رویای خیس

fluids reworking the syngenetic to diagenetic mineralization (Cailteux and Kampunzu, 1995, Cailteux, 1997 and Kampunzu and Cailteux, 1999).

Brine oversaturation in metals resulting in the deposition of the syngenetic to early diagenetic sulphides was controlled by evaporation, probably within sabkha basins (Cailteux, 1986 and Garlick, 1989). Changes in Eh–pH conditions may explain the consistent sulphide zonation in the Copperbelt orebodies (Cailteux, 1986).

The model of primary concentration of copper in the Copperbelt stratiform orebodies involves the mixing of oxidized mineralising brines from the hypersaline lagoon with interstitial reducing water rich in organic compounds. This model is supported by: (1) the occurrence of relict evaporitic beds at the top of the R.A.T. Subgroup and by the collapse dissolution breccias in the Kambove Formation in Congo. This implies at least two periods of prograding sabkha-type lagoons, the first generating the lower and upper orebodies and the second forming the third, minor orebody; (2) the strong lithostratigraphic control on the position of the orebodies in Zambia and in Congo (more than 700 km along strike) and the close link between the mineralisation and anoxic shallow-water intertidal to supratidal host-sedimentary rocks. These data and the stable isotopic compositions discussed above indicate that sedimentary processes played a key-role in the metallogenetic processes.

Orebodies or sub-economic Cu sulphide ores are hosted in footwall sedimentary rocks deposited under oxidized conditions in the Zambia-type siliciclastic Mutonda Formation (e.g. Muliashi-South in the Luanshya district, Chingola; Binda, 1997). They are lenticular bodies located at different stratigraphic levels within the Mutonda unit (Fig. 6). Van Eden and Binda, 1972 and Binda, 1987 suggest a downward migration of mineralising diagenetic brines remobilising copper from the Ore Shale to the porous footwall formation. However, the occurrence in these peculiar orebodies of pyrite-I included in copper sulphides-II at Kinsenda (Ngoyi and Dejonghe, 1997) suggests an early diagenetic biogenic reduction of seawater sulphates in the precursor sediments. This could indicate that local redox fronts existed in these sediments, and that the same mineralising process as in the Ore Shale acted in the Mutonda Formation.

The model proposed in this paper links the influx of metals to the Katangan sedimentary basin to the erosion of pre-Neoproterozoic basement terrains (transportation in solution). Geochemical and geological data indicate that the Archaean Zimbabwe craton and the Palaeoproterozoic basement complex in the Bangweulu Block and within the Copperbelt represent the main sources of sediments and metals accumulated in the Katangan basin. The basement terrains particularly host lithological units with potential for the supply of large amounts of copper and cobalt and containing the required additional metal association Ni, Au, Ag, PGE (e.g. several Cu occurrences in the basement complex, cobalt in Ni-laterites formed on Archaean low–grade Ni–Co–Au–PGM deposits in the Zimbabwe craton). Some metals (e.g. U, Sn, Ta, W) most probably originated from the erosion of post-orogenic Kibaran tin-granites (Caron et al., 1986) and this is supported by the presence of detrital cassiterite, wolframite, tantalite in the orebodies in the Kolwezi mining district (Jedwab, 1997) and late Mesoproterozoic zircons in the Mwashya tuffs (Rainaud et al., 1999 and Rainaud et al., 2003).

In the case of the Lower Mwashya orebodies at Shituru, the Cu–(Co) mineralization is hosted in dolomite displaying lithological similarities with those hosting the Lower orebody in the Mines Subgroup. Minor ore in the pyroclastic rocks at Shituru probably originated from remobilisation of local sediment-hosted mineralization.

11. Conclusions

The central African Copperbelt represents a Neoproterozoic stratiform sediment-hosted province >700 km long and <100–150 km wide. It contains >140 Mt copper and >6 Mt cobalt (mined out production, ore reserves and resources). Although the Congo-type and Zambia-type orebodies show some differences (e.g. clastic vs. carbonate host rocks), they display a large number of analogies, including: (1) their location in laterally correlative lithostratigraphic units deposited in supratidal to sabkha-type highly saline environments; (2) similarities of ore textural features with predominantly disseminated sulphides closely linked to structures such as sedimentary lamination, cross-bedding, etc.; (3) identical sulphide parageneses; (4) identical zoning of sulphides related to primary variation of Eh–pH parameters during ore deposition; (5) relatively low (<100 °C) crystallisation temperature of primary syngenetic/early diagenetic sulphides versus higher ( 200 °C) crystallisation temperature for late diagenetic/metamorphosed sulphides. The low crystallisation temperatures for the primary sulphides are similar to temperatures recorded during bacterial mediated sulphate reduction, producing sulphur required for growth of sulphides under reducing conditions, e.g. by bacterial reduction of seawater sulphate ions.

The following are therefore critical parameters for the development of world class sediment-hosted stratiform deposits in the central African Copperbelt: (1) availability of large tonnage of metals in the hinterland (Palaeoproterozoic and Archaean basement), subjected to erosion during the early Neoproterozoic; (2) arid climate in the depositional area where the evaporation induced a natural pre-concentration of metals; (3) development of reducing conditions during the deposition of the Mines Subgroup (Congo-type) and its correlative the Musoshi Subgroup (Zambia-type) rock association. This reducing environment triggered the crystallisation of syngenetic and early diagenetic sulphides, and therefore the copper–cobalt deposits in the central African Copperbelt are typical syngenetic-early diagenetic deposits; (4) late diagenetic, metamorphic and relatively recent oxidation processes reworked the orebodies enhancing metal grades in some deposits.

Acknowledgements

This research is a contribution to the UNESCO-IUGS/IGCP-450 project and it has been supported by Forrest Group and Gécamines Mining Companies (Congo). A.B.K. acknowledges the support of the University of Botswana for his research activities. We acknowledge the constructive comments of P. Binda and an anonymous reviewer that allowed us to improve or clarify some points in the text.

References

Annels, 1974 Annels, A.E., 1974. Some aspects of the stratiform ore deposits of the Zambian Copperbelt and their genetic significance. In: Bartholomé, P. (Ed.), Gisements stratiformes et provinces cuprifères. Centenaire Société Géologique de Belgique, Liège, pp. 235–254.

Annels, 1979 A.E. Annels, The genetic relevance of recent studies at Mufulira Mine, Zambia, Annales Société Géologique de Belgique 102 (1979), pp. 431–449. View Record in Scopus | Cited By in Scopus (2)

Annels, 1989 Annels, A.E., 1989. Ore genesis in the Zambian Copperbelt with particular reference to the northern sector of the Chambishi basin. In: Boyle, R.W., Brown, A.C., Jefferson, C.W., Jowett, E.C., Kirkham, R.V. (Eds.), Sediment-Hosted Stratiform Copper Deposits. Geological Association of Canada, Special Paper, vol. 36, pp. 427–452.

Annels and Simmonds, 1984 A.E. Annels and J.R. Simmonds, Cobalt in the Zambian Copperbelt, Precambrian Research 25 (1984), pp. 75–98. Abstract | PDF (1523 K) | View Record in Scopus | Cited By in Scopus (9)

Annels et al., 1983 A.E. Annels, D.J. Vaughan and R.J. Craig, Conditions of ore minerals formation in certain Zambian deposits with special reference to the role of the cobalt, Mineralium Deposita 18 (1983), pp. 71–88. View Record in Scopus | Cited By in Scopus (4)

Armstrong et al., 1999 R.A. Armstrong, L.J. Robb, S. Master, F.J. Kruger and P.A.C.C. Mumba, New U–Pb age constraints on the Katangan sequence, central African Copperbelt, Journal of African Earth Sciences 28 (1999), pp. 6–7.

Audeoud, 1982 Audeoud, D., 1982. Les minéralisations uranifères et leur environnement à Kamoto, Kambove et Shinkolobwe (Shaba, Zaïre). Pétrographie, géochimie et inclusions fluides. Thèse de 3^ème cycle, Lyon (France), 211p.

Audeoud et al., 1984 Audeoud, D., Moine, B., Poty, B., 1984. Minéralisations uranifères et milieux confinés au Shaba. Gréco 52, Société de Géologie Appliquée, Paris.

Bartholomé, 1962 Bartholomé, P., 1962. Les minerais cupro-cobaltifères de Kamoto (Katanga-Ouest), I. Pétrographie—II. Paragenèses. Studia Universitatis “Lovanium”, Faculté des Sciences, University of Kinshasa, vol. 14, 40 p. and vol. 16, 24 p.

Bartholomáaaebbb, 1963 Bartholomé, P., 1963. Sur la zonalité dans les gisements du Copperbelt de l’Afrique Centrale. In: Symposium on Problems of Postmagmatic Ore Deposition, Prague, vol. 1, pp. 317–321.

Bartholomé, 1966 P. Bartholomé, Sur l’abondance de la dolomite et de la sépiolite dans les séries sédimentaires, Chemical Geology 1 (1966), pp. 33–48. Abstract | PDF (978 K) | View Record in Scopus | Cited By in Scopus (2)

Bartholomé, 1969 Bartholomé, P., 1969. Kamoto (Katanga) et White Pine (Michigan): deux gisements stratiformes de cuivre. Académie Royale des Sciences d’Outre-Mer, Bulletin des Séances, pp. 397–410.

Bartholomé, 1974 Bartholomé, P., 1974. On the diagenetic formation of ores in sedimentary beds, with special reference to Kamoto, Shaba, Zaïre. In: Bartholomé, P. (Ed.), Gisements stratiformes et provinces cuprifères. Centenaire de la Société Géologique de Belgique, Liège, pp. 203–213.

Bartholomé et al., 1972 P. Bartholomé, P. Evrard, F. Katekesha, J. Lopez-Ruiz and M. Ngongo, Diagenetic ore-forming processes at Kamoto, Katanga, Republic of the Congo. In: G.C. Amstutz and A.J. Bernard, Editors, Ores in Sediments, Springer-Verlag, Heidelberg (1972), pp. 21–41.

Bartholomé et al., 1971 P. Bartholomé, F. Katekesha and J. Lopez-Ruiz, Cobalt zoning in microscopic pyrite from Kamoto, Republic of the Congo, Mineralium Deposita 6 (1971), pp. 167–176. View Record in Scopus | Cited By in Scopus (4)

Berner, 1964 R.A. Berner, Iron sulfides formed from aqueous solution at low temperatures and atmospheric pressure, Journal Geology 72 (1964), pp. 293–306.

Berner, 1970 R.A. Berner, Sedimentary pyrite formation, American Journal of Science 268 (1970), pp. 1–23.

Binda, 1972 Binda, P.L., 1972. Zircons of the Nchanga Granite and overlying metasediments, Zambia. In: 24th International Geological Congress, Section 1, pp. 179–186.

Binda, 1987 P.L. Binda, Depositional environment and copper mineralization of the footwall rocks at Muliashi South, Zambian Copperbelt. In: G. Matheis and H. Schandelmeier, Editors, Current Research in African Earth Science, Balkema, Rotterdam (1987), pp. 397–400.

Binda, 1994 P.L. Binda, Stratigraphy of Zambian orebodies. In: A.B. Kampunzu and R.T. Lubala, Editors, Neoproterozoic Belts of Zambia, Zaire and Namibia, Journal of African Earth Sciences 19 (1994), pp. 251–264. Abstract | PDF (1494 K) | View Record in Scopus | Cited By in Scopus (24)

Binda, 1997 Binda, P.L., 1997. Cenozoic analogies for the Katangan copper deposits of Zambia. In: Charlet, J.-M. (Ed.), Colloque International Cornet, Gisements stratiformes de cuivre et minéralisations associées, Mons (1994). Académie Royale des Sciences d’Outre-Mer, pp. 199–215.

Binda and Mulgrew, 1974 Binda, P.L., Mulgrew, J.R., 1974. Stratigraphy of copper occurences in the Zambian Copperbelt. In: Bartholomé, P. (Ed.), Gisements stratiformes et provinces cuprifères. Centenaire de la Société Géologique de Belgique, Liège, pp. 215–233.

Binda and Van Eden, 1972 P.L. Binda and J.G. Van Eden, Sedimentological evidence on the origin of the Precambrian Great Conglomerate, (Kundelungu tillite) Zambia, Palaeogeography, Palaeoclimatology, Palaeoecology 11 (1972), pp. 151–168. Abstract | Article | PDF (2729 K) | View Record in Scopus | Cited By in Scopus (9)

Brandt et al., 1961 R.T. Brandt, C.C.J. Burton, S.C. Maree and M.E. Woakes, Mufulira. In: F. Mendelsohn, Editor, The Geology of the Northern Rhodesian Copperbelt, Macdonald, London (1961), pp. 411–464.

Brock, 1961 B.B. Brock, The structural setting of the Copperbelt. In: F. Mendelsohn, Editor, The Geology of the Northern Rhodesian Copperbelt, Macdonald, London (1961), pp. 81–89.

Brown and Chartrand, 1986 A.C. Brown and F.M. Chartrand, Diagenetic features at White Pine (Michigan), Redstone (N.W. Territories, Canada) and Kamoto (Zaire). In: G.H. Friedrich, A.D. Genkin, A.J. Naldrett, J.D. Ridge, R.H. Sillitoe and F.M. Vokes, Editors, Geology and Metallogeny of Copper Deposits, Part IV, Sediment-Hosted Deposits. 27th International Geological Congress, Moscow (1984), Special Publication 4, Society for Geology Applied to Mineral Deposits, Springer-Verlag, Heidelberg (1986), pp. 390–397.

Buffard, 1988 Buffard, R., 1988. Un rift intracontinental du Précambrien Supérieur: le Shaba méridional (Zaïre). Evolution sédimentaire et tectonique du Supergroupe de Roan au Groupe du Kundelungu inférieur (Supergroupe du Kundelungu). Thèse de Doctorat en Sciences, University of Maine, France, 316 p.

Byamungu et al., 1979 B.R. Byamungu, R. Giordano, A.B. Kampunzu and C.N. Male, A propos du Kibarien (Précambrien Moyen) de la région du barrage de N’Zilo (Shaba, Zaïre), Annales de la Faculté des Sciences, Lubumbashi 2 (1979), pp. 39–48.

Cahen, 1954 Cahen, L., 1954. Géologie du Congo Belge. Vaillant Carmanne, Liège, 577 p.

Cailteux, 1973 J. Cailteux, Minerais cuprifères et roches encaissantes à Musoshi, Prov. Shaba (Rép Zaïre), Annales de la Société Géologique de Belgique 96 (1973), pp. 495–521.

Cailteux, 1974 Cailteux, J., 1974. Les sulfures du gisement cuprifère stratiforme de Musoshi, Shaba, Zaïre. In: Bartholomé, P. (Ed.), Gisements stratiformes et provinces cuprifères. Centenaire de la Société Géologique de Belgique, Liège, pp. 267–276.

Cailteux, 1978a J. Cailteux, Particularités stratigraphiques et pétrographiques du faisceau inférieur du Groupe des Mines au centre de l’Arc cuprifère shabien, Annales de la Société Géologique de Belgique 100 (1978) (1977), pp. 55–71.

Cailteux, 1978b J. Cailteux, La succession stratigraphique du C.M.N. (ou R-2.3) au centre de la sous-province cuprifère shabienne, Annales de la Société Géologique de Belgique 100 (1978) (1977), pp. 73–85.

Cailteux, 1979 J. Cailteux, L’origine du talc dans le C.M.N. (ou R.2.3) de Kambove (Shaba-Zaïre), Annales de la Société Géologique de Belgique 102 (1979), pp. 213–221. View Record in Scopus | Cited By in Scopus (1)

Cailteux, 1983 Cailteux, J., 1983. Le Roan shabien dans la région de Kambove (Shaba-Zaïre). Etude sédimentologique et m

contain NaCl-saturated fluids with varying liquid–vapour–solid ratios suggesting heterogeneous trapping. (3) Primary and secondary inclusions from mineralised K-feldspar–quartz–biotite-anhydrite assemblage veins cross-cutting the orebodies in the Nkana synclinorium, and representative of post-tectonic fluids, contain H₂O–NaCl solutions which reflect two end-members fluids: (A) low salinity (8–14 wt.% NaCl equiv.) and high Th (300–400 °C); (B) high salinity (14–23 wt.% NaCl equiv.) and low Th (100–150 °C). The authors concluded that (1) low temperature—high salinity fluids may be characteristic of basinal—early diagenetic brines, whereas high-temperature—low salinity fluids are possibly derived from later regional metamorphic events.

10. Discussion

There are no data to support the genetic ore model involving a widespread circulation of hydrothermal fluids ascending along rift fractures and deriving metals from deep-seated mafic rocks, as proposed by Annels, 1974, Annels, 1979, Annels, 1989, Annels and Simmonds, 1984 and Lefebvre, 1989. Indeed, fracture-filling ores expected to mark mineralising fluid flow paths are unknown and there is no link between Cu–Co distribution and Upper Roan/Dipeta igneous mafic rocks (Sweeney et al., 1991a, Sweeney et al., 1991b, Kampunzu et al., 2000 and Kampunzu et al., this volume) in Congo and Zambia. No large plutonic body able to supply the amount of metals known in the Copperbelt has ever been detected by gravity and aeromagnetic surveys beneath the mineralised section of the Katangan belt (Sweeney et al., 1991a, Sweeney et al., 1991b, Sebagenzi, 1993, Sebagenzi, 1997a and Sebagenzi, 1997b; Gecamines, unpublished aeromagnetic data).

Similarly, there are no data to support the hypothesis of Unrug, 1988 and Unrug, 1989, assuming the existence of two pulses of hydrothermal ore-forming fluids that supplied metal to the Roan Subgroup “aquifers”. This author suggests a two-stage model including: (1) Co–Ni-PGM hydrothermal fluids linked to the emplacement of mafic magmas in the Copperbelt during the deposition of Nguba Group sediments; (2) convective circulation of basinal ore solutions driven by a thermal gradient and with leaching of metals from Nguba pelites, incorrectly inferred by the author to contain 50% of igneous mafic material. Several authors (e.g. Sweeney and Binda, 1989a and Sweeney and Binda, 1989b) stressed already that this model is untenable. The bulk of the ore sulphides were deposited at the same time as the deposition of the Mines Subgroup and thus predate the deposition of the Nguba Group. The lack of spatial relations between Cu–Co mineralisation and igneous mafic rocks has already been stressed above. The sulphide zoning reported above cannot be explained by this model.

Several authors (e.g. Brown and Chartrand, 1986, Haynes, 1986, Rose, 1989 and Walker, 1989) suggested that dewatering of “red beds” located stratigraphically beneath the sediment-hosted ore deposits could be the main source of copper-(cobalt) mineralizing fluids in the Copperbelt stratiform orebodies. However, available data do not support this attractive model. For example, geochemical data (Kampunzu et al., this volume) show that: (1) the transition metals content in the footwall sedimentary rocks is higher than average content in normal clastic rocks and there is no geochemical evidence to support loss of these metals in the footwall; (2) there is no evidence of widespread diagenetic copper-bearing dewatering-veins in the footwall rocks; (3) accumulation of at least 1850 million tons of copper contained in the Copperbelt requires the erosion and the deposition in the Katangan basin of 10¹⁴ m³ of source rocks, which is about 100 times the volume of footwall sedimentary rocks in the Lufilian Arc. On the other hand, occurrences of copper are known in the basement: (1) the Samba deposit granitoids (50 million tons at 0.7 wt.% Cu); (2) the Muva phyllites south of the Nkana basin (3.6-m wide zone at 3.6 wt.% Cu); (3) the Nchanga Red Granite (1.5 million tons at 1.5 wt.% Cu); >5000 ppm Cu over a large area below the Chingola-Nchanga orebodies; several pockets reported from a number of mines and prospects in the Copperbelt (Binda, 1997 and references therein; Hitzman, 2000).

The Copperbelt primary mineralisation displays very fine sulphide layering and the ore and host rock grain sizes are strongly positively correlated. The ore sulphides occur along foresets of cross-bedding and bedding planes and they are affected by pre-consolidation sedimentary structures such as sedimentary truncation and slumping (Garlick, 1961b and Garlick, 1989). The orebodies are affected by the oldest Lufilian compressional structures such as thrusts related to the Kolwezian tectonic event (Kampunzu and Cailteux, 1999). Therefore, these orebodies cannot be linked to fluids from syn-orogenic metamorphic dewatering even if they display features indicating partial reworking of primary ores during Lufilian tectonic-metamorphic events. The most important among these features are the following: (1) one set of low-salinity fluid inclusions yield equilibration temperatures 200 °C, contrasting with lower temperatures (<100 °C) obtained on high-salinity inclusions preserving primary features (Pirmolin, 1970, Audeoud, 1982, Audeoud et al., 1984, Sweeney, 1987, Richards et al., 1988, Annels, 1989 and Greyling et al., 2002); (2) local syn-kinematic hydrothermal leaching (Ngongo, 1975a, Lefebvre, 1976b, Cailteux and Kampunzu, 1995 and Cailteux, 1997) indicates small scale metamorphic remobilisation of hypersaline fluids (cf. fluid inclusion composition) inducing recrystallisation of both gangue and ore minerals and local development of barren and mineralised veins; (3) kyanite–serpentine–florencite and paragonite–phengite assemblages in veins indicate temperatures up to 400 °C in the presence of hypersaline fluids (Lefebvre and Patterson, 1982 and Cluzel, 1986). The genetic model developed below puts emphasis on the mineralising processes that were the most important in the formation of the primary orebodies.

The Mines/Musoshi orebodies are lithologically/stratigraphically bound to more or less evaporitic tidal flat/subtidal shales-carbonates in Congo and to their lateral correlatives in Zambia. All rocks hosting Cu–Co sulphide in Congo-type orebodies (Grey R.A.T.–D.Strat.–R.S.F.–S.D.B.–Kambove Formation) were deposited under reducing conditions, in a tidal flat/subtidal environment, during a major transgressive–regressive event. They overlie continental siliciclastic sedimentary rocks (Red R.A.T. in Congo-type and Mutonda in Zambia-type orebodies) deposited under an oxidized, hot, arid to semi-arid environment. Variegated R.A.T. represents the transition zone between the oxidized footwall sedimentary rocks and the main orebodies. This transitional lithology indicates the existence of a reducing/oxidizing front (Cailteux, 1978a and Cailteux, 1994). A thinner but similar transition zone occurs over tens of centimetres between the Zambian Ore Shale and its Footwall (e.g. Musoshi; Cailteux, 1973).

The earliest sulphides (pyrite-I, Co–Ni-pyrite-II, chalcopyrite-I, bornite-I) were deposited before the lithification of the host rocks, i.e. they are syngenetic to early diagenetic. This interpretation is based on the supposition that early diagenetic processes start when sedimentation is ongoing. Framboidal pyrite-I is a syngenetic mineral since such texture marks direct precipitation of FeS₂ from solutions (Berner, 1964 and Berner, 1970) or bacterial reduction of seawater sulphates (Annels, 1974 and Sweeney et al., 1987). Inclusions of copper sulphides-I in framboidal pyrite indicate that the earliest Cu-sulphides precipitated early, before some framboidal pyrite and thus represent also syngenetic minerals. The increase of Cu–Co–Ni content from the centre (pyrite-I) to the margin (pyrite-II) of grains reflects an increase of transition metal concentration in the interstitial water, possibly indicating the first steps of evaporation in the sedimentary basin, yielding metal-rich brines. High transition metal content in host rock primary minerals such as dolomite (Sweeney and Binda, 1989a, Sweeney and Binda, 1989b and Loris, 1996) indicates that the transition elements were readily available and even concentrated in the depocentre water during sediment deposition.

Pyrite group sulphides (pyrite, cattierite, vaesite) were deposited before Cu–Co sulphides-II. According to Craig and Vaughan (1979), the pyrite group forms a crystallisation sequence pyrite → Co-pyrite → cattierite + thiospinel or vaesite (e.g. at Luiswishi), marking Fe–Co–Ni compositional changes of the ore-forming fluids.

The second generation of sulphides (Cu–Co sulphides-II) replaces or includes sulphides of the first generation (i.e. bornite replacing pyrite, chalcopyrite including framboidal pyrite). Intergrowth of Cu–Co sulphides-II with diagenetic minerals (e.g. chlorite, leucoxene-rutile) indicates that they grew during diagenesis. Diagenetic conversion of detrital ilmenite into leucoxene and rutile released at least part of the iron required for sulphides-II growth in reducing conditions that prevailed in rocks hosting most orebodies.

Replacement textures within the external rim of copper sulphides-II are also related to diagenetic reactions. An increase of cobalt concentration in the brine and its reaction with copper sulphide grains led to the growth of carrollite at their rims. The reactions involved are as follows (Cailteux, 1983 and Cailteux, 1986):

The copper released during these reactions enhanced, at the contact with carrollite, the conversion of chalcopyrite and bornite into digenite. Small pyrite crystals within the carrollite fringe could be linked to the iron released during the above reactions. Excess iron was probably released to the interstitial fluid. Positive cobalt anomalies in the hangingwall of the Congo-type orebodies indicate that the brines were still enriched in cobalt after the deposition of Co-sulphides in the orebodies.

Sulphur isotopic data on nodular and lenticular anhydrite from the orebodies and barren rocks from the same unit indicate that anhydrites formed by evaporation of seawater under supratidal or sabkha conditions, requiring a major sulphate super-saturation of the mineralising brines. The sulphide isotopic data show that the source of sulphur was seawater sulphate ions. They also support growth of sulphides at low temperature (less than 100 °C) by bacterial reduction of sulphate ions in the brines and possibly by reaction with earlier anhydrite (Annels, 1974, Sweeney and Binda, 1989a and Sweeney and Binda, 1989b). This process liberated the sulphur necessary for sulphide growth. Isotopic data also identify the occurrence of two generations of sulphides (Hoy and Ohmoto, 1989): the first (50–75 vol.% sulphides) is syngenetic to early diagenetic whereas the second (25–50 vol.% sulphides) is attributed to super-saturated copper-bearing fluids generated at the depositional site during synkinematic metamorphic processes (Cailteux and Kampunzu, 1995 and Kampunzu and Cailteux, 1999). Fluid inclusion data reviewed above (e.g. Greyling et al., 2002) coupled with field and petrologic observations show that: (1) the first group of sulphides grew probably at less than 100 °C, and this is compatible with sulphide crystallisation at low temperature by bacterial reduction of sulphate ions in brines (Annels, 1974, Sweeney and Binda, 1989a and Sweeney and Binda, 1989b); (2) the second group of sulphides grew during the Lufilian orogenesis from metamorphic fluids reworking the syngenetic to diagenetic mineralization (Cailteux and Kampunzu, 1995

sulphur.

 

 

Full-size image (28K)

 

Fig. 15. Sulphur isotope values for sulphides and sulphates from Congo-type deposits (Kamoto, Kambove-Ouest, Luiswishi, Etoile, Ruashi; data from Okitaudji, 1989 and Lerouge et al., 2004), sulphides from the Zambian Ore Shale, Footwall and veins at Konkola, and sulphates from Mufulira ore horizons (data from Sweeney et al., 1986). CP = chalcopyrite; CR = carrollite.

View Within Article

 

 

A detailed investigation of the relations between δ³⁴S of individual sulphides and the lithostratigraphic position at Konkola in Zambia (Fig. 16a; Sweeney et al., 1986) and Luiswishi in Congo (Fig. 16b; Lerouge et al., 2004) indicates a strong stratigraphic control of δ³⁴S in sulphides. These relations suggest an introduction of sulphur to the sediments during sedimentation and early diagenesis. According to Ohmoto and Rye (1979), variations of δ³⁴S values may be interpreted in terms of transgression–regression events, i.e. δ³⁴S values become lighter (high negative values down to −70‰) during transgressive events, whereas they become heavier (high positive values up to +70‰) during regressive events. Consequently, both in Congo and Zambia, the high δ³⁴S sulphide values from the base of the orebodies were probably produced by δ³⁴S-depletion in a main reservoir during a regressive period, marking a basin closed from the seawater. The decrease of δ³⁴S values, down to −15‰ in the Ore Shale in Zambia, suggests that the system was progressively open to a SO₄-rich source, marking a transgressive event during the deposition of the lower part of the Ore Shale (Units A and B), followed by a regressive regime during the deposition of its upper part (Units C–E). In Congo the transgressive regime was persistent during the deposition of the sediments hosting the Lower and Upper orebodies (Kamoto and Dolomitic Shales Formations). These results are in agreement with the inferred lithological transgressive–regressive evolution of the ore-hosting sediments from the Mines Subgroup in Congo (Bartholomé et al., 1972, Cailteux, 1983 and Cailteux, 1994) and the Ore Shale Formation in Zambia (Sweeney et al., 1986).

 

 

Full-size image (25K)

 

Fig. 16. Sulphur isotope composition of sulphides along vertical sections through (a) the Zambian Ore Shale (Sweeney and Binda, 1989a and Sweeney and Binda, 1989b), (b) the Congo-type orebodies in the Luiswishi deposit (Lerouge et al., 2004); the δ³⁴S sulphide values show the same transgressive–regressive events as those indicated by the host-rock lithologies (Sweeney and Binda, 1989a, Sweeney and Binda, 1989b and Cailteux, 1994).

View Within Article

 

 

The sulphur isotopic data of coexisting sulphide pairs at Konkola (Sweeney et al., 1986) and Luiswishi (Lerouge et al., 2004) show heterogeneous isotopic fractionations between the sulphides, indicating crystallization at disequilibrium. In the Luiswishi deposit, chalcopyrite–carrollite isotopic fractionations are systematically reversed when organic matter is abundant. This, along with the large range of δ³⁴S values, confirms that sulphide δ³⁴S values are controlled by the sedimentary environment, but also by the presence of organic matter, and probably by complex kinetic processes.

The range of δ³⁴S values of sulphides in veins is very close to δ³⁴S values of early associated stratiform sulphides (chalcopyrite, carollite, bornite) indicating a good preservation of the primary sulphur isotopic composition. This strongly suggests a local scale reworking of the early stratiform sulphides, i.e. local recrystallization and neocrystallization of sulphides in veins during the Lufilian tectono-metamorphic events. The preservation of disequilibrium between coexisting sulphides in veins suggests that the reworking was not strong enough to re-equilibrate the ³⁴S/³⁴S ratios.

8. Thermometry

François, 1973 and François, 1974 indicated that the total thickness of Katangan sedimentary rocks above the stratiform orebodies was 6 km. Assuming a geothermal gradient of 10 °C/km which is a minimum value deduced from P, T estimates in high-pressure metamorphic rocks in the Katangan/Zambezi belt (Massone et al., 1994 and John et al., 2003) and 50 °C/km which is a maximum value calculated by Cluzel (1986) in rocks from Luishia, the temperature at the 6-km burial depth should be between 60 and 300 °C. The regional metamorphism related to the Lufilian orogeny increases from the north to the south, evolving from zeolite and greenschist facies in Congo to the north up to amphibolite facies in Zambia to the south (e.g. Mendelsohn, 1961b, Oosterbosch, 1962, Drysdall et al., 1972, François and Cailteux, 1981 and Tembo, 1994). Most of the Zambian-type deposits in Zambia (e.g. Nchanga, Chambishi, Muliashi, Mufulira, Nkana, Luanshya) and some deposits in Congo (e.g. Musoshi, Kinsenda) evolved under the greenschist facies, possibly around 350–400 °C (Richards et al., 1988). Particularly at Musoshi, late quartz + biotite + microcline + carbonates + sulphides ± anhydrite ± barite related to the Lufilian orogeny recorded temperatures around 400 °C (fluid inclusion data in quartz, Richards et al., 1988). In northwestern Zambia (Domes area), peak metamorphic conditions at F1 (D1 Kolwezian phase of Kampunzu and Cailteux, 1999) are around 650 °C and 13 kbars, followed by a decompression at 6 kbars (Cosi et al., 1992 and Steven and Armstrong, 2003).

9. Fluid inclusions

Fluid inclusions found in gangue minerals (dolomite, magnesite and quartz) from Congo-type Kamoto-Principal, Shinkolobwe, Kambove-Ouest and Luiswishi orebodies (Pirmolin, 1970 and Ngongo, 1975b) are of two types: (1) two phase- (liquid–gas) inclusions of small size in dolomite or quartz, CO₂-free, with yield temperatures around 70 °C and salinities in the range 7–10 wt.%; (2) three phase-(solid–liquid–gas) inclusions in dolomite grains from R.S.C., hosting one to three different types of solid phases (NaCl, KCl and CaSO₄) and containing CO₂, with yield temperatures of 200 °C and salinities estimated around 40 wt.%.

Audeoud, 1982 and Audeoud et al., 1984, working on fluid inclusions contained in dolomitic veinlets of the R.A.T. formations and on identical inclusions associated with the anchimetamorphic recrystallization, showed that the aqueous phase is the most important one, and confirmed the very high salinity of this phase, i.e. containing more than 60 wt.% of dissolved salts (MgCl₂, CaCl₂). The NaCl and KCl concentrations in these inclusions are relatively low, which is consistent with the R.A.T. composition (Kampunzu et al., this volume).

In Zambia, Sweeney (1987) found that: (1) fluid inclusions in quartz veins cross-cutting the Ore Shale at Konkola show post-trapping alteration features; (2) inclusions of the same vein system are characterized by varying fluid chemistry; (3) hydrocarbon liquid inclusions are present in several samples; (4) fluid chemistry variation corresponds to diagenetic changes in the different lithologies during the basin evolution. The author concluded that the veins represent a post-formational tectono-thermal event, and formed by lateral secretion of fluids during late diagenetic dewatering at temperatures of 120 °C.

Fluid inclusions in quartz from quartz-hematite veins cutting the Footwall at Musoshi (Richards et al., 1988) contain halite-saturated fluid, with a minimum salinity of 28–39 wt.% NaCl and 15–17 wt.% KCl, minor amounts of CO₂, and also contain Fe, Ca, Mn. They yielded temperatures of 275–397 °C. These authors concluded that the hydrothermal event post-dates the stratiform copper deposition and may have been linked to compressional deformation and metamorphism during the Lufilian orogeny.

Fluid inclusions from early quartz associated with pyrite and carrollite (pseudomorph after anhydrite) in the Chambishi orebody, yielded salinities in the range of 9–16 wt.%, whereas fluid inclusions from syn-kinematic veins yielded higher salinities between 16 and 22 wt.% (Annels, 1989).

Recent studies by Greyling et al. (2002) on various tectonic settings (pre-deformational, syn-tectonic, post-deformational) showed the following scenario. (1) There are primary and secondary inclusions of H₂O–NaCl–CO₂ ± CH₄ compositions with a salinity of 23 wt.% NaCl equivalent in mineralised and non-mineralised quartz veins, formed prior to deformation and folding from deposits in Zambia (e.g. Chambishi). (2) Representative of syn-tectonic fluids, primary inclusions in quartz veins in the Nchanga open pit contain NaCl-saturated fluids with varying liquid–vapour–solid ratios suggesting heterogeneous trapping. (3) Primar

dolomite, quartz, pyrite in barren rocks.

The relationship between sulphides and leucoxene-rutile were documented in both Congo-type and Zambia-type deposits. In the Musoshi (Zambia-type) deposit (Cailteux, 1973 and Cailteux and Dimanche, 1973), the barren rocks below and above the orebodies are marked by detrital ilmenite (Ilm) showing partial to complete conversion into leucoxene (Lx), with intermediate products (Ilm+Lx or Rt-Lx). These minerals are associated with diagenetic hematite. In the Musoshi orebodies, the conversion of ilmenite into Lx-Rt is complete, and 65% of the leucoxene-rutile grains show associations and intergrowths with pyrite-I, -II and/or copper sulphides-II, whereas 25% of the leucoxene-rutile and 10% of the sulphides occur in isolated grains (Cailteux and Dimanche, 1973). A similar diagenetic mineral association occurs in clastic rocks from the Congo-type orebodies (S.D.S. and S.D.B./upper orebody; Fig. 13), e.g. at Kipapila-Kimpe (Cailteux and Lefebvre, 1975), Etoile (Lefebvre and Cailteux, 1975) and Kambove-Ouest (Cailteux, 1978a and Cailteux, 1983). In the S.D.S., this association coexists with framboidal pyrite-I.

 

 

Full-size image (47K)

 

Fig. 13. Relict Leucoxene (Ti–Fe oxide, LX) associated with pyrite (Py), chalcopyrite (CP), bornite (BN) assemblage; note the diagenetic destabilisation of detrital ilmenite releasing iron presumably fixed into copper sulphides. Upper orebody (S.D.B.), sample No. 822, D.H. Kwf-1148, Kambove-Ouest deposit (Cailteux, 1983).

View Within Article

 

 

In the Kambove Formation, carrollite is the major sulphide in talc-dolomite bearing beds and layers (Fig. 4 and Fig. 5), whereas dolomite–quartz bearing beds host mainly copper sulphides (Cailteux, 1983). Talc in these rocks is diagenetic and presumably results from an early sepiolite that coexisted with calcite (converted into diagenetic dolomite) instead of growing from the reaction between dolomite and quartz (Bartholomé, 1966 and Cailteux, 1979). However, a metamorphic origin for this talc remains possible.

6.3. Relations between orebodies and deformation events

In Congo, the orebodies were tectonically dismembered, forming part of thrust sheets (e.g. Derriks and Vaes, 1956, Derriks and Oosterbosch, 1958, Mendelsohn, 1961b and Demesmaeker et al., 1963), related to the first Lufilian compressional deformation event known as the Kolwezian tectonic event (Kampunzu and Cailteux, 1999 and references therein). The heterogenous distribution of strain during this event with the maximum strain focused along thrust sense shear zones, possibly controlled by evaporitic layers (Cailteux, 1994 and Cailteux and Kampunzu, 1995), explains the absence of strong fabric within the rocks and thus the good preservation of sedimentary/diagenetic textures. In Zambia, a stronger fabric occurs in some deposits and is registered in both sulphides and gangue minerals (e.g. Nkana, Luanshya, Mufulira; Mendelsohn, 1961b and Brandt et al., 1961).

Metamorphism and/or hydrothermal alteration (fluids escape during compressional tectonics) generated variable re-equilibration, remobilization and secretion of sulphides into late- to post-kinematic veins both in Zambia and Congo (e.g. Garlick, 1961b, Garlick, 1964, Mendelsohn, 1961c, Cailteux, 1983, Cailteux, 1997, Sweeney, 1987, Cailteux and Kampunzu, 1995 and Loris, 1996). In the orebodies, minor remobilisation of stratiform ores is shown by a few cross-cutting mineralised veins surrounded by centimetre-wide zones within which stratiform sulphides have been depleted. A few centimetres away from these veins, the fine primary compositional zoning of the disseminated stratiform sulphides is well preserved. In the Musoshi deposit, the copper content in the “depleted” orebody rocks affected by fractures related to the Lufilian orogeny is 0.03 wt.% Cu, whereas the adjacent undepleted orebody contains more than 3.0 wt.% Cu (Lefebvre and Tshiauka, 1986 and Richards et al., 1988).

7. Isotopic geochemistry

The δ¹⁸O and δ¹³C values for Footwall and Ore Shale dolomites in Zambia define two fields in Fig. 14 (Sweeney et al., 1986). δ¹⁸O values for Footwall dolomites are between +20.82 and +26.38‰ SMOW. The Konkola Ore Shale carbonate pseudomorph after anhydrite nodules or lenticles yielded δ¹⁸O values between +14.56 and +16.16‰ SMOW. For comparison, δ¹⁸O present-day mine waters from the Copperbelt yielded values of −6.2‰ SMOW (Sweeney et al., 1986).

 

 

Full-size image (13K)

 

Fig. 14. Values of δ¹⁸O plotted against δ¹³C for carbonates from Zambian Footwall and Ore Shale rocks (Sweeney et al., 1986).

View Within Article

 

 

Sulphur isotopic data on sulphides from Cu–Co mineralization in Congo and Zambia are dispersed (Fig. 15). However they show a consistent large range of δ³⁴S values (Jensen and Dechow, 1962, Dechow and Jensen, 1965, Sweeney et al., 1986, Okitaudji, 1989, Hoy and Ohmoto, 1989, McGowan et al., 2003 and Lerouge et al., 2004), from high negative to high positive values, which characterize the sediment-hosted deposits (e.g. Ohmoto and Rye, 1979, Krouse, 1980, Misi et al., 2000 and McGowan et al., 2003 and others). The range of δ³⁴S values (Fig. 15) are in agreement with values of sulphides resulting from a bacterial reduction of marine sulphates at superficial temperatures (<50 °C). However, for deposition in Congo, Hoy and Ohmoto (1989) suggested that the high positive δ³⁴S values originated from an input of hydrothermal sulphur characterized by a δ³⁴S   +9‰. The same hypothesis is also proposed for the Meso- and Neoproterozoic lead–zinc deposits of the São Francisco Craton (Misi et al., 2000). Rare δ³⁴S sulphate analyses from host rocks are +17‰ at Mufulira in Zambia (Sweeney et al., 1986) and +22.6‰ in the Mines Subgroup (average) at Kolwezi (Okitaudji, 1989). These values are quite close to the reference value of Neoproterozoic seawater (Claypool et al., 1980), and tend to confirm a largely marine origin for the sulphur.

 

 

Full-size image (28K)

 

Pyrite, chalcopyrite and bornite grains (sulphides-III) include diagenetic gangue minerals (chlorite, dolomite, quartz) and disseminated copper sulphides-II, indicating late-stage formation of the sulphide grains. Pyrite-III rims copper sulphides-II.

Bornite (-II) grains include digenite in bornite-dominant beds (Cailteux, 1986). In these beds, carrollite grains include bornite-II in the centre and digenite-II towards the rim. This indicates that carrollite grew before and after the conversion of bornite into digenite.

Replacement carrollite forms the external rim of chalcopyrite and/or bornite (-II or -III) grains (Fig. 12). The transition between carrollite rims and chalcopyrite or bornite cores is marked by a digenite fringe (Fig. 12a, b), and small pyrite grains occur within the carrollite rims (Fig. 12c). Pyrite (III, IV), chalcopyrite and bornite (-IV) overgrow these parageneses (Fig. 12d). Some carrollite grains include copper sulphides showing replacement textures by carrollite; others show microfractures filled by chalcopyrite or bornite (-IV). In Zambia-type deposits (e.g. Musoshi), bornite frequently includes chalcopyrite both as lattice or irregular exsolutions (Cailteux, 1973 and Cailteux, 1974). Similar parageneses occur in the Shituru lower orebody, forming also several generations of sulphides (Lefebvre, 1974).

 

 

Full-size image (104K)

 

Fig. 12. (a) Incomplete transformation of a chalcopyrite grain into carrollite and digenite; relict chalcopyrite (CP) occurs in the centre of the grain; digenite (DG) occurs between chalcopyrite and carrollite (CR); early pyrite (Py-1) grains are included in copper sulphides; (b) Detail of the same paragenesis; (c) Occurrence of pyrite grains (Py-2) within carrollite. Grey R.A.T., sample No. 1291, D.H. Kwf-1150, Kambove-Ouest deposit (Cailteux, 1983); (d) Late pyrite (Py-3) partly surrounded by late bornite (BN). S.D.B., sample No. 1511, D.H. Kwf-1138, Kambove-Ouest deposit (Cailteux, 1983).

View Within Article

 

 

6.2. Relations between sulphides and gangue minerals

Abundant nodules or beds of anhydrite occur in the Zambia-type Ore Shale (e.g. Nkana, Mufulira) and there is evidence for replacement of evaporitic minerals by calcite–dolomite, quartz, bornite and chalcopyrite (Annels, 1974). This author showed that, in the Mufulira deposit, mineralised zones correspond to areas in the orebodies with little or no anhydrite content, whereas high anhydrite contents mark barren or sparsely mineralised zones. Pseudomorphs after anhydrite nodules occur in S.D.B. and at the base of the Kambove Formation within the Congo-type deposits (Bartholomé et al., 1972, Annels, 1974, Katekesha, 1975, Cailteux, 1978a and Cailteux, 1978b). These nodules were completely replaced by dolomite, quartz, pyrite and copper–cobalt sulphides in the orebodies and, within the same stratigraphic units, by dolomite, quartz, pyrite in barren rocks.

The relationship between sulphides and leucoxene-rutile were documented in both Congo-type and Zambia-type deposits. In the Musoshi (Zambia-type) deposit (Cailteux, 1973 and Cailteux and Dimanche, 1973), the barren rocks below and above the orebodies are marked by detrital ilmenite (Ilm) showing partial to complete conversion into leucoxene (Lx), with intermediate products (Ilm+Lx or Rt-Lx). These minerals are associated with diagenetic hematite. In the Musoshi orebodies, the conversion of ilmenite into Lx-Rt is complete, and 65% of the leucoxene-rutile grains show associations and intergrowths with pyrite-I, -II and/or copper sulphides-II, whereas 25% of the leucoxene-rutile and 10% of the sulphides occur in isolated grains (Cailteux and Dimanche, 1973). A similar diagenetic mineral association occurs in clastic rocks from the Congo-type orebodies (S.D.S. and S.D.B./upper orebody; Fig. 13), e.g. at Kipapila-Kimpe (Cailteux and Lefebvre, 1975), Etoile (Lefebvre and Cailteux, 1975) and Kambove-Ouest (Cailteux, 1978a and Cailteux, 1983). In the S.D.S., this association coexists with framboidal pyrite-I.

 

 

Full-size image (47K)

 

Fig. 13. Relict Leucoxene (Ti–Fe oxide, LX) associated with pyrite (Py), chalcopyrite (CP), bornite (BN) assemblage; note the diagenetic destabilisation of detrital ilmenite releasing iron presumably fixed into copper sulphides. Upper orebody (S.D.B.), sample No. 822, D.H. Kwf-1148, Kambove-Ouest deposit (Cailteux, 1983).

View Within Article

 

 

In the Kambove Formation, carrollite is the major sulphide in talc-dolomite bearing beds and layers (Fig. 4 and Fig. 5), whereas dolomite–quartz bearing beds host mainly copper sulphides (Cailteux, 1983). Talc in these rocks is diagenetic and presumably results from an early sepiolite that coexisted with calcite (converted into diagenetic dolomite) instead of growing from the reaction between dolomite and quartz (Bartholomé, 1966 and Cailteux, 1979). However, a metamorphic origin for this talc remains possible.

6.3. Relations between orebodies and deformation events

In Congo, the orebodies were tectonically dismembered, forming part of thrust sheets (e.g. Derriks and Vaes, 1956, Derriks and Oosterbosch, 1958, Mendelsohn, 1961b and Demesmaeker et al., 1963), related to the first Lufilian compressional deformation event known as the Kolwezian tectonic event (Kampunzu and Cailteux, 1999 and references therein). The heterogenous distribution of strain during this event with the maximum strain focused along thrust sense shear zones, possibly controlled by evaporitic layers (Cailteux, 1994 and Cailteux and Kampunzu, 1995), explains the absence of strong fabric within the rocks and thus the good preservation of sedimentary/diagenetic textures. In Zambia, a stronger fabric occurs in some deposits and is registered in both sulphides and gangue minerals (e.g. Nkana, Luanshya, Mufulira; Mendelsohn, 1961b and Brandt et al., 1961).

Metamorphism and/or hydrothermal alteration (fluids escape during compressional tectonics) generated variable re-equilibration, remobilization and secretion of sulphides into late- to post-kinematic veins both in Zambia and Congo (e.g. Garlick, 1961b, Garlick, 1964, Mendelsohn, 1961c, Cailteux, 1983, Cailteux, 1997, Sweeney, 1987, Cailteux and Kampunzu, 1995 and Loris, 1996). In the orebodies, minor remobilisation of stratiform ores is shown by a few cross-cutting mineralised veins surrounded by centimetre-wide zones within which stratiform sulphides have been depleted. A few centimetres away from these veins, the fine primary compositional zoning of the disseminated stratiform sulphides is well preserved. In the Musoshi deposit, the copper content in the “depleted” orebody rocks affected by fractures related to the Lufilian orogeny is 0.03 wt.% Cu, whereas the adjacent undepleted orebody contains more than 3.0 wt.% Cu (Lefebvre and Tshiauka, 1986 and Richards et al., 1988).

7. Isotopic geochemistry

The δ¹⁸O and δ¹³C values for Footwall and Ore Shale dolomites in Zambia define two fields in Fig. 14 (Sweeney et al., 1986). δ¹⁸O values for Footwall dolomites are between +20.82 and +26.38‰ SMOW. The Konkola Ore Shale carbonate pseudomorph after anhydrite nodules or lenticles yielded δ¹⁸O values between +14.56 and +16.16‰ SMOW. For comparison, δ¹⁸O present-day mine waters from the Copperbelt yielded values of −6.2‰ SMOW (Sweeney et al., 1986).

 

 

Full-size image (13K)

 

Fig. 14. Values of δ¹⁸O plotted against δ¹³C for carbonates from Zambian Footwall and Ore Shale rocks (Sweeney et al., 1986).

View Within Article

 

 

Sulphur isotopic data on sulphides from Cu–Co mineralization in Congo and Zambia are dispersed (Fig. 15). However they show a consistent large range of δ³⁴S values (Jensen and Dechow, 1962, Dechow and Jensen, 1965, Sweeney et al., 1986, Okitaudji, 1989, Hoy and Ohmoto, 1989, McGowan et al., 2003 and Lerouge et al., 2004), from high negative to high positive values, which characterize the sediment-hosted deposits (e.g. Ohmoto and Rye, 1979, Krouse, 1980, Misi et al., 2000 and McGowan et al., 2003 and others). The range of δ³⁴S values (Fig. 15) are in agreement with values of sulphides resulting from a bacterial reduction of marine sulphates at superficial temperatures (<50 °C). However, for deposition in Congo, Hoy and Ohmoto (1989) suggested that the high positive δ³⁴S values originated from an input of hydrothermal sulphur characterized by a δ³⁴S   +9‰. The same hypothesis is also proposed for the Meso- and Neoproterozoic lead–zinc deposits of the São Francisco Craton (Misi et al., 2000). Rare δ³⁴S sulphate analyses from host rocks are +17‰ at Mufulira in Zambia (Sweeney et al., 1986) and +22.6‰ in the Mines Subgroup (average) at Kolwezi (Okitaudji, 1989). These values are quite close to the reference value of Neoproterozoic seawater (Claypool et al., 1980), and tend to confirm a largely marine origin for the sulphur.

 

 

Full-size image (28K)

 

Etoile and Kambove-Ouest deposits (Congo) display a different sulphide zonation in the same lithostratigraphic units. At Etoile, two sequences from chalcocite to bornite and chalcopyrite were observed, one from Grey R.A.T. to D.Strat., the other from R.S.F. to S.D.B. (Lefebvre and Cailteux, 1975). At Kambove-Ouest (Cailteux, 1983 and Cailteux, 1986), chalcopyrite is dominant at the base (Grey R.A.T.) and at the top of S.D.B, whereas a chalcocite–digenite–bornite zone occurs in the middle (D.Strat., R.S.F., base of S.D.B.). This illustrates the variability of the sulphide zonation within the same sedimentary units, implying that this zonation is not controlled by the lithological characteristics of the host rock.

The third “orebody” (Kambove Formation) is hosted in dolomitic shales and dolomitic laminites at Kambove-Ouest, with a remarkable decimetric to metric vertical zoning marked by the recurrence of pyrite–chalcopyrite–bornite–chalcopyrite–pyrite ores (Fig. 8). Carrollite (cobalt-copper sulphide) is irregularly distributed in the orebodies. It is associated with chalcopyrite and pyrite in copper-poor zones (e.g. Kambove-Ouest, Chibuluma, Nkana). In the Kambove-Ouest deposit, a 0.5–1.5-m thick carrollite–chalcopyrite orebody frequently occurs in the pyritic Grey R.A.T. below the base of the lower orebody. Centimetric grains of carrollite occur in the poorly mineralised R.S.C. and define orebodies in talcose dolomites of the Kambove Formation (Fig. 4 and Fig. 5).

 

 

Full-size image (35K)

 

Fig. 8. Decimetric sulphide zoning in the Kambove Formation (third orebody) at Kambove-Ouest (Cailteux, 1986). Minor carrollite is associated with the copper sulphides. Py = pyrite; CP = chalcopyrite; BN = bornite; DG = digenite.

View Within Article

 

 

The lower orebody sulphide mineralisation in the Lower Mwashya Shituru deposit includes disseminated pyrite, chalcopyrite, bornite, minor carrollite/linnaeite and supergene copper sulphide (digenite, chalcocite) associations, similar to those of the Mines Subgroup orebodies (Lefebvre, 1974). However, there is no zoning in these sulphides. The upper orebody contains finely disseminated pyrite in the unweathered zone.

In the Congo-type Mines Subgroup deposits, the highest Co:Cu ratio occurs in the upper part of the orebodies, and shows local positive anomalies in the hangingwall (Fig. 9a and b). In the Zambia-type sequence, linnaeite and carrollite occur in the Ore Shale, mainly within its lower part (e.g. Nchanga, Chambishi, Mindola, Nkana, Chibuluma, Muliashi, Baluba; Mendelsohn, 1961c, Annels et al., 1983 and Annels and Simmonds, 1984; Fig. 9d).

 

 

Full-size image (89K)

 

Fig. 9. Cu and Co distribution diagrams in the orebodies and hangingwall of representative Copperbelt ore deposits (data from Oosterbosch, 1962 for Kamoto and Shinkolobwe and from Annels and Simmonds, 1984 for Chambishi Southeast).

View Within Article

 

 

Cobalt or cobalt–nickel disseminated sulphides in copper-rich stratiform ores from both Congo- and Zambia-type deposits include mostly cattierite and cobaltite at Kinsenda and Luiswishi (Dejonghe, 1995, Ngoyi and Dejonghe, 1997 and Loris et al., 1997), siegenite at Kambove-Ouest and Luiswishi (Cailteux, 1997 and Loris et al., 1997), Co-pentlandite at Chambishi (Annels et al., 1983 and Annels and Simmonds, 1984). Other deposits contain nickel/cobalt-rich and copper-poor stratiform orebodies, i.e. Ni-cattierite, Co-vaesite and siegenite in Shinkolobwe and Swambo (Derriks and Vaes, 1956, Derriks and Oosterbosch, 1958 and Oosterbosch, 1962), siegenite and violarite in Kalumbila (Kabompo Dome; Steven and Armstrong, 2003).

Co, Ni, Cu diagrams show variable inter-element ratios in the orebodies and the hangingwall (Fig. 9a–d). Microprobe analyses showed that nickel content is <1 wt.% in sulphides, i.e. in carrollite and cattierite, chalcopyrite, chalcocite and Co-pyrite from Kamoto, Kambove-Ouest, Shinkolobwe, Musoshi-Konkola, Baluba, Chibuluma, Chambishi (Bartholomé et al., 1971, Cailteux, 1974, Cailteux, 1997, Craig and Vaughan, 1979, Annels et al., 1983 and Sweeney et al., 1986). Sulphides from some deposits (e.g. Kamoya Sud-II in the Kambove area) contain several hundred, up to 1000 ppm Ni in the oxide ores, but Ni:Co ratio remain very low ( 0.01). Microprobe analyses identified a complete solid solution between pyrite (FeS₂), cattierite (CoS₂) and vaesite (NiS₂) (Loris, 1996 and Loris et al., 2002). Sulphides of the linnaeite group were also documented, with chemical compositions indicating siegenite, linnaeite, carrollite or polydymite. In the richest cobalt deposits (e.g. Shinkolobwe, Swambo), Ni:Co ratios is higher (e.g. 1:4) since nickel concentration reaches 0.5 wt.% in a disseminated ore containing Ni-cattierite, thiospinel, siegenite, vaesite in addition to pyrite and chalcopyrite (Derriks and Vaes, 1956 and Derriks and Oosterbosch, 1958). Linnaeite (Co₃S₄)-carrollite (Co₂CuS₄) thiospinel solid solutions contain significant amounts of Ni and Fe. Zambian thiospinels yielded 0.45–2.53 wt.% Ni (Annels et al., 1983), whereas higher values (11.4–12.2 wt.% Ni) were reported at Shinkolobwe (Craig and Vaughan, 1979). These data indicate that nickel is closely linked to primary Cu–Co ore in the central African Copperbelt.

Since 1903, gold, platinum–palladium and silver were mined from Congo-type Cu–Co stratiform deposits, i.e. Mutoshi (formerly Ruwe), Musonoi (Kolwezi area) and Shinkolobwe; their concentration in ores from the oxidized zone reached 26 g/t Au, 36 g/t Pd, 10 g/t Pt (du Trieu de Terdonck, 1956 and Jedwab, 1997). These precious metals occur in the lower orebody (mainly in R.S.F.) and form pure nuggets and alloys (involving Cu and/or Se) or are included in oxides (e.g. palladinite, Fe-Co–Ni–Cu–Mn hydroxides), As/Se/Te metallic compounds (e.g. oosterboschite, moncheite, trogtalite) or sulphides (e.g. linnaeites, Se–As sulphides). Sub-economic occurrences were recorded in oxide ores from a few Cu–Co deposits, e.g. gold at Kalongwe, Fungurume, Kamoya (Kambove area), Likasi (du Trieu de Terdonck, 1956), Pd–Au at Mindigi (Jedwab, 1997). The grades in these occurrences may reach 0.3 g/t Au, 2.9 g/t Ag and 0.3–0.6 g/t Pt-Pd. Platinum and palladium are generally hosted in heterogenites. In Zambia, Au also occurs in Cu–Co ores since it is recovered from electrolytic Cu refining.

6. Ore petrology

6.1. Sulphide parageneses

Parageneses of the stratiform disseminated sulphides are well documented both in Congo- and Zambia-types deposits, e.g. at Kamoto-Principal (Bartholomé, 1962, Bartholomáaaebbb, 1963, Bartholomé et al., 1971, Bartholomé et al., 1972 and Dimanche, 1974), Musoshi (Cailteux, 1973 and Cailteux, 1974), Kambove-Ouest (Cailteux, 1983 and Cailteux, 1986), Luiswishi (Loris, 1996; Loris et al., 2002) and Kinsenda (Ngoyi and Dejonghe, 1997).

Framboidal-pyrite (pyrite-I) grains (Fig. 10a) occur mainly within zones adjacent to the orebodies (below, above, and laterally). Sometimes, chalcopyrite (-I) or bornite (-I) form the core of this early pyrite (Cailteux, 1974). Microprobe analyses indicated the presence of copper in framboidal pyrite-I and revealed cobalt–nickel rich pyrite (-II) forming the outer rims of pyrite-I grains, e.g. at Kamoto (Bartholomé et al., 1971), Musoshi (Cailteux, 1974) and Kinsenda (Ngoyi and Dejonghe, 1997). Parageneses with pyrite-I—(Co,Ni) pyrite-II (bravoite)—pyrite-III concentric zones were reported in the Luiswishi deposit (Loris, 1996 and Loris et al., 2002). Framboidal and small isolated grains of pyrite (I, II, III) are included in diagenetic quartz and dolomite (Fig. 10b), e.g. at Kamoto (Bartholomé et al., 1971) and Kambove-Ouest (Cailteux, 1983).

 

 

Full-size image (55K)

 

Fig. 10. (a) Framboidal pyrite (Py) associated with later bornite (BN). Bornite includes framboidal pyrite. Upper Orebody (S.D.B.), sample No. 822, D.H. Kwf-1148, Kambove-Ouest deposit (Cailteux, 1983). (b) Framboidal pyrite (Py) included in dolomite (D). Grey R.A.T., sample No. 1874, D.H. Kw-236B, Kambove-Ouest deposit (Cailteux, 1983).

View Within Article

 

 

Primary chalcopyrite (-II) and bornite (-II) are the main copper sulphides in the orebodies (e.g. Kambove-Ouest); they grew at the same time as separate or coalescent grains. Carrollite and pyrite-III coexist with copper sulphides-II (mainly chalcopyrite). Chalcopyrite-II includes framboidal pyrite-I, e.g. at Kinsenda (Ngoyi and Dejonghe, 1997). Bornite-II replaces pyrite-I (and -II), e.g. at Kamoto (Bartholomé et al., 1972), as shown by carrollite grains including well-preserved aligned pyrite-I (-II), whereas pyrite grains outside carrollite have been completely replaced by bornite-II (Fig. 11). The textural relations indicate that bornite-II grew after the development of carrollite grains. In the Luiswishi deposit (Loris, 1996 and Loris et al., 2002) copper sulphides-II and sulphides of the linnaeite group (linnaeite–siegenite–carrollite/polydymite) formed after those of the pyrite group (pyrite–cattierite–vaesite).

 

 

Full-size image (19K)

 

Fig. 11. Grain of carrollite in dolomitic shale of the Upper Orebody (Kamoto-Principal), including a lamina with well-preserved pyrite-I (-II), and showing that this pyrite is replaced by bornite outside carrollite; the sketch also indicates that bornite grew after carrollite (Bartholomé et al., 1972).

 

 

Full-size image (21K)

 

Fig. 5. Schematic reconstruction of the Fe-Cu–Co sulphide distribution in the Kambove area (Cailteux, 1994). (A) Kw-236 low-grade deposit; (B) and (C) northern and southern folds, respectively, Kambove-Ouest deposit. The distances between A–B and B–C profiles are both 500 m.

View Within Article

 

 

Other economic to sub-economic Cu–Co mineralisations in the Menda and Luishia facies (e.g. Kambove-Ouest, Luishia, Luiswishi) are hosted stratigraphically higher up, in the Kambove Formation (upper part of the Mines Subgroup), i.e. 60–100 m above the classical upper orebody described above (Table 1 and Table 2; Fig. 4 and Fig. 5). For clarity, the small orebodies in the Kambove Formation will be called globally the third Congo-type orebody in this paper. The stratiform disseminated sulphides of this third orebody (4–20 m thick, 10–100 m long bodies) are hosted in tidal and reef lithologies similar to the host rocks in the lower and upper orebodies (Cailteux, 1978b, Cailteux, 1986 and Cailteux, 1994). However, the host rocks are this time part of a regressive sequence. A major point from these data is that Cu–Co ore deposits in the central Africa Copperbelt are closely linked to tidal and reef sedimentation.

3.2. Musoshi Subgroup Zambia-type deposits

The Zambia-type deposits are characterized by one or several orebodies, called “Ore Shale”, and hosted in the Ore Shale Formation (Darnley, 1960, Mendelsohn, 1961c, Garlick, 1961b, Garlick and Fleischer, 1972, Van Eden and Binda, 1972, Cailteux, 1973, Annels, 1974, Binda and Mulgrew, 1974, Clemmey, 1974, Van Eden, 1974 and Cailteux and Lefebvre, 1975). The Ore Shale Formation is a sedimentary unit ranging in lithology from quartzo-feldspathic wacke (Mufulira) to siltstone (e.g. Konkola, Chambishi), to finely laminated argillite (e.g. Nkana) and to shaley or siliceous dolomite (e.g. Baluba, Muliashi South) (Binda and Mulgrew, 1974 and Binda, 1994). The Musoshi Subgroup includes also lithologies (e.g. Nchanga, Nkana, Mufulira, Baluba, Mindola) transitional to those of the Congo-type lower and upper orebodies (Cailteux et al., 1994 and Binda, 1997). The lack of a reef sedimentary package in the Zambian orebody marks the predominantly clastic sedimentation in the Zambia-type Roan sequence. However, several reef occurrences were described or mentioned (e.g. Mufulira, Pitanda/NW and Kitwe/SE sides of the Chambishi-Nkana basin, Luanshya; Malan, 1964 and Clemmey, 1974). The Ore Shale Formation is marked by evaporitic conditions, as shown by preserved blebs and beds of anhydrite (Brandt et al., 1961, Annels, 1974 and Clemmey, 1974), and by tidal flat/subtidal regressive and transgressive sequences grading into stromatolitic carbonates (e.g. Kitwe; Clemmey, 1974).

The Ore Shale includes one (e.g. Musoshi-Konkola, Nkana; Jordaan, 1961, Schwellnus, 1961 and Cailteux, 1973) or several orebodies separated by barren (<1 wt.% Cu) or low grade (1–1.5 wt.% Cu) mineralised beds (e.g. Nchanga and Nchanga-West orebodies; Lower and Upper Orebodies at Luanshya; A, B, C Orebodies at Mufulira and Mimbula; Brandt et al., 1961, McKinon and Smit, 1961, Mendelsohn, 1961d, Smit, 1961 and Freeman, 1988b). The Ore Shale cumulative thickness is 5–50 m (average: 20–25 m), i.e. of the same order as the cumulative thickness of orebodies in the Congo-type deposits. Sulphides occur along foresets of cross-bedding, within troughs of ripples and along shale laminae; erosional channels interrupt mineralised beds; ore and its host-rocks display sedimentary deformation structures such as slumping or compaction cracks (Garlick, 1961b). Evaporitic conditions are supported by preserved blebs and beds of anhydrite (Brandt et al., 1961 and Annels, 1974). Clemmey (1974) documented a strong relation between mineralisation trends and the sedimentary context at Kitwe: (1) copper grades are highest toward the inferred palaeo-land and decrease away from it; (2) the alignment of copper grades is in shoots parallel to deducted ebb and flood tidal directions; (3) copper grades are controlled by facies distribution.

Lenticular quartzites, feldspathic quartzites, and dolomitic argillite in the hangingwall host a few small copper mineralisations forming the topmost orebodies, including The Feldspathic Quartzite (T.F.Q.) at N’changa and the weakly mineralised Glassy Quartzite at Mufulira (Binda and Mulgrew, 1974). These may be potential Zambian correlatives of the third orebody hosted in the Kambove Formation in the Congo-type deposits.

Several ore occurrences or deposits lie below the Ore Shale (Fig. 6), including four major orebodies at Kinsenda (Ngoyi and Dejonghe, 1997), three at Lubembe (Lefebvre, 1989 and Tshiauka, 2001), several orebodies in Nchanga (Voet and Freeman, 1972 and Freeman, 1988b), minor occurrences at Chambishi (Garlick, 1961c), one orebody at Nkana (Jordaan, 1961), three mineralised lenses at Chibuluma (Winfield, 1961) and Muliashi-South in the Luanshya district (Van Eden and Binda, 1972). These are not in the classical lithostratigraphic position of orebodies in the Copperbelt since these stratiform orebodies occur in arkoses of the commonly barren siliciclastic Mutonda Formation.

 

 

Full-size image (10K)

 

Fig. 6. Chingola B, C, D, E, F footwall orebodies (Binda, 1997).

View Within Article

 

 

3.3. Mwashya Subgroup deposits

The Mwashya Subgroup is exposed for several hundred kilometres along major Lufilian thrust faults between Kolwezi to the west and Kimpe to the southeast (Fig. 1). Several copper deposits and sub-economic occurrences (<1.0 wt.% Cu) were recorded in the Lower Mwashya: e.g. Shituru and Mulungwishi-Kampina (Likasi district), Kipoï (Luishia district), Kifumashi and Kasonta (Lubumbashi area) (François, 1974; Gécamines, unpublished data).

The Shituru deposit, located 1 km away of the Likasi Shituru plant (Fig. 1), is the only mined deposit hosted in the Lower Mwashya. It occurs on the southern flank of an anticline faulted along the fold axial plane. Mineralisation forms two stratiform orebodies (upper and lower) with high grades in the supergene zone (≈10.5 wt.% Cu and up to 2.0 wt.% Co; François, 1974 and Lefebvre, 1974), and with lower grades ( 2 wt.% Cu and 0.1 wt.% Co) at the deeper level (>80 m depth) (Lefebvre, 1974). Most ores are hosted in dolomitic laminites and dolomitic shales, lithologically similar to R.S.F./D.Strat-type and S.D.-type rocks of the Mines Subgroup, and interbedded with sub-economic (<1.0 wt.% Cu) stromatolitic massive dolomite (Lefebvre, 1974). No direct link has been found between the pyroclastic rocks interbedded in the Lower Mwashya and this copper mineralization (Lefebvre, 1974).

4. Copper–cobalt distribution in the central Africa Copperbelt

Based on present day mined out production, ore reserves and resource evaluation using cut off grades of 1 wt.% Cu, the economic orebodies host 82.2 Mt copper and 1.4 Mt cobalt in Zambia-type deposits, against 58.0 Mt copper and 4.6 Mt cobalt in Congo-type deposits. This adds up to 140 Mt copper and 6 Mt cobalt for the whole central African Copperbelt (Freeman, 1988a and Freeman, 1988b; Gécamines, unpublished data). Therefore, the Zambia-type orebodies contain 59% of the copper whereas the Congo-type orebodies host 77% of the cobalt in the Copperbelt. The known ore deposits in Western Zambia represent less than 1% of the total evaluated copper. The overall Co:Cu ratio is 1:57 in Zambia-type against 1:13 in Congo-type deposits, although cobalt-rich deposits reach Co:Cu ratios of 1:15 in Zambia-type and 3:1 in Congo-type orebodies. Geochemical studies (Kampunzu et al., unpublished work) indicate that the total amount of copper and cobalt contained in the Katangan economic orebodies represents ±8% and 0.8%, respectively, of the total (i.e. 1850 Mt copper and 750 Mt cobalt, respectively) metal contained in Roan sedimentary rocks.

The distribution of copper and cobalt deposits is related to the regional tectonic control of the distribution of the Mines Subgroup and lateral correlative units along the Lufilian Arc (François and Oosterbosch, 1968, François, 1973 and François, 1974). The Long and Kilamusembu facies exposed between Kolwezi and Tenke contain 26% and 19% of Katangan belt copper and cobalt resources, respectively; they are marked by low copper grades (1.0–2.0 wt.% Cu) and relatively low cobalt contents (0.1–0.4 wt.% Co). The Musonoi and Kalumbwe facies exposed between Kolwezi and Kakanda-Fungurume host copper-rich (>2.0 wt.% Cu) and cobalt-poor to cobalt-rich (<0.1–0.5 wt.% Co) ores, representing 56% and 61% of Katangan belt copper and cobalt resources, respectively. The southern Menda and Luishia facies (18% copper and 20% cobalt resources) is exposed from Kalongwe to Etoile and host copper- and cobalt-rich (>2.0 wt.% Cu and 0.4–0.6 wt.% Co) ores. Substantial nickel (from several hundred ppm up to 0.5 wt.% Ni) is associated with cobalt in the Menda and Luishia facies, forming Ni–Co sulphide deposits (e.g. Shinkolobwe). Between Lupoto and Lubembe, the Luishia facies host cobalt-poor (<0.1–0.4 wt.% Co) orebodies.

In Zambia, cobaltiferous deposits occur southwest of the Kafue Anticline (Fig. 2), e.g. Konkola-Chililabombwe, Nchanga, Chambishi Southeast, Chibuluma (West of Nkana), Luanshya and Baluba deposits (Annels et al., 1983, Annels and Simmonds, 1984 and Freeman, 1988b). Cobalt contents are generally between 0.1–0.2 wt.% Co (e.g. Chambishi, Nkana), with local higher values (e.g. up to 0.44 wt.% Co at Nchanga), matching the concentration range reported in Congo. There is almost no cobalt in the copper deposits northeast of the Kafue Anticline, e.g. Kinsenda (Congo), Mufulira, Bwana Mkubwa (Zambia). This pattern compares to that documented in Congo where the richest cobalt deposits occur along the southern fringe of the Copperbelt.

The Co:Cu ratio defines two geochemical groups of stratiform copper deposits in the central African Copperbelt (Fig. 7): (1) the first group represents cobalt-poor copper deposits marked by low Co:Cu ratio (0–0.02). This group includes most Zambia-type copper deposits and a few Congo-type deposits (e.g. Mutoshi in the Kolwezi area, Kakanda-Nord, Kalengwa in western Zambia); (2) the second group represents cobalt-rich copper deposits marked by high Co:Cu ratio (0.02–2.80), including most Congo-type deposits and some deposits in Zambia (e.g. Nkana-Mindola, Nchanga, Baluba). The highest Co grades within the total resources in Zambia (Co:Cu = 0.05–0.07) occur in the Baluba, Nkana-Mindola, Chibuluma deposits (Freeman, 1988b), which show transitional lithofacies between Zambia-type and Congo-type sedimentary sequences. Local high grades are documented in the Nchanga deposit. The Mwashya Subgroup Shituru deposit is part of the low Co:Cu ratio (0.01) deposits.

 

 

Full-size image (40K)

 

Fig. 7. Distribution of the Congo–Zambia Copperbelt Mines/Musoshi Subgroups deposits according to their Cu and Co tonnages.

View Within Article

 

 

5. Metal and sulphide distribution within the orebodies

Most Copperbelt deposits display vertical and lateral zoning of disseminated copper sulphides. In Zambia-type deposits, the vertical zoning starts with chalcocite–digenite–bornite at the bottom, followed by bornite–chalcopyrite and chalcopyrite-dominant zones, and pyrite at the top (e.g. Chambishi, Chibuluma, Baluba, Nchanga, Musoshi; Garlick, 1961c, Lee-Potter, 1961, McKinon and Smit, 1961, Cailteux, 1973 and Cailteux, 1974). A similar trend marks some Congo-type deposits. For example, the Kamoto deposit is characterized by chalcocite–digenite–bornite in the lower orebody (Grey R.A.T., D.Strat., R.S.F.) and in S.D.B., chalcopyrite and minor bornite in B.O.M.Z., chalcopyrite–pyrite in S.D.2a and pyrite in S.D.2b (Oosterbosch, 1962, Bartholomé, 1962, Bartholomáaaebbb, 1963 and Bartholomé, 1969). Outside the orebodies, pyrite is common and coexists sometimes with a few chalcopyrite grains.

grading into copper-poor zones and to pyritic-barren zones; e.g. Kambove-Ouest (Cailteux, 1983, Cailteux, 1986 and Cailteux, 1994) and Nchanga (McKinon and Smit, 1961).

 

 

Full-size image (19K)

 

Fig. 3. Kakanda-Nord: distribution of Cu and Co average concentrations on cross-sections from X: 0 to X: 1600 (LOB = lower orebody; UOB = upper orebody).

View Within Article

 

 

Some Cu–(few Co) primary sulphide mineralisations occur in the Mwashya Subgroup in Congo, and also are stratigraphically controlled in dolomites of the Lower Mwashya.

3.1. Mines Subgroup Congo-type deposits

The Congo-type stratiform deposits stretch from Kolwezi up to Kimpe (Fig. 2) and are generally characterized by two major Cu–Co orebodies, the “lower” and “upper” orebodies, totalling 15–55 m cumulative thickness (average: 20–25 m). The mineralisation is hosted in a transgressive supratidal to subtidal sedimentary sequence deposited under quiet, shallow-water conditions (Bartholomé et al., 1972, Cailteux, 1978a, Cailteux, 1983 and Cailteux, 1994). The host rocks contain blebs, nodules and lenticular beds of dolomite–quartz pseudomorphs after anhydrite and gypsum, and high contents of Mg, Ba, Sr, Li, B, Br can be linked to the deposition of sediments under saline evaporitic conditions (Bartholomé et al., 1972, Katekesha, 1975, Cailteux, 1978a, Cailteux, 1983, Cailteux, 1994 and Moine et al., 1986).

The lower orebody host-rocks include (Table 2): (1) a massive chloritic–dolomitic siltite known as Grey R.A.T. (“Roches Argilo-Talqueuses”); (2) a fine-grained stratified dolostone (D.Strat. “Dolomie Stratifiée”); (3) silicified-stromatolitic-dolomites forming laminites alternating with thin chloritic–dolomitic silty beds (R.S.F. “Roches Siliceuses Feuilletées”). The Upper Orebody host-rocks include (Table 2): (1) the basal Dolomitic Shales (S.D.B., “Shales Dolomitiques de Base” also called S.D.1a); (2) an overlying coarse grained impure dolostone (B.O.M.Z., “Black Ore Mineralised Zone” also called S.D.1b) which is sometimes missing in the succession (e.g. in the Kambove area). A generally “barren” reef-type stromatolitic dolomite (R.S.C., “Roches Siliceuses Cellulaires”) occurs between the two orebodies. Ores are known on 0.4–1.0 m thickness along the contact between this reef dolomite and both lower and upper orebodies. The chloritic-silty-dolomitic lenses or layers locally interbedded within the R.S.C. are also mineralised (e.g. Kamoto). In some deposits (e.g. Kambove-Ouest), the primary stratiform mineralisation extends to the overlying carbonaceous dolomitic shales S.D.2a, up to the base of the S.D.2b. The organic matter content is variable, generally low, although local high contents have led to the development of black shales and dolomites in R.S.F.-R.S.C.-S.D.B units (Cailteux, 1983).

The Congo-type mineralised succession is very regular along strike (Fig. 2), showing the same lithological succession for >350 km, from Kolwezi (Demesmaeker et al., 1963, François, 1973 and Katekesha, 1975), to Tenke-Fungurume (Oosterbosch, 1950 and Oosterbosch, 1951), Kambove-Kakanda (Cailteux, 1978a and Cailteux, 1983), Kabolela (Lefebvre, 1976a and Lefebvre, 1976b), Etoile (Lefebvre and Cailteux, 1975) and Lubembe (Lefebvre and Tshiauka, 1986 and Tshiauka et al., 1995). However, there is a clear across-strike lithofacies variation marking a progressive evolution from more near-shore (north) to more reefal (south) environments (François, 1973, François, 1974, Lefebvre, 1979, Cailteux, 1978a, Cailteux, 1978b, Cailteux, 1983 and Cailteux, 1994). This palaeo-environmental variation seems to correlate with different copper–cobalt grades in the rocks (François, 1973 and François, 1974, and details below). The northern (present coordinates) near-shore sequences (“Long” and “Kilamusembu” facies) are characterized by the absence of stromatolites, the occurrence of dolomites and arenites in the Dolomitic Shales Formation and of arenites in the Kambove Formation. In these two sequences, the lithostratigraphic units usually hosting the orebodies are barren or poorly mineralised (e.g. Dipeta Syncline between Tenke and Fungurume), except in the Tenke deposit. The Kilamusembu facies occurs only in the Kolwezi area and represents a transitional facies between Long and Musonoï facies. The southern sequences (“Musonoï” and “Kalumbwe” facies) are marked by: (a) clasts of stromatolites; (b) stromatolites in R.S.C.; (c) lack of arenites in Dolomitic Shales and Kambove Formations. There are no dolomites in the Kalumbwe facies Dolomitic Shales Formation (e.g. Kakanda-Nord). This sequence hosts the most important copper–cobalt deposits (e.g. Kamoto, Fungurume), with only a few barren or poorly mineralised zones. The southernmost reef sequence (“Menda” and “Luishia” facies) is marked by algal bioherms in R.S.C. and in the Kambove Formation. The lithostratigraphic units usually hosting the orebodies are barren or poorly to well mineralised (e.g. Kambove-Ouest, Luishia, Luiswishi).

Sub-economic orebodies (generally <1 wt.% Cu) and small economic deposits (locally >2 wt.% Cu) occur in dark-grey to black carbonaceous metapelites forming the S.D.2d and 3b (Fig. 4 and Fig. 5; Table 2). However, the metals in these units are strictly bound to thin organic matter-rich horizons indicating deposition under strong reducing conditions.

 

 

Full-size image (73K)

 

Fig. 4. Cross-section X: 90 E through the Kambove-Ouest deposit and Roan breccia below the oxidized zone (modified from Cailteux, 1983); KTO = Kamoto Formation, SD = Dolomitic Shale Formation, KVE = Kambove Formation; R.A.T. = Roches Argilo-Talqueuses.

View Within Article

 

 

Full-size image (69K)

 

Fig. 2. Location of the main stratiform Cu–Co deposits in the central African Copperbelt (KA = Kafue Anticline); modified from François, 1974 and Cailteux, 1994.

View Within Article

 

 

The Katangan supracrustal sedimentary succession is 5–10 km thick and commonly sub-divided into three major lithostratigraphic units (François, 1974 and François, 1995): Roan, Nguba and Kundelungu Groups (Table 1). The Roan Group is made up of siliciclastic and carbonate sedimentary rocks (fluviatile and lacustrine sediments; Buffard, 1988, Cailteux, 1994 and Cailteux et al., 1994), and volcanic and plutonic mafic rocks emplaced in a continental rift (Kampunzu et al., 2000 and references therein). The Nguba supracrustal assemblage is made up of siliciclastic and carbonate sedimentary rocks (François, 1974 and Buffard, 1988) and includes mafic igneous rocks emplaced in a proto-oceanic rift similar to the Red Sea (Kampunzu et al., 1991, Kampunzu et al., 1993, Manteka et al., 1985 and Kapenda et al., 1998). Kundelungu sedimentary rocks represent syn- to post-orogenic sedimentary deposits (Kampunzu and Cailteux, 1999). The tabular-shaped Kundelungu is a continental clastic molasse sequence extending into the lower Palaeozoic (Kampunzu and Cailteux, 1999). The Katangan basin closed during the Lufilian Orogeny leading to the development of predominantly north-verging folds, thrusts and nappes. In Congo, all exposed Roan (except for the Nzilo basal conglomerate), Nguba and folded Kundelungu (excluding tabular Kundelungu) sedimentary rocks are part of allochthonous tectonic sheets. The Mwashya Subgroup rocks conformably overly the lowest Roan rocks (Cailteux et al., 1994), and are conformably overlain by the Grand Conglomérat (Table 1).

Grujenschi, 1979 and Wendorff, 2000a interpreted occurrences of megabreccia in the Katangan succession as sedimentary syntectonic conglomerates (olistostromes), and Wendorff (2000b) further questioned the lithostratigraphic succession of the Roan Group. Although the existence of synorogenic sedimentary rocks in the Katangan is not a matter of debate (cf. Grujenschi, 1979), several claims in Wendorff (2000b) are not supported by available data (e.g. Cailteux et al., this volume and Kampunzu et al., this volume). In all cases, the debate on the lithostratigraphy of the Katangan is outside the objectives of this paper, and only the common lithostratigraphic position of the Mines Subgroup and related ore deposits is discussed further here below.

The stratiform Copperbelt copper–cobalt orebodies occur in the Roan Group (i.e. in Mines and Mwashya Subgroups; Table 1). The Roan Group sedimentary rocks display a regional lateral variation of facies between Zambia-type and Congo-type successions. In Zambia and SE Congo, deposits are mainly hosted in para-autochthonous siliciclastic rocks close to basement terrains. The main ore deposits define two parallel trends, north-east and south-west of the Kafue anticline (Fig. 2). The known deposits which lie off these two trends (e.g. Western province in Zambia) are assumed to be of smaller economic importance (Freeman, 1988a), although this could be a conclusion biased by inappropriate exploration coverage of this area.

The lowest Roan fluviatile sedimentary rocks rest unconformably on the pre-Katangan basement (Mendelsohn, 1961a and Binda and Mulgrew, 1974). In Congo, Cu–Co deposits and their host rocks define thrust sheets, nappes and klippen formed during the Lufilian Orogeny (Demesmaeker et al., 1963, Cailteux and Kampunzu, 1995 and Kampunzu and Cailteux, 1999). The dominant lithological units are dolomites and dolomitic shales (Oosterbosch, 1962, Demesmaeker et al., 1963, François, 1974, François, 1987 and Cailteux, 1994).

The lowest formations of the Roan Group (R.A.T.—“Roches Argilo-Talqueuses”—and Mindola Subgroups; Table 1) were deposited in an oxic environment. In Zambia, the Mindola Subgroup includes (Binda, 1994, Cailteux et al., 1994 and Tshiauka et al., 1995): a scree-type boulder conglomerate at the base (Chimfunsi Formation), followed by aeolian quartzites (Kafufya Formation) and by immature braided stream/alluvial fan conglomerates, arkoses and upward-fining sandstone sequences (Mutonda Formation). In Congo, the base of the R.A.T. Subgroup is unknown (François, 1974), but a boulder conglomerate, probably correlative of the Chimfunsi Formation, occurs at Nzilo above the Kibaran basement. R.A.T. sedimentary rocks, laterally correlative of the Mutonda Formation (Cailteux et al., 1994), include red chlorite-rich dolomitic siltstones, dolomitic fine-grained sandstones, silty dolostones and dolomitic silty chloritites (Oosterbosch, 1950, Katekesha, 1975, Cailteux, 1978a, Cailteux, 1983 and Cailteux, 1994).

Wendorff (2000b) claims that Red and Grey R.A.T. are syn-orogenic sedimentary rocks younger than the Roan Group and deposited in the Katangan foreland basin after the deposition of the Nguba Group. However, field observations and geochemical data invalidate this interpretation (Cailteux et al., this volume and Kampunzu et al., this volume). Furthermore, the same author suggested that the Nzilo conglomerate is part of the Mwashya Subgroup, but there is no field evidence supporting this interpretation (e.g. Byamungu et al., 1979 and Madi, 1985).

Musoshi (Zambia) and Mines (Congo) Subgroups (Table 2) represent a transgressive succession deposited in a reducing evaporitic environment. They include a succession of arenites, silty-sandy argillites and shales exposed north of the Kafue “Anticline” (Zambia), dolomitic shales and dolomites in Congo and south of the Kafue “Anticline” in Zambia (Bartholomé et al., 1972, Annels, 1974, Binda and Mulgrew, 1974, Cailteux, 1978a, Cailteux, 1994, Cailteux et al., 1994 and Tshiauka et al., 1995). A carbonate unit marks the top of the laterally correlative mineralised successions in Congo and Zambia. The copper–cobalt orebodies occur in the lower part of these successions and the stratiform mineralisation was deposited before the Lufilian compressional tectonics both in Congo and Zambia, as shown by folds and thrusts affecting the orebodies (Garlick, 1940, Reynolds, 1959, Mendelsohn, 1961c, Demesmaeker et al., 1963, François, 1973, Katekesha, 1975, Cailteux, 1983 and Cailteux and Kampunzu, 1995).

 

Table 2.

Lithostratigraphy of the Mines Subgroup in Congo (modified from François, 1987 and Cailteux, 1994)

Sub-group	Formation	Member	Lithology
Mines R-2	Kambove R-2.3 (up to 190 m)	Upper R-2.3.2	White to pink massive dolomites and more or less talcose finely bedded dolomites, with interbedded grey to pink-red chloritic–dolomitic siltstones, occasional evaporitic-type collapse breccias and intraformational conglomerates
			More or less carbonaceous, massive dolomites with occasional stromatolites, more or less talcose finely bedded dolomites with interbedded chloritic–dolomitic siltstones, occasional evaporitic-type collapse breccias and intraformational conglomerates	}	Third Orebody (lenses)
			Pink-brown to white massive dolomite
		Lower R-2.3.1	More or less carbonaceous, talcose, massive or finely bedded dolomites, with occasional oolitic or cryptoalgal beds
			More or less carbonaceous laminitic dolomites with tabular stromatolites, talcose to the top
			More or less carbonaceous, massive, stromatolitic dolomites with interbedded dolomitic shales and laminitic dolomites
	Shales dolomitic R-2.2 (up to 110 m)	S.D.-3b	Black carbonaceous weakly dolomitic shale
		S.D.-3a	Highly dolomitic shales, with occasional stromatolitic dolomite bed at top or at base
		S.D.-2d	Black carbonaceous weakly dolomitic shale
		S.D.-2c	Highly dolomitic shales; occasional black carbonaceous shale at base
		S.D.-2b	Dolomitic shales, with frequent stromatolitic dolomite bed at base
		S.D.-2a	Black carbonaceous weakly dolomitic shale
		S.D.-1b (B.O.M.Z.)	Silty and chloritic dolomite, coarse crystalline dolomite and dolomitic shales, with nodules and concretions pseudomorph after anhydrite	}	Upper Orebody
		S.D.-1a (S.D.B.)	Dolomitic shales, with lenticular beds and nodules pseudomorph after anhydrite
	Kamoto R-2.1 (up to 50 m)	R.S.C.	Massive, stromatolitic dolomites, with interbedded dolomitic siltstones
		R.S.F.	Siliceous finely bedded dolomites with laminitic stromatolites; interbedded dolomitic siltstones or shales	}	Lower Orebody
		D.Strat.	More or less silty and chloritic stratified dolomites
		R.A.T. grises	Grey chloritic–dolomitic massive siltstone (up to 10 m)

Full-size table

View Within Article

 

 

The laterally correlative Kirilabombwe (Zambia) and Dipeta (Congo) Subgroups display strong similarities, e.g. the lithological succession includes arkoses, conglomerates, siltstones, dolomitic shales, dolomites, and these lithologies show a similar succession. The Mwashya Subgroup is characterized by platform carbonates in the Lower Mwashya, grading to more open marine dolomitic shales, black shales or sandstones in the Upper Mwashya. Gabbros intruding the Upper Roan/Dipeta formations (but not the Mwashya Subgroup) and mafic lavas and pyroclastic rocks in the Lower Mwashya belong to a single syn-Lower Mwashya igneous event (Kampunzu et al., 2000 and references therein) dated at 760 ± 5 Ma by U–Pb SHRIMP technique (Key et al., 2001). The upper Mwashya is overlain by a glacial diamictite, called “Grand Conglomérat” which starts the Nguba succession (Cahen, 1954 and Binda and Van Eden, 1972).

3. Lithostratigraphic control of copper–cobalt ores

Major primary deposits and most primary copper occurrences are stratigraphically controlled (Table 1 and Table 2), i.e. they occur in the Kamoto Dolomite and Dolomitic Shales Formations of the Mines Subgroup in Congo (Oosterbosch, 1962 and Cailteux, 1994 and references therein), and in lateral correlative units known as the Ore Shale Formation (Binda and Mulgrew, 1974 and Cailteux et al., 1994) at the base of the Musoshi Subgroup in Zambia. Within these lithostratigraphic units, the orebodies extend for hundreds of metres (e.g. Kakanda-Nord; Fig. 3) to several kilometres (e.g. Dikuluwe-Mashamba at Kolwezi; Luanshya in Zambia) along strike, except where they are interrupted by compressional structures related to the Lufilian orogeny (Demesmaeker et al., 1963 and Kampunzu and Cailteux, 1999). The lateral variation of sulphides in the orebodies shows copper-rich zones

Genesis of sediment-hosted stratiform copper–cobalt deposits, central African Copperbelt

J.L.H. Cailteux^{a, ,} , A.B. Kampunzu^b, C. Lerouge^c, A.K. Kaputo^d and J.P. Milesi^c

^aDépartement Recherche et Développement, E.G.M.F., Groupe G. Forrest International, Lubumbashi, D.R. Congo

^bDepartment of Geology, University of Botswana, Private Bag 0022, Gaborone, Botswana

^cB.R.G.M., BP 6009, 45060 Orléans cedex 2, France

^dDépartement Géologique, GECAMINES, Likasi, D.R. Congo

Received 1 October 2003; 

accepted 1 May 2004. 

Available online 27 October 2005.

 

Abstract

The Neoproterozoic central African Copperbelt is one of the greatest sediment-hosted stratiform Cu–Co provinces in the world, totalling 140 Mt copper and 6 Mt cobalt and including several world-class deposits ( 10 Mt copper). The origin of Cu–Co mineralisation in this province remains speculative, with the debate centred around syngenetic–diagenetic and hydrothermal-diagenetic hypotheses.

The regional distribution of metals indicates that most of the cobalt-rich copper deposits are hosted in dolomites and dolomitic shales forming allochthonous units exposed in Congo and known as Congolese facies of the Katangan sedimentary succession (average Co:Cu = 1:13). The highest Co:Cu ratio (up to 3:1) occurs in ore deposits located along the southern structural block of the Lufilian Arc. The predominantly siliciclastic Zambian facies, exposed in Zambia and in SE Congo, forms para-autochthonous sedimentary units hosting ore deposits characterized by lower a Co:Cu ratio (average 1:57). Transitional lithofacies in Zambia (e.g. Baluba, Mindola) and in Congo (e.g. Lubembe) indicate a gradual transition in the Katangan basin during the deposition of laterally correlative clastic and carbonate sedimentary rocks exposed in Zambia and in Congo, and are marked by Co:Cu ratios in the range 1:15.

The main Cu–Co orebodies occur at the base of the Mines/Musoshi Subgroup, which is characterized by evaporitic intertidal–supratidal sedimentary rocks. All additional lenticular orebodies known in the upper part of the Mines/Musoshi Subgroup are hosted in similar sedimentary rocks, suggesting highly favourable conditions for the ore genesis in particular sedimentary environments. Pre-lithification sedimentary structures affecting disseminated sulphides indicate that metals were deposited before compaction and consolidation of the host sediment.

The ore parageneses indicate several generations of sulphides marking syngenetic, early diagenetic and late diagenetic processes. Sulphur isotopic data on sulphides suggest the derivation of sulphur essentially from the bacterial reduction of seawater sulphates. The mineralizing brines were generated from sea water in sabkhas or hypersaline lagoons during the deposition of the host rocks. Changes of Eh–pH and salinity probably were critical for concentrating copper–cobalt and nickel mineralisation. Compressional tectonic and related metamorphic processes and supergene enrichment have played variable roles in the remobilisation and upgrading of the primary mineralisation.

There is no evidence to support models assuming that metals originated from: (1) Katangan igneous rocks and related hydrothermal processes or; (2) leaching of red beds underlying the orebodies. The metal sources are pre-Katangan continental rocks, especially the Palaeoproterozoic low-grade porphyry copper deposits known in the Bangweulu block and subsidiary Cu–Co–Ni deposits/occurrences in the Archaean rocks of the Zimbabwe craton. These two sources contain low grade ore deposits portraying the peculiar metal association (Cu, Co, Ni, U, Cr, Au, Ag, PGE) recorded in the Katangan sediment-hosted ore deposits. Metals were transported into the basin dissolved in water.

The stratiform deposits of Congo and Zambia display features indicating that syngenetic and early diagenetic processes controlled the formation of the Neoproterozoic Copperbelt of central Africa.

Keywords: Metallogeny; Copper–cobalt; Sedimentary rocks; Katangan; Central African Copperbelt; Congo-Zambia

Article Outline

1. Introduction

2. Geological setting

3. Lithostratigraphic control of copper–cobalt ores

3.1. Mines Subgroup Congo-type deposits

3.2. Musoshi Subgroup Zambia-type deposits

3.3. Mwashya Subgroup deposits

4. Copper–cobalt distribution in the central Africa Copperbelt

5. Metal and sulphide distribution within the orebodies

6. Ore petrology

6.1. Sulphide parageneses

6.2. Relations between sulphides and gangue minerals

6.3. Relations between orebodies and deformation events

7. Isotopic geochemistry

8. Thermometry

9. Fluid inclusions

10. Discussion

11. Conclusions

Acknowledgements

References

1. Introduction

The Neoproterozoic Katangan Copperbelt of central Africa stretches on both sides of the border between Zambia and Democratic Republic of Congo (DRC; hereafter Congo). It hosts one of the world’s greatest concentration of stratiform copper–cobalt deposits, representing more than half of the world’s mineable cobalt and includes world-class Cu–Co deposits, e.g. Kolwezi, Tenke-Fungurume, Konkola-Chililabombwe, Nchanga, Nkana, Mufulira, each containing 10 Mt copper. Total copper hosted in the Katangan basin of central Africa is close to 200 Mt if sub-economic (Cu 1 wt.%) occurrences are included (data from Gécamines Mining Company for DRC and from Freeman, 1988 for Zambia). Copper and cobalt are associated with iron, and sometimes with anomalous concentrations of other metals (e.g. Ni, U, Ag, Au, PGM, Se, Mo, V, Te, As, Th). The ore is mainly made of disseminated sulphides forming stratiform orebodies hosted in fine-grained siliciclastic or dolomitic sedimentary rocks.

Since the discovery of the Copperbelt in the early 1900s, several metallogenic hypotheses were proposed to explain the primary source of metals and the mineralisation process. The historical review of these genetic theories is given in Sweeney et al., 1991a, Sweeney et al., 1991b and Sweeney and Binda, 1994 for the Zambian Copperbelt. The epigenetic hypothesis suggests the introduction of hydrothermal mineralizing solutions after the deposition, lithification and deep burial of sediments. In this model, the hydrothermal fluids are supposed to originate from the emplacement of granite–granodiorite–tonalite bodies in the Copperbelt (Gray, 1929, Davidson, 1931, Jackson, 1932, Thoreau and du Trieu de Terdonck, 1932, Derriks and Vaes, 1956, Derriks and Oosterbosch, 1958, Darnley, 1960 and Vaes, 1962). The existence of minor sulphide veins or veinlets within a few sediment-hosted copper deposits in Zambia (e.g. Nchanga) and in Congo (e.g. Shinkolobwe) and within a few Zambian granites was taken as a support for this interpretation. However, an unconformable erosional contact occurs between the granitoids and the overlying Katangan sedimentary succession in Zambia (Garlick, 1961a and Binda, 1972). This is supported by U–Pb zircon geochronological data (Armstrong et al., 1999, Rainaud et al., 1999 and De Waele and Mapani, 2002) indicating that the granitoids exposed in the Copperbelt and surrounding areas are older than the Katangan sedimentary succession, i.e. Palaeoproterozoic (2.05–1.65 Ga), Mesoproterozoic (predominantly 1.05–1.0 Ga) or early Neoproterozoic, e.g. 0.88 Ga for the Nchanga granite which is unconformably overlain by the oldest Katangan sedimentary rocks.

Emerging in the 1930s, the syngenetic theory linked the deposition of metals to the deposition of host-sediments (Schneiderhöhn, 1931, Schneiderhöhn, 1932, Schneiderhöhn, 1937, Garlick, 1945, Garlick, 1961b, Garlick, 1967 and Garlick, 1989). Metals were sourced from continental erosion and transported in solution by rivers to the sedimentary depocentres. Ore sulphide precipitation occurred in reducing stagnant water under high bacterial activity and decomposition of organic matter. This hypothesis was based on: (1) the existence of sulphide zonal distribution parallel to the palaeo-shorelines inferred to mark marine transgression–regression events; (2) the coincidence between the polarity of the sulphide zonation and the sedimentary palaeocurrent directions. However, the lack of a systematic correlation between all transgressive/regressive events and lateral/vertical zonation of sulphides, and the discontinuity of the mineralisation within a single lithostratigraphic unit invalidated this model (e.g. Annels, 1974, Renfro, 1974 and Sweeney and Binda, 1994).

Studies related to diagenetic processes in sedimentary rocks triggered the diagenetic model for the central African copper orebodies. Two sulphide generations were documented in the orebodies: (1) the earliest copper–(cobalt)-sulphide generation (hereafter sulphide I) grew during the deposition and the early diagenetic stage of the host-sediments; (2) the second copper–(cobalt)-sulphide generation was inferred to form during a large scale chemical reaction between the host-sediment interstitial water and a metal-bearing brine (Bartholomé, 1962, Bartholomáaaebbb, 1963, Bartholomé, 1969, Bartholomé, 1974 and Bartholomé et al., 1972). However, the model does not address the origin of solutions, the primary source of metals, and the exact timing of mineralisation (early, late diagenesis). These unknowns led to a hydrothermal-diagenetic model linking the mineralizing fluids to late diagenetic hydrothermal fluids of undefined origin (Cluzel and Guilloux, 1986) or originating from mafic igneous rocks or rift related processes (Annels, 1974, Annels, 1979, Annels, 1989, Annels and Simmonds, 1984, Lefebvre, 1989 and Unrug, 1988).

Cailteux et al. (1994) showed that stratiform copper–cobalt orebodies in Zambia and Congo are hosted in laterally correlative formations (Table 1). Therefore, the aim of this paper is to review data from both countries showing striking similarities between Congo-type and Zambia-type deposits, and allowing us to further constrain the mineralizing processes.

Table 1.

Lithostratigraphy of the Katangan succession in Congo and Zambia (modified from François, 1974, François, 1995, Cailteux, 1994 and Cailteux, 2003; maximum age of the Katangan based on SHRIMP U–Pb zircon dating by Armstrong et al., 1999)

	Group	Sub-group	Lithologies
±500 Ma	Kundelungu (prev. Upper Kundelungu) Ku	Plateaux Ku 3	Arkoses, conglomerates, sandstones, shales
		Kiubo Ku 2	Sandstones, carbonated siltstones or shales, limestones
		Kalule Ku 1	Ku 1.3: Carbonated siltstones and shales; grey to pink oolitic limestone at base (“Calcaire Rose Oolitique”)
			Ku 1.2: Carbonated siltstones and shales; pink to grey dolomite at base (“Calcaire Rose”)
±620 Ma			Ku 1.1 “Petit Conglomérat”: glacial diamictite
	Nguba (prev. Lower Kundelungu) Ng	Monwezi Ng 2	Dolomitic sandstones, siltstones or shales
		Likasi Ng 1	Ng 1.3: Carbonated siltstones and shales
			Ng 1.2: Dolomites, limestones, dolomitic shales and siltstones
±750 Ma			Ng 1.1 “Grand Conglomérat”: glacial diamictite
Congo				Zambia
Group	Sub-group	Formation	Lithologies	Lithologies	Formation	Sub-group
Roan	Mwashya R-4	Upper R-4.2	Shales, carbonaceous shales or sandsones	Dolomitic shales, grey to black carbonaceous shales, feldspathic sandstones		Mwashia
		Lower R-4.1	Dolomites, jasper beds, pyroclastics and hematite; local stratiform Cu–Co mineralisation
	Dipeta R-3	R-3.4	Dolomites interbedded with argillaceous to dolomitic siltstones and feldspathic sandstones; intrusive basic bodies	Dolomites interbedded with dolomitic shales; gabbroic bodies	Kanwangungu	Kirilabombwe RU 1 - RU 2 RL 3
		R-3.3
		R-3.2
		R.G.S. R-3.1	Dolomitic siltstones	Shales with grit	Kibalongo
	Mines R-2	Kambove R-2.3	Laminitic, stromatolitic, talcose dolomites and dolomitic siltstones; local stratiform Cu–Co mineralisation	Dolomites or argillaceous dolomites; local stratiform Cu mineralisation	Chingola	Musoshi RL 4 - RL 6
		Dolomitic shale R-2.2	Dolomitic shales, carbonaceous shales, dolomites and occasional sandstones or arkoses	Arenites, argillites and dolomitic argillites; occasional dolomites at the base; main stratiform Cu–Co mineralisation in the lower part (Ore Shale)	Pelito-arkosic
			Dolomitic shales, sandy dolomite at top; stratiform Cu–Co (Upper Orebody)		Ore Shale
		Kamoto R-2.1	R-2.1.3 “Roches Silicueuses Cellulaires”: stromatolitic dolomite with interbedded siltstones; Cu–Co at top & base
			R-2.1.2: bedded dolomites with siltstones; silty dolomite in the lower part; stratiform Cu–Co (Lower Orebody)
			R-2.1.1 “R.A.T. grises”: dolomitic siltstone; Cu–Co at top
	R.A.T. R-1	R-1.3	Pink-lilac, hematitic, chloritic–dolomitic massive siltstones	Conglomerates, coarse arkoses and argillaceous siltstones; occasional Cu–Co mineralisation	Mutonda	Mindola (Footwall) RL 7
		R-1.2	Pink to purple-grey, hematitic, chloritic siltstones; sandstones in the lower part; stromatolitic dolomite at top
		R-1.1	purple-red, hematitic, slightly dolomitic bedded siltstones
	base of the R.A.T. sequence unknown			Quartzites	Kafufya
<900 Ma	Basal conglomerate			Pebble and coble conglomerate	Chimfunsi

Full-size table

View Within Article

 

 

2. Geological setting

The Neoproterozoic Katangan belt forms a north-directed thrust-and-fold arc, called the “Lufilian Arc”, located between the Congo and Kalahari cratons (Fig. 1). It is more than 150 km wide and stretches for 700 km from Mwinilunga in the west (e.g. Brock, 1961 and Steven, 2000), to Kolwezi in the northwest, up to Luanshya (previously Roan Antelope) and Lonshi in the southeast of the belt (Fig. 2). It is commonly assumed that this copper metallogenic province is bounded by the Mwembeshi Dislocation Zone, but ongoing investigations (IGCP-302 and 450 projects) suggest that it possibly forms a vast copper metallogenic province extending into the Zambezi belt to the east, and linking southwestwards (Fig. 1) with the Kalahari Copperbelt of Botswana (Ghanzi-Chobe belt) and Namibia (Damara/Otavi belt).

 

 

Full-size image (40K)

 

Fig. 1. Location of the central African Copperbelt between the Congo and Kalahari cratons.

View Within Article

 

 

CLOSE

Fig. 7. Distribution of the Congo–Zambia Copperbelt Mines/Musoshi Subgroups deposits according to their Cu and Co tonnages.

CLOSE

Cailteux et al. (1994) showed that stratiform copper–cobalt orebodies in Zambia and Congo are hosted in laterally correlative formations (Table 1). Therefore, the aim of this paper is to review data from both countries showing striking similarities between Congo-type and Zambia-type deposits, and allowing us to further constrain the mineralizing processes.

Table 1.

Lithostratigraphy of the Katangan succession in Congo and Zambia (modified from François, 1974, François, 1995, Cailteux, 1994 and Cailteux, 2003; maximum age of the Katangan based on SHRIMP U–Pb zircon dating by Armstrong et al., 1999)

	Group	Sub-group	Lithologies
±500 Ma	Kundelungu (prev. Upper Kundelungu) Ku	Plateaux Ku 3	Arkoses, conglomerates, sandstones, shales
		Kiubo Ku 2	Sandstones, carbonated siltstones or shales, limestones
		Kalule Ku 1	Ku 1.3: Carbonated siltstones and shales; grey to pink oolitic limestone at base (“Calcaire Rose Oolitique”)
			Ku 1.2: Carbonated siltstones and shales; pink to grey dolomite at base (“Calcaire Rose”)
±620 Ma			Ku 1.1 “Petit Conglomérat”: glacial diamictite
	Nguba (prev. Lower Kundelungu) Ng	Monwezi Ng 2	Dolomitic sandstones, siltstones or shales
		Likasi Ng 1	Ng 1.3: Carbonated siltstones and shales
			Ng 1.2: Dolomites, limestones, dolomitic shales and siltstones
±750 Ma			Ng 1.1 “Grand Conglomérat”: glacial diamictite
Congo				Zambia
Group	Sub-group	Formation	Lithologies	Lithologies	Formation	Sub-group
Roan	Mwashya R-4	Upper R-4.2	Shales, carbonaceous shales or sandsones	Dolomitic shales, grey to black carbonaceous shales, feldspathic sandstones		Mwashia
		Lower R-4.1	Dolomites, jasper beds, pyroclastics and hematite; local stratiform Cu–Co mineralisation
	Dipeta R-3	R-3.4	Dolomites interbedded with argillaceous to dolomitic siltstones and feldspathic sandstones; intrusive basic bodies	Dolomites interbedded with dolomitic shales; gabbroic bodies	Kanwangungu	Kirilabombwe RU 1 - RU 2 RL 3
		R-3.3
		R-3.2
		R.G.S. R-3.1	Dolomitic siltstones	Shales with grit	Kibalongo
	Mines R-2	Kambove R-2.3	Laminitic, stromatolitic, talcose dolomites and dolomitic siltstones; local stratiform Cu–Co mineralisation	Dolomites or argillaceous dolomites; local stratiform Cu mineralisation	Chingola	Musoshi RL 4 - RL 6
		Dolomitic shale R-2.2	Dolomitic shales, carbonaceous shales, dolomites and occasional sandstones or arkoses	Arenites, argillites and dolomitic argillites; occasional dolomites at the base; main stratiform Cu–Co mineralisation in the lower part (Ore Shale)	Pelito-arkosic
			Dolomitic shales, sandy dolomite at top; stratiform Cu–Co (Upper Orebody)		Ore Shale
		Kamoto R-2.1	R-2.1.3 “Roches Silicueuses Cellulaires”: stromatolitic dolomite with interbedded siltstones; Cu–Co at top & base
			R-2.1.2: bedded dolomites with siltstones; silty dolomite in the lower part; stratiform Cu–Co (Lower Orebody)
			R-2.1.1 “R.A.T. grises”: dolomitic siltstone; Cu–Co at top
	R.A.T. R-1	R-1.3	Pink-lilac, hematitic, chloritic–dolomitic massive siltstones	Conglomerates, coarse arkoses and argillaceous siltstones; occasional Cu–Co mineralisation	Mutonda	Mindola (Footwall) RL 7
		R-1.2	Pink to purple-grey, hematitic, chloritic siltstones; sandstones in the lower part; stromatolitic dolomite at top
		R-1.1	purple-red, hematitic, slightly dolomitic bedded siltstones
	base of the R.A.T. sequence unknown			Quartzites	Kafufya
<900 Ma	Basal conglomerate			Pebble and coble conglomerate	Chimfunsi

Full-size table

View Within Article

 

 

2. Geological setting

The Neoproterozoic Katangan belt forms a north-directed thrust-and-fold arc, called the “Lufilian Arc”, located between the Congo and Kalahari cratons (Fig. 1). It is more than 150 km wide and stretches for 700 km from Mwinilunga in the west (e.g. Brock, 1961 and Steven, 2000), to Kolwezi in the northwest, up to Luanshya (previously Roan Antelope) and Lonshi in the southeast of the belt (Fig. 2). It is commonly assumed that this copper metallogenic province is bounded by the Mwembeshi Dislocation Zone, but ongoing investigations (IGCP-302 and 450 projects) suggest that it possibly forms a vast copper metallogenic province extending into the Zambezi belt to the east, and linking southwestwards (Fig. 1) with the Kalahari Copperbelt of Botswana (Ghanzi-Chobe belt) and Namibia (Damara/Otavi belt).

 

 

Full-size image (40K)

 

Fig. 1. Location of the central African Copperbelt between the Congo and Kalahari cratons.

View Within Article

 

 

 

Genesis of sediment-hosted stratiform copper–cobalt deposits, central African Copperbelt

J.L.H. Cailteux^{a, ,} , A.B. Kampunzu^b, C. Lerouge^c, A.K. Kaputo^d and J.P. Milesi^c

^aDépartement Recherche et Développement, E.G.M.F., Groupe G. Forrest International, Lubumbashi, D.R. Congo

^bDepartment of Geology, University of Botswana, Private Bag 0022, Gaborone, Botswana

^cB.R.G.M., BP 6009, 45060 Orléans cedex 2, France

^dDépartement Géologique, GECAMINES, Likasi, D.R. Congo

Received 1 October 2003; 

accepted 1 May 2004. 

Available online 27 October 2005.

 

Abstract

The Neoproterozoic central African Copperbelt is one of the greatest sediment-hosted stratiform Cu–Co provinces in the world, totalling 140 Mt copper and 6 Mt cobalt and including several world-class deposits ( 10 Mt copper). The origin of Cu–Co mineralisation in this province remains speculative, with the debate centred around syngenetic–diagenetic and hydrothermal-diagenetic hypotheses.

The regional distribution of metals indicates that most of the cobalt-rich copper deposits are hosted in dolomites and dolomitic shales forming allochthonous units exposed in Congo and known as Congolese facies of the Katangan sedimentary succession (average Co:Cu = 1:13). The highest Co:Cu ratio (up to 3:1) occurs in ore deposits located along the southern structural block of the Lufilian Arc. The predominantly siliciclastic Zambian facies, exposed in Zambia and in SE Congo, forms para-autochthonous sedimentary units hosting ore deposits characterized by lower a Co:Cu ratio (average 1:57). Transitional lithofacies in Zambia (e.g. Baluba, Mindola) and in Congo (e.g. Lubembe) indicate a gradual transition in the Katangan basin during the deposition of laterally correlative clastic and carbonate sedimentary rocks exposed in Zambia and in Congo, and are marked by Co:Cu ratios in the range 1:15.

The main Cu–Co orebodies occur at the base of the Mines/Musoshi Subgroup, which is characterized by evaporitic intertidal–supratidal sedimentary rocks. All additional lenticular orebodies known in the upper part of the Mines/Musoshi Subgroup are hosted in similar sedimentary rocks, suggesting highly favourable conditions for the ore genesis in particular sedimentary environments. Pre-lithification sedimentary structures affecting disseminated sulphides indicate that metals were deposited before compaction and consolidation of the host sediment.

The ore parageneses indicate several generations of sulphides marking syngenetic, early diagenetic and late diagenetic processes. Sulphur isotopic data on sulphides suggest the derivation of sulphur essentially from the bacterial reduction of seawater sulphates. The mineralizing brines were generated from sea water in sabkhas or hypersaline lagoons during the deposition of the host rocks. Changes of Eh–pH and salinity probably were critical for concentrating copper–cobalt and nickel mineralisation. Compressional tectonic and related metamorphic processes and supergene enrichment have played variable roles in the remobilisation and upgrading of the primary mineralisation.

There is no evidence to support models assuming that metals originated from: (1) Katangan igneous rocks and related hydrothermal processes or; (2) leaching of red beds underlying the orebodies. The metal sources are pre-Katangan continental rocks, especially the Palaeoproterozoic low-grade porphyry copper deposits known in the Bangweulu block and subsidiary Cu–Co–Ni deposits/occurrences in the Archaean rocks of the Zimbabwe craton. These two sources contain low grade ore deposits portraying the peculiar metal association (Cu, Co, Ni, U, Cr, Au, Ag, PGE) recorded in the Katangan sediment-hosted ore deposits. Metals were transported into the basin dissolved in water.

The stratiform deposits of Congo and Zambia display features indicating that syngenetic and early diagenetic processes controlled the formation of the Neoproterozoic Copperbelt of central Africa.

Keywords: Metallogeny; Copper–cobalt; Sedimentary rocks; Katangan; Central African Copperbelt; Congo-Zambia

Article Outline

1. Introduction

2. Geological setting

3. Lithostratigraphic control of copper–cobalt ores

3.1. Mines Subgroup Congo-type deposits

3.2. Musoshi Subgroup Zambia-type deposits

3.3. Mwashya Subgroup deposits

4. Copper–cobalt distribution in the central Africa Copperbelt

5. Metal and sulphide distribution within the orebodies

6. Ore petrology

6.1. Sulphide parageneses

6.2. Relations between sulphides and gangue minerals

6.3. Relations between orebodies and deformation events

7. Isotopic geochemistry

8. Thermometry

9. Fluid inclusions

10. Discussion

11. Conclusions

Acknowledgements

References

1. Introduction

The Neoproterozoic Katangan Copperbelt of central Africa stretches on both sides of the border between Zambia and Democratic Republic of Congo (DRC; hereafter Congo). It hosts one of the world’s greatest concentration of stratiform copper–cobalt deposits, representing more than half of the world’s mineable cobalt and includes world-class Cu–Co deposits, e.g. Kolwezi, Tenke-Fungurume, Konkola-Chililabombwe, Nchanga, Nkana, Mufulira, each containing 10 Mt copper. Total copper hosted in the Katangan basin of central Africa is close to 200 Mt if sub-economic (Cu 1 wt.%) occurrences are included (data from Gécamines Mining Company for DRC and from Freeman, 1988 for Zambia). Copper and cobalt are associated with iron, and sometimes with anomalous concentrations of other metals (e.g. Ni, U, Ag, Au, PGM, Se, Mo, V, Te, As, Th). The ore is mainly made of disseminated sulphides forming stratiform orebodies hosted in fine-grained siliciclastic or dolomitic sedimentary rocks.

Since the discovery of the Copperbelt in the early 1900s, several metallogenic hypotheses were proposed to explain the primary source of metals and the mineralisation process. The historical review of these genetic theories is given in Sweeney et al., 1991a, Sweeney et al., 1991b and Sweeney and Binda, 1994 for the Zambian Copperbelt. The epigenetic hypothesis suggests the introduction of hydrothermal mineralizing solutions after the deposition, lithification and deep burial of sediments. In this model, the hydrothermal fluids are supposed to originate from the emplacement of granite–granodiorite–tonalite bodies in the Copperbelt (Gray, 1929, Davidson, 1931, Jackson, 1932, Thoreau and du Trieu de Terdonck, 1932, Derriks and Vaes, 1956, Derriks and Oosterbosch, 1958, Darnley, 1960 and Vaes, 1962). The existence of minor sulphide veins or veinlets within a few sediment-hosted copper deposits in Zambia (e.g. Nchanga) and in Congo (e.g. Shinkolobwe) and within a few Zambian granites was taken as a support for this interpretation. However, an unconformable erosional contact occurs between the granitoids and the overlying Katangan sedimentary succession in Zambia (Garlick, 1961a and Binda, 1972). This is supported by U–Pb zircon geochronological data (Armstrong et al., 1999, Rainaud et al., 1999 and De Waele and Mapani, 2002) indicating that the granitoids exposed in the Copperbelt and surrounding areas are older than the Katangan sedimentary succession, i.e. Palaeoproterozoic (2.05–1.65 Ga), Mesoproterozoic (predominantly 1.05–1.0 Ga) or early Neoproterozoic, e.g. 0.88 Ga for the Nchanga granite which is unconformably overlain by the oldest Katangan sedimentary rocks.

Emerging in the 1930s, the syngenetic theory linked the deposition of metals to the deposition of host-sediments (Schneiderhöhn, 1931, Schneiderhöhn, 1932, Schneiderhöhn, 1937, Garlick, 1945, Garlick, 1961b, Garlick, 1967 and Garlick, 1989). Metals were sourced from continental erosion and transported in solution by rivers to the sedimentary depocentres. Ore sulphide precipitation occurred in reducing stagnant water under high bacterial activity and decomposition of organic matter. This hypothesis was based on: (1) the existence of sulphide zonal distribution parallel to the palaeo-shorelines inferred to mark marine transgression–regression events; (2) the coincidence between the polarity of the sulphide zonation and the sedimentary palaeocurrent directions. However, the lack of a systematic correlation between all transgressive/regressive events and lateral/vertical zonation of sulphides, and the discontinuity of the mineralisation within a single lithostratigraphic unit invalidated this model (e.g. Annels, 1974, Renfro, 1974 and Sweeney and Binda, 1994).

Studies related to diagenetic processes in sedimentary rocks triggered the diagenetic model for the central African copper orebodies. Two sulphide generations were documented in the orebodies: (1) the earliest copper–(cobalt)-sulphide generation (hereafter sulphide I) grew during the deposition and the early diagenetic stage of the host-sediments; (2) the second copper–(cobalt)-sulphide generation was inferred to form during a large scale chemical reaction between the host-sediment interstitial water and a metal-bearing brine (Bartholomé, 1962, Bartholomáaaebbb, 1963, Bartholomé, 1969, Bartholomé, 1974 and Bartholomé et al., 1972). However, the model does not address the origin of solutions, the primary source of metals, and the exact timing of mineralisation (early, late diagenesis). These unknowns led to a hydrothermal-diagenetic model linking the mineralizing fluids to late diagenetic hydrothermal fluids of undefined origin (Cluzel and Guilloux, 1986) or originating from mafic igneous rocks or rift related processes (Annels, 1974, Annels, 1979, Annels, 1989, Annels and Simmonds, 1984, Lefebvre, 1989 and Unrug, 1988).

Genesis of sediment-hosted stratiform copper–cobalt deposits, central African Copperbelt

J.L.H. Cailteux^{a, ,} , A.B. Kampunzu^b, C. Lerouge^c, A.K. Kaputo^d and J.P. Milesi^c

^aDépartement Recherche et Développement, E.G.M.F., Groupe G. Forrest International, Lubumbashi, D.R. Congo

^bDepartment of Geology, University of Botswana, Private Bag 0022, Gaborone, Botswana

^cB.R.G.M., BP 6009, 45060 Orléans cedex 2, France

^dDépartement Géologique, GECAMINES, Likasi, D.R. Congo

Received 1 October 2003; 

accepted 1 May 2004. 

Available online 27 October 2005.

 

Abstract

The Neoproterozoic central African Copperbelt is one of the greatest sediment-hosted stratiform Cu–Co provinces in the world, totalling 140 Mt copper and 6 Mt cobalt and including several world-class deposits ( 10 Mt copper). The origin of Cu–Co mineralisation in this province remains speculative, with the debate centred around syngenetic–diagenetic and hydrothermal-diagenetic hypotheses.

The regional distribution of metals indicates that most of the cobalt-rich copper deposits are hosted in dolomites and dolomitic shales forming allochthonous units exposed in Congo and known as Congolese facies of the Katangan sedimentary succession (average Co:Cu = 1:13). The highest Co:Cu ratio (up to 3:1) occurs in ore deposits located along the southern structural block of the Lufilian Arc. The predominantly siliciclastic Zambian facies, exposed in Zambia and in SE Congo, forms para-autochthonous sedimentary units hosting ore deposits characterized by lower a Co:Cu ratio (average 1:57). Transitional lithofacies in Zambia (e.g. Baluba, Mindola) and in Congo (e.g. Lubembe) indicate a gradual transition in the Katangan basin during the deposition of laterally correlative clastic and carbonate sedimentary rocks exposed in Zambia and in Congo, and are marked by Co:Cu ratios in the range 1:15.

The main Cu–Co orebodies occur at the base of the Mines/Musoshi Subgroup, which is characterized by evaporitic intertidal–supratidal sedimentary rocks. All additional lenticular orebodies known in the upper part of the Mines/Musoshi Subgroup are hosted in similar sedimentary rocks, suggesting highly favourable conditions for the ore genesis in particular sedimentary environments. Pre-lithification sedimentary structures affecting disseminated sulphides indicate that metals were deposited before compaction and consolidation of the host sediment.

The ore parageneses indicate several generations of sulphides marking syngenetic, early diagenetic and late diagenetic processes. Sulphur isotopic data on sulphides suggest the derivation of sulphur essentially from the bacterial reduction of seawater sulphates. The mineralizing brines were generated from sea water in sabkhas or hypersaline lagoons during the deposition of the host rocks. Changes of Eh–pH and salinity probably were critical for concentrating copper–cobalt and nickel mineralisation. Compressional tectonic and related metamorphic processes and supergene enrichment have played variable roles in the remobilisation and upgrading of the primary mineralisation.

There is no evidence to support models assuming that metals originated from: (1) Katangan igneous rocks and related hydrothermal processes or; (2) leaching of red beds underlying the orebodies. The metal sources are pre-Katangan continental rocks, especially the Palaeoproterozoic low-grade porphyry copper deposits known in the Bangweulu block and subsidiary Cu–Co–Ni deposits/occurrences in the Archaean rocks of the Zimbabwe craton. These two sources contain low grade ore deposits portraying the peculiar metal association (Cu, Co, Ni, U, Cr, Au, Ag, PGE) recorded in the Katangan sediment-hosted ore deposits. Metals were transported into the basin dissolved in water.

The stratiform deposits of Congo and Zambia display features indicating that syngenetic and early diagenetic processes controlled the formation of the Neoproterozoic Copperbelt of central Africa.

Keywords: Metallogeny; Copper–cobalt; Sedimentary rocks; Katangan; Central African Copperbelt; Congo-Zambia

Article Outline

1. Introduction

2. Geological setting

3. Lithostratigraphic control of copper–cobalt ores

3.1. Mines Subgroup Congo-type deposits

3.2. Musoshi Subgroup Zambia-type deposits

3.3. Mwashya Subgroup deposits

4. Copper–cobalt distribution in the central Africa Copperbelt

5. Metal and sulphide distribution within the orebodies

6. Ore petrology

6.1. Sulphide parageneses

6.2. Relations between sulphides and gangue minerals

6.3. Relations between orebodies and deformation events

7. Isotopic geochemistry

8. Thermometry

9. Fluid inclusions

10. Discussion

11. Conclusions

Acknowledgements

References

1. Introduction

The Neoproterozoic Katangan Copperbelt of central Africa stretches on both sides of the border between Zambia and Democratic Republic of Congo (DRC; hereafter Congo). It hosts one of the world’s greatest concentration of stratiform copper–cobalt deposits, representing more than half of the world’s mineable cobalt and includes world-class Cu–Co deposits, e.g. Kolwezi, Tenke-Fungurume, Konkola-Chililabombwe, Nchanga, Nkana, Mufulira, each containing 10 Mt copper. Total copper hosted in the Katangan basin of central Africa is close to 200 Mt if sub-economic (Cu 1 wt.%) occurrences are included (data from Gécamines Mining Company for DRC and from Freeman, 1988 for Zambia). Copper and cobalt are associated with iron, and sometimes with anomalous concentrations of other metals (e.g. Ni, U, Ag, Au, PGM, Se, Mo, V, Te, As, Th). The ore is mainly made of disseminated sulphides forming stratiform orebodies hosted in fine-grained siliciclastic or dolomitic sedimentary rocks.

Since the discovery of the Copperbelt in the early 1900s, several metallogenic hypotheses were proposed to explain the primary source of metals and the mineralisation process. The historical review of these genetic theories is given in Sweeney et al., 1991a, Sweeney et al., 1991b and Sweeney and Binda, 1994 for the Zambian Copperbelt. The epigenetic hypothesis suggests the introduction of hydrothermal mineralizing solutions after the deposition, lithification and deep burial of sediments. In this model, the hydrothermal fluids are supposed to originate from the emplacement of granite–granodiorite–tonalite bodies in the Copperbelt (Gray, 1929, Davidson, 1931, Jackson, 1932, Thoreau and du Trieu de Terdonck, 1932, Derriks and Vaes, 1956, Derriks and Oosterbosch, 1958, Darnley, 1960 and Vaes, 1962). The existence of minor sulphide veins or veinlets within a few sediment-hosted copper deposits in Zambia (e.g. Nchanga) and in Congo (e.g. Shinkolobwe) and within a few Zambian granites was taken as a support for this interpretation. However, an unconformable erosional contact occurs between the granitoids and the overlying Katangan sedimentary succession in Zambia (Garlick, 1961a and Binda, 1972). This is supported by U–Pb zircon geochronological data (Armstrong et al., 1999, Rainaud et al., 1999 and De Waele and Mapani, 2002) indicating that the granitoids exposed in the Copperbelt and surrounding areas are older than the Katangan sedimentary succession, i.e. Palaeoproterozoic (2.05–1.65 Ga), Mesoproterozoic (predominantly 1.05–1.0 Ga) or early Neoproterozoic, e.g. 0.88 Ga for the Nchanga granite which is unconformably overlain by the oldest Katangan sedimentary rocks.

Emerging in the 1930s, the syngenetic theory linked the deposition of metals to the deposition of host-sediments (Schneiderhöhn, 1931, Schneiderhöhn, 1932, Schneiderhöhn, 1937, Garlick, 1945, Garlick, 1961b, Garlick, 1967 and Garlick, 1989). Metals were sourced from continental erosion and transported in solution by rivers to the sedimentary depocentres. Ore sulphide precipitation occurred in reducing stagnant water under high bacterial activity and decomposition of organic matter. This hypothesis was based on: (1) the existence of sulphide zonal distribution parallel to the palaeo-shorelines inferred to mark marine transgression–regression events; (2) the coincidence between the polarity of the sulphide zonation and the sedimentary palaeocurrent directions. However, the lack of a systematic correlation between all transgressive/regressive events and lateral/vertical zonation of sulphides, and the discontinuity of the mineralisation within a single lithostratigraphic unit invalidated this model (e.g. Annels, 1974, Renfro, 1974 and Sweeney and Binda, 1994).

Studies related to diagenetic processes in sedimentary rocks triggered the diagenetic model for the central African copper orebodies. Two sulphide generations were documented in the orebodies: (1) the earliest copper–(cobalt)-sulphide generation (hereafter sulphide I) grew during the deposition and the early diagenetic stage of the host-sediments; (2) the second copper–(cobalt)-sulphide generation was inferred to form during a large scale chemical reaction between the host-sediment interstitial water and a metal-bearing brine (Bartholomé, 1962, Bartholomáaaebbb, 1963, Bartholomé, 1969, Bartholomé, 1974 and Bartholomé et al., 1972). However, the model does not address the origin of solutions, the primary source of metals, and the exact timing of mineralisation (early, late diagenesis). These unknowns led to a hydrothermal-diagenetic model linking the mineralizing fluids to late diagenetic hydrothermal fluids of undefined origin (Cluzel and Guilloux, 1986) or originating from mafic igneous rocks or rift related processes (Annels, 1974, Annels, 1979, Annels, 1989, Annels and Simmonds, 1984, Lefebvre, 1989 and Unrug, 1988).

سلام دوستان خوبم
منم خوبم
هوا باد گرمیه و کمی مارو عذاب میده  و مدام لباسها رو بهم میریزه
صبح برای رفتن به مدرسه اماده شدم شوشو تا جلوی در همراهم اومد ومنو با بوس و مهربونی بدرقه کرد و از بالای پنجره وقتی من اون پایین منتظر تاکسی بودم نگام میکرد بالاخره من رفتم و بعد از یه مسیر نسبتا طولانی به دبیرستان رسیدم فقط یکشنبه ها مسیرم دوره و من کمی شاید عذاب بکشم پارسال که همون مدرسه میرفتم ماشین میبردم اما امسال ماشین نداریم بععله مدرسه هم با دانش آموزان بد و خوبش و درس خون  درس نخونش گذشت و اما یه شایعه ای توی شهر ما شده و اونم اینه که نوزادی دهان به سخن گشوده و گفته فلان تاریخ شهر ما کن فیکون میشه من که متعجبم از این مردم و از این طرز فکر
راستش دوستان من خیلی سرم شلوغه طوری که واقعا وقت کم میارم واسه کارهام هر چه قدر هم برنامه ریزی میکنم باز باید بدووم اما وقتی داشتم میاومدم از مدرسه یه دفعه سوار یه ماشین شدم که راننده اش از همون ادم مشکوکا بود من بقیه روزها سرویس دارم که از میدون محله مون میاد دنبالمون تا دم مدرسه اما فقط یکی از روزها واسه رفتن به یه مدرسه باید خودم برم که امروز بودش و من خیلی ترسیدم به شوشو زنگ زدم و گفت چرا سوار اون ماشین شدی و پشت  ماشینو از اون پرده سیاه زده بود دور تا دورش عکس د  خ تر و چشماشو و ..بود و از همه بدتر خود راننده که سرتاسر دستشو خالکوبی کرده بود و منو حسابی ترسوند ومن هی صلوات میدادم چون یه مسیر کاملا خلوت بود و من با اون راننده یه مرد دیگه تنها بودم و حسابی ترسیده بودم بالاخره به خیر گذشت . وقتی رسیدم شوشو پیشنهاد داد که من برم خونه مون و خودش میره خونه شون و بعد میخواست بره به مراسم ختم یکی از فامیلاشون و منم قبول کردم و بدون اطلاع رفتم خونه مون و خواهرام از دیدنم خیلی خوشحال شدند و راستی ما در حال تعمیر اساسی خونه مون خونه مامانم اینا هستیم و  ومن حالا تازه رسیدم خونه و شوشو جون هنوز نیومده خیلی کار دارم غذای فردا خوندن و مطالعه و  تصحیح ورقه های بچه ها و از همه مهمتر کارهای دانشگاه خدایا کمکم کن شکرت ای مهربون

راستی فائزه جون دلم برات تنگ شده بهت فکر میکنم ادرستو واسم بذار عزیزم