Subduction of oceanic plates can transport oceanic crust and sediment to the deep mantle, profoundly affecting the chemical and lithological composition of the mantle source region of ocean island basalts (OIB). Over the past few decades, geochemical studies on OIB have revealed a high degree of chemical heterogeneity in the deep mantle, and identified deep mantle chemical reservoirs such as EM1 (enriched mantle type 1), EM2 (enriched mantle type 2), HIMU (high μ; μ = 238U/204Pb), and FOZO (focus zone). Although we now have a good understanding of the geochemical properties of these deep reservoirs, their petrological properties are still poorly known. In recent years, stable isotope systems of major elements (e.g., Fe, Ca, etc.) have developed rapidly and shown sensitivity to different mineral assemblages in the mantle, providing us with the possibility of revealing in depth the petrological properties of the aforementioned mantle chemical reservoirs. Ca is one of the major components in the mantle, mainly occurring in clinopyroxene in typical mantle rocks such as lherzolite. Due to the small fractionation of Ca isotopes (usually expressed as δ44/40Ca) between clinopyroxene and basaltic melts, the degree of Ca isotope fractionation during peridotitic melting is relatively small (typically ∆44/40Camelt-source value is about -0.1‰, indicating that the melt is about 0.1‰ lighter in δ44/40Ca compared to the source region). For eclogite transformed from subducted oceanic crust, its Ca isotopes are lighter than those of peridotitic mantle (light Ca isotopes enter the oceanic crust during pyroxenite melting to produce basaltic crust), and garnet and clinopyroxene are the main minerals containing Ca. Since garnet has a significantly heavier Ca isotope composition compared to the melt, the melt produced by partial melting of eclogite should have a significantly lighter Ca isotope composition compared to the melt directly derived from peridotitic mantle. Therefore, studying the Ca isotope composition of OIB is expected to reveal the petrological properties of the aforementioned mantle chemical reservoirs.
Based on the above scientific hypothesis, the “Roking Mantle Group” research group selected classic end-member OIB (referring to Sr, Nd, Pb and other radiogenic isotopic compositions similar to EM1, EM2, HIMU, and FOZO mantle chemical reservoirs; Figure 1A-B) for high-precision Ca stable isotope analysis. The obtained Ca isotope data shows that the Ca isotope compositions of HIMU-type and FOZO-type OIB are similar (with average δ44/40Ca values of 0.77 ± 0.09‰ and 0.78 ± 0.06‰, respectively) and are close to the Ca isotope composition of mid-ocean ridge basalt (MORB) (MORB δ44/40Ca is 0.84 ± 0.09‰); EM1 and EM2-type OIB, on the other hand, have significantly lighter Ca isotope compositions compared to MORB (δ44/40Ca ranges from 0.65 to 0.72‰ and 0.61 to 0.71‰, respectively) (Figure 1C). Since the Ca isotope fractionation during the crystallization differentiation process of basaltic magma is very small (δ44/40Ca variation is usually less than 0.07‰), and the Ca isotope compositions of EM1 and EM2-type OIB are still significantly lighter than HIMU and FOZO-type OIB at the same MgO content (Figure 1C), the difference in Ca isotope compositions among different end-member OIB is not caused by magma crystallization differentiation process. Detailed quantitative calculations show that partial melting of garnet peridotite can well explain the Ca isotope compositions of HIMU and FOZO-type OIB (Figure 2), while the lighter Ca isotope compositions of EM1 and EM2-type OIB are difficult to be generated by partial melting of garnet peridotite. The coupling of light Ca isotopes and high Gd/Yb ratios indicates a contribution of eclogite components derived from garnet-rich to the mantle source regions of these OIBs.(Figure 2).
Figure 1 (A)Sample location map of ocean island basalts (OIBs);(B)The Sr, Nd, and Pb isotopic compositions for the studied OIB samples;(C)δ44/40Ca versus MgO for OIBs.
Figure 2 δ44/40Ca versus Gd/Yb for OIBs
The negative correlation between the Ca and Fe isotopes of the OIB samples in this study (Figure 3) further confirms the conclusions obtained based on quantitative calculations. The melts produced by the melting of peridotite usually have Fe isotope compositions similar to MORB, while the melts produced by the melting of eclogite transformed from subducted oceanic crust have significantly heavier Fe isotope compositions than MORB. HIMU and FOZO-type OIBs have Ca and Fe isotope compositions similar to MORB, consistent with the melts produced by the melting of garnet peridotite, while EM1 and EM2-type OIBs have light Ca and heavy Fe isotope compositions, indicating the contribution of eclogite components in their mantle source regions. It is due to the differences in the lithological compositions of these different end-member OIBs in their mantle source regions that their average Ca and Fe isotope compositions exhibit a significant negative correlation (Figure 3).
The above research results indicate that Ca isotopes can effectively identify the lithology of mantle source regions in basalt (olivine /eclogite), and have the potential to become a new “link” connecting mantle chemical heterogeneity and lithological heterogeneity. In addition, since the mantle plumes generating the aforementioned OIB (Ocean Island Basalt) mostly originate from the interior of the Large Low Shear Velocity Province (LLSVP) beneath the Pacific plate (Figure 1A), the lithological heterogeneity of these OIB mantle source regions also implies a high degree of heterogeneity within the LLSVP.
Figure 3 Negative correlation between average δ44/40Ca and δ57Fecorr (fractionalcrystallization-corrected Fe isotopic compositions, which can be regarded as the δ57Fe values of primary magmas) values for OIBs.
The related study, titled "Calcium isotopic variability in hotspot lavas controlled by partial melting and source lithological heterogeneity," was published in the well-known journal "Chemical Geology" in the field of geochemistry. Associate Professor Wang Xiaojun from the "Roking Mantle Group" research group is the first and corresponding author of the paper, with Professor Chen Lihui and team members Zhou Zhongbiao, Liu Jianqiang, Zeng Gang, among others, participating in the research. This study was financially supported by the National Natural Science Foundation of China (grants 42130310, 41973001).