TitaniQ in reverse: backing out the equilibrium solubility of titanium in quartz

Abstract:

There is close agreement among three of the four experimental studies that have ‘calibrated’ the P–T dependencies of Ti-in-quartz solubility. New experiments were conducted to identify potential experimental disequilibrium, and determine which Ti-in-quartz solubility calibration is most accurate. Quartz and rutile were synthesized from SiO₂- and TiO₂saturated aqueous fluids in a forward-type experiment at 925°C and 10 kbar in a piston-cylinder apparatus. A range of crystal sizes was examined to determine if growth rate affected Ti incorporation in quartz. Cathodoluminescence (CL) images and electron microprobe measurements show that intercrystalline and intracrystalline variations in Ti concentrations are remarkably small regardless of crystal size. The average Ti-in-quartz concentration from the forward-type experiment is 392±1 ppm Ti, which is within 95% confidence interval of data from the 10 kbar isobar of Wark and Watson (2006) and Thomas et al. (2010). Quartz from the forward-type experiment was used as starting material for reversal-type experiments. The high-Ti quartz starting material was recrystallized at 925°C and 20 kbar to reduce the solubility of Ti in recrystallized quartz to the equilibrium solubility concentration of the reversed P–T condition. The ‘dry’ and ‘wet’ reversal experiments produced polycrystalline quartzites. Rutile occurs as inclusions in quartz, and as individual crystals dispersed along quartz/quartz grain boundaries. Quartz that recrystallized during the reversal-type experiment has substantially lower Ti concentrations than the quartz starting material because Ti solubility at 20 kbar is significantly lower than at 10 kbar. Dark cathodoluminescent quartz with low Ti concentrations shows that extensive quartz recrystallization occurred at the reversal P–T condition. The average Ti concentration in quartz from reversal experiments is 94±2 ppm Ti, which is within the 95% confidence interval of a linear fit to the 20 kbar data of Thomas et al. (2010).

Thomas JB, Watson EB, Spear FS, Shemella FS, Nayak SK, Lanzirotti A (2010) Contrib Mineral Petrol 160:743–759

Wark DA, Watson EB (2006) Contrib Mineral Petrol 152:743–754