Experimental investigation on the fracture surface features of heat-treated red sandstone containing fissure under constant amplitude low cycle impact using 3D digital reconstruction

Fractured rocks as typical engineering materials are commonly exposed to temperatures and dynamic perturbations. In this paper, constant amplitude low cycle (CALC) impact tests were carried out on red sandstone containing a single fissure (0 degrees, 30 degrees, 60 degrees and 90 degrees) after thermal treatment (25 degrees C, 200 degrees C, 400 degrees C, 600 degrees C and 800 degrees C) utilizing a modified split Hopkinson pressure bar (SHPB). Subsequently, the fracture surfaces of the failure specimens were digitally reconstructed based on point cloud data captured by a three-dimensional (3D) laser scanning. And scanning electron microscopy (SEM) was also employed as a complementary tool to study the fracture micromorphology. The fractal and rough features of fracture surfaces based on the box counting method and image segmentation method were investigated, respectively. Meanwhile, four second-order statistics in the grayscale co-occurrence matrix (GLCM) of the digitized fracture surface after grayscale processing were calculated to describe the fracture surface texture. The best-fit multiple regression models between the fracture surface feature parameters and the interaction of temperatures and fissure angles were constructed and analyzed. Moreover, the relationships between the fracture surface features and the mechanical parameters as well as the intrinsic mechanisms were analyzed and discussed in detail. The results demonstrate that the fractal dimension (D-C), joint roughness coefficient (JRC), maximum elevation difference (M-D), roughness characterization factor (omega), and GLCM second-order statistics possess specific response patterns to the temperatures and fissure angles. Whether the sensitivity in portraying the fracture surface features to the interaction of temperatures and fissure angles or the reliability of the best fit multiple regression models, the shadow method based on image segmentation behaves more promisingly. Furthermore, conspicuous correlations can be found between both the feature parameters including the GLCM second-order statistics and the failure strength, strain rate as well as life. The high temperature activates thermal activation mechanisms that alter the potential crack sources and pathways in the grain and rock skeleton, affecting the microscopic fracture face morphology. Finally, the maximum tensile stress at the edge of the prefabricated fissure might be responsible for the angle effect exhibited by the feature parameters.