Immiscibility between carbonic fluids and granitic melts during crustal anatexis: A fluid and melt inclusion study in the enclaves of the Neogene Volcanic Province of SE Spain

In the restitic crustal enclaves in the Neogene volcanics of El Hoyazo and Mazarrón (SE Spain), associations of fluid and silicate melt inclusions indicative of immiscibility are frequently observed in the Bt-poor, Crd-rich graphitic metapelites. These occurrences, extremely rare for anatectic crustal rocks, have been studied by microthermometry, micro-Raman spectroscopy and EMP analysis. Both at El Hoyazo and Mazarrón the immiscible FI and MI are hosted in plagioclase and cordierite, with microstructures that suggest primary trapping. The FI hosted in cordierite are monophase, whereas they may contain crystals of calcite when hosted in plagioclase, indicating fluid–host reaction during cooling. Decrepitation microstructures have not been observed. The MI contain fresh glass of peraluminous, felsic rhyolitic composition, typical of anatectic S-type melts. In all samples the fluids are CO2-dominated (> 85 mol%), with minor amounts of N2 and CH4, and traces of CO and H2. The nucleation of graphite, induced by the laser beam in some inclusions, demonstrates that the fluids are graphite-saturated. Large scatter in microthermometric behaviour of FI indicates significant density variations. Only in one sample from El Hoyazo are fluid densities compatible with the estimated P–T conditions of trapping (5–7 kbar, 850 ± 50 °C), whereas in the remaining samples the extremely low densities, suggesting trapping pressures < 1 kbar, are in contrast with the microstructural indications of primary entrapment and little evidence of decrepitation. In addition, although some measured compositions are compatible with C–O–H–N fluid speciation at the estimated P–T conditions of anatexis, the lack of H2O in all FI is inconsistent with the slightly hydrated character of coexisting glasses. The possible mechanisms accounting for departure of densities and compositions of FI from expected values are discussed, and leakage of H2O suggested as the most plausible.