Siderophile and chalcophile metals as tracers of the evolution of the Siberian Trap in the Noril'sk region, Russia

Abstract

In this study Cu, Ni, and platinum-group elements (PGE) were determined in a sequence of basaltic and picritic lavas from the Siberian Trap in the Noril'sk area of Russia to constrain genetic relationships between the basalts and the petrogenesis of Ni-Cu-PGE sulfide deposits associated with the Talnakh and Noril'sk intrusions. In the most primitive basalts (8–19 wt% MgO) of the Tuklonsky (Tk) suite, Pt and Pd concentrations range from 4–13 ppb, increasing with decreasing MgO content; whereas Ir contents decrease with MgO from 0.8-0.05 ppb. The contrasting behavior of these elements, which all have very high sulfide-silicate partition coefficients, as well as the primitive mantle-like ratios of Cu/Y and Pd/Y, suggests that these magmas were not sulfide-saturated. The high PGE abundances imply that their parental magmas were also not sulfide saturated during partial melting in the mantle. Due to sulfide segregation, the overlying basalts of the Nadezhdinsky (Nd) series are low in Cu and Ni (52 and 38 ppm, respectively); highly depleted in all PGE; and have very low Cu Y , Pd Y , and Pd Cu ratios. However, in stratigraphically higher levels, Cu, Ni, and PGE concentrations increase systematically through the Morongovsky (Mr) suite to reach a concentration plateau in the uppermost Mokulaevsky (Mk) suite (Pt 8 ppb; Pd: 9 ppb; Ir: 0.12 ppb; Rh: 0.4 ppb). At the same time, ratios such as Cu Y increase and approach primitive mantle values. However, ratios involving PGE, such as Pd Y , remain low, suggesting the removal of small amounts of sulfide (0.01–0.03%). The compositional variations in the basalts and the sulfide liquids can be quantitatively described by fractional segregation of a sulfide liquid in an open- or closed-system magma chamber. The latter model suggests that the basalts represent the eruption products of a zoned magma chamber in which light magma, with crustal components contaminated, overlies less contaminated, denser magma. Crustal contamination caused sulfide saturation, and the resulting sulfide liquids settled through a magma column and accumulated at the bottom of the chamber. In this model, the sulfide liquid is not in equilibrium with the whole magma mass, and sulfide segregation is compared with the zone-refining process of metallurgy. The sulfides become more enriched as they move through the magma; and although the magma left behind is depleted in PGE, Cu, and Ni, their concentrations also increase with depth. Eventually, the magma chamber is emptied from the top to the bottom, producing the flood basalt sequence and the associated intrusions and ore deposits. In the open-system model, sulfide saturation was initially caused by assimilation of crustal material by the Tuklonsky magma. Continuous and simultaneous replenishment, assimilation, and crystallization processes formed the lower Nd lavas. The concurrent removal of 0.5-1% sulfide strongly depleted these magmas in chalcophile and siderophile metals. Due to the continuous replenishment of the magma chamber with uncontaminated PGE-rich magma, succeeding lavas (Mr, Mk) show diminishing signs of crustal contamination and become less sulfide-saturated, as indicated by the increasing Ni, Cu, and PGE abundances. During the evolution of the chamber, the magma remained sulfur-saturated, and sulfides accumulated at the base. The composition of the sulfide ores could be regarded as a mixture consisting of low Ni-, Cu-, and PGE-sulfides derived with a low silicate/sulfide ratio (100) from the Tk-Nd magma and high Ni-, Cu-, and PGE-sulfides formed with a high ratio (10,000) from the Mr-Mk magma.