Search Documents

By J. L. Bowlby

OWING to the pressure of routine analyses at the laboratory during the past year, work of an investigational nature on water problems has been greatly curtailed. Favourable reports by the boiler inspectors indicate that the operators are giving satisfactory attention to feed-water treatment at the several . power plants. Some of the operators prefer lime and soda, others the phosphate and soda treatment. Springhill is still on the same commercial treatment as they were using last year, and satisfactory results are being obtained. This individual preference for a particular treatment, although it may be different for feed waters of similar nature, indicates that so long as proper conditions of alkalinity and concentration are maintained within the boiler, it is immaterial what chemicals are used in the treatment. This statement refers, of course, to the low-pressure boilers. In high-pressure boilers, a treatment such as lime and soda cannot be satisfactorily employed, on account of the precipitated calcium carbonate unduly increasing the amount of suspended matter. Only one case of impending trouble was reported during the year. That was at Seaboard, shortly after the new unit was placed in operation. Dissolved oxygen, considerably in excess of the permissible 0.030 cubic centimeters per liter, was detected at various points on the feed line. It was suspected that a de-aerating heater was being operated at too low a temperature, and chemical tests, made before and after the necessary adjustments, proved that this was the source of the trouble. While it has been impossible to give mine waters the attention they require, progress can be reported in the work carried out by the Mechanical Department. Past investigation of the action of mine water on various acid-resisting alloys is now bearing fruit, and it can safely be said that repair work on mine pumps has been reduced by nearly one-half.

Jan 1, 1938

A typical large surface mine in Scandinavia has hard rock, a cold cli- mate, poor ore, a complicated vein type orebody and high labour costs. In order to survive, these mines have to maximise their ore production, minimise waste extraction and get the most out of every man hour. This article describes two examples of how inclined blastholes drilled with hydraulic percussive drills can help to meet these requirements. Both of these Scandinavian mines, Kemira Ltd's Siilinjarvi mine in Finland and Sydvaranger A/S's Bjornevattn iron mine in Norway, use inclined blastholes of 200 mm in diameter. The Siilinjarvi quarry is using inclined holes in production drilling. The Bjornevattn iron mine is blasting the final pit walls with inclined holes. The benefits gained when using inclined instead of vertical blastholes are better selectivity, better bench drilling conditions, steeper final pit walls and increased safety.

Jan 1, 1989

By Sirish L. Shah

It is now common to have archival history of thousands of sensors sampled every second over long time. Yet we frequently have process engineers complain:??.We are drowning in data but starving for information??. How can these rich data sets be put to use? This paper addresses the issue of information and knowledge extraction from data with emphasis on process and performance monitoring including fault detection and isolation. Most of the major plant, factory, process, equipment and tool disruptions are avoidable, and yet preventable fault detection and diagnosis strategies are not the norm in most industries. It is not uncommon to see simple and preventable faults disrupt the operation of an entire integrated manufacturing facility. For example, faults such as malfunctioning sensors or actuators, inoperative alarm systems, poor controller tuning or configuration can render the most sophisticated control systems useless. Such disruptions can cost in the excess of $1 million per day and on the average they rob the plant of 7% of its annual capacity. Over the last decade the fields of multivariate statistics and data visualization and analysis methods have merged to develop powerful sensing and condition-based monitoring systems for predictive fault detection and diagnosis.These methods rely on the notion of sensor fusion whereby data from many sensors or units are combined with process information, such as physical connectivity of process units, to give a holistic picture of health of an integrated plant. Such methods are at a stage where these strategies are being implemented for off-line and on-line deployment. This article attempts to outline the field of sensor fusion - the application of signal processing methods, in the temporal as well as spectral domains, on a multitude and NOT singular sensor signals to detect incipient process abnormality before a catastrophic breakdown is likely to occur Keywords: Process analytics, process and performance monitoring, multivariate statistical analysis, sensor-fusion

Sep 1, 2012

Jan 1, 1949

By T. B. Jarboe

The 150 MWe circulating fluidized bed (CFB) power plant described in this paper is the product of recent Bechtel experience in the industrial power and steam market, where the economic feasibility of fluidized bed projects has been proven. The inherent ability of CFB boilers to burn low grade fuels makes it possible to also burn low ranking coals or coal wastes, including anthracite culm, coal mine rejects, and coal wash plant waste slurry. The impact of fuel characteristics on plant design is addressed. This includes solids handling, steam generation, sulfur capture, and auxiliary load requirements.

Jan 1, 1988

By J. F. Elliott

PRESENT knowledge of the usual metallurgical slags indicates that they are, for the most part, rather complex in behavior and as yet there is no ready means for describing, in a simple manner, the behavior of any one of them. One of the best known slag systems is the iron oxide-silica-lime ternary which is the basic "solvent" in a number of important metallurgical refining operations, the basic open hearth being one of the most important. In this operation, the slag dissolves such components as sulphur, phosphorus, manganese oxide, and magnesia. Considerable study of this slag system and the behavior of these additions has been carried out in the past by a number of authors, as has been summarized in several critical reviews.','2 However, except for determination of the activity of iron oxide, only a limited amount of effort has been directed towards developing, from these data, an understanding of the general behavior of the basic solvent. Reported here are the results from a series of calculations based on data from the literature which permit a semiquantitative evaluation of the activities of iron oxide, silica, and lime (plus magnesia) in the ternary system at 1600°C. The preliminary results, which were reported briefly at a symposium held by AIME in 1953, have been revised and are completed. The steps in the calculation are as follows:* I—establish the activity curves and the curve of the excess molar free energy of mixing at 1600°C for each of the binary systems, 2—construct the activity surface of iron oxide for the ternary from the data on the binary systems and information available in the literature for the ternary area, 3—determine the surface of excess molar free energy of mixing for the ternary system from the activity surface of iron oxide and from the molar curves obtained for the binary system, and 4—differentiate the ternary surface of the molar excess free energy of mixing to obtain the ternary surfaces for the logarithm of the activity coefficients for silica and lime (log rslo, and log rc.~). Si0,-Fe,O: Schuhmann and Ensio have measured the activity of iron oxide in simple iron oxide-silica slags when in equilibrium with y iron. Their data recalculated to 1600°C are shown in Fig. 1. Also included is a point representing a measurement by Gokcen and Chipmana of the activity of iron oxide at 1600°C at the point of saturation with solid silica. For convenience and in accordance with other treatments,' the calculations are based on the hypothetical component, FelO, which is obtained by converting all the analyzed iron in the slag to FeO. In spite of Schuhmann and Ensio's conclusion that the activity of iron oxide in the system does not vary with temperature over the experimental range of 1258" to 1407"C, the data are corrected to 1600°C assuming that temperature does have an effect. It was felt to be most reasonable to assume that the term log rr.10 is a linear function of the reciprocal of the temperature. Reyu has indicated that an effect of temperature on the activities in this system is to be expected from the Schuhmann and Ensio data. In essence, the correction consists of multiplying the experimental value of log rf,,o by the ratio of the experimental temperature in Kelvin to 1873°K. The magnitude of the correction is not large, being approximately 11.5 pct of the experimental value of log rve10. A very minor correction was necessary to compensate for the fact that the slags were in equilibrium with y iron in the experiment, while at steel-making temperatures they would be in equilibrium with liquid iron. Data for the correction were obtained from Darken and Gurry. The standard states established are pure liquid iron oxide (FelO) in equilibrium with pure liquid iron (with the appropriate amount of oxygen in solution) and pure liquid silica. The method of plotting in Fig. 1 is convenient for the calculation of the activity of liquid silica and permits a reasonable extrapolation for the activity of Fe,O in the ranges where no experimental data are available. The uncertainty in the extrapolation to infinity at one terminal where Nvelo = 1 for the usual Gibbs-Duhem integration is reduced considerably by this method. The region of two coexisting liquid phases is estimated to range from 1.8 to 41.7 mol pct Fe,O. The nature of the activity curve for the single-phase region indicates that the activity of iron oxide across the two-phase region is very close to 0.39. Computation of the function log ~F,,o/(1— NF,,o)' for this region (dashed line) in conjunction with the curve through the adjusted experimental data indicate the best probable value of 0.382 for alPe,o in the two-phase area. The line from 0 to 0.018 Nf~~o is obtained by assuming that the component follows Henry's law. In this range, the value for log rveto is 2.59. Appropriate mathematical manipulation of the plotted linet yields the activity curves for the The curve AF", the excess molar free energy of mixing (actual minus ideal), as shown in Fig. 3 is also computed from Fig. 1. This curve is required for subsequent calculations. CaO-Fe,O: The phase diagram for the lime-iron oxide system when in equilibrium with liquid iron is not well known but there appears to be no intermediate compound present. This fact as well as the activity values for Fe,O extrapolated to the CaO-Fe,O binary from Taylor and Chipman' tend to indicate somewhat negative deviations from ideality for the activity curves for the two components. Strong indication of this is evident in Fig. 1 where are plotted the points computed from the estimated activities of Fe,O for the binary system.' It appears that the best line through the data is a horizontal straight line. Because of the general indication of the slight negative departure from ideality, the line is extrapolated horizontally to NF~,o = 0. It is con-

Jan 1, 1956

By Ian Misich

Surface subsidence resulting from "total extraction" of coal in the Collie Basin first occured in 1987 when Western Collieries Ltd trialed the Wongawilli mining method. Because very little quantitative subsidence data existed for the weak, Permian sediments, a research program was set up prior to the extraction of the first panel. The objective of the study was to be able to predict subsidence characteristics and effects prior to mining future total extraction panels. Attempts at predicting subsidence began using existing subsidence models from other coal mining regions, however, as more site information became available, it became clear that it was necessary to develop a separate, site specific predictive model for the Collie Coal Basin. Analysis of extensive field monitoring data for both surface and subsurface subsidence has produced a series of empirical models which can predict the shape of surface subsidence troughs and corresponding tills, and strains for rectangular panels and describe the sequences of ground movement from the mine root through to the ground surface. The fact that both subsurface and surface subsidence characteristics have tied in well, gives greater confidence for predictions made from these models

Jan 1, 1992

By P. R. Kalischer

In the production of alloys by powder metallurgical processes it is often necessary or desirable to include one or more cornponents that tend to form very stable oxides. Included in this group of metals are aluminum, manganese, magnesium, silicon, and a few others. In many cases it is not necessary for the sintered compact to have full density and the other physical characteristics of a cast alloy and in that event no undue precautions need be taken. If, however, it is required that the sintered part be entirely homogeneous, and have maximum density and physical properties, some method must be found to reduce the surface films of oxide, which are always present on the metals, and for the metals mentioned above this is always extremely difficult, if not impossible. since the production of high-density alloys is of particular interest to powder metallurgists, some experiments were undertaken to devise methods of reducing the refractory oxides. For this work, alloys of iron, nickel, and aluminum were chosen because aluminum oxide is one of the most stable oxides and because these alloys in the right compositions form excellent permanent magnets. All of the pertinent properties of the magnets could be accurately and delicately measured by density measuremerits and magnetic tests of sufficient scope to give a complete demagnetization curve. From such a curve the maximum coercive force, residual magnetism, and energy products could be obtained and compared with the same properties for a cast material. Any lack of homogeneity, density, or solid diffusion would be immediately apparent as a decrease in the magnetic properties. Nature and Removal of Surface Barriers The early work of the present investigation was based upon published processes.* In the published information it was claimed that by intimately mixing iron, nickel, and iron-aluminum alloy (50 Per cent Fe, So Per cent Al) powders, compressing such a mixture into the desired shape and size, and then sintering (heating to a temperature below the melting point of the component metals) in hydrogen at 1200°C. or higher, a sound, homogeneous article could be made having the desired physical size, shape, and strength, and good magnetic characteristics. However. this early work showed clearly that something in the powder compress was inhibiting diffusion and the process was almost entirely unworkable. In the initial phase of the investigation samples of iron powder, nickel powder, and iron-aluminum alloy powder were pressed into separate bars and sintered in hydrogen. Both the iron and nickel powders sintered

Jan 1, 1941

By Robert McClure, Frederic Leclercq, Ricardo Chavez

With significant vibration control and fragmentation challenges facing them, the CBS quarry (department of the Nord) successfully integrated a technical partner and mine to mill optimisation to deal with a difficult environment. For the last three years, CBS has been conducting a global production monitoring program similar to the mine to mill concept used in the mining industry. This concept consists of optimising the drilling and blasting while monitoring the effects on the downstream chain (vibration control, loading, hauling, crushing, plant output). Blast optimization is ensured by systematic control of the blast geometry (volumes, patterns, bench and face anomalies, drill deviation) and by the use of constant specific energy blasting. Constant specific energy optimises the work of the explosive in a blasting operation and, when combined with a high accuracy programmable electronic blasting system, reduces the impact of blasting on the environment and enhances blast performance. These techniques ensure consistent blasting results and hence improved site productivity. The improvements achieved have been identified thanks to an effective system designed to monitor the key production parameters.

Jan 1, 2007

By Eliodor Stoian

PRESSURE information has earned a place of recognition to almost everyone interested in oil and gas production. This interest is due chiefly to the multitude of uses to which it can be put. A sub-surface pressure test, conducted for as little as $60, may yield data worth many times its cost. To the uninformed, the testing techniques may seem complex. Fundamentally, however, they consist merely of the application of common sense. At one time, sub-surface pressure data were used only in identifying individual well problems and in finding appropriate solutions. Today, it is generally acknowledged that sub-surface pres-sure .data are essential for the proper evaluation and exploitation of oil and gas pools. The following is a review of significant applications of sub-surface data with regard to: (A).-Development (B) .-Completion and work-overs (C) .-Production (D) .-Reservoir engineering (E).-Proration (F) .Other applications of pressure data of a more general nature.

Jan 1, 1958

By R. W. Lewis

The Federal Bureau of Mines Boulder City Metallurgy Research Laboratory conducted experiments in order to develop a spectrochemical method for determining the concentrations of impurities In high-purity tungsten. This report presents a procedure, developed by these experiments, for determining 19 impurities in high-purity tungsten. Metal samples were thermally converted to the oxide, mixed with graphite, which was prepared and mixed with an internal-standard element, and excited in a d.c. arc. Sixteen of the impurity elements were determined with one set of conditions, using germanium as an internal standard. Lanthanum was used as an internal standard in the tests for boron and chromium. Molybdenum required the substitution of sodium chloride for the graphite and the use of copper electrodes. Data on the precision of the method are included in this report. The coefficients of variation vary considerably but average about 15 percent. Higher concentrations of impurities than those investigated may be determined with some modifications of the testing procedure, provided attention is given to the possibility of interelement interferences.

Jan 1, 1961

By Belcher C. B

A procedure was required for the rapid determination of silicon in low-alloy irons and a suitable method used the formation and reduction of silicomolybdic acid at 50°C. The procedure gives results for elemental silicon and this enables the correct computation of oxidation requirements in basic oxygen steelmaking. The method could be useful in iron foundries which have not installed direct reading spectrometers.INTRODUCTIONOxygen steelmaking experiments required rapid determinations of 0,5-1,6 per cent silicon in iron (3-4 per cent C) within 10 min of sampling; a reproducibility better than ±0.03 per cent Si was desired. Only elemental silicon (Si°) is of interest as a fuel in basic oxygen steelmaking technology, and the proposed procedure determines exclusively this form of silicon. The method is described as it may prove useful to other workers although the interference study would be incomplete for some applications.The analysis of silicon in iron is frequently obtained using direct reading spectrometers, but for the proposed experiments the spectrometers could not be used because of the prohibitive distance from the furnace. It should be noted that spectrometers determine both the elemental silicon and that present as silicon dioxide in occluded slag and inclusions.The silicon content of iron and steel is determined in this laboratory by spectrophotometric measurement of a "molybdenum blue" colour. The reactions involved have been extensively investigated; Strickland (1952) showed that a...

Jan 1, 1965

By Robert Hopler

HUDSON MAXIM A Noted Inventor of High Explosives ALTHOUGH the subject of the present sketch has been a tireless worker in many fields of activity, his name is best known in connection with his work in the development of high explosives and their application to modern ordnance

Jan 1, 2012

By Stone. Jon P.

INTRODUCTION Countries that are currently developing their mineral potential have unique information systems requirements to support these activities. The mining laws usually grant the State considerable authority over the process. In many instances, the mining laws give a government agency the responsibility for granting and controlling exploration permits and development concessions. This is the case in Costa Rica. The Direcion of Geologia, Minas, y Hidrocarburos (DGMH) of the Ministerio de Industria, Energia y Minas (MIEM) is responsible for the management and operation of the National Mining Register. This paper describes the results of two studies of the information system requirements to support the National Mining Registry as well as those supporting some of the other function of the DGMH. The first of the studies reviewed the situation before the installation of computer equipment. The second study evaluated the installed systems. BACKGROUND Historical Perspective Mining in Costa Rica is at present a relatively minor part of the economy. It consists primarily of industrial minerals and gold production. Gold production is largely by small-scale miners and may be more substantial than reported. There is considerable interest on the part of the Government in both having a better understanding of the existing mining industry and promoting future development. Mining law was first provided by the Ordinanza de Mineral (of 1830). This was replaced by the first Mining Code of 1953. Apparently, one of the first problems caused by the 1953 law was the granting of concessions which were not worked. A 1968 agreement with ALCOA also resulted in political conflict. The Government worked on a new Mining Code for many years and in 1982 the current Mining Code was enacted. Draft implementing regulations were prepared in April 1984. Mining Code of 1982 The Mining Code of 1982 addresses most aspects of mining in Costa Rica including the declaration that all mineral resources are to be the property of the State. The State may grant exploration permits and exploitation permits. The exploitation of hydrocarbons, radioactive minerals and certain water resources was reserved to the state. The DGMH was given authority to settle placer claim disputes, maintain a registry of exploration permits, and exploitation concessions. and generally supervise mining activity. Towards this, the Mining Code of 1982 includes rather detailed requirements for the processing of application for permits and concessions. Mining Code Application Requirements The following briefly describes the requirements to give the reader a feeling for the complexity of the process. Currently, over 1,000 active applications, exploration permits and exploitation concessions are being monitored under the process. The number remains relatively consistent in that the number of new applications is usually about the same as the applications that are put aside. The time and date of the filing must be certified by the DGMH. The application must include all the information specified in the Mining Code. If incomplete, DGMH must notify the applicant of the deficiencies or omissions and direct that the applicant correct the deficiencies or omissions within 8 business days (a single 30 business-day extension is allowable). If the applicant does not respond as directed, the application is deemed as not having been filed and the pertinent information is placed in the files. The DGMH must reject the application outright and direct it be placed in the files in the following cases: (a). If the new application covers the area of a previous permit or concession, either being processed or already granted.

Jan 1, 1987

By D. P. Schlick

The German experience in dealing with the health hazard of respirable dust in underground mines is reported from first-hand observation in this Bureau of Mines paper. The report covers-- 1. The control of dust by water infusion, water spraying, and ventilation. 2. The sampling devices (tyndalloscope , konimeter , and BAT-I) and methods used to determine the dust concentrations in mines. 3. The systems of classifying hazardous and safe dust levels in mines. 4. The research facilities devoted to theoretical and applied work on mine dust and pneumoconiotic disease. Tile major research projects on dust and disease now underway in West Germany are outlined in an appendix.

Jan 1, 1970

By Y. L. Su, S. Vongpaisal, H. A. Smith

"It is extremely difficult for a mine's central control room operator to control peakpower demands and to maintain normal materials handling operations at the same time. An investigation of the Falconbridge Strathcona Mine 's central control room has revealed that there are more than eight thousand possible combinations of ""situations"" that can arise. For such a circumstance, an expert system can become a very valuable tool to assist the operator in taking the most appropriate and timely action.A forward chaining expert system module has been developed for this application. The module is based on Level 5 ObjectTM software. It is graphics oriented and user-friendly. Training requirement is minimal.ForewordThis paper has arisen out of work carried out by YLS Associates and Falconbridge Limited as part of a contract between Falconbridge and the Department of Energy, Mines and Resources, Government of Canada, on the application of expert systems in mining operations.IntroductionThe electrical bill charged to large consumers by Ontario Hydro is based upon two elements: demand and energy consumption. The demand is the peak kilowatt level that the consumer has required during a monthly period. It is not uncommon for the demand portion to be as much as one third of the monthly bill. Energy consumption is the total kilowatt-hour s (kWh) consumed during the period.Ways to reduce the energy consumption part of the electrical bill involve utilizing energy-efficient equipment and operating machines in off-peak hours when a lower energy rate is charged. Maintaining the peak power below a critical level may require that simultaneous usage of major pieces of equipment be avoided. A 3000 kW skip run at an inopportune time could cost a company an extra $30,000 in that month's power cost.Potential disadvantages associated with peak power control are lost production due to work stoppage and increased operating costs due to the need to employ extra personnel. Most companies find these matters relatively easy to deal with and with incentives provided by Ontario Hydro, actively pursue this facet of energy savings (Harju , 1991)."

Jan 1, 1993

By Z. Halifa, P. G. Geyer

Tapping a blast furnace is necessary not only to deliver the product, but is also essential for the safety of the furnace and workers in the cast-house. This paper will highlight the practices at ArcelorMittal’s blast furnaces in the Vanderbijlpark Works, with specific focus on tap-hole design and monitoring as well as casting practice. Extension of the tap-hole life and hearth campaign will be illustrated with reference to the methods of maintenance of the tap-holes, troughs, runners, and cast-house equipment. The importance of proper cooling, good thermocouple installation, and temperature monitoring will be explained.

Jan 1, 2014

By George H. Holmes

This report describes the results of a reconnaissance survey of areas in California and Nevada considered favorable to beryllium mineralization. Pegmatite dikes, veins, intrusive sills, contact metamorphic, and disseminated deposits as well as tactite deposits in the former major tungsten-producing districts were investigated. A general geographic pattern of beryllium mineralization was found to be concentrated along a wide belt in eastern Nevada extending from the Ruby Mountains, Elko County, south through the Fish Creek Range, Eureka County, the Snake Range, White Pine County to the Virgin Mountains in Clark County, Nev., and Mohave County, Ariz. Sporadic occurrences of beryllium are in central and southern Nevada, and in the pegmatite dikes of San Bernardino, Riverside, and San Diego Counties, Calif., and the granite masses of the southern Inyo Mountains, Inyo County, Calif. Principal company activities are centered in the Snake Range, Fish Creek Range, and the Virgin Mountains, Nev., and the Inyo Mountains, Calif. Other areas in California and Nevada have been superficially explored with results indicating either limited or widespread, but submarginal, mineralization. Although beryllium was found to occur, to some extent, in many of the deposits, mining has been hindered by the lack of a market for beryllium ores and concentrates other than standard-grade beryl and also by the erratic pattern of mineralization and the high exploration and beneficiation costs required to produce a marketable product.

Jan 1, 1963

By Harley G. Pyles

Rotary Percussion drilling has shown great improvement over rotary drilling, when hard roof is encountered. In addition, Rotary Percussion drills have eliminated the need for air drills in many applications. Improved drilling rates have been achieved but roof bolting is still a bottleneck in many operations and the demand for further improvement will continue. Drilling rates beyond current performance is limited by the capabilities of existing drill steels and bits to resist increased horsepower input. The most apparent immediate solution to improved drilling is to increase the strength and rigidity of the drill steels. Drilling experience indicates that a l-1/2" diameter or 1-5/8" diameter hole can be drilled as rapidly or 'more rapidly than the standard 1-3/8" diameter hole. Greater torque and thrust capacities can be achieved through the use of larger O. D. and I. D, steels, however, this is only possible if the larger hole sizes can be made allowable. Large diameter anchorage shells can be made available at a slight extra cost.

Jan 1, 1965

By D. I. Chalmers

The Brockman rare-metals and rare earth resource is located within volcanic and volcanogenic sediments of the later Proterozoic Halls Creek Group in the East Kimberley region of Western Australia. Exploration including surface trenching, percussion and diamond drilling has defined a major resource of rare-metals and heavy rare earth elements. The host&apos; is a steeply dipping ash-flow tuff, described informally as the Niobium Tuff, enriched in zirconium, hafnium, yttrium, niobium, tantalum, rare earth elements, gallium and beryllium. The mineralization is fine- grained (<20 micrometres) and is the result of alteration and re- mobilization of comparatively simple precursor minerals such as columbite and zircon by fluorine-rich deuteric solutions that were retained in the tuff following eruption. The ultra fine-grained structure of the tuff precludes mineral liberation and separation by physical methods. However, values can be converted to water soluble sulphates by reacting with concentrated sulphuric acid at 250°C. The solution containing aluminium, niobium, tantalum, zirconium, hafnium, yttrium, rare earths and gallium can then be processed to pure products in a sequence of solvent extraction and ion-exchange steps. As a result of these processing techniques, the commercial products obtainable from the resource include mixed rare earth and yttrium concentrates, zirconium-hafnium concentrate, niobium-tantalum concentrate and ceramic grade alumina. Further separation could result in high purity products suitable for ceramic and electronic applications.

Jan 1, 1992