Investigation on Mechanical Behaviors of Sandstone with Two Preexisting Flaws under Triaxial Compression

• Published in: Rock mechanics and rock engineering Vol. 49; no. 2; pp. 375 - 399
• Main Authors: Huang, Da, Gu, Dongming, Yang, Chao, Huang, Runqiu, Fu, Guoyang
• Format: Journal Article
• Language: English
• Published: Vienna Springer Vienna 01.02.2016; Springer Nature B.V
• Subjects: Civil Engineering; Coalescence; Coalescing; Cracks; Defects; Earth and Environmental Science; Earth Sciences; Geophysics/Geodesy; Original Paper; Rock; Rock deformation; Sandstone; Shear; Shear strength; Stone; Stress-strain curves; Stresses; Stress–strain; Triaxial compression; Crack coalescence; Sandstone; Preexisting closed flaws; Strength; AUTODYN
• ISSN: 0723-2632; 1434-453X
• DOI: 10.1007/s00603-015-0757-3

Triaxial compression experiments on sandstone samples with two preexisting closed non-overlapping flaws were performed to investigate the deformation and strength behaviors. Three types of preexisting closed flaw pair in sandstone samples, i.e., parallel low-dip (type B), parallel high-dip (type C), and composite high- and low-dip (type D), were considered as the typical arrangements of the non-overlapping crack pair. A general rule has been found that the arrangement of the flaw pair has greater impact on the rock deformation, strength, and crack coalescence pattern than the confining pressure (5–20 MPa). Experimental results showed that, compared with intact sandstone samples, the postpeak stress–strain curves of flawed samples distinctly demonstrate stress fluctuation. In particular, the unique prepeak stress–strain curves of the specimens with a low-dip flaw pair (type B) present oblique Z-shape with a double-peak stress. The stress for crack initiation σ ci , the critical stress of dilation σ cd , and the peak strength σ c of precracked sandstone samples are significantly lower than those of intact rock. The present numerical study, which is an extension of the test analysis, focuses on identifying the crack nature (tensile or shear) and coalescence process. These simulated crack coalescence patterns are in good agreement with the laboratory test results. The cracks of the precracked samples that contained flaws with small inclination angle (associated with either type B or type D) generally initiate at the inner flaw tips and eventually lead to simple direct shear coalescence. However, complex indirect shear coalescence appears in the model containing a steep preexisting flaw pair (associated with type B specimen), even though no coalescence occurs when σ 3  = 5 MPa.