Phalaborwa (Palabora)

The Phalaborwa carbonatite complex is probably, in economic terms, the most important carbonatite in the world, being not only one of the world's major sources of copper, but of a wide range of other valuable commodities besides - as detailed below. The complex is emplaced in Archaean granite gneisses which are fenitized along some, but not all, contacts. The central complex measures about 7 km north-south and varies between about 1.5 and 3.5 km in width, with an area of 15 km2. There are also numerous associated plugs and dykes of syenite and carbonate-bearing breccias. The irregular form of the main complex seems to have been produced by three overlapping centres of injection, each of which has a concentric distribution of rock types. The most abundant rocks are pyroxenites which grade along most of their margin to a microcline-bearing pyroxenite, also referred to as shonkinite, which is in places in contact with fenite. Towards the centre of the complex at the former hill of Loolekop (now completely removed by mining operations) a 400x900 m vertical pipe of carbonatite is surrounded by a broad zone of phoscorite (spelled foskorite in some publications) and this in turn by 'outer zone pegmatoid' with similar rocks forming an extensive area in the northern part of the complex and a smaller area in the south. Within the large area of outer zone pegmatoid in the north is an inner zone pegmatoid which is being worked at present for vermiculite. A crescentic area of hydrobiotite/glimmerite lies to the west of the carbonatite between pyroxenite and shonkinite and has gradational contacts with them. An area of similar rock also lies to the southwest of Loolekop. A large number of syenite plugs, many of which form conical hills, cut the country rocks and are particularly abundant in the vicinity of the southern half of the complex. The pyroxenites constitute about two thirds of the complex and their distribution is shown in Fig. ??, based on the work of Hanekom et al. (1965), but significant differences are shown by later work, notably in the central northern pegmatoid area (Eriksson et al., 1985; Fourie and Jager, 1989). The pyroxenites comprise clinopyroxene, phlogopite, apatite and microcline and vary from rocks consisting solely of pyroxene to glimmerites consisting only of phlogopite. These rocks are referred to as pegmatoids when the grain size exceeds 5 mm, but in many areas it can be greater than 1 metre. Eriksson et al. (1985) further divide the rocks into apatite-poor (<1% P2O5 and no visible apatite in hand specimen) and apatite-rich, the former group being concentrated in the central northern area amongst which phlogopite-pyroxene and serpentine-phlogopite pegmatoids are widespread. The massive pyroxenites consist of pale green diopside-salite, apatite, minor phlogopite, calcite and magnetite. They grade through phlogopite pyroxenites to glimmerites with <25% pyroxene. These rocks are texturally very variable and may contain alternating bands, on a scale of a few centimetres, of phlogopite and pyroxene-apatite, which persist laterally for tens of metres, generally have a curved form and are described and illustrated by Moore (1984), who describes some of the structures as 'orbicular'. The pyroxenites are characterised by their heterogeneity with banding, inclusions and cross-cutting relationships common, as illustrated by Eriksson et al. (1985, Figs 6-12). The narrow peripheral band of feldspathic pyroxenite varies from zero to several hundred metres wide and consists of subhedral to euhedral diopside crystals enclosed poikilitically by microcline. When feldspar is abundant the rims of the pyroxenes may be a deeper green. The same rock type occurs in three plugs around the complex (Hanekom et al., 1965) in one of which, Guide Copper Mine, the pyroxene is multiply zoned from diopside to aegirine-augite (Eriksson, 1985). The pegmatoids consist of phlogopite-pyroxene-apatite and serpentine-phlogopite varieties according to Eriksson et al. (1985) of which the latter consists of phlogopite dunite in which the olivine is serpentinised, and dyke-like bodies of phlogopite and serpentinised pyroxenite, but drilling indicates the pyroxene to decrease with depth. Pegmatoids of the central and southern areas are rich in apatite, which is absent from that of the northern area, while much of the phlogopite in the pegmatoids is altered to vermiculite. Analyses of the micas are given by Hanekom et al. (1965) and of micas and pyroxenes by Eriksson (1989). Between the carbonatite and the surrounding phlogopite-pyroxene-apatite pegmatoid is a zone of phoscorite, a rock first named at this locality and consisting of olivine, apatite, magnetite, carbonate minerals, phlogopite and serpentine. The proportions of the minerals vary widely from pure olivine to pure magnetite rocks with the minerals distributed approximately in vertical bands. Irregular areas of carbonate occur within the phoscorite and these become more abundant as the central carbonatite is approached. Olivine, which forms grains up to several centimetres across, is variably fresh or partly serpentinised and may be rimmed by phlogopite or chondrodite. Phlogopite has reversed pleochroism and magnetite is of several generations; grains range from <1 mm to tens of centimetres in diameter. Apatite averages 25% by volume (Hanekom et al., 1965) and forms stubby prisms several millimetres in length, as well as larger aggregates. Calcite forms veins and lenses; clinohumite, baddeleyite and a wide range of sulphides are present. The central carbonatite is of two generations: an earlier 'banded' carbonatite and a later 'transgressive' carbonatite. The banded carbonatite is intimately interdigitated with, and grades into, the phoscorite and forms numerous arcuate sheets around the central transgressive carbonatite, as clearly shown on the detailed map of the Palabora Mining Company Limited Mine Geological and Mineralogical Staff (1976, Fig. 3). Drilling has indicated that the carbonatite has a vertical, pipe-like structure that persists to a depth of more than a kilometre. The banded carbonatite is relatively coarse-grained and averages about 20% magnetite which forms continuous layers, lines of crystals and aggregates that define a vertical, elliptical banding parallel to magnetite-rich layers in the adjacent phoscorite. Calcite of the banded carbonatite contains 7.5% MgCO3 (Lombaard et al., 1964) and often shows exsolution lamellae of dolomite. Olivine is rare and may be partly replaced by phlogopite, clinohumite or monazite while there are a range of sulphide minerals the most widespread being bornite. Analyses of calcite, exsolved dolomite, apatite, magnetite and humite are given in Dawson et al. (1996). The transgressive carbonatite, which is elongated east-west, comprises a central body and numerous dykes and veins emanating from it that extend into the surrounding phoscorite and pyroxenite. There are also two arcuate sheets of transgressive carbonatite within the phoscorite east of the main body. The carbonate is more magnesian (14% MgCO3) than that of the banded carbonatite, but magnetite is equally abundant. Phlogopite is plentiful and, like that of the phoscorite, displays reverse pleochroism. Apatite forms anhedral grains and is more abundant than in the banded carbonatite; olivine is rare but chondrodite and clinohumite form aggregates of crystals up to several centimetres across. Chalcopyrite is the principal sulphide phase and contains lamellae of cubanite and bornite. An account of the copper minerals, which in an upper oxidation zone include malachite, azurite, chrysocolla, plancheite, cuprite and native copper, is given by Lombaard et al. (1964). The numerous syenite plugs cutting the basement around the complex have been described in some detail by Frick (1975). The syenites include porphyritic, hypidiomorphic, granular and gneissose varieties with a wide range of grain sizes. They consist of orthoclase, which may form phenocrysts, albite, quartz and aegirine-augite, which may be partially replaced by sodic amphibole; titanite, apatite and magnetite are accessory. Frick (1975) gives numerous analyses of the syenites which indicate that they are highly potassic rocks (up to 14% K2O) and generally peralkaline. He also describes fenites from the western margin of the central complex which involve the development of orthoclase and sodic amphibole in the basement granites from which quartz is removed. Analyses indicate that these rocks are also highly potassic. Analyses of all rock types are to be found in Hanekom et al. (1965), of the main silicate rocks in Eriksson (1989) and of phoscorite, pyroxenite and glimmerite in Fourie and Jager (1986). Data on Rb-Sr, Sm-Nd, U-Pb, O and C isotopes will be found in Eriksson (1989) and on S isotopes of carbonatites in Mitchell and Krouse (1975) and Gehlen (1976).