Issue 49

C.Y. Liu et alii, Frattura ed Integrità Strutturale, 49 (2019) 557-567; DOI: 10.3221/IGF-ESIS.49.52 559 However, there is currently no standard or widely accepted concept for the evaluation of the rock brittleness index. Researchers hold different views for different research purposes. Morley [6] and Hetényi [7] argue that rock brittleness is characterized by low elongation or low strain value due to the lack of ductility or compressibility. Ramasy [8] defines brittleness as the loss of cohesion in a rock as it deforms within the elastic range. Similarly, Obert and Duaval [9] consider brittleness to be a phenomenon in which a material (such as cast iron or rock) breaks or only slightly exceeds the yield stress. Tarasov and Potvin [10] proposed that brittleness is the ability to self-maintain the macroscopic damage in the post-peak area under the compressive load due to the accumulation of elastic energy. With the development of rock mechanics, a lot of research has been done on the evaluation of rock brittleness, and the commonly used evaluation methods are shown in Tab. 1. Brittleness evaluation method based on the rock stress-strain curve In Tab. 1, the brittleness indices B 1 - B 4 are proposed based on the relationship between UCS (uniaxial compressive strength) and uniaxial tensile strength. According to the experimental results, Kahraman [11] believed that the penetration rate of the rotary drill bit has a strong exponential relationship with the brittleness indices B 1 and B 2 . Altindag [12] proposed the brittleness indices B 3 - B 4 based on the tensile-compressive strength curve to quantitatively evaluate the brittleness of rocks. These two indices are often used to predict the drillability of rocks. However, the brittleness indices B 1 - B 4 based on UCS and uniaxial tensile strength is not applicable to the analysis of rock brittleness under complex stress conditions. According to the post-peak stress drop rate and peak strength, Li Qinghui et al [13] proposed the brittleness index B 5 , which is based on the three standard coefficients for the post-peak stress drop, but only applies to a certain type of rocks, and what is more, a lot of experiments need to be done to obtain an accurate value. At the same time, Tarasov and Potvin [14] proposed the brittleness indices B 6 - B 7 based on the post-peak secant modulus and the pre-peak elastic modulus. Xia Yingjie believed that these two brittleness indices cannot effectively distinguish the brittle characteristics of different stress-strain curves [15]. R. Altindag [16] proposed the brittleness indices B 7 and B 8 based on the peak stress-strain and residual stress-strain relations. Hucka and Das [17] proposed the brittleness indices B 10 based on the ratio of recoverable strain to peak strain before the peak; however, these methods consider only a few mechanical parameters and cannot fully reflect the strain-strain process of the entire rock. Therefore, these methods require further improvement. Meng et al. [18] proposed the brittleness index B 11 based on the relative magnitude and absolute rate of the post-peak stress drop, and further verified the accuracy of the index by doing comparison test on different types of rocks under different surrounding rock pressures, but this index does not represent the pre-peak mechanical characteristics. Xia Yingjie et al. proposed the brittleness index B 12 based on the post-peak stress drop rate and the ratio of the elastic energy released by the instability failure to the total energy stored before the peak. Chen Guoqing et al. [19] proposed the rock brittleness index B 13 based on the post-peak stress drop rate and the stress growth rate between the pre-peak initiation stress and the peak stress. The method uses the stress growth rate between the initiation stress and the peak stress to characterize the pre-peak brittle state. At present, there are two effective ways to determine the initiation stress, one being acoustic emission and the other based on the strain inflection point of the crack volume [20]. Acoustic emission is often affected by noise, making it difficult to determine the moment of crack initiation. The second method, which determines the initiation stress through the strain inflection point of the crack volume, often depends on the mineral composition and particle size, so it is also difficult to determine an accurate value. In order to further verify the accuracy of the method proposed in this paper, the following section will make comparisons with the experimental data by Chen Guoqing et al. Brittleness evaluation method based on internal friction angle Hucka and Das [21] proposed evaluating the brittleness indices B 14 and B 15 of rocks considering the internal friction angle. At the same time, Tarasov and Potvin [22] found through experiments that the brittleness index B 13 is positively correlated with the rock fracture angle. However, the brittleness indices B 14 and B 15 only apply to the same type of rocks; and it is also difficult to obtain an accurate rock fracture angle. Brittleness evaluation method based on elastic modulus and Poisson's ratio Rockman et al. [23] proposed the brittleness index B 16 based on shale reservoir; however, the brittleness index B 16 only takes into account the elastic modulus and Poisson’s ratio, but ignores many important mechanical parameters. In order to obtain accurate parameters, mechanical experiments still need to be conducted on a lot of rock samples. All these factors limit the development of the brittleness index B 16 .