9:45 AM - 10:00 AM
[SMP29-03] Spectroscopic investigation on high-pressure behavior of KAlSi3O8-hollandite under the lower mantle condition
Keywords:K-hollandite, Raman spectroscopy, phase transition, pressure-transmitting medium
K-hollandite is a high-pressure polymorph of potassium-rich feldspar (KAlSi3O8) existing in the Earth's mantle [1,2]. There are two polymorphs: K-hollandite-I and K-hollandite-II. K-hollandite-I has a tetragonal structure (space group I4/m) and stable at lower pressure (10-20 GPa) [3]. K-hollandite-I undergoes a second-order phase transition into K-hollandite-II with a monoclinic structure (space group I2/m) without any discontinuous changes in cell volume. K-hollandite-II is stable at higher pressure corresponding to the entire lower mantle condition [2,3,4]. These crystal structures have large square tunnels formed by double chains of edge-shared (Al,Si)O6 octahedra, and K resides in the tunnel and serves as a host mineral for large-ion lithophile elements (LILEs) [2,3]. The compressibility of K-hollandite has been studied via in-situ X-ray diffraction whereas the changes in the phase transition were not clearly observed in Raman spectra in previous studies. Liu et al. 2009 [5] has reported Raman spectra up to 31 GPa, but used water as a pressure-transmitting medium. In this study, therefore, Raman spectra of synthesized K-hollandite were investigated up to 51.0 GPa at room temperature using a diamond anvil cell. The new bands previously reported to be derived from K-hollandite-II were examined under more hydrostatic condition. Various pressure-transmitting media (water, methanol-ethanol mixture, argon and helium) were used to clarify the effect of hydrostaticity on the pressure behavior of K-hollandite. In-situ synchrotron X-ray diffraction observation was also performed at BL18C, PF, KEK to investigate the possibility of heavy noble gases incorporation into the tunnel structure.
Observed Raman bands increased with increasing pressure. The pressure dependance of each Raman mode was linear, but the slope changed before and after the phase transition. The isothermal mode Grüneisen parameter, which is important to quantitively describe the elasticity and anharmonicity under high P-T conditions, was also calculated for both K-hollandite phases. The Raman shift derived from the B1g splitting mode only showed a notable change at different pressure depending on the hydrostaticity. The change in this B1g mode directly linked to the structural distortion due to the phase transition between K-hollandite-I and K-hollandite-II. No obvious difference in the compressibility was observed regardless of the pressure-transmitting medium, suggesting that the K-hollandite does not incorporate substantial amount of LILEs or noble gases in the lower mantle.
[1] Irifune et al. (1994) Earth Planet. Sci. Lett. 126, 351–368
[2] Nishiyama et al. (2005) Phys. Chem. Minerals. 32, 627–637
[3] Kawai and Tsuchiya (2013) Am. Min. 98, 207–218
[4] Hirao et al. (2008) Phys. Earth Planet. Inter. 166, 97–104
[5] Liu et al. (2009) Phys. Chem. Minerals. 36, 143–149
Observed Raman bands increased with increasing pressure. The pressure dependance of each Raman mode was linear, but the slope changed before and after the phase transition. The isothermal mode Grüneisen parameter, which is important to quantitively describe the elasticity and anharmonicity under high P-T conditions, was also calculated for both K-hollandite phases. The Raman shift derived from the B1g splitting mode only showed a notable change at different pressure depending on the hydrostaticity. The change in this B1g mode directly linked to the structural distortion due to the phase transition between K-hollandite-I and K-hollandite-II. No obvious difference in the compressibility was observed regardless of the pressure-transmitting medium, suggesting that the K-hollandite does not incorporate substantial amount of LILEs or noble gases in the lower mantle.
[1] Irifune et al. (1994) Earth Planet. Sci. Lett. 126, 351–368
[2] Nishiyama et al. (2005) Phys. Chem. Minerals. 32, 627–637
[3] Kawai and Tsuchiya (2013) Am. Min. 98, 207–218
[4] Hirao et al. (2008) Phys. Earth Planet. Inter. 166, 97–104
[5] Liu et al. (2009) Phys. Chem. Minerals. 36, 143–149