Adaptation of the SELECTOR-C program package for solving petrogenetic problems of metamorphic rocks

The paper presents thermodynamic models for mineral solid solutions used in physicochemical simulations with the SELECTOR-C program package (PP) in application to metamorphic mineral-forming processes. It is demonstrated that the simulated FeO and MgO distribution in the mineral pairs garnetbiotite, garnet-orthopyroxene, orthopyroxene-biotite, orthopyroxene-olivine, garnet-cordierite, garnetclinopyroxene, and clinopyroxene-orthopyroxene in the model samples satisfactorily corresponds to available experimental and empirical data. Simulations of naturally occurring mineral associations are employed to demonstrate the capabilities of the new version of the SELECTOR-C PP as a tool for studying the evolution of mineral assemblages at varying P-T conditions and fluid regime, the perfectly mobile and inert behaviors of certain fluid components during the origin of mineral associations are demonstrated, the pseudosection method applied over a broad P-T range is used to trace systematic variations in the composition of mineral associations in granulite-facies metabasites and metapelites, and the upper limit of plagioclase stability is estimated for these rocks at pressures of 11-12 kbar. Principal differences are elucidated in the effect of rocks rich and poor in Fe3+ on the percolation of metamorphic fluid through them: Fe3+-rich rocks retain their own redox potential at a certain level by buffering reactions, whereas Fe3+-poor rocks rapidly exhaust their buffer capacity and acquire the redox potential of the inflowing external fluid. This allowed us to evaluate the logfO(2) at no higher than -17 (at T = 700A degrees C and P = 6.8 kbar). Our simulation of the equilibrium of natural rock samples provides good reasons to believe that natural mineral assemblages can be formed at low fluid/rock ratios of no higher than 0.01-0.06.