Integrating acoustic emission into a percolation model to evaluate crack distribution characteristics of heated granite subjected to rapid cooling

• Published in: Results in physics Vol. 38; p. 105600
• Main Authors: Zhou, Chunbo, Gao, Feng, Cai, Chengzheng, Wang, Zekai, Zheng, Wenqi, Gao, Xianghe
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.07.2022; Elsevier
• Subjects: Acoustic emission events; Crack distribution; Liquid nitrogen cooling; Percolation model; Spatial connectivity; Acoustic emission events; Crack distribution; Spatial connectivity; Percolation model; Liquid nitrogen cooling
• ISSN: 2211-3797; 2211-3797
• DOI: 10.1016/j.rinp.2022.105600

•Neighbor distances between AE events are evaluated by percolation model.•Main AE events with highest spatial connectivity are identifiedby percolation model.•Number, density, and complexity of cracks are increasedby liquid nitrogen cooling. To compare the distribution of cracks formed in the failure process of heated granite subjected to rapid cooling, compression tests (0–60 MPa) and acoustic emission (AE) monitoring were performed on heated granite (25, 200, and 400 °C) after water cooling and liquid nitrogen (LN2) cooling. Integrating AE events into the percolation model and counting the AE events with a cube covering method, the main AE events with the highest spatial connectivity were identified among the whole set of AE events. The count, density, and fractal dimension of the whole set of AE events and the main AE events were discussed to indirectly evaluate the crack distribution in the failure process. The results show that the most frequent, maximum, and minimum neighbor distances between AE events can be evaluated by the covering cubes with various lengths. The main AE events with the highest spatial connectivity should be characterized by short neighbor distances and most frequent neighbor distances. When the length of covering cubes is large enough to exceed the most frequent neighbor distances, the corresponding length is defined as the characteristic length. When the characteristic length is determined as the length of the covering cubes, the main AE events can be identified among the whole set of AE events. According to the distributions of all the AE events and the main AE events, LN2 cooling is proven to result in a more number, density, and complexity of cracks than water cooling, under high temperatures and high confining pressures.