Large amounts of H2O are carried into trenches via subduction of the sediments, basaltic crust and uppermost mantle that make up the oceanic lithosphere. A major open question is how much of this subducted H2O is released into the overlying mantle wedge and how much is carried deeper into the mantle. This depends on whether the fluid is able to form an interconnected network among the mineral grains that make up the rock down to very low fluid fractions. In many cases, a minimum amount of fluid (critical porosity) is required in order to achieve connectivity, such that some fluid will remain trapped in the rock.
This proposal seeks support for an experimental determination of the 3-D grain scale distribution of aqueous fluid in three types of subducted sediment: siliceous, carbonate, and clay. We will exploit synchrotron X-ray microtomographic techniques developed by Zhu et al. (2011) for the study of 3-D melt distribution in olivine-basalt aggregates to obtain 3-D data on the distribution of fluids in subducted sediment. This nondestructive 3-D imaging technique has a spatial resolution of 0.7 um and provides quantitative information on geometrical parameters of fluid topology, such as the dihedral angles, melt channel sizes and connectivity. We plan to initially investigate the distribution of water in simplified sediment analogs (i.e. quartz for siliceous sediments; calcite for carbonate sediments; smectite for clays). Fluid fraction will be varied from ~10% down to ~1% to determine the porosity at which connectivity is lost for each sediment type. Once we have characterized fluid distribution in simplified systems, we will investigate more realistic scenarios by adding more solid phases and altering the composition of the fluid to more closely approximate seawater. We anticipate that the proposed experiments will lay the necessary groundwork for an NSF proposal to determine 3-D grain scale distribution of fluids in a range of subducted lithologies, including sediments, basalt (eclogite), and peridotite. Results from the proposed study will provide important new insights into the amount of fluid that can be transported into the deep mantle by subduction.