[S-CG46] New Insights of Fluid-Rock Interactions: From Surface to Deep Subduction Zone

[SCG46-P10] Nondestructive measurement of water content from neutron imaging of rock samples using RANS

*Misaki Hasegawa¹, Tomoaki Morishita¹, Tomohiro Kobayashi², Takaoki Takanashi², Yasuki Okuno² (1.Kanazawa University, 2.RIKEN)

Water content and distribution of water components in rocks are fundamental data for understanding rock-water reactions. The conventional methods for measuring water content in rocks is to measure the difference in weight before and after sample heating. In order to perform these analyses, the sample is generally powdered, which results in the loss of information on the distribution of water content in the sample. Neutrons interact with and attenuate the nuclei of materials as they penetrate them. Since neutrons are characterized by their high attenuation for hydrogen, the amount of neutrons in a rock sample after neutron irradiation is expected to reflect the amount and distribution of water components, including hydrogen. In this study, I investigated the measurement of water content in rocks using neutron imaging obtained on plate-like samples (rock sample with a few cm squares and about 1 cm thick). In this study, the RIKEN compact neutron source system (RANS) of RIKEN was used as the neutron source. First, water samples of different thicknesses (0.1, 0.2, and 0.5 mm) in quartz containers were examined with respect to water thickness and neutron transmission rate. Next, neutron imaging was performed on seven samples of serpentinite, chlorite rock, two types of hornblendite, potterite, and two types of peridotite. The relationship between the neutron absorption coefficient (the logarithm of the neutron transmission coefficient divided by the thickness of the sample) and the loss on ignition (LOI) was investigated for each sample. In neutron transmission imaging, it is necessary to take into account the effects of neutron scattering and other factors influenced by the size and shape of the transmitted sample: region of interest (ROI) to use in the neutron imaging image. I performed the neutron scattering situation and several ROI regions from neutron transmission images. It was assumed that a decrease in neutrons occurred at the outer edge of the sample for the measured sample size, and that as the water content increased. The difference in neutron transmission values between using the entire sample area in the image (entire ROI) and using the area near the center (center of ROI) became larger. The analysis was performed by setting the ROI at the center of the sample. The transmittance decreases linearly as the thickness of the water in the quartz container increases (0.1~0.5 mm). The results for natural samples are also basically correlated with the neutron absorption coefficient for each sample and the LOI. I used the results of neutron transmission coefficients of water in peridotite and serpentinite samples to estimate the water content in serpentinite in a preliminary manner. At the present stage, the water content estimated by neutron imaging is not in good agreement with the conventional measurement of the strong thermal water content. This might be partly due to the difference between the water thickness estimated for the natural rock samples and that of the standard sample in quartz container. Further study is needed for the water content measurement by neutron imaging.