2002 Denver Annual Meeting (October 27-30, 2002)

Paper No. 11
Presentation Time: 4:00 PM

GOLD SOLUBILITY IN MAGNETITE


SIMON, Adam C.1, CANDELA, Philip A.1, PICCOLI, Philip M.1, PETTKE, Thomas2 and HEINRICH, Christoph A.2, (1)Laboratory for Mineral Deposits Research, Department of Geology, University of Maryland, College Park, MD 20742, (2)Isotope Geochemistry and Mineral Resources, Federal Institute of Technology, ETH Zentrum NO, CH-8092, Zurich, Switzerland, asimon@geol.umd.edu

The orthomagmatic hypothesis for the formation of Au-rich porphyry-type ore deposits includes all chemical and physical processes operative during the evolution of ore-generative magma (melt + crystals + magmatic volatile phase (MVP)).  In the literature, attention is continuously given to processes that occur after a melt saturates with and exsolves a MVP, the magmatic Au-transporting entity responsible for focusing Au deposition.  Whereas the physico-chemical properties of the MVP exert the dominant control on the efficiency with which Au may be transported from a given melt, the more fundamental control on the total tonnage of Au in any subsequently formed mineralized rock is the absolute quantity of Au available in the melt at the time of MVP saturation.  Any crystallizing phase that depletes the melt in Au prior to MVP saturation can potentially reduce the ultimate total Au tonnage of mineralized rock.  To quantify the effect that magmatic magnetite (Mt) can have on the Au budget of a magmatic system, we grew synthetic Mt crystals, from 4 to 12 µm, in a S-free, Au-saturated haplogranite-vapor-brine-magnetite-gold capsule system at 800ºC, 140 MPa and fO2 ~ NNO.  Au concentrations in Mt, floating in silicate glass, were quantified by <i>in-situ</i> LA-ICP-MS, using a 193nm ArF Excimer laser with a homogenized energy beam profile.  Individual Mt crystals were drilled out of glass and the mixed signal was deconvoluted as per Halter <i>et-al</i> (2002, Chem. Geol. 183, 63-86).  Au concentrations in Mt vary from 1.6 to 13.5 ppm.  Analytical signals with a homogeneous Au distribution cluster about a mean of 2.5 ppm.  Higher Au concentrations may reflect chemisorption onto the Mt surface as evinced by distinct Au peaks.  Using an accepted Au concentration in felsic melt, at saturation, on the order of 1 ppm (Frank <i>et-al</i>., 2002, GCA, in press) yields a Mt-melt partition coefficient, Kd, on the order of 2.5.  A lower Au concentration in melt, as evinced by LA-ICP-MS, indicates that this Kd may be a minimum.  Although relatively low, this Kd indicates that Mt can sequester ~11% of the total Au content of a sulfide-undersaturated, oxidized calc-alkaline melt experiencing 30% isobaric Rayleigh crystallization.  Our data suggest that crystallization of Mt can have a moderate effect on the Au budget of an evolving porphyry system.