A-type Medina batholith and post-collisional anatexis in the Ara?ua? orogen (SE Brazil).

The Medina batholith and its host granitic migmatites record intriguing plutonic processes in the northern Ara?ua? orogen (SE Brazil). This orogen shows a long lasting (630?480 Ma) succession of granite production events from the earliest pre-collisional plutons to the latest post-collisional intrusions. The Medina batholith includes granite intrusions ascribed to the post-collisional stage. They show high alkali and halogen contents, low CaO (at SiO2=71%:Na2O+K2O=7 to 9%; CaO=1.6%), and high FeOt/(FeOt+MgO) ratios (0.78 to 0.92). The Medina granites are metaluminous to weakly peraluminous, with ASI (molecular ratio Al/(Ca-1.67P+Na_K)) values of 1.76 to 2.07, and have high concentrations of high field strength elements (Zr+Nb+Ce+Y N 700 ppm), as well as high Ga/Al ratios. Accordingly, the Medina intrusions are typical ferroan A-type granites. U\\Pb ages fromzircon (501?2 Ma) and monazite (497?2 Ma) constrain the emplacement timing of theMedina batholith. Surprisingly, all monazite ages from host rocks also cluster around 500Ma, despite their nature and distance from the batholith, suggesting that they would have shared a same thermal process. The studied host rocks are granitic migmatites varying from patch metatexite to nebulitic diatexite, comprising paleosome of foliated sillimanite-garnet-biotite metagranite to gneiss, and non-foliated garnet-cordierite neosome poor to free of biotite. A metatexite (R14) located relatively far from the Medina batholith, and a diatexite (M26) found at the batholith contact were sampled for detailed studies. The paleosome of foliated metagranite (R14A) only shows zircon grains with igneous features and Th/U ratio from1.64 to 0.26. Although the spreading of zircon spots, themain cluster yields a Concordia age at 556?6 Ma, constraining the protolithmagmatic crystallization. A minor cluster furnishes a Concordia age at 499?7 Ma, in agreement with the U\\Pb monazite age at 501?2Ma. Extracted from the same metatexite sample, the non-foliated garnet-cordierite neosome (R14B) shows both igneous and metamorphic zircon domains with Th/U ratios ranging from 1.47 to 0.00. Again, the U\\Pb spots cluster at two distinct Concordia ages (562?3Ma and 499?3Ma). The youngest of them, fitting with themonazite age (495?3Ma), constrains melt crystallization,while the oldest age suggests paleosome inheritance. The nebulitic diatexite (M26) showsmonazite (497?2Ma) and zircon (Th/U=1.7 to 0.0; Concordia ages at 564?2Ma and 507?3Ma) populations similar to themetatexite neosome, alsowith the youngest ages bracketing themelt crystallization process around 500 Ma. Accordingly, all those ages at around 500Ma disclose a partial melting episode coeval with the Medina batholith emplacement. Phase equilibrium modeling on a garnet-cordierite neosome furnished P-T conditions of 750?840 ?C at 2.4?3.5 kbar for that post-collisional anatexis. Evidence for such a late thermal event are common in the Ara?ua? orogen, even far from the post-collisional batholiths. Thus, a possible major heat source can be envisaged, like a mantle plume triggering crustal anatexis and regional fluid circulation during the gravitational collapse of the Ara?ua? orogen