Experimental observations from the rock scratch test in sedimentary rocks present compelling evidence that the uniaxial compressive strength (UCS) can be estimated from the specific energy (SE). However, large variations of the SE/UCS ratio are also observed in various types of rocks which is related to the different failure mechanisms during the test. The failure mechanism transition from ductile to brittle is controlled by the critical cutter depth which varies among different types of rocks. It is essential to quantify the correlations between the types of rock characterized by UCS/UTS and the critical cutter depth before using the scratch test to estimate the compressive strength. In this work, the scratch test is modeled numerically using the three-dimensional discrete element method (DEM) with an innovative displacement softening contact model to simulate rocks with different compressive to tensile strength ratios ranging from 5 to 30. Then a baseline simulation of the scratch test is conducted to benchmark geometric parameter. The failure mechanism transition from brittle mode to ductile mode is numerically modeled. The same loading configuration is then applied to investigate the effect of strength ratios on the failure mechanism transitions.
In the past decades, experimental observations (Richard, 1999; Richard et al., 2012) from the scratch test or the rock cutting test in sedimentary rocks present compelling evidence that there exists a correlation between the uniaxial compressive strength σc and the specific energy ε determined from the slope of the mean cutting force versus the cross-sectional area of the cut, i.e., the input energy per unit volume of cut, when the test is conducted in a ductile-dominant mode of failure associated with continuous decohesion and plastic flow at shallow cut. With the benefits of the test being non-destructive and can yield continuous results along the length of cut, the scratch test has been employed in evaluating various mechanical properties of rock, such as elastic modulus (Schei et al., 2000), rock heterogeneity (Suàrez-Rivera et al., 2002), and strength parameters (Mitaim et al., 2004).