Search Documents

By H Austrheim, A K. Engvik, A Putnis

How aqueous fluids move through rocks and minerals is a question of possible pathways and mechanisms. Existing pathways such as fractures and grain or mineral boundaries provide obvious channels for fluid movement. We show how minerals reactive in the presence of an interfacial fluid can be replaced by new product phases through a process of interface-coupled dissolution-reprecipitation. A resulting porosity in the new phase provides a new pathway for fluid movement and therefore mass transport through the mineral, which progressively reacts at a moving interface. Eventually fluids of new and changing composition can pass through originally solid minerals and move on further, taking remobilised elements with them, to be reprecipitated whenever the chemical and/or physical conditions allow. One such example is albitisation of rocks associated with mineralisation.

Jan 1, 2008

Stope filling practice on the Great Boulder Mine was described in -a paper published in Aus.I.M.M., Proceedings No. 137, in 1945. When the paper was presented the process had emerged from the experimental stage, most of the initial operating difficulties had been overcome, and it had been established that this method of filling was suitable for the working of ore bodies in the Kalgoorlie field: Since 1945, it has fulfilled its early promise, -and during 1947, the tonnage of fill placed was sufficient to cover current operating requirements as called for by the mine ore breaking programme. The object of this paper is to outline current operations. In so doing it will refer' to some ground covered by the previous paper, but will omit details of the initial experimental work. It will include, however, a description of further experimental work on the use of hydraulic filling for worked-out shrinkage stopes, arid will provide more accurate data on costs than were available previously.ESTIMATE OF TONNAGE OF FILL REQUIRED.Early in 1946, the mine ore reserves were analysed in order to estimate the relative tonnages to be mined per four-weekly period by shrinkage stope methods -and by cut and fill stope methods respectively. This analysis was extended beyond the blocks already prepared for stoping, and inclqded the majority of the major ore blocks in the mine ore reserves. It gave the following results:-For a mill tonnage of 28,000 tons per period, 6,933 tons were required from cut and fill stopes, and a possible future maximum programme of 35,000 tons per period would require...

Jan 1, 1948

By H R. Watling, W van Bronswijk

Of the various copper sulfides in the EarthÆs crust, chalcopyrite is by far the most common. This mineral is also the most difficult to leach, with hydrometallurgical extraction processes characterised by prohibitively slow kinetics. While a number of processes, including leaching at atmospheric pressure and temperature, pressure leaching at higher temperature, and bioleaching of concentrates in stirred tanks and whole ores in heaps have been demonstrated to some extent, there is currently no full-scale unit process for the hydrometallurgical extraction of copper from chalcopyrite ores. Titan Resources NL operated a test bacterial heap leach at Radio Hill, Western Australia in 2001, to extract nickel and copper from a mixed sulfide ore that contained significant amounts of chalcopyrite, pentlandite and pyrrhotite. A proprietary mixed bacterial culture was used to inoculate the heap, and the nickel and copper were precipitated from the resultant solution as mixed hydroxides. Although 65 per cent of the copper from the Radio Hill heap was eventually recovered, the rate of extraction followed an unexpected trend, as the initial rates of copper recovery were much slower than column tests had suggested. It was postulated that this was due to dissolved copper precipitating as covellite or chalcocite in the most poorly aerated zones within the heap, ie furthest from the aeration pipes at the base. It was further hypothesised that the lack of oxidants (oxygen, ferric iron) and the likely inactivity of bacteria in these areas limited the re-dissolution of such secondary copper precipitates. In this study the above hypothesis was tested in column experiments using the same mixed sulfide ore and a synthetic leach solution. One set of columns was aerated and another set was operated without aeration. Results showed that precipitation of copper occurred regardless of whether the column was aerated or not. The results suggested that pH of the leach solution played a significant part in the precipitation of the copper. Further column tests revealed this to be the case. These results are described here in detail.

Jan 1, 2004

By Kershaw M. G

At the Risdon plant of the Electrolytic Zinc Company of Australasia Limited, impurities toxic to the electrolytic production of zinc metal, notably arsenic and antimony, have traditionally been coprecipitated when precipitating iron as ferric hydroxide. With the development of a new lead/zinc mine at Elura, New South Wales, it was anticipated that the concentration of impurities, particularly arsenic, in the zinc sulphide concentrates treated at Risdon would increase. This suggested that an increase in the capacity of the iron purification step in the Risdon plant would be required to cope with the higher input of these impurities.Laboratory investigations showed that the additional capacity would be achieved, either by a large increase in the amount of iron used as the "collector" in the process, or by a two-stage process. The latter approach was preferred as a increase in the amount of iron precipitated would lead to much higher operating costs and overloading of the existing equipment.In the two stage procedure, acidic iron bearing solution is added to solution feeding the first stage and most of the arsenic removed from solution when iron is precipitated from solution by neutralisation with limestone. Mter separation of the precipitated solids, acidic iron bearing solution is added to the solution, and remaining arsenic removed almost completely following neutralisation by limestone.This paper describes the laboratory and pilot plant results and the plant modifications required. Abrief account is also given of the operator training and acceptance in the use of the digital control facilities installed in the new plant.The new plant is working well and good solution purity is being maintained despite a substantial increase in the concentration of impurities in the raw materials.

Jan 1, 1990

THE MORWELL BROWN COAL DEPOSITSTHIS deposits of brown coal in the neighbourhood of Morwell, are of tertiary age, and are represented on several horizons, but, owing to lack of evidence, the exact age cannot be fixed definitely. It is chiefly by reference to the basalt flows of the Older Volcanic series that the brown coal beds have been classified as of Pre-Volcanic and of Post-Volcanic age. The Pre-Volcanic brown coals are less widely distributed, and the seams generally are thin, though near Thorpdale a thickness of 40 tt. has been recorded, and from 10 ft. to 12 ft. in the vicinity of Narracan. Both of these localities lie to the south-west of Yallourn, and it has not been proven as yet whether these older coals and accompanying basaltic flows pass under the Morwell beds.The brown coal deposits have accumulated in an area where extensive faulting in early tertiary times has taken place with oscillations of elevation and depression of the land surface. In the vicinity of Yallourn, Paheozoic strata (probably Silurian) are exposed on either bank of the Latrobe River about a mile above the powerhouse weir, where the river flows through steep banks rising a couple of hundred feet on either side. The Silurian strata are highly inclined, and consist mostly of slates and Illudstones. Resting unconformably on the upturned edges of the Silurian is a conglomerate bed of Mesozoic age, above which occur beds clipping to the south-east of felspathic sandstones, with alternating mudstones at times carbonaceous, and containing thin seams of black coal.The brown coal seams and associated lignitic Clays, sands, gravels, etc, rest on the Mesozoic strata (as proved in the vicinity of the power-house and by boring north of the Latrobe River, near the old open cut), but do not appear on the surface of the Mesozoic, where these beds attain an altitude of more than 50 ft. above the present level of the Latrobe River.Half a mile upstream from the power-house a layer of basalt of the Older Volcanic series, almost resting upon the Silurian strata, is...

Jan 1, 1924

Agitation leaching data has generally not been analysed to extract the maximum amount of information that is available. Analysis of heap leaching systems has been developing over the last five years to include the underlying rate determining process. This has been found to be diffusion control for all copper leaching; irrespective of whether they are oxide or sulfide minerals. Extension of the diffusion model has enabled better scale-up protocols to be developed. Acid consumption in copper heap leaching operations has been found to be a first-order chemical controlled system. The analysis of the data is relatively straightforward. However the interpretation and scale-up is not as direct as for agitation leaching. The differences in the residence time distributions (RTDs) are required to properly account for the scale-up between test columns and field. The leaching of oxide copper in agitated systems has also been found to be diffusion controlled. However, there are significant differences in the control mechanisms. Initially the leaching control is the supply of acid to the mineral surface. This is limited by the diffusion through the surface boundary layer and is thus subject to the level of agitation. Higher agitation provides thinner boundary layers and faster diffusion rates. Similarly, higher grade material creates a thicker concentration boundary layer and relatively slower leaching. Acid consumption in agitated leach systems is also a first-order system. However, the interpretation of the RTD is easier from batch testing to continuous multiple CSTR series plant operations. Altering the method of acid delivery can significantly reduce acid consumption rates. Predilution of the acid minimises the local acid concentration near the addition point and can reduce the consumption by up to 50 per cent. Modelling of the combined leach rate(s) and acid consumption rates in agitated systems allows comparison of various leaching strategies and selection of the most appropriate. Similarly, the model allows the identification of economic limits to leach recovery; where the cost of acid consumption is greater than the value of the copper recovered. From this analysis selection of residence times for agitation leaching can be determined. The modelling tools are useful in identifying optimum conditions for the leaching process.

Jan 1, 2005

By J Rapkoch, M Cerna

As long as the tendency of Chilean copper ores continues to feature high levels of impurities, the smelters will tend to produce high-impurity mattes. The most notable impurities include As, Sb, Pb and Zn. Producing high-quality copper with such materials is a major challenge to metallurgical engineers and operators. More investigations into process optimisations and/or alternatives are needed, using both theoretical knowledge and operational skill. Converting of matte in Chile depends on the matte grade using established technologies. Mattes up to 66 per cent Cu, such as those from Outokumpu Flash furnaces, are converted in two stages, while typical Teniente Converter mattes, ranging from 70 - 74 per cent Cu are converted in one blow, directly to copper. In this paper thermodynamic data are used to evaluate possible variations in the use of traditional technologies applied to smelt copper concentrates with high impurity levels. The results of this analysis help to predict the impurity distributions between matte-slag-gas during the slag blow, and between copper-slag-gas during the copper blow. For converting high-impurity white metal, the effectiveness of volatilisation in the slag blow is analysed. In the copper blow, the type and the amount of flux (SiO2, CaO and others) are altered with the aim of optimising impurity distribution to the slag and gas.

Jan 1, 2005

Australian coal mines have spent considerable capital on dragline improvements over the last two years. This has included ~$30 M on UDD conversions, >$20 M on new buckets, boom upgrades, electrical upgrades, etc. This is a natural part of the response of mining companies to technology and replacement programs. Capital expenditure is a normal part of the ongoing success of most companies. There is however, a tendency for some people to rely on the capital alone to provide the ongoing improvements in equipment productivity. Implementing new technology through capital expenditure is only part of the equation in continuous improvement. For 80 - 90 per cent of the year, the operator controls the productivity achieved by the dragline. Variation between operators is huge. The average standard deviation in productivity is 12 per cent and maintenance impact is over 40 per cent. Robbins 2003, states: Contrary to what we were taught in grade school, we werenÆt all created equal. Most of us are to the left of the median on some normally distributed ability curve. Further, he states: The issue is knowing how people differ in abilities and using that knowledge to increase the likelihood that an employee will perform his or her job well. There are two options for reducing variability between operators; improving operator ability and getting the machine to take over what the operator is doing (automation). Dragline automation will be discussed, however, this paper will focus more on the æhuman factorÆ and how to establish a dragline with minimum variability between operators.

Jan 1, 2005

High-sulphidation (HS) gold, silver and/or copper deposits are generated in both the epithermal and the upper parts of the underlying porphyry environments over vertical intervals of up to 2 km. The HS deposits are generated in advanced argillic lithocaps, which are products of the absorption of acidic magmatic volatiles by voluminous groundwater systems. Mineralisation styles in HS systems reflect depth of formation as well as the interplay between structural, lithological and hydrothermal parameters. The deep parts of HS systems, at depths of >1000 m, are typified by disseminated copper¦gold mineralisation comprising digenite, chalcocite and covellite in pervasive advanced argillic as well as underlying sericitic alteration. In highly telescoped systems, suchmineralisation may overprint porphyry stocks and associatedquartz-veinlet stockworks. Intermediate levels of HS systems commonlycontain fault-controlled copper-gold mineralisation, typically as enargite in bodies of vuggy residual quartz, silicification and/or massive pyritic sulphide. The shallow parts of HS systems, at depths of The exploration focus is on four principal HS mineralisation styles: 1. copper (eg Chuquicamata, Monywa) or copper-gold (Wafi) in thedeep porphyry-hosted parts of systems preferably, in the case of thelatter, where supergene oxidation is limited and, hence, flotationmay be used for metal recovery,2. copper¦gold-bearing replacement mantos and pipes hosted bycarbonate wallrocks in the deep parts of systems (eg Bisbee),3. high-grade gold in late-stage veins or hydrothermal breccias thatoverprint the intermediate to shallow levels of systems (eg El Indio,Goldfield), and4. large, bulk-mineable gold deposits in the shallow parts of systemsthat were subjected to supergene oxidation, thereby permittingheap-leach treatment (eg Yanacocha, Pierina).To these preferred HS styles may be added the low-sulphidation veinor disseminated gold¦silver mineralisation that is commonplacealongside many HS systems (eg Victoria at Lepanto). * Originally published in the Proceedings of PACRIM æ99,published by The AusIMM.

Jan 1, 2000

The control of industrial grinding circuits in Australian concentrators has been practised for many years in one form or another. Some have established simple feedback loops in areas such as fine ore bin feeders (tonnage control), mill pump sump levels (flow control) and water additions (density control). There are also some sophisticated systems which infer circuit product size using complex measurements of mass flaws and water additions in the circuit and attempts are made to control those flaws to produce a constant circuit product size. The inferred size method in Australian concentrators has become firmly entrenched in the last ten years and control techniques involving the use of direct size measurements with reliable on-stream sizers have been Ignored.

Jan 1, 1983

By Edwards A. B

In their description of the post-mine leaching of galena and marmatite at Broken Hill, Messrs. Garretty and Blanchard1 find themselves unable to account satisfactorily for the fact that the more usual order of solubility of these two minerals is here reversed, and that the galena is leached more rapidly than the marmatite. An explanation of this somewhat unusual phenomena is perhaps provided by the recent work of Garrels.2 Garrels showed that the presence of Cl' ions in Pb-Zn solutions, in concentrations in excess of 0·1 N, rendered the zinc less soluble than the lead. It seems possible that Cl' ions may be present in the underground waters of the North Broken Hill Mine in sufficient concentration to cause such reversal in the relative solubility of the lead and zinc sulphides.Thus the analysis of "water believed to be free from contamination by mine drainage" given by Messrs. Garretty and Blanchard (p. 158 of their paper) is as follows:Total solids in solution 878 grains per gallon...164 grams of Cl per gallon is equivalent to 2· 3394 grams of Cl' per litre, which is equivalent to 0066 N.This concentration is approaching a value sufficient to cause a reversal of the relative solubility of zinc and lead salts; and it seems probable that there will be parts of the ore-body where locally the Cl' concentration will greatly exceed this figure. At such points the lead salts would be more soluble than the zinc salts, so that the galena would be leached more rapidly than the marmatite. With the dilution of such Cl'-rich waters with water containing less Cl', the Cl' concentration would fall below 0·1 N, and the lead would be reprecipitated, while the zinc would stay in solution, as, for example, in the water analysis quoted. The other point on which the authors speculate is the reason why the Northern Ore-body in the mine has undergone greater leaching than the...

Jan 1, 1942

IN carrying out experiments to determine the amount of metal taken away by fume or fine dust the apparatus used requires to be strong and compact, as it is generaIIy necessary to work on a staging fixed at considerable distance above the ground and against a smoke stack or flue. It is essential that the test be carried out at a part of the stack welI away from any bends, to prevent eddies. The method described below has been used for estimating the dust losses from Dwight & Lloyd sintering machines, antimony drossing furnaces, pot sintering trials, etc., and has been founc! satisfaetory. It is welI to bear in mind that the longer the period over which dust coIIection is extended, the more valuable the result is likely to be. The rateof aspiration of the gases through the apparatus is kept as regular as possible by changing the filter thimbles frequently, and the rate of flow should correspond closely with the rate of flow of the gases up the stack.The apparatus required for carrying out a test is as folIows :-{a) Thermometers : two are required, one for taking the temperature of the gas in the stack and one for giving the temperature of the gas in the aspirating drum. If the temperature in the stack is over 400° C., it is advisable to use a pyrometer such as the "Bristol" the couple being standardized before use at about the temperature of the stack.(b) Pilot Tube: The foIIowing sketch (Fig. 1) iIIustrates the principle of the pitot tube. The readings are to be taken in a straight flue, or stack, preferably 20 diam. away from any bend. A, which is open, faces the air current and...

Jan 1, 1922

The advent and easy availability of fast and powerful digital computers has marked a transition from one era to another in the field of process control in many industries. Mineral processing is no exception. Aims, practices and problems associated with process control of mineral beneficiation plants in the 'old days' are compared with those available at present. Although basic processes remain generally the same, there have been simultaneous social and economic developments, which directly relate to the development and present value of process control. Lower head grades are now treated, plant capacities have increased, and a large proportion of future capital in the mineral industry will be associated with large capacity plant, treating still lower grades. Environmental considerations have pressurised moves to better control in some areas. With these changes, economic operating incentives have changed, and process control has, in many plants, become more critical. Finally, while computer technology has developed recently at a very rapid rate, developments in the fields of instrumentation, basic on-line measurement and sampling have in general lagged behind, limiting the progress possible in process control. Hence, the benefits of computer control vary depending on

Jan 1, 1977

By Y Peng, D J. Bradshaw, X Chen

At NewcrestÆs Telfer gold mine, a strategy of regrinding pyrite rougher flotation concentrates has been proposed to increase gold recovery and reduce cyanide consumption in the CIL circuit. The regrinding will be followed by an additional copper/gold flotation stage, where copper minerals need to be separated from pyrite. However, although the minerals are well liberated, the separation of chalcopyrite from pyrite in the cleaner stage is not always straightforward. The surfaces of the fine particles, produced during regrinding, are more easily oxidised and contaminated than coarse particles, directly influencing the particle floatability. The depression of pyrite is difficult in the cleaner stage as well due to copper activation, a process which is also highly dependent on the grinding environment. In this study, different grinding media, including mild steel and stainless steel, was used in the regrinding to produce different regrinding chemistry, such as, pH, Eh, and iron hydroxide contamination. The effects of regrinding conditions on chalcopyrite flotation and its separation from pyrite were studied. It was found that the stainless steel grinding media, which produced an oxidising environment, was more beneficial for the separation of chalcopyrite from pyrite after regrinding. The mild steel grinding media had a detrimental effect on both chalcopyrite and pyrite flotation. This study addresses the importance of choosing suitable regrinding conditions in the pyrite and copper flotation circuit.CITATION:Chen, X, Peng, Y and Bradshaw, D J, 2012. Effect of regrinding conditions on the flotation of chalcopyrite and its separation from pyrite in the cleaner stage, in Proceedings 11th AusIMM Mill OperatorsÆ Conference, pp 279-286 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Oct 29, 2012

By Johnstone RD, Rattenbury MS

RACE Quartz-calcite veins containing Au, As, W, Cu, Pb, Zn and Fe cross-cut Alpine Fault- related structures in the Alpine Schist of South Island, New Zealand. These veins, and some other related veins have been deposited by metamorphic fluids at 250¦C to >320¦C at pressures of a few hundred bars. The fluids were very variable in composition, with 0 to 50 mole % CO2 and 2 to 18 wt. % NaCl equivalent. Penetration of hot metamorphic fluids to shallow levels in the crust was facilitated by rapid uplift of hot rocks at a compressive plate margin. Fluid immiscibility at shallow levels caused concentration of salts and metals into the aqueous phase. Gold precipitation was caused by reduction by pelitic country rock and/nr dluILion by other aqueous fluids.

Jan 1, 1987

Australia, with its fast-growing economy and high standard of a~ living, has a high and rapidly increasing per capita consumption of energyIn total terms, Australia's energy requirements are doubling every 15 years, and among mature Western economies Australia ranks behind only Italy-and Japan in the rate of increase of energy consumption.The use of energy is often accepted as a yardstick of development and on this basis Australia might well congratulate itself on its high energy consumption. However, it is less well recognised that the resources absorbed in the production of energy represent a real cost to the* community, and that just as we strive for industrial efficiency or farm productivity we should strive for economy in the use of energy fuels.Australians spend annually an estimated ú500 million.on all forms of energy, a sum which represents about seven per cent of the Gross National Product. In all, energy production and distribution absorb the employment of more than 100,000 persons. In addition, the annual capital investment by.the oil industry amounts to some E55 millions, by the coal industry about ú12 millions, and by electrical and gas authorities some E108 millions, in total representing more than ten per cent of annual investment in other than private motor cars and housing,

Jan 1, 1962

By L Grigorov, G J. Jame, A V. Nguyen

The dewetting kinetics between a small air-liquid interface and a silica (negatively charged in water) planar surface in solutions of dodecylammonium chloride (cationic surfactant) has been investigated using the Scheludko cell and digital high-speed video microscopy. The gas-liquid interface was created at the bottom of a small silica capillary of the cell and then was moved towards the silica planar surface. After the rupture of the liquid films between the interfaces, the expansion of the gas-liquid-solid three-phase contact (TPC) line was observed and recorded using a digital high-speed video microscope system, operating at 1000 frames per second. The surface tension of the surfactant solution was measured using the pendant drop technique. The equilibrium contact angle was measured using the Scheludko-Minings method with a silica micro-sphere. The TPC radius was determined as a function of time and compared against the available hydrodynamic and molecular-kinetic models. The experimental data are described very well by the molecular-kinetic model for the TPC line motion. The hydrodynamic model is in agreement with the experimental data only for a short time after inception of the three-phase contact expansion.

Jan 1, 2005

By R Balusu, P Humphries

In combination with detailed field studies and goaf gas characterisation, computational fluid dynamics (CFD) models have been used to develop optimum and effective strategies for inertisation during longwall sealing operations to achieve goaf inertisation within a few hours of panel seal-off operation. This study has combined detailed analysis of the performance of various inertisation field trials together with CFD modelling results of different inertisation operations in order to develop the optimum inertisation strategies. A number of parametric studies were conducted on the base case CFD models that had been calibrated and validated based on the information obtained from previous inertisation studies and goaf gas monitoring. These studies included changes in inert gas injection locations, inert gas flow rates, seam gradients, and different inertisation strategies to investigate their effect on goaf inertisation. Studies indicate that the strategy of inert gas injection through the MG seal was not as effective as the alternative strategy of inert gas injection at 200 m behind the face. Innovative inertisation strategies have been developed and subsequently implemented at an underground coal mine in Australia. Further investigations have been carried out in the development and demonstration of proactive inertisation strategies with the objective to suppress the occurrence of spontaneous heatings in active longwall faces, in particular under unexpected scenarios such as during slow retreat/face stoppage due to difficult geological conditions. Initial trials at two Australian longwall panels have demonstrated the great potential of this practice to contain the onset of heatings in the goaf.

Jan 1, 2005

By Waladan M, Elliot B, Rankin W. J

A survey was undertaken on the Kalgoorlie Nickel Smelter in which samples of slag and matte were taken along the length of the integrated flash smelting furnace and analysed to determine the variation of nickel and iron concentration. At the same time the oxygen potential of the slag and matte was measured using disposable tip solid-state oxygen probes. The results show that the oxygen potential of the slag is higher than that of the matte and that both decrease from the reaction shaft end to the slag tap hole end of the furnace. The slag and matte are not at equilibrium except towards the tap hole end. Reduction of nickel and magnetite from the slag occurs before the slag enters the slag-cleaning appendage by reaction with iron in the matte. Thermodynamic modelling and laboratory experiments showed that the equilibrium nickel and magnetite contents of the slag in the absence of a carbonaceous reducing agent are about 0.4 per cent and 7.5 per cent, respectively. A carbonaceous reductant is required to obtain a lower nickel content in the slag.

Jan 1, 1991

By Oversby B, Wellman P, Wyborn D, McPhie J, MacKenzie D

The Bureau of Mineral Resources (BMR) and the Queensland Mines Department are engaged in integrated regional- and deposit-scale studies in the Burdekin Falls-"Conway" area, about 100km southeast of Charters Towers (Fig. 1). The regional work augments recent mapping in the Mount Coolon area farther south (Hutton et al., in prep.). These studies are aimed at improving understanding of gold mineralisation and its geological setting in the Drurunond Basin and associated rocks. Knowledge of the regional stratigraphic- structural framework and setting of mineralisation has until now been based on investigations dating from the early to mid 1960's (mainly Bowen 1:250 000 Sheet; e.g. Paine et al., 1974; Wyatt et al., 1971), and on regional syntheses (Olgers, 1972; Levingston, 1981). Some recently explored epithermal gold occurrences in the region have been interpreted on the basis of this early work as being hosted by the Late Carboniferous Bulgonunna Volcanics. The objective of our field research was to refine the regional geology, paying particular attention to the Bulgonunna Volcanics. The results summarised here indicate that unnamed Devonian-Carboniferous volcanics (DCv) of the northern Drummond Basin and mid to late Palaeozoic granitoids are much more extensive, and the Bulgonunna Volcanics much less so, than was previously recognised.

Jan 1, 1989