Search Documents

By Jerome Strauss, Carl L. Shapiro

THIS is a critical examination of the theory that the amount of carbon necessary to form the iron-carbon (carbide) eutectoid is lowered by the addition of any carbide-forming element. Although this theory has been built up over a long period of years,, it appears, in the light of recent published investigations upon the ternary constitutional diagrams of iron with carbon and other alloying elements, to be no longer tenable. A survey of these new results, summarized below, shows that (as is known) some metals lower the carbon content of the iron-carbon (carbide) eutectoid, whereas others raise it, but (as not heretofore clearly set forth) some first move the eutectoid in one direction and then, at higher concentrations, shift it in the opposite direction. Moreover, these differences in behavior exist even among the common alloying metals classed as carbide formers. PART I INFLUENCE OF BODY-CENTERED CUBIC ELEMENTS The effects of the various body-centered cubic metals (chromium, tungsten, molybdenum, vanadium, columbium and tantalum) are reviewed individually. Iron-carbon-chromium System.-In the monograph on Alloys of Iron and Chromium, Kinzel and Crafts' concluded that the carbon content of the eutectoid is lowered by increasing the chromium content. The quantitative influence of chromium on the eutectoid point of iron-carbon alloys was shown by Monypenny2 (Fig. I) and no data have been forthcoming to refute his findings. Iron-carbon-tungsten System.-Gregg,3 in the monograph on Alloys of Iron and Tungsten, accepted the data showing that tungsten decreases the amount of carbon required to form the eutectoid. The influence of tungsten on the eutectoid percentage of carbon as determined by Oberhoffer and Daeves4 is shown in Fig. a; these results have received general approval. Iron-carbon-molybdenum System.-In the monograph on Alloys of Iron and Molybdenum, Gregg5 reviewed the work of Guillet,6 Swinden,7 and Reed,8 all showing that molybdenum lowered the carbon content of the eutectoid, whereas Takei9 contradicted this conclusion. Svetchnikoff and Alferova,10 a little later, showed that large additions of molybdenum shift the eutectoid to the right. These last-named investigators found that molybdenum, up to 1.2 per cent, shifts the pearlite point and the point of maximum solubility of cementite in the gamma phase only slightly in the direction of lower carbon content; on the other hand, increasing amounts of molyb-

Jan 1, 1943

By J. J. Egan, A. B. Kinzel

Much work has been done recently in an attempt to analyze the physicochemical mechanism involved in the production of steel by the open-hearth process. This has resulted in reducing the process to a series of interrelated phenomena obeying the usual laws of mass action and diffusion. Of these, the reaction of the iron oxide and carbon in the molten steel is of most interest from a technical point of view as it constitutes the main action in the usual steelmaking processes. A. L. Feild' has formulated a basic differential equation expressing the open-hearth reactions and Herty has done considerable experimental work on the determination of a number of constants involved in the basic open-hearth reaction. In considering the application of the law of mass action to the reaction FeO + C + CO + Fe, the iron will be taken as constant, as is the case in the open hearth and the carbon monoxide pressure will be assumed constant at one atmosphere. From this xy = m at equilibrium, where x is the percentage weight of carbon, y percentage weight of iron oxide and m the weight-concentration equilibrium constant. While it is customary for physical chemists to express equilibrium constants in terms of molar concentrations, the constant is here expressed as the product of the percentage-weight concentrations for ease of comparison with the constants quoted by Herty and by Beild, and for more ready applicability to the open-hearth reactions. Herty and his associates have attempted to calculate the value of this constant from data on open-hearth heats. So many variables affect these results that an adequate calculation presents very great difficulties. A value of 4 X lop2 was first arrived at, but this was later changed to 1.6 X 10-2. Feild, in order-to illustrate the operation of the differential equation of the open hearth, has assumed a value considerably lower than Herty's, namely, 1.0 X 10-2, the largest value consistent with certain theoretical

Jan 1, 1929

By I. N. Knapp

Introduction.-The special object of these notes is to describe the mixing, testing, and use of mud-ladened water for rotary drilling in such a way as to make them helpful to the driller, the operator, or the engineer in solving his own special drilling problems. The structures, apparatus and tools used are indicated in a general way. No attempt is made to describe the art of rotary drilling; only such description as are necessary to make plain the use of mud are given. The information is the result of actual experience in drilling in Coastal Plain formations. The materials encountered in the wells drilled were unconsolidated sands, gravels, and clays, in which thin layers of sandstones, shell conglomerates, and shales began to appear at about 1,200 ft. in depth, although one well was drilled to 3,018 ft. without encountering any cemented or indurated materials. Unusual Conditions.-The general surface of the ground where the drilling was done being hardly a foot above mean tide, and the daily tidal variation being about 15 in., it was found necessary to plank over the sod surface with 3-in. plank to work from. A 30 by 30 ft. planked area was sufficient to carry the derrick with equipment and 2,500 ft. of 4-in. drill pipe stacked in it. The derrick was set up 3 ft. on cribbed blocking arranged to spread the load over the planking. The engine, boilers, and pipe yard were also carried on 3-in. planking. Drilling Outfit.-This consisted of a derrick, 20 by 20 ft. and 84 ft. high, two 35 h.p., "oil country" boilers, one double 8 ¼ by 10 in. reversible engine, two 10 by 6 by 12 in. duplex mud pumps, one 15-in. rotary, "hoisting works" with chain drive, hoisting block with line, crown block with pulleys, rotary' jetting swivels, hose, complement of tools, pipes and connections, and. a mud mixer, with engine.

Jan 12, 1914

By Carl L. Shapiro, Jerome and Strauss

This is a critical examination of the theory that the amount of carbon necessary to form the iron-carbon (carbide) eutectoid is lowered by the addition of any carbide-forming element. Although this theory has been built up over a long period of years, it appears, in the light of recent published investigations upon the ternary constitutional diagrams of iron with carbon and other alloying elements, to be no longer tenable. A survey of these new results, summarized below, shows that (as is known) some metals lower the carbon content of the iron-carbon (carbide) eutectoid, whereas others raise it, but (as not heretofore clearly set forth) some first move the eutectoid in one direction and then, at higher concentrations, shift it in the opposite direction. Moreover, these differences in behavior exist even among the common alloying metals classed as carbide formers. PART I Influence of Body-centered Cubic Elements The effects of the various body-centered cubic metals (chromium, tungsten, molybdenum, vanadium, columbium and tan. talum) are reviewed individually. Iron-carbon-chromium System.—In the monograph on Alloys of Iron and Chromium, Kinzel and Crafts1 concluded that the carbon content of the eutectoid is lowered by increasing the chromium content. The quantitative influence of chromium on the eutectoid point of iron-carbon alloys was shown by Monypenny2 (Fig. I) and no data have been forthcoming to refute his findings. Iron-carbon-tungsten System.—Gregg,3 in the monograph on Alloys of Iron and Tungsten, accepted the data showing that tungsten decreases the amount of carbon required to form the eutectoid. The influence of tungsten on the eutectoid percentage of carbon as determined by Oberhoffer and Daeves4 is shown in Fig. 2; these results have received general approval. , I ron-carbon-molybdenum System.—In the monograph on Alloys of Iron and Molybdenum, Gregg5 reviewed the work of Guillet,6 Swinden,7 and ReedI8 all showing that molybdenum lowered the carbon content of the eutectoid, whereas Takei9 contradicted this conclusion. Svetchnikoff and Alferova,10 a little later, showed that large additions of molybdenum shift the eutectoid to the right. These last-named investigators found that molybdenum, up to 1.2 per cent, shifts the pearlite point and the point of maximum solubility of cementite in the gamma phase only slightly in the direction of lower carbon content; on the other hand, increasing amounts of molyb-S

Jan 1, 1944

By Jerome and Strauss, Carl L. Shapiro

This is a critical examination of the theory that the amount of carbon necessary to form the iron-carbon (carbide) eutectoid is lowered by the addition of any carbide-forming element. Although this theory has been built up over a long period of years, it appears, in the light of recent published investigations upon the ternary constitutional diagrams of iron with carbon and other alloying elements, to be no longer tenable. A survey of these new results, summarized below, shows that (as is known) some metals lower the carbon content of the iron-carbon (carbide) eutectoid, whereas others raise it, but (as not heretofore clearly set forth) some first move the eutectoid in one direction and then, at higher concentrations, shift it in the opposite direction. Moreover, these differences in behavior exist even among the common alloying metals classed as carbide formers. PART I Influence of Body-centered Cubic Elements The effects of the various body-centered cubic metals (chromium, tungsten, molybdenum, vanadium, columbium and tan. talum) are reviewed individually. Iron-carbon-chromium System.—In the monograph on Alloys of Iron and Chromium, Kinzel and Crafts1 concluded that the carbon content of the eutectoid is lowered by increasing the chromium content. The quantitative influence of chromium on the eutectoid point of iron-carbon alloys was shown by Monypenny2 (Fig. I) and no data have been forthcoming to refute his findings. Iron-carbon-tungsten System.—Gregg,3 in the monograph on Alloys of Iron and Tungsten, accepted the data showing that tungsten decreases the amount of carbon required to form the eutectoid. The influence of tungsten on the eutectoid percentage of carbon as determined by Oberhoffer and Daeves4 is shown in Fig. 2; these results have received general approval. , I ron-carbon-molybdenum System.—In the monograph on Alloys of Iron and Molybdenum, Gregg5 reviewed the work of Guillet,6 Swinden,7 and ReedI8 all showing that molybdenum lowered the carbon content of the eutectoid, whereas Takei9 contradicted this conclusion. Svetchnikoff and Alferova,10 a little later, showed that large additions of molybdenum shift the eutectoid to the right. These last-named investigators found that molybdenum, up to 1.2 per cent, shifts the pearlite point and the point of maximum solubility of cementite in the gamma phase only slightly in the direction of lower carbon content; on the other hand, increasing amounts of molyb-S

Jan 1, 1944

By A. B. Kinzel

MUCH work has been done recently in an attempt to analyze the physicochemical mechanism involved in the production of steel by the open-hearth process. . This has resulted in reducing the process to a series of interrelated phenomena obeying the usual laws of mass action and diffusion. Of these, the reaction of the iron oxide and carbon in the molten steel is of most interest from a technical point of view as it con-stitutes the main action in the usual steelmaking processes. A. L. Feild1 has formulated a basic differential equation expressing the open-hearth reactions and Hefty has done considerable experimental work on the determination of a number of constants involved in the basic open-hearth reaction. In considering the application of the law of mass action. to the reaction FeO +C ?? CO +Fe, the iron will be taken as constant, as is the case in the open hearth and the carbon monoxide pressure will be assumed' constant at one atmosphere. From this xy = m at equilibrium, where x is the percentage weight of carbon, y percentage weight of iron oxide and m the weight-concentration equilibrium constant. While it is customary for physical chemists to express equilibrium constants in terms of molar concentrations, the constant is here expressed as the product of the percentage-weight concentrations for ease of comparison with the constants quoted by Herty and by Feild, and for more ready applicability to the open-hearth reactions. Herty and his associates have attempted to calculate the value of this constant from data on open-hearth heats. So many variables affect these results that an adequate calculation presents very great difficulties. A value of 4 X 10-2 was first arrived at, but this was later changed to 1.6 X 10-2. Feild, in order to illustrate the operation of the differential equation of the open hearth, has assumed a value considerably lower than Herty's, namely, 1.0 X 10-2, the largest value consistent with certain theoretical

Jan 1, 1929

By David Segura

THE Huitzuco mercury deposits have been described by C. W. Vaupell,1 therefore the general information such as location, history, production, etc. will not be repeated here. The present owners acquired the property when the deep levels of the Cruz mine were flooded, and it was believed that a relatively large tonnage of low-grade ore and fills would be encountered in the flooded section, because of the early mining methods used in the Cruz mine. The mine was partly unwatered and about 187,000 tons of fills was pit-sampled in the San Cayetano stope. These fills average 2.4 kg. Hg per ton. During the period of maximum production at Huitzuco, the ores were sorted in the mine and cobbed into 3-in. pieces, which were treated in Scott furnaces. Large pieces of raw ore are found in the sinter, and it is indicated that both recovery and costs were not very satisfactory. Recent practice has been to treat the ores in closed retorts made of 10-in. pipes. Because of the presence of combined antimony in the mineral livingstonite (HgS.2Sb2S3) and considerable native sulphur, little or no metallic mercury is recovered in the condensers and a second retorting of the condenser muds is necessary. This makes the treatment of the ores expensive and tends to low recoveries of mercury. The principal costs of retorting are, in their relative order: (1) fuel, (2) low recovery of mercury, (3) replacement of retort pipes, (4) labor. METHODS TESTED Three series of experiments have been made, with the object of reducing the costs of recovering mercury and antimony from the ores, by: (1) wet metallurgical methods, (2) bulk roasting in large mechanical furnaces, (3) concentration to reduce the bulk to be followed by roasting in some mechanical furnace. The wet metallurgical methods did not appear feasible because of the high cost of the alkaline solvents used, and investigations of these were not carried beyond the preliminary stage.

Jan 1, 1938

By A. W. Schlechten, Ernest J. Breton Jr.

Experiments on the separation of copper and zinc ions by selective action of ion exchange resins showed the carboxylic type to be more effective than the sulphonic resins. The latter demonstrated a greater capacity over a wider pH range. Data show the effectiveness of resins as a means of concentration. IN recent years the restrictions of stream pollution laws and the high price of metals have created an interest in ion exchange as a means for metal recovery. Some applications have proved successful. In Germany during World War 11, 17 tons of copper per day were recovered from rayon mill wastes by means of ion exchange resins;' and for some time in this country a large ion exchange unit has been in operation for the recovery of copper from rayon waste water. The possibilities of applying ion exchange to the recovery of metals occurring in plating rinse water is particularly promising. In most of these applications only the metal being recovered occurs in the waste. The ion exchange resins act merely as a means of concentrating the metals to a point where they can be recirculated. It would be highly desirable to use ion exchange as a means of not only concentrating but also of separating metals. With the exception of the impressive separations accomplished in connection with the atomic energy program, very little has been done on metal separations.' Therefore, an investigation was undertaken at the Missouri School of Mines and Metallurgy to determine if either of the two main types of ion exchange resins could be used to separate metal ions in solution. The selective removal of copper ions from a mixture of copper and zinc on carboxylic and sulphonic-type resins was investigated as a function of flow rate, pH, copper-zinc ratio, and concentration. It was shown that zinc can be separated from copper and that very large ratios of concentration can be obtained using ion exchange resins. Since ion exchange is relatively new to the field of metallurgy, a brief review of the subject will be included. Theory of Ion Exchange A comprehensive theory for ion exchange has not been developed as yet, but the mechanisms are analogous to metathetical reactions: R Na + Cu++ *=? K(SO3)2 Cu + 2Na+ R is the designation for the ion exchange resin. If a copper solution is passed over a resin bed in the sodium form, two ions of sodium will be released for every ion of copper removed. For the most part this reaction follows the laws of mass action and of electrical neutrality. Consequently, if an excess of sodium ions is passed over a bed containing copper, the reactions will be reversed, and the resin will be regenerated to its original form. A few empirical rules governing the exchange reaction have been set forth: 1—In general ions with a high valence will replace ions with a lower valence. 2—Ions having higher activity coefficients have a higher replacement potential. 3—In a series of mono-valent ions, those with the smallest radii of hydra-tion will tend to replace those having larger radii of hydration. 4—Where ions are similar in most respects, those with the higher atomic weight sometimes will take precedence. This last rule is not as definite as some of the others. These rules apply to rather dilute solutions at moderate temperatures and assume all ions to be present in about equal concentrations. Higher concentrations and temperatures may in some cases reverse the normal exchange reactions. Ion exchange materials are unique in that their efficiency increases as the concentration of the solution decreases. For many exchangers, most efficient operation is obtained at concentrations in the order of one thousandths of a percent. Most applications, though, are made in solutions containing considerably higher concentrations than this. Coste9 as shown that ion exchange resins will remove aluminum and iron effectively' from solutions of up to 10 pct chromic acid. Ion Exchange Resins Ion exchange resins are insoluble, porous, resinous structures to which active groups have been attached. Active groups such as (—SO,,)- and (COO)- pick up cations; hence structures saturated with groups such as these are called cation exchangers. Structures saturated with groups such as (—NH,)' which pick up anions, are referred to as anion exchangers. The resinous structure of necessity is resistant to strong acids, bases, oxidizing, and reducing agents, and most of the common organic solvents. An idea of the stability can be gaged from the fact that resins last for many years under constant use without detectable chemical or physical breakdown. The ion exchange reaction is not confined to the surface of these synthetic resins. Its porous structure permits active groups in the center of a particle as well as those on the surface to remove ions. A high capacity resin such as Amberlite IR-120 will remove up to 3.3 lb Cu per cu ft of resin. In this investigation several approaches to the problem of separating copper from zinc by ion exchange were considered. First, if a reagent could be found which would complex one of these metals and not the other, then by passing this reagent through a bed of exchanger containing copper and zinc, the

Jan 1, 1952

By W. L. DeMoulin

A rather detailed description of the entire plant and leaching process was given in a paper recently presented to the Institute,I so this paper will cover briefly only the crushing practice of the New Cornelia Copper Co. for the year 1918. The ore, which is mined by steam shovels and loaded in side-dump cars, passes through two crushing plants. These crushing plants were constructed to crush the ore required by a leaching plant of 5000 tons daily capacity, during a crushing period of 16 hours. The primary plant reduces to less than 3 in. (76 mm.) and the secondary plant, to 1/4 in. (6.35 mm.), which is sufficiently fine for satisfactory percolation of solution through the ore in the leaching tanks, and to give-good extraction. The ore is then conveyed to the leaching tanks by a system of 28-in. (71-cm.) belt conveyors. On the way to the leaching tanks, it passes through an automatic sampling plant, where a 1-per cent. sample is taken. From the primary crushing plant, the ore is conveyed by two 36-in. (91-cm.) belt conveyors to a 10,000-ton steel storage bin, with a reinforced-concrete flat bottom. The ore discharges from the bottom of this' bin onto four 20-in. belt conveyors, which deliver it to the four units of crushers in the secondary crushing plant. There is no storage bin between the mine and the primary crushing plant. The ore breaks very coarse, is very hard, and contains a great many boulders, some as large as 4 1/2 by 4 1/2 by 10 ft. (1.3 by 1.3 by 3 m.). Jams form in the bowl of the coarse crusher in the primary crushing plant, as a consequence, and are freed by means of an immense steel hook operated from a 40-ton electric traveling crane. Primary Crushing Plant The primary plant consists of one No. 24 Gates, style K, gyratory crusher, followed by four Gates No. 8, style K, gyratory crushers. All

Jan 1, 1920

By J. B. Seabrook

The accompanying Fig 1 is a graphical scheme of intermediate accuracy for expediting interconversions of mol, weight, and volume percents. This chart consists of a family of curves of mol or atomic or volume percent plotted against weight percent, each curve in the family providing for a certain ratio of molecular or atomic weights or densities. Thus, the whole range of variations is presented in a very familiar format, and the conversion is effected simply by choosing the proper curve and reading off the required values. An entire series of conversions for a given binary system can be found by following the appropriate curve. It is interesting to examine the analytical geometry of the plots. The curves, which must fit into a square and all join at the 0 and 100 pet points, may be shown to be hyperbolas with their centers on the extension of the diagonal of the square which they intersect. Multiplying out the expression x A atomic fraction of A = y = ------ where x is the weight fraction of A, and A and B are atomic weights, we get xy+A/By- A/Bxy=x Upon rotating and translating axes and avoiding imaginaries, it is found that the angle of rotation is —45°, the origin moves upv2/2 units andv2/2(A/B+1/A/B-1) units to the right, and the standard form is Thus the curves are equilateral hyperbolas, their centers are units away from their intersection with

Jan 1, 1950

By Thomas T., Read

E DUCATION for the mineral industry was at first a single comprehensive curriculum, but it was early recognized that the main basis of mining is physics, while that of metallurgy is chemistry. The first school to offer curricula leading to degrees in mineral industry education in the United States (1853) had separate ones in mining and metallurgy. The Columbia School of Mines (the first really successful one) had only a single curriculum when it opened in 1864, but within a few years had so reorganized its work that it offered curricula in mining, metallurgy, geology, chemistry, and civil engineering. Interest in the metallurgy or geology curricula was slight at first, and for the next 30 years the great majority of Columbia students took the mining curriculum, probably partly because it contained nearly as much geology as the geology curriculum, practically all the metallurgy, and most of the chemistry, and partly because of the degree awarded for it. Other schools kept their mining curricula equally broad, and as late as 1900-1910 men were graduating as mining engineers who were practically as well prepared to make geology or metallurgy their life work, as they were for mining. Gradually, however, the subject matter of mining geology and metallurgy became so enriched through research and development that separate curricula in them became worth while; their progressive development is the subject of another chapter. Mining curricula also began to be somewhat diversified, and schools in areas where coal mining was of outstanding importance began to offer options in coal mining, the earlier curricula having been chiefly aimed at metal mining. The petroleum industry began its meteoric rise about the same time as mineral instruction in this country, but little more of applied science attended its birth than had attended such ancient practical arts as copper and iron

Jan 1, 1941

By W. B. Daly, A. S. Richardson

The application of artificial refrigeration, or air-conditioning, to the ventilation of deep, hot mines has long been a subject of interest to the operators of such properties. Artificial cooling of the air in theaters, railroad coaches, and various industrial plants has been so successful within recent years that it has been natural to assume that the adoption of similar methods would afford a ready solution to the difficult problem of mine ventilation. It should however, be obvious, that although there may be nothing technically impossible in the application of this process to mine ventilation there is a great economic distinction between the maintenance of a high standard of comfort for a large number of people in a relatively small space such as a theater and the cooling of air to improve working conditions for miners scattered throughout the extensive workings of a large mine. Many methods of air-conditioning have been considered by the management of the Anaconda Copper Mining Co. at various times, but until the development of the process to be described in this paper none has seemed economically practicable. The Ventilating Problem at Butte The main obstacle to satisfactory ventilation in the Butte mines is the normally high temperature of the granite country rock. In the present operating zone, rock temperatures as high as 120' F. have been noted, and in some parts of the district the rate of increase of temperature with depth has ranged as high as 3.5" F. per 100 ft. Other factors, such as the decay of mine timber and the oxidation of sulfide minerals in the ore bodies, contribute to raise the temperature of the air in the working places but, particularly in the development of the mines to greater depth, the naturally high temperature of the rock is of predominant importance. In few other places, probably, are the ordinary ventilating measures more hindered than at Butte. At an operating depth now exceeding 3000 ft., all of the air required for ventilation must pass through timbered shafts which, because of heavy ground, are of small size and offer great resistance to the flow of air. Shaft compartments are rarely more than 4.5 X 5 ft., and the shaft area open to the flow of air usually does not exceed 65 sq. ft.; in only two shafts is the area as great as 85 sq. feet.

Jan 1, 1934

By W. J. Babyak, F. N. Rhines

It has long been supposed that, regardless of the mechanism by which grain boundaries increase the hardness of a ductile metal, the effect should be linearly proportional to the grain boundary area. At the present time two methods for determining grain boundary area in an isotropic aggregate are available. The first, derived by Angus and Summers,' presumes that the grain boundary area is equal to 3vn, where n is the number of grains per sq mm in a random section through the metal. This estimate is based upon the assumption that the grains are equivalent to cubes whose faces correspond with the grains as observed in the micro-section. Measurements on copper, bronze, and a brass showed that a nearly linear relation exists between 3vn and hardness. This is illustrated, for the case of brass, in Fig. 1. The second method, developed independently by Smith and Guttman2 and by Duff in, Meussner, and Rhines,3 is absolute and employs a simple linear intercept count for measuring grain boundary area. It involves no assumptions with respect to the shape and distribution of the grain boundaries. The total surface in unit volume is 2 NL , where NL is the number of intercepts between grain boundary and unit length of a test line passed randomly through the sample. In practice, NL,, the average number of grain boundaries intercepted by unit length of traverse across a prepared microsection, is ob-

Jan 1, 1961

By Edwin H. Olson, Morton Smutz, Charles J. Baroch

IN 1949 an orebody containing some 10 billion lb of recoverable rare earth metals was discovered in the Mountain Pass district of San Bernardino County, California.1 The following year Molybdenum Corp. of America purchased a number of claims at the deposit and a mill on the property which had been constructed to process the district's gold ores. Average composition of ore from the San Bernardino deposit is less than 0.1 pct ThO2, 25 to 35 pet calcite, 10 pet rare earth oxides, 15 to 20 pet silica, and 30 to 40 pet barite.2 Rare earth content of the ore analyzes 50.7 pet CeO2, 4.2 pet Pr6O11, 11.7 pet Nd2O3, 1.3 pet Sm2O3, and 34.3 pet La2O3, and others.2

Jan 3, 1959

By R. L. Perrine

A stability theory is developed for miscible liquid-liquid displacement within a porous medium. In the usual case considered, a high-density high-viscosity "oil" is displaced downdip by a low-density low-viscosity "solvent'! Perturbation methods are used to find the conditions under which the spreading mechanism changes from the stable dispersion process to unstable viscous fingering. We find that instability is conditional, that there is a dependence on the shape of a disturbance leading to a"diameter" effect and that very difficult experimental scaling problems may result. A useful consequence is the definition of a minimum "slug size" for stable miscible displacement. This should make possible optimum use of the solvent process for oil recovery. The results may apply to many situations in which one fluid displaces another of somewhat different fluid properties within a porous medium. INTRODUCTION The process of miscible liquid-liquid displacement looked very favorable when it first received serious consideration as a means to increase petroleum recovery. Some early laboratory results were interpreted to mean that only a small "slug" of solvent was needed, perhaps 2 to 3 per cent of a hydrocarbon pore volume. This "slug" could recover the oil in an entire reservoir provided the fluids remained miscible.1,2 Only a slow growth of the "mixing zone", or gradual transition from oil to solvent, should occur. This would follow naturally from mixing by a dispersion mechanism such as that described by Scheidegger.3 And, indeed, laboratory cores have shown this kind of behavior provided solvent viscosity was at least as great as that of the oi1.2,4-6 The same kind of behavior has also been observed with solvent viscosity lower than that of the oil after the distance traversed became large. Not all laboratory results have been this favorable, however.7-9 Mixing zone growth may become proportional to the distance traveled rather than the square root of this quantity. Such behavior accompanies high rates in short systems of large diameter, provided the viscosity ratio is adverse. If this more rapid spreading were to persist in a reservoir, the solvent requirement for miscible "slug" displacement could exceed 30 per cent of a pore volume. There is considerable economic importance in the difference between 3 and 30 per cent solvent. A 1ogical conclusion is that two different kinds of flow behavior are possible, with each kind leading to a different result. One kind gives efficient displacement, the other does not. In the efficient case, the solvent bank spreads out only by the mechanism we have termed dispersion. Under these conditions, displacement is as near piston-like as possible. In the second kind of flow, found only with adverse viscosity ratios, viscous fingering occurs. That is, permeability variations cause a small finger of low-viscosity solvent-rich fluid to move ahead of its average position within the mixing zone. This creates a path of low resistance to flow, and an even greater amount of solvent-rich fluid follows. Thus, the process is autocatalytic. Once started, the fingering mechanism rapidly becomes dominant. The question to be answered is this. Under what conditions will each of these two different kinds of flow occur? In particular, what conditions denote the transition from dispersion to viscous fingering as the solvent spreading mechanism? The answer may tell us whether or not the desirable, near piston-like miscible displacement process is practical. An answer to this problem can be obtained from theory by the use of perturbation methods. The procedure is as follows. We first formulate a mathematical representation of the system. Then a small disturbance (or perturbation) in solvent concentration profile is introduced and observed to see what happens. Unstable flow is indicated when a small disturbance will grow larger. This will lead to eventual viscous fingering. If on the other hand the disturbance dies out, the displacement is stable. In this latter case, with dispersion as the spreading mechanism, near piston-like displacement is possible. Thus, this paper presents the development of a stability theory for miscible liquid-liquid displace-

By R. J. Piros, Brusenback, D. A. Dahlstrom

DEWATERING fine coal has been a serious problem to many operators who desired to wet-wash finer sizes and maintain high recovery. Centrifugal driers have become popular for this purpose for sizes between 3/8 in. and 14 mesh. Below 14 mesh the perforated screen centrifuges lose a large per- cent in the effluent solid bowl types that do retain -14 mesh coal, produce excessive degradation, tend to retain the highest ash fraction in any size fraction, and are subject to abrasion, which results in higher maintenance. In 1949 Truax-Traer Coal Co. planned a cyclone installation at their Ceredo preparation plant to recover coal down to 150 mesh with approximately a 14-mesh top size. Since available centrifugal equipment was not considered satisfactory, an investigation of rotary vacuum filters was begun in cooperation with the Peterson Filters and Engineering Co. to determine the suitability of the Peterson Top Feed Reservoir (TFR) filter. This investigation, which is described below, has resulted in a highly satisfactory installation of two 6x10-in. Peterson TFR filters that have been in operation over a year. After some preliminary laboratory batch vacuum filter tests, a pilot plant double-drum Peterson TFR filter was installed at Ceredo to prove the feasibility of vacuum filtration and determine optimum conditions. The full scale units were then installed on the basis of the pilot plant results. This was followed by an extensive research program on vacuum filtration in the laboratory at Northwestern University to investigate basic design and operating variables and their influence on final moisture con- tent of the filter cake. A method of scale-up from laboratory batch tests to the full-size unit was also proposed and proved. Because of the importance of dewatering fine coal today, it was felt that the results of the three steps should be made available for the use of others facing this problem. The work at Northwestern is given in a theoretical section following the discussion of the pilot plant and full- scale operation. Fig. 1 is a flow diagram of the cyclone and filter installation. The feed is drawn from below the

Jan 1, 1952

By E. Brunner, E. S. Mardock

A method of measuring oil saturations in cores or sand packs by neutron scattering is described. This method permits local SatUrations to be measured in a core enclosed in a steel pressure vessel without interrupting the flow. calibration data and a saturation distribution measurement obtained with the apparatus are included by way of illustration. Introduction In laboratory flow experiments designed to study the mechanism of oil production it is frequently necessary to measure the oil saturation in a core or sand pack containing Oil, gas, and perhaps brine. It is sometimes desirable that the saturation measurement refer to a relatively small portion of the core, so as to eliminate end effects; that it be made without interrupting the flow, and that it be applicable to a core enclosed in a steel Pressure vessel capable Of withstanding the Pressures encountered in petroleum reservoirs. None of the usual methods of measuring saturation meets all Of these requirements. To measure the gas saturation by observing the apparent compressibility of the gaseous Phase1 would require the interruption of the flow and would give an average saturation for the whole core. An electrical conductivity method is available for measuring brine saturations,2,3 but it is not applicable to oil. A radioactive tracer might be used if a suitable substance could be found that would dissolve in the oil, emit gamma radiation, and not be adsorbed on the sand, the steel core holder, or in an aqueous- phase. The large gamma-ray background from the mass of radioactive oil in other parts of the apparatus might, however, be undesirable. There remain for the saturation measurement methods involving the absorption or scattering of some penetrating radiation produced outside the apparatus. It might be possible to use X-rays or gamma rays; but these radiations would be so much less strongly absorbed and scattered by the oil than by the steel walls of the core holder and by the sand that it seems doubtful that much precision could be attained. In a beam of neutrons, however, there is a radiation that is very penetrating and, moreover, its scattering by substances containing hydrogen, such as oil, is qualitatively different from that occasioned by other materials. It would seem possible, and has proved to be quite practicable, to measure oil saturations accurately under the required conditions by means of neutron scattering. The first, to our knowledge, to suggest neutron scattering as a means of measuring oil saturation were F. Brans and ,. A. Bottema,4 of the Amsterdam laboratories of the Bataafache Petroleum Maatschappij, who made a few preliminary experiments to test the method while they were engaged in exploring the possibilities of neutrons for well-logging purposes. A comprehensive account of neutrons and their behavior will be found, together

Jan 1, 1946

By E. S. Mardock, E. Brunner

A method of measuring oil saturations in cores or sand packs by neutron scattering is described. This method permits local SatUrations to be measured in a core enclosed in a steel pressure vessel without interrupting the flow. calibration data and a saturation distribution measurement obtained with the apparatus are included by way of illustration. Introduction In laboratory flow experiments designed to study the mechanism of oil production it is frequently necessary to measure the oil saturation in a core or sand pack containing Oil, gas, and perhaps brine. It is sometimes desirable that the saturation measurement refer to a relatively small portion of the core, so as to eliminate end effects; that it be made without interrupting the flow, and that it be applicable to a core enclosed in a steel Pressure vessel capable Of withstanding the Pressures encountered in petroleum reservoirs. None of the usual methods of measuring saturation meets all Of these requirements. To measure the gas saturation by observing the apparent compressibility of the gaseous Phase1 would require the interruption of the flow and would give an average saturation for the whole core. An electrical conductivity method is available for measuring brine saturations,2,3 but it is not applicable to oil. A radioactive tracer might be used if a suitable substance could be found that would dissolve in the oil, emit gamma radiation, and not be adsorbed on the sand, the steel core holder, or in an aqueous- phase. The large gamma-ray background from the mass of radioactive oil in other parts of the apparatus might, however, be undesirable. There remain for the saturation measurement methods involving the absorption or scattering of some penetrating radiation produced outside the apparatus. It might be possible to use X-rays or gamma rays; but these radiations would be so much less strongly absorbed and scattered by the oil than by the steel walls of the core holder and by the sand that it seems doubtful that much precision could be attained. In a beam of neutrons, however, there is a radiation that is very penetrating and, moreover, its scattering by substances containing hydrogen, such as oil, is qualitatively different from that occasioned by other materials. It would seem possible, and has proved to be quite practicable, to measure oil saturations accurately under the required conditions by means of neutron scattering. The first, to our knowledge, to suggest neutron scattering as a means of measuring oil saturation were F. Brans and ,. A. Bottema,4 of the Amsterdam laboratories of the Bataafache Petroleum Maatschappij, who made a few preliminary experiments to test the method while they were engaged in exploring the possibilities of neutrons for well-logging purposes. A comprehensive account of neutrons and their behavior will be found, together

Jan 1, 1946

By L. G. Schulz

THE possibility of using reflected X-rays to produce images of crystals was explored by Barrettl who developed simple experimental procedures for X-ray microscopy and who showed by numerous examples the great range of information that could be made available. In his method characteris radiation is used to produce an image of a limited area of the sample on fine-grained film, this image subsequently being enlarged approximately X100. A somewhat different method will be described here;' white radiation from a fine-focus X-ray tube-"' is used to produce an image of a relatively large area of a single-crystal surface. part of the sample; therefore, white radiation is needed. The resolution in the image formed on the photographic film will depend on the size of the focal spot of the X-ray tube. If the sample is made

Jan 1, 1955

By Jacob T. Wainwright

This stove has been designed to meet the requirements of a fur nave that must be operated with either a reducing or a neutral flame ; and more particularly to make feasible the operating of re duction-furnaces with the cheaper grades of gas-fuel, which has heretofore been impracticable. In general, cheap gas-fuels result from incomplete combustion, and, being suited for further combustion, answer very well for heat ing purposes; and high temperatures may be attained with them, when they are used in conjunction with regenerative stoves. But, in operating a reduction-furnace, a high temperature is not the only requirement. It is also necessary that the fuel shall not be cansumed beyond a certain degree, in order to insure a re ducing flame for the reduction-process ; consequently, it becomes necessary to dispense with combustion as much as possible previous to the use of the fuel-gases in the reduction-furnace. Moreover, these gases must be preheated to a high degree, in order that a sufficient temperature may be maintained in the furnace; and, since this preheating cannot be accomplished by a previous, partial combustion of the gases themselves, it follows that the stoves for pre heating these gases should be heated to as high degree as it is pos sible to attain with an inferior fuel. This result can best be accom plished by complete combustion of the heating gases, and is further facilitated by preheating all of the air and gas used for this com plete combustion. It is more particularly desirable to preheat the air, since the combustible gases used may generally be obtained directly from the furnace, or gas-producer. The reducing gases, escaping from a gas-fuel reduction-furnace, are capable of considerable further dilution with air to secure com plete combustion, and consequently may be utilized for heating the stove to a high temperature, provided the air be sufficiently preheated.

Jan 1, 1889