Fluid inclusions from the epigenetic Zn-Cu deposit of Kipushi, central African copper belt, Katanga, Democratic Republic of Congo, have been subjected to an integrated microthermometric, bulk crush-leach and laser ablation inductively coupled plasma mass spectrometric (LA-ICP-MS) study to elucidate the physicochemical conditions of ore formation and the origin of the mineralizing fluids. The Kipushi deposit is hosted in carbonates and partly in siliciclastic rocks that have undergone regional, low-grade greenschist facies metamorphism and which were strongly deformed during the Lufilian orogeny (main deformation phase ~550 Ma). Investigation of host-rock, gangue, and ore samples enabled reconstruction of the paragenesis of ore deposition.
The main Zn-Cu (+Ge, Pb) sulfide mineralization was accompanied by three generations of hydrothermal dolomite and a generation of coarse-grained quartz. Primary and pseudosecondary fluid inclusions in these minerals contain predominantly halite-saturated, aqueous brines (~30–43 wt % NaCl equiv), with locally unidentified trapped solids. These inclusions mostly decrepitate upon heating before total homogenization, which could suggest that they also trapped excess carbonic vapor. They show petrographic and microthermometric evidence for postentrapment reequilibration, prohibiting their use as accurate geothermometers and geobarometers. Previous studies on chlorite and sphalerite geothermometry framed the formation temperature of the Kipushi Cu-Zn deposit between ~290° to 380°C. Younger fluids are present as secondary inclusions in the gangue minerals and as primary fluid inclusions in sphalerite from a minor secondary mineralization phase. These two-phase aqueous inclusions have a slightly lower salinity (~23–31 wt % NaCl equiv) and have homogenization temperatures ranging from ~170° to <80°C in successive fluid inclusion assemblages. Sporadically, some mixed aqueous-carbonic and gaseous CO2 + CH4 inclusions are present, but their genetic position is un-clear and they are insignificant with regard to the mineralization process as a whole.
Crush-leach (bulk) analysis of the fluid inclusions show that the mineralizing fluids had a molar Cl/Br ratio close to, or somewhat higher than, the seawater value, which rules out a major contribution from bittern brines and suggests that the fluids gained their salinity primarily by the dissolution of evaporites. Younger fluids that circulated through the deposit after the main phase of ore formation have Cl/Br and Na/Br signatures indicative of a greater involvement of halite-dissolution brines (values much higher than that of seawater). LA-ICP-MS analyses of the mineralizing fluids reveal that they have molar Li/Na, Ca/Na, Mg/Na, Mn/Na, and Sr/Na in the range of natural high-salinity fluids in metamorphic environments worldwide. However, they have relatively high molar Fe/Na, Zn/Na, Pb/Na, and especially very high Ba/Na ratios. Their molar K/Na is also somewhat higher than for other low-grade metamorphic fluids but is nevertheless in agreement with the values predicted by the K/Na geothermometer at the conditions of ore formation. This suggests that the major element composition of the fluids was buffered by fluid-mineral equilibria. Their relatively high metal contents (Ba, Zn, Fe, and Pb) suggest that they interacted with and leached these elements from felsic basement rocks at higher temperatures.
Whatever the origin of the fluids, their salinity and the metals, their uniformly high Ba (and ore metal) content together with the absence of barite and the occurrence of Ba silicates at Kipushi indicates that sulfide mineralization occurred when the metal-bearing fluids encountered a reducing, H2S-rich reservoir. The ubiquitous presence of preexisting, metamorphosed carbonic material (shungite) in the sulfides and host rock, and a sulfur isotope signature of the ores in agreement with thermogenic sulfate reduction, suggests that this condition was likely created due to the high initial reduced-carbon content of the host rocks at Kipushi, which seems to be exceptional in the copper belt. Therefore, we postulate that a critical factor in the formation of major, vein-type Zn-Cu mineralization was the presence of a hydrocarbon-rich facies of the lower Nguba carbonates at Kipushi.