*Yuna Arai1, Haruki Miwa1, Yoshihiro Konno1,2, Yusuke Jin3, Takaomi Tobase4
(1.Systems Innovation,Faculty of Engineering,The University of Tokyo, 2.Graduate School of Frontier Science, the University of Tokyo, 3.National Institute of Advanced Industrial Science and Technology, 4.Electric Power Development Co.,Ltd.)
![](https://confit-sfs.atlas.jp/customer/jpgu2023/web/JpGU2023_OSPA_entrysign.png)
Keywords:CCS, CO2 hydrate, self-sealing, sand particle size, core flooding tests
Carbon dioxide capture and storage (CCS) is widely acknowledged as an effective measure to combat climate change. However, in Japan, it is uncertain whether a structural reservoir with cap rocks (aquifer storage) alone will provide sufficient storage capacity. Therefore, as a new storage method to complement aquifer storage, the possibility of using hydrate CCS, whereby CO2 is injected beneath the seafloor to form hydrate that serves as a cap rock, is being investigated. In most of the previous studies on hydrate CCS, CO2 was injected into a core made of silica sand that simulated a reservoir, and the formation behavior of hydrate was observed. These studies were conducted under ideal conditions, using only one type of silica sand, and the length of the cores was relatively short (5-15 cm), resulting in one-dimensional analyses. In contrast, this study utilized a more realistic sediment grain size distribution that is closer to that of offshore reservoirs and lengthened the cores to 75 cm in order to conduct an experimental analysis of the three-dimensional hydrate formation behavior and development of sealing function.
At a pressure of 7.5 MPa and a temperature of 8.5 °C, liquid CO2 was injected from the lower end of the core, and temperature and pressure were measured at various points in the core. The experiment involved using a mixture of Toyoura standard sand and Tohoku silica sand No. 8 for the simulated sediments and comparing the results with the use of only Toyoura standard sand. The findings of the experiment showed that the standard sand stopped fluid penetration, whereas the mixed sand could not stop fluid penetration. Furthermore, the position of hydrate formation in the core cross section was heterogeneous in the mixed sand. These results indicate that in mixed sands with a wide grain size distribution, the diameter distribution of CO2 flow path is also wide, and CO2 selects the thicker flow path, resulting in a heterogeneous in the position of hydrate formation and fluid penetration from areas where hydrate is not formed. Thus, to achieve hydrate CCS in real marine reservoirs, it is necessary to repeatedly inject CO2 to expand areas of hydrate formation and form a strong sealing layer to prevent leakage.