Search Documents

By J M. F Clout

The low-grade clay-rich -1.0 mm fraction of fine iron ores have often proved difficult to upgrade and capital intensive to process when using traditional wet beneficiation technology including single or two stage hydrocyclone desliming, wet high intensity magnetic separation (WHIMS), spirals and final dewatering using fine screens. Very fine (<0.010 mm) clay-rich slimes simply follow the water split to the hydrocyclone underflow where it often impedes the subsequent processing performance of any downstream spirals and can result in troublesome direct dewatering of hydrocyclone underflow. The -1.0 mm fraction in goethitic iron ores is typically considered as waste. A new two-stage wet gravity separation flow sheet, referred to as the ædesandÆ process, has been designed to overcome these issues for the upgrading of low-grade clay-rich goethitic or hematite-quartz fine iron ores. The desand process is designed to treat the -1.0 mm range from -8.0 mm natural fines, to recover a -1.0 +0.038 mm higher-grade fraction by rejecting to tailings any low-grade clay-rich ultrafines and low density sand-size fine chert/quartz gangue and shale.The desand process consists of wet screens with a nominal 0.7-1.0 mm cut point which direct the undersize to hydrocyclones. The hydrocyclone underflow is then fed to upcurrent classifiers where the classifier middlings is sent to spirals and classifier and spirals concentrates are then dewatered using vacuum belt filters or fine screens. The three product upcurrent classifier is employed to further deslime the hydrocyclone underflow, producing firstly a coarse final concentrate, and then a deslimed middlings for feeding to spirals as well as a 0.100 mm clay and fine quartz or banded iron formation (BIF) sand in final overflow for feeding to the residue thickener. The upcurrent classifier middlings provides the spirals with a significantly cleaner feed material than traditional methods, with the end result that the spirals do not silt up with slimes and therefore can continue to run at their optimum. The upcurrent classifier and spirals form a complimentary pair, with the upcurrent classifier better able to send coarser, acceptable grade particles to concentrate and fine slimes to tailings whilst allowing the spiral to better reject coarser, subgrade low specific gravity particles. Test results for the Mayoko-Moussondji iron project, Republic of Congo, using upcurrent classifier-Wilfley Table (simulating spirals) gave acceptable mass recovery from run-of-mine (ROM) and Fe recovery from hydrocyclone underflow for HEMCOL (6.2 per cent mass recovery from ROM/43.1 per cent Fe recovery) and HEMBIF2 (5.1 per cent mass recovery from ROM/22.3 per cent Fe recovery) mineralisation types. The desand process also enables improved dewatering of -1.0 +0.038 mm product from a clay-rich plant feed.A number of new wet plants in the Pilbara use the desand flow sheet including FMGÆs Cloudbreak and their two Christmas Creek plants. Many potential new applications are being explored, with new iron ore plants in India as well as those being proposed for West Africa. Operational experience has highlighted the importance of providing clean process water to the upcurrent classifiers as well as ongoing design improvements including increased wear life, reduced maintenance and enabling a more steady feed rate to spirals.CITATION:Clout, J M F, 2013. Beneficiation of fine iron ores using the desand process, in Proceedings Iron Ore 2013 , pp 337-348 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Aug 12, 2013

By Caivin JS, Freeman CJ

The Fairbanks Mining District in central Alaska was discovered in 1902 and since then approximately 7.5 million troy oz of placer gold, 250,000 troy oz of lode gold, and several thousand short tons of antimony and tungsten hap been produced from an area of about 1000 km . Historic lode production was from high grade gold-quartz veins that averaged 1.3 OPT free milling gold. Five main types of mineral deposits' are recognized in the district. I. Conformable, stratabound volcanogenic zinc, antimony, lead, and copper sulfides in places containing gold, silver and tungsten in bimodal metavoleanics (Cleary Sequence). II. Complex precious metal-bearing lead sulfosalt-quartz veins in Cretaceous granitic intrusives. III. Tungsten skarns adjacent to granitic intrusives. IV. Gold-bearing polymetallic sulfide-quartz veins crosscutting metavolcanics and metasedimentary units of the Cleary Sequence. V. Antimony-bearing gash veins along axial plane shears in various lithologies of the Cleary Sequence. Type I and Type IV mineral occurrences are currently being evaluated as low. grade bulk tonnage (target model 16-20 'million tons of 0.05 to 0.15 OPT)'and high grade moderate tonnage (target model 0.5 to 2.5 million tons of 0.5 OPT) gold deposits. The host rocks for Type I mineralization occur over a 30 km strike length and are at least 300 meters thick. Stratigraphic, structural, fluid inclusion, and isotopic data suggest that the Type IV mineralization is related to remobilization of Type I

Jan 1, 1987

By G W. Croft, L E. Krause

The current copper ore handling systems in use at Xstrata CopperÆs Mount Isa Copper Operations are up to 33 years old and have hoisted over 160 Mt at a rate of around 6 Mtpa. The mineÆs production profile is increasing in line with the optimum life of mine plan to over 6.6 Mtpa, which requires a system utilisation and reliability above that previously achieved. Copper ore is hoisted up the U62 shaft, which forms the central part of the U62 ore handling system. In theory this system can cope with the duty planned if the bottleneck is only the U62 winder. To achieve the planned production rate will require increased focus on reliability of each component in the ore handling system so that ore is available at the shaft for hoisting whenever the shaft is available, and the shaft is available for the maximum practical time. Increased availability will be achieved through the outcomes of the study of each component in the U62 system, and the internal ore handling systems that deliver ore from as far as 2 km laterally and 700 m vertically from the U62 shaft. The true capacity of each component is being determined and current performance is being measured with the aim of improving performance through root cause analysis and engineering out the causes of delays. The program began in March 2004 and this paper describes the progress achieved as of April 2005.

Jan 1, 2005

By Lacy JC

Human habitation of the Pacific Rim began when mankind began migrations out of its place of origin in the Rift Valley of Central Africa c. 4million years ago. This migration was erratic and largely controlled by advances and retreat of the various Ice Ages which, while limiting advances in areas of intense cold, at the same time aided migration thro-ugh the creation of land bridges as ocean levels declined during periods of glacial advance (Fig.l).Mankind began its trek to the Pacific from Africa across India then north into southern China and south into Indonesia (Fig.2). In part, these migrations stimulated early technological develop-ments to fulfill needs and adapt to environmental conditions; thus, these nomads developed stone and bone tools, animal skin clothing and the use of caves and other simple shelters. The earliest evid-ence of mankind's use of fire can be found at Zhou-koudaian near Beijing, c.400,000 BP.From China, the completion of the Pacific Rim's population can be attributed to the last Ice Age that dropped the ocean level by more than 300 feet and created land bridges about 40,000 BP allowing migration across the Bering Strait into Alaska and to Australia from Indonesia. The American migration reached Mexico about 30,000 BP, and by 6000 BP the occupancy of the Pacific Rim was complete when the first human settlers reached the southern tip of Chile.

Jan 1, 1990

By McAllister J

Oxygen was in restricted use at Port Kembla until tonnage oxygen became available for steelmaking in the open hearth furnaces. In adopting the new process various modifications to furnace design and operation were necessary and some of these, including increase in roof height and variation in firing and draft control, are discussed.The design of oxygen lances is also discussed and various steelmakingtechniques considered.Steel production increased with oxygen roof lancing-one- furnace producing 8,507 tons in one week-but new problems, such as rate of refractory wear, and disposal of fumes, were created.Design of future open hearth oxygen furnaces is predicted.INTRODUCTIONIn the post-war era, the availability of "tonnage" oxygen caused a major revolution in steelmaking techniques. Various pneumatic processes showed that steel can be madeat a very fast rate and that oxygen steelmaking can be applied to open hearth furnaces. Methods used are surface impingement with roof lances, bath injection with door lances, flame enrichment with fuel firing, mixed gas and oxygen firing, and enrichment of normal combustion air by oxygen injection into the flue systems.Owing to its previous restricted availability and high cost, oxygen was used in rather small quantities in the open hearth furnaces at Australian Iron and Steel Pty. Ltd.(A.I. & S.), Port Kembla, until about the middle of 1960.At that time tonnage oxygen became available and roof lancing commenced on No. 33 furnace at No. 2' open hearth shop.

Jan 1, 1962

By Whiten WJ

Although cone classifiers have received little attention in the modelling of mineral processing units they are one of the oldest non-mechanical classifiers. Their use has enjoyed a revival in recent years because of their low cost of installation and operation. In this paper the development of models for cone classifiers is reviewed and a new model is proposed. The experimental work, data collection and analysis are briefly described, and the new model is presented and discussed. The data used in the model development are made up from three distinct experiments using different units in each case. There are 21 data sets from a 3 m cone treating the tailings in a copper concentrator, 24 data sets from a 0.3 m pilot-scale cone, and 29 data sets from a laboratory glass funnel (0.(Y73 m diameter) used as a cone classifier, giving a total of 74 data sets. The model structure was developed from an analysis of particle settling in classifiers, using specialised model building software. The size of a particle settling in a separator and the forced flow through the apex of the cone were related via hydrodynamic dimensionless groups to the operating conditions of the unit. The model predicts the separation efficiency using the following four parameters: efficiency curve shape factor, cut-point, underflow density (per cent solids) or underflow pulp velocity. The model covers a range of feed flowrates from 1.2 to 5000 1/min.

Jan 1, 1993

By Lawson F

Arsenic-containing waste was dissolved into molten slags and the leaching behaviour of the quenched arsenic-doped slags (containing up to 2.3 wt per cent arsenic) was ascertained to evaluate the potential of this method as a means for arsenic disposal. The leaching behaviour was evaluated using acetic and sulphuric/nitric acid based solutions.From short-term (18 hour) leach tests, the amounts of arsenic in the leach solutions were found generally to decrease with an increase in the Fe(III)/Fe(total) ratio of the slag or an increase in pH of the leach solution. The amount of arsenic in solution increased with an increase in the slag basicity, the slag arsenic concentration or the surface area of the slag exposed to the leach solution. From longer term leach tests (up to 40 weeks) in the pH range 5 to 8, the arsenic remaining in the leach solution was believed to be controlled by sorption processes. The sorbent was an iron oxide-rich material which had the capacity to lower the arsenic concentration in the leach bottles to well below 1 ppm in some cases. At a leach solution pH over 9, the arsenic concentration in the solution increased with time, although the rate of solubilisation was much slower due to the high pH.Slags have the potential to stabilise arsenic and prevent it entering the environment provided attention is paid to the slag chemical composition and oxygen potential.

Jan 1, 1996

Single station group velocities of fundamental made Rayleigh waves over the period range 15-100 s are carguted for five provincial profiles of the West Pacific. Based on recordings of earthquakes around the West Pacific by the W~ (World Wide Standard Seismic Network) stations AFI (Afianalu), HNR (Honiara), WEL (Wellington), PAR (Raratonga) and RAB (Rabaul). The inversion of the derived dispersion curves for average shear velocity-depth structure of the crust, LID and LVL shows that (i) the over-all structure beneath the profile 'westernSamoa-Micronesia-Mariana' is in between that of regions 3 and 4 of Pacific Ocean (Yu and Mitchell, 1979), except that the velocity is slightly lower, (ii) the uppermost mantle beneath the 'Vitiaz Trench' profile between AFI and HNR is similar to that of region 2 of the Pacific - in other words, thin LID and thick LVL; (iii) the structure beneath the ' Ibnga-Keradec Ridge' profile has a thick crust with thick sediments, and LID-LVL velocity and thickness are similar to region 3 of the Pacific; (iv) beneath the profiles ' Tonga/ Kermadec Trench events-RAR station' the structure is intermediate between regions 3 and 4 of the Pacific Crean; (v) from decomposition of the 'Solomon-meriana' profile, the structure beneath the Cntong Java Plateau has a thick crust with thick sediments and the uppermost mantle is similar to that of region 3 of the Pacific clean.

Jan 1, 1990

By Woodcock JT, La Broody SR

As part of a project on the application of thiourea leaching to a variety of Australian ores, the systematic variation of pH, redox po- tential, and thiourea concentration were in- vestigated on an oxidized gold ore from the Nevoria mine, WA. Stirred open beaker tests were conducted with a pH-stat and Eh-stat system to automati- cally control pH and redox potential at selected values, while thiourea concentration was main- tained manually. Increasing the acidity of the leach liquor from pH 3.6 to pH 1, increased the gold extrac- tion from 37 per cent to over 90 per cent in S h, but there was only a limited improvement below pH 1.5. Silver extraction and thiourea consumption increased as the acidity was in- creased. Increasing the redox potential from 150 to 200 mV (i.e. potential of Pt electrode vs SCE) increased the initial rate of gold and silver extraction, but the effect on cumulative gold extraction by 5 h was slight. Thiourea consump- tion was greatly increased as the redox poten- tial was increased. An increase in thiourea concentration from 1.2 to 4.8 per cent increased the initial rates of gold and silver extraction, but, as for the change in Eh, the improvement in cumulative ex- traction by 5 h was marginal. Thiourea consump- tion increased with thiourea concentration.

Jan 1, 1984

By Milner SC

Four quartz latite rheoignimbrite units (units I to IV) mapped in the Goboboseb Mountains have been traced to their eruptive source, the Messum intrusive complex; this is a complex multistage ring structure (18 km diameter) with well defined gabbroic, monzonitic, and syenitic intrusive phases, and a central ring zone containing mixed rhyolitic to latitic agglomerates. Three of the four quartz latite units are correlated with the upper and lower (2 units) Springbok quartz latites outcropping some 90-130 km north of McIsum. This implies minimum volumes of 300-500 km for wits I and III (Lower Springbok) and 2200-2600 km for unit IV (Upper Springbok). This latter unit is still ti 300 m thick at its outcrop limit, some 130 km from Messum. Each of the four units represents single cooling units. Lithologically, each can be divided into three zones: A basal, dark grey to black flinty or glassy zone, often with highly attenuated fiamme (ressembling flow banding); a central massive and strongly recrystallised zone; and an upper variable zone which exhibit breccias, contorted flow banding, and amygdaloidal to pumiceous clasts. Their overall characteristics tend to be intermediate between ash flows and lava flows. The rheoignimbrites are interpreted as the oproducts of dense, high temperature (950¦-1100 C) ash flows which underwent rheomorphism and lava-like flow prior to their final cooling.These large scale eruptions are related to the initial rifting of Africa and S. America, possiby in response to a major hot spot.

Jan 1, 1990

By O Ostrovski

Iron ore reduction and iron cementation by CH4-H2-Ar gas was investigated in a laboratory isothermal fixed bed reactor. Iron ore was first reduced to metallic iron by hydrogen, then carburised to cementite by methane. Increasing temperature and hydrogen content accelerated the reduction process. Methane had almost no effect on the reduction process. Meanwhile, higher temperatures increased the rate of cementation and also the rate of free carbon deposition. The optimal temperature for cementite formation was found to be 750¦C. At this temperature, reduction of iron ore to cementite by gas containing 55 vol per cent hydrogen, 35 vol per cent methane and 10 vol per cent argon was completed in about 15 minutes. Addition of sulphur into the reducing/carburising gas strongly retarded the process of cementite formation and free carbon deposition. Cementite is an unstable phase which decomposes to iron and carbon. The rate of cementite decomposition increased with temperature up to 600¦C and above 750¦C . In the temperature range from 600¦C to 750¦C the rate of cementite decomposition decreased with temperature. Cementite was the most stable at 750¦C and least stable at 600¦C. The two-interface grain model was used to describe the process of iron ore reduction and the volume reaction model was employed to model the iron cementation process. The simulated results for both reduction and cementation are consistent with the experimental data.

Jan 1, 2000

By J Kleynhans, D Vermeulen, V Cogho

A number of rehabilitated open cast coalmines in SA are already decanting. The paper focuses on the situation at one of these mines where the following information exist: monthly decant rates and rainfall; sulfate concentrations of the decanting water; water level measurements inside the mine prior to decanting; ABA analyses; information about the rehabilitation (drainage slope, vegetation, soil used). A 2D-model was constructed for the mine and calibrated against six years of flow data. About 85 per cent of the decant outflow rate comes from direct recharge on top of the rehabilitated mine while the other 15 per cent is derived from lateral groundwater inflows. The estimated recharge value was 17 per cent of the rainfall. Measured sulfate values showed that the sulfate generation rate for the mine was in the order of 7 kg/ha/d. Conclusions from the study are:most of the water decanting from rehabilitated open cast coal mines is derived from recharge via rainfall; given a specific sulfate generation rate, then: the lower the recharge the less the decant rate but the higher the sulfate concentration of the decanting water and vice versa; future prediction of the decant rate must be stochastic according to expected percentile monthly rainfall values; and as a first assessment of the predicted sulfate concentration of decant water from rehabilitated open case mines, the 7 kg/ha/d sulfate generation rate can be used.

Jan 1, 2003

At moderate depths this question is of relatively small importance, but at depths over 2,000 feet it becomes a matter of some consequence. The rate of increase of temperature with the depth varies in different localities, but as a rough average we may take it to be about 1° Fahr. for every 50 feet of depth. The variation of temperature of the earth's surface from summer to winter penetrates to a comparatively shallow depth. At 50 feet the variation becomes practically insensible, and the temperature of the strata will be about the mean temperature at the given locality, from 60° to 75° F. in this country or the comfort of men working in the mines it is desirable to maintain a temperature of about 70° F. in the workings, or approximately the temperature of the strata near the surface.In order to obtain some idea of the magnitude of the undertaking involved in cooling a mine, I have examined the effect of placing in a long drive an apparatus by means of which we can abstract heat at a uniform rate from the drive, keeping, thetemperature nearly the same at all points of the drive at any given time.The lowering of temperature 70 F. at any distance feet from the centre of the drive (treated as a cylinder of the same sectional area as the actual drive), after an interval of time days from the commencement of the operation is given by the formula...

Jan 1, 1907

I offer some comments which might complement the valuable recent paper by De Angelis et al (1987), on the newly discovered nickel-copper deposit at Radio Hill.The authors have accepted subdivisions and correlations by Hickman (1977, 1983) although they are disputed (Fitton, Horwitz and Sylvester, 1975; Horwitz, 1979; Horwitz and Guj, 1986).Also, in my opinion, more plausible correlations of the Sholl Belt with Whim Creek Belt have been shown, but not discussed, in Hudson and Horwitz (1986, Figure 1). Our suggested correlations and subdivisions have appeared in publications and reports which are not readily available, but it is gratifying to note that many of the observations and conclusions of De Angelis et al are consistent with our conclusions of the Whim Creek Belt.The units of the Sholl Belt (of Hickman, 1983), between the Sholl Fault (Shear Zone) and the overlying Mt Bruce Supergroup to the south, equate well with units of the Whim Creek Belt. A basalt, characterised by an abundance of vesicles and similar in many ways to the Wararnbi Basalt of Fitton, Horwitz and Sylvester (I975), is well developed at the base of the succession, particularly in the southern part of the area (Figure 1). It is overlain by acid tuffs (locally metamorphosed, such as at Whundoo Mine) and associated acid and mafic lavas, similar to those of the Mons Cupri area. The mafics in the succession are marked by the absence of the numerous cherty sedimentary intercalations, such as the Marble Bar Chert, which occur in the Warrawoona Group; and they also lack the intense chertification that is prevalent in all parts of the proven areas of Warrawoona Group rocks.The Whim Creek Group is overlain, locally unconformably (two to three km east of Mount Negri), by the Negri Volcanics which include here the Louden Volcanics (unit Abu of Hickman, 1977), (Horwitz, 1979). The negri volcanics are an alternating sequence of variolitic and quench textured lavas not dissimilar to the highest units of the Sholl Belt (Figure 1). They range from high-Mg basalt to tholeitic andesites (Glikson et al, 1986, pp 79 and 84). The whole succession of the Sholl Belt is estimated to be of the order of four km in thickness, more consistent with the two km thick (Horwitz, 1979, Figure 1) Whim Creek area sequence than with the much thicker Warrawoona Group sequences. The Whim Creek Group (including the Negri Volcanics) is considered to be a widespread grouping (Fitton, Horwitz and Sylvester, 1976; Horwitz, 1979; Horwitz and Guj, 1986) which is prevalent in the northern part of the Pilbara Craton. It is considered by these authors to include the Mallina Formation turbidites of the trough area of Ryan (1965). One component, the Mons Cupri Volcanic, is about 2.95 Ga (Richards and Blockley, 1984). Between the Sholl Fault and Egina, the Whim Creek Group is intruded by large ultrarnafic to granophyric ills of the Millindinna Complex (Fitton, Horwitz and Sylvester, 1975), dated at 2.860 ± 002 Ga by Gulson and Korsh (1983) near the Sherlock River (south side of Caines Well Granite area). The Millindinna Complex has been equated to the same magmatic episode as the layered complexes of the Sholl Belt by Fitton, Horwitz and Sylvester (1975, p 20).

Jan 1, 1988

By Sinha DK

Coal Mining in India had been confined in the past primarily to two or three coalfields. Unscientific mining has led to huge degradation of land including massive subsidence of strata and disturbances of water tables to a wide extent. This has caused problems not only in flooding of mines from surface water but also in further planning of the surface drainage for current mining activities. Now, therefore, surface hydrology and land-use potentialities of the area have been endangered inflicting severe damage to the overall environment and ecology of the area. The paper Discusses certain investigative aspects of the problems highlighted as above in one of the coalfields, and encompasses mainly the following areas : 1) Quantitative estimation of the damage already done to the natural surface drainage system. For this purpose damage to topography and drainage for the last fifty years have been made out through the aerial photographs and topographic map sheets 2) Analysis and comparison of different states of such photographs are carried through to indicate the extent of damage to the hydrological conditions inducing undesirable changes in its land-use pattern. Based on the analysis, concrete suggestions are made to restore the water drainage system which will be most suitable for future mining in the area containing high grade coal. The design and planning of the drainage system is based on the total system approach including the cost benefit analysis as well as hydrological considerations.

Jan 1, 1988

A summary of the more important work by previous workers is followed by brief descriptions of the rock formations and regional structure. Type examples of the various forms of mineralization are described and the conclusion is drawn that there has been one main period of ore formation, namely, during the "Viti Revolution" in the mid or late Miocene.INTRODUCTIONThe Crown Colony of Fiji is comprised of a group of some 250 islands situated between latitude 160 sand 21 0 S and longitudes 1770 E and 1780 w: By far the greater, part of the total area, 7,083 sq. miles, is contained within the two main islands, Viti Levu 4,053 sq. miles, and Vanua Levu 2,128 sq. miles, and it is to these that most geologists have directed their studies. The most detailed work has been in the Tavua Basin, north coast of Viti Levu, centering about the mining town of Vatukoula and the geology of this area forms the subject of a separate paper by Blatchford (this volume p. 13).The present contribution summarizes geological knowledge of Fiji, at the time of writing (Feb. 1952), with particular reference to the known mineralIzation, but excluding the Tavua Basin. For general descriptions of Fiji's topography reference should be made to the recent works of Derrick and Dunkin.PREVIOUS WORKERSPeriod 1840-1930Prior to the gold rush of the early 1930s the visiting geologists concerned themselves in the main with the struc...

Jan 1, 1953

THE writer of this paper well realizes that the majority of engineering and metallurgieal works have from time to time evolved mechanical devices which are of great value as regards application in an operating plant.Many of these deviees would not be apparent to an individual making a brief inspection of works such as these, and it is for this reason that the writer proposes to outline a few of the devices which have been evolved by the engineering department of the Broken Hill Associated Smelters Pty. Ltd.NOVEL METHOD OF MACHINING A SPHERICAL JOINTFig. 1 represents one half of a casing for a ball-joint coupling. In the ordinary methods of boring an external spherical casing, that is with a radius bar attached to the tool rest of a lathe, there is always an error in the bore. It has two flat sides, and this error is inherent in the ordinary orthodox method and cannot be corrected. The engineers of the Broken Hill Associated Smelters Pty. Ltd. undertook a job that had to be extremely accurate, and the tool shown in Fig. 2 was evolved. The tool is not exactly fool proof, but in the hands of an average mechanic an absolutely perfect FIG. 1. sphere can be bored. The casing has to be held in the lathe in such a manner that one half can be removed without affecting the setting of the work. The business end of the tool is then placed inside the...

Jan 1, 1922

By F Jiao, J Wang

By the year 2000, China was the sixth largest gold producer in the world, with about one-quarter of total production being derived from refractory or semi-refractory deposits. Although bacterial oxidation plants had been previously tried in China using local technology, these plants had not been commercially successful. The Laizhou Gold Enterprise (Tarzan BioGold Co Ltd) located in Shandong Province, undertook to modernise their gold processing facility in 2000 with the introduction of Bacterial Oxidation (BACOX) Technology. In 1997 BacTech Australia and Mintek of South Africa had formed a technology partnership, combining forces in the provision of bioleach technology. Under a cooperation agreement, BacTech/Mintek were chosen as the preferred technology providers for this first application of commercially proven bacterial oxidation technology within China. The paper describes the successful operation of this plant over the last three years. The events of project implementation are described, from the metallurgical testing of concentrate; through to the commissioning at the design throughput of 100 tonnes per day resulting in 80 000 ounces (2.5 tonnes) per year of refined gold. Process aspects of the operation encountered over the last three years are discussed which illustrate the robustness of the operation. The success of the bioleach facility has led to an advancement of plans to double plant capacity and some of the innovations, which may be employed in this expansion phase, are discussed.

Jan 1, 2004

The Writer's knowledge of this region began in 1912, while he was with the Geological Survey of South Australia, and, investigated the principles governing the occurrence of fresh and saline ground waters in the area then being opened for closer settlement by the railways under construction from Port Lincoln.These field investigatons for water were extensive rather than intensive, but other examinations covered certain mineral aspects. The Writer also has' had the benefit in the preparation of this paper, of later work done by others, as well as of the monumental earlier work in South Australia by the late H. Y. L. Brown.AREAThe area dealt with, approximately 21,000 square miles, lies to the south of the latitude of Port Augusta.TOPOGRAPHYAs a whole the region is one of low relief, but the Olinthus Range, west of Cowell and north of Cleve, reaches an elevation of 2,000ft., while there is als0 a high area west of Port Lincoln and Tumby Bay. At the north end of the region the Gawler Ranges are formed by the dissection of a thick sheet of felspar porphyry extending over two degrees of longitude, by a width of 20 to 70 miles, with fairly uniform maximum elevations of the undissected portions of the order of 1,000 ft. above sea level. South of the Gawler Ranges are many isolated ridges up to 1,000ft. in height, generally sedimentary and trend...

Jan 1, 1951

By C Aldrich

In this paper, the use of topology preserving projections (Sammon maps) and neural networks are considered as a basis for the construction of multivariate statistical process control charts for a semi-autogenous grinding circuit on a gold plant. Comminution circuits are often operated in a quasi-steady state mode, where the mills tend to move rapidly from one state to another. This makes them difficult to monitor by conventional means designed for continuous steady state processes. As a first step, process data reflecting normal operating conditions are mapped with SammonÆs algorithm. Only one feature is extracted at a time and each mapping is captured by a neural network model to allow online application. Reverse neural network models are used to reconstruct the data prior to the extraction of subsequent features from the data residuals. The ensemble of neural network models can then be used for both fault detection and identification. The approach compared favourable with both linear and non-linear principal component models. It was able to detect two simulated fault conditions both with a reliability of 100 per cent, as opposed to 57.8 per cent and 95.2 per cent detected by an approach based on the use of linear principal component analysis and 84.1 per cent and 98.0 per cent detected by an approach based on non-linear principal components extracted with auto-associative neural networks.

Jan 1, 2007