Pontificia Universidad Católica de Chile Pontificia Universidad Católica de Chile
Oliveira I.L., Brod J.A., Cordeiro P.F.O., Dantas E.L., Mancini L.H. (2017)

Insights into the late-stage differentiation processes of the Catalão I carbonatite complex in Brazil: New Sr–Nd and C–O isotopic data in minerals from niobium ores

Revista : Lithos
Volumen : 274-275
Páginas : 214-224
Tipo de publicación : ISI Ir a publicación

Abstract

The Late Cretaceous Catalão I carbonatite complex consists of ultramafic silicate rocks, phoscorites, nelsonites and carbonatites. The latest stages of the evolution of the complex are characterized by several nelsonite (magnetite–apatite rock) and carbonatite dykes, plugs and veins crosscutting earlier alkaline rocks. The interaction between the latter and late-stage carbonatites and/or carbo-hydrothermal fluids, converted the original dunites and bebedourites to metasomatic phlogopitites. Late-stage nelsonites (N1), pseudonelsonites (N2) and various types of dolomite carbonatites (DC) including norsethite-, magnesite- and/or monazite-bearing varieties show significant whole-rock Nd and Sr isotopic variations. To elucidate whether magmatic or metasomatic processes, or both, were responsible for these isotope variations we characterized the Nd and Sr isotope compositions of major mineral phases (i.e. apatite, dolomite, norsethite, pyrochlore and tetraferriphlogopite) in these late-stage rocks. Mineral isotope data recorded the same differences observed between N1 and N2 whole-rocks with N2 minerals showing more enriched isotopic signatures than minerals from N1. Sr isotopic disequilibrium among minerals from N2 pseudonelsonites and spatially related dolomite carbonatite pockets implies formation from batches of carbonate melts with distinct isotopic compositions. A detailed investigation of Nd and Sr isotopes from whole-rocks and minerals suggests that the most evolved rocks of the Catalão I complex probably derive from two different evolution paths. We propose that an earlier magmatic trend (path A) could be explained by several batches of immiscible and/or residual melts derived from carbonated-silicate parental magma (e.g. phlogopite picrite) contaminated with continental crust to a variable extent, in an AFCLI-like process. A second trend (path B) comprises highly variable 143Nd/144Ndi at nearly constant 87Sr/86Sri coupled with high δ18O in carbonates. This is interpreted here as the result of the interaction of previously-formed dolomite carbonatites with carbo-hydrothermal fluids.