Search Documents

Knowledge of the crystal chemistry of carbonate apatites is very important for the development of the concept of biomineralisation and the creation of a scientific base for new biocompatible materials for medical purposes. The present work is devoted to research results on various synthetic and biological carbonate apatites obtained at Saint Petersburg University over many years. All-important problems will be considered such as the location of CO32- ions in the apatite structure, the ionic substitutions in various biological apatites and the development of synthesis methods for their analogues. Syntheses of apatites were realised by precipitation methods and under hydrothermal conditions (from solid compounds and by calcite treatment). Biological samples used in this study were physiogenic and pathogenic apatites formed in the human organism, eg in teeth and in renal and salivary stones and dental calculi as well as apatites from fossils of different taxonomic families (brachiopods, conodonts, fishes, reptiles, mammal) with different geological ages and burial conditions. The study of the crystal chemistry of carbonate apatites has been carried out by powder (including high-temperature measurements) and single-crystal X-ray diffraction methods, IR-spectroscopy, thermal analysis and various chemical analytical techniques.

Jan 1, 2008

By A Singh, C Da Corte

Banded Iron Formations or more commonly known as BIF are sedimentary rock formations with alternating silica-rich layers and iron-rich layers that are typically composed of iron oxides (hematite and magnetite), iron-rich carbonates, and/or iron-rich silicates. Considering the depletion of highgrade iron ore reserves in South Africa, there is a need to maximise resource utilisation through the exploitation of sterile resources in particular BIF material which makes up greater than 65 per cent of current reserves. BIF, like most resources, is non-renewable and the deposition of BIF is thought to have come about by the reaction of Fe2+ ions in the ocean water with dissolved oxygen billions of years ago. BIF ore derived from the Limpopo region in South Africa (SA) located in the Penge Iron Formation is according to Cairncross, Kramers and Villa (2018) equivalent to that contained within the Asbestos Hills Iron Formation of the Transvaal Supergroup in the Northern Cape, SA. Considering that these are one of the largest iron ore deposits in SA, the case study involved mineralogical characterisation followed by beneficiation of BIF from these two regions to ascertain the comparability. Coarse and fines gravity separation comprising of Heavy Liquid Separation/Dense Media Separation on the +1.18 mm size fraction and shaking table on the -1.18 mm size fraction was conducted to ascertain whether an on-grade product or blended feedstock could be produced. The Limpopo mixed and Northern Cape conglomerate BIF samples achieved similar overall responses due to the conglomerate sample containing 15.3 per cent additional -1.18 mm fines. A combined coarse and fines product grading at 60 per cent Fe could be achieved at a yield and recovery of 43.9– 53.0 per cent and 62.2–63.8 per cent respectively. Increasing the grade to 63 per cent Fe reduces the overall recovery by 13.3 per cent. At the higher target grade, the contaminant specifications could be achieved for both samples. The Northern Cape laminated BIF sample is the most promising of the three samples studied. At the two iron target grades overall yields and iron recoveries of 72.9–65.4 per cent and 85.2– 80.2 per cent was achieved. In both instances the silica grade exceeded the product specification, however if the silica content is the driver the target iron grade should increase to 63.32 per cent. This will result in a minimal decrease in recovery of 0.4 per cent. The promising results of this study indicate that coarse beneficiation of BIF material is possible. These ore types could be blended with superior grade products to improve overall yields and recoveries or sold directly at a 63 per cent Fe product grade whilst sacrificing 0.4–13.3 per cent recovery.

Nov 8, 2021

By A M. Myrvang, B Nøst, N Q. Trinh, I S. B Bergh

The iron ore producer Rana Gruber AS is situated in northern Norway just below the Arctic Circle, and has been producing hematite and magnetite concentrates for more than 60 years. Since the late 1990s, the major part of the production comes from a sublevel stoping underground operation after many years of open pit mining. Despite moderate mining depth, the operations have experienced considerable rock stress problems due to quite high tectonic, horizontal stresses. To manage the problems, comprehensive rock mechanics investigations have been carried out by SINTEF, a major contract research organisation. This has included in situ rock stress measurements, laboratory investigations and numerical modelling. Based on this and practical experience, satisfactory design of vertical and horizontal pillars was achieved. At depth, the ore width increases from 30–40?m to more than 100?m. This, together with a major planned increase in production, called for a change of the mining method. After thorough investigations and considerations, it was decided to change to sublevel caving, very much like the LKAB system. The development of the new system has been closely followed up by rock mechanics measures, including 3D numerical modelling and comprehensive monitoring of deformations and rock stress changes. To achieve a satisfactory caving of the hanging wall, all vertical and horizontal pillars in the sublevel stoping area had to be removed. This has been a challenging task, and the development of the caving has been slow until recently. However, now the caving develops satisfactorily.CITATION:Bergh, I S B, Trinh, N Q, Myrvang, A M and Nøst, B, 2014. From sublevel stoping to sublevel caving at Rana Gruber, Norway – rock mechanical aspects, in Proceedings AusRock 2014: Third Australasian Ground Control in Mining Conference , pp 323–330 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Nov 5, 2014

By J Mauk, R Ruddock

The Taharoa ironsand deposit is 1600 hectares in size and is located 144 km south of Auckland on the west coast of the North Island, New Zealand. The sand predominantly contains heavy mineral-rich sediments, quartz, and feldspars. One hundred and two samples were collected from all areas of the mine, from the stockpile, black sand areas, tailings and Te Akeake Sands. Samples were collected with the primary focus on the next predicted mining path and secondly on gaining a good geographic spread throughout the mine. All samples were analysed for the dry, wet and particle density, the porosity and the magnetic, non-magnetic and clay minerals content.Stockpile samples had an average magnetic mineral content of 92 percent. Dry, wet and particle density values for stockpile material exceed all values from untreated ironsand samples with average values of 2.8g/cm3, 2.9g/cm3 and 4.4g/cm3 respectively. Black sand samples from the mine came from various areas such as the central area, the forest, the southern area and the benches. All black sand samples are remarkably consistent, with dry and wet density values between 2.3 û 2.5g/cm3 and particle densities that range from 3.7 to 3.9g/cm3. The porosity values are comparable to all other areas of the mine (average 37 per cent). The tailings and Te Akeake Sands have different physical properties due to their different mineral contents. The dry, wet and particle densities are all lower than that of all other areas with averages of 2.1g/cm3, 2.2g/cm3 and 3.3g/cm3 respectively. These lower values reflect the lower magnetic mineral content (average 18 per cent) and the associated increase in non-magnetic mineral content (average 78 per cent).

Jan 1, 2002

By Boyes B. A, Haydon D. J

The Paddington project is a gold mining and processing operation located 35 km north of Kalgoorlie in the Eastern Goldfields region of Western Australia (Figure 1). The project, which is operated by Pancontinental Goldmining Areas Pty Ltd, treats 1.25 Mtpa of oxidised gold ore, derived mainly from open cuts adjacent to the plant, with minor tonnages from other pits in the region treated on a custom-milling basis. Gold is recovered using a conventional CIL process. The process raw water supply of 3 500 kUd is derived from the local groundwater system. Average salinity of the raw water is around 90 000 - 100 000 mg/L total dissolved solids. A small portion of the raw water is passed through a vacuum distillation desalination plant to produce the project's 200 kUd requirement for high quality water for potable use and gold recovery.

Jan 1, 1988

By Armstrong G. J, Bowling K. McG, Gray C. H

An experimental induration facility has been developed and used for conducting iso- thermal tests and plant simulation studies on the thermal induration of iron ore pellets. The influence of time, temperature, lime content, ore type, shale content and particle-size dis- tribution (e.g. Blaine Index) is being studied in a continuing programme. The present paper deals with experiments relating to the first three of these parameters and illustrates the use of an experimental design suitable for statistical analysis. The relations between isothermal data and the results of complete simulation rests are discussed.

Jan 1, 1973

By Ricketts N. J, Smith G. D J

A hydrometallurgical process was developed for recovery of copper from copper-lead matte formed as an intermediate product at the BHAS lead smelter at Port Pirie. A plant utilizing this process was commissioned in 1984. The leach step involves oxidation of the copper-lead matte in an oxygenated acidic chloride-sulphate leachant. The results of laboratory scale testwork from which the chemistry of leaching was deduced, are presented in this paper. The rate of recovery of copper to solution from copper-lead matte depends largely_on the availability of oxygen to the sulphide surface. The optimum leach temperature is approximately 85¦C. Free acid must be supplied according to the stoichiometry of the leach reaction: Cu1.91 S + PbS + 2.91 H2SO4 + 1.455 02 = 1.91 Cu2+ + 1.91 SOj- + PbSO4(s) + 2.91 H2O + S¦ The grind size of the matte is of minor significance. Chloride concentration in leachant should be at or above a minimum of 10.0 g/l. Chloride acts to complex cuprous ions formed during leaching.

Jan 1, 1987

By S Meffre, S Golding, K Zaw, J McPhie, R James, S Suphananthi

The Chatree deposit is located in the central portion of Loei Fold Belt, which extends from Lao PDR in the north through centralThailand to Cambodia in the south. Epithermal gold-silver mineralisation at Chatree occurs as veins/veinlets and stockworks with minor breccia in volcaniclastic and volcano-sedimentary rocks of andesitic to basaltic andesite composition and Permo-Triassic age. This preliminary oxygen isotope study mainly focuses on vein quartz of Au-Ag bearing mineralization (Stage 2) and a few samples of late Stage 4 quartz ¦ calcite veins. Quartz from Au-Ag bearing mineralisation has oxygen isotope values ranging from 8.1 to 12.3ë with the lowest values generally occurring at the deeper level (140 m below the surface). The late barren quartz ¦ calcite veins (Stage 4) tend to have higher oxygen isotope values ranging from 11.9 to 12.0ë. The oxygen isotope composition of the Chatree ore fluids were calculated at 250¦C, 300¦C and 350¦C based on homogenisation temperatures of fluid inclusions. The calculated oxygen isotopic composition of the Chatree ore fluid ranges between -1 and 6ë that overlaps with both meteoric and magmatic compositions. This shift in oxygen isotope compositions may relate to processes such as water rock-interaction and/or mixing of meteoric and magmatic fluids. It is interesting to note that the lowest quartz oxygen isotope values are spatially more or less confined to the prominent structures in the centre of the major ore zones. These structures may have been acted as conduits for higher temperature magmatic fluids. The quartz oxygen isotope values increase outwards from the centre of the main ore zones to the margin of the pits possibly due to mixing of magmatic fluids with lower temperature meteoric water. Whole rock oxygen and hydrogen isotope studies are in progress to investigate isotopic trends in the wall rock away from the ore zone. The systematic variation of oxygen isotopes in quartz from the mineralised veins and adjacent wall rocks could potentially be used as a vector to ore. An EXTENDED ABSTRACT is available for download. A full-length paper was not prepared for this presentation.

Jan 1, 2008

Scheduling the transformation of in situ coal to run of mine coal and its subsequent beneficiation to a saleable product involves the use of many `parameters' or `adjustment factors'. These include in situ density, moisture adjustments, mining

Jan 1, 1998

By Milawarma E, Cusack C

The Sapan Dalam coal deposit of Indonesia consists of ten steeply dipping seams with thicknesses varying between 0.1 and 3.5 m and split by shale, clay, silt and sandstone. The area contains four faults dipping about 75¦ with four to 85 m vertical displacements. The complex nature of the deposit coupled with limited 27 exploration drillholes at an average spacing of 200 m made it difficult to create a realistic geological model. A three-dimensional gridded seam model gave a fair representation of the deposit. Attempt to create an optimum pit using the WHITTLE 3D pit optimiser proved unsuccessful since the pit optimiser is restricted to standard regular or fixed blocks. The gridded seam model was converted to a fixed block model prior to the optimum pit design. Smoothing and pit base controlling were also introduced to obtain a smoother pit. A mining scenario produced 3.6 million tonnes of coal, 28.4 million bcm of waste and US$18.1 million revenue.

Jan 1, 1995

The Que River Mine is a small under-ground base and precious metals mine located in north-western Tasmania. Current production rate is 220,000 tonne/yr from reserves grading 0.37% copper, 7.1% lead, 12.2% zinc, 195 g/tonne silver and 3.4 g/tonne gold.Ore is mined using longhole open stoping techniques. Cemented fill is placed in primary stopes to permit safe and efficient recovery of pillars. The operation is trackless with both shaft and decline access from surface.Ore produced is trucked 40 km by road to be sold to the Electrolytic Zinc Company of Australia Ltd., for treatment at their Rosebery concentrator.

Jan 1, 1985

By H Ishikawa, H Zhou, M Spagnolo, R Halomoan, S Palaniandy

The Karara Mine concentrator commenced operation in November 2012 to produce high-grade magnetite concentrate. The circuit consists of three stages of comminution plus magnetic separation at each stage for early gangue rejection. High-Pressure Grinding Rolls (HPGRs) are used for fine crushing, followed by ball mills and TowerMills in secondary and fine grinding duties respectively. In recent times, the plant throughput rate has had to exceed design by 18% to maintain the targeted concentrate productivity. Meanwhile, the ore hardness has increased by 25% to 15 kWh/t, leading to changes in the circuit operation and design. These changes led to coarser and harder feed to the TowerMill circuit that affects the subsequent separation processes. Nippon Eirich (the TowerMill manufacturer) has partnered with Karara Mining Limited (Karara) to optimise the TowerMill circuit operation to achieve a circuit product with an 80% passing (P80) of 35 µm. Several activities, including a TowerMill circuit audit and performance evaluation, multi-component analysis, modelling and simulation, were carried out as part of the optimisation exercise. The circuit audit results revealed that the TowerMill circuit feed size F80 had increased and ranged from 67–75 µm. The circuit product P80 ranged between 42–55 µm. The average circuit specific energy was 3.4 kWh/t. Moreover, the differences in density between magnetite and gangue minerals had affected the hydrocyclone performance where the efficiency curve exhibits coarse end deflection indicating that low-density coarse material was bypassing to the overflow stream. A multi-component analysis was carried out to further investigate the coarse end deflection of the hydrocyclone efficiency curve. The analysis showed that the magnetite and gangue minerals had different cut sizes of 32 and 64 µm respectively. Furthermore, 33% of the magnetite (below 35 µm) was being recycled back to the mill due to the density effect. Based on these findings, several recommendations for process improvement were identified and implemented including:• converting to a bottom feeding configuration to improve the grinding efficiency in the mill• additional grinding power in the fine grinding duty to compensate for the increase in plant throughput and ore hardness• improved classification efficiency by converting to smaller size hydrocyclones. Simulation studies carried out to evaluate the impact of these implementations had indicated that a P80 close to 35 µm with a steeper size distribution could be achieved by using the bottom feed configuration with additional grinding power. The TowerMill circuit was converted to bottom feed configuration and the performance was evaluated. The operating work index, size specific energy and circulating load were 30% lower than the top feed configuration, indicating that the bottom feed configuration had a higher grinding efficiency. This paper discusses the outcome of the circuit audit, multicomponent analysis, circuit simulations, improvement strategies and quantification of the improvements. CITATION:Palaniandy, S, Spagnolo, M, Halomoan, R, Zhou, H and Ishikawa, H, 2017. Fine grinding circuit process improvement at the Karara mine concentrator, in Proceedings MetPlant 2017, pp 101–120 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 11, 2017

By P Cleary, G K. Robinson, J Hilton, P Wilson

Concern has been expressed that Vezins can be biased when used for sampling fine materials if the cutter speed is as high as the 0.6 m/s which is often used. Using modelling techniques that consider airflow as well as particles, we have endeavoured to investigate sampling bias as a function of the size of fine particles, cutter aperture, cutter speed, shape of cutter blades, and the size of the gap between the inlet pipe and the cutter blades. Using crude estimates of the rate of airflow and a simple model of a single particle moving through air led to an approximate rule that bias will not occur due to mechanisms related to airflow for particles larger than about 200 microns; and bias will always occur for particles smaller than about 40 microns. For particles between these size limits, the extent of bias depends on particle density, cutter speed and cutter geometry as well as particle size.CITATION:Robinson, G K, Wilson, P, Hilton, J and Cleary, P, 2012. Avoiding bias when using vezins for sampling dusty materials, in Proceedings Sampling 2012 , pp 109-114 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Aug 21, 2012

Evidence for Middle-Late Triassic large-scale, silicic volcanism characteristic of extensional continental terranes is widespread in the Tasman Orogenic Zone in Eastern Australia. Detailed study of an area of south-east Queensland has defined the eroded remnants of a large collapse caldera and an associated area of rhyolite dome building and regional downsag which may be a second caldera (Stephens, 1986). The silicic volcanics and minor basaltic to andesitic products have been assigned to the Aranbanga Volcanic Group (Cranfield, in prep.) and the associated granite complex has been named the Mungore Complex (Ellis, 1968). Published isotopic age dates on the Mungore Complex are 212-219 m.y. (K-Ar Biotite; Ellis, 1968) and 226 m.y. (pooled Rb-Sr on several similar plutons in the region; Webb and McDougall, 1967). Basalts within the area have been dated at 219 m.y. (GSQ Annual Rpt., 1986). The collapse caldera (Mungore Caldera) lies on the intersection of 4 major structural features: 1) the Mt. Perry Lineament, a meridional fault zone extending for over 600 kilometres in a NNW-SSE direction and which has faulted the eastern part of the caldera 5 kilometres sinistrally. The lineament has been active at least from mid-Permian to Late Triassic and has localised activity into the Tertiary. Granite plutons of the same age as'the Mungore Complex- are topographically prominent and largely concentrated east of the Mt. Perry Lineament and its extensions (Nash, 1986); 2) the extension of the Darling River Lineament, a continental-scale, ENE trending feature which correlates with regional structural discontinuities but which lacks expression as faults within the Tasman Orogenic Zone;

Jan 1, 1987

Although the diesel engine is normally considered a tried and proven power source, many new and exciting diesel design concepts are being evaluated in engine development laboratories around the world. Three such programs currently underway at Cummins Engine Company are discussed. The turbocompound engine converts into power exhaust heat energy which is normally wasted. It does this by incorporating after the turbocharger a low pressure gas turbine which is gear driven to the crankshaft. The adiabatic engine has no conventional cooling system. It converts into useful power most of the heat energy normally rejected to the coolant. The series boost turbocharged engine uses seperate low and high pressure turbochargers to increase engine output and improve transient performance. These design concepts make possible increases in diesel engine thermal efficiency from the current 33 to 38 per cent range to the 50 percent range.

Jan 1, 1982

By Christie AB

Stream sediment geochemical analyses from 140 mining company exploration surveys of the Coromandel, Northwest Nelson, Marlborough and Westland regions of New Zealand have been compiled from open file reports and entered on a database called REGCHEM (for Regional Exploration GeoCHEMistry). REGCHEM is supported by GEOEXPL, a custom-written computer software application which operates under the Microsoft Access relational database mangement system on a PC-type computer. The data are plotted as maps using the Maplnfo geographic information system (GIS) package. Thesholds separating anomolous values from background values are determined using the PROBPLOT univariant statistical package, which enables the descrimination and modelling of several populations. Relevant REGCHEM data are being incorporated into a GIS for the Coromandel region as part of ongoing studies to sevelop mineral deposit and metallogenic models for the region, and to assess regional prospectivity.]

Jan 1, 1995

By P Jones

Fosterville Gold Mine is an underground operation located approximately 20 km northeast of Bendigo and 170 km north of Melbourne. Fosterville is the most productive gold mine in Victoria, producing over 100 000 ounces of gold annually. Access to the underground is via two portals cut into previously mined open pits. Fosterville is 100 per cent owned and operated by Northgate Minerals, a gold and copper producer with mining operations, development projects and exploration properties in Canada and Australia. (Northgate Minerals Corporation, 2010).Underground operations commenced in 2006 with ore production focused on three orebodies – Phoenix, Ellesmere and Falcon. Primary ventilation to the existing operation is provided by wall mounted exhaust fans located underground. This system provides approximately 220 m3/s of primary ventilation to the operations with intake achieved through the portals. Exploration drilling during recent times has uncovered another lode, the Harrier orebody, located 2.2 km to the south of the existing workings. Due to the location of the new Harrier orebody, vent analysis revealed the requirement for a separate source of primary ventilation to service Harrier. Figure 1 shows a current underground long section and the location of Harrier in relation to the existing workings. A number of options were considered, including a parallel decline from one of the existing two portals, however, a 4 m diameter, 350 m deep surface ventilation shaft was selected as the most suitable and feasible option for the mine.In December 2009, geotechnical drilling commenced to determine the most suitable location for the shaft. Site locations were constrained by a number of factors including interaction with current and planned surface infrastructure as well as proximity to nearby community members at the southern end of the mining lease. Results from the geotechnical drilling indicated that the majority of the proposed shaft would be suitable for a 4 m raise bored shaft. However, the top 40 m of the proposed shaft were deemed unsuitable for raiseboring due to zones of weakness that would likely influence the stability of a 4 m diameter shaft (Robinson, 2010). Following independent geotechnical advice and input from experienced raiseboring contractors, it was decided to construct a 40 m presink prior to raiseboring activities.A crane mounted auger was chosen as the preferred method to construct the presink. This method differs substantially from traditional blind sinking methods which are highly labour intensive. The auger, or drill, is designed and built to be attached to a crane base using engineered pivoting cradles. The auger is suspended from the crane jib by pennant ropes attached to hydraulic rams. The auger applies rotation and axial loads to the drilling tools via a telescopic steel bar (Kelly Bar). Drill tools are attached to the bottom of the Kelly Bar and comprise an assortment of drilling buckets and reaming barrels. Drilling buckets cut in situ material and collect the spoil for removal and are used for the excavation of the ‘pilot hole’. Reaming barrels are used to expand the diameter of the excavation by directing cuttings into the pilot hole which are then removed by the drilling bucket. In June 2010, the presink was constructed utilising a crane mounted auger with support to the shaft walls applied via a remote fibrecrete unit. The excavation method involved cutting a series of smaller diameter holes (2.4 m wide) before reaming out to the final diameter (4.7 m wide) in 4 - 5 m vertical increments. Fibrecreting activities were carried out on the backshifts, to ensure the stability of the excavation and minimise interaction between the two units. The presink was completed on time and offered many advantages compared to traditional blind sinking methods, including safety, costs and timing. This paper will examine the benefits of this method as well as lessons learnt and future applications in the mining industry.

Mar 21, 2011

Geomorphology is one of the most basic elements in geographic enviroments. Geomorphological mapping is significant in the spatial distribution of representing land form and understanding various features of geomorphology. Geomorphological mapping began very early in China, as the map, showing conekarst(karst pinnacles) of Juiyishan, Hunan province, it was drawn in 168B.C., which was excavated from Han tomb. Mawangdui, Changsha. Its original idea of map-making is rather similar to that of modern map and could undoutedly be the oldest map of tropical karst discovered in the world.Before liberation, landform mapping was done piecemeally and only a small number of maps were prepared by a few editor as insets.

Jan 1, 1990

By D A. Williams, J C. J Boshoff

The properties of tailings that emanate from laterite nickel processing plants are significantly different to most other mine tailings. They typically comprise a mixture of fine materials that are made up of altered clays, chemical residues and salts. Neutralisation processes that are undertaken prior to tailings discharge result in a high proportion of gypsum in the tailings. The salinity of the pore fluid is commonly higher than 140 g/L of salts, dominated by magnesium sulfate. Hydraulically-placed tailings that undergo solar evaporation typically reach dry densities of less than 1 t/m3 of insoluble solids. In situ water contents after many months of drying may be as high as 80 per cent, and peak undrained shear strengths are unlikely to be greater than 20 kPa.This type of material is not conducive to upstream raising, from both a practicality and overall slope stability perspective. Even at rates of rise of less than 1 m/a, it is difficult to effect upstream raising on unaltered laterite nickel tailings. However, the authors have successfully designed and implemented upstream-raised laterite nickel tailings storage facilities. The upstream raises have been formed from both imported natural materials and air-dried and compacted tailings borrowed from the tailings beaches.Drying and associated strength gain has been achieved through various means of mechanical intervention, repeatedly exposing a zone of tailings to the atmosphere for evaporation. Field trials have been undertaken to optimise the approach, and the findings from the trials have been implemented in practice to allow stable upstream raises to be constructed. The cost savings and reduced environmental impact compared to high-volume downstream raises have been significant.CITATION:Williams, D A and Boshoff, J C J, 2015. Upstream raising of laterite nickel tailings storage facilities, in Proceedings Tailings and Mine Waste Management for the 21st Century , pp 271–280 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Jul 27, 2015

By Y Li

An ilmenite ore from the Murray Basin was initially characterised using combined X-ray Fluorescence Spectroscopy (XRF), X-Ray Diffraction (XRD), Scanning Electron Microscope (SEM) and Electron Probe Microanalysis (EPMA) techniques. XRF analysis showed the ilmenite sample contained 58.7 wt per cent TiO2 and 26.2 wt per cent Fe2O3. Major impurities included: Cr2O3 (4.5 wt per cent), Al2O3 (2.7 wt per cent), SiO2 (1.9 wt per cent), Mn3O4 (1.5 wt per cent) and MgO (1.1 wt per cent). Manganese and magnesium were present as solid solution components within the ilmenite lattice while the elevated Al2O3 and SiO2 levels were due to separate aluminosilicate gangue minerals. Chromium was associated with discrete chrome spinel gangue phases of variable composition, (Fe,Mg)(Cr,Al)2O4. Quantitative XRD analysis of the sample identified rutile (12 per cent), ilmenite (11 per cent), and pseudorutile (67 per cent) as the main titanium-containing phases and chrome-rich spinels (seven per cent) as the major impurity phase. Minor quartz, aluminosilicates, zircon and monazite were also present. The ilmenite was mixed with synthetic graphite, pressed into pellets and reduced carbothermally in nitrogen at 1200 and 1400°C with C/O molar ratio of 1.2. Titanium oxide in the ilmenite was converted to titanium oxycarbonitride. Iron oxides were reduced to metallic iron. The Changes of impurities as a result of reduction was examined by SEM and EPMA analyses. Iron and chromium oxides within Fe-rich chromite were reduced to form a Fe-Cr-C alloy. Mg-rich chrome spinels were partly reduced at 1200°C. Zircon particles were not reduced at 1200°C, but were significantly reduced at 1400°C, with the particles becoming porous and the silicon being removed.

Oct 5, 2011