Search Documents

By D. C. Bailey, D. I. Hodson

Paper presented at the annual joint meeting of the IMM and IMinE held in Harrogate, 31 January 1991. The Neves-Corvo copper-tin mine is the highest-grade large-tonnage copper operation in the world and is now arguably the most significant producing base-metal mine in Europe, rating in the top ten world producers of both copper and tin. A description is given of the initial exploration work and the development of the project from the orebody discovery in the Iberian Pyrite Belt of southern Portugal in 1977 to full operating capacity in 1990. The chronological account includes discussion of the main organisational, financial, technical and environmental decisions and of the various difficulties encountered, and the challenges of the future are identified. The project was brought in on schedule and within budget

Jan 1, 1991

By K. M. Kinealy, A. R. Ramsden, D. H. French

A negative correlation exists between sphalerite composition and gold content when gold is transported as Au(HS)2(-); however, there are spatial and temporal differences within a single deposit. A previous model for the formation of volcanogenic massive-sulphide deposits within thermally intensifying and waning hydrothermal sytems in much the same way as is observed around modern sea-floor vents is supported. The results also emphasise that recognition of the mechanisms of gold transport in ancient deposits depends critically on an understanding of their mineralogy and stratigraphy

Jan 1, 1991

By I. R. Hanby

A simple solution to the problem of whether to round the calculated number of trucks up or down to the nearest whole number is to adopt as criterion the minimum cost per unit volume/mass moved, assuming all trucks in the fleet are of the same type. When the effects on journey time of varying haul-road slope are calculated for several rolling resistances, it is shown that the optimum angle of 7.5 degrees is moderately independent of both rolling resistance and height climbed; the lower the resistance, the lower the sensitivity to ramp angle, but if haul-road gradients do not exceed 1 in 10 (5.7 degrees) for safety reasons then travelling time will never be more than 7% greater than its theoretical minimum. Further calculations result in a graph of the maximum payload values to incur the minimum cost, for different values of total effective resistance

Jan 1, 1991

By L. G. Billington

"The Agnew nickel deposit occurs in a major linear greenstone belt in the Archaean Eastern Goldfields Province of Western Australia, and has been subjected to mid-amphibolite facies nietamorphism of dynamic style. The largest dunitic ultramafic body in the Perseverance Block is host to the bulk of the nickel mineralization. The Perseverance Ultramafic is mineralogically zoned and has an embayed western margin against a mixed metasedimentarv-metavolcanic wallrock sequence.Three categories of nickel sulphide mineralization are concentrated on this western margin at the thickest part of the ultramafic exposure. Massive sulphides characteristically occur as contact mineralization, internal mineralization and as remobilized concentrations in vein form structures. Disseminated ore constitutes the largest volume of mineralized materials in the deposit and occurs in three separate shoots in en-echelon relation. Weakly mineralized 'cloud sulphide' zones surround and envelop the shoots and persist much further along strike. In the upper mine levels the shoots are truncated by a shear or crush zone, which inhibits geological knowledge to the south but marks a major fundamental structural break.Most of the ore types have suffered supergene alteration and there is a complex variable distribution of the mineralogical facies. The ore types have been metamorphically retextured, but the oreshoots can be characterized by their Ni tenor and some chemical ratios. S/Se values, PGF. and precious metal values are distinctly different from those of other nickel sulphide ores extracted.Despite intense structural and metamorphic modification of the mineralizations, some features support an original magmatic derivation for the sulphides. These include the localization of massive sulphide ore types on one marginal contact within an emhayment structure, the general distribution of disseminated mineralizations located immediately above the massive sulphides and the Ni/Cu, Ni/Co, Fe/Ni ratios of the oreshoots being similar to those of deposits that typify the intrusive dunite- associated class of nickel sulphide deposit. Varying S/Ni ratios between different shoots suggest a revision of the single magmatic intrusion model in favour of a changing composition between successive magmatic episodes."

Jan 1, 1984

By K. C. Misra, W. Wrightson Jr.

The Lick Fork prospect shows appreciable nickel sulphide mineralization in a narrow, steeply dipping, layered sequence of metamorphosed hornblende peridotite, gahhro and hornblende gahhro, intrusive into the Pedlar Formation (charnockitic pyroxene granulite) of the Virginia Blue Ridge Complex (Precambrian). Rocks of the intrusion have equilibrated to granulite- facies metamorphism, as indicated by pervasive polygonal texture of silicate aggregates, development of biotite- and hornblende-preferred orientation, compositional equilibration of the silicate minerals, the Mg-Fe2+ (orthopyroxene-dino- pvroxene) distribution coefficients (average 0.57), (MgO + FeO + Fe20 3)/Al20 3ratios in orthopyroxenes, and Ti contents and AI(i\)/A I(\i) ratios of hornblendes. The sulphides, consisting of pyrrhotite-pentlandite (with up to 4 wt% Co)-chalcopyrite-pyrite-marcasite-violarite- bornite—mackinawite, are virtually restricted to the ultramafic unit of the intrusion and constitute 20-30% of the samples of hornblende peridotite studied. Although the basal accumulation of the sulphides may suggest magmatic segregation, sulphide- silicate textures typical of sulphide liquid immiscibilitv are not present, even in the sparsely mineralized gahbro units. On the other hand, the transgressive and relict textures are strongly suggestive of replacement of the silicate minerals by sulphides. These include chaleopyrite ± pyrrhotite ± pentlandite (violarite) ± pyrite ±magnetite veins crosscutting silicates, sulphide penetration into silicates at silicate-sulphide contact zones, rounding of silicate grains against sulphide matrix, and islands of silicate minerals (olivine, hornblende, pyroxene, biotite) in sulphide masses. Metamorphic mobilization of pre-existing sul­ phides is a possibility, but involvement of post-metamorphic, hydrothermal fluids is suggested by the following features: paucity of annealing and deformation textures in the sulphides, replacement of metamorphic and deformed silicates, lack of differential sulphide mobilization and the common occurrence of sulphides along grain boundaries and triple junctions of poly gonal silicate aggregates. The consistent association between sulphides and hydrothermal alteration products of silicates suggests that they are cogenetic. The hydrothermal fluids may have merely mobilized and concentrated the pre-existing, disseminated, magmatic sulphides, or there may have been significant addition of sulphides during this process. This question cannot be resolved unequivocally with the existing data.

Jan 1, 1984

By E. C. I. Hammerbeck

"Geological controls and constraints in space and time define the milieu of the nickel deposits of Southern Africa. Primary, magmatic deposits are of two types, e.g. synvolcanic or syntectonic deposits confined to certain orogenic domains, and those related to intracratonic, post-orogenic magmatism, e.g. the Great Dyke, the Bushveld Complex and the Insizwa intrusion. Epigenetic and supergene deposits form a separate, unrelated group and are of minor economic importance.The metallogenic evolution of syntectonic nickel deposits is particularly well illustrated in the Zimbabwean Province, where the relationships between Archaean greenstone development and nickel mineralization are best illustrated. Volumetrically subordinate, nickel-poor greenstones in the 3500 m.y. old Sebakwian Group are followed by more common komatiite development in the Lower Greenstones (3000 m.y.), and, finally, by very significant komatiites with important nickel mineralization in the 2700 m.y. old Upper Greenstones of the Bulawayan Group. Comparable nickel mineralization is not known in the Raapvaal Province, in spite of ubiquitous komatiitic magmatism of the Onverwacht Group (3500 m.y.) and other greenstone sequences. No komatiitic rocks younger than about 3200 m.y. are known in this region, orogenic activity in the Kaapvaal tectonic province having ceased at about 3100 m.y., whereas the Zimbabwe Province remained tectonically active until at least 2600 m.y.The tectonic environment, host-rock petrology and mode of emplacement, as well as some details of the nickel mineralization, are the parameters employed to classify the Southern African magmatic nickel deposits. A rating of the deposits is introduced that reflects the original resources of the deposit, its average nickel content, its mining history and potential to facilitate ready comparison with and assessment of the nickel deposits listed in the classification, as well as some important nickel deposits worldwide."

Jan 1, 1984

By B. H. Smith

"Most of the discoveries of nickel sulphide deposits in Western Australia have followed the search for gossans. The chances of further discoveries depend on improving our understanding of the cover of strongly leached and transported materials that obscure approximately 85% of the area of the Yilgarn Block that has nickel sulphide mineralization potential. The lateritic profiles that are observed on the Yilgarn Block today were formed over a period of more than 100 m.y. of warm and humid climates. This lateritized land surface has been preserved by induration during subsequent arid climates as well as by the structural stability of the Yilgarn Block.In geochemical surveys of lateritied terrain samples have to be taken from at least the saprolite zone usually by drilling so that confident predictions can be made of the significance of analytical data. Account must be taken of the mineralogy of samples, as well as the nature of the horizon of a laterite profile that is being sampled, owing to the concentrating effects on trace elements of Fe and Mn oxides.Multivariate techniques have been applied successfully to analytical data in gossan recognition when textural evidence is ambiguous. Pathfinder geochemistry has also been used in gossan recognition and the most significant recent advance has come from studies of platinum group metals in nickel sulphides and their dispersion during weathering."

Jan 1, 1984

By R. E. T. Hill

"A laboratory study was undertaken to determine the condensed equilibrium phase relations at 600°C in that portion of the quaternary Fe-Ni-Cu-S system which lies in the neighbourhood of the monosulphide solid solution (MSS). The study sought to determine to w hat extent copper affects the temperatures of stability and compositions of iron-nickel sulphides in bulk compositions similar to those of massive Fe-Ni-Cu ores. Experimental data, including stable phase assemblages and phase compositions, are presented.In the quaternary system at 600°C MSS forms stable phase assemblages with pentlandite, chalcopyrite solid solution, pyrite and vaesite. Over the range of bulk compositions displayed by natural ores copper as well as metal/sulphur ratio have a significant effect on the thermal stability of pyrite and pentlandite. Errors therefore accompany the widespread practice of ignoring copper or chalcopyrite when the mineralogical and compositional changes that have accompanied the initial cooling of primary magmatic sulphides or their subsequent homogenization during isochemical metamorphism are being deciphered. These errors are most significant for ores the bulk compositions of which lie tow ard the sulphur-deficient side of the cupriferous MSS. Temperatures as much as 300°C above the copper-absent value are possible for the interpreted upper stability limits of pentlandite in these ores.The experimental data show that up to 2.0 at% copper can occur in pentlandite stable at 600°C in a bulk composition equivalent to natural magmatic ores. This is equivalent to nearly 10 vol% chalcopyrite. Pentlandite contains more copper than coexisting MSS and the copper content of both phases increases with increasing iron."

Jan 1, 1984

Jan 1, 1984

By Le Thac Xinh, van Chu Nguyen

In northern Vietnam Ni-Cu sulphide deposits associated with ultramafic rocks are located in several continental rift zones of Permo-Triassic age. In recent years a number of ultramafic bodies have also been discovered in southern Vietnam, but detailed work on sulphide occurrences within these bodies has still to be undertaken. A brief description is given of some N i-C u sulphide deposits that are located in the northern part of the country. Of the depos­ its that have been examined in any detail, those in the Takhoa region are of the greatest economic interest. Ni and Ni-Cu sulphide bodies have, however, also been found in association with the ultramafic rocks.

Jan 1, 1984

Jan 1, 1984

"The sulphide nickel resources of Australia are concentrated in the Western Australian Shield. This shield is considered to have developed by extensive, but incomplete, Proterozoic reworking of once-continuous Archaean crust. It now constitutes relic Archaean cratons and intervening orogenic mobile belts transected by narrow, curvilinear tectono-thermal mobile belts of younger Proterozoic age. Remnants of the Archaean cratons are exposed in the Yilgarn and Pilbara Blocks. In the former an ancient (to ca 3600 m.y.) high-grade gneiss terrain (Western Gneiss Terrain) occurs along the western margin of the block and is interpreted to flank younger (ca3000-2800 m.y.) greenstone belts to the east. Ancient (to ca3500 m.y.) granitoid-greenstone terrains dominate the Pilbara Block.Proterozoic mobile belts contain only traces of nickel mineralization, except for one locality in the Halls Creek Province. The Archaean high-grade gneiss terrains contain no important mineralization, and the Pilbara granitoid-greenstone terrains contain only scattered examples of less important types of mineralization.The important deposits (e.g. Kambalda, Windarra, Agnew, Mt Keith) are all of komatiitic affinity and are best developed in the youngest greenstone belts of the eastern part of the Yilgarn Block. In particular, they are concentrated within, and adjacent to, the Norseman-Wiluna Belt, which contrasts in age, lithofacies and structural style with all other greenstone belts in the shield. It is interpreted to have been a rift zone. Within this belt the two main deposit types—the volcanic peridotite-associated and intrusive dunite-associated deposits—are concentrated into spatially separate zones. Precise controls on this heterogeneous distribution cannot be determined unequivocally from available data. One possibility is that the major volcanic peridotite-associated deposits were developed selectively in the actively subsiding axial part of the rift zone during the early extensional phase, whereas intrusive dunite-associated deposits were formed during a subsequent (less active?) stage of volcanic basin development. Distinctive sub-types (facies) of the volcanic peridotite-associated deposits form clusters that are generally within the rift zone. These may represent deposits related to discrete eruption centres, and at least part of their variation is due to relative position within extensive, topographically confined, linear flows."

Jan 1, 1984

By A. J. Naldrett

"Synorogenic mafic intrusions are a common feature of the northern Appalachians. Two sulphide-bearing intrusions in central Maine that have been investigated revealed a number of important features that illustrate sulphide-silicate interaction that relates to magmatic sulphides. The Moxie Pluton is a large mafic intrusion emplaced during the Devonian Acadian orogeny. The Katahdin gabbro is a small mafic body that is probably genetically and temporally related to the Moxie Pluton. Both intrusions were formed bv the intrusion of variably fractionated magmas into irregular chambers and feeder zones. The southern half of the Moxie Pluton contains three small sulphide accumulations, whereas the Katahdin gabbro hosts a large, predominantly massive, sulphide body. Textural evidence clearly documents the magmatic origin of these sulphides.The three Moxie Pluton sulphide bodies are located at or very close to the intersection between the intrusion and the sulphidebearing Hildreths Formation. Sulphur assimilation, suggested by this relationship, is supported by the isotopically light (634S = —8 to —25) and the selenium-poor (S/Se - 25 000-48 000) character of the Moxie sulphides and correlation with the Hildreths Formation sulphides (S34S = —25, S/Se = 25 000-48 000). Isotopically light sulphur (534S = —20 to —21) and high S/Se ratios indicate that the majority of sulphur in the Katahdin sulphide body was also derived from an unexposed country rock source. The volume of sulphide in this example illustrates the potential efficiency of the sulphur assimilation process—so efficient in this case that marked dilution of the economic chalcophile elements resulted from the addition of so much external sulphur.The three Moxie Pluton sulphide occurrences are all impoverished in Ni and Cu (1.87-0.44% Ni; 0.72—0.32% Cu in massive sulphide), but contain normal Co contents (0.22-0.13%) relative to typical magmatic sulphides. The Katahdin sulphides show a greater Ni and Cu impoverishment (0.2% Ni; 0.1% Cu in massive sulphide) with similar Co (0.17%). The Ni content of the Moxie Pluton olivines document the fractional removal of sulphides during crystallization of the host magma, which resulted in the progressive depletion of the chalcophile metals in this magma. Modelling that involved the fractional crystallization of the magma coupled with the fractional removal of a specified proportion of sulphide has successfully duplicated the Moxie Pluton sulphide compositions. A similar magmatic model has been postulated for the Katahdin sulphides with a greater degree of sulphide fractionation. This latter model is again supported by extremely low Ni/forsterite ratios in olivines, though the compositional range is insufficient to document fractionation. These magmatic models are not unique, but they provide a relatively constrained and consistent explanation for the sulphide compositions.The Ni contents of olivines are the most important indicator of sulphide-silicate interaction in mafic magmas and have exploration potential. They provide a means of assessing the timing and extent of sulphur saturation in an intrusion and are a guide to the Ni content of a sulphide accumulation in equilibrium with olivine. In the Maine examples they provide strong evidence of the magmatic origin of the sulphide accumulations and justification for magmatic modelling."

Jan 1, 1984

By T. Piirainen, T. Alapieti

"The Koillismaa layered igneous complex is the result of igneous activity that occurred 2436±5 m.y. ago. Its crystallization took place in three separate but connected magma chambers, the total volume of which is estimated at more than 2000 km'. These chambers were located partly in a feeder system within Archaean granitoids and partly in an unconformity between these grani­ toids and Early Proterozoic supracrustal rocks. The composition of the parental magma that initially filled the chambers is thought to have been that of a relatively MgO-rich tholeiite. Sulphides and platinum-group minerals are mainly concentrated in the marginal series of the western part, where they are encountered over a distance of 25 km. This marginal series is an originally nearly horizontal series of rock types that displays a reversed differentiation trend, grading inwards from gabbroic to ultramafic rocks, followed bv rocks of the layered series. The chilled margin is not encountered anywhere in its original position at the contact, probably as a consequence of magmatic erosion, but remnants are to be found in the form of elongated xenoliths. Partly melted siliceous fragments from gneissic country rock typical of the lower part of the marginal series indicate pronounced contamination.Chalcopyrite, pentlandite and pyrrhotite±pyrite are the predominant sulphides. The accessory sulphide minerals include covellite, violarite, marcasite, mackinawite, sphalerite and argentian pentlandite. Platinum-group minerals are represented by froodite, michenerite, merenskyite and sperrylite. Other accessories that have been identified are electrum and native copper and gold. The mean of the Cu/(Cu+Ni) ratio in the mineralizations is 0.67 and Pt/(Pt+Pd) 0.31.It is suggested that the sulphide phase was formed in the marginal series as the solubility of sulphur in the basic magma was lowered by contamination, possibly assisted by some other factors. The sulphur isotope data support the juvenile origin of the sulphur."

Jan 1, 1984

By D. L. Buchanan, M. J. Jones

In May, 1982, the University of Western Australia in Perth hosted a three-day international conference on nickel sulphides, followed by a ten-day field excursion to the Western Australian nickel sulphide deposits. The conference was arranged by Professor D. I. Groves, C. M. Lesher and Dr. R. E. T. Hill under the auspices of the International Geological Correlation Programme's Projects 161 ("Sulphide deposits in mafic and ultramafic rocks") and 91 ("Metallogeny of the Pre- cambrian"). This volume contains a selection of the papers that were presented at the Perth conference. Most of the papers are related to the nickel sulphide deposits of Western Australia, but contributions with settings in North America, Finland, Vietnam and Southern Africa have also been included.

Jan 1, 1984

By M. L. Schmulian

"The Windarra nickel deposits display many of the characteristics that typify Archaean volcanic peridotite-associated nickel sulphide ores of Western Australia. The Fe-Ni-Cu sulphide mineralization, which comprises dominantly disseminated sulphides, occurs at the base of a differentiated ultramafic volcanic pile grading upwards from massive meta-peridotitic flows to thin, locally spinifex-textured meta-picritic units. The entire mafic-ultramafic belt has been subjected to mid-amphibolite- facies metamorphism. The ore-bearing ultramafic is immediately underlain by dominantly sulphide-facies banded iron formation (BIF), which commonly contains nickel sulphides adjacent to its contact with mineralized ultramafic. BIF-hosted ores are formed both by physical remobilization of ultramafic-hosted sulphides and by migration of Ni and Cu into pre-existing iron sulphides.At Mt. Windarra the mineralization comprises eight steeply dipping, steeply plunging elongate oreshoots, which display a marked parallelism with major and minor fold structures in the underlying BIF. Seven of the shoots occur at or close to the base of the ultramafic pile, and are hosted by at least three overlapping flows. The zone of overlap occurs at the northern end of each flow, where picritic ultramafic, representing a dominantly liquid portion of the flow, occupies the basal contact. A gross south- north ultramafic flow direction is inferred. The eighth shoot occurs in a stratigraphically higher position within the ultramafic sequence.At South Windarra at least two overlapping mineralized flows occupy the basal contact, the ore zone having a strike extent three to four times its downdip extent. There is no obvious structural control to the mineralization."

Jan 1, 1984

By D. J. Esdale, D. F. Pridmore, F. W. Lindman, J. H. Coggon

Almost all the presently viable nickel reserves in the Yilgarn Block contain nickel in a sulphide phase. Both volcanic peridotite- associated (VPA) deposits and intrusive dunite-associated (IDA) deposits are characterized by massive, matrix and disseminated sulphides (typically, pyrrhotite, pentlandite, pyrite, chalcopyrite, magnetite and other spinels), though in the IDA deposits the volume of disseminated sulphides is far greater than that in the VPA deposits. Potentially all geophysical techniques have application in the direct detection of these deposits'gravity for the detection of the dense massive and matrix ore; magnetics for the detection of magnetite and pyrrhotite in the ores; electromagnetics for the detection of the massive and matrix ore; and induced polarization for the detection of disseminated and to some extent matrix and massive ore. In practice the application of geophysical techniques is limited to a varying extent by (1) the relatively small size of an economic accumulation of sulphides, particularly the VPA deposits, (2) the extreme depth of weathering and lack of outcrop on the Yilgarn Block and (3) the abundance of geological noise or anomalous responses caused by sources other than economic mineralization.

Jan 1, 1984

By L. R. Chimimba

"The Trojan nickel deposit occurs within a pile of ultramafic lavas subjected to a lower amphibolite grade of metamorphism. The lavas form part of the Upper Greenstones (Bulawayan Group) of the Mazoe greenstone belt. The nickel mineralization occurs as disseminated, massive and near-massive sulphides. The massive and near-massive ore occurs along the basal contact of the serpentinite units and is overlain by the more abundant disseminated ore, which occurs within the serpentinite. The main sulphide minerals are pyrrhotite. pentlandite and chalcopvrite. Minor amounts of millerite and pvrite are present. The Ni : Cu ratio in all three types of ore averages 15:1.The intercumulus nature of sulphides in the near-massive ore and similar textures occasionally seen in the disseminated ore indicates a magmatic origin of the mineralization. The basal position of the near-massive and massive ore suggests that the sul­ phides settled out under gravity during the emplacement and crystallization of a peridotitic magma. The deposit has been deformed and some of the primary massive ore has been remobilized into veins and breccias. The disseminated mineralization now occurs mainly as blebs of sulphides intergrown with silicate minerals or as stringers of sulphides between foliation planes in the serpentinite."

Jan 1, 1984

By H. L. Paterson, R. R. Keays, M. J. Donaldson, D. J. Boyack, M. F. Smith, J. J. Gresham

"Albite-rich sedimentary rocks that contain abundant Fe sulphides occur in the lower portion of the ultramafic sequence at Kambalda and their broadly antipathetic relationship to most of the nickel orebodies is well documented. These sediments generally contain less than 0.1% Ni, but nickeliferous equivalents with 0.3 to >6% Ni occur locally in close spatial association with more typical volcanic peridotite-associated massive, matrix and disseminated ores. Nickeliferous sediments have been mined at several locations at Kambalda but, collectively, they constitute less than 2% of the total Kambalda nickel reserve. The nickeliferous sediments occur both on the ultramafic-footwall basalt contact and within the lower part of the ultramafic sequence, and are usually less than 5 m thick. They are commonly finely laminated rocks with a cherty appearance, composed essentially of layers of albite+quartz±amphibole, alternating with pyrrhotite+pentlandite±pyrite layers. Some contain a minor amount of carbon. The sediments are only of ore grade (> 1% Ni) immediately adjacent to massive Fe-Ni-Cu sulphides, which are interpreted as normal magmatic ores on the basis of geochemistry and the presence of distinctive ferrochromites. The nickel content of the sediments decreases to background levels (<0.1%) within about 20 m of the massive sulphides. In most locations the massive sulphides clearly have been emplaced tectonically within or adjacent to the sediments, which are commonly highly deformed. Typical nickeliferous sediments contain about 1% Ni, 0.1% Cu, 0.1% Zn, 400 ppb Pd, <0.2 ppb Ir and 10-200 ppb Au. They have lower Ni/Cu (average 10), higher S/Ni (average 6) and higher S/Se (average 11 200) than associated massive sulphides (average 40, 4.8 and 7400, respectively). The Pd/Ir ratio of the nickeliferous sediments (typically > 1000) contrasts markedly with that of associated massive sulphides (average ca 16).The high Pd/Ir values are similar to those which have been reported for hydrothermal Fe-Ni sulphides and barren sulphidic sediments of probable volcanic exhalative origin. They preclude a direct magmatic origin for the sulphides in the nickeliferous sediments, though S/Se and S isotope data indicate a magmatic source for the sulphur. The restriction of nickeliferous sediments to within about 20 m of normal magmatic sulphides and the lack of a widespread Ni dispersion halo suggest local redistribution of originally magmatic Ni. This probably occurred by diffusion and hydrothermal mobilization during deformation and amphibolite-facies metamorphism. Although not totally ruled out, a volcanic exhalative origin for the Ni seems unlikely."

Jan 1, 1984

By J. E. Rouse, D. L. Buchanan

"During the emplacement of the Bushveld Complex in the Potgietersrus area magma reacted with the original floor rocks— Archaean granite or Malmani dolomite and overlying Penge handed ironstone or sedimentary rocks of the Pretoria Series. The complex sequence of sulphide-rich mafic rocks that formed along the contact is known as the Platreef. Associated with the sul­ phide mineralization are important concentrations of platinum-group metals, copper and nickel. Xenoliths of Malmani dolo­ mite have been incorporated into the Platreef and previous work has suggested that anhydrite is likely to have been present in the dolomite. The sedimentarv-derived sulphur thus introduced into the magma exercised an important control on the forma­ tion of the sulphides.Petrographic studies on the Platreef rocks from the Tweefontein area indicate that the original magma reacted with the banded ironstone to form iron-rich pyroxenites. In the Turfspruit area, to the south, the footwall rocks consist of sediments of the Timeball Hill Series. Original argillaceous rocks have been metamorphosed to cordierite-, corundum- and spinel-bearing hornfelses. Graphite was present in the contact zone in both the Tweefontein and Turfspruit areas. Anhydrite-bearing hornfels rocks are also present on Turfspruit.The Platreef rocks that overlie the corundum- and spinel-bearing hornfelses on Turfspruit are peridotitie, and pyroxenites are present in areas underlain by cordierite-bearing hornfels. The absence of orthopyroxene in the peridotite is probably indicative of low silica activity in the original magma. The pyroxenites have, however, undergone silica contamination from the underlying more siliceous cordierite hornfels.Carbon isotope studies indicate that the graphite in the Tweefontein and Turfspruit samples represents organic material orig­ inally associated with the footwall sediments. Sulphur isotope determinations on the anhydrite-bearing floor rocks indicate that this phase is clearly of sedimentary origin and was probably derived from an evaporitic unit within the carbonate sequence. This provides firm support for a model of ore genesis that involved the introduction of additional sulphur into the Bushveld magma. Isotope studies on sulphides associated with the banded ironstone and graphite indicate that these units were not the source of additional sulphur. Experimental studies predict, however, that a magma contaminated with iron- and carbon-rich material would have a high sulphur-carrying capacity. Where additional sulphur is available the magma would achieve saturation early in its crystallization history and precipitate an immiscible sulphide phase. Experimental work has also shown that sulphide precipi­ tation can be triggered by the introduction of salic material into a magma close to sulphur saturation. This mechanism probably played a role in the control of the sulphide mineralization in the Turfspruit area."

Jan 1, 1984