Application of multigrain crystallography to mineralogy of the deep Earth

Laser-heated diamond anvil cell (DAC) technology coupled with x-ray diffraction (XRD) probe has become a powerful tool to explore the mineralogy of the deep Earth. However, powder diffraction alone has the intrinsic limitations when the sample consists of multiple different phases; only a couple of low angle diffraction rings are unique to each of the phases; most overlapping diffraction rings cannot be used for unique identification or accurate determination of crystallographic parameters. Such powder refinements at extreme P–T conditions also suffer from unreliable intensity measurements caused by texturing of the sample and spottiness due to crystal growth under high T. In addition, new structures and minor phases are often overshadowed by the diffraction peaks of major phases.

The crystal structure and unit-cell parameters of Fe-bearing minerals change with pressure/temperature, spin transition, and Fe redistribution. Precise structure analysis is required to understand the role of Fe in the geophysical and geochemical processes in the deep Earth. To overcome the difficulties, we introduced a multigrain crystallography method to determine crystallographic orientations of individual crystallites within a polycrystalline sample in-situ in a DAC. Using this method, we are able to separate individual crystallites and treat them as individual single crystals. The high-brilliance x-ray beam available at the 3rd generation synchrotron facilities has made it possible to collect diffraction spots in a powder sample comprised of hundreds of submicron-sized crystallites. Once separated, the data set for each crystallite can be treated with the standard single-crystal refinement program, resulting in excellent statistics in refinement and maximal coverage of the reciprocal space. The combination of powder diffraction and multigrain crystallography is very powerful in unequivocal determination of symmetry and unit cells, solving new structures, and picking out minor phases. The method has been demonstrated for solving crystallographic problems above megabar pressures.