The Oka Carbonatite Complex, Quebec: Aspects of Carbonatite Petrogenesis.

Despite intensive study of carbonatites, there remain uncertainties about the original compositions of carbonatite magmas, the origins of carbonatites (including many proposed mantle and crustal sources), and the potential importance of physical processes, including fractional crystallization and separation of immiscible liquids. To investigate these questions, thermodynamic calculations and physical models were applied to a well-exposed carbonatite, the Oka complex (Quebec), which is a composite hypabyssal pluton including intermixed silicate (ijolite) rocks and carbonatite. Chemical processes and equilibria in carbonatite magmas can best be studied by thermodynamic calculations, once minerals in buffering assemblages are chemically analyzed. Essential to thermodynamic calculations is a consistent thermochemical data base, so free energies and entropies of minerals in CaO-SiO(,2)-CO(,2), which includes potentially important carbonatite phases, were derived by evaluation of phase relations and thermochemical measurements. Calculated equilibria in CaO-SiO(,2)-CO(,2)-H(,2)O suggest (contrary to previous speculation) that spurrite is not important in carbonatite petrogenesis, and allow determination of the vapor phase composition for a carbonatite dike at Oka. Calcite-dolomite thermometry is consistent with an inferred eutectic at 640(DEGREES)C, 1 kbar, and mineral equilibria imply that the vapor was water-rich {X(H(,2)O) = 0.88, X(CO(,2)) = 0.11} with oxygen fugacities near QFM. Equilibrium immiscibility between silicate and carbonate melts has been suggested as the origin of the silicate and carbonate rocks at Oka, and immiscibility can be tested by field relations, by comparison with experiments, and by thermochemical constraints. Intimately associated silicate and carbonate rocks contain possible ocelli of each other, and compositions are similar to experimentally produced immiscible melts. The most stringent test of immiscibility is comparison of mineral assemblages and composition, both of which must be identical in rocks derived as equilibrium immiscible fractions. Mineral assemblages are identical in the associated rocks, and minerals (nepheline, melilite, pyroxene) have similar compositional ranges. Physical processes in carbonatites cannot be considered without knowledge of carbonate melts, which are inferred to be ionic liquids (e.g., NaCL). Viscosity near that of water implies rapid crystal settling (< 0.25) cm/sec for 0.01 cm. diam. grains), and crystal cumulation will not yield the coarse-grained rock seen at Oka. In situ growth of crystals is a likely alternative mechanism. . . . (Author's abstract exceeds stipulated maximum length. Discontinued here with permission of author.) UMI