Evidence for SiO2-NaCl complexing in H2O-NaCl solutions at high pressure and temperature

摘要

Experimental studies reveal complex dissolution behavior of quartz in aqueous NaCl solutions at high temperatureand pressure, involving variation from salting-in to salting-out that changes with temperature, pressure, and salt concentration. The behavior is not explainable by traditional electrostatic theory. An alternative hypothesis appeals to complexing of SiO2 with NaCl and can explain the observations. However, the hypothesis of complexing, as previously applied, is inadequate in several respects: it neglects polymerization of solute silica, regards the SiO2-NaCl hybrid complex(es) as anhydrous, which seems unlikely, and invokes an incorrect stoichiometry of the hydrated silica monomer, now known to be Si(OH)4•2H2O. These neglected features can be incorporated into the complexing model in a revised formulation based on a simple thermodynamic analysis using existing quartz solubility data. The analysis leads to a quasi-ideal solution model with silica monomers, dimers, and two distinct hydrous SiO2-NaCl hybrid complexes with overall NaCl:H2O=1:6, one Na-bearing and one Cl-bearing. Their (equal) molar concentrations (Xhc) are governed by a pressure- and temperature-dependent equilibrium constant, Khc=Xhc2/(aNaclaH2O6), where aNacl and aH2O are the respective activities of the solvent components. The stability of the hybrid complexes (i.e., their concentration) is very sensitive to H2O activity. The entire set of experimental quartz-solubility data at 700°C, 1–15kbar, is reproduced with high fidelity by the expression logKhc=−4.585+0.2691P−2.023×10−3P2 (P is pressure in kbar), including the transition from low-pressure salting-in to high pressure salting-out. The results indicate that hybrid SiO2-NaCl complexes are the main hosts for dissolved silica at NaCl concentrations greater than 6 wt%, which are likely common in crustal fluids. At higher temperatures, approaching the critical end point in the system SiO2-H2O, the model becomes progressively inaccurate, probably because polymers higher than the dimer become significant as SiO2 concentration increases.