Search Documents

By Alden B. Greninger

Although the crystallography of lattice transformations has been studied extensively during the past few years, these studies have been limited, with few exceptions1'2, to specimens in which the transformed lattice appears as a segregate phase. A realization of the technical importance of the segregation mechanism, especially in the formation of martensitic structures and in age-hardening phenomena, has undoubtedly stimulated research in this direction. In the crystallographic studies of segregation, the general problem has been to relate the orientation of the parent crystal to that of the segregate crystals. The technique has been complicated by the fact that it has been necessary to deal with the presence of as many as 24 ideal segregate orientations within one parent lattice; furthermore, the segregate lattices are invariably scattered (seldom less than ±2º, often as much as 15" or 20') about these ideal positions. It is probable that a more complete picture of phase transformations may result from a study in which complete lattice relationships between individual grains are evaluated, thereby providing an additional parameter which when it is necessary to work with multiple segregate orientations, cannot be evaluated. The present paper presents the results of such a study, ? wherein a peritectic reaction is made use of to produce the transformed lattices. Matebials and Methods Pem'tectic Reaction.—In Fig. 1 is reproduced a portion of the copper-zinc equilibrium diagram, which includes the peritectic reaction a + liq. —ß. A liquid of composition between B and D upon solidifying will

Jan 1, 1937

By Walter Crafts

Tests carried out at the Union Carbide and Carbon Research Laboratories, Inc., on the effects of grain-refining deoxidizers are in substantial agreement with the data presented by Mr. Comstock. Notable is the observation that, although boron and aluminum increased hardenability, the steel was of poor quality in comparison with the other boron-treated steels to which at least two grain refiners were added. The effect of zirconium in improving the ductility of the chromium-nickel steel is in agreement with our experience that zirconium is specifically beneficial to chromium steels. In order to amplify Mr. Comstock's results and to indicate some of the basic principles of special deoxidation, the primary observations of our study have been summarized. The effect of boron in increasing hardenability was described by R. Walther, U. S. Patent No. 1519388, Dec. 16, 1924, but effective amounts in incompletely deoxidized steel are sufficient to harm hot workability and room-temperature ductility. Work at the Union Carbide and Carbon Research Laboratories, Inc., has shown that in order to obtain deep hardening without using such large amounts of boron as to result in impairment of ductility, the steel must be thoroughly deoxidized. This may be accomplished by a high silicon content or a grain-refining addition of strong deoxidizers. The degree of grain refinement and hardenability are inclined to be erratic when only one grain-refining element is used, although aluminum, zirconium, titanium, and vanadium have been used singly and resulted in increased. although somewhat erratic, hardenability. The ductility likewise was inclined to be inconsisterlt when only one grain-refining deoxidizer was added. By combinations of grain-refining elements, it was found possible to combine the specific virtues and counteract some of the defects of each clement. In addition, the complex deoxidizers gave more consistent grain refinement and hardenability. When used in combinations, each element adds its own specific influence as well as modifying the character of the other elements. The function of aluminum is primarily to give consistent results and to minimize abnormalities resulting from too

Jan 1, 1943

By James Norman, E. W. Gieseke

Spodumene is a lithium-bearing pyroxene, and is an important source of lithium compounds. Because of its high alumina and lithia content, it might be a desirable constituent of glass batches. The use of spodumene in ceramics will undoubtedly increase, as it has been shown that a mixture of the proper proportions of this mineral with soda feldspar and potash feldspar has a much lower maturing temperature upon firing than has any one of the three components. The mineral spodumene seldom occurs as pure mineral, but ordinarily with feldspars, mica and quartz. All spodumene-bearing rocks tested contained these minerals as gangue. Small amounts of iron and manganese oxides also were present, usually as stains on the silicate grains. Mica associated with the spodumene carried a considerable amount of lithia. This was discovered in flotation tailings, which, although virtually free of spodumene grains when examined microscopically, showed 0.4 to 0.5 per cent Li²O on chemical analysis. The lithia content of micas picked from the North Carolina spodumene rock was variable. Some samples contained little lithia, others more than 3 per cent. In 1938, under the supervision of Oliver C. Ralston,S the Bureau of Mines, in cooperation with the Tennessee Mineral Products Corporation, undertook to study the beneficiation of North Carolina spodumene by froth flotation. The Bureau had previously published investigations on the beneficiation of spodumene.2,3The methods described in the second publication3 did not give good results with badly weathered spodumene rock, so that more thorough investigation of the subject was warranted. Study of Method of Flotation Separation Most of the work on concentration of spodumene was done with rock from near King's Mountain, North Carolina. Rock from two

Jan 1, 1942

By David H. Browne

Nickel has always been a fruitful mother of problems. Previous to the year 1906 nickel was regarded as an element replacing iron in copper-mattes, and it was believed that the same laws which governed the elimination of iron could be applied to nickel. This conception has lately undergone a complete change, and nickel in all its stages is now regarded as an element replacing copper; in other words, copper-nickel is considered not as two elements, but as one metal. When we commenced our investigations on Monel metal we made a great many experiment* on the conversion of mattes, both high copper and high nickel. We studied closely the composition of the mattes and the changes produced by the blow. We found in all cases certain agreements which me believe are due to some law. Briefly stated, our conclusions are as follows : 1. Nickel is not an element replacing iron in matte. 2. Nickel-copper alloys act in the matte-blow like one metal. 3. Nickel-copper alloys follow, during the matte-blow, exactly the same laws that govern the behavior of copper alone. This brings up the question: What are the laws governing the production of metallic copper, and what are the relations of copper to sulphur and iron during the Bessemer blow?

Jan 1, 1911

By S. Frederick Ravitz

IN recent years, two theories of the mechanism of flotation have received a great amount of attention. One of these is Ostwald's theory of linear flotation,1 according to which most collectors contain three active groups: a hydrophilic or water-loving, a metallophilic or mineral-loving, and an aerophilic or air-loving group. Under the action of these three groups, Ostwald considers a single ring of collector molecules at the three-phase air-liquid-mineral boundary sufficient to bring about the flotation of the mineral particle. Siedler, Moeller and Reddehase2 objected to this theory immediately after it was presented, but most of their objections were answered by Ostwald.3 Very recently, however, Wark and Cox4 have more adequately criticized this theory, and their criticism makes its validity appear doubtful. The other important recent theory is the "chemical" theory of flotation, proposed by Taggart and his coworkers,5 who maintain that "all dissolved flotation reagents, which, in flotation pulps, either by action on the to-be-floated or on thenot-to-be-floated particles affect their floatability, function by reason of chemical reactions of well recognized types between the reagent and the particle affected." In general, they appear to consider most flotation results to be brought about by purely chemical reactions between the collectors and the products of oxidation present

Jan 1, 1933

By G. M. Swinkels, A. Vizsolyi, E. Peters

A process is described utilizing ferric chloride leaching followed by reduction of the cupric content to cuprous by cement copper, with CuCl obtained thereafter by crystallization. The residual liquor is cemented with sponge iron to produce cement copper and a copper free ferrous chloride solution. The latter is oxidized in an autoclave with oxygen to regenerate ferric chloride and to precipitate excess iron as a granular, filterable hematite. Cuprous chloride crystals are reduced to high purity metal with hydrogen, while sulphur reports almost quantitatively in residues in the elemental form. Pyrite and molybdenite are not leached. This paper describes the important variables in the process, as well as the deportment of minor elements as silver, gold, selenium, tellurium, bismuth, nickel, lead, zinc, magnesium, etc. An elegant method of isolating silver from CuCl crystals is described, as well as recovery of gold from residues by a chloride reagent The process is described as non-polluting and capable of producing refined copper, ore grade iron oxide, elemental sulphur, and yields of minor metals as good or better than obtainable by conventional smelting, without the use of solvent extraction or electrochemistry.

Jan 1, 1981

By F. O. Garrabrant

Power is furnished to the Climax Molybdenum Co. by the Public Service Co. of Colorado over two 100,000-volt lines to a bank of three 3333-kva. transformers 100/13.8 kv. These transformers are so designed that with the addition of forced cooling their capacity may be increased 331/3 Per cent at 12,000-ft. altitude. Under present conditions, these transformers are operated at 140 per cent of their rating with forced air cooling. On the secondary side of this bank there is a 15-kv., 1200-amp. Westinghouse oil circuit breaker provided with automatic reclosing relays allowing three successive short circuits before lockout. From this breaker 13.8-kv. power is transmitted to a central switching station (Fig. I) where all circuits radiate to the different operations of the property. Each circuit is protected by a 23-kv., 600-amp. oil circuit breaker with by-pass and isolating disconnecting switches.

Jan 1, 1947

By M. F. Smith

BELOW the nose of the bainitic C curves in isothermal transformation diagrams, the rate of the austenite-to-bainite transformation usually decreases with decreasing temperature. However, in some steels and iron-carbon alloys, there is a noticeable acceleration in the initial reaction rate at temperatures just above Ms. Such speeding-up of the bainitic transformation would be understandable at temperatures just below Ms because of the catalytic effect of the martensite formed,' but the reversal in isothermal kinetics just above Ms seems quite anomalous. This behavior was observed by Howard and Cohen' in a 1.35 pct C steel and, on the basis of metallo-graphic examination, it was attributed to the formation of a thin plate-like product not belonging in the bainitic sequence. More recently, schaaber2 found two-stage reactions occurring in the vicinity of Ms in a number of steels, and suggested that the first stage might be the isothermal formation of marten-site, preceding the isothermal formation of bainite in the second stage. A decrease in the bainitic incu-

Jan 1, 1960

By Stanley Herold

As an accompaniment to the interesting papers which discuss the practical aspects of forced drive, or flooding, in our oil fields, and to others which deal with laboratory or field tests on the same subject, a few remarks on the theoretical principles concerned may not be considered inappropriate at the present time. Without claiming to have had the field experience which some of you certainly possess I venture to present some of the results of an analytical investigation upon this method of increasing the recovery of oil from our natural reservoirs. I firmly believe that these results will accord with your experience. I shall confine the present discussion to the consideration of a perfectly ideal situation Let us assume that the productive formation lies perfectly horizontal, that it is of uniform thickness, of perfectly homogeneous texture, and that it contains oil with gas in solution, and no water, throughout a definite, although undefined, lateral extent. We shall furthermore assume that the pressure existing within the reservoir as thus specified, is such that some of the gas is in the free state, spread uniformly throughout, in the form of minute bubbles within the pores of the formation. We now have an ideal reservoir subject to the peculiar action of capillarity as described by the French physicist Jules Jamin. In all directions within the reservoir we see these bubbles of gas, separated by globules of oil which tend to remain in the capillary channels that connect the pores. The action of these bubbles and globules under the application or release of pressure has been fully described by Jamin.1 Most of our oil fields in the Mid-Continent and Eastern districts are subject to this action of capillarity: The fact that the actual reservoir is often far from being ideal in the manner specified above does not preclude the, influence of this phenomenon. The oil, existing under a pressure exerted by the compressed gas, is securely locked in place

Jan 11, 1926

By F. F. Lines

IT is not the intention of the writer to enter into a discussion of the relative merits of the duplex process as compared with the straight scrap and pig iron process, working under the same conditions, but rather to present in a general way several modifications of the former process, and then describe the method adopted by the Maryland Steel Co., as best suited to local conditions, and also to meet the metallurgical problem presented in the removal from the iron of an unusual constituent-chromium. The principal factors to be considered in deciding whether or not duplexing should be employed are: (1) Chemical constituents of pig iron. (2) Variety and quantity of steel to be produced. (3) Relative cost of pig iron and scrap. (4) Regularity of operation. (5) Capital to be invested. (6) An existing Bessemer plant. Having decided to adopt the duplex process, several combinations may be used, viz.: (1) Acid-lined converter with basic open hearth. (2) Acid-lined converter with electric furnace. (3) Basic-lined converter with basic open hearth. (4) Basic-lined converter with electric furnace. (5) Basic open hearth with electric furnace. The combination of an acid-lined converter with a basic open-hearth furnace is the one most commonly employed, and as it is the one best adapted to the conditions encountered at the plant of the Maryland Steel Co., we will briefly outline the advantages and disadvantages of the different methods of working.

Jan 4, 1915

By A. L. Holley

THOSE who are familiar with working gas furnaces will perhaps admit that the ordinary producer is the least satisfactory feature of the- whole system, chiefly by reason of its great waste of fuel, both above and below the fire. The waste of fuel in cleaning the fire has not, as far as I am aware, been carefully measured ; it has been roughly estimated at various works as averaging twenty per cent. to twenty-five per cent. The coke that falls into the ash-pit is largely mixed, if not stuck together with clinker; when this coke is not recovered and utilized (and where coal is cheap it does not pay to recover it), the waste is certainly as high as above estimated. Cleaning a fire on a grate (especially a fire of dirty coal, like much of our Western coal), by inserting a few temporary grate-bars, then withdrawing the regular grate-bars and letting the whole lower part of the fire tumble into the ash-pit, must necessarily be wasteful of fuel, but it has been found to be the most practical system. The fire under a boiler should be thin, to insure perfect combustion; hence it may be moved from one side to the other, so that .the clinker and ashes may be uncovered and removed. But a producer fire should be so thick that carbonic oxide only shall rise from it ; a thick fire cannot be thus manipulated. In several of the French works the ordinary producer without grates is employed ; the whole body of the fire rests on the floor of the ash-pit. It did not seem to me that less fuel was wasted here than with grates in cleaning fires. The bottom layer of the fire had to be drawn out with hooks, and there was nothing to prevent much coke from coming away with the clinker and ash. A producer shaped like a cupola furnace has many advantages, but if grates are employed the waste of coke in cleaning is not lessened. But the waste at the bottom of the fire is only a part of the waste in the ordinary producer. Mr. Emmerton, chemist to the Joliet Iron and Steel Company, has lately made a number of gas analyses from the producers of that company, worked with impure Illinois

Jan 1, 1880

By V. M. Lovell

The recovery of apatite from the phoscorite ores occurring in the Transvaal, Republic of South Africa, involves a flotation process that is particularly difficult to characterize from a fundamental point of view. However, the reagents used, (tall oil fatty acid, nonylphenyltetraglycol ether, sodium silicate, and sodium hydroxide) can successfully bring about a separation of apatite from dolomitic calcite, which is the most troublesome gangue component. Recently, it has been demonstrated with some flotation feeds that the addition of certain polysaccharides can improve the flotation significantly. It has been shown that the metallurgy obtained is closely linked to the nature of the froth, and this, in turn, is influenced, often dramatically, by the presence or absence of certain of the reagents. The character of the froth depends less on the inherent frothing ability of these reagents than on the presence or absence of fine mineral particles in the froth, and on their surface characteristics and state of aggregation. The polyglycol ether and the polysaccharides have been shown to affect these properties, and it is suggested that a major function of these reagents is to modulate the properties of the fine particles in the pulp, thereby changing the character of the froth and giving rise to improved results.

Jan 1, 1976

By R. L. Frantz, R. W. Bouman

The authors investigated the comminution of quartz and limestone mixtures in a ball mill, showing that the distribution modulus of a comminuted material is the same whether it is ground separately or as part of a mixture. Analysis of the ground products indicates that quartz and limestone consume energy during ball milling in proportion to their mass fraction in the mixtures. Recently Schuhmann' made a significant contribution to comminution theory by proposing that complex comminution operations comprise the summation of many individual comminution events. By means of the single comminution event hypothesis, Schuhmann derived the Charles energy-size reduction relationship by considering that fragments of every particle crushed during a single impact have a size distribution characterized by the Schuhmann size distribution. In this paper, the size reduction of mixtures of solids in ball mills is analyzed by considering that size reduction occurs as many single comminution events. If a mixture of two materials is being ground in a ball mill, each successful impact of a ball with a particle of one material should yield an assembly of particles whose size distribution is characteristic of that material. Considering both materials to follow the Schuhmann size distribution equation, the size distribution of material 1 in the product is

Jan 1, 1962

By S. Usui, I. Matsuoka, H. Sasaki, J. Shimoiizaka

Depression of galena by sulfite or chromate ion in xanthate flotation was investigated by means of adsorption measurements, electron diffraction and flotation tests. Flotation of galena was depressed by sulfite ion only when galena had been oxidized; no depression took place with unoxidized clean galena. A similar behavior was found when chromate ion was used as a depressant, with the exception that the chromate ion depressed unoxidized clean galena around neutral pH. Depression was attained with galena on which xanthate still remained in appreciable amounts, indicating that the depression is not due to a decrease in xanthate adsorption. The galena surface produced sparingly soluble films under conditions in which depression of galena took place. These films were identified by electron diffraction as lead sulfite and lead chromate on oxidized galena and as chromic oxide on unoxidized clean galena. These observations in conjunction with contact angle measurements, indicate that the depressing action of sulfite or chromate ion arises primarily from film deposition resulting from the reaction between depressants and the galena surface. The thickness of the film was analyzed by means of electrochemical polarization.

Jan 1, 1976

By Samuel Gibson Frantz, G. W. Jarman

The purpose of this paper is to relate briefly the development of magnetic separation and its extension from the separation of iron into its present use in the nonmetallic field, to suggest possible future extensions of its field of usefulness and its possible cooperation with other separating methods. The engineering design of separating machines, and statistics as to the number of machines in operation, types, their location and production are outside the scope of this paper. Specific data will be given, however, on certain selected applications. A large part of human endeavor consists in separating one thing from another; the good from the bad, the desired from the undesired. In fact, a philosopher might plausibly claim that separating things into categories and choosing between them is the object of thought itself. In dealing with matter, separation is nowhere more inherently necessary to attain our objects than in mining and metallurgy. Both our desire and our ability to make separations are results of the physical differences between things. Gross separations such as removing ore from worthless rock or coal from slate are usually made visually and depend upon form and color, which are physical properties. When the desired and the undesired things are intimately mixed, however, it becomes impractical to select visually even if the individual particles are readily recognizable to the eye, and some automatic way of making the particles separate themselves on the basis of their own inherent differences in physical properties must be resorted to. Some of the physical properties which have been used as a basis of separation are: color, shape, hardness, size, density, elasticity, surface propertics (surface tension in contact with liquids, etc.), electrical conductivity, dielectric constant, and magnetic susceptibility. Differences in these properties are used in the following methods of separation: color sorting, sifting, settlement in air and in liquids, float-sink methods, flotation, electrostatic and magnetic separation.

Jan 1, 1932

By Christopher W. Schram, Allan B. Tanner, Frank E. Senftle, George R. Boynton, Philip W. Philbin

A borehole sonde using a 252Cf neutron source and a Ge(Li) gamma-ray detector was field tested in a coal bed in Belmont County, Ohio, to ascertain whether such a device could be useful as a coal-exploration tool. As the sonde was lowered into the borehole, successive spectra were accumulated over short time intervals. The intensities of the capture gamma rays from hydrogen and of decay gamma rays from aluminum were determined simultaneously in each spectrum. A log of the intensities versus depth shows the depth and thickness of the coal bed. With the sonde stationary at points above, in, and below the coal bed, capture gamma rays were measured to show the variation in concentration of certain elements. From the spectrum recorded in the bed, a complete ultimate analysis, the percent ash content, and the heat content were determined without recourse to an external standard. Delayed gamma-ray analyses of the core were also made on the surface in the vicinity of the borehole in a fixed-geometry irradiation facility wing the detector in the borehole sonde. The concentrations of 10 elements were determined by this method on a coal split from the drill core. These experiments show that it is feasible to use a 252Cf- Ge(Li) borehole sonde as an aid in coal exploration.

Jan 1, 1979

By Harold H. Haman

In view of Twentieth Century advances in mineral dressing technology, most modern metallurgists would be inclined to class jigging as an early, somewhat primitive, representation of their art. The practice of metallic sulfide concentration by jigging is becoming rare. This is so, in part, because the type of ore amenable to the process is becoming rare. It is appropriate, however, that this symposium recognize that a significant contribution of domestic zinc remains dependent upon coarse preconcentration by jigging, and that the state of the art be described even though the theory and applications are already well delineated in technical literature. One of the few areas still using jigging as a major concentrating process is The Upper Mississippi Valley. AREA HISTORY AND JIGGING The Upper Mississippi Valley (UMV) has produced one billion dollars worth of zinc and lead at current prices. The district includes the southwest part of Wisconsin, the northwest corner of Illinois and an adjoining narrow strip of Iowa. Within a total area of 4,000 square miles, important sub-districts cover about 400 square miles. As is widely known, early mining in this district involved shallow lead---crevice deposits of galena that were very selectively mined, hand sorted and cleaned, and then smelted at or near the site. The district has produced almost a million short tons of metallic lead, a very large part of this during the fifty- year period following 1830. Zinc mining began at the end of the civil war, but at a very low rate until about 1905.

Jan 1, 1970

By Donald Barton

THE purpose of this torsion balance survey and the quantitative calculations was to delimit the extension of the cap rock at the Hoskins Mound salt dome. The sulfur ore in the Gulf Coast sulfur-bearing salt domes occurs in the lime rock, which is the outer layer of the cap-rock mass, and also in the underlying gypsum zone. The status of the exploration and exploitation of the Hoskins Mound dome at the date of this survey is shown by Fig. 1. The mound had been covered by the very early prospect wells of the Mound Oil Co., and the later and more numerous prospect wells of The Texas Company. The Freeport Sulphur Co. was producing from the flank wells in the southwest quadrant. The limits of the cap were surmised to lie inside the centers of quad-rangles 4F, 3F, the center of the NE. ¼ of 2E, and the north line of 2D. That surmise was based partly on the data of Fig. 1 and partly on the position of the edge of the topographic mound. The management of the Freeport Sulphur Co. wished to block out the extent of its ore reserves with a minimum of expenditure. The management was informed by the writer that no geophysical method could map the presence or quantity of sulfur directly, or determine the presence and quantity of the lime-rock cap independently of the rest of the cap, but that a reasonable probability existed that the approximate position of the edge, the depth and the thickness of the cap-rock mass as a whole could be determined through quantitative calculations based on the data of a very careful torsion balance survey. Considerable lime rock would probably be present, of course, wherever much cap rock was present. This paper presents the data of the survey, our calculated prediction of the configuration of the cap rock, and the degree of confirmation by the subsequent drilling. The survey was made by the Geophysical Research Corporation, under the direction of the writer. The work of the calculations was done by Andrew Gilmour, then assistant to the writer.

Jan 1, 1936

By Foster Fraas

MOST commercial electrostatic separators utilize the electrical property of conductivity, but although based on the same principles, they are constructed in a variety of forms, a common one being the simple roll type. Some information obtained during an investigation of that type of separator, and which indirectly is applicable to other types, is presented here. NUMERICAL RELATIONSHIP In Fig. I material discharged from hopper 6 is fed onto grounded rotating roll 2. Electrode 3 is at a high electrical potential with respect to roll 2, and any conductive particle in contact with roll 2 therefore acquires an electrical charge of such polarity that, when present in the electrical field, the force on it is in the direction of electrode 3. A detailed description of the charging by conductance as utilized in electrostatic separators may be found in a previous publication.1 As a result of this force, the path of the projected particle is changed in increasing amounts in the direction toward region C as the magnitude of the charge is increased. With the dividing edge 4 the projected particles may be divided into two fractions -those in region C designated as "deflected" or " deflectant " and those in region B designated as "residue." [ ] When material is passed over the roll only once, a complete separation is seldom obtained, therefore most commercial separators contain several roll-separator units in series, and some practical machines use as many as 12. In this paper a "pass" means the passage of material over a single-roll unit. The procedure is to repass either residue B with a low potential on electrode 3 or deflectant C with a high potential on 3. Most electrostatic separations follow a flowsheet of roughing and cleaning either the concentrate or the tailing, on both. In a specific instance, a dry, heated mixture of wet cleaned chromite and garnet from an

Jan 1, 1942

By Ernest R. Darby

If the correct balance between fuel and air is maintained in an open-flame furnace,' little chemical action may be expected between the products of combustion and the metal being melted. Physical' changes in the metal are the object of the melting operation, and indirectly, under certain conditions, may be the cause of subsequent chemical actions. If sufficient heat has been introduced into the metal to bring about the change from the solid phase to the liquid phase, and still further from the liquid to the vapor phase, the metal vapor in all probability will be carried from the furnace by the exhaust and rapidly combined with the oxygen of the outside air. The charge being melted in the furnace may contain elements which are chemically inactive at normal temperatures but which will form compounds at the elevated temperatures of the metal bath, or even at a lower degree. Such indirect chemical actions are very confusing and often lead to incorrect conclusions regarding furnace atmospheres. In the open-flame furnaces used in the melting of bronze, the furnace atmosphere is spoken of as being either neutral, oxidizing or reducing in its effect upon the metal. A neutral atmosphere, as mentioned, is generally considered as the atmosphere of the furnace that will produce melted bronze, neither oxidized nor gassed. It is in effect an atmosphere which is not directly the cause of any chemical action. In such an atmosphere absolutely pure metals may be melted and poured without the absorption of gases or the formation of metallic oxide. The metal has undergone a simple, physical change. An oxidizing atmosphere is one in which the air introduced to the furnace is in excess of that required to complete combustion of the fuel used. The oxygen of the excess air may combine with the metal being melted to form metallic oxide, which may or may not be dissolved by the metal. If an absolutely pure metal is melted in a furnace operating with such an atmosphere, a portion of the metal charged may be converted into metallic oxide, the amount depending upon the duration of the

Jan 1, 1930