Università di Genova, Laboratorio di Geochimica

A Conjugated Toop-Samis-Flood-Grjotheim (CTSFG) model is developed by combining the framework of the Toop-Samis polymeric approach with the Flood-Grjotheim theoretical treatment of silicate melts and slags. Electrically equivalent ion fractions are computed over the appropriate matrixes (anionic and cationic) in a Temkin notation for fused salts, and are used to weigh the contribution of the various disproportionation reactions of type: M2/pO(melt)+ 1/2S(gas)+M2/pS(melt)+1/2O2(gas) M2/po(melt)+1/2S2(gas)+3/2O2(gas)-M2/pSO4(melt)v being the charge of the generic Mp-1 cation. The extension of the anionic matrix is calculated in the framework of a previously developed polymeric model (Ottonello et al., 2001), based on a parameterization of Lux-Flood acid-base properties of melt components. Model activities follow the Raoultian behavior implicit in the Temkin notation, without the needs of introducing adjustable parameters. The CTSFG model is based on a large amount of data available in literature and exhibits a satisfactory heuristic capability, with virtually no compositional limits, as long as the structural role given to each oxide holds. The model may be employed to compute gas-melt equilibria involving sulfur and allows computing sulfide and sulfate contents of silicate melts whenever the fugacity of a gaseous sulfur species and oxygen are known. Alternatively, the model calculates the oxidation state of the system (i.e., oxygen fugacity), whenever an analytical determination of either sulfide/sulfate or ferrous/ferric ratios in the melt is provided. Calculated sulfide and sulfate capacities allow the estimates of sulfur abundance in various melts of geological interest, both under anhydrous and hydrous conditions or, alternatively, of fS2, given fO2 and the bulk sulfur content. In this case, fSO2 and fH2S may be eventually computed along the water-sulfur-melt boundary provided fH2O is known.

The Thermodynamics of quasi-chemical and polymeric models are briefly reviewed. It is shown that the two classes are mutually consistent, and that opportune conversion of the existing quasi-chemical parameterization of binary interactions in MO-SiO2 joins to polymeric models may be afforded without substantial loss of precision. It is then shown that polymeric models are extremely useful in deciphering the structural and reactive properties of silicate melts and glasses. They not only allow the Lux-Flood character of the dissolved oxides to be established, but also discriminate subordinate strain energy contributions to the Gibbs free energy of mixing from the dominant chemical interaction terms. This discrimination means that important information on the short-, medium- and long-range periodicity of this class of substances can be retrieved from thermodynamic analysis. Lastly, it is suggested that an important step forward in deciphering the complex topology of the inhomogeneity ranges observed at high SiO2 content can be performed by applying SCMF theory and, particularly, Matsen-Schick spectral analysis, hitherto applied only to rubberlike materials.

The bulk solubility and speciation of H2O in silicate melts of virtually any composition is predicted from first principles with a satisfactory precision. The solubility of molecular water is first predicted from the Scaling Particle Theory (SPT) coupled with an ab initio assessment of the electronic, dispersive and repulsive energy terms based on the Polarized Continuum Model (PCM). The Silver-Stolper ideal homogeneous speciation model is then applied to compute the fractional molar amount of neutral hydroxyl functionals [OH]0 in the melt and the computed [OH]0 amount is added to the molecular form [OH2]. The Hydrogen Bonding (HB) electrostatic contributions to the stabilization of molecular water [OH2] in solution are then resolved through an inverse non-linear minimization procedure on the basis of an extended dataset (970 samples) of experiments concerning the H2O saturation hyper- surface carried out in the last half century. The Gibbs free energy of solution ΔGS, the ΔGHB contributions and the energy terms involved in the homogeneous speciation reaction are shown to be consistent with first principles. The procedure is fully predictive (i.e. no need of an initial hint about approximate bulk amounts of H2O in solution) and sufficiently accurate to be proposed as an exploratory tool (mean absolute accuracy of ~2.1 kJ/mol in terms of energy and ~1.7% in terms of fractional molar amount XH2O per unit mole of liquid).