Evidence for Paleoproterozoic anatexis and crustal reworking of Archean crust in the Sao Francisco Craton, Brazil: A dating and isotopic study of the Kinawa migmatite

New SHRIMP U-Pb zircon dating indicates that, contrary to previous thought, part of the Archean crust in the southern Sao Francisco Craton (SFC), NW of the Jeceaba Born Sucesso Lineament was affected by the Paleoproterozoic Transamazonian orogeny. The Kinawa migmatite is a reworked part of the Archean crust in the southern SFC, and occurs within the Claudio Shear Zone, a NE-NW crustal-scale structure. The protolith of the migmatite is a metagranodiorite that has an U-Pb zircon crystallization age of 2704 +/- 3 Ma coeval with the voluminous potassic magmatism in the region. Moreover, the epsilon Hf-(t=2704) data suggests a slightly evolved crust(-1.5 to -6.1) and thus a relatively long crustal residence (similar to 700 Ma). Surprisingly, the migmatite itself yielded much younger ages of 2034 +/- 32 Ma and 2048 +/- 25 Ma, which coincide with the ages of the Mineiro Belt and Mantiqueira Complex. The distribution of ages using data from the literature and this study show that within the Archean domain the occurrence of Paleoproterozoic ages is restricted to the Claudio Shear Zone. Inherited zircon crystals from all samples yield Neoarchean ages (similar to 2.7 Ga), and similar Hf signatures to the metagranodiorite protolith (epsilon Hf(t-2704) = -5.5 to -0.5). The epsilon Hf-(t) signatures of the Kinawa migmatite are heterogeneous and record two events, one is interpreted to be the formation of the protolith with lower Hf-176/Hf-177 (at similar to 2.7 Ga) and the other with higher Hf-176/Hf-177 (at similar to 2.05 Ga) as the anatexis that produced the migmatite. An important result is that the migmatite had two sources. Two samples are consistent with derivation from an evolved crust comparable with the metagranodiorite protolith (e.g. Lu-176/Hf-177 similar to 0.01 and epsilon Hf-(t=2048) -13.4 to -17), but the other sample has much less negative values (epsilon Hf-(t=2048) = -4.6 to -8.7), which requires a different source with much lower Lu-176/Hf-177, and therefore younger, possibly Post-Archean material. (C) 2017 Elsevier B.V. All rights reserved.