Linking low δ44/40Ca basalts to recycled carbonates through coupled chemical and Ca-Zn-Mg isotopic covariations

Stable calcium (Ca) isotopic studies have revealed that some mantle-derived magmas, such as basalts, kimberlites, and carbonatites, generally exhibit varying degrees of lower δ^44/40Ca (δ^44/40Ca = [(⁴⁴Ca/⁴⁰Ca)_sample / (⁴⁴Ca/⁴⁰Ca)_{SRM 915a} − 1] × 1000) values compared to the peridotitic mantle. However, it remains controversial whether the low δ^44/40Ca signature can represent a fingerprint of isotopically light recycled carbonates in the mantle source or is merely a result of isotope fractionation during mantle partial melting. In this study, we further explore this issue by presenting the coupling of chemical and Ca–Zn–Mg isotopic compositions in a suite of Cenozoic intraplate basalts (Abaga and Dalinor) from Northeast China. These basalts exhibit strong correlations among various chemical indices indicative of a carbonated mantle source; specifically, the samples with low SiO₂ contents (e.g., ≤ 45 wt%) are characterized by high CaO/Al₂O₃ (≥ 0.8) and low Hf/Hf* (≤ 0.6) and Ti/Ti* (≤ 0.8) values. This observation indicates that such basalts may have been variably contributed by a carbonated mantle source. Notably, the δ^44/40Ca values (0.63–1.04 ‰) of the studied samples display positive correlations with SiO₂, Hf/Hf*, and Ti/Ti* values, while showing a negative correlation with CaO/Al₂O₃ ratios. These correlations cannot be explained by potential effects of secondary alteration or magmatic processes, such as magma differentiation and partial melting, but should also be attributed to varying extents of contributions from a carbonated mantle source. Furthermore, the δ^44/40Ca values exhibit a positive correlation with δ²⁶Mg (−0.50 ‰ to −0.34 ‰) and a negative correlation with δ⁶⁶Zn (0.34–0.50 ‰) values. Such tightly coupled covariations of Ca, Mg, and Zn stable isotopes demonstrate that recycled carbonates in the mantle source have contributed to the low δ^44/40Ca basalts. Binary mixing calculations indicate that adding 5–10 % dolomite to the peridotitic mantle can explain the Ca–Zn–Mg isotopic compositions of most samples in this study. Therefore, this study suggests that the low δ^44/40Ca signature of some mantle-derived magmas can be regarded as a fingerprint of recycled carbonate in the mantle source, rather than being exclusively attributed to the effects of partial melting. The integrated application of Ca–Zn–Mg isotopes can provide deeper insights into the origin of the low δ^44/40Ca signature in mantle-derived magmas and deep carbonate recycling.