EXPERIMENTS, EQUILIBRIA, AND ORE METALS: FROM MAGMAS TO PORPHYRY SYSTEMS

In the Laboratory for Mineral Deposits Research, hydrothermal cold-seal methodologies, involving both rapid and conventional quench, have been used for over 20 years to determine Nernst-type partition coefficients (D) for ore and related elements in magmatic systems. Silica tube (ST), and most recently, diamond cell techniques, have also been used. Experiments have employed noble metal (Au, Pt, Ag/Pd) capsules (or ST) loaded with subsets of the polyphase assemblage: silicate melt, aqueous vapor, brine, sulfide melt, sulfide/oxide mineral assemblages, and rock-forming minerals. Important issues in experimentation involve: attaining analytical, but geologically meaningful, concentrations of target (ore) elements; minimizing or taking advantage of interactions between the charge and the capsule; attaining sufficient phase separation for analysis; quenching vs. formation of synthetic fluid inclusions; attainment of equilibrium; and testing hypotheses for equilibria comprising independently variable components of phases.

For example, if the D for Cu partitioning between pyrrhotite (po) and melt (m) is studied without consideration of the controlling equilibria, then results may be misapplied. The following example is drawn from three studies performed at Maryland: First, Lynton et al. (1993 Econ Geol p. 901) determined D (po/m) = 550± 220, @ lgfO2 = -13.4, lgfS2 = -2; Then, Jugo, et al., (1999, Lithos, p. 573) hypothesized that: ½ S2 + CuO0.5(m) + FeS (po) = CuFeS2 (po)+ ¼ O2 was a controlling equilibrium in the system, suggesting a functional dependence for D upon the fugacities for O2 and S2, and found D= 2600 ± 300±107, @ lgfO2 = -14.3, lgfS2 = -1; Finally, in the present study Simon et al., (in review, GCA) found D = 174±107, @ lgfO2 = -14, lgfS2 = -3. These values are consistent with the above equilibrium, and show that D's cannot be evaluated independent of equilibria, and that equilibria must be written in terms of independently variable components. These variations represent an order of magnitude difference in the partition coefficient over geologically reasonable fugacities of O2 and S2. Further, given that the fraction of po in the hypersolidus assemblage can be on the order of 0.001, variations of D around 1000 are critical in the sequestering of Cu in magmatic systems.