed maps showing the structure of the deposits were prepared, and geological findings were closely correlated with data on specific yields and costs of mining and processing. The result is information that permits a reasonably accurate appraisal of the mica deposits of New England and the mining industry that they support. At a time when the future of the industry is in doubt, it seems desirable to submit this appraisal for consideration. Domestic mica deposits supply only a small fraction of the country's needs in normal times, for only the richest domestic deposits can compete with the low-cost mines of India and Brazil. However, the possibility that this country may be cut off from its supplies in time of war gives domestic deposits an importance much greater than their peacetime production would imply. New England supplied about 22 per cent of domestic output in 1943 and 1944, the peak years of wartime production. Its actual contribution to the war effort was greater, for the proportion of sheet mica of high quality in the New England product is higher than the average for domestic mica. The data analyzed below were collected for the Geological Survey by D. M. Larrabee, A. H. McNair, J. J. Page, V. E. Shainin, G. W. Stewart, and the co-authors of this paper. The cooperation of officials of Colonial Mica Corporation and of mine operators and owners in furnishing information is gratefully acknowledged. D. M. Larrabee, A. H. McNair, and V. E. Shainin have made various suggestions based on critical reading of the manuscript. ANALYSIS OF WARTIME OPERATIONS In considering the data presented in this report, it must be remembered that the scale of mica operations is very small compared with that of many other mining industries. The typical New England mica mine of 1943 and 1944 was worked by a crew of five or six men. Equipment was varied but on the whole extremely simple. The first stage of operation at a deposit was normally opencut mining. Drilling was done mostly with jackhammers powered by portable compressors driven by gasoline engines. Muck was shoveled by hand into buckets with a capacity of 1/3 to 2/3 ton and commonly hoisted by derrick to a dump adjacent to the cut. As operations were carried to greater depth, and the transition to underground mining was made, these methods were modified to a degree depending largely on the attitude of the deposit mined. At steeply dipping deposits,

Jan 1, 1946

By B. H. Caudle, A. B. Dyes, R. A. Erickson

The study of the influence of fluid mobilities on the sweepout pattern resulting from the injection of gas or water has been extended to cover the production period which follows breakthrough of the injected material. Mobility ratios over the range common in field operations (0.1 to 17) were studied for several pattern floods (five-spot, staggered, and direct line drive). The experimental data required for these studies were obtained by the use of the x-ray shadowgraph technique using miscible oil phases of different viscosities in porous plate mode1s of a reservoir element. From the shadowgraph pictures obtained before and after breakthrough of the injected fluid, flowing ratios at the producing well and cumulative volumes injected were calculated. The method for applying such data in predicting field behavior is illustrated for a water flood of a five-spot. For this case a range of mobility ratios of 0.5 to 5.0 results in: (I) nearly complete (95 to 100 per cent) sweepout pattern efficiencies at abandonment conditions, (2) production after breakthrough being responsible for as much as one-third of the total recovery at the lower mobility ratios, and (3) a twofold variation in the operating life of the reservoir. In determining the merits of a secondary recovery operation, the fraction of the reservoir which is ultimately to be swept by the injected fluid is of major importance. Because of its significance a large amount of research has been applied to the study of areal sweepout efficiency. Muskat, Wyckoff and Botset1,2 have presented the classic analysis of the pattern efficiency at breakthrough for several injection patterns in which the mobility of the fluids ahead of the injected phase front was equal to the mobility of the fluids behind the front. Aronofsky3 as extended this analysis of recovery at breakthrough through numerical computations and by the stepwise use of the potentiometric model to include the effect of different fluid mobilities ahead of and behind the invasion front on the line drive performance. Fay and Prats' have also employed numerical computations to show the influence of one other condition of fluid mobilities on the five spot flood. In a previous paper' a new method of attack on this problem which employed x-ray and porous plates was described. At that time experimental results were also presented which described the effect of fluid mobilities on the reservoir area contacted by the injected fluid at breakthrough for the five spot and direct line drive patterns. Most of the literature pertaining to sweepout pattern efficiency has emphasized the pattern which is obtained at breakthrough of the injected phase. The increase in the sweepout pattern which occurs after breakthrough

Jan 1, 1955

By Walter Kauenhowen

The production of crude oil in Germany during 1933 totaled about 1,669,521 bbl., a slight increase over the 1,608,558 bbl. produced in 1932. This is an extension of a yearly increase which has continued uninterruptedly since 1928. The 1933 increase in production is due chiefly to developments in the Nienhagen field. The production of the Wietze and Oberg fields remained rather stable, while the Oelheim-Eddesse and Volkenroda fields showed decreases. Nienhagen Field.—This field furnished about 61 per cent of the total German oil production in 1933. Most of the drilling activity centered the north district near the village of Nienhagen, where the trend of

Jan 1, 1934

By Hsun Hu, R. S. Cline

The annealing behavior and the mechanism of re-orientation during recrystallization of iodide titanium cold-rolled 94 pct have been studied in detail. Results indicate that recrystallization occurs by the nucleation and growth of new grains, as in other common metals. Recrystallization nuclei form by the coalescence of subgraim, and the change in texture as a result of recrystallization is largely due to selective growth among the nuclei formed. The annealing of titanium is characterized by a wide range of overlap of the various stages of the annealing process, which may be responsible for a range of activation energies observed, and for the apparently gradual change in the annealing texture as a function of time or temperature. The deformation and recrystallization characteristics of titanium and zirconium are very similar. In cold-rolled strip, the deformation texture consists of two symmetrically oriented components, each having the basal plane laterally tilted at about 30 deg from the rolling plane and the [1010] direction parallel to the rolling direction. Upon annealing for recrystallization, the change in texture can be described, for simplicity,* as rotations around [0001].2'6'8 According to McGeary and Lustman,' recrystallization occurs in zirconium through normal growth of the subgrains, which they called "domains", without the nucleation of new grains; and the magnitude of rotation around the [0001] axis increases gradually during the progress of recrystallization. If these conclusions were true, the mechanism of recrystallization in zirconium would be basically different from that in most metals, since it is commonly known that recrystallization with reori-entation always involves the migration of high-angle boundaries. In an attempt to clarify the situation, the mechanism of reorientation during recrystallization in iodide titanium cold-rolled 94 pct was studied in detail. The structural and textural changes upon annealing at various temperatures were examined by optical and transmission-electron microscopy, X-ray pole figures, pole density distribution measurements, and micro-beam techniques. EXPERIMENTAL PROCEDURE Material and Specimen Preparation. An iodide titanium crystal bar was are-melted and solidified in a cold-hearth crucible under a purified argon atmosphere. The solidified ingot had dimensions of approximately 3 by 1/2 by 3 in. One face of the ingot was somewhat uneven, but was as clean and shiny as the remaining parts of the ingot. Large grains with a Widmanstatten internal structure were clearly shown on the shiny surfaces, indicating the occurrence of P — a transformation upon rapid cooling from the melt. Analysis of the are-melted ingot indicated C 0.033, N 0.010, H 0.013, 0 0.002 in weight percent, and traces of iron, copper, and silicon as detectable impurities. The ingot was cold-rolled -40 pct to 0.300 in. thick with a reduction of 0.005 in. per pass. The defects on the uneven side of the ingot were then removed by machining. This reduced the thickness to 0.285 in. The piece was then recrystallized by annealing at 800°C for 1 hr in a fused silica boat charged into a fused silica tube furnace under a vacuum of 10~5 mm Hg. To refine the grain size, the recrystallized metal was again cold-rolled 40 pct to 0.170 in., then annealed at 700°C for 1 hr. These treatments yielded a strip with a uniform equiaxed grain structure, having a penultimate average grain diameter of 0.04 mm and a hardness of approximately 90 Dph. Final rolling reduced the thickness from 0.170 to 0.010 in., corresponding to a reduction of 94 pct. The strip was rolled in both directions by reversing end for end between passes. Surface lubrication was provided by oil-soaked pads attached to both rolls. Specimens of 1 in. length (for X-ray examinations) and +in. length (for hardness and microstructure examinations) were cut from the rolled strip, and a width of & in. was cut from the edges of each specimen by a jeweler's saw. These specimens were then etched in a solution of 10 cu cm HN03, 5 cu cm HF, and 50 cu cm H,O to 0.008 in. thick to remove the surface metal, as well as the distorted metal at the saw cuts, prior to annealing or measurements. To minimize any surface reaction with the atmosphere, all specimens were kept in an evacuated desiccator. Isothermal Anneals. All annealing treatments were conducted in vacuum in a fused silica tube furnace as described earlier. The temperature of the furnace was controlled to within *2"C. The specimen was placed in a fused silica boat, then pushed into the hot zone of the furnace. It took about 5 to 6 min for the specimen to reach the furnace temperature. After the specimen was held at temperature for a desired length of time the boat was pulled to the cold zone of the furnace; the heating-up period was excluded from the isothermal annealing time. Thus, the uncertainty in annealing time is higher for very short anneals, but negligible for long anneals.

Jan 1, 1969

By E. C. Wright

THE following article is based on observation of college graduates entering the steel industry in technical work made during the past 25 years, the first five of which were spent as a college instructor in two engineering schools, and the last 20 in the employ of a large steel company that has hired hundreds of engineering graduates for steel-mill operations. The college graduates encountered during this period have come from many, colleges representing schools in Eastern States, Midwestern Universities, many Southern colleges, and many smaller colleges and normal training schools mostly in western Pennsylvania and Ohio. The college training of these students for engineering work naturally has varied widely and the educational qualifications of the entrants have been in many ways different. It was found at an early date that graduates of standard engineering colleges were much more adaptable than graduates in science from nontechnical schools. The engineering students seemed to be more suited to actual work in steel' plants and in general were better able to grasp a knowledge of the operations they observed. However, although the graduates of engineering schools have had somewhat better training -in physics, mathematics, and engineering, it has been a general and continual observation that most of the students employed have had inadequate training in the fundamental sciences. Since this discussion relates entirely to metallurgical engineers, it should be emphasized that students taking metallurgical engineering courses in college seem to be poorly trained in organic chemistry, physics, physical chemistry, thermodynamics, and any higher form of mathematics than simple calculus. It has been noted frequently that young engineers faced with problems in a mill that involve some mathematics generally flounder badly, and that they have little conception of physical chemistry. Another common deficiency lies in the young graduate's lack of knowledge of where to look for the more detailed literature on any subject. He reverts instinctively to some college textbook whose coverage is usually very elementary. It is recommended that such textbooks contain a bibliography of the source literature, and that students be taught to use theoriginal data. Many of the students had had previous courses in college dealing with metallurgical work such as steel melting, metallography, etc., but it was found that their grasp of these subjects was extremely sketchy. It was soon realized that an ordinary engineering graduate could be taught more about steel melting and manufacturing processes in a, few weeks in a. plant than he could learn in a college course on such subjects. A similar statement may be made regarding the ability of students to conduct laboratory work. Their lack of technique and skill in handling equipment and setting up experiments was soon apparent. It became neces-

Jan 1, 1946