Origin of Variations in Nickel Tenor Along the Strike of the Edwards Lode Nickel Sulfide Orebody, Kambalda, Western Australia

Abstract

Nickel tenor (Ni concentration in 100% sulfides) varies from 12.5 to 14.5 wt% on the eastern flank, and from 16 to 18 wt% on the western flank of the Edwards lode orebody, Kambalda, Western Australia. The center of the orebody exhibits a large internal variation in tenor, from 9.6 to 18.7 wt% Ni. Tenor variation of this magnitude within a single orebody has not previously been documented at Kambalda. Three models were evaluated for the observed variation. (1) Variation in R factor. This model is supported by a higher nickel tenor within the matrix ores than the massive ores. Pyrolite-mantle-normalized abundances of the platinum-group elements (PGE) are consistent among massive, matrix and disseminated ores. This does not support an R-factor model for tenor variation, because sulfide liquid / silicate melt partition coefficients for the PGE are higher than for Ni (D PGE >> D Ni ). (2) Variations in oxygen fugacity f(O 2 ). The f(O 2 ) control on Ni tenor is not supported by geochemical or petrographic observations. The proportion of magnetite is relatively consistent within the orebody, demonstrating no correlation with Ni tenor.(3) Partial redistribution and sulfidation of the nickel ores during upper-greenschist-to lower-amphibolite-facies metamorphism. Evidence includes increased pentlandite abundance correlating with an increase in the abundance of secondary metasomatic pyrite. Some pentlandite grains demonstrate a genetic relationship with carbonate veining, indicating nickel mobility. The large variation in Ni tenor observed within the central domain of the Edwards lode orebody correlates with the ore surface that exhibits the greatest degree of metamorphic heterogeneity. A laser-ablation ICP-MS study of the PGE distribution within sulfides shows that the Ir-group PGE (IPGE) are homogeneously distributed, in contrast to the Pt-group PGE (PPGE). The distribution of Pd reflects the distribution of pentlandite, and the other PGE (Rh, Ru, Pt) and Au seem controlled by PGM and Au-bearing phases. The IPGE show a strong primary magmatic control with distance along the strike of the orebody. Nickel does not demonstrate a correlation with the IPGE, indicating that post-volcanism processes of remobilization partly control the distribution of Ni along strike. SOMMAIRELes teneurs en nickel (concentration en nickel sur 100% de sulfure) varient de 12.5 à 14.5% (poids) sur le flanc est, et de 16 à 18% sur le flanc ouest du gisement d'Edwards, à Kambalda, en Australie occidentale. Le centre du gisement montre des variations internes importantes des teneurs en nickel, depuis 9.6 jusqu'à 18.7%. De telles variations en teneur n'avaient jamais été 656 THE CANADIAN MINERALOGIST décrites auparavant à Kamblada. Trois modèles ont été testés pour évaluer les variations en teneur observées. (1) Variation du facteur R. Ce modèle est en accord avec les teneurs plus élevées des minerais en matrice par rapport aux minerais massifs. Les abondances en éléments du groupe du platine, EGP, normalisées par rapport à un manteau pyrolitique, son...

Oxygen solubility and speciation in sulphide-rich mattes

“…The spinel is localised in the sulphide matrix adjacent to komatiite inclusions in the massive ore and it is concentrated near, but not on, sulphide-komatiite contacts. Subsequent observations by Heath et al (2001) showed that, in addition to the Cr-rich spinel described by Frost and Groves (1989), primary magnetite is found in the matrix ore (1-9 wt%) and to a lesser extent in the massive ore (less than 1 wt%). Magnetite enrichment is more pronounced along the boundary between massiveand disseminated-ore types (Woolrich et al, 1981;Heath et al, 2001), which is explained by Naldrett (1969) as the result of fO 2 enhancement due to differences in the diffusion coefficients of O in mattes and silicate melt.…”

“…Subsequent observations by Heath et al (2001) showed that, in addition to the Cr-rich spinel described by Frost and Groves (1989), primary magnetite is found in the matrix ore (1-9 wt%) and to a lesser extent in the massive ore (less than 1 wt%). Magnetite enrichment is more pronounced along the boundary between massiveand disseminated-ore types (Woolrich et al, 1981;Heath et al, 2001), which is explained by Naldrett (1969) as the result of fO 2 enhancement due to differences in the diffusion coefficients of O in mattes and silicate melt. These observa- Nagamori and Kameda (1965) 59 All b 1.05-2.6 9.4-12.0 1300 H extraction Kaiser and Elliott (1986) 71 None 1.62-2.98 8.33-9.56 1195 H extraction Somsiri and Gaskell (1995) 36 None 1.6-2.3 8.5-13.72 1300 IR analysers Blatov et al (1997) 25 All c 1.06-2.26 7.53-11.45 1300 Unknown Kress (1997) 15 2 1.24-3.05 8.5-13.72 1097-1348 EMP Brenan and Caciagli (2000) 6 3 2.06-4.0 8.1-10.1 1300 EMP Nagamori and Yazawa (2001) 12 None 2.65-3.48 9.6-12.0 1200 EMP Rose and Brenan (2001) 29 7 1.6-4 8-11.1 1300 EMP Andrews and Brenan (2002) 6 None 1.7-3 8.1-8.6 1300-1400 EMP Brenan and Rose (2002) 11 4 1.6-1.9 8.1-9.5 1300 EMP Mungall et al (2005) Nagamori and Yazawa, 2001;Kress, 1997, respectively), only the dataset of Nagamori and Kameda (1965) plots at consistently higher O concentration than other datasets, probably due to an analytical bias.…”

“…Other ore deposits, do not display such a clear picture with regards to the fate of the matte's O abundances. For example, the Kambalda massive Ni-sulphide deposit features, in addition to rare disseminated magnetite (Heath et al, 2001), has Cr-rich magnetite within the massive sulphide ore, in close proximity to komatiitic melt inclusions (Frost and Groves, 1989). Moreover, in some sections of the Sudbury igneous complex (e.g.…”

“…The magnetite present in the ore as well as the Cr-rich spinel, are thought to be primary in origin (Frost and Groves, 1989;Heath et al, 2001). It is necessary to characterise the mechanisms that allowed the crystallisation of these two different types of primary magnetites in order to understand how O behaves during the fractional crystallisation of sulphide-rich mattes.…”

The Alkaline Solution to the Emergence of Life: Energy, Entropy and Early Evolution

“…The metal ions in this ocean were derived from the main two types of submarine crust. The dominant and primordial crust comprised silicates (MgO [[ FeO [ CaO$Al 2 O 3 i.e., Mg-rich basalt, komatiite and peridotite) with dispersed and interspersed pyrrhotite (Fe 7 S 8 ), pentlandite ((Fe,Ni) 9 S 8 ) and pyrite (FeS 2 ) (Lesher and Stone 1997;Heath et al 2001). Awaruite (Ni 3 Fe) filaments would also have been a feature of the serpentinized ocean crust (Krishnarao 1964).…”