[S-EM13] Geomagnetism, paleomagnetism and rock magnetism

[SEM13-11] Relationship between the abundance of the exsolved magnetite and the valence state of iron in single plagioclase crystal separated from Doshi gabbro

*Yujiro Tamura¹, Masahiko Sato², Ryoichi Nakada³, Chie Kato⁴, Shinji Yamamoto¹ (1.Yokohama National University, 2.The University of Tokyo, 3.Japan Agency for Marine-Earth Science and Technology, 4.Kyushu University)

Keywords:magnetite exsolution, plagioclase, iron valence

Paleomagnetic measurements using single silicate crystals such as quartz (Tarduno et al., 2010), zircon (Sato et al., 2015), and plagioclase (Kato et al., 2018) play a key role on the geomagnetic field study. Among them, the plagioclase crystal sometimes contains fine-grained and rod-shaped magnetite as exsolution lamellae, and the paleomagnetic record of exsolved magnetite likely provides reliable paleomagnetic information for a long era. However, the exsolved magnetite does not commonly exist in plagioclase crystals of various rock samples, and its formation mechanism has not been understood. It is expected that the formation of magnetite exsolution lamellae is caused by the decrease of maximum solubility of iron in plagioclase crystals during the cooling process of rocks under subsolidus conditions. Additionally, important factors controlling the exsolution of magnetite are the cooling speed of crystal, iron concentration, and valence state of iron in plagioclase should be important factors controlling the exsolution of magnetite. Then, to understand the exsolution mechanism, investigations of iron valence state and abundance of exsolved magnetite within a single plagioclase crystal are suitable because these factors are not significantly different in a single crystal. This study investigated the plagioclase crystal from the Doshi gabbro, northwest of the Tanzawa tonalitic complex, for microscopic observation and measurements of iron valence state. Thin sections with double-side polished were made for the observation by polarizing microscope and electron microscope. Subsequently, we analyzed the composition of plagioclase by Electron probe Micro-Analyzer and the valence state of iron within plagioclase by K-edge X-ray absorption fine structure analysis. The abundance of exsolved magnetite has estimated from backscattered electron images in the image analysis using the Image J software. The plagioclase crystal shows the clear zonal structure of exsolved magnetite parallel with the chemical zoning structure. In order to investigate the formation mechanism of magnetite exsolution, we compared the iron valence within plagioclase with the abundance of exsolved magnetite and iron concentration around the measured spot. The iron valence of plagioclase in each spot was approximately 2.5 to 2.6, and there is no clear relationship between the iron valence of plagioclase and the abundance of exsolved magnetite. We additionally estimated the iron valence state in plagioclase crystal before the magnetite exsolution from iron valence measurement and the abundance of exsolved magnetite. As a result, it was suggested that exsolved magnetite tends to form in the areas that originally had a high iron valence. The causes of the differences in iron valence in plagioclase will also be discussed.