Elemental and isotopic insights into fluid mobility accompanying the subduction of oceanic lithosphere

This study presents a detailed elemental and isotopic investigation of element mobility related to high-pressure, subduction related metamorphism of oceanic lithosphere. Particular respect is paid to the mobility of key redox mediating components (e.g. multivalent elements such as iron (Fe) and sulphur (S)). The overarching aim of this work is to elucidate the contributions of different subducting lithologies to the redox budget of the sub-arc mantle. Furthermore, a comprehensive petrological characterisation of different meta-mafic and meta-ultramafic lithologies is presented. The primary geochemical tools utilised here are iron (Fe) and zinc (Zn) stable isotopes. Iron stable isotopes are sensitive to small losses of Fe and, as such, serve as excellent tracers of Fe mobility within a high-pressure subduction setting. In addition they are fractionated in the presence of S in the form of Fe-SOx(aqueous) bearing complexes. Consequently Fe isotopes can be utilised to trace the mobility of oxidising S bearing fluids during subduction zone metamorphism and slab dehydration. Zinc stable isotopes are strongly fractioned by SOx and COx bearing fluids, and have been utilised here to trace the mobility of S and/or C bearing fluids within subduction zones. Whole-rock metabasalts and metagabbros data presented here demonstrate that there is no resolvable fractionation of either Fe or Zn isotopes across prograde metamorphic facies (greenschist to blueschist to eclogite). Rather our Fe and Zn isotope data for basaltic eclogites (mean δ56Fe = +0.12 ± 0.11 ‰ and δ66Zn = +0.27 ± 0.09 ‰) overlaps with both measurements of low-grade, seafloor metabasites (mean δ56Fe = +0.11 ± 0.04 ‰ and δ66Zn = +0.20 ± 0.04 ‰) and estimates of MORB (δ56Fe = +0.11 ± 0.17 ‰ and δ66Zn = +0.28 ± 0.03 ‰; Teng et al., 2013 and Wang et al., 2017 respectively. It is noteworthy that the blueschist facies samples show trace element evidence in the form of strong enrichments in alkali fluid mobile elements, for interactions with external, sediment derived fluid(s), which has driven the Fe isotope compositions of these metagabbros towards light values. These factors together suggest that the prograde dehydration of the mafic oceanic crust within subduction zones is associated with minimal release of S or C bearing fluids. It is also apparent that the associated meta-gabbro/basalt dehydration fluids are relatively poor carriers of dissolved Fe. In order to understand the role of subducting slab serpentinites on the redox budget of subduction zones, an exhumed section of eclogite-facies serpentinised ultramafics from the Zermatt-Saas ophiolite have been studied, and as part of this investigation a network of high-pressure metamorphic veins have also been documented. These veins predominantly comprise of olivine and Ti-clinohumite, and are interpreted to be high permeability zones of fluid release and migration formed during the partial dehydration of the antigorite serpentinite host. Building on the petrographic investigation of these serpentinite dehydration features it is shown that whole rock antigorite + olivine bearing serpentinites, which are representative of serpentinites undergoing partial dehydration, display a strong negative correlation between δ56Fe values and S contents. This suggests that the release of S from these rocks that controls their Fe isotope composition. Furthermore, both vein forming olivine and Ti-clinohumite display markedly light δ56Fe values (δ56Fe = -0.24 ± 0.28 ‰ and -0.12 ± 0.11 ‰), demonstrating the mobility of isotopically light Fe within serpentinite-derived fluids in subduction zones. We thus conclude that serpentinite dehydration is associated with the liberation and mobilization of SOX bearing fluids, which are isotopically light with respect to Fe. An entire segment of oceanic lithosphere, from mafic crustal units to underlying serpentinised peridotite, has been studied. The data presented here has enabled a greater understanding of the principal controls on fluid release and element mobility during subduction zone metamorphism. While the dehydration of metabasites clearly involve the release of fluid mobile elements it is likely that the dehydration of these lithologies play a relatively small role in the transfer of redox mediating components during subduction. Of greater significance is the process of serpentine mineral breakdown and fluid release, which is observed to play a key role in the release of highly oxidizing species within associated serpentinite derived fluids