Search Documents

China has abundant bauxite resources, but these resources are mostly of a diaspore, which has a feature of low Al2O3 to SiO2 mass ratio, typically ranging from four to six. Compared to other materials, this kind of raw materials consumes more alkaline and energy in the process of metallurgy, and incurred a high alumina production cost. To solve this problem, researchers in China have done a lot of work on removing of the silicon from bauxite. Under the impetus of these works, in 2003, the first flotation plant of diasporic bauxite in the world was constructed in Zhongzhou, Henan, China. Seven years of operating experience shows that this technology can signifi cantly reduce the costs of following-metallurgy processing. These efforts are an important reference for the processing of similar resource. In general, for Western scholars it is diffi cult to fully understand these works because most of them are published in Chinese. In this paper, a detail introduction for the research progress on China bauxite de-silica flotation was presented, these works involved crystal chemistry, wettability of bauxite minerals, collector and modifi er in bauxite fl otation and flotation practice, etc.

Sep 13, 2010

By A Logan, B Cuthbert

Brown and Chitombo (2007) indicated that during the last two decades, the international mining industry has been using mainly mass mining methods to extract safely and economically some of the important commodities such as copper, diamonds, gold, iron and nickel. Mass mining may be implemented either on the surface through large open pits or underground using mining methods such as sublevel, block and panel caving. Developments in the last ten years have demonstrated that caving methods can be applied successfully in strong orebodies at greater depths and in increasingly challenging geotechnical environments for which more traditional mining methods would be unsuitable, unsafe and/or uneconomic. These underground mass mining challenges require æfront endÆ investigation and analysis, robust engineering design and disciplined implementation and operation. Newcrest Mining Limited has focused its efforts on the development of gold/copper projects of substantial scale and margin. To help achieve this, the company is applying and further developing mass mining methods. These include large-scale open pits (OP) and underground mining methods such as sublevel (SLC), block (BC) and panel caving (PC). Block and panel caving now offer economies of scale opportunities for Newcrest to achieve high production rates at lower production costs; thus enabling the exploitation of mineral endowments at higher margins. This paper discusses the practical aspects of caving technologies that Newcrest continues to develop through its expertise at Telfer and Ridgeway SLC operations, Ridgeway Deeps BC and Cadia East PC projects.

Jan 1, 2008

By Avraamides J, Ritchie I. M

Gold Tellurides are important sources of gold at Kalgoorlie, Western Australia, Cripple Creek, Colorado and Vatukoula, Fiji. Unlike free gold, the tellurides of gold are rather refractory and the usual treatment practice has been to oxidize the tellurides, either chemically or by roasting prior to cyanidation of the liberated free gold. There is a large measure of disagree- ment as to the actual cause of the refractory nature of gold tellurides. In order to investigate the leaching chemistry of gold telluride, a study was undertaken on the thermodynamics and kinetics of its dissolution using electrochemical methods. From this study suitable direct leaching condi- tions are suggested. In acidic chloride solutions gold telluride can be leached using iron(III) as the oxidant. At lower acid concentrations the reaction is controlled by the formation of a passivating film of H2TeO3. In alkaline solutions containing cyanide the rate of gold telluride dissolution increases with increasing pH and strength of oxidant. However, this can result in excessive consumption of cyanide in the presence of stronger oxidants.

Jan 1, 1991

Microorganisms have been used throughout history for the benefit of mankind with some of the earliest applications being in the production of fermented foods and beverages. Reactions involving inorganic substances such as metals, sulphur and sulphide compounds were hardly ever used until the early 1900s when it was discovered that Thiobacillus ferrooxidans was capable of promoting the oxidation of metal sulphides in ores. Today, it is widely recognised that biotechnology has great potential in the minerals industries, particularly in such areas as the leaching of low-grade ores of metals such as copper, gold and uranium; removal of heavy metal ions from process streams and from liquid process wastes; and in the desulphurisation of high-sulphur fuels. Although widespread commercialisation of mineral biotechnology has not yet been achieved, it is estimated that by the Year 2000 mineral biotechnology will have a significant impact on the minerals and metal industry and will probably account for tens of billions of dollars worth of annual production world-wide. This paper examines the actual and possible future applications of biotechnology in the minerals, metal refining and fossil-fuel processing industries. Firstly, the important chemical (and biochemical) principles involved in bio-dissolution of metals from ores and solid wastes, and in bio-accumulation of metals from process and waste solutions are reviewed, including leaching practices - past, present and future. Microbial desulphurisation of coal, and microbially-enhanced oil recovery are also discussed. The practical aspects of genetics and genetic engineering, the types and characteristics of those microorganisms important to mineral biotechnology, and the current research trends are reviewed as well. KEYWORDS: Biotechnology, microbial leaching, bioaccumulation, biosorption, bioleaching, biohydrometallurgy.

Jan 1, 1993

By V Lickfold, A J. Crawford, A J. Wilson

A total of 12 alkalic porphyry centres of Late Ordovician age have been discovered in two mineral districts within NSW. The alkalic porphyry deposits are hosted within shoshonitic volcanic centres that formed where a major cross-arc basement structure (the Lachlan Transverse Zone) cut the Ordovician Macquarie Arc. The Macquarie Arc was amalgamated with the back-arc basin sediments of the Wagga Marginal Basin and with the proto-Pacific margin of Gondwanaland to form the Lachlan Fold Belt after the cessation of arc magmatism in the Late Ordovician. Within the Ordovician volcanic belts of NSW, the alkalic porphyries are the most significant mineral deposits economically. The combined resource of the Cadia district porphyries (729 Mt @ 0.79 g/t Au and 0.30 per cent Cu) makes Cadia the worldÆs seventh largest known porphyry gold system and the largest known gold resource in an alkalic porphyry system (574 t Au). Hydrothermal alteration in the alkalic porphyries comprises a complex sequence of potassic, calc-potassic, sodic, calc-sodic, propylitic and late-stage, typically fault-controlled phyllic assemblages. Hematite dusting is common in many of these alteration assemblages, giving the intrusions and the altered volcano-sedimentary host sequences a distinctive pink-orange colouration. Intensely developed hydrothermal alteration assemblages are localised to within a few hundred metres of the composite intrusions that sourced the mineralising fluids, making these deposits difficult exploration targets. Sulfides are typically zoned, with several of the deposits having bornite-rich cores, chalcopyrite-dominant annuluses and pyritic outer haloes. Gold is well correlated with bornite in most of the deposits, and with chalcopyrite at Cadia Hill.

Jan 1, 2002

By Daxian Z, Youdi Z, Shucai Y

The Huolinhe surface mine is a very large-scale surface coal mine, located in the north-east area of China with a production rate of 1.2 Mtpa. Its coal haulage system has a large capacity, long transportation line and high degree of automation. The system comprises two crushing plants, many belt conveyors, four surge bunkers, one loading bunker, one coal storage and five end product bunkers. In this paper, the network patterns of semi-continuous systems in surface mines are studied, the reliability calculation methods and a simulation model are established and the software for computing the reliability of a semi-continuous system is worked out. By considering the whole coal haulage system in the Huolinhe surface mine, the authors developed the system reliability-numerical simulation method. The steps and the computing results of this method are described in detail. Through the calculation, we can master the present situation of system reliability and the main factors impairing system reliability, expose the system weak links, put forward the ways of enhancing system reliability and furnish the basis for improving the system and equipment. The way of increasing capacity and methods of enhancing system reliability are further raised.

Jan 1, 1995

By D Jordan, R Wilmoth, J LeFever

Exposed, open pit mine highwalls contribute significantly to the production of acid mine drainage (AMD) thus causing environmental concerns upon closure of an operating mine. Four innovative technologies were evaluated under the US Environmental Protection AgencyÆs, Mine Waste Technology (MWTP), Prevention of AMD Generation from Open Pit Highwalls Demonstration Project. This project is funded by the US Environmental Protection Agency (EPA) and jointly administered by the EPA and the US Department of Energy (DOE) through an Interagency Agreement. The objective of the field demonstration was to evaluate the technologies ability to decrease or eliminate acid generation from treated areas of the highwall in comparison an untreated highwall area. The field demonstration was performed at the Golden Sunlight Mine (GSM) a subsidiary of Placer Dome US, located near Whitehall, Montana. GSM is an operating gold mine. The highwall at GSM was divided into five plots measuring 15.24 by 15.24 m and each technology was assigned a plot with the fifth plot being used for comparison as a background control plot. The four technologies were spray-applied. The applied in situ treatment technologies included: a modified, furfuryl alcohol resin sealant, utilising wood and agricultural by-products to coat the rock surface; EcobondTM ARD, a phosphate based process that forms a stable, insoluble compound, coating the rock surface; and two AMD passivation technologies, magnesium oxide and the Dupont technology developed to create an inert layer/coating on the pyrite, preventing oxidation. The technologies were evaluated using the following methods: ASTM D 5744-96, accelerated weathering of solid materials using a modified humidity cells; mine wall residual wall rinse sampling; microscopy; and core drilling followed by water injection testing. The following paper describes the technologies used for passivation of acid generation from the highwall, the field demonstration, and the testing results from the background and each technology plot.

Jan 1, 2003

By N Slade, D Kennedy

Optimal throughput is a key objective for the successful operation of any manufacturing or processing business. In the mining industry in particular, in order to obtain maximum return on the significant capital investment in the physical assets involved, ensuring continuous throughput is a major focus for the complete leadership team. Failure to meet throughput objectives becomes a major source of distraction for management which too often attempts to solve shortfalls by æforcingÆ and æenforcingÆ production targets. This focus on the short term and the lack of attention to addressing root causes frustrates operators who attempt to satisfy unattainable short-term targets and lose sight of the long-term goal. Mount Isa Copper Operations, Xstrata Copper, North Queensland, had ambitious plans to increase throughput volume by ten per cent within a four month period to allow budgets to be achieved within the financial year. To achieve this, the process had to overcome the interplay of a multi-node (æsiloedÆ) production system and its associated inventories, each node, of which on the face of it at least, had more than enough ænameplateÆ capacity to meet these plans. However, the actual volume achieved at the end point of the system was well below this nameplate capacity and the system operated with an extraordinarily high standard deviation. The key issues in terms of making or exceeding plan were: how to substantially reduce the variability of the system and managing the inventories between nodes, how to synchronise throughput between the production nodes, and how to improve the decision-making ability within the process. In this paper, the implementation team has shown that a revolutionary process view of the operation together with appropriate planning decisions allowed increased throughput to be achieved without further capital investment.

Jan 1, 2006

By Muchi I

To clarify the characteristics of non-uniform flow of gas through layered burdens of ore and coke, flow equations of continuity and motion were solved numerically taking account of the geometrical effect of a bed. The results are compared with that predicted by an approximate solution. A mathematical-kinetic model for determining the distributions of process variables over the region from the top to the melting level of blast furnace with the horizontal or the inclined layers of burdens is developed in consideration of the uneven gas flow. By the use of this model, the longitudinal and the radial distributions of process variables in an existing blast furnace in Japan were calculated with the aid of a digital computer. Computed results show that the process variables and the melting level change remarkably over the cross-section of the furnace and that the patterns of these distributions are similar to the results observed by other investigators.

Jan 1, 1975

The common notion that the Box Hole Bore mineralization represents a variant of the Mississippi Valley type, has not been supported by the present investigation. Most of the common characteristics of Mississippi Valley type, namely 1. saturated saline brines; 2. low temperature domain; 3. mineralization in intraformational breccias and collapsed anticlinal structures; 4. hydrocarbon-rich or foetid carbonates, and 5. very simple geochemistry seem not to be applicable. The Box Hole Bore mineralization features: 1. moderately saline brines (18% NaCl); 2. very high temperature (495¦-520¦C) and low density brines, of critical vapour:water ratio and low pressure of formation (700 bars); 3. mineralization confined by faults and ex- tending laterally into carbonates of high pore volume such as stromatolitic dolo- mites; 4. ore environment characterized by oxidizing conditions, such as ferruginous carbonate and silica sands of high energy environment (heavy mineral banding) and 5. more complex geochemistry than is normal for Mississippi Valley type deposits (copper sulphosalts, and appreciable silver content). Additionally, the ore has un- usual textures formed by the replacement of carbonates and barite by large idiomorphic galena crystals. The pressure and temperature domains of the Box Hole Bore mineralization indicate post- Devonian emplacement and local high heat flow, possibly related to the Alice Springs Orogeny. Because of high lead and zinc trace element con- tents (100-600 ppm) of the carbonates, even out- side mineralized areas, it is possible that the mineralizing fluids have derived their metal con- tent from a local source, possibly aided by evap- oritic concentrations of the tidal to supratidal sequences. The new model for Box Hole Bore gives treater importance to the confining faults.

Jan 1, 1984

By Mercer J. K

Shaft guide reaetions in No. 2 Ore Shaft, Mount Isa Mines Limited, have been reduced to a low order of magnitude by the use of self aligning slippers and spring loaded guide rollers. Before commissioning the new 4500 h.p. winder at No. 2 Ore Shaft in January, 1958, it was realised that the increase in hoisting speed from 1800 to 2650 ft/min. would cause high shaft guide reactions which would be increased further when the elongated 12 ton skips replaced the existing 7 ton skips.The Cambridge accelerometer was used to measure the shaft guide reactions and to confirm that progressive skip modifications were resulting in reduced reactions. The skip modifications were completed in 4 weeks.INTRODUCTIONIn 1956 an expansion programme was initiated to increase production from 5000 to 11000 tons per mining day. At No. 2 Ore Shaft the 1500 h.p. winder operated at 1800 ft./min. and hoisted a 7 ton capacity skip. The new 4500 h.p. winder operates at 2650 ft./min. and hoists a 12 ton capacity skip. No. 2 Ore shaft is equipped with two 5 ft. by 5 ft. hoisting compartments (Fig. 1) and any increase in skip capacity meant an abnormally long skip. This feature, coupled with the increase in hoisting speed, could have caused very high shaft guide reactions. Before commissioning the 4500 h.p. winder, a Cambridge accelerometer was borrowed from the University of Adelaide and was used to measure the shaft guide reactions. Rather than carry out an extensive shaft timbering programme, the skips were modified by fitting cushioned guide shoes and spring loaded guide rollers (Fig. 1). Testing commenced in February, 1958, and was completed in March, 1958.

Jan 1, 1961

By Y Peng, D J. Bradshaw, G Long

As copper reserves deepen, they are becoming increasingly less pure. Penalty elements such as arsenic, antimony and bismuth are occurring in increasing concentrations. Environmental issues with smelters are leading to increases in penalty rates for these elements such that some concentrates may become difficult to sell. Arsenic concentration in Rosebery copper concentrates during 2010 varied between 3000 and 8500 ppm. Arsenic levels above 5000 ppm can preclude import to some customer countries. Optical mineralogy has indicated that arsenic in Rosebery copper concentrates is most commonly associated with tennantite (Cu12As4S13) in solid solution with tetrahedrite (Cu12As4S13). In addition, silver reports to concentrate as freibergite (Cu12As4S13 ). A reduction in arsenic content will therefore lead to a concurrent reduction in the silver content.A literature review by Long, Peng and Bradshaw (2012) indicated that arsenic bearing minerals could be removed using potential controlled flotation. Some success in reducing the amount of arsenic had also been reported using either roasting or chemical depression using reagents such as lime and cyanide. Due to the location of the Rosebery concentrator adjacent to world heritage listed temperate rainforest, the use of a roaster or cyanide was immediately discounted. Lime is already used as a reagent in the Rosebery flow sheet, so could be considered as a potential arsenic depressant.In order to determine a method of separating out the high arsenic bearing minerals, two series of flotation tests were conducted. The first series investigated the effect of pulp potential on the recovery of copper, silver and arsenic minerals; a second series investigated the effect of pH, froth depth and regrind.CITATION:Long, G, Peng, Y and Bradshaw, D J, 2012. Production of a low arsenic copper concentrate at MMG Rosebery, in Proceedings 11th AusIMM Mill OperatorsÆ Conference, pp 347-354 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Oct 29, 2012

By Peng S. S

Measurement of pressure distribution under the base plate of the Hemscheidt 2-leg 800-ton shield was performed underground for two shifts using 12 pressure cells installed in two rows, one each under each split base plate. Pressure distribution was uneven depending on contact condition. In general the highest base pressure occurred at 1/3 of the base plate contact length from the rear edge. The resultant was near vertical, and the pressure under the tail side split base plate was always higher than the head side. For comparison, pressure distribution on the canopy and under the base plate was also measured under controlled conditions at the USBM's Mine Simulator Testing Facilities in Pittsburgh, PA. A total of 45 tests were conducted under several simulated contact conditions. The results showed that the actual contact area is highly dependent on the compliance of the immediate roof and floor due to the sloped canopy tip and a base lifting device under the base tip. The loading distribution tends to change in magnitude and profile as the leg pressure increases. The base lifting device exaggerates the high toe loading on the base plate characteristics of this type of shield support.

Jan 1, 1992

Using a Jarrell-Ash "Compact" direct reading spectrometer, analyses were made of a wide variety of rocks ranging from acidic to ultrabasic types. Statistical evaluation suggests that accuracies comparable with the values obtained by the rapid chemical methods used by the United States Geological Survey will be obtained.The spectrographic method involves fusion of the rock with lithium tetraborate, strontium being incorporated in the flux as the internal standard. The resulting bead is ground with graphite and pressed into a pellet. The counts obtained on two standard pellets are used to correct for instrumental drift.Background readings are made at 4109 Å and corrections for background are made to all lines. Corrections for matrix interferences, both by chemical suppression and optical enhancement, are made. The three corrections of drift, background, and matrix are incorporated into one general expression which has been programmed on magnetic cards, each element requiring a separate card. The cards are placed individually into a desk top computer (an Olivetti Programma 101), which enables true counts to be calculated from the "readout" counts, and the element concentration is then obtained from computer-printed tables by referring to the appropriate value of true counts versus concentration. These tables were prepared by making a polynomial regression analysis of the spectrographic results from 50 standard and analysed samples, a curve of best fit being chosen for each element.The regression coefficients, usually of a fourth or fifth degree equation, together with other relevant data, were fed into a large computer, programmed to print out tables of concentration versus true counts. This technique enables the results obtained to be based on mathematically-derived values rather than on hand-drawn calibration graphs, giving greater inherent accuracy. The technique has been shown to give results of good accuracy and acceptable precision suitable for many geological and geochemical exploration programmes involving the analysis of many rock samples of different types.

Jan 1, 1969

By D Mather

The Resources Boom of the 1960s in Australia provided a strong foundation for AustraliaÆs economic prosperity for the remainder of the twentieth century. In combination with sound economic management, our world-class minerals and energy sector has played a key role in the nationÆs competitiveness, and it has been the major source of our exports since the early 1970s. Even with the emergence since 1990 of major mineral provinces in Latin America and elsewhere, Australia has increased its percentage of world production of several minerals. Between 1990 and 2002, our share of iron ore production rose from 13 per cent to 20 per cent, nickel mine production from eight per cent to 17 per cent, copper mine production from four per cent to seven per cent and zinc mine production from 13 per cent to 17 per cent. Whether this situation will continue during the coming decades motivates the discussion of this reflective paper. The approach is to consider fundamental supply-side and demand-side issues in arriving at a view about the state of the world, the world minerals sector and the Australian resources sector in the coming decades. The argument draws from the experiences of other mature mineral producing countries such as Canada and the United States. In such nations, efficient producers have often sustained competitiveness for extended periods, despite falling grades and other challenging operating conditions. Yet several forces are changing the economic geography of the minerals industry in significant ways. These include the re-emergence of major global mining corporations, the revision of mineral policy in over 100 nations since 1985, the mobility of technology and more efficient transport both for mineral productions and mining professionals. Given these trends the paper argues that the Australian mining industry must continue to change and evolve significantly in the foreseeable future. Its leaders will need to be even more internationally focused if it is to maintain its strong competitive position.

Jan 1, 2006

By R Skirrow, E Bastrakov, A Mauger, P

A new rapid spectroscopic core and chip logging and imaging system developed by CSIRO was evaluated by PIRSA and GA on a wide variety of mineralised properties in South Australia, including the Emmie Bluff, iron oxide copper-gold-style prospect on the Stuart Shelf. During a three-month campaign in South Australia in 2003 the system logged over 50 000 metres of core from the State Core Library, company prospects and mines. The new system uses automated core tray handling, continuous visible and infrared spectroscopy, and digital imaging, to characterise and identify dominant mineral species on core, chips and pulps, at spatial resolutions of ~1 cm (spectral data) and ~0.1 mm (image data). The system can log up to 1000 m of core per day and several thousand metres of drill chips. Mineralogy is pre-interpreted using specialised identification software trained on a selected suite of minerals showing characteristic absorption features within the measured spectral range. On core from the Emmie Bluff iron oxide copper-gold prospect, 470 km NW of Adelaide and 80 km SSE of the Olympic Dam Cu-U-Au deposit, the spectral mapping system was able to characterise a number of alteration zones enveloping the sub-economic Cu-Au mineralisation. These consist of distal white mica phases showing decreasing Al content towards mineralisation, a zone of increased iron oxide content in the hangingwall to mineralisation, and a zone proximal and coincident to Cu-Au mineralisation that is enriched in chlorite with variable Mg/Fe chemistry. Additionally the results show background spectral characteristics that mimic the magnetite distribution.

Jan 1, 2004

By Taylor L. M

Accurate computer prediction of the muck pile produced by a conventional blast requires modeling of the physics that occurs during the rock motion phase of a blast. The ability to predict the motion and the subsequent muck pile can have a significant economic impact on the mining industry. Some of the computer programs previously used to predict rock motion employed polygonal elements with comers and sides to model the rock (Cundall, 1971; Walton, 1982; Taylor, 1983). This approach allows the elements to bulk because of the corners and the aspect ratio. However, tracking the many interactions of the comers and sides of the elements is a computationally intensive process. Several other rock motion computer programs were developed that employed spherical elements, (Schamaun, 1984; Gorham-Bergeron et al, 1987) but they lacked correct bulking mechanisms or could not be vectorized for efficient execution on vector processing machines. A new spherical element code called DMC (Distinct Motion Code) (Taylor and Preece, 1989a, 1989b) has been developed this past year. DMC was designed and written to take advantage of super-computer vector processing capabilities. The efficiency has also been increased through the use of spherical elements, explicit time integration and a new contact search algorithm. When spherical elements were chosen for efficiency, some physically correct mechanisms for capturing the bulking behavior of the rock mass also had to he developed. This paper will treat several mechanisms that were developed to allow spherical elements to exhibit correct hulking behavior.

Jan 1, 1990

By C Menzies

Hydraulic Monitor mining at Solid EnergyÆs Spring Creek Mine has proven to be a safe and effective method of extracting steeply dipping thick seams at production capabilities approaching one million tonnes per annum.   This paper traces the development of Hydraulic Monitor mining from its initial introduction into old workings at Strongman Mine, its refinement at Strongman 2 in shallow, thick and steep workings, through to its introduction into the deep, very thick seams at Spring Creek Mine. This last step of the progression came with all the added complications that come with depth, particularly flammable gas.   This paper will describe the various changes in mining conditions that dictated the performance of the monitor, how the mine layouts were changed to adapt, and the various operational systems that were put in place, to allow a safe and productive operation.

Jan 1, 2007

By Rixon LK, Walton KP

Evidence will be presented to show that borescope inspection of mine roofs is one of the most useful techniques available for stability assessments. ACIRL currently has four borescopes, comprising two fibre-optic units of 2.7 m and 4.5 m length and two optical rod units (introscopes) 6 m and 8 m in length. (The 8 m unit is not yet fully operational because of modifications needed to meet safety regulations). Using the fibre-optic units, photographs of hole walls or ends can be taken to record features. Roof lithology and crack patterns can be observed in roof holes from 25 - 60 mm in diameter. Several crack parameters are useful: type, location in the roof with respect to height and lithological boundaries, presence of individuals or zones containing crack clusters and the estimated amount of dilation. These data are readily computer- plotted to form "borescope logs". Provided that the holes remain open, a hole or group of holes can be re-surveyed at intervals to monitor crack development across or along a roadway, for example with respect to an advancing blind-ended roadway face.

Jan 1, 1986

Naturally occurring rutile contains approximately 96 per cent titanium dioxide (Ti02), whereas the product from the ilmenite reduction - aeration upgrading process, "sub- rutile", contains only about 92 per cent Ti02. Therefore, there is an-incentive to lift the quality of sub-rutile by removing transition metal impurities, in order to supply a more valuable and marketable product, particularly for the chloride-route pigment producers. Previous studies have shown that the sub- rutile residual iron content (of approximately 4 per cent) and residual manganese content (of approximately 1.5 per cent) are chemically bound within the titanium-oxygen mineral lattice, and are most difficult to remove by simple leaching methods. In attempts to improve their eventual removal, a simple process modification has been successfully investigated in laboratory scale trials. Transition metal sulphides have been formed in the reduced ilmenite by using a "sulphur-doping" technique during reduction (metallization) conducted at controlled temperatures. These compounds give rise to greatly enhanced dissolution of the unwanted transition metal constituents during the sub- sequent aeration and acid washing steps. Thus, the manganese content of the final sulphur- doped, reduced-aerated, acid washed product can be reduced by about 90 per cent to 0.2 per cent, and the iron content by about 70 per cent to less than 1.5 per cent, in comparison with those levels remaining after unmodified pro- cessing. Additions of sulphur as low as 2 per cent of the weight of the ilmenite have proven satisfactory. Preliminary studies reveal that the modified process is economically feasible. These results suggest that the sulphur content of the Collie coal char used in an unmodified reduction step must exert some influence on the eventual removal of iron and manganese during aeration and acid washing.

Jan 1, 1973