The boron and lithium isotopic composition of mid-ocean ridge basalts and the mantle

© The Author(s), 2017. This is the author's version of the work. It is posted here under a nonexclusive, irrevocable, paid-up, worldwide license granted to WHOI. It is made available for personal use, not for redistribution. The definitive version was published in Geochimica et Cosmochimica Acta 207 (2017): 102–138, doi:10.1016/j.gca.2017.03.028.A global selection of 56 mid-ocean ridge basalt (MORB) glasses were analysed for Li and B abundances and isotopic compositions. Analytical accuracy and precision of analyses constitute an improvement over previously published MORB data and allow a more detailed discussion of the Li and B systematics of the crust-mantle system. Refined estimates for primitive mantle abundances ([Li]=1.39±0.10[Li]=1.39±0.10 μg/g and [B]=0.19±0.02[B]=0.19±0.02 μg/g) and depleted mantle abundances ([Li]=1.20±0.10[Li]=1.20±0.10 μg/g and [B]=0.077±0.010[B]=0.077±0.010 μg/g) are presented based on mass balance and on partial melting models that utilise observed element ratios in MORB. Assimilation of seawater (or brine) or seawater-altered material beneath the ridge, identified by high Cl/KCl/K, causes significant elevation of MORB δ11Bδ11B and variable elevation in δ7Liδ7Li. The B isotope ratio is, hence, identified as a reliable indicator of assimilation in MORB and values higher than −6‰ are strongly indicative of shallow contamination of the magma. The global set of samples investigated here were produced at various degrees of partial melting and include depleted and enriched MORB from slow and fast-spreading ridge segments with a range of radiogenic isotope signatures and trace element compositions. Uncontaminated (low-Cl/KCl/K) MORB show no significant boron isotope variation at the current level of analytical precision, and hence a homogenous B isotopic composition of δ11B=-7.1±0.9‰δ11B=-7.1±0.9‰ (mean of six ridge segments; 2SD). Boron isotope fractionation during mantle melting and basalt fractionation likely is small, and this δ11Bδ11B value reflects the B isotopic composition of the depleted mantle and the bulk silicate Earth, probably within ±0.4‰. Our sample set shows a mean δ7Li=+3.5±1.0‰δ7Li=+3.5±1.0‰ (mean of five ridge segments; 2SD), excluding high-Cl/KCl/K samples. A significant variation of 1.0–1.5‰ exists among various ridge segments and among samples within individual ridge segments, but this variation is unrelated to differentiation, assimilation or mantle source indicators, such as radiogenic isotopes or trace elements. It, therefore, seems likely that kinetic fractionation of Li isotopes during magma extraction, transport and storage may generate δ7Liδ7Li excursions in MORB. No mantle heterogeneities, such as those generated by deeply recycled subducted materials, are invoked in the interpretation of the Li and B isotope data presented here, in contrast to previous work on smaller data sets. Lithium and boron budgets for the silicate Earth are presented that are based on isotope and element mass balance. A refined estimate for the B isotopic composition of the bulk continental crust is given as δ11B=-9.1±2.4‰δ11B=-9.1±2.4‰. Mass balance allows the existence of recycled B reservoirs in the deep mantle, but these are not required. However, mass balance among the crust, sediments and seawater shows enrichment of 6Li6Li in the surface reservoirs, which requires the existence of 7Li7Li-enriched material in the mantle. This may have formed by the subduction of altered oceanic crust since the Archaean.This study was in part financially supported by the NSF ocean sciences program (OCE grant #1232996 to VDW and HRM), and by a WHOI internal grant provided by the Andrew W.Mellon Foundation to HRM and NS (#27005261). TE was partially supported by NERC grant NE/M000427/1