• journal_title：American Mineralogist
• Contributor：Li Chung Ming ; Young-Ho Kim ; T. Uchida ; Y. Wang ; M. Rivers
• Publisher：Mineralogical Society of America
• Date：2006-
• Format：text/html
• Language：en
• Identifier：10.2138/am.2006.1930
• journal_abbrev：American Mineralogist
• issn：0003-004X
• volume：91
• issue：1
• firstpage：120
• section：Articles

The phase transformation from ilmenite to perovskite in FeTiO₃ was directly observed using synchrotron-based X-ray diffraction and a large-volume press. The perovskite phase is temperature quenchable at 20 GPa and converts into the LiNbO₃ phase at pressures below 15 GPa at room temperature. The LiNbO₃ phase transforms into the ilmenite phase at 10 GPa and 673 K. However, the back-transformation from the ilmenite to the LiNbO₃ phase was not observed, thus strongly suggesting that the LiNbO₃ phase is not thermodynamically stable but rather a retrogressive phase formed from perovskite during decompression at room temperature.

By cycling the pressure up and down at temperatures between 773 and 1023 K, the perovskite-ilmenite transformation could be observed in both directions, thus confirming that perovskite is the true high-pressure phase with respect to the ilmenite phase at lower pressures. The phase boundary of the perovskite-ilmenite transformation thus determined in this study is represented by P (GPa) = 16.0 (± 1.4) – 0.0012 (± 0.0014) T (K), which is inconsistent with P = 25.2 – 0.01 T (K) reported previously (Syono et al. 1980). The discrepancy could be attributed to the different experimental methods (i.e., in situ vs. quench) used in the two studies. The ilmenite-perovskite phase boundary with such a small slope would potentially serve as a useful geobarometer for ilmenite-bearing rocks derived from the deep mantle or for those shocked in meteor craters.