Institute of Mineralogy, Petrology, & Economic Geology, Tohoku University^* The Tohoku University Museum, Japan^** Institute of Mineralogy, Petrology, & Economic Geology, Tohoku University, Japan^*** BL15XU/SPring8, National Institute for Materials Science, Japan^****
○Koichi Momma^* Toshiro Nagase^** Yasuhiro Kudoh^*** Takahiro Kuribayashi^*** Masahiko Tanaka^****

Brazil twin in microcrystalline quartz was studied by using Rietveld analysis of X-ray powder diffraction and Raman spectroscopy. Genesis of Brazil twin and moganite in microcrystalline quartz varieties has long been argued. However, effects of surface energy on their occurrence have never been studied so far.
Crystallite size and anisotropic lattice strain of natural and synthetic quartz was refined by Rietveld analysis of synchrotron X-ray powder diffraction data. The lattice strain along <101>* is largest in all microcrystalline samples, which are different in origins, crystallite size, and texture. On the other hand, the lattice strain is isotropic in macroscopic samples. The anisotropic lattice strain increases with decreasing size of the crystallite. Axial ratio (c/a) also decreases with an increase in lattice strain and with a decrease in crystallite size. Quantitative analysis of moganite contents in microcrystalline silica samples was carried out by Raman spectroscopy. The result reveals that the samples having large lattice strain contain high amount of moganite, and that the lattice strain along <101>* is caused by Brazil twin. A sample synthesized from Al-doped silica gel has larger lattice strain than those of non-doped samples. A synthesis from Fe-doped silica gel has the same lattice strain as those of non-doped samples. The correlation between the Brazil twin and crystallite size shows that the surface energy is the driving force for the formation of Brazil twin in microcrystalline quartz. Although ferric iron is believed to be a cause of Brazil twin, its role in microcrystalline quartz is negligibly small compared to the surface energy.