Search Documents

By Paul M. Jr. Musgrove, Donald C. Moore

Conventional Smelting Practice Conventional copper smelting practice varies from smelter to smel¬ter, but generally consists of some or all of the following unit processes: roasting, smelting, converting, and fire refining. Roasting. Copper sulfide concentrates can be smelted directly or after an initial roasting step. Roasting is used in some smelters because roasting prior to smelting increases smelting capacity, less energy is required to melt hot roaster calcines than wet sulfide concen¬trates, roaster off gases are high in Sot concentration, 5-15% SO2, and some volatile impurities are removed from the concentrate prior to smelting. However, many smelters do not use roasters, because the problems associated with handling hot dry calcines outweigh the advantages mentioned. Concentrate roasting is performed in multiple hearth or fluid bed roasters. If the moisture is low, roasting can be performed autogenously, usually at 500-600°C. High roasting temperatures are avoided because excess oxidation of the iron compounds may lead to magnetite formation. Magnetite is detrimental to reverb operation because mag¬netite can combine with refractory minerals to form a highly viscous slag. This slag prohibits efficient matte-slag separation and leads to excessive copper losses. Also, magnetite can settle through the matte layer, deposit on the furnace bottom, and consequently reduce furnace capacity. Roasting is carried out only on sulfide concentrates prior to smelt¬ing in reverb or electric furnaces. For smelting processes, such as the flash and continuous that rely on the exothermic heat of oxidation of the sulfur minerals, roasting is not practiced. Reverberatory Smelting. The predominate copper smelting fur¬nace for the past 50 years has been the reverb. These furnaces are typically 100-120 ft long, 30-35 ft wide, and 12-15 ft high. A typical furnace layout is shown in Fig. 2. Refractory brick linings cover all internal surfaces of the furnace. Originally the flame was directed to reverberate or reflect off the furnace ceiling and melt the feed material. Current practice is to direct the flame down the furnace length to melt the concentrate. A method of charging the concentrates or calcines, generally along the side walls to minimize refractory erosion, is incorporated in the furnace design. The copper concentrates, calcines, and fluxes charged into the reverb undergo a series of complicated reactions as the temperature of the mixture increases. The reaction of the iron and copper sulfides with the oxygen in the furnace produces a molten Cu25-FeS mixture called matte. Copper smelting metallurgy is based on the fact that sulfur has a greater affinity for copper than for iron and most other common metals. Therefore, in a system containing copper, the copper will preferentially remain as a sulfide compound until all of the other metals have been oxidized. The oxidized metals combine with silica to form a silicate slag that floats on the matte and is removed from the system. Reverberatory furnace smelting chemistry can be approximated by the following chemical equations: FeS2 + O2 - FeS+ SO2 (1) The formation of FeS ensures that any copper present other than as sulfides will be reduced by the relationship: CuO2 + 2FeS + O2 - CuS + 2FeO+SO2 (2) or 2Cu +FeS - Cu2S + Fe (3) As the molten charge travels down the furnace, continued oxida¬tion of the iron minerals and sulfurization of the copper minerals occurs. When all of the copper has been converted to sulfides, the iron sulfides can then be further oxidized as: FeS + (3)2 O2 FeO + SiO2 (4) The FeO reacts with the silica added as flux in the furnace charge. A simplified equation is: FeO + SiO2 -FeO SiO2 (5) The iron silicate slag formed is skimmed from the surface at the end opposite the burners. The copper content of reverb slag is usually less than 0.6% Cu and is discarded. Matte is removed along the side wall and is taken to the converter for oxidation of the remaining sulfur and iron. The main objectives in reverberatory smelting are to produce a molten Cu2S-FeS matte containing 30-60% Cu and a throwaway slag. Production of matte permits complete conversion of all copper minerals into copper sulfides, which can migrate because of specific gravity differences, through the lighter slag layer. Also, the molten matte droplets collect the noble metals, gold and silver, as the matte settles in the furnace. The large settling area of the furnace provides enough separation time to produce a low grade slag, which can be discarded without further processing. High heat losses are associated with reverberatory smelting be¬cause of the large volume of gases sweeping through the furnace. Therefore, an outside source of heat is required to keep the smelting reaction going. Natural gas, fuel oil, or pulverized coal are used as this heat source.

Jan 1, 1985

Solvent extraction-electro- winning appears to be one of the most important processes for the future in the North American copper industry and for this reason both the SX and EW steps are furtive areas for development. This paper has the objective of reviewing the present state of the electrowinning technology. It discusses some recent equipment developments and concludes by discussing the direction the technology will likely go fn the next decade.

Jan 1, 1985

By H. F. King

In 1982, the result of a 3-year study was published as "A Guide to the Understanding of Ore Reserve Estimation". The principal conclusions were that computation formed only part, and perhaps not the most important part, of an estimate and that the estimate should recognize that geological, mining engineering and metallurgical judgements formed an essential part of the eventual result. Subsequent expectation/realization studies of several mines have produced case histories which form the principal basis of the paper. These imply that geology is even a larger factor than thought earlier.

Jan 1, 1985

By D. B. Stephens

Milling operations throughout the world have created vast amounts of waste by-products, or tailings, which are often disposed on the land surface. The wastes may be disposed behind dams, on untreated ground, or on compacted clay or synthetic liners of impoundments and trenches. Often one of the principle concerns of environmental regulatory agencies is whether seepage from the waste pile could move through the vadose zone to the water table and possibly contaminate an aquifer. The seepage may be generated by the drainage of liquids initially deposited along with the tailings or by infiltrating meteoric water which leaches solutes from the tailings. The purpose of this article is to discuss some of the commonly held assumptions regarding storage of seepage wastes in the unsaturated zone. The significance of recent studies of water movement in dry climates which pertain to tailings site selection will be presented. The paper is relevant not only to seepage from impoundments at mill sites, but to many other fluid waste impoundments as well in part this article reinforces earlier work by Stephens and Siegel (1980) and an excellent review paper by McWhorter (1985). The present report addresses considerations in porous media; for a discussion of the effects of fractures and anisotropy in the unsaturated-zone, refer to the paper by McWhorter (1985).

Jan 1, 1985

By Kenneth R. Hancock

The importance of "Mineral Pigments" can best be shown by noting that the world production is estimated to be in excess of 47 billion kilograms (52 million st). This quantity if pack- aged in the common 22.6 Kg (50 Lb.) bag and placed end to end would circle the world over 500 times. This world production figure is based on a broad concept that pigments, extenders, and fillers have essentially similar functions. In this paper mineral pigments are products produced synthetically or from selected minerals and through pulverization have characteristics of pigments.

Jan 1, 1985

Tunnel damage from explosion and earthquake excitation is correlated with strain and particle motions. Critical strains and motions are then correlated with focal distance, which is so small that permanent strains are highly likely. The shaking process is then numerically modeled to investigate the effects of excitation frequency and joint orientation. The models confirm that near cavern blocks slide intermittently only at relatively high particle velocities during periods of low joint normal stress. Larger displacements occur with relatively small shear wave lengths or relatively high frequencies. Variation in joint orientation and rock mass dilation result in changes in displacement.

Jan 1, 1985

By C. Basilio

Thermodynamic calculations based on a rigorous mass balance technique have been carried out for the chalcocite-potassium ethyl xanthate (KEX) system. The results of the calculations are used to: i) construct Eh-pH diagrams for KEX additions ranging from 0 to 10-4 M, ii) determine the amounts of cupric ions and copper xanthate formed at different Eh and pH values to make three-dimensional diagrams, and iii) determine the minimum KEX addition required to form CuX and CuX2.. This allows prediction of the optimum conditions for flotation and comparison of this with microflotation test data. In addition, voltammetry experiments have been conducted on chalcocite electrodes at pH 9.2 and pH 6.8 in order to gain information on the role of metastable species in flotation. On the basis of these experiments, possible reaction mechanisms are discussed and comparisons are made with what is expected on the basis of thermodynamics.

Jan 1, 1985

By E. P. Lunghofer

Silicon Carbide and Aluminum Oxide were two of the early forms of man-made abrasives often referred to as synthetic and were invented at the turn of the century. Edward G. Acheson invented Silicon Carbide in 1893 and Charles F. Jacobs filed a patent for the manufacture of fused alumina from bauxite in 1900. Jacobs' patent provided a method for tempering the alumina by reduction and fusion to a degree of hardness and toughness suitable for abrasive purposes. These two inventions are the basis of the synthetic abrasive industry today. Improvements have been made in the equipment used in the process, however the process itself remains basically the same. In 1904 A. C. Higgins observed that hot spots developed on the shell of the pot furnace used in the reduction of bauxite. He found that by spraying water on the hot spots they could be prevented and, in addition, they failed to return in the same place. Higgins reasoned that the cooling effect of the water caused the melt contacting the metal shell to freeze as an insulating autogenous refractory. This suggested that an improved furnace design could be obtained by using a metal shell which is continuously flooded with water on the exterior surface. Paradoxically, a furnace required to contain tons of molten Aluminun Oxide was thus constructed of nothing but a sheet steel shell with no refractory lining at all. A test of such an arrangement was tried by Higgins with considerable trepidation and was found to be extremely satisfactory.

Jan 1, 1985

By G. Norman Owen, Roger E. Scholl

INTRODUCTION The analysis and design of underground structures to resist vibratory motion due to earthquakes has been of interest to engineers for several decades. The dynamic stability of a powerhouse cavern was considered in the 1960s (Campbell and Dodd, 1967). About the same time, the designers of the San Francisco Bay Area Rapid Transit system developed procedures for analyzing the subaqueous tunnels placed under San Francisco Bay and the tunnels excavated along the land portions of the alignment (Kuesel, 1969). Japanese designers developed fairly sophisticated procedures for analyzing subaqueous tunnels during the late 1960s and early 1970s (Kuribayashi et al., 1974). In the past decade, proposals to site important critical underground facilities in regions with moderate to high seismic activity have increased the interest in earthquake engineering of underground structures. Seismic designs or seismic studies have been conducted for rock caverns for storage of petroleum or location of power plants (Yamahara et al., 1977; Glass, 1976), buried tanks for fuels or rad¬ioactive waste (Fukushima et al., 1977; URS/Blume, 1976), tunnels for disposal of nuclear waste (Yanev and Owen, 1978; Wahi et al., 1980), and urban mass transit tunnels (Monsees and Merrit, 1984). One of the greatest drivers of the interest in seismic analysis and design of underground structures has been the proposals to dispose of high-level nuclear waste in mined repositories some 300 to 600 m (approximately 1,000 to 2,000 ft) below the ground surface. In 1979, the authors, in conjunction with Professor Tor Brekke, presented a paper on an evaluation and design procedure used in a study of a proposed nuclear waste repository (Owen, Scholl, and Brekke, 1979).

Jan 1, 1985

By H. Nakazawa

Galvanic contact, occurring between electricall y conducting minerals in flotation pulps, may affect the nature of mineral surfaces and influence their floatabilities. Rest potential and galvanic current measurements were made on an electrode combination of pyrite and pyrrhotite under nitrogen, air and oxygen aeration conditions, and the results were correlated with their oxygen-consuming characteristics and flotation behaviors. The galvanic contact improved the floatability of pyrrhotite, but adversely affected that of pyrite. Auger electron spectroscopy, x-ray photoelectron spectroscopy and scanning electron microscopy indicated that the formation of hydroxide or oxide and sulfate species of iron on pyrrhotite decreased whereas that on pyrite increased when these minerals were held in contact. An electrochemical mechanism was proposed for the observed behaviors.

Jan 1, 1985

By Sam M. Pickering

Most high, purity silica is produced from crystalline quartz-generally from sand. However, there is a small area of southern Illinois where a porous, friable white chert rock is found which is very different in physical form from quartz sand. This microcrystalline material, commonly referred to as tripoli, is more easily ground to micron sizes than sand. Although of similar Moh's hardness, microcrystalline tripoli is less abrasive than either ground quartz sand or many synthetic abrasives. Location The southwestern corner of Illinois is underlain by extensive deposits of very porous chert; in some places the material is quite white. Two companies - Illinois Minerals Co. and Tammsco, Inc. - mine and process this white material. The Illinois Minerals plant is located at Elco, and the Tammsco plant is in Tamms, Illinois. Both companies mine from various locations within a ten mile truck haul of their plants. Production from this district has been continuous since at least 1915. Malvern Minerals Company mines and processes a generally similar microcrystalline, friable novaculite near Hot Springs, Arkansas.

Jan 1, 1985

By R. L. Roller

In recent years, there have been many articles and papers written on the use of specially designed belt conveyor systems in mining operations, either economic studies or actual experience. These papers generally conclude that the use of these systems can reduce overall costs from 10% to 40% over conventional methods using general purpose equipment (truck/shovel, scrapers, etc.). However, they also generally conclude that conveyor systems are inflexible and capital intensive. With the increasing cost of mobile equipment, the capital cost conclusion is losing steam, particularly when the costs of maintenance shops, road and ramp construction and other related start-up costs are thrown in. Attention here will be focused on answers to flexibility related questions, but first, a look at costs in general. Basic production economics tell us that when we have a small job to do, it is cost effective to use simple, low cost, general purpose tools which will have a minimum impact on total cost. It also tells us that when we have a large project, we can employ methods using special, relatively expensive production tools. Figure 1 illustrates what occurs on projects, of different sizes, with various quantities of production. With small quantities of work, the cost per unit of work starts out much lower using general purpose equipment, rather than using special purpose methods and equipment. However, at a relatively low quantity of work, the cost hits a lower limit and, regardless of how great the quantity of work, the unit cost can never go any lower.

Jan 1, 1985

By F. C. Bond

History The breaking and shaping of rock was one of man's earliest occupations. In the Paleolithic Age long before the dawn of history, arrow¬smiths and the makers of stone axes, hammers, knives, scrapers, spears, and borers were highly respected members of society. In early historical times stones for building blocks, roads, and city walls were shaped by slaves and convicts, who also did most of the mining. However, great artists erected beautiful stone sculptures, while gifted architects planned imposing temples and monuments. Until well into the 19th century nearly all rock was broken laboriously by hand. The small rock required by John MacAdam for his macadamized roads in England in the 1820s was produced by women and boys seated alongside the roadside with hand hammers and legs wrapped with rags. Eli Whitney Blake, a nephew of the Eli Whitney who invented the cotton gin, developed the first successful jaw crusher before 1870. The gyratory conical crusher soon followed. Comparative tests established its large capacity advantage over the jaw crusher, as well as its greater cost for a given feed size. Both types have been in use for more than 100 years. Crushing rolls appeared before 1900. Thomas A. Edison made very large diameter rolls which were excessively long; they failed because of shaft deflection. Various types of disk crushers and edge runner rolls appeared about this time. The older methods of reducing rock were adaptations of other processes. The stamp battery of dropping weights effected crushing by simulating heavy hammer blows; the much earlier arrastre, in which heavy stones were dragged in a circular path over the ore by animal power, came from the prehistoric method of grinding grain between two rubbing stones, while the jaw crusher was adapted from simple squeezing devices. But the tumbling grinding mill was not just an adaptation; it was an invention, because it required thinking on a somewhat higher order-there was no prototype. Its nearest antecedents were probably the small closed tumbling drums used in England more than a century ago for cleaning and polishing small iron castings. The date of the first tumbling mill actually used to grind rock is unknown, but it was later than the American Civil War (1861¬1865). It was almost certainly a closed or batch mill in which rock was placed and rotated until it reached the desired particle size. It could have been operated either wet or dry. The first published refer¬ence to such a batch mill was one introduced by Alsing in England (1870) for the grinding of calcined flints for pottery work.21 There are several rather indefinite reports of grinding mills in the early 1890s, including an overflow ball mill in the Helena and Livingstone reduction plant in Montana which may have been the first of its kind. 11 Many of the first mills, which were called tube mills, used hard rock both as grinding media and as mill lining. The rock used was preferably stone from the Normandy and Danish beaches, when it could be imported. This remarkable siliceous stone was already widely used for grist mills throughout America, and its resistance to wear was greatly respected. The decade of the 1890s saw the development of tumbling mills with continuous feed and discharge and their extension into different industrial uses. By 1895 some experience had been accumulated. Iron grinding balls were being tested and the proper speed of rotation was being determined. The Clark Patent tube mill was featured in an E. P. Allis bulletin of 1890, which may have been the first published description of a tube mill. More than 1,000 Gates tube mills had been built by Allis-Chalmers before 1913. Many of these were used in gold mining, espe¬cially in South Africa. The 5 x 22-ft size was particularly favored for grinding portland cement; the use of tumbling mills in the manufac¬ture of cement began about 1900. A great deal of attention was paid to the mill lining. Metal was then relatively expensive, and the general approach was to trap some of the rock grinding media into mill lining pockets. This rock would then absorb the wear and protect the metal lining. In the first ten years of the 20th century there were several different types of pocketed liners, with different manufacturers advancing the superior claims of their patented arrangements. The Osborne liner, developed in South Africa, was probably the most successful. 21 Another item which attracted much attention was size classifica¬tion within the mill and in ancillary equipment attached to the mill and rotating with it. The Krupp type, with interior screens protected inside the mill lining, was developed quite early in Germany, possibly before 1890. The Dorr reciprocating rake classifier (1907) had not yet been invented, and many strange and impractical screening and classifying devices were proposed. In these unsatisfactory machines the two separate processes of size reduction and classification were combined into one operation. It was many years before recognition came that a machine is most efficient when it is designed for one specific purpose. There was much industrial wastage before the opera¬tions of grinding and classification were finally separated. After 1900 the grinding of portland cement raw material and of cement clinker required large numbers of tumbling mills. Most of the raw material was then ground dry. This was also the heyday of gold mining. The old stamp mills that were used in great numbers for grinding gold ore did not grind sufficiently fine to liberate all of the gold, and the new tube mills were installed following the stamps. After 1910 larger diameter tum¬bling mills with larger grinding media were developed. These could receive the finely crushed ore directly, and the inefficient stamp batter¬ies were gradually eliminated. The Rand in South Africa was the greatest gold producer, treating immense quantities of rather low-grade but consistently free milling gold ore. The first tumbling mills, or tube mills, went into operation there in 1904. They were so successful that within a few years no new stamp batteries were installed in the district, even though old ones continued to pound away until after 1950.3 The early tube mills on the Rand all employed Normandy or Danish pebbles, which had to be imported at considerable expense. Their reported wear was as low as 4 lb per ton ground. 4 Many of the mills were lined with the same tough Danish stone cemented into place, while others used the pocketed steel Osborne liner. It was in 1907 on the Rand that an important test was made using hard native ore for grinding media in place of the expensive imported pebbles. 3 This ore, called banket, did not wear as well as the renowned Danish pebbles, but the cost per ton of grinding was definitely reduced. Many of the tube mills on the Rand were soon grinding with native ore. This was the beginning of the development now called autogenous grinding, in which the ore grinds itself. This is treated under a separate heading. See Subsection 3C, Chapter 4. Gold mining was important in America also, and its grinding history follows that of the Rand. Danish pebbles were replaced by native ore in Santa Gertrudis, Mexico, in 1913, 4 and in Consolidated Gold Fields, Nevada, in 1914. 5 Other properties followed suit. However, it gradually became apparent that the capacity of a given mill could be almost doubled if rock grinding media were re¬placed by cast iron or steel grinding balls. In order to increase plant grinding capacity many rock media mills were converted to iron grind¬ing media in the second decade of the present century. In some mills the motor size was doubled; other mills were cut in two and another motor was provided for the second half. Grinding mills began to assume a more modern appearance. Crushing rolls were formerly much used following crushing in jaw or gyratory crushers and preceding grinding in ball mills. How¬ever, the roll surfaces wore rapidly, and skilled maintenance was required to obtain even wear. Rod mills could take feed of the same crusher product size and reduce it finer. The first rod mill was con¬structed by Mine and Smelter Supply Co. and was tested in Canada

Jan 1, 1985

By N. Rowlands

The importance of mineralogical and chemical characterization of fine particles in the fields of mineral and metallurgical processing is discussed. Emphasis is placed on electron beam analysis techniques including electron diffraction and energy dispersive analysis with special reference to the scanning trans- mission electron microscope. The correct choice of instrument depending upon particle size, composition and nature of sample is stressed and the analytica1 capabilities of different types of instruments are evaluated. Recent applications of electron beam equipment to processing problems are presented and the paper concludes with a review of new developments in the fields of surface analysis and micro-diffraction.

Jan 1, 1985

By R. A. Cummings

From before the turn of the century until as recently as the 1950s, relatively shallow coal reserves near the margins of the Illinois Basin were extracted by room- and pillar methods. Subsequent development has, in many cases, extended onto ground with high subsidence potential. The same state of affairs exists in many other historical coal mining districts, as well as in Illinois. Pit subsidence in Illinois is particularly common in the towns of Streator, Carterville, and Herrin, although there are others. Pit subsidence is characterized by distinct circular or oval depressions and normally is associated with shallow mining and thick soil relative to the thickness of rock.

Jan 1, 1985

By R. D. Rein

The Henderson mine is a large continuous panel caving operation in central Colorado. The orebody is roughly 760 m by 910 m and will be extracted from two production levels. Each level is divided into three north - south panels 244 m to 317 m wide. The general arrangement of a production level and its associated undercut drifts is shown in Figure 1. The production drifts (3.6 m by 3.7 m) are spaced 24 m apart center to center and have undercut drifts (3.4 m by 3.8 m) 17 m directly above them. The production drifts are interconnected with drawpoint crosscuts (4 m by 4.3 m) driven on 12 m centers. Drawpoint crosscuts receive substantial brow steel and concrete support prior to undercutting. The brow steel is a 15 cm by 46 cm wide flange beam with a 2.5 cm wear plate to box in the 'cave' side of the beam. The concrete is a minimum of 0.3 m thick and the batch design results in 27 500 kPa strength. The finished drawpoint opening is 3.7 m by 3.2 m with a choke area of 3.7 m by 2.4 m. (Figure la) Undercutting and bell formation is accomplished by blasting rings of 7.6 cm holes drilled from the undercut drifts. The rings have a burden of 2.0 m and six rings develop one complete bell.

Jan 1, 1985

By John F. MacDonald

This paper describes the driving of a 400 m long by 25 m diameter highway tunnel through sensitive, overconsolidated silts and clays. The design concept requires the construction of the permanent support system, consisting of 24 concrete filled perimeter drifts, before excavating the mass core. General construction procedures used during the construction of the perimeter drifts are described.

Jan 1, 1985

By M. K. McCarter

A unique telemetry system was developed to monitor potentially unstable natural slopes above a residential area. Continuous interpretation of data transmitted from remote movement-sensitive devices and immediate communication of signficant changes to emergency personnel were made possible by a readily available "home" computer. Even though the computer based system was originally developed for a non- mining application, it could be used with little or no modification to monitor ground control instrumentation in open pit and underground situations.

Jan 1, 1985

By A. H. Morris, W. Harrison

1. INTRODUCTION Roadheaders have a shadowy origin - probably in Eastern Europe - but were introduced to Europe and the West by the importation of a Russian PK3 machine by the National Coal Board in 1961. A number of features rendered this type of machine an attractive tool for coal mine operators: its flexibility of operation, its combined cutting and loading ability, the opportunity it provided of making any shape of roadway, as well as its ability to cut some rocks. Encouraged and sponsored by the National Coal Board, the development of roadheaders was taken up in Great Britain, notably by Anderson Strathclyde and Dosco; in Europe, predominantly in numerical terms by Voest Alpine and Paurat. Over the 24 years since the importation of the first PK3, evolutionary development has taken place to the extent that machines are now available in many forms up to 120 tons in weight. Machine design has altered, partly as a result of increased power requirements and partly in line with the use of a ripping or a co-axial head. However, the basic concept remains: a boom containing a motor and transmission gear, with a driven cutting head at its extremity and capable of universal movement urged by hydraulic rams mounted on the boom turret and on the machine deck. Generally, the turret, hydraulic powerpack and all other necessary equipment are mounted on a chassis, which contains a discharge conveyor, the whole being carried on caterpillar tracks so that a complete machine entity is provided. Figure 1 shows how the weight, power and - non-proportionally - the cutting ability of machines have increased over the years. From

Jan 1, 1985

By T. J. Jr. DeMull, F. G. Miller, J. P. Matoney

For some time a need had existed in the minerals processing field for a relatively efficient separator that would treat high tonnages of particles in the intermediate size range, i.e., those particles too large for froth flotation and too small for conventional gravity-type separa¬tors. Among those devices developed to meet this need are the centrifu¬gal specific gravity separators. These devices employ centrifugal acceleration to assist gravitational acceleration in separating light¬density minerals from heavy-density minerals. In the category of centrifugal specific gravity separators are the heavy-media centrifugal separator and the water-only cyclone. The two major centrifugal heavy-media separators, i.e., the heavy-media cyclone and the DynaWhirlpool, as well as the water-only cyclone, are discussed in terms of: design features, operating variables, operat¬ing data, and flowsheet design criteria. Examples of plant applications are given in the field of coal processing as well as the processing of other minerals such as iron ores, potash, and tin. Finally, the subject of the staging of centrifugal separators and their use in combination with other separators is discussed. PRINCIPLES For coarse sizes of minerals, efficient specific gravity separations have been possible for many years with open-bath vessels using the natural settling velocity or buoyancy of the particles. These bath ves¬sels process ore by utilizing micron-size solid particles suspended in the slurry fed to the separator. The inclusion of these particles in the slurry increases the effective density of the separating fluid to allow particle separations to be made at densities greater than that of water. However, if vessel size is to remain within economical limits, the particles processed in the bath vessel must have high settling rates in a IG gravitational field. Because of this requirement, heavy¬medium bath vessels are usually restricted to processing +V4-in. sizes. To extend efficient specific gravity separation to smaller sizes, the gravitational acceleration of particles is replaced by centrifugal acceleration. The settling of a small particle in a fluid in a centrifugal force field is similar to that found in a static bath except that the acceleration due to gravity, g, is replaced by a centrifugal acceleration where v, is the tangential velocity at radius r: V=kdm(P-P,), V'. (J) µ In more practical terms where the particles settle in a suspension of finer particles comprising the heavy media and with an effective suspension density p" V = kdm' (P P ) . v'. (2) U r To date the most effective use of this principle has been obtained with devices that rotate a liquid or suspension within a stationary enclosure in order to create centrifugal force. Cyclones are the most common devices used for this purpose, because they generate centrifu¬gal forces far greater than the force of gravity and therefore not only have high capacities but can treat finer sizes than bath-type vessels can. The two main types of cyclones used by industry are the heavy-media cyclone and the water-only cyclone. Also quite widely used is the DynaWhirlpool, which, though based on the same princi¬ple, differs in design from the conventional cyclone. HEAVY-MEDIA CENTRIFUGAL SEPARATORS Like the bath vessels, the heavy-media centrifugal separators em¬ploy media composed of micron-size particles suspended in water. However, the centrifugal force generated in these separators accentu¬ates the difference in settling rate between particles of different density and thus makes possible separations of finer size particles than can be treated in bath vessels. The two most common heavy-media centri¬fugal separators are the heavy-media cyclone and the DynaWhirlpool. Heavy-Media Cyclone Although cyclones were originally developed for use as classifiers or thickeners, it was later found that they could also effectively serve as heavy-media separators.63. 64, Design Features Fig. I I is a schematic of a typical cyclone developed to serve for any one of the following purposes: as a classifier, thickener, or specific gravity separator. The cyclone consists of a cylindrical section joined to a conical section, usually having an included angle of between 14° and 25°. Feed enters the cyclone tangentially through an orifice attached to the cylindrical section. The overflow orifice is located in the base plate of the cylindrical section. The vortex finder, a tube attached to the overflow orifice, extends into the cyclone from the base plate of the cylindrical section. The underflow orifice is located at the apex of the conical section. As some medium together with mineral particles is fed through the feed orifice, a vortex with a hollow air core extending from the overflow to the underflow orifice forms in the cyclone while hollow spray discharges form at each of these orifices. Under the influence of the centrifugal force, high specific-gravity particles move through the medium to the wall of the cyclone and descend in a spiral flow pattern to the underflow orifice. Those particles in the feed stream, lower in specific gravity than the feed medium, follow the major portion of the flow to the center of the core where they are caught in the high-velocity upward central current and are carried out through the overflow orifice. Fig. 12 shows a family of curves that illustrates how materials of varying specific gravity are recovered by a cyclone. Since the specific gravity of the medium is 1.40, the particles of 1.40 sp gr actually act as part of the medium and, regardless of the particle size, split between the underflow and overflow of the cyclone in proportion to the volume split of the medium. Particles higher in specific gravity than 1.40 are recovered in the underflow of the cyclone at increasing rates as the difference in specific gravity increases and the particle size increases. Particles lower than 1.40 in specific gravity are dis¬charged through the overflow orifice at increasing rates as the specific¬gravity difference increases and the particle size increases. However, all the curves originate at the fluid-flow ratio point for the finest particles of any gravity. The fluid-flow ratio is defined as the ratio of the rate of fluid flowing from the underflow to the rate of fluid

Jan 1, 1985