Search Documents

By B. A. Wilcox, A. H. Clauer

DURING the course of an investigation on the high-temperature creep behavior of TD Nickel* (Ni + 2) vol pct ThO2), it was observed that the creep fractures were similar in appearance to low-temperature ductile fibrous fractures and creep ruptures by cavitation. Microfractographs of fibrous ruptures are typically characterized by a dimpled appearance, with second-phase particles frequently found near the bottom of the dimples.'-4 These somewhat resemble creep ruptures by cavitation, in that fracture occurs by void formation, with the subsequent growth and final merging of the voids to form a complete fracture.' It is the purpose of this communication to describe the creep fracture of TD nickel and suggest a mechanism for the rupturing process. The discussion is relevant to material crept to fracture in vacuo at a constant stress of 18,000 psi, over the temperature range 900" to ll00°C. The material used in this investigation was 1/2-in.-diam bar stock supplied by Du Pont, contain- ing 2.3 vol pct Tho2. The final fabrication treatment consisted of -95 pct reduction by swaging followed by a 1-hr anneal at 1000°C. Transmission and replica electron microscopy revealed that the material had a fibered structure with a transverse grain size of -1 µ and a longitudinal grain size of 10 to 15 µ. Selected-area diffraction indicated that the fiber axis was parallel to (00l), in agreement with the results of Inman et a1.6 The thoria particle size varied from 80 to 1000Å, with an average size of 400Å, and the mean particle spacing was about 1200Å. The remarkable structural stability7 was confirmed by annealing for 6 hr at 1300°C (about ' 0.9 the absolute melting point). Transmission electron microscopy revealed that the only structural change was a slight decrease in dislocation density. Furthermore, recrystallization or grain growth during creep has not been observed. All specimens were annealed in vacuo at 1300°C for 3 hr prior to creep testing. The following times to rupture were obtained for vacuum creep tests at a constant stress of 18,000 psi: 900°C, 854 hr; 930°C,

Jan 1, 1965

By D. W. Fuerstenau, S. Raghavan

This paper presents a brief summary of some of the thermo- dynamic aspects of flotation processes. Thermodynamic considerations that control interfacial and wetting behavior in mineral/water/air systems are discussed. Particular attention is given to the thermodynamics of collector adsorption.

Jan 1, 1976

By Michael R. Wilder

This paper discusses the two major problems of computerization currently confronting the coal mining industry, specifically, user education and standardization of software practices. Each problem has a unique set of criteria that require addressing individually and collectively. This paper addresses these problems and additionally, a plausible method of overcoming these problems.

Jan 1, 1983

By James E. Dunn

For several years, ground support at the Pitch uranium mine has been a serious headache to the management. The "moving" ground raised havoc with standard timber sets as well as steel sets. The author discusses this problem which has led to the adoption of heavy cribbing practices in drifts and raises.

Jan 6, 1961

William Clinton Brown, Brooklyn, N. Y.: The demand for an apparatus for cooling water for condensers, refrigeratingmachinery and air-compressors, as well as water-jacketed furnaces, has led manufacturers to put such an apparatus on the market, so that it is now possible to purchase a cooling-tower, properly designed and constructed, by simply stating the amount of water to be cooled and the temperatures of the hot and the cold water. The principle is similar to that described by Mr. Henrich, but wood is not used in the best apparatus. There are two great objections to wood: First, it cannot last long under the very trying conditions to which it is subjected; and second, it absorbs the organic matter with which water in constant use becomes impregnated, and the resulting odor is very disagreeable. A few days ago I saw a cooling-tower and condenser, built by the Henry R. Worthington Co., in operation at the Edison electric-lighting station in Brooklyn, N. Y., cooling the water for a 700 horse-power condenser. It consisted of a large stack of iron, about 15 feet in diameter by about 30 feet high. This stack was filled with tiling through which the water falls. A whirligig distributes the water evenly over the top, and a large fan, about 6 feet in diameter, driven by a 10 horse-power electric motor, blows air into the bottom, thus giving a positive circulation. The water, in running down the sides of the tiling, is met by the blast of air, and enough is evaporated to cool the rest. The vacuum-gauge on the condenser showed 26 inches; so the condenser was undoubtedly getting sufficient cool water.

Jan 1, 1896

By Harry Seltz, J. C. DeHaven

InteRmetallic compounds are formed in many binary metal systems. Some compounds are stable to their melting points, and others decompose at lower transition temperatures. Even those of the first class are dissociated in the liquid state, as indicated by the continuous form of their solubility curves at their melting points1. While the lattice structures of some of these compounds have been determined, their thermodynamic properties have not been investigated. In this paper is described the determination of the free energy and heat of formation of the compound CdSb, which is of the first class mentioned above, melting congruently at 455' C. This compound shows a rhombic structure, and the lengths of the edges of the elementary cell containing four molecules are: a = 6.52, b = 8.60, c = 4.16 A.' There are three possible methods of determining the free energy of formation of a compound of this type. The first method, applying the third law of thermodynamics, requires the measurement of its heat capacity preferably to liquid hydrogen temperatures, and a calorimetric determination of its heat of formation. The second method involves the study of the equilibrium of the compound in some gas reaction, such as: CdSb + CO² = CdO + CO + Sb From the known free energies of formation of CdO, CO² and CO the free energy of formation of CdSb could be calculated. The third method measures the electromotive force of a galvanic cell in which the cell reaction is the reversible formation of the compound from the elements. The first method is generally applicable, but requires elaborate experimental equipment. The second method can not be used for compounds melting at low temperatures, owing to the difficulty in establishing equilibrium. The electromotive force method, therefore, seemed to offer

Jan 1, 1935

By J. C. DeHaven, Harry Seltz

InteRmetallic compounds are formed in many binary metal systems. Some compounds are stable to their melting points, and others decompose at lower transition temperatures. Even those of the first class are dissociated in the liquid state, as indicated by the continuous form of their solubility curves at their melting points1. While the lattice structures of some of these compounds have been determined, their thermodynamic properties have not been investigated. In this paper is described the determination of the free energy and heat of formation of the compound CdSb, which is of the first class mentioned above, melting congruently at 455' C. This compound shows a rhombic structure, and the lengths of the edges of the elementary cell containing four molecules are: a = 6.52, b = 8.60, c = 4.16 A.' There are three possible methods of determining the free energy of formation of a compound of this type. The first method, applying the third law of thermodynamics, requires the measurement of its heat capacity preferably to liquid hydrogen temperatures, and a calorimetric determination of its heat of formation. The second method involves the study of the equilibrium of the compound in some gas reaction, such as: CdSb + CO² = CdO + CO + Sb From the known free energies of formation of CdO, CO² and CO the free energy of formation of CdSb could be calculated. The third method measures the electromotive force of a galvanic cell in which the cell reaction is the reversible formation of the compound from the elements. The first method is generally applicable, but requires elaborate experimental equipment. The second method can not be used for compounds melting at low temperatures, owing to the difficulty in establishing equilibrium. The electromotive force method, therefore, seemed to offer

Jan 1, 1935

By Cyril Stanley Smith

The method of thermal analysis, so important in the development of metallographie science, has of recent years been falling into disuse owing to the development of other physical methods which give result's easier of interpretation and which can be used for the study of either equilibrium or transitional conditions because they do not require a continuously changing temperature. Nevertheless thermal analysis remains useful in the study of reactions that take place rapidly, and the simplicity of the experimental technique and the general availability of the apparatus involved are point's in its favor. Theoretically, thermal analysis is capable of giving specific and latent hat measurements and therefore should permit the study of reactions on a more fundamental basis than any other method. It is, however,difficult or impossible to obtain quantitative heat measurements from cooling curves obtained in the usual way. Even if the furnace containing the sample is cooled or heated at a linear rate, or if a gradient furnaec is employed to maintain a constant rate over a large range of temperature, the heat flow from or to the specimen depends on its previous history, for the difference of temperature between the specimen and its surroundings varies. The emissivity of the specimen surface, moreover, is unlikely to be constant. The able mathematical analysis by Russrll,1 although it prevent,s misinterpretat,ion of data, has an effect rather the reverse of that intended, and leaves the reader doubtful of the possibility of ever obtaining quantitative results from thermal curves. Nothing can be done to apply thermal analysis t'o the study of equilibrium conditions of reactions that arc not completed at rates of temperature change above about 1° C;. per minute, and this unfortunatetly excludes most solid transformations in alloys.† For the determination of liquidus

Jan 1, 1940

By Cyril Stanley Smith

The method of thermal analysis, so important in the development of metallographie science, has of recent years been falling into disuse owing to the development of other physical methods which give result's easier of interpretation and which can be used for the study of either equilibrium or transitional conditions because they do not require a continuously changing temperature. Nevertheless thermal analysis remains useful in the study of reactions that take place rapidly, and the simplicity of the experimental technique and the general availability of the apparatus involved are point's in its favor. Theoretically, thermal analysis is capable of giving specific and latent hat measurements and therefore should permit the study of reactions on a more fundamental basis than any other method. It is, however,difficult or impossible to obtain quantitative heat measurements from cooling curves obtained in the usual way. Even if the furnace containing the sample is cooled or heated at a linear rate, or if a gradient furnaec is employed to maintain a constant rate over a large range of temperature, the heat flow from or to the specimen depends on its previous history, for the difference of temperature between the specimen and its surroundings varies. The emissivity of the specimen surface, moreover, is unlikely to be constant. The able mathematical analysis by Russrll,1 although it prevent,s misinterpretat,ion of data, has an effect rather the reverse of that intended, and leaves the reader doubtful of the possibility of ever obtaining quantitative results from thermal curves. Nothing can be done to apply thermal analysis t'o the study of equilibrium conditions of reactions that arc not completed at rates of temperature change above about 1° C;. per minute, and this unfortunatetly excludes most solid transformations in alloys.† For the determination of liquidus

Jan 1, 1940

By N. A. Zeigler

Considerable work has been and is being done on the changes of physical properties that take place in alloys at elevated temperatures, and much information on this subject is published. Much less is known, as pointed out by Rickett and Woodll about reactions between the alloys and atmospheric gases and still less is known abut changes within the alloy itself under the influence of heat and given atmosphere. Moreover, most of the work published has been done on the alloys of "heat-resisting" type, in which chromium is an essential ingredient, the behavior of all other alloys being quite unknown, except perhaps of decarburization produced at the surface by oxidizing atmospheres. It has been shown2 that iron, with total amount of foreign ingredients not exceeding 0.01 per cent, if heated in an oxidizing atmosphere at above 900" C., may be saturated with oxygen to 0.1 per cent. It was suspected that something of a similar nature may take place in iron-silicon alloys (transformer material), the mechanism of such processes being either a direct diffusion of oxygen into the alloy, or oxygen combining with silicon to form SiO2 or more complicated (silicate) inclusions. Another possibility to be considered was that during annealing in oxidizing atmospheres, when scale is being formed, silicon may oxidize preferentially at the surface, thus impoverishing the alloy in silicon. In collaboration with Dr. T. D. YensenP3 the writer set out to investigate this problem about two years ago in order to settle these questions. Testing Procedure In the first part of the investigation, two pieces of 4 per cent silicon iron, prepared in hydrogen atmospheres and forged to 3/4-in. square bar,

Jan 1, 1934

By F. R. Morral, E. P. Miller

Hot-dipped heat-treated zinc coatings on sheet steel were examined with X-rays. The phases identified were compared with those listed in the modern literature 01 the zinc-iron system. The zeta (FeZn13) phase is shown to correspond to the phase found by Schueler (1925) in heat-treated zinc coatings. An investigation was undertaken to determine the phases present in hot-dipped heat-treated zinc coatings on steel sheets and wire. This kind of coating on wire was studied some years ago by Schuelerl by microscopic and chemical methods. He reported a new phase, FeZn10. Later, Clark and Stillwell,2 using X-ray crystal analysis, identified a new phase in ordinary galva. nized wire, which was assumed to be the FeZn10 phase. Since the extensive review of the iron-zinc alloys presented a few years ago,3 much new work has been published, some of which5-l6 was too late to be included in the recent edition (1939) of the American Society for Metals' Handbook.4 Figs. 1 and 2 are reproductions of the latest iron-zinc diagram proposed by Schramml1 as a result of his investigations by microscopic, thermal, magnetic susceptibility and X-ray methods. Table 1 summarizes all the known published X-ray data of the phases Present. The solubility of iron in zinc has been determined by x-ray methods.10 In the same manner the homogeneity range of the P phase was studiedlo using the principle that the parameter of an intermetallic compound changes while the concentration of One of the constituents changes. Schramm10 has found some 80 alpha lines in X-ray powder photograms of phase d1, and from these he believes that this phase is related to the P phase. Previous investi-

Jan 1, 1941

By E. P. Miller, F. R. Morral

Hot-dipped heat-treated zinc coatings on sheet steel were examined with X-rays. The phases identified were compared with those listed in the modern literature 01 the zinc-iron system. The zeta (FeZn13) phase is shown to correspond to the phase found by Schueler (1925) in heat-treated zinc coatings. An investigation was undertaken to determine the phases present in hot-dipped heat-treated zinc coatings on steel sheets and wire. This kind of coating on wire was studied some years ago by Schuelerl by microscopic and chemical methods. He reported a new phase, FeZn10. Later, Clark and Stillwell,2 using X-ray crystal analysis, identified a new phase in ordinary galva. nized wire, which was assumed to be the FeZn10 phase. Since the extensive review of the iron-zinc alloys presented a few years ago,3 much new work has been published, some of which5-l6 was too late to be included in the recent edition (1939) of the American Society for Metals' Handbook.4 Figs. 1 and 2 are reproductions of the latest iron-zinc diagram proposed by Schramml1 as a result of his investigations by microscopic, thermal, magnetic susceptibility and X-ray methods. Table 1 summarizes all the known published X-ray data of the phases Present. The solubility of iron in zinc has been determined by x-ray methods.10 In the same manner the homogeneity range of the P phase was studiedlo using the principle that the parameter of an intermetallic compound changes while the concentration of One of the constituents changes. Schramm10 has found some 80 alpha lines in X-ray powder photograms of phase d1, and from these he believes that this phase is related to the P phase. Previous investi-

Jan 1, 1941

By Lyle L. Marsh, David L. Douglas

CORRELATION was made between the heat treatment and hardness of three U-Ti alloys ranging in composition from 8.5 to 50 atomic pct Ti. The following important observations were made: 1) A direct quench of y U-Ti solid solution resulted in the formation of a hard martensitic phase in low-titanium alloys. 2) Subsequent tempering initially increased the hardness by the precipitation of a fine dispersion of U2Ti; further precipitation of U2Ti with tempering time eventually resulted in softening. 3) High-hardness levels were obtained also by isothermal transformation of an 8.5 atomic pct Ti alloy at 400º and 500°C. 4) The crystal structure of the phases present at the solution-treating temperature (one region in the phase diagram excepted) and the temperature within a given phase region affected the hardness of both quenched and isothermally transformed alloys. Introduction It has been known since 1887 that the mechanical properties of metals were dependent upon their microstructure. Sorby pointed out in that year that annealed gray iron castings were softer than chilled iron castings by virtue of the graphite which had replaced hard cementite in the microstructure. Over the years, interest has increased in the effect of microstructure on the mechanical properties of metals, and now high-strength levels may be achieved by bringing about the dispersion of a second phase or by other microstructural adjustments. The nature of a second phase in an alloy strongly influences the properties. For example, the size, shape, amount, and distribution of a second phase plays a very important role in the ultimate properties of an alloy. Microstructural control is achieved by heat treatments. Hence, any alloy system of potential use should be understood in terms of its response to heat treatment and the resultant effects upon the properties.

Jan 1, 1958

By K. M. Rolls, R. M. Rose, H. B. Shukovsky

Studies of the phase structure, critical current density, and resistive critical magnetic field of a 45 pct Nb-55 pct Zr superconducting alloy after final-size heat treatments are reported in this paper. A combination of optical metallography, X-ray diffraction, and electrical-resistivity measurements were used to determine the phase structure. Superconductivity measurements, made with heat-treated wires, are correlated with microstructure. Although no quantitative relationship between microstructure and critical current density is found, it is shown that SINCE Kunzler1 as well as Berlincourt, Hake, and Leslie2 first described the high-field, high-current-density superconductivity of Nb-Zr alloys, it has be- the critical current density is highest when the microstructure is finely divided by fibering, coring, precipitation, or fine-scale phase dispersions. The resistive critical field is very sensitive to the compositions of the phases present. Since the alloy investigated here is near the maximum resistive critical field for the Nb-Zr system, any phase decomposition results in a lower H,. Elimination of the effects of cold work lowers H, also. A possible interpretation of the latter in terms of internal surface superconductivity is suggested. come apparent that the resistive critical field and the current-carrying capacity of such alloys are, in large measure, dependent on composition and processing. The resistive critical magnetic field (Hr) for low current densities has been found to increase progressively with zirconium content until a flat maximum is reached between 50 and 65 pet Zr.3 For any single composition, a characteristic maximum value of H, is attained by heavily cold-drawn wire.4 critical current density (J,) depends not only upon cold work but also, apparently, upon microstructure. Heat treatments of cold-worked wires generally

Jan 1, 1965

By Ruder, W. E.

CONSIDERABLE interest has been shown lately in various types of furnaces for the production of high, temperatures, both for laboratory purposes and for small industrial uses. Dr. J. A. Harker1 described certain modifications of the Arsem graphite resistance helix for high-temperature work at the National Physical Laboratory. Dr. Northrup ,also has developed two interesting forms of high-temperature furnaces. In 1911, Messrs. Winne and Dantsizen2 described two forms, of resistance furnaces using ductile molybdenum or tungsten as resistors. Since the discovery, by this laboratory, of methods for producing these metals in ductile form, their use as resistance elements in the research laboratory of the General Electric Co. has grown to such an extent that they are now almost indispensable. These furnaces are used here for alloy research, annealing, heat-treating, and practically all f the thermal processes requiring temperatures above 900° C. Various types of furnaces have been developed to meet the general needs of laboratory work, and it is the purpose of this paper briefly to describe some of these furnaces. A tubular furnace wound with a metallic resistor is the simplest type to construct, and almost every laboratory has made such furnaces. The only essential difference between a furnace wound with platinum or base metal and one wound with tungsten or molybdenum is that the coils of the latter must be heated in a neutral or reducing atmosphere. It has long been our custom here to build our resistance furnaces with a metallic casing from which the tube could be easily removed for rewinding. The heat-insulating material-usually calcined magnesia, alumina, silica, or silex, depending upon the nature of the resistor-is used in powdered form so that it can he readily removed and replaced. In order to adapt this method to tungsten or molybdenum windings, it

Jan 2, 1918

By J. P. Neumann, Y. A. Chang

Lattice parameter measurements of the hcp phase in the system Ag-A1 near the stoichiometric composition Ag2Al were carried out for specimens quenched from 825°K, annealed for 10 days at 453°K, and annealed for 100 days at 413°K. The three heat treatments resulted in the same lattice parameters at room temperature. None of the specimens showed any superlattice lines. The lattice parameters of this alloy were also measured at temperature from 77° to 750°K. An anomaly in the temperature variation of the lattice paramete.rs occurs in the region near 400°k, which may be associated with local ordering. FROM measurements of the diffuse scattering of X-rays, it is known that the [ phase in the system Ag-A1 near the stoichiometric composition Ag2A1 ex- hibits a considerable degree of short-range order at room temperature.' Moreover, based on the position of the short-range order peaks a superlattice model had been proposed in the same study and it was suggested that lowering the annealing temperature to 420°K might result in long-range order. In the present study, the diffraction patterns and the lattice parameters of three differently heat-treated specimens, i.e., quenched from 825°K, annealed for 10 days at 453°K, and annealed for 100 days at 413°K, were determined from 77" to 750°K. The purpose of these measurements was to investigate 1) whether long-range order could be established by prolonged annealing and 2) whether the lattice parameters at room temperature depended on the heat treatment, since it had been proposed earlier2 that the lattice parameters in this phase are related to the atomic arrangement. 1) EXPERIMENTAL TECHNIQUES Silver and aluminum, both 99.99 pct purity, were melted in a zirconium oxide crucible under helium in

Jan 1, 1969

By Henry M. Howe

The transformation which steel undergoes in slow cooling, from the condition of austenite whelk above the transformation rage into that of pearlite plus either ferrite or cementite below that range, is subject to great lag. This transformation is essentially the pearlitizing of the austenite. Indeed the hardenillg of steel is due to this lag, which is made so great by extremely rapid cooling that a very large part of the transformation itself is restrained, so that the rapid cooling catches the metal in an only partly transformed state, that of martensite, and under favorable contditions even in an almost wholly untransformed state, as in the cases of Maurer's austenitic manganese steel. Hadfield's austemtic. manganese steel, 25 per cent. nickel steel, and other anstenitic alloy steels, arc austenitic even after slow cooling. It has long been known that this lag increases (1) with the rapidity of cooling, (2) with the presence of certain retarding elements, notably manganese and nickel, and (3) with the temperature from which cooling starts. The present inquiry touches the last of these influences, but incidentally it throws light on the first also. This influence of higher heating ill increasing lag manifests itself in different ways which, though they may at first seem unrelated, are in fact hut different aspects of one and the same thing. Thus, among heatings to different high temperatures, all of them above the transformation range: the higher the temperature to which the steel is heated (1) the lower is the temperature at which the transformation occurs in cooling slowly down, (2) the shorter is the time occupied in the actual cooling from above the transformation range to the room Temperature under given external conditions, Whether of slow or

Jan 1, 1914

By Henry Howe

(New York Meeting, February, 1913.) THE transformation which steel undergoes in glow cooling, from the condition of austenite when above the transformation range into that of pearlite plus either ferrite or cementite below that range, is subject to great lag. This transformation is essentially the pearlitizing of the austenite. Indeed the hardening of steel is due to this lag, which is made so great by extremely rapid cooling that a very large part of the transformation itself is restrained, so that the rapid cooling catches the metal in an only partly transformed state, that of martensite, and under favorable conditions even in an almost wholly untransformed state, as in the cases of Maurer's austenitic manganese steel. Hadfield's austenitic manganese steel, 25 per cent. nickel steel, and other austenitic alloy steels, are austenitic even after slow cooling. It has long been known that this- lag increases (1) with the rapidity of cooling, (2) with the presence of certain retarding elements, notably manganese and nickel, and (3) with the temperature from which cooling starts. The present inquiry touches the last. of these influences, but incidentally it throws light on the first also. This influence of higher heating in increasing lag manifests itself in different ways which, though they may at first seem unrelated, are in fact but different, aspects of one and the same thing. Thus, among heatings to different high temperatures, all of them above the transformation range, the higher the temperature to which the steel is heated (1) the lower is the temperature it which the transformation occurs in cooling slowly down, (2) the shorter is the time occupied in the actual cooling from above the transformation range to the room temperature under given external conditions, whether of slow or

Jan 3, 1913

By W. E. Ruder

Considerable interest has been shown lately in various types of furnaces for the production of high temperatures, both for laboratory purposes and for small industrial uses. Dr. J. A. Harkerl described certain modifications of the Arsem graphite resistance helix for high-temperature work at the National Physical Laboratory. Dr. Northrup also has developed two interesting forms of high-temperature furnaces. In 1911, Messrs. Winne and Dantsizen2 described two forms of resistance furnaces using ductile molybdenum or tungsten as resistors. Since the discovery, by this laboratory, of methods for producing these metals in ductile form, their use as resistance elements in the research laboratory of the General Electric Co. has grown to such an extent that they are now almost indispensable. These furnaces are used here for alloy research, annealing, heat-treating, and practically all of the thermal processes requiring temperatures above 900' C. Various types of furnaces have been developed to meet the general needs of laboratory work, and it is the purpose of this paper briefly to describe some of these furnaces. A tubular furnace wound with a metallic resistor is the simplest type to construct, and almost every laboratory has made such furnaces. The only essential difference between a furnace wound with platinum o base metal and one wound with tungsten or molybdenum is that the coils of the latter must be heated in a neutral or reducing atmosphere It has long been our custom here to build our resistance furnaces with : metallic casing from which the tube could be easily removed for rewind ing. The heat-insulating material—usually calcined magnesia, alumina silica, or silex, depending upon the nature of the resistor-—is used i: powdered form so that it can be readily removed and replaced. I order to adapt this method to tungsten or molybdenum windings,

Jan 1, 1918

By J. M. Hartman

ONE of the most important subjects to the blast-furnace engineer is a thorough knowledge of the conditions affecting the temperature in the different portions of the furnace. All efforts to decrease the consumption of fuel and improve the working of the furnace must be based upon it, and I may, therefore, be permitted to place before you the results of my observations, and detail the conclusions to which they have led me. Tracing the thermic conditions from below upwards we have at the bottom of a blast furnace making No. 3 iron a temperature of 2900' F., which increases slightly to a point a little below the tuyeres. In the immediate vicinity of the tuyeres the temperature is somewhat lower, owing to the entering blast; but a short distance above the tuyeres, where all the oxygen of the blast has been converted into carbonic acid, the highest temperature in the furnace must be attained. This carbonic acid is, however, almost as soon as formed, converted by the glowing coal into carbonic oxide, a process which absorb, heat, reduces the temperature, and provides the active agent for the reduction of the iron ores. In its ascent the hot carbonic oxide gradually parts with its heat, first melting the descending iron and earthy materials, which drop down to the hearth of the furnace. The limit of this zone of fusion is rather sharply defined, and the temperature above this zone suddenly decreases. In its further ascent the hot gas drives off the carbonic acid from the limestone, which causes a further absorption of heat, and finally the gas escapes at the top, at a ternperature of about 250' F. if the furnace is working well. These changes of temperature I have attempted to represent graphically,

Jan 1, 1880