Sharp et al. [2007, Nature] reported 37Cl values for MORB (EPR, PARC, Iceland, Macquarie), and other primitive materials, that might suggest an isotopically uniform mantle source of chlorine (37Cl=0±1). Nevertheless, the isotopic distribution of chlorine in the diverse tectonic regimes that produce basaltic magmas has not yet been fully explored.
New determinations on a broad selection of oceanic basalts, representative of isotopic mantle end members (HIMU, EMI, EMII) suggest that there may be a systematic and discernible variation in the ranges of Cl isotopes in these sources. Improvements to the SIMS method for 37Cl determination originated by Layne et al. [2004, Chem Geol] and Godon et al. [2004, Chem Geol] have enabled determinations in basalt glass fragments to better than 0.3 per mil (1s) precision and 0.7 per mil overall reproducibility for Cl contents greater than 150 ppm.
Replicate determinations in a subset of the samples characterized by Stroncik & Haase [2002, Geology] yielded mean 37Cl values of -3.5 (range -5.2 to -1.6; n=4) for EMI/II and -0.8 (-2.6 to 0.8; n=6) for HIMU. The observation that EM-type basalts have the most negative 37Cl supports the model that the mantle source of these lavas contains recycled sedimentary material. The moderately negative 37Cl values of some HIMU-type lavas overlap with those of altered basalts, which are generally assumed present in the HIMU source.
Basalts from the northern Kermadec island arc (Monowai) display slightly heavier, positive 37Cl (0.3 to 0.5) - compared to dacites from the southern Kermadec arc (Brothers; -0.3). The former could be explained by a strong flux of Cl from subducting serpentinite whereas the latter agree with the observed increasing influence of subducting sediments in the southern Kermadec arc.