DI32A-01:
Water and Slabs in the Transition Zone - Hydrous Ringwoodite in Diamond

Wednesday, 17 December 2014: 10:20 AM
D. Graham Pearson1, Frank E. Brenker2, Fabrizio Nestola3, John McNeill4, Lutz Nasdala5, Mark Hutchison6, Sergei Matveev1, Kathy Mather4, Laszlo Vincze7, Sylvia Schmitz2 and Bart Vekemens7, (1)University of Alberta, Edmonton, AB, Canada, (2)Geothe University, Frankfurt, Germany, (3)Department of Geosciences, University of Padova, Padova, Italy, (4)Durham University, Durham, United Kingdom, (5)Universitat Wien, Vienna, Austria, (6)Trigon GeoServices Ltd, Las Vegas, United States, (7)Ghent University, Ghent, Belgium
Abstract:
Theory and experiments have shown that the Earth’s Transition Zone (TZ) could be a major repository for water, due to the ability of the higher-pressure polymorphs of olivine - wadsleyite and ringwoodite - to host up to ~2.5wt. % H2O. Despite experimental demonstration of the water-bearing capacity of these phases, geophysical probes such as electrical conductivity have provided conflicting results, and the issue of whether the TZ contains abundant water remains highly controversial. We report X-ray diffraction, Raman and infra-red spectroscopic evidence for the first terrestrial occurrence of any higher pressure polymorph of olivine: ringwoodite, included in a diamond from Juína, Brazil. The ringwoodite occurs with a Ca-walstromite phase that we interpret to be retrogressed Ca-silicate perovskite. The most likely interpretation of this two-phase assemblage is that it represents a partially retrogressed portion of a somewhat Fe-rich peridotitic mantle, in which hydrous ringwoodite, and former CaSiO3- perovskite co-existed above 15GPa. The ringwoodite has a Mg# of ~ 75, suggesting that it may be mantle hybrised with a more fertile component such as subducted oceanic crust. The water-rich nature of this inclusion (~1.5 wt%), along with the preservation of ringwoodite, is the first direct evidence that, at least locally, the TZ is hydrous, to about 1 wt%. As well as being in agreement with recent magnetotelluric estimates of the TZ water content, this amount of water helps to reconcile measured TZ seismic velocities with those predicted from lab experiments. The finding also indicates that some kimberlites must have their primary sources in this deep mantle region. The high water content of the ringwoodite suggests that it was not close to the mantle geotherm when trapped in the diamond. This may be an indication that the the assemblage was part of a water-rich subducted slab out of thermal equilibrium, within the transition zone. The water-rich nature of the ringwoodite inclusion provides support for the notion that the lower P and S wave velocities expected from hydrous peridotite based on laboratory experiments are consistent with a hydrous transition zone.