Partial melting of lower crust at 10-15 kbar: Constraints on adakite and TTG formation

The pressure-temperature (P-T) conditions for producing adakite/tonalite-trondhjemite-granodiorite (TTG) magmas from lower crust compositions are still open to debate. We have carried out partial melting experiments of mafic lower crust in the piston-cylinder apparatus at 10-15 kbar and 800-1,050 °C to investigate the major and trace elements of melts and residual minerals and further constrain the P-T range appropriate for adakite/TTG formation. The experimental residues include the following: amphibolite (plagioclase + amphibole ± garnet) at 10-15 kbar and 800 °C, garnet granulite (plagioclase + amphibole + garnet + clinopyroxene + orthopyroxene) at 12.5 kbar and 900 °C, two-pyroxene granulite (plagioclase + clinopyroxene + orthopyroxene ± amphibole) at 10 kbar and 900 °C and 10-12.5 kbar and 1,000 °C, garnet pyroxenite (garnet + clinopyroxene ± amphibole) at 13.5-15 kbar and 900-1,000 °C, and pyroxenite (clinopyroxene + orthopyroxene) at 15 kbar and 1,050 °C. The partial melts change from granodiorite to tonalite with increasing melt proportions. Sr enrichment occurs in partial melts in equilibrium with <20 wt% plagioclase, whereas depletions of Ti, Sr, and heavy rare earth elements (HREE) occur relative to the starting material when the amounts of residual amphibole, plagioclase, and garnet are >20 wt%, respectively. Major elements and trace element patterns of partial melts produced by 10-40 wt% melting of lower crust composition at 10-12.5 kbar and 800-900 °C and 15 kbar and 800 °C closely resemble adakite/TTG rocks. TiO₂ contents of the 1,000-1,050 °C melts are higher than that of pristine adakite/TTG. In comparison with natural adakite/TTG, partial melts produced at 10-12.5 kbar and 1,000 °C and 15 kbar and 1,050 °C have elevated HREE, whereas partial melts at 13.5-15 kbar and 900-1,000 °C in equilibrium with >20 wt% garnet have depressed Yb and elevated La/Yb and Gd/Yb. It is suggested that the most appropriate P-T conditions for producing adakite/TTG from mafic lower crust are 800-950 °C and 10-12.5 kbar (corresponding to a depth of 30-40 km), whereas a depth of >45-50 km is unfavorable. Consequently, an overthickened crust and eclogite residue are not necessarily required for producing adakite/TTG from lower crust. The lower crust delamination model, which has been embraced for intra-continental adakite/TTG formation, should be reappraised.