MINOR ELEMENT SOLID SOLUTION IN GARNET AT ULTRA-HIGH PRESSURE METAMORPHIC CONDITIONS

The chemistry of metamorphic garnet at ultra-high pressure (UHP) can depart significantly from the familiar solid solution involving the divalent cations Ca, Fe, Mg, and Mn in the 8-coordinated X site, limited trivalent cations Fe and Cr replacing Al in the 6-coordinated (octahedral) Y site, and complete occupancy of the (tetrahedral) Z site by Si. UHP crustal metamorphic garnets can contain volumetrically significant exsolved phases such as hydroxyapatite, ilmenite, TiO₂ and SiO₂ as well as silicate phases not limited to clinopyroxene. Additionally, abnormal concentrations of P, Na, HREE, Y and Zr are known to remain in the garnet structure.

Very little is known of the crystal chemistry of high-valence substituents in silicate garnet, and though many models have been postulated, few provide balanced equations to explain the natural observations, particularly the exsolution mechanisms. UHP metamorphic conditions can allow the well-known majoritic substitution of Mg + Si in the Y site (see Liou et al. 2009: J. Asian Earth Sci. 35: 199-231 for summary). Coupled sodium (Na) and phosphorus (P) substitution in garnet has been constrained experimentally (Brunet et al. 2006: Am Min 91: 211-215). The experiments of Zhang et al (2003: EPSL 216:591-601) showed increasing Ti content in garnet coexisting with clinopyroxene as pressure increased from 5 to 15 GPa. Other elements in garnet (Ca, Mg, Si, Na) increased with pressure but limited experiments did not quantify this effect.

Many UHP crustal terranes have significantly exceeded the coesite and diamond phase boundaries, but beyond these often rare and imperfectly-preserved minerals there are few reliable and experimentally-constrained indicators of more advanced P-T conditions. UHP garnet chemistry as revealed by exsolved phases and relic substituents may provide qualitative information but experimental determination of the pressure and temperature dependence of minor element solid solution is necessary. Characterization of site occupancy of all cations will be needed for balanced exchange equations. If pressure dependence of Na, P, Ti, Zr and octahedral Si solid solution in garnet can be quantified by experiments, this could be a guide to conditions once experienced by these rocks.