Search Documents

A study was conducted to evaluate the effectiveness of the three types of sample grinders routinely used in the mining industry for the sample preparation of gold ores with particular emphasis placed on testing their capacity to produce a representative assay sample of coarse gold ores. The grinders used in this study are the Labtechnics LM5, the Mixermill 4010 ringmill and the Keegor vertical spindle pulveriser, commonly called a Keegormill. The study was based on the preparation of samples from two sources of known coarse gold ore, the Union Reefs Gold Mine and the Brocks Creek Gold Mine in the Pine Creek region of the Northern Territory of Australia, and a fine gold ore which was from the Macraes Gold Mine in New Zealand. The Union Reefs Mine contains visible coarse gold with 30 per cent to 40 per cent of the gold in the plus 75 μm fraction of milled ore. Discreet particles up to 5 mm have been recorded, Hellsten (1994). Union Reefs Gold Mine recovers 40 per cent of its bullion production through a gravity circuit, Riches (1997). Brocks Creek ore contains visible coarse gold with gravity bullion production fluctuating between 50 per cent and 65 per cent, depending on the mill feed. Average gravity bullion production from the Brocks Creek Mine is 55 per cent, Wilke (1997). The average gold grain size of Macraes Mine ore is 20 μm with 40 per cent of the gold in sub 10 μm particles. No gravity bullion is produced from the Macraes Mine, O'Keefe (1997). The testwork demonstrated conclusively that it is possible to prepare a representative sample for assay, of a coarse gold ore, when a Keegormill is used to pulverise the sample to a nominal 100 μm particle size (90 per cent passing a 100 μm screen by wet sieving). Both the Mixermill and the Labtechnics LM5 failed to prepare a representative sample, even when ground to a nominal 75 μm particle size.

Jan 1, 1998

By Dodd SF

Valuation of mining companies and their underlying assets prepared by brokers' analysts are, by necessity, approximations. The reliability of the valuations is entirely proportional to the quality of available data, the reliability of forecasts and the nature of underlying commodity price assumptions. The latter is perhaps the greatest single cause of gross variations in company and asset earnings estimates and- valuations. Assumptions are often necessary regarding commodity price achievement based on commodity hedge positions. Public disclosure from listed companies provides a sound base from which to project earnings, cashflows and balance sheets. The uncommon disclosure of projected production and cost schedules often reduces the reliability of future estimates. The application of reserve and resource estimates is contentious with practitioners applying differing weightings to importance of resources in estimating mine lives (and therefore cashflow streams). Valuation methods applied in equity markets usually focus on relative value and the identification of pricing anomalies. The commonest ""rules of thumb"" applied in the valuation of mining companies (and therefore their underlying assets) is with the use of year specific price/earnings (P/E) and price/cashflow (P/CF) ratios, with low multiples suggesting undervalued assets, high multiples overvalued situations (but these rules can often be clouded by market expectations of underlying commodity price movements). Another simply applied valuation tool is to examine the examine the value placed on reserves and resources by the market. Inexpensive gold in the ground (expressed as $ of market capitalisation per ounce of reserves or resources) might indicate a buying opportunity, but the market does discount the value of gold in the ground prior to mine development and where development risk is perceived.

Jan 1, 1994

This paper describes the technical development of the Newman beneficiation plant over its 23 years of operation and how this development relates to the mine ore reserves of high-grade and low-grade ore. It also discusses the importance of the beneficiation product grades in the overall blending of shippable products from a range of mines to optimise overall ore resource utilisation. Also discussed is the importance of the ore reserve reconciliations in resource utilisation. Data is provided on the capacity and quality improvement results of the plant upgrade, carried out in 1999. This shows how the original capacity upgrade targets have been exceeded and the additional quality benefits that have been obtained from this capacity.

Jan 1, 2002

By van der Walt TJ, van Deventer JSJ, Reuter MA

Owing to their complexity, many metallurgical processes are difficult to model fundamentally, and as a result more attention has been focused on the use of artificial intelligence techniques to simulate these processes. In this paper recent advances in the use of neural nets to model processes not amenable to conventional methods are discussed. These include the application of a new modelling methodology based on the use of neural nets for the dynamic simulation of metallurgical processes, the use of conventional techniques in conjunction with neural nets to simulate gravity separation circuits, as well as the detection of gross errors in variable measurements by means of a neural net representation of the relation between measurement and constraint residuals.

Jan 1, 1993

Longhole open stoping was introduced in the zinc-rich 'B' lode at New Broken Hill Consolidated Limited (NBHC) during 1968. Since then one longitud- inal and a series of transverse longhole open stopes have been developed. The longholes are drilled using Gardner Denver airtrac drilling jumbos and the broken ore is loaded from drawpoints and transported to chutes using Wagner ST5 scoop- trams. The holes were drilled from sill stopes 160 ft apart. The width of an emptied transverse longhole stope was successfully widened by blasting rings of blastholes drilled parallel to the stope wall. Proposed stope and pillar designs for a series of transverse longhole open stopes were assessed using a method of finite element analysis, applied to a longitudinal section through the area. Blastholes have been surveyed for dip and bearing using an Eastman borehole survey camera. The data produced has provided the basis of a study of blast- hole deviation. Deviation of boreholes is considered to be the major cause of the formation of benches in longhole stopes.

Jan 1, 1971

By A V. Ozsever

The selectivity in flotation columns involving the separation of particles of varying degrees of floatability is based on differential flotation rates in the collection zone, reflux action between the froth and collection zones, and differential detachment rates in the froth zone. Using well known theoretical models describing the separation process and experimental data, froth zone and overall flotation recovery values were quantified for particles in an anthracite coal that have a wide range of floatability potential. For highly floatable particles, froth recovery had a very minimal impact on overall recovery while the recovery of weakly floatable material was decreased substantially by reductions in froth recovery values. In addition, under carrying-capacity limiting conditions, selectivity was enhanced by the preferential detachment of the weakly floatable material. Based on this concept, highly floatable material was added directly into the froth zone when treating the anthracite coal. The enriched froth phase reduced the product ash content of the anthracite product by five absolute percentage points while maintaining a constant recovery value.

Jan 1, 2005

By J Mukhopadhyay, J Gutzmer

Most world-class high-grade (60 - 67 wt per cent Fe) hematite iron ore deposits are the product of enrichment of Precambrian iron-formations but processes responsible for enrichment are still unclear. Different models, ranging from syngenetic and diagenetic (King, 1989), deep-seated hydrothermal (Powell et al, 1999; Barley et al, 1999; Taylor et al, 2001; Gutzmer et al, 2002) and ancient supergene (Morris, 1980; 1985; Van Schalkwyk and Beukes, 1986) to modern supergene (MacLeod, 1966) have been suggested. The high-grade hematite deposits stand in contrast to martiteûgoethite ores which are generally accepted to be the result of supergene enrichment of iron-formation because they are found along Cretaceous to Tertiary lateritic weathering profiles (Morris, 1980; Harmsworth et al, 1990). Uncertainties about the origin of the high-grade hematite ores appear largely due to the monomineralic composition of the ores. They are almost exclusively composed of hematite (as microcrystalline hematite and martite), a mineral with wide stability field and very simple chemical composition that reveals little about its origin. Second, many of the deposits have been overprinted by later deformation and chemical weathering obscuring primary ore characteristics and zonation (Taylor et al, 2001). Third, surprisingly few detailed geological studies have been undertaken on the deposits with the aim of unraveling ore genesis. Most studies focus on the present-day structure of orebodies and their general physical-chemical composition for mining and beneficiation purposes. A more thorough understanding of the origin of high-grade hematite iron ore deposits may be gained by combining available mining-orientated data with systematic petrographic-geochemical analyses of different ore generations and hematite phases. This paper presents a first summary of results of an ongoing study, started some two years ago, of high-grade iron ore deposits in South Africa, India and Brazil, including a comparison with the rather well studied deposits of the Hamersley Province in Australia.

Jan 1, 2002

Chlorination of organic matter by direct injection of chlorine gas or hypochlorite has been investigated metallurgically worldwide with little understanding of the basic chemistry of the passivation mechanisms in- volved. In this paper, chlorination of carbon is attributed to oxidation-reduc- tions, resulting in surface halogenation of graphitic or amorphous carbon. It is proposed that chlorine and carbon interact to form thin coatings of chioro-carbon complexes that block the adsorption of gold cyanide com- plexes, thus passivating the gold adsorption sites of carbonaceous matter. Analytical, spectrographic, microscopic and electron-optical methods have been applied to demonstrate the proposed mechanism of chlorine-carbon bonding, as well as to evaluate the advantages and limitations of chlorination in the pretreatment of refractory carbonaceous gold ores.

Jan 1, 1991

By C Ausa, Nyunt Htay, A, L Deiparine

The main tectonic elements in Myanmar are a late Mesozoic and Cenozoic magmatic arc system in the west, repeated in northern Myanmar by more than 300 km of dextral movement on the Sagaing Fault, and the back-arc continental region affected by early and late Mesozoic orogenies, in the east. Historically most lode gold production has been from latest Cretaceous high-grade mesothermal veins and derived placers in the Wuntho and Tagaung-Myitkyina segments of the magmatic arc. Placer deposits in the Ayerwaddy River north of Myitkyina are currently worked by more than 100-bucket dredges. South of Wuntho, the magmatic arc and Pegu Yoma fold belt host epithermal gold with associated As, Sb and Zn in mid-Tertiary marine sediments. MyanmarÆs most productive lode gold deposit, the Kyaukpahto mine, lies within a north-trending district east of the magmatic arc and north of Mandalay. Here mineralisation and associated jasperoidal silica occur in Eocene sandstones and Permian and Triassic carbonates. Possibly this district continues southeastwards, across the Sagaing Fault, as sediment-hosted Sb Au and locally skarn Au mineralisation within a 1100 km long fold-thrust belt. Also present in the back-arc region are sandstone-hosted Au Cu in a Palaeozoic slate belt, and gold and stratabound gold-base metal mineralisation in the medial metamorphic belt. While untested targets remain in the magmatic arc, particularly for high sulphidation Au, the best potential is clearly within the back-arc slate and fold-thrust belts in the western part of the Shan Plateau and in northeastern-most Myanmar.

Jan 1, 1999

By H Zhang, M Karimi, M Hassan, S Saydam, B Li

With the prevalence of Mine Internet of Things (MIoT), different types of sensors have been widely deployed in mine sites for environmental monitoring and operational management. Currently, sensors in mine sites are mainly powered by cables or batteries. However, the deployment and maintenance of a power supply network can be difficult and expensive, and batteries have a limited lifespan. The vibration energy harvesting (VEH) technology can be used to provide power for lowpower sensors in mine sites. This presentation briefly introduces three widely used VEH methods (ie piezoelectric, electromagnetic, and electrostatic). The performance of a commercial electromagnetic energy harvester (ReVibe model D) is evaluated based on the vibration of a lab vibrator (3 g, 50 Hz) and an idling SUV automobile (0.3 g, 25 Hz). Experimental results indicate that a ReVibe energy harvester can generate adequate electricity to power a sensor, and its power output depends on the load resistance as well as the vibration frequency and amplitude of vibration sources. To apply VEH to mining applications, this study also analyses the characteristics of vibration sources at the mine site based on the vibration data collected from various operating machines in a coal handling and preparation plant (CHPP). The vibration analysis results indicate that all the measured vibration sources in the CHPP can generate stable and abundant vibrations, characterised by substantial amplitudes (up to 5.55 g) and low dominant frequencies (ranging from 14 Hz to 25 Hz), which are sufficient for VEH.

Sep 1, 2024

By K G. van den Boogaart, U Mueller

"An accurate prediction of benefit for some types of ore may require not just the ore grade, but a whole compositional characterisation of the estimates; that is, its full mineral composition including waste composition and nuisance elements. An example is Nickel laterite ores, where Ni/Co content must be complemented by estimates of many other elements, as these might have a notable impact on ore processing.Estimates are often obtained with (co-)kriging or co-simulation; however, geostatistics applied to compositional data may yield spurious and inconsistent results because of the constant sum constraint. The log-ratio approach avoids such problems. It proposes a three-step procedure: data are mapped to a set of log ratios of components; for example, the additive log-ratio transformation (alr) transformed scores are modelled with an appropriate fully multivariate geostatistical toolbox; for example, direct/cross-semivariograms and co-kriging or co-simulation results are back-transformed to the original units.Some practical aspects may make the application of this technique difficult: the assumption of Gaussianity and the high dimensionality of the compositions combined with the need for using multivariate methods. This contribution compares several ways of treating compositional data (combining normal score transformation, log-ratio transformations, and minimum/maximum autocorrelation factor (MAF) decomposition) with respect to their ability to generate sensible simulations. The various approaches considered are illustrated with a Nickel laterite data set for which ten variables are available. Both theoretical considerations and illustration results suggest that the best combination is (log)ratio-normal score transform-MAF, in this order. The factors so obtained are closer to Gaussian, approximately spatially decorrelated and can be simulated independently. Simulations can later be recombined to alr-simulations, which in turn may be converted back to point compositions. The resulting simulated compositions are compared with a full Gaussian co-simulation of the raw data. The log-ratio methods show reasonable to good reproduction of mean values, and distribution of the data set and by construction satisfy the total sum constraint, in contrast to the simulations based on raw data for which the total sums fluctuate strongly.CITATION:Mueller, U, Tolosana-Delgado, R and van den Boogaart, K G, 2014. Approaches to the simulation of compositional data – a nickel-laterite comparative case study, in Proceedings Orebody Modelling and Strategic Mine Planning Symposium 2014, pp 61–72 (The Australasian Institute of Mining and Metallurgy: Melbourne)."

Nov 24, 2014

By D Giddings, S A. Silvester, S Pickerin

Underground fires represent a constant threat to the safety of underground personnel. Miners in the immediate vicinity may face intense heat, blinding smoke, toxic fumes, fall of ground and other direct effects of a fire. However, the vast majority of victims never actually see the fire, and are overcome either by deadly fumes in the ventilating current or by asphyxiation. The mine ventilation system, which maintains a sustainable atmosphere to the working places, can transport products of combustion with equal efficiency. Miners who are remotely situated from a fire may be forced to evacuate considerable distances, through dense smoke and fumes, or even become cut-off from escape. Additionally, miners may become confused by unfamiliar ventilation characteristics caused by fires. Conveyor belts are widely used for transporting minerals in mines as well as in many other industrial situations. Because they contain large amounts of polymeric materials their use in certain environments, especially coal mines, but also in steel works, power stations and other enclosed areas must be controlled to minimise the fire risk. Over last nine years there have been 60 fires in underground mines in the UK alone and that of these, half have been associated with conveyors. This paper summarises the results obtained from a recent research project that has developed a preliminary computational thermal and fluid dynamic model to represent the spread of flame along a typical conveyor belt used in UK deep coal mines. The results of these models have been calibrated against a series of experimental tests performed within a surface fire test gallery. An improved knowledge of the burn characteristics and rate of release of the products of combustion from conveyor belt fires will assist in the development of better fire detection and fire fighting techniques.

Jan 1, 2005

By C A. Foss, D A. Clark, D Hillan

While magnetic surveys are the most common geophysical method used in mineral exploration, magnetics and geology remain to be fully integrated quantitatively in iron ore provinces. Although high-resolution aeromagnetic data has been collected over prospective areas of the Hamersley Province, much of the information in the data sets has hitherto not been fully exploited due to a lack of suitable interpretation methods. Recent advances in the processing and modelling of magnetics in highly magnetic terrains, constrained inversion and 3D visualisation and understanding the magnetic properties of banded iron formations and iron ores potentially create the capability to generate realistic 3D mapping of magnetite and hematite concentrations over large areas. Initial processing of conventional total magnetic intensity data produces true potential field maps that are unaffected by local perturbations of the regional geomagnetic field direction by intense local anomalies. The true potential field can be converted into vector and tensor components, each of which emphasises different structural features of the geology. Scalar quantities derived from these components simplify interpretation, using new methods, developed by CSIRO and elsewhere, that are less sensitive to magnetisation direction. Preliminary interpretations allow interpolation between mapped units and structures and evaluation of remanence and anisotropy effects. Accurate modelling requires rigourous treatment of self-demagnetisation and magnetic interaction effects. Semi-automated constrained inversion, combining magnetics with available geological information and other geophysical data, can be utilised to generate models of disposition of major magnetic units and the 3D magnetisation distribution within them. The final stage of interpretation is to exploit the understanding of magnetic properties (susceptibility, anisotropy and remanence) developed over many studies by CSIRO, the University of Western Australia and others to convert the interpreted magnetisation distributions into 3D maps of magnetite and hematite contents, which are the quantities of direct exploration interest.CITATION:Clark, D A, Foss, C A, Austin, J and Hillan, D, 2015. Three-dimensional mapping of magnetite and hematite concentrations from reprocessing of detailed aeromagnetic data, in Proceedings Iron Ore 2015, pp 229–240 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Jul 13, 2015

Research cruises dedicated to seafloor hydrothermal activity along the Kermadec arc first began with the Sonne-135 expedition in 1998. This was followed by the worldÆs first systematic survey for venting along an arc during the 1999 NZAPLUME I expedition. Since then, cruises dedicated to mapping plumes along the Kermadec arc include NZAPLUME II (2002), NZAPLUME III (2004) and ROVARK (2007). In addition, the manned submersible cruises SWEEP Vents (2004) and NZASRoF (2005) utilised the Shinaki 6500 and Pisces V, respectively, to sample in detail select hydrothermalsystems for mineralised samples, rocks and vent fluids. The ROVARK (2007) cruise was the first to deploy an autonomous underwater vehicle (ABE) along the arc which surveyed Brothers volcano at a far greater resolution than had been done previously by ships on the sea surface. While these expeditions have largely been led by GNS Science, together with its collaborators from National Oceanic and Atmospheric Administration, Pacific Marine Environmental Laboratory (NOAA/PMEL), significant contributions have been made by other groups from various nations, exemplifying the global interest in seafloor hydrothermal activity associated with intraoceanic arcs. This is an ASTRACT only. A full-length paper was not prepared for this presentation.

Jan 1, 2008

By J Pan, G Qiu

In this paper, an analysis is made of the present situation in the Chinese steel industry and its demand for iron ores. It is shown that the demand for iron ores has been growing substantially with the rapid development of the steel industry all over the world, especially in China. However, the iron ore market is currently in the situation where supply exceeds demand due to the economic crisis, with more investment in mines and less output of steel for several years ahead all over the world. Gigantic pressure is confronting the Chinese steel mills due to the too high and irrational price of iron ores. It is time to return the price of iron ores to a normal position much closer to their value. More attention has therefore been paid to developing low-grade and complex iron ore resources in China due to the irrational price of imported iron ores. More innovative mineral processing processes have been developed to tackle these complex resources. A systematic introduction to the development of beneficiation processes for low-grade iron ores is provided, including upgrading low-grade siderite, limonite, micron sized haematite and magnetite resources.

Jan 1, 2009

By A Madera, R Watt

The Tampakan deposit is a strongly telescoped high-sulphidation-epithermal/ porphyry deposit pair which is hosted by a sequence of probable Pliocene age subaerial andesite flows. These host units lie on the western flank of a deeply dissected andesitic stratovolcano within the northernmost portion of the Sangihe Arc in southern Mindanao. The deposit was discovered by WMC geologists in 1992 and is currently undergoing resource evaluation. Preliminary resource estimations indicate a total inferred metal content of 12 million tonnes of copper metal and 16 million ounces of gold. Using a 0.2 per cent Cu cut-off grade, the deposit has a current mineral resource of approximately 2.5 billion tonnes with an estimated grade of 0.48 per cent copper. At a higher cut-off grade of 0.5 per cent Cu, the mineral resource is approximately 900 million tonnes with an estimated grade of 0.75 per cent copper. Mineralisation is open both to the west and at depth. The Tampakan district is centrally located with respect to a 100 km wide zone of left-lateral strike-slip deformation represented by the trans-Mindanao Cotabato Fault Zone. The district is transected by several WNW-striking wrench faults which represent strands of the regional Cotabato Fault Zone. High-sulphidation epithermal mineralisation within the Tampakan deposit is strongly controlled by NNE-trending faults which lie along a dilational orientation within the Pliocene stress field associated with the Cotabato strike-slip fault zone. The mineralisation is broadly hosted by a gently-dipping, tabular zone of partial to massive silicification which displays multi-phase brecciation, acid-leaching, related vuggy porosity and which is developed within a district-scale advanced-argillic and argillic lithocap exceeding 90 km 2 in area. High-sulphidation epithermal mineralisation is associated with a silica-pyrophyllite-dickite-alunite-diaspore¦sericite alteration assemblage which is transitional downward to sericite-chlorite alteration and relict potassic biotite-chlorite-magnetite-anhydrite alteration associated with pervasively developed but weakly mineralised porphyry quartz stockwork veins. The high-sulphidation mineralisation is dominated by disseminated, vein and vug-filling enargite-bornite-digenite-chalcocite-covellite ¦ molybdenite which has overprinted an earlier phase of porphyry copper stockwork veins associated with high-level porphyritic hornblende diorite stocks. The distribution of high-sulphidation alteration and mineralisation reflects strong stratigraphic and structural controls. The high-sulphidation mineralisation covers a surface area of approximately 1.6 km by 2.0 km and forms a tabular, flat-lying to gently-dipping body between 200 and 500 metres thick. It is intersected at RL-1200 m ASL (at surface) in the northern portion of the deposit and extends down to RL-600 m ASL at the southern end. The deposit has been diamond-drill tested to a depth of 400 - 500 metres where low-grade chalcopyrite-bornite-pyrite mineralisation is hosted by pervasive quartz stockwork veins. This deep-level mineralisation represents the outer portion of a porphyry Cu system hosted by both andesite flows and the uppermost portions of high-level hornblende diorite stocks. Deep drilling to test for higher tenor porphyry Cu mineralisation is planned.

Jan 1, 1999

By Senanayake G

A sample of crude copper powder produced by the CLEAR Process was refined by dissolution and recovery in sulfate media using the Parker Process. The Parker Process offers an alternative method of copper recovery to electrorefining and allows a range of Eh and pH conditions to be used to provide further discrimination of impurities. High purity copper powder was recovered by the Parker Process even in the presence of added impurities when the Eh and pH was maintained near constant throughout the refining step.

Jan 1, 1984

By M Fishwick Tapia

This paper was prepared and translated from the final report of the investigation into the Codelco Chile Salvador DivisionÆs air blast which occurred in the Inca-West area. In this section of the block cave mine, a stable arch formed covering an area of approximately 15 000 m2 . This zone subsequently collapsed producing a massive air blast. The collapse followed a series of planned extraction activities and occurred on 5 December 1999. This paper discusses the cave back arching event and the measures that were taken in order to induce the controlled collapse. This was to minimise risk to personnel, equipment and installations. The results obtained to date are presented.

Jan 1, 2000

By E R. Ramanaidou, M A. Wells

Asian steel mills are demanding iron ore with specific chemical, mineralogical, metallurgical and environmental properties to warrant that the ore is of the greatest quality. As a result, a toolkit of analytical techniques becomes important. Hyperspectral analyses of iron ore cores and chips using a HyLogging™ system are now routinely undertaken by BHP Billiton Iron Ore in Western Australia for banded iron formation-hosted iron ores and channel iron deposits to better characterise their iron ores. Objective information can now be expanded by adding new sensors. The current HyLogging™ system provides point analysis reflectance spectra and a red–green–blue high-resolution image, while recent developments, such as hyperspectral imaging using the Corescan system, now provide full spectral image and surface topography. The addition of Raman spectroscopy provides the ability to detect and quantify quartz and magnetite (or maghemite) and cross validate hyperspectral information on hematite and goethite and aluminium substitution in the iron oxides. Micro-X-ray fluorescence mapping also adds chemical information appropriate for cross validation with the aforementioned methods. This paper demonstrates the value of combining these techniques for an optimum characterisation of iron ore.CITATION:Ramanaidou, E R and Wells, M A, 2015. A new toolkit for iron ore characterisation, in Proceedings Iron Ore 2015, pp 587–590 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Jul 13, 2015

By N Bezuidenhout

The performances of ten field-scale experimental final covers were evaluated over an eight-year period. The covers were evaluated as part of the rehabilitation of acid generating coarse coal reject dumps and opencast coalmines in the Northern KwaZulu Natal province in South Africa. The main purpose of the experimental covers was to restrict rainfall infiltration into acid generating coal discard. The mean average rainfall over the monitoring period was 839 mm with 84 per cent occurring between October and April. The average evaporation was 2100 mm. The site is described as a ædry siteÆ since the evaporation exceeds rainfall by a significant margin. Approximately 30 per cent of rainfall infiltrated the uncovered cells. Rainfall infiltration into the fine coal discard was reduced to 20 per cent and 15 per cent for a 300 mm and 500 mm cover respectively. Infiltration was reduced to ten and 12 per cent of rainfall for fine coal discard provided with a 1000 mm double soil cover and a compacted bottom layer. This relatively high infiltration was probably caused by moisture being entrapped in the bottom retentive layer, and not being released by evapotranspiration. The best performing cover comprised 1000 mm single soil with a compacted bottom layer and infiltration was reduced to four per cent of rainfall. Rainfall infiltration into sloped cells varied between seven per cent and 14 per cent of rainfall. The computer software, SoilCover, was used to predict infiltration and moisture content in the cover. Predictions of rainfall infiltration were accurate, but were less accurate for covers where the hydraulic properties of field compacted soils were not adequately characterised. Oxygen ingress was significantly reduced for cells provided with a 1000 mm cover. Oxygen concentrations were generally below one per cent, with the exception during exceptionally dry periods when the cover dried out, resulting in increased oxygen ingress. For other covers, and the uncovered cells, oxygen concentrations were close to atmospheric concentrations. During the eight-year monitoring period, acid breakthrough occurred in five of the cells, including three uncovered cells, one cell with a 300 mm cover and one sloped cell. In contrast, cells covered with 1000 mm, bottom compacted soils shows a marked improvement of leachate quality, indicating that sulfide oxidation processes have been slowed. There was a distinct relationship between oxygen ingress and leachate quality for all of the cells. No cover degradation was evident over the eight-year period, except for some erosion observed on a sloped cell. The results indicate that the onset of acid drainage can be delayed by perhaps hundreds of years and salinity loads can be reduced by more than 90 per cent, if coarse coal reject dumps in the KwaZulu-Natal provinces in South Africa are provided by an adequate cover system. These research outcomes can be extrapolated and applied to other dry site regions where similar materials are being considered.

Jan 1, 2003