Search Documents

By Zhi-yong Liu

A novel process, which includes hydrochloric acid leaching, iron powder cementation, sulphide precipitation and spray pyrolysis, was proposed to treat iron concentrate recovered from zinc kiln slag for comprehensive utilization. Treated by two-stage hydrochloric acid leaching, the leaching extents of Ag, Pb, Cu, Fe and Zn are 96.78%, 95.14%, 97.72%, 94.51% and 87.74%, respectively. More than 99% Cu and Ag are recovered from the leach liquor when iron powder is 25% higher than the stoichiometric requirement. More than 96% Pb and Zn is removed with three times of theoretical consumption of FeS, and the concentration of impurities of the final solution is less than 500 mg/L. Fe2O3 powder, which is spherical with a mean size of 12 μm and purity of 99%, is prepared with final solution at 700 °C. In this process, not only the metal values can be recovered effectively, but also the iron resources of zinc kiln slag can be converted into high purity Fe2O3, thus, the comprehensive utilization of iron concentrate recovered from zinc kiln slag can be realized.

By B. L. Landrum, P. B. Crawford

The rise of a new laboratory model for studying tran-sient fluid flow problems, is described. The theory of he model is based on the analogy between the equa-ions which describe the flow of compressible fluids in porous media and the conduction of heal m solids. The "thermal model" is used to study the Iransient Pressure distributions during fluid displacement in five Standard well patterns. The studies indicate quantita-ively how die pressures change with lane at any poit in the reservoir. It was found that, in general, the equipressure contours are more radial about the wells early in the program than after steady-state conditions are established. In addition two studies of fractured five-spot patterns are presented to illustrate the opplicability of the model for a wide range of reservoir conditions. INTRODUCTION In reviewing the oil recovery techniques involving fluid displacement it was found that a very considerable quantity, if not all of the fluid within the pattern, mav be displaced (during a time-dependent or transient period rather than the usually applied steady-state conditions. Since the oil is displaced during a transient period, it was believed desirable to study the nature of the transients for several of the commonly used fluid displacement patterns. A literature review indicated fairly complete studies of the streamlines and isopotentials for steady-state: yet few data were available on the transient displacement process. Several simulated steady-state multiwell displacement programs were carefully mapped by Foster and Lodge on an electrical conducting sheet as early as 1875, hut the more recent work of Muskat provides the best source for steady-state phenomena having direct petroleum production applications. In order to study the transient phenomena a new lab-oratory model was developed which is based on the similarity of the equations that describe the flow of fluids and the conduction of heal. The equations describing heat flow were developed early in the nineteenth century. Regorous solutions tor most transient problems are in gcneral complex and usually limited to fairly restricted houndary conditions.2 the thermal model may be used to obtain approximate solutions for many transient problems without some of the restrictions required for analytical solution. The application of the analogy between heat conduc-. tion in solids. electrical flow and fluid flow, in a porous media has long been recognised. Techniques originally developed for the solution of heat conduction prol?lems have in many instances proved useful in ohtaining solutions to fluid flow prohlcms. Ir is not surpri5ing then that a heat conduction or thermal model might bf usefull in sludying transient fluid flow. As an example ol the problems suitable for the study with the thermal model, transient pressure distributions during fluid displacement programs are studied for the five-spot, line-drive, staggered line-diibe. seven-spot 211 two fractured patterns. A model was constructed of an iron plate of geometry similar to the reservoir patterns under study. The principle of mirror images as suggested by Bryan was employed in shaping the model. A temperature differential was obtained by soldering two sections of copper tubing to the plate at points analogous to the well location in the field. Hot and cold water was passed through the tubing. The temperature distribution, analogous to the pressure distribution in the field. was obtained a a function of time from thermocouplcs distributed over the plate.

By J. L. Howell, D. E. Pollock, A. B. Cecala, A. D. O’Brien

Dust from worker,s clothes has been shown to be a problematic source of personal dust exposure. A recently completed effort between NIOSH and the Unimin Corporation has resulted in a quick and effective way to clean worker clothing. The process involves a booth under negative pressure with an air spray manifold to supply compressed air to blow off the dust from the clothing. The overall system is designed to meet MSHA and OSHA requirements. Results of field testing indicated that the manifold cleaned the clothes 10 times faster and removed 50% more dust than cleaning methods used by workers today.

By Eckley B. Coxe

IN the anthracite coal regions of Pennsylvania the custom has been to sight either at an open light (generally a mine-lamp), or at the string of a plumb-bob. If the station was intended to he a permanent one, a spud, as it is called, that is, a nail resembling a horseshoe nail with a hole in the head, is driven into the timbers over the station, or, if there be no timber, a hole is drilled in the coal or rock roof, into which a wooden plug is driven, which serves to hold the spud. The first operation in making a survey is to lay out the stations, that is, to mark the place where the holes are to be drilled for the plugs or the points on the timbers where the spuds are to be driven in. This should be done before any instrumental work is begun, as much labor can generally be spared, and the use of very short sights can often be avoided, by carefully laying out the stations beforehand. When the stations were laid out, a plumb-bob was hung from the innermost spud, which I will call No. 1, the instrument was put in position at No. 2, by plumbing down and putting a centre-pin under the spud, and then setting up over the centre-pin, and another plumb-bob was suspended from No. 3. If great accuracy was not required, a mine-lamp was set up under the plumbbobs, at No. 1 and No. 3, and the engineer sighted at them. If great accuracy was required, a lamp or some white sbface was held by an assistant behind the strings of the plumb-bobs. To work with any speed by the latter method (i. e., the accurate one), it was necessary for the engineer to have three assistants on whom he could rely even when the chaining was done afterwards; viz. : one to hold the light behind the string at No. 1, one at No. 3, and an assistant at the instrument to hold the light while levelling, reading the in

By E. Pothas, P. R. Jochens, L. B. McRae, D. D. Howat

SYNOPSIS The phase equilibria, viscosities and sulphur capacities of titaniferous slags were determined in the composition range of CaO = 30 to 0 per cent, MgO = 0 to 30 per cent, SiO2 = 19.69 per cent, AI.O. = 13.12 per cent and TiO2 = 37.19 per cent. The phase equilibria were determined by high temperature microscopy, the viscosities were measured using a rotating outer cylinder viscometer and the sulphur capacities were evaluated from the results obtained by equilibrating the slag in a gas stream of CO, CO2 and SO2. The behaviour of these titaniferous slags under reducing conditions was examined. Possible use of the slags for aggregate, or slag-cement, and the possibility of extracting the TiO2 are discussed. CONTENTS INTRODUCTION PHASE EQUILIBRIA VISCOSITIES DISCUSSION OF THE PHASE EQUILlBRIA AND THE VISCOSITY VALUES EFFECT OF REDUCING CONDITIONS SULPHUR CAPACITY DISCUSSION OF THE SULPHUR CAPACITY VALUES POSSIBLE USES CONCLUSIONS ACKNOWLEDGEMENTS REFERENCES INTRODUCTION The latest phase in the exploitation of the mineral wealth of the Bushveld Igneous Complex is the erection of a plant by the Highveld Steel and Vanadium Corporation Limited near Witbank to produce steel and vanadium-rich slag from the titaniferous magnetites occurring in this rock formation. These titaniferous magnetites occur as segregations in the gabbro body and the TiO2 is mainly present within the structure of other constituent minerals but also occurs in smaller amounts as ilmenite and ulvospinel. The vanadium is thought to be present in the iron oxide phase due to isomorphous replacement of the iron. Until very recently, it was feared that the high titania content of these titaniferous magnetites would militate against their value as useful iron ores. However, the design of an integrated plant to produce both steel and vanadium-rich slag, and in which the iron-making process is conducted in submerged arc furnaces instead of in

By H. Cui

Rapid technological development has caused an increase in the demand for rare earths. Ancylite, a strontium rare earth carbonate, is a potentially significant source of rare earths. In this research paper, essential aspects of the magnetic separation and flotation processes for Bear Lodge ore are presented. A flowsheet is proposed based on an optimization of the variables and flotation simulation. The economic feasibility of the proposed processing operation is analyzed.

By Robert F. Chaken

The National Institute of Occupational Safety and Health (NIOSH) is dedicated to achieving "safety and health at work for all people -- through research and prevention." An important procedure in pursuing this goal is the NIOSH surveillance studies where the number of injuries are monitored according to demographic, employment, and injury characteristics. These numeric data are then used to identify injury risk factors, develop safety programs, and monitor implemented remediation methods. In essence, ongoing surveillance data form the bases for directing the resources of NIOSH in setting of research and prevention priorities, and in evaluating safety procedures. This surveillance techmque, which is based on the Public Health Service approach to identification and control of disease (1, 2, has been used quite successfully in dealing with relatively high number injuries such as hearing loss, musculoskeletal disorder, respiratory disorder, falls, etc. (2). However a problem arises when extending this paradigm to lower frequency traumatic events; particularly those often referred to as disasters (i.e., high impact events like mine explosions, airplane crashes, etc.). One can ask the question - there hasn't been a coal mine fire fatality in the U.S. for over ten years, to what extent should we be concerned with mine fire safety (resource-wise)? - The National Traumatic Occupational Fatalities (NTOF) system (4J which NIOSH uses to survey occupational fatalities annually, would suggest that underground mine fires would not be a high priority item in occupational safety. Yet, miners' (i.e., the exposed workers) do consider mine fires and explosions to be major safety concerns. I believe that we need to rethink our evaluation of work place disasters, what is worker exposure, and what are the underlying assumptions (both explicit and implicit) as to the impact that economic, societal and degree of hazard factors make on risk evaluation. The answers to all these problems, will require much broader considerations than those in current surveillance safety and health models. This paper discusses some facets of those considerations.

By J. E. Chilton

Regular monitoring for methane gas is required near working faces in gassy underground mines where the potential for methane ignitions is greatest. However, high concentrations of methane can also accumulate outby the face where methane levels are monitored less frequently. Wearing a personal methane monitor equipped with an alarm could protect persons working in these outby areas from injury or death due to methane ignitions. The National Institute for Occupational Safety and Health (NIOSH) has conducted studies to evaluate methanometers that could be used as person-wearable monitors. For the present study, seven different person-wearable monitors that are not currently approved for underground use were evaluated. This paper describes test procedures used to evaluate performance of methane monitors that could be used underground for continuous personal monitoring of methane. The information in this report provides data that allows the reader to make side-by side comparisons of these instruments.

By Stéphane Bourg

The SCRREEN project gathers more than 50 European initiatives, associations, organisations or independent experts working on Critical Raw Materials (CRM) into a long-lasting network including stakeholders, public authorities, and civil society representatives. SCRREEN contributes to improve the CRM strategy in Europe by (i) mapping primary and secondary resources and substitutes of CRMs, (ii) estimating the expected demand of various CRMs in the future and identifying major trends, (iii) providing policy and technology recommendations for actions improving the production and the potential substitution of CRM, (iv) addressing specific waste electrical and electronic equipment (WEEE) and other End-of-Life (EoL) product issues related to their mapping and treatment standardization, and (vi) identifying the knowledge gathered over previous years and making the data accessible to those beyond the project. As a first step, SCRREEN worked at providing stakeholders with a comprehensive analysis of the current use of the CRM, the mapping of potential primary and secondary resources, and identification of the key technologies used in their production. Substitution profiles, analysis of European business and policy issues pertinent to CRMs, and an identification of the standards, policies, and regulatory frameworks concerning these CRM was also completed. In a second step, to be conducted between mid-2018 and mid-2019, SCRREEN will provide guidance for improving the CRM market in Europe by identifying opportunities from research, development, and innovation (R&D&I) that could reduce the supply and/or economic risks around the most relevant CRM for Europe.

By K. Leutwyler

The authors present an interesting review of various applications of rather well known theories. Their rearrangement of the original working equations is strictly computer oriented and serves well to that effect; however, the new appearance will be confusing to the slide rule-wielding members of the profession. In reviewing discrepancies existing between calculations made by Baker Oil Tools, Inc. field engineering staff members and computer output submitted based on the authors' program, the following error in concept was discovered in Eq. 4. The authors quote the length change due to helical buckling in the following form We appreciate Kurt Leutwyler's awareness of Archimedes' principle, and concur that the final term in the denominator of Eq. 4 results in a small error. Examples checked revealed less than I ft error, and corrective expressions for the four possibilities of displacement follow. The portion of the denominator within extra large brackets represents the original term W,, = weight of outside fluid displaced per unit length. The authors interpreted the expression "displaced" in general oilfield terminology as a fluid displacement, such as "mud displaced with water, etc." A review of the theory forming the basis for these helical buckling calculations discloses that this displacement is meant to be interpreted in the Archimedean sense, i.e., fluid displaced by the solid body. here the tubing and the seal nipple. Lubinski* illustrated the point by summing moments due to mechanical and hydraulic forces acting on a free body immersed in fluid. This moment summation, always in the Archi,medean sense, yields effects of the real forces, including W,, and the fictitious force on the bending mechanism. I would be interested to hear the authors' comments regarding this theoretical point, which nevertheless is not mute since its lack of recognition seems to result in erroneous calculations. We must not remain mute to Leutwyler's suggestion that these equations, although computer oriented, are confusing to the slide rule engineers. Such input information as density in pounds per gallon, pipe diameters in inches, pumping rate in barrels per minute and pipe length in feet is not confusing to most of us. Yet there may be some rugged individuals in field positions who carry tables of conversion constants and have committed to memory such items as coefficients of thermal expansion and Poisson's ratio; however, even this type of individual would welcome a simplified equation occasionally. M

The May General Meeting of the Institute was held in Kelvin House, Johannesburg on 20th May, at 4.30 p.m. Mr V. C. Robinson (Vice-President) was in the Chair in the absence, overseas, of the President, Mr J. K. E. Douglas. There were also present 16 Fellows: Messrs J. A. Adamson, R. S. J. du Toit, J. A. Drummond, D. F. A. Grave, T. L. Gibbs, G. H. Grange, D. V. Howat, P. H. Kitto, Dr P. J. LIoyd, Messrs B. H. L. Leach, J. de V. Lambrechts, Dr J. T. McIntyre, Messrs C. S. MacPhail, P. W. van Rensburg, S. G. Taussig, W. S. Rapson. Six Members: Messrs E. L. Atkinson, R. Hemp, Dr J. Martinson, Messrs A. H. Munro, G. J. van Jaarsveld and J. H. Qwilliam. Two Associate Members: Messrs B. H. van Rensburg and H. Triiter. Four Visitors: Prof J. E. Kerrich, Dr G. K. Sluis-Cremer, Messrs D. Kisbey-Green and L. W. Isserow. For Secretariat: Mrs S. de Gersigny. Total present: 27. OBITUARY The Vice-President: "It is my sad duty to announce the death of Stephen Ring, a member who joined the Institute in ]961, and passed away on the 31st of December, 1969. I would ask you all to rise and observe a moment's silence. Thank you." MINUTES The Vice-President: "May we confirm the minutes of the Special General Meeting and Monthly General Meeting, held on 20th February, 1970, as published in the April issue of the Journal?" MEMBERSHIP The Vice-President: "I have much pleasure in announcing that the names of the undermentioned candidates, having been published in accordance with By-Law 5 2.2, Council has elected them to membership of the Institute in the following grades: Fellows: Denis Ian Field, Arthur Cecil Waiter Comyn, John Hall, Leon VercueiI. Members: Jan McLellan Vosloo, Jan van Velze. Graduates: Richard Darnley Beck, John Edward Andrews, Roderick Nowell Randell. Associates: Alexander Duncan Ross, Ian James Terrill. Students: Hendrik Jacobus Oostenwald Prins Smith, Graham Peter Gooding, Rudannes Raath Homan. "I welcome the newly, elected members to the Institute." GENERAL BUSINESS The Vice-President: "Are there any matters which you wish to raise under 'General'? There being no matters to raise under 'General', we will proceed to the next item on the Agenda." PAPER FOR PRESENTATION The Vice-President: "It is my pleasant duty to call on Dr du Toit to deliver his paper, entitled, 'The functional relationship between dust hazard and the rate of collecting funds to pay compensation for pneumoconiosis'." After the presentation of the paper (published in the March Journal) the Vice-President said: "This paper is a very important addition to the considerable volume of work applications on this subject of pneumoconiosis. The disease, in all its various forms, has received attention from many, many great men, both in the medical world, and engineers. The fact that there is this big volume of publications of literature on it still doesn't mean that we can get anywhere beyond the old, simple injunction, that the prevention of pneumoconiosis does not depend on compensation, levies or anything else. It simply depends on allaying dust as quickly as possible. "This paper, however, is another scientific landmark and I think, in a cold and dispassionate way, it sets out some important points in this whole subject of the relationship of the levy - the money of compensation to the dust which causes it. "I have, personally, very little to criticise in this paper but when I listened to Dr du Toit's final conclusions, I wondered whether the words, 'Within the data available', aren't the key to this whole situation. I agree that he has demonstrated a workable relationship between the current industries for levy and dustiness, but among the factors which he lists as being important, I know that there is one which is missing. This is extremely difficult to evaluate. It's a factor which has destroyed, on so many occasions, the work which has been done by so many genuine, intent, honest people working in this field, and this is this unknown factor, this hidden factor of political opportunism, which has so often played a very important part in pneumoconiosis compensation. "How many of us have seen the grubby, grasping fingers of power-mad demagogues, and I can name them for you - but I won't - and also the blatant vote-catching of certain politicians interfering in this field. How, at the end of all that, the poor scientist, the poor dedicated man in the Mines' Department, who has to write the law of compensation, can operate, I do not know. "From this picture the ordinary layman recoils in horror. Among the casualties in this particular war, which has been fought over so many years, I am perfectly certain, are some of the criteria of certification, especially in '56/,57, when that extraordinary condition, known as 'pulmonary disability' was introduced. I am sure there were casualties then, in the criteria of certification. "Nevertheless, this paper is a worthy addition to the long list of scientific work done on this so-important

By D. R. Parrish, V. J. Berry

Reverse combustion is one thermal method of recovering hydrocarbons from porous undergrortnd formations containing oil or tar. In applying this method, air is introduced via an injection well and the mixture of air and hydrocarbons is ignited in the production well. A combustion zone then recedes toward the injection well, counter-current to the air flow. If the combustion-zorie temperature is sufficiently high, the oil or tar irz place is distilled arid cracked. The hydrocnrhon flows as a vapor to the production well and subsequently is condensed at the surface. Maximum temperature and velocity of morcment arc the two dependent variubles defining the progress of the combustion zone. A theoretical analysis has been made of heat flow in the reverse-comhustion process as.srtrning linear flow in a homogeneous system. The differentirll equatious, which include the oxygen-hydrocar-hon rctrction rate, have been solved numerically. Rescllts indicate that the maximum temperature reached and he cornbustion-zone relocity both increase with an incrense in air-injection rote. Heat loss to surroundings has little effect on the maximum comhustion-zone ten?perature achieved, but it is reflected in a reduced com-bustion-zorzc, velocity. It is also predicted that an increase in the oxygen-hydrocarbon reaction rate results in a reduction in the maximum temperature reached. The calculated results are in agreement with results from reverse-combustion experinments using samples of a tar sand. INTRODUCTION Reverse combustion is one thermal method of recovering hydrocarbons from porous underground forma-tions containing oil or tar.' In applying this method, air is introduced into the underground formation via an in-jection well. In one or more adjacent production wells, the mixture of air and hydrocarbons is ignited at the sand face. The combustion zone thus formed recedes toward the injection well, counter-current to the air flow (Fig. 1). Under certain conditions, the rate of recession of the combustion zone is sufficiently rapid that only a small fraction of the hydrocarbon at any point in the formation is consumed by combustion. If the combustion-zone temperature is sufficiently high, the uncon-sumed oil or tar in place may be partially distilled or cracked. This material then can flow as a vapor with the gaseous combustion products to the production well and subsequently be condensed at the surface. Under other conditions, the combustion-zone temperature may be so low that little, if any, of the hydrocarbon is vaporized. Nevertheless, the viscosity of the hydrocarbon may be reduced enough so that other recovery methods subsequently can be applied to produce oil at attractive rates. The maximum temperature achieved in the combustion zone and the velocity of movement are the two dependent variables defining the progress of the combustion zone. Presumably. knowledge of these factors will permit estimates to be made of oil recovery efficiency, oil and gas producticn rates and average amount of injection air required per barrel of oil recovered. The maximum temperature and combustion-zone vclocity are, in turn, expected to be dependent upon air flow rate, heat losses to surrounding formations, physical properties of the rock and the hydrocarbon. the extent to which the porous formation is saturated with the hydrocarbon, the rate of oxidation of the hydrocarbon and the amount of heat released. It is desirable that the relative importance of each of these factors be understood so that the results of laboratory tests of oil recovery by reverse combustion can be properly interpreted in terms of expected field performance. It is particularly desirable to assess the importance of heat lcsses because these are difficult to control or eliminate

By Oswald J. Heinrich

This question having been repeatedly raised, and particularly revived in a discussion at the last meeting of the Institute, I beg to submit the following remarks, based partly upon personal experience, partly upon the results of large works in Europe, establishing, years ago, the practicability of certain methods, which may therefore be considered to have passed the stage of mere experiment. Before entering upon the merits of one system or the other, it is indispensable to lay down some recognized principles, which ought to serve as guides, in arriving at fair judgments in questions of this character. In a profession like that of mining, locally so much influenced by combinations of circumstances too various to mention, it is difficult to do more than lay down the main features of a system based upon general laws and recognized facts, leaving the details to the sound judgment and experience of the engineer, who, by virtue of his education, ought to be able to make such choice and such modifications in the plans of a mine as the particular case demands. Mineral resources, in some countries the sole property of the owner of the soil, in others controlled by the strong hand of a government, are nevertheless a gift of nature, in which, indirectly, the whole of the human race has as real an interest as the immediate owner himself. This interest requires, at least, that no unnecessary waste of

By Dalene Joalet Steenkamp

Excavation of industrial-scale furnaces allows for the systematic study of reaction sequences by identifying the different reaction zones within the furnace. In 2013, Transalloys excavated a 48 MVA submerged arc furnace that was used for silicomanganese production using the ore-based route. The excavation method was reported elsewhere as was observations made in terms of refractory wear and modes of electrical energy dissipation prior to excavation. The paper presented here, reports on the process reaction zones observed during the excavation and subsequent phase chemical analyses of a number of process samples obtained during the excavation. The zones identified were a loose burden zone, a dry coke-bed zone, a wet coke-bed zone, a hard build-up zone, and an alloy zone. The results are compared to observations made in the excavation of an industrial-scale SAF and a pilot-scale SAF in Norway. The presence of the hard build-up zone below one of the electrodes and absence of a slag zone below all three electrodes are unique features of the SAF excavated at Transalloys.

By G. A. Christie

This paper presents a vision system capable of providing an accurate estimate for the volume of stockpiles using 2D images. By imaging a stockpile from multiple viewpoints with a two-camera system, a 3D point cloud of known scale is generated using image-based 3D reconstructions. With a small amount of user interaction, the points that correspond to the surface of the stockpile are extracted and used to measure the volume. The hardware required for the two-camera system is light enough to be carried by a small unmanned aircraft and is a cheaper alternative to LiDAR while still being competitive in terms of accuracy. When testing the volume measurement algorithm on a stockpile of known volume, the measured output was found to have less than a 3% error from the ground truth.

By Fengyu Ren, Ruslan Puscasu, Kunpeng Yu, Gideon Chitombo, Liangliang Kang

Weak rock mass and poor flow of material can cause excessive ore loss and dilution in sublevel caving. This paper proposes a method to establish design parameters for an inclined deposit within a weak rock mass and evaluates its effectiveness through a case study. In order to solve the problem of low production capacity and poor safety conditions of sublevel caving mining at the Yingfangzi silver mine in China and to ensure a smooth production from the refractory ore body, it is necessary to select appropriate stope structure parameters. Based on random medium theory of gravity flow and the improved equation of ore discharge in random medium, an end wall drawing experiment is carried out to determine the optimum design and flow parameters. It was observed from the experiment that, due to the influence of the randomness of particle movement, there is a mismatch in the flow parameter measurement. The flow characteristics and distribution of the marker particles are different from that of the tested particles resulting in large drawbody shape distortion. The study finds that the flow parameters obtained by the new method are more practical and account for drawpoint width and recommends that the new method for determining the sublevel stope height is used when the drawbody is a non-standard ellipsoid. The outcomes have shown a significant reduction in the dilution and improvement in the recovery rates, and the safety and efficiency of mining have been greatly improved.

By E. R. Darby

When bronze is melted in open-flame furnaces a considerable amount of slag is formed during the melting operation. This slag may be incidental to the melting practice or it may be formed intentionally by slag-forming constituents added to the furnace charge to provide a covering for the melted metal When this slag is removed from the furnace, it contains, in addition to the free metal carried with it by the skimming operation, a considerable amount of metal as oxide either combined or mechanically mixed with it. By the usual crushing and washing methods this metallic oxide generally goes to the tailings pile, so diluted by other foundry waste that its recovery is economically impossible. If this furnace slag is not inixed with other foundry waste, practically all of the metal oxide as well as the free metal may be recovered by some form of smelting operation. The first side of the problem to be considered is the amount of slag produced. It may not be true that the majority of foundries have no record of the amount of slag made, but it is certainly true that a very great many either have no record or pay little attention to it. Usually the collection of furnace slag for one normal week's run will be sufficient to determine the advisability of its being handled in a different manner from the crushing and concentrating generally employed. If a large representative sample of the accumulation is crushed and concentrated and the recovery compared with the results of a careful assay, it will be found that a large percentage of the total metal content was lost in the washing process. The percentage of loss will be greater than the metallic oxide content, as some free metal in extremely fine particles always is carried away in grains of slag. From such a comparison, it may be determined whether or not the slag should be melted, sold or treated as before. An example of an actual test of this kind is given below. In a foundry melting 50,000 to 80,000 Ib. of bronze per day, a week's accumulation of furnace slag and ladle skimmings amounted to 16,500 1b.—over three times as much as the foundry superintendent estimated before the accumulation was made. Two tons of this material was taken as representative of the lot. This was ground through a thoroughly cleaned ball-mill and concentrated over a Wilfley table. The

By J. Downie, H. A. Kendall, G. H. F. Gardner

Some cases of the motion of two miscible fluids in uniform linear models are discussed. There is no bulk flow through the models, and the convection currents are caused solely by density gradients. Horizontal, vertical and inclined positions of the models are treated separately. Analytical formulas for the motion of a narrow transition zone between the fluids are obtained and compared with measurements, and the limitations are discussed. Observations of the ultimate stabilizing effect of diffusion on the motion me compared with the theory. It is shown that fluid motion in an inclined model with a constant rate of bulk flow can be directly deduced from the cases discussed, provided the viscosity ratio is small. INTRODUCTION In some reservoirs, convection currents caused by variations in the fluid densities may be significant. For practical purposes it may be desirable either to minimize the effect, particularly in miscible displacement, or to maximize the effect as in some schemes for the production of heavy crude oils by combustion. Experiments by Craig, et al,l have shown the effectiveness of gravity in reducing the efficiency of a miscible displacement when the mixed zone between the displacing and displaced fluids is narrow compared with the height of the system. The presence of a board transition zone from one fluid to the other has been shown by Perkins, et al,2 to nullify considerably the influence of a difference in density. Indeed, the theoretical discussion given by Perrine3 has shown that when the transition zone exceeds a certain critical length the efficiency approaches 100 per cent. Many additional results 495 have been published on miscible and immiscible displacements which show the importance of density difference and also mobility ratio. The aim of the present paper is to discuss some cases of gravity segregation of miscible fluids in linear models such as are often used in the laboratory to evaluate recovery processes, and to supply additional experimental evidence in support of the conclusions. An attempt is made to calculate simply the magnitude of the fluid motion caused by density pdients and mobility ratios, and to quantify the modifications in the fluid motion produced by the combined effect of molecular and convective dispersion. However, it should be pointed out that the uniform permeability of the models and the use of only a single pair of fluids (whereas several pairs of fluids may be employed in a recovery scheme) severely restricts direct conclusions about natural reservoirs. For conciseness, the motion generated by density gradients is emphasized for the case of no bulk flow through the models. This is not a drastic limitation because, as shown in Appendix A, a fluid motion with a constant bulk flow through any linear model is equivalent to a fluid motion with no bulk flow in the same model inclined at a different angle to the horizontal, provided only that the density is a linear function of the viscosity of the fluid mixture. HORIZONTAL MODEL In this case the model is assumed to have a rectangular cross-section with a horizontal breadth b, vertical height and a great length L. Throughout.the paper it will be assumed that x-axis points along the length, the y-axis along the width and the z-axis along the height. Thus, in the present case the x, y plane is horizontal. Initially, the transition zone between the fluids is assumed to be thin and parallel to the y, z plane. An example of this situation in a practical problem arises in a scheme for using air for cushion gas in the storage of fuel gas in aquifers. 6 With the passage of time the denser fluid sinks and spreads along the bottom of the model while the lighter fluid rises. The interface extends from the top plane of the model to the bottom. Experiments have shown that for practical purposes this interface may be treated as a plane surface at all times. The surface is significantly curved, however, for viscosity ratios greater than about 10. Other complications arise with large viscosity ratios because of the nonlinear dependence of the viscosity on the composition of the mixture, so that the simple results given here are restricted to fairly low viscosity ratios.