11:00 AM - 1:00 PM
[SCG49-P02] Experimental study on grain growth of synthetic labradorite polycrystals under high temperature and high pressure
Keywords:Plagioclase, Grain growth, Synthetic polycrystals, Griggs type deformation apparatus
The flow law of rocks varies by strain rate, temperature, grain size, and water fugacity (e.g., Karato and Jung, 2003 for olivine, Rybacki et al., 2006 for plagioclase). The usage of natural rock samples has difficulty to control those parameters (Chopra and Paterson, 1984). In addition, rocks exhumed from depth often have weathered and weakened grain boundaries by exhumation. Hence, it is important to prepare synthetic polycrystals with controlled composition and grain sizes to extract rheological properties of the rock. The intermediate composition of plagioclase has the lowest melting point, suggesting that the composition of plagioclase in the lower crust under the island arc is lower than An100 (Ishikawa et al., 2014). In this study, we prepare synthesis plagioclase (labradorite) aggregates that are worthy of investigating the rheology of the lower crust.
As starting material, we prepared fine-grained powder of labradorite single crystals (An50-70, Madagascar) by milling (Multi-Beads Shocker). Mean grain size is 0.29 µm. The milling was carried out at 3000 rpm for 180 seconds under conditions inhibiting agglomeration of fine particles and damage due to heating by milling. The powder was kept in an oven at 110℃ for more than one week to obtain dry powder. We used Griggs type deformation apparatus at Tohoku University for sintering under high pressure. Experimental conditions was 2 - 240 hours for 900℃, 1GPa, 1000℃, 1.35GPa to maintain the stability of plagioclase (Stünitz and Tullis, 2001). Also atmospheric annealing at 900℃ was also conducted. We observed microstructures using scanning electron microscope (FE-SEM) and electron backscattered diffraction (EBSD).
The grain size variation for 2 hours annealing time follows a lognormal distribution, but not for samples with an annealing time longer than 2 hours. The average grain size for 2 hours of annealing time was 2.58 µm and for 240 hours of annealing time was 2.81 µm, indicating that the very subtle grain size increase with time. So, a transition from normal grain growth to abnormal grain growth or no growth was observed at the boundary of 2 hours of annealing time. On the other hand, as the annealing time increased, the grain size distribution became narrower. The porosity decreases with annealing time and pressure, and the minimum porosity is 4.37 % at 240 hours. There was no change in the grain aspect ratio by annealing, but the grain boundaries became straight and curved after annealing. There is no lattice preferred orientation (LPO).
Blackening was observed in all the samples prepared under high pressure. In previous study, the blackened samples suggested the presence of graphite at grain boundaries (Brooker et al., 1998). Impurities of graphite at grain boundary may inhibit grain growth resulting in an early transition from normal grain growth to an abnormal grain growth and no growth. Annealing time for 2 - 48 hours, the entire samples turned black. The upper half of the sample is white and only the lower half is blackened within the same sample at annealing time of longer than 173.5 hours. Grain size of the white part is bigger than one of the blackened part. Porosity of the white part is lower than one of the blackened part. This difference in grain growth may be related to the distribution of graphite within the sample and likely reflecting pressure gradient within the deformation apparatus. To synthesize homogeneously dense aggregates, we need to control impurity distributions in the aggregates.
Reference
-Brooker et al., 1998, Am. Mineral., 83, 985-994.
-Chopra and Paterson, 1984, J. Geophys. Res. Solid Earth, 89, 7861-7876.
-Karato and Jung, 2003, Philos. Mag., 83, 401-414.
-Ishikawa et al., 2014, Japanese Mag. Mineral. Petrol. Sci., 43, 100-107.
-Rybacki et al., 2006, J. Geophys. Res., 111, B03203.
-Stünitz and Tullis, 2001, Int. J. Earth Sci., 90, 136-148.
As starting material, we prepared fine-grained powder of labradorite single crystals (An50-70, Madagascar) by milling (Multi-Beads Shocker). Mean grain size is 0.29 µm. The milling was carried out at 3000 rpm for 180 seconds under conditions inhibiting agglomeration of fine particles and damage due to heating by milling. The powder was kept in an oven at 110℃ for more than one week to obtain dry powder. We used Griggs type deformation apparatus at Tohoku University for sintering under high pressure. Experimental conditions was 2 - 240 hours for 900℃, 1GPa, 1000℃, 1.35GPa to maintain the stability of plagioclase (Stünitz and Tullis, 2001). Also atmospheric annealing at 900℃ was also conducted. We observed microstructures using scanning electron microscope (FE-SEM) and electron backscattered diffraction (EBSD).
The grain size variation for 2 hours annealing time follows a lognormal distribution, but not for samples with an annealing time longer than 2 hours. The average grain size for 2 hours of annealing time was 2.58 µm and for 240 hours of annealing time was 2.81 µm, indicating that the very subtle grain size increase with time. So, a transition from normal grain growth to abnormal grain growth or no growth was observed at the boundary of 2 hours of annealing time. On the other hand, as the annealing time increased, the grain size distribution became narrower. The porosity decreases with annealing time and pressure, and the minimum porosity is 4.37 % at 240 hours. There was no change in the grain aspect ratio by annealing, but the grain boundaries became straight and curved after annealing. There is no lattice preferred orientation (LPO).
Blackening was observed in all the samples prepared under high pressure. In previous study, the blackened samples suggested the presence of graphite at grain boundaries (Brooker et al., 1998). Impurities of graphite at grain boundary may inhibit grain growth resulting in an early transition from normal grain growth to an abnormal grain growth and no growth. Annealing time for 2 - 48 hours, the entire samples turned black. The upper half of the sample is white and only the lower half is blackened within the same sample at annealing time of longer than 173.5 hours. Grain size of the white part is bigger than one of the blackened part. Porosity of the white part is lower than one of the blackened part. This difference in grain growth may be related to the distribution of graphite within the sample and likely reflecting pressure gradient within the deformation apparatus. To synthesize homogeneously dense aggregates, we need to control impurity distributions in the aggregates.
Reference
-Brooker et al., 1998, Am. Mineral., 83, 985-994.
-Chopra and Paterson, 1984, J. Geophys. Res. Solid Earth, 89, 7861-7876.
-Karato and Jung, 2003, Philos. Mag., 83, 401-414.
-Ishikawa et al., 2014, Japanese Mag. Mineral. Petrol. Sci., 43, 100-107.
-Rybacki et al., 2006, J. Geophys. Res., 111, B03203.
-Stünitz and Tullis, 2001, Int. J. Earth Sci., 90, 136-148.