Mantle heterogeneity through Zn systematics in oceanic basalts: Evidence for a deep carbon cycling

Subduction at convergent margins introduces a range of sedimentary and crustal materials into the mantle, providing the most dominant form of heterogeneity in the source of oceanic basalts. Yet, the relationship between geochemical variability and lithologic heterogeneities in the Earth's mantle remains controversial. In this paper, we comprehensively review Zn, δ⁶⁶Zn and Sr-Nd isotope systematics in near-primary basalts erupted at mid-ocean ridges (MORB) and ocean islands (OIB) to help constrain the nature and proportion of the carbon (C) bearing slab-derived component in their mantle sources. We show that Zn elemental and isotopic composition of oceanic basalts differs according to their tectonic settings, increasing from MORB (Zn = 62 ± 10 to 73 ± 11 ppm; δ⁶⁶Zn = +0.24 ± 0.01 to +0.31 ± 0.02‰) to OIB (Zn = 74 ± 9 to 124 ± 7 ppm; δ⁶⁶Zn = +0.21 ± 0.07 to +0.40 ± 0.04‰). Unlike MORB, the high Zn and δ⁶⁶Zn recorded in OIB cannot be explained by partial melting of a fertile peridotite mantle source only. Importantly, global correlations between Zn content and Sr-Nd isotopes in oceanic basalts suggest that the Zn enrichment in OIB is inherited from a recycled component in their mantle source rather than melting processes. We demonstrate that involvement of neither typical MORB-like oceanic crust nor subducted sediments can achieve the whole range of Zn composition in OIB. Instead, addition of ≤6% C-bearing oceanic crust to a fertile peridotite mantle fully resolves the Zn heterogeneity of OIB, both in terms of magnitude of Zn enrichment and global trends with Sr-Nd isotopes. Such scenario is corroborated by the elevated δ⁶⁶Zn of OIB relative to MORB and mantle peridotites, reflecting the contribution of isotopically heavy C-bearing phases (δ⁶⁶Zn = +0.91 ± 0.24‰) to the mantle source (δ⁶⁶Zn = +0.16 ± 0.06‰). Our study thus emphasizes the use of Zn and δ⁶⁶Zn systematics to track the nature and origin of mantle carbon, highlighting the role of subduction in the deep carbon cycle. Finally, the positive correlation between Zn content and temperature of magma generation of oceanic basalts suggests that hotter mantle plumes are more likely to carry a higher proportion of dense C-bearing eclogite. Zinc systematics therefore may provide evidence that the presence of heterogeneous domains in the source of OIB is, at least partly, linked to plume thermal buoyancy, bringing new insights into mantle dynamics.