A model for the saturation of C-O-H-S fluids in silicate melts

Abstract

Abstract The behaviour of volatile components in magmas is crucial for magmatic and volcanic processes, from the deep regions of magma generation and storage to the shallow regions of magma eruption and emplacement. Water, carbon dioxide, and sulphur compounds are the main volatile components in natural magmas, generally comprising more than 99% of the volcanic gases released before, during, and after eruption. We have set up a method to calculate the chemical equilibrium between a fluid phase in the C-O-H-S system and a silicate liquid with a composition defined by ten major oxides. The method is based on previous models for the saturation of H 2 O-CO 2 fluids (Papale 1997, 1999) and sulphur solubility (Moretti 2002) in silicate liquids, and for the fugacities of components in fluids with complex composition (Belonoshko et al. 1992). The model calculations provide estimates of the partitioning of H 2 O, CO 2 and S between the silicate liquid and the coexisting fluid, and the composition of the fluid phase in terms of H 2 O, CO 2 , SO 2 , and H 2 S, as a function of pressure, temperature, volatile-free liquid composition, oxygen fugacity, and total amount of each volatile component in the system. Model calculations are presented for silicate liquids with tholeiitic and rhyolitic composition, oxygen fugacities in the NNO ± 2 range, and pressures from a few hundred MPa to atmospheric, with the simplifying assumption that no reduced or oxidized sulphur-bearing solid or liquid phases nucleate or separate from the liquid-gas system. The results are in good agreement with the bulk of experimental data from the literature, and show the well-known minima in sulphur saturation contents as a function of oxygen fugacity, the mutual effects of volatiles on their saturation contents, and the complex relationships between the saturation surface of multi-component fluids and the liquid composition, volatile abundance, and pressure-temperature-oxygen fugacity conditions. The new model therefore represents a powerful tool for the prediction of multi-component gas-liquid equilibria in natural magmatic systems, for the simulation of magmatic and volcanic processes, and for the interpretation of data on the degassing of magma bodies and composition of volcanic plumes, provided that the assumptions on which the model rests, first of all that no separation of additional S-bearing phases occurs, are satisfied.