Search Documents

By Garrett R. Hyde, Aleksandar Stojsic

Introduction Economic pressures in the international marketplace have spurred advances in almost every facet of flotation. New developments have occurred in flotation chemicals, flotation machines, and in the application of automated process control technology to grinding and flotation. Many of these advances are the results of concerted efforts to apply the knowledge obtained from basic flotation research over the past 15 to 20 years, to develop industrial technology for the more efficient separation of sulfide minerals in real ores. Mining company executives and equipment manufacturers are being called on to make decisions for incorporating some of this technology into present operations. This discussion is limited to two developments now at the stage of being engineered into industrial scale operations. They are control of the electrochemical environment or redox potential of the slurry, and control of bubble size in column flotation. Potential economic impacts are noted. Froth flotation Froth flotation makes use of differences in the surface properties of finely ground minerals to obtain separation. If the mineral surface is water repellent, the mineral particle will stick at the air-water interface of a bubble and float. If a mineral surface is not naturally water repellent, a chemical collector is added to the slurry to make it so. The collector must adsorb on, and make water repellent only the mineral to be recovered, not the gangue minerals. To achieve this degree of selective adsorption, chemical modifiers may have to be added to the slurry. Modifiers include pH regulators, activators, and depressants. Electrochemical control One of the most significant developments in recent years in the area of mineral processing technology is the recognition that collector adsorption of sulfide minerals is an electrochemical process. Consequently, flotation of these minerals depends on the electrochemical or redox potential of the system. Many of the modifiers used to control various flotation processes are actually potential controlling reagents. For example, the role of well known depressants, such as cyanide, sodium hydrosulfide, and sulfur dioxide, is essentially to

Jan 7, 1987

By John R. Thalimer

For the past four years the U.S. Bureau of Mines has investigated the early detection of insulation failure in squirrel cage induction motors. Data modeling techniques based on values calculated from a motor's voltage and current phasors were used to produce a pair of polynomial equations that calculate a motor's insulation leakage current and power. This paper describes a prototype system using these models that monitors six motors on a continuous miner for insulation leakage in order to anticipate insulation failure. The system consists of a motor-data system for each motor on the machine, a control computer located some ' distance from the miner, and a serial bus to interconnect these items. Each motor-data system consists of an analog interface to the motor's voltages and currents, a single-board computer that reads digitized data and calculates voltage and current phasors, and a bus node that interfaces the single-board computer with the serial bus. The control computer can request and receive phasor data from any motor-data system on the bus. Using models developed at the Bureau, the control computer calculates and displays two deterioration values for each motor, one for leakage current and the other for leakage power. These values are then stored into a database from which the user can display them over time.

Jan 1, 1992

By Alain GSELL

The aluminous nature of the laterites capping the Minim Martap plateaus - named after two neighbouring villages - and the small nearby upland of Ngaoundal, was discovered by the Cameroon Mines and Geology Division in 1958. These deposits of worldwide importance, are located in the Ngaoundere region, some 500 km (300 miles) north-east from the port of Douala (as the crow flies). They constitute the major part of the Cameroonese mining potential. The study of the deposits and their economic development has been entrusted to SEBACAM, Societe d'Etudes des Bauxites du Cameroun, a company in which the Cameroon government participates. The bauxites originate from the transformation into laterite of lavas which covered a Precambrian basement during the Upper Cretaceous period and the Lower Tertiary era. The entire Minim Martap deposit comprises bauxite plateaus ranging in area from a few acres to several square miles. The average thickness of the bauxite layer varies from 10 metres at Minim Martap to 17 metres at Ngaoundal. The mineral constituents of the bauxite are gibbsite trihydrate, kaolinite and the usual iron and titanium oxides. Treatment of the ore does not involve any particular technical problems. Proven exploitable reserves amount to 560 million tonnes of dry bauxite with a reactive silica content of 1.80% and an available alumina content of 43.70% at Minim Martap, and 120

Jan 1, 1984

By A. E. Hall, D. M. Mchaina

This paper outlines a proposed research project to investigate secondary sulphide ignitions. Sulphide dust explosions have resulted in loss of production and considerable capital and equipment replacement cost in Canadian mines. The practical application of laboratory test results to actual mine situations presents a complex problem and is a field requiring immediate investigation. The factors contributing to the outbreak of sulphide ignitions are not clearly understood. This indicates a need for future research to study the influence of blasting parameters, exhaust temperatures, and dust characteristics on secondary ignitions at source. The aim is to establish realistic quantities for each of the parameters and develop a model that will incorporate field and laboratory quantities. It may be possible to predict outcomes of sulphide ignitions when the parameters are known. It is also planned to study the techno-economic feasibility of the preventive and control methods. One aspect of planned future research requires quantification of some terms by obtaining measurements of pollutant concentrations and size distribution of the airborne dust in an underground environment immediately after a main blast. A blast-proof aerosol sampling package to measure the source term will be designed. This sampling package will consist of a sampling section and a shield which will be capable of withstanding a blast explosion. The accuracy in measuring the variables and combination of variables will be the main focus in this proposed study.

Jan 1, 1987

By L. H. Munson, E. H. Carbaugh, D. E. Hadlock, L. F. Munson, R. L. Gilchrist

As a result of the Three Mile Accident and because of evidence indicating radiation protection problems, the U.S Nuclear Regulatory Commission (USNRC) undertook a major effort, with the assistance of the Pacific Northwest Laboratory (PNL), to analyze radiation protection programs at 47 commercially operated nuclear power plants. This effort, the health physics appraisal program (HPAP), using a modified management oversite and risk tree analytical logic methodology (MORT) which consisted of analytical trees with applicable questions for each tree, was subsequently extended to include USNRC licensed uranium mills. In the uranium mill appraisal, a team of four senior radiation safety professionals spent a week at each mill evaluating the total health physics program. Prior to the HPAP, the USNRC inspection efforts were designed to determine if the licensees were fulfilling the requirements of the code of federal regulations and the mills source material license. The HPAP, however, was designed and structured to facilitate an integrated look at the total program. It is also directed into areas for which explicit NRC regulatory requirements may not currently exist. The latitude of the HPAP was designed to allow professional judgement in the evaluation of program adequacy, because merely meeting the explicit regulatory requirements does not necessarily ensure an adequate and effective health physics program. The results of the first eight site visits and the generic health physics problems identified within the uranium milling industry will be identified in this paper.

Jan 1, 1981

By G. T. Lineberry, B. R. Yama

Artificial intelligence research has produced several tools for commercial application. Some of the techniques that are widely used today include neural networks, fuzzy logic and expert systems. Artificial neural networks (ANNs) are excellent predictive, pattern recognition and data analysis tools. In the mining industry, ANN techniques are being used commercially for real-time process-control applications. Modeling of spatial data, ore-reserve estimation, tunnel design, long wall-stability prediction and geologic roof classification are additional applications in which neural networks have been applied successfully. In this study, a standard back propagation algorithm was used to train a series of neural networks for a real-world predictive task. After training and optimizing the neural network architecture, the performance of the network is measured on an independent validation set. Results indicate a mean error of less than 1 % between the actual and predicted values. A neural network model was developed for learning the spatial continuity of a mineral field and, consequently, for predicting sulfur values for given coordinates. The neural network not only performed satisfactorily, but in some cases performed even better than the kriging model.

Jan 1, 2000

By E. J. Peuraniemi, J. T. Yli-Penttilä, A. Jokilaakso, K. M. Riihilahti

Dust formation and the reaction mechanisms of copper matte particles were investigated in simulated flash-oxidation conditions. The granulated and ground matte consisted of 72% Cu, 4% Fe and 21 % S (by weight). Experiments were conducted by using a laminar flow furnace with a reaction-gas composition ranging from 20% to 75% O2 (by volume) in nitrogen. The particle feeding was carried out by a single-particle method, and samples were quenched after a specified reaction time ([~] 0.05 to 0.4 s) in a cold water film. Experimental temperatures ranged from 800o to 1,300o C. Chemical analysis was used to determine the degree of oxidation, and optical and scanning electron microscopes, combined with energy dispersive analysis, were used for characterization of the reacted particles.

Jan 1, 1999

By Oleg N. Tikhonov

A new extraction method which includes a combination of gravity and alternating electromagnetic fields with interfriction of mineral particles was proposed (Tikhonov O.N., 1979) and tested in lab conditions at Mineral Processing Department of St. Petersburg Mining Institute. At the beginning of the work the main goal was to find an efficient alternative to traditional extraction methods (like shaking table, etc.) used for concentration of fine alluvial gold particles. After the method and a series of gradually improved lab separators (Tikhonov, et al., 1983) were positively tested for gold (Tikhonov, et al.. 1986) the efficiency of the method was also proved for extraction of and other heavy minerals from alluvial and ground ores in experimental work at the St. Petersburg Mining Institute. The fine heavy minerals were separated to concentrates with high technological indices (Andreev et al., 1987), (Tikhonov, Gladkov, 1987), (Tikhonov, Gladkov 1989). In parallel a semi-industrial gravi-electro- magnetic separator was made and several technological flowsheets were designed for testing the method in industrial conditions (Tikhonov O.N., Miazin B.N., 1987), (Tikhonov O.N., Tolstunov S.A., Jebelev V.V.,1989). A series of industrial tests accompanied with complete sampling and standard technological measurements were performed at several alluvial deposits, with good technological indices. Thus the method and the flowsheets displayed high additional recovery of fine alluvial gold particles during elkrated industrial experiments in Siberia (Aldan province). At higher concentrate grade this method, even in one-stage operation, increased gold recovery more than by 25% compared with shaking table (for particles -0.10 mm gold recovery was 82% vs. 55% by shaking table, for particles - 0.25mm the recovery was 97.5% vs. 61.8%). Experiments were performed by joint team of St. Petersburg Mining Institute and the Irkutsk Institute of Rare Metals (Tikhonov O.N., Gladkov S.A., Tarasova T.B., 1991).

Jan 1, 1995

By J. E. Gebhardt, P. E. Richardson

External control of the electrochemical potential of a bed of mineral particles was investigated by the US Bureau of Mines as a method for controlling particle interaction with organic collectors and, thus, flotation response. A cylindrically shaped, 200- to 300-g bed of sulfide mineral particles was used as the working electrode in an electrochemical-flotation cell. In a pH 9.2 sodium tetraborate buffer, chalcocite floated with ethylxanthate at applied potentials [=] - 0.25 V (versus the standard hydrogen electrode) ; pyrite floated at potentials [=] 0.15 V. Differential separation of a chalcocite-pyrite mineral mixture was obtained by controlling the potential, but the flotation response of the mixture did not follow the behavior predicted from measurements on the individual mineral beds. Chalcocite dissolution products had a deleterious effect on chalcocite-pyrite separation by activating pyrite so that it floated below the potential (~ 0.13 V) at which xanthate is oxidized to dixanthogen. A pyrite bed activated with a copper(II) sulfate solution exhibited a flotation response with a potential dependence between that of chalcocite and pyrite. Pyrite dissolution products had no effect on chalcocite flotation.

Jan 1, 1988

By Wu Chao, Wang Pinglong, Meng Tingrang

This paper discusses the research results of sulphide ore properties with respects on self-heating and the assessment techniques on mine spontaneous combustion for several mines of sulphide ores in China. A group of experimental devices for testing the chemical behaviour of sulphide minerals, the geological structures, the rate of oxygen absorption, the initial self-heating temperature, and the ignition point of ore samples, etc. are also described in detail. These methods have been proved to be effective by several sulphide mines which all have the history of spontaneous combustion of sulphide ores. INTRODUCTION A fire disaster from the spontaneous combustion of sulphide ores in a mine can produce large quantities of toxic gases and heat to pollute the working environment, lead to large economic losses and loss of human life, and even compell the mine to be closed. Some mines of copper, lead, zinc, tin and iron in China are of high content of sulphur. It is of great importance to evaluate the potential of spontaneous combustion of sulphide ores before beginning to design and mine the ore deposits. At present, the mechanism of spontaneous combustion accepted commonly is that sulphide ores undergo slow oxidation on exposure to moisture air at ambient temperature, with the evolution of heat, gases and moisture. The heat generated, if not dissipated, gives rise to an increase in the temperature of the sulphide ores. As the temperature of the sulphide ores rises, the rate of oxidation increases. If this is allowed to proceed unchecked it can eventually result in the ignition of the sulphide ores. For pyrite (FeS2), it reacts with oxygen and water according to the following equations.

Jan 1, 1993

By R. L. Neman

Southern Oklahoma (including the Arbuckle Mountains) has outcrops of rocks representing Precambrian, Paleozoic, and Cretaceous ages. Thus, rocks of every geologic age in the Paleozoic and Mesozoic Eras, except the Jurassic and Triassic Periods, are present in south-central Oklahoma. With rocks of such an enormous age range present in a relatively small geographic area, it is little wonder that a wide variety of economic rock and mineral resources are locally present. This report will describe the occurrence, mining, processing, and end-uses (dictated by chemical and physical properties) of the Precambrian Troy Granite, the Cambrian Royer Dolomite and Fort Sill Limestone, Ordovician silica sands of the Simpson Group, Ordovician limestones of the Viola Springs and Welling Formations, and Ordovician Sylvan Shale. In general, the more the end-use depends upon chemical composition and purity, the greater the price that can be demanded for such-minerals. The not-so-successful attempts to mine zinc, manganese, copper, and iron from the Arbuckle Mountains also will be discussed briefly, and I will provide a sermonette on analytical chemistry and its usefulness to corporate CEOs. I may be one of the few persons at this Forum on the Geology of Industrial Minerals who is not a professional geologist. I am an analytical chemist specializing in geochemistry. In particular, I investigate trace elements in rocks and their consequences in industrial processes. As such, I have consulted to the mining industry (and their clients) for two decades, always with the intent of improving service and product quality.

Jan 1, 1999

By C. McLean, S. G. Kolb, A. J. Erickson, S. C. Moore

The Grossschloppen vein-uranium deposit, Bavaria, West Germany, was examined with underground workings during 1980-82 by Esso Erz GmbH, an affiliate of Exxon Minerals Company (EMC). Geologic evaluation entailed dense drilling of a portion of the deposit from workings constructed specifically for the program. Discovered in 1977, the deposit was initially explored by surface diamond drillholes which allowed definition of a 30-60 m wide vein system discontinuously mineralized along a 1000 m strike length and to at least a 450 m depth. The underground program was conceived as a cost-effective procedure to answer questions on vein correlation, grade continuity and variability. A 1200 m decline allowed access for detailed sampling of approximately 10% of the known area of mineralization. Fanned drillholes, logged by gamma probe, were spaced to provide intersections of veins at 10 to 20 m intervals. Six cross-cuts also penetrate the pitchblende and uranophane mineralization which occurs in 0.1 to 2.5 m thick quartz veins. Detailed cross-sections and level plans were constructed for resource estimates of the intensively studied portion of the vein system. The program resulted in the discovery of local, high grade areas and an average grade in the evaluated area nearly double that expected from surface drilling.

Jan 1, 1992

By Eric Forssberg, Mats Hallin

INTRODUCTION Evaluation and optimization of mineral processes has been going on since the the first plant started hundreds of years ago. A huge mass of knowledge has been collected since then. The main problem with this type of empirical knowledge of complex processes is that it is very difficult to formulate and document. The methods used for collecting the knowledge is not consistent but-varies with "he person" that does it. In the late 60's and early 70's efforts were made to use computer based techniques to evaluate process performance and to formulate process models for mineral processes. Models were developed, such as the flotation model by R. P. King (1973) which is still used to simulate flotation circuit behaviour. During the 70's. a number of mass balancing packages were developed together with process simulation packages. The industrial sector also made great efforts to develop and utilize computer based systems for process control and evaluation. The expert system that BOLIDEN created in 1974 at the Kristineberg plant was a remarkable achievement considering the hardware (CDC 3600) and the software tools (FORTRAN) available at that time. The main obstacle to successfully implement optimizing computer control of mineral processes was and still is the limited knowledge about the dynamics of complex processes. There are, however. many activities going on in this area today and it seems that the technology is very near a break through. Process evaluation is the most important tool to improve and maintain the optimal performance of a process. Today dedicated computer based techniques are used for fast and reliable screening of the processes. The stages in process evaluation consist of three basic steps: - Data acquisition, manually or by data logging. - Data refining, mass balancing and parameter calculation. - Data evaluation. The refined data can then be used for model construction/calibration and subsequently for process simulation. This paper describes two studies where the techniques mentioned above are used to evaluate and simulate

Jan 1, 1991

By Ted D. Johnson

INTRODUCTION The report of the workshop on Alternate Energy Strategies estimated that the world's reserve of uranium oxide is about 700,000 tonnes. In order to extract these reserves, it is estimated that between 100 and 200 million tonnes of solid waste and of the order of 1 billion cubic meters of liquid waste will be generated. In addition, radium within tailing will continue to generate radon gas as it decays for many years after closure. Hence, the handling, storage and possible use of uranium mine waste is a matter which deserves serious consideration and planning. Since the properties of the waste will largely dictate the method of handling and storage or use of the waste, the properties of the waste should be determined, to the degree possible, prior to the design and subsequent operation of the mine. Uranium mine waste has both solid and fluid constituents. The solids include soil and rock removed during site preparation, similar material removed during development of the mine and uranium tailing generated by the mill. Soils can be stockpiled for reuse during reclamation or used as earth materials for construction of fills, roads, dams and soil linings. Rock can be stockpiled, used as mine backfill or can be used for earth construction in fills, roads and dams. The tailing is usually placed in a tailing storage area. The fluid constituents are both liquid and gaseous. The liquid waste is generally composed of the mill water; however, additional waste water can be generated by dewatering processes and diversion of surface runoff. The gaseous constituent of the uranium mine waste is radon generated by the radioactive decay of radium in the ore tailing. DEFINITION OF PROPERTIES The following presents a brief definition of some of the more important properties discussed later. The definitions are not intended to be comprehensive, but are useful as a basis for understanding.

Jan 1, 1980

By George M. Potter

During the late 1960s the mining industry became interested in USBM's efforts to develop the gold-silver heap leaching process. In the last 10 years, successful commercial heap leaching applications increased. The process provides a relatively low-capital means of processing treatable ores with reasonable operating costs. Present gold and silver prices compared to capital and operating costs favor the continued growth of the heap leaching method. The heap leaching process is defined as the percolation leaching of relatively coarse gold-silver ore piled on an impervious surface. To date, dilute alkaline cyanide solution is the lixiviant of choice. Other leaching reagents such as thiourea and hypo have been suggested, although they are not currently used in heap leaching practice. Ores may he run-of-mine in size or crushed as fine as about 10 mm. Gold and silver-bearing so¬lutions draining from the heaps may be processed in activated carbon columns to recover the metals, followed by stripping and metal recovery, or treated by the Merrill-Crowe zinc dust precipitation method. All solutions are generally recycled to the leaching step, and the heap leaching system is relatively easy to control from an environmental viewpoint. One advantage of the heap leaching process is its flexibility. Plants may be very simple, comprising leaching of uncrushed ore on a locally-obtained clay-lined pad and shipment of gold and silver recovered in the form of loaded carbon or zinc precipitates; such a plant would have a relatively low capital investment. More elaborate plants for larger ore bodies may be planned to include crushing, cyclic pad use, and done gold-silver recovery. General planning is similar in each case. This article outlines some key factors in designing heap leaching operations and discusses how these factors are applied to develop a working operation. Since individual design requirements and local costs vary so much, only a few general comments on costs are included.

Jan 3, 1981

By C. L. Schneider, R. P. King

The ability to predict the liberation spectra that will be produced by a closed-continuous grinding circuit is of fundamental importance for the design, optimization and control of most operations in an ore dressing plant. The only known approach through which such a calculation can be performed is the population-balance-model equation. The latest, and certainly most significant, advance made in this area is the solution proposed by King (1990), which requires one textural parameter measured directly from the ore being studied. The main advantage of King's solution is that, unlike the conventional solutions, it does not require the estimation of hundreds of arbitrary parameters that cannot be independently measured. However, King's solution relies heavily on a knowledge of the detailed internal structure of the Andrews-Mika diagram (Andrews and Mika, 1975), which has never been measured for any ore. The characterization of the internal structure if the Andrews-Mika diagram requires the measurement of the liberation spectra at several progeny sizes produced by the breakage of monosize-monograde particles at several different grades. In this paper, a procedure based on magnetic fluid fractionation for generating seven narrow-size, narrow-grade samples is presented. Each of these samples was then broken into six progeny size classes in an ultrasonic mill. The one- dimensional liberation spectra of each of the 42 progeny and seven parent particle samples were then measured by image analysis. The measured linear-grade distributions are presented in this paper. The calculation of the correspondent volumetric-grade distributions is discussed.

Jan 1, 1995

By S. A. Kupferman

Cement raw materials used in southern California and Arizona cement plants occur in a wide variety of geologic environments. Limestone, the primary material making up 75-80% of the raw materials mix used in cement manufacture, is mined solely from Paleozoic Era deposits. During this time, 570-245 million years ago, vastly more limestone was formed in California and Arizona than at any other time (continent was in southern latitudes). The general geology of each limestone deposit associated with the eight southern California and two Arizona cement plants will be reviewed. Many of these deposits are geologically complex, having been subjected to Mesozoic deformation, metamorphism and instrusive events, along with Cenozoic uplift and deformation. All of the cement plants in this region require the use of secondary additives in order to meet the chemical requirements for cement. The kinds of additives vary greatly, depending upon the chemistry of the limestone deposit, cement specifications, and the cement manufacturing process. In some cases, clay deposits with suitable chemistry on or near the limestone sources are utilized. Where alkali limitations or need for alumina-rich materials exist, then specialty clay deposits are used. Iron-rich sources in the region include iron ore deposits and slag or mill scale from local steel mills. The geologic setting and location of these additives, along with gypsum and silica-rich deposits presently in use, will be mentioned.

Jan 1, 1987

By Byron Belew, Juris R. Harlamovs, D. B. Dreisinger, John L. Watson, G. A. Kordosky, Thomas O. O’Keefe

Unwanted impurities, present in raw materials, are often coextracted along with desired constituents during leaching. Due to its abundance and ease of extraction, the presence of ferric iron in leach solutions, solvent extraction circuits, and subsequently in electrowinning circuits, has always been a problem. In the case of zinc hydrometallurgical processing, iron is frequently an impurity in leach solutions. Purification of such solutions by solvent extraction, using di (2-ethylhexyl) phosphoric acid (D2EHPA) in a kerosene diluent, results in ferric ions being readily extracted along with zinc into the organic phase. Ferric iron has proven difficult to strip from the organic phase, but it is known that ferrous iron may be easily stripped from D2EHPA, and hence a process to effect reduction of iron present in the organic would have great potential. This research investigated a process to directly reduce ferric to ferrous iron in the organic phase, and to strip the ferrous iron using a dilute acid solution. The process utilized metallic zinc as the reductant and a low concentration sulfuric acid stripping solution. The results of the laboratory tests illustrated that iron could be readily removed from the organic phase by simultaneous direct reduction and acid stripping. The reductive stripping process was subsequently utilized in industry at the laboratory and pilot plant levels. The results of the industrial tests were encouraging and demonstrated the potential of this novel concept.

Jan 1, 1993

Sweden is beginning to reap the rewards of a process that has transformed a country with a history of metal mining into one of Europe's most vibrant exploration regions. Exploration and mining companies from North America, Europe and Australia have joined local firms in the search of gold, base metals and diamonds. This has been due to changes in Sweden's mining laws in the early 1990s. These changes opened the way for unhindered private sector investment in the country's mineral industry. More than 20,000 km2 (4.9 million acres), mainly in northern Sweden, is now held under granted or applied exploration claims. Exploration permits diamonds, applied or granted covers more than half of Sweden's surface. This is a greater proportion of the country than ever before. Major mining companies and junior explorationists are evaluating a geological potential that bears favorable comparison to other major mining centers. Historically, Sweden's gold re- sources occur in greenstone belts in northernmost Sweden and in association with massive sulfide deposits in the Skellefte district. These are receiving particular attention by exploration companies. In addition, new areas in Sweden are also being explored for gold. This increases the probability that Swedish gold production - already the highest in Europe - will increase in the future (Table 1). Sweden's recorded mining history goes back for much of the last large pool of expertise in exploration, mining, equipment manufacturing and minerals industry research. In addition, the Geological Survey of Sweden provides sup- port for exploration in terms of data archives, processing and publications. Other benefits for exploration companies are Sweden's communications infrastructure, its member- ship in the European Union, bringing access to one of the world's major markets and the high level of educational, technical and language skills held by the country's work- force. And deregulation of the economy and the introduction of competitive corporate tax rates have made Sweden a favorable destination for investment in Europe.

Jan 1, 1997

Compared to alternative ways of dealing with coal fines, oil agglomeration has always been a commercial bridesmaid and never yet a bride. Yet interest is high in the process, as shown by a number of pilot plants in operation that are funded by industry as well as government. In the past in the US, alternative ways to handle coal finer than 100 to 200 mesh included flotation, or no treatment except dewatering, or discarding the fines as waste. There are now a number of factors that tilt the economics more toward fine coal recovery. Today, the cost of mining a ton of underground coal runs from $22-30, and discarding even 10% as unrecovered fine coal is a real economic loss. A number of attempts (notably by EPRI) have been made to quantify the effect of coal cleaning on utility boiler operations. Transportation and ash handling costs, frequency of unscheduled outages, fouling and slagging problems, and boiler tube erosion are all lowered by cleaning coal. The advantage of coal cleaning on these parameters has been shown, but the dollar amount varies with many assumptions and is difficult to pin down. The message to a utility, however, has not been lost and the trend toward cleaner steam coal is accelerating. This may lead to fewer installations that only wash the coarse coal, and add back the unwashed fine coal until a maximum contractual ash limit is reached. And, as the proportion of coal fines continues to increase by continuous and longwall mining, it may become impossible to clean the coal sufficiently without cleaning the fines. Flotation and oil agglomeration are the only processes that presently deal directly with cleaning the very finest coal fractions. Flotation has proven its effectiveness in cleaning blackwater and in separating fine coal from mineral matter. It has the achievable potential for far greater pyrite rejection than is now obtained in commercial preparation plants. While flotation will continue to be the predominant method for fine coal cleaning, there are conditions and opportunities where oil agglomeration can be a

Jan 8, 1980