Investigation into the mercury intrusion porosimetry (MIP) and micro-computed tomography (μCT) methods for determining the pore size distribution of MX80 bentonite pellet

The pore size distribution (PSD) plays an important role in the hydro-mechanical behaviour of porous materials. In this study, the PSD of MX80 bentonite pellets under various imposed suctions from 262 to 1.5 MPa was determined by both mercury intrusion porosimetry (MIP) and micro-computed tomography (μCT). The void ratios e18–344 in the overlapping pore range (18–344 μm) identified by MIP and μCT were compared to clarify the difference between the two techniques. Results showed that the MIP-detection e18–344 nearly overlapped with the μCT-detection one at suction s ≥ 57 MPa, indicating the reliability of MIP and μCT results for pellets at s ≥ 57 MPa. However, the μCT-detection e18–344 became larger than the MIP-detection one at 9 MPa ≤ s ≤ 38 MPa, while the changes of MIP-detection and μCT-detection e18–344 with suctions were opposite at s ≤ 4.2 MPa. As the difficulty of interpreting the variation in μCT attenuation coefficient at low suctions undermined the reliability of μCT thresholding, the segmentation of μCT images was suggested to be determined by making the μCT-detection e18–344 equal to the MIP-detection one. Because the reliability of MIP results at s ≤ 4.2 MPa was reduced by the difficulty of keeping the integrity of freeze-dried samples, the MIP-detection e18–344 determined in a previous study without the sample integrity problem was chosen as reference for the μCT thresholding at 1.5 MPa ≤ s ≤ 4.2 MPa. The corrected μCT-detection e18–344 calculated by the new thresholds matched well with the MIP-detection e18–344, completing the μCT-detection at suctions smaller than 57 MPa. Since the MIP-detection PSD only covers the pores smaller than 344 µm, the μCT-detection PSD with pores larger than 344 µm was then integrated into the MIP-detection PSD, enlarging the MIP-detection to millimetric pores.