Gold in sulfide fluids revisited

Abstract

Gold solubility was measured at temperatures of 350, 400, 450, and 490 °C and pressures of 500 and 1000 bar in an ’oxidized sulfide’ system, as a function of pHT (2 – 10) and sulfur concentration (m(Stotal) = 0.03 – 1.2 [mol·(kg H2O)-1]). In this system, sulfur primarily exists as H2S, H2SO3, H2SO4, their dissociation products, and radical species such as S2- and S3-. The complexes Au(HS)2-, Au2S22-, AuHS(aq), AuHS(H2S)3(aq), and AuOH(aq) were identified as the primary gold species in the experimental fluids, varying with pH and m(Stotal). The solubility constants for Au(HS)2-, a critical (hydro)sulfide complex, align excellently with literature for ’reduced sulfide’ fluids, where sulfur predominantly exists in the 2- oxidation state. New experimental data from the ’oxidized sulfide’ system were regressed along with reliable literature data from ’reduced sulfide’ systems to calculate standard thermodynamic properties and parameters of the Helgeson-Kirkham-Flowers (HKF) model. The solubility constants for charged complexes, Au(HS)2- and Au2S22-, increase sharply with temperature, whereas those for neutral species, AuHS(aq) and AuHS(H2S)3(aq), show a pronounced peak near 300 °C. These (hydro)sulfide complexes account for gold solubility ranging from a few tens of ppb to a few tens of ppm in natural sulfide fluids, depending on the fluid pH. Thermodynamic calculations also indicate that, in addition to (hydro)sulfide species and the hydroxide complex, AuCl2- significantly contributes to Au mobility in high-temperature acidic chloride fluids. Based on new experimental data and prior studies using solubility and X-ray absorption spectroscopy methods, other gold complexes including mixed Au-HS-Cl, Au-HS-S3- species, and complexes with alkali metal cations, are deemed redundant. Above 250 °C, the influence of chloride salts on gold solubility can be accurately modeled using a simple extended Debye-Hückel equation with the term bγ·I = 0. The Setchenov coefficient bn = 0 suffices for calculating the activity coefficients of neutral species. This streamlined thermodynamic model aligns closely with earlier experimental work by Terry Seward and his team, effectively describing the state of Au in all types of natural fluids where Au exists in the 1+ oxidation state, under any set of P-T-f(O2)-f(S2)-compositional parameters.