Experimental study on tensile strength and splitting failure characteristics of rocks-like with eccentric cracks
In order to explore the mechanism of eccentric cracks influence on the tensile mechanical properties of rock,some rock-like samples containing pre-crack with different eccentric distances and dip angles are prepared by a self-designed processing device.The flattened Brazilian splitting test and non-contact full-field strain monitoring test on these rock-like samples are carried out.Based on the load-dis-placement curve and strain field evolution characteristics,the effects of eccentric distances and dip angles of pre-crack on deformation,tensile strength and splitting failure are analyzed to reveal the deterioration law of tensile strength and the causes of splitting failure.The results show that the surface strain of the sample is basically unchanged in the compression and approximate linear elastic deformation stages,which changes from slow increase to sharp increase occurring in the stable and unstable crack propagation stages.Under the same crack dip angles,the tensile strength of the rock-like samples contai-ning eccentric pre-crack increases with the increase of the eccentric distance.In addition,the effect of the eccentric distances on the deterioration of tensile strength is much greater than that of the crack dip angles.The eccentricity of prefabricated cracks is the main control factor affecting the splitting failure of the sample.With the increase of eccentric distance,the less horizontal and vertical secondary damage and failure outside the main damage and failure,and the simpler the failure mode,which can be classi-fied into three types:Z-shaped,one-shaped and central splitting failure.The tensile stress induces the splitting main failure and the vertical secondary failure,while the horizontal secondary failure is induced by the compressive stress.Moreover,the smaller the eccentric distance,the more significant the effect of stress concentration on the tip of the pre-crack on the splitting failure.
万方数据
维普
