On the Lux-Flood basicity of melts in solving their chemical reactivity

Reactivity of silicate melts is due to charged functional groups (cations, free anions and polymeric units or structons), which govern mutual interactions between constituting oxides. The main problem arises when defining thermodynamic oxide ion activities. This difficulty is overcome if we adopt the Fincham and Richardson (1954) formalism coupled with a Toop and Samis (1962a,b) polymeric description of the anion matrix, based on the principle of equal reactivity of co-condensing groups and involving singly bonded, doubly bonded and free oxygen, that is oxygen species in three different polarization states. In a chemically complex melt the capability of transferring fractional electronic charges from the ligand to the central cation depends in a complex fashion on the melt structure, affecting the polarization state itself. Therefore, the mean polarization state of the various ligands (mainly oxide ions in silicate melts) and their ability to transfer fractional electronic charges to the central cation are conveniently represented by the optical basicity of the medium. On this basis, Ottonello et al. (2001) related linearly the optical basicity to the extent of the anionic matrix and hence to the polymerization constant. Although the adopted functional form is rather brutal, not accounting for temperature effects, it allows an accurate description of the anionic matrix which enable us to study the iron oxidation state in both anhydrous and hydrous melts, then solving some controversies present in literature, to define a consistent model for water speciation which accounts for the amphoteric behavior of this component and to assess sulfur speciation and solubility. The latter, in particular, was possible only through the adoption of the Flood and Grjotheim (1952) thermochemical cycle, which accounts for the standard state transposition between the Temkin standard state of completely dissociated components and that of pure component at T and P of interest. Recently, we applied the Hybrid Polymeric model (Ottonello, 2001) to assess silicate melt energetics, distinguishing chemical interaction terms from strain energy contributions (Ottonello and Moretti, 2004). Lux-Flood basic oxides give rise to purely endothermic effects when admixed to silica along simple bianry joins, whereas acidic Lux-Flood oxides originate thermal admixtures and amphoteric oxides promote both enthalpic and entropic (non configurational) chemical interactions. This makes the Lux-Flood acid-base character of the various oxides consistent with experimental determinations of nephelauxetic properties of the limiting oxide components in the mean donor ligand field, represented in terms of optical basicity. A linear proportionality is observed between endothermic heat of mixing and optical basicity which allows us to predict the polymerization extent in molten MO-SiO2 binaries. The extension of this proportionality to complex systems requires the application of the Flood-Grjotheim (1952) approach and allows us to shift the previously developed models for iron, water and sulfur to a more rigorous treatment of the anionic matrix of silicate melts. REFERENCES Fincham and Richardson (1954) Proc. Roy. Soc. London, 223A, 40. Flood and Grjotheim (1952), J. Iron Steel Inst., 171, 64. Toop and Samis (1962a) Can. Met. Quart., 1, 129. Toop and Samis (1962b) Trans. Met. Soc. AIME, 224, 878. Moretti (2003) Ann. Geophys. (submitted). Moretti and Ottonello (2003a) Metall. Mat. Trans. B, 34B, 399. Moretti and Ottonello (2003b) J. Non Cryst. Sol., 323, 111. Ottonello (2001), J. Non Cryst. Sol., 282, 72. Ottonello and Moretti (2004) J. Phys. Chem. Sol. (submitted). Ottonello et al. (2001), Chem. Geol., 174, 157.