Japan Geoscience Union Meeting 2021

Presentation information

[J] Oral

S (Solid Earth Sciences ) » S-MP Mineralogy & Petrology

[S-MP25] Deformed rocks, Metamorphic rocks and Tectonics

Sun. Jun 6, 2021 9:00 AM - 10:30 AM Ch.20 (Zoom Room 20)

convener:Yumiko Harigane(Geological Survey of Japan, National Institute of Advanced Industrial Science and Technology (AIST)), Yoshihiro Nakamura(Geological Survey of Japan, National Institute of Advanced Industrial Science and Technology), Chairperson:Yumiko Harigane(Geological Survey of Japan, National Institute of Advanced Industrial Science and Technology (AIST)), Yoshihiro Nakamura(Geological Survey of Japan, National Institute of Advanced Industrial Science and Technology)

9:45 AM - 10:00 AM

[SMP25-04] Crystal size distribution of garnet formed by two-stage growth in the Kotsu eclogite, Sanbagawa belt

*OTGONBAYAR DANDAR1, Atsushi Okamoto1, Masaoki Uno1 (1.Graduate School of Environmental Studies, Tohoku University)

Keywords:Garnet, Crystal size distribution, nucleation, growth, Kotsu eclogite

The nucleation and growth history of crystals during metamorphism are recorded both as crystal size distribution (CSD) and compositional zoning from core to rim. In particular, the crystal size distribution of garnet has been investigated and modeled by relative rate of nucleation and growth, and coupled with diffusion. As garnet is commonly formed during prograde metamorphism, the incremental amount of the garnet growth at individual P-T conditions is essential information to elucidate the water release at the subduction interfaces. Conventionally, there are two types of model on crystal size distribution; one is the nucleation and growth rates are defined as a function of time (e.g., Spillar and Dolejs 2013), and the other one is growth rate is defined as the size (e.g., Eberl et al., 1998). However, it is still uncertain which model is more appropriate, as the data aquations of CSD from thin sections and the composition of “true” core are very difficult due to the cut effects. In this contribution, we showed the detailed data showing the relationship between crystal size distribution and compositional zoning of garnet in the Kotsu eclogite in the Sanbagawa metamorphic belt, Japan. This eclogite sample is suitable to this trial, because garnet grains show euhedral shape and record two-stage growth history composed of the core containing abundant mineral inclusions (epidote, plagioclase, chlorite, carbonate, titanite, quartz, and omphacite) and poor-rim.

The detailed analyses were performed in the following steps. (1) We crashed the ~120 grams of eclogite samples by the system of high voltage pulse fragmentation (Selfrag), and picked up all garnet grains (n>4342). (2) We sieved the grains into five classes (>2 mm, >1 mm, >0.5 mm, >0.25 mm, and <0.25 mm). (3) In each class, garnet grains are embedded in epoxy resin, and we took the X-ray CT images for getting the crystal size distribution. (4) After taking X-ray CT, we polished the sample to section the real cores. (5) The two-stage growth is obtained based on the back-scattered images and get the compositional zoning from core to rim.
Combining the results of the crystal sizes taken from the X-ray CT images in all classes, we reveal that the garnet population shows pseudo-lognormal CSDs, which has a small shoulder at the medium grain size (~0.75 mm). The crystal growth is divided into two stages. The garnet grains are euhedral grains composed of the inclusion-rich core (stage 1) and poor-rim (stage 2). The core shows prograde zoning, characterized by decreases of XMn (0.025 to 0.005) and increase of XMg (0.06 to 0.12) from core to rim, and almost constant XFe (0.55-0.65) and XCa (0.27-0.35). The smaller grains have lower Mn and higher Mg contents, suggesting that nucleation at the later retrograde stages. All grains have an inclusion-poor rim showing Mn increase. One of the interesting observations is a positive relationship between the diameter of the inclusion-rich core and the growth width of the inclusion-poor rim, meaning that the growth rate is proportional to the grain size. We will construct the nucleation and growth model to reproduce the CSD and the trend of growth zoning, and discuss the mechanism and water production history.