An experimental study of Ti and Zr partitioning among zircon, rutile, and granitic melt

摘要: In order to evaluate the effect of trace and minor elements (e.g., P, Y, REEs) on high-temperature solubility Ti in zircon (zrc), we conducted 31 experiments a series synthetic natural granitic compositions [enriched TiO_2 ZrO_2; Al/(Na + K) molar ~1.2] at pressure 10 kbar temperatures ~1,400 1,200 °C. Thirty produced zircon-saturated glasses, which 22 are also saturated rutile (rt). seven experiments, quenched glasses coexist with quartz (qtz). SiO_2 contents liquids range from 68.5 82.3 wt% (volatile free), water concentrations 0.4–7.0 wt%. rutile-saturated melts positively correlated run temperature. Glass ZrO_2 (0.2–1.2 wt%; volatile free) show broad positive correlation temperature and, given T, strongly parameter (Na K 2Ca)/(Si·Al) (all cation fractions). Mole fraction (X^(rt)_(ZrO2)) quartz-saturated runs coupled other 10-kbar qtz-saturated experimental data literature (total 675 °C) yields following temperature-dependent expression: ln(X^(rt)_(ZrO2))+ln(ɑ_(SiO2))=2.638(149)−9969(190)/T(K) , where silica activity ɑ_(SiO2) either coexisting polymorph or silica-undersaturated melt is referenced α-quartz P T each experiment best-fit coefficients their uncertainties (values parentheses) reflect X^(rt)_(ZrO2) . NanoSIMS measurements overgrowths yield values ~100 800 ppm; Coupled for ɑ^(SiO2) ɑ_(TiO2) experiment, (ppm) can be related over °C by ln(Ti ppm)^(zrc)+ln(ɑ_(SiO2))−ln(ɑ_(TiO2))=13.84(71)−12590(1124)/T(K) After accounting differences bulk based granite overlap measured using compositions. literature, this suggests that ≥ 1,100 °C, levels “granitic” do not appear influence zircon. Whether true magmatic crustal hydrous silica-rich 800–700 remains demonstrated. Finally, D^(zrc/melt)_(Ti) (calculated weight basis) presented here 0.007–0.01, relatively independent temperature, broadly consistent determined glass pairs.