The 131–134 Ma A-type granites from northern Zhejiang Province, South China: Implications for partial melting of the Neoproterozoic lower crust

Abstract

Although Mesozoic granites are widely distributed in the Gan-Hang Belt in Southeast China, their petrogenesis and geodynamic settings are still matters of dispute. Here, the major and trace elements, Sr-Nd isotopic compositions, zircon U‐Pb dating and Hf isotopes of three late Mesozoic granite plutons from Machebu, Shenzhongwu and Daixi in northern Zhejiang Province were analyzed to investigate their petrogenesis. These granite plutons are featured by an A2-type granite geochemical signature (e.g., high SiO2 (71 to 78wt.%), total alkalis (Na2O+K2O=7.57 to 9.12wt.%), rare earth elements (total REE=174 to 519ppm) and HFSE contents, with mostly high FeOT/(FeOT+MgO) (0.82 to 0.93) and Ga/Al ratios (2.49 to 5.07) and low Ce4+/Ce3+ ratios in zircons (1 to 90)), suggesting they were formed in an extensional tectonic setting. Combining whole rock Zr contents, mineral assemblages and the An content of plagioclase, it was suggested that these granites could have formed at a high temperature (>850°C) with a low H2O content (~2.5wt.%). Assuming the granite with the lowest SiO2 and high CaO contents as the “primary granite melt”, the melting pressure was estimated to be 2.5kbar based on model calculations using MELTs. Taking into account the effect of plagioclase fractional crystallization during the granite emplacement into shallow crustal levels, the initial melting pressure could be >2.5kbar at the stable field of plagioclase. This agrees well with the very low and variable Sr and Eu contents of these granite plutons.Zircon U–Pb dating yields a weighted mean 206Pb/238U age of 130.9±1.0Ma for the Machebu granite, 134.4±2.1Ma for the Shenzhongwu granite and 131.9±1.2Ma for the Daixi granite. The εHf(t) values of the zircons in these plutons gradually increase from southwest to northeast, i.e., from −14.5~−3.5 for the Machebu granite to −1.5~−0.2 for the Shenzhongwu granite and from 5.1~8.6 for the Daixi granite, suggesting the origin of different parent rocks. These granites have two-stage Nd model (TDM2(Nd)) ages of 1099Ma–838Ma, and zircons from these granites and the Neoproterozoic basement in the Gan-Hang Belt plot on the same evolutionary trend in the εHf(t)-age diagram. It is interesting to note that the collection of literature data shows a positive correlation between SiO2 and εNd(t) for the late Mesozoic A-granites in the Gan-Hang Belt, and the Neoproterozoic A-granites in the Gan-Hang Belt cluster in two groups of the high-SiO2-εNd(t) group and low-εNd(t) group. The positive correlation of SiO2-εNd(t) demonstrated by the late Mesozoic A-granites can be well explained by a high-degree of melting of mixtures between the two groups of Neoproterozoic A-granites. We thus suggest that the late Mesozoic A-type granites in the Gan-Hang Belt could have been derived from the rejuvenated Neoproterozoic rocks rather than directly from the Mesoproterozoic metamorphic basement as a result of subduction.