Partial melting of fertile peridotite fluxed by hydrous rhyolitic melt at 2–3 GPa: implications for mantle wedge hybridization by sediment melt and generation of ultrapotassic magmas in convergent margins

Ananya Mallik,Jared Nelson,Rajdeep Dasgupta
DOI: https://doi.org/10.1007/s00410-015-1139-2
IF: 4.1068
2015-05-01
Contributions to Mineralogy and Petrology
Abstract:We investigated the melting behavior of peridotite fluxed with 25 wt% of H2O-bearing rhyolitic sediment melt (1.8 wt% bulk H2O), by performing experiments from 1100 to 1300 °C at 2 GPa and 1050–1350 °C at 3 GPa. The apparent solidus of our bulk composition lies between 1100 and 1150 °C at both pressures, which is at a higher temperature than the vapor-saturated solidus and close to the pargasite dehydration solidus of peridotite. With increasing temperature, reacted melt fraction increases from 20 to 36 wt% from 1200 to 1300 °C at 2 GPa and 7 to 24 wt% from 1225 to 1350 °C at 3 GPa. Orthopyroxene is present as a residual phase in all the experiments, while olivine is present as a residual phase in all the experiments at 2 GPa only. Amphibole is absent above 1100 °C at both pressures, clinopyroxene disappears above 1200 and 1300 °C at 2 and 3 GPa, respectively, and garnet (only present at 3 GPa) melts out above 1300 °C. Upon reaction with the mantle wedge and subsequent melting of the hybrid rock, subducted sediment-derived rhyolites evolve in composition to a nepheline-normative ultrapotassic leucitite, similar in major element composition to ultrapotassic lavas from active arcs such as Sunda and inactive arcs such as in the Roman Magmatic Province. Fluxing peridotite with H2O versus H2O-bearing sediment melt at similar pressures does not appear to have an effect on isobaric melt productivity, but does have significant effect on melting reactions and resultant melt composition, with influx of sediment melt adding K2O to the system, thereby stabilizing phlogopite, which in turn buffers the reacted melt to ultrapotassic compositions. Previous experimental studies, along with this study, find that phlogopite can be stable near the hotter core of the mantle wedge and, hence, is likely to be subducted to deeper mantle, thereby influencing deeper cycling of volatiles and large ion lithophile elements. Also, because DRbphl/melt ≫ DSrphl/melt and DNdphl/melt, DSmphl/melt ≪ 1, long-term stability of phlogopite in the mantle can create ‘enriched mantle’ domains (εSr and εNd ≥ 0).
geochemistry & geophysics,mineralogy
What problem does this paper attempt to address?