Cr-spinel/olivine and Cr-spinel/liquid nickel partition coefficients from natural samples

The Nernst partition coefficient of nickel (DNi) between Cr-spinel and silicate melt in natural systems has been investigated using mid-ocean ridge basalts (MORB) and other volcanic rocks. The Cr-spinel/olivine DNi values in volcanic rocks are between 1.2 and 0.3, indicating that the Cr-spinel/liquid DNi values vary from slightly higher to significantly lower than the olivine/liquid DNi values in natural systems. The Cr-spinel/liquid DNi values from the MORB samples vary between 6 and 11, slightly higher than those from the S-bearing experiments of Satari et al. [Satari P., Brenan J. M., Horn I. and McDonough W. F. (2002) Experimental constraints on the sulfide- and chromite–silicate melt partitioning behavior of rhenium and platinum-group elements. Economic Geology97, 385–398]. The results of the MORB samples and the experiments of Satari et al. (2002) indicate a negative correlation between the Cr-spinel/liquid DNi and the XCr values in Cr-spinels (Cr cation number on the basis of 3 total cations in the spinel structure). Variations of Cr-spinel/liquid DNi values with Cr-spinel compositions can be estimated from an empirical equation based on the results of the MORB samples and the experiments by Satari et al. (2002). The choice of Cr-spinel/liquid DNi = 10 for numerical modeling by Righter et al. [Righter K., Leeman W. P. and Hervig R. L. (2006) Partitioning of Ni, Co, and V between spinel-structured oxides and silicate melts: importance of spinel composition. Chemical Geology227, 1–25] is reasonable for basaltic systems. For picritic and komatiitic systems a lower value of ∼5 is more appropriate.