Subjectivity in Crack Initiation Measurements Using Conventional Stress-Strain Based Methods

Abstract

ABSTRACT Stress-strain based methods are frequently adopted in the literature to estimate crack initiation (CI) strength in rock. However, despite their widespread use, these methods can introduce subjective interpretations of the stress-strain curves, particularly in rocks with pre-existing microcracks or high porosity. This study aims to examine the subjectivity inherent in strain-based methods and its potential impact on consistent measurement of rock CI. A survey is then conducted using experimental data obtained from sandstone and granite specimens. Results suggest a wider scatter of CI values for sandstone samples, which aligns with expectations due to their inherent pore structure heterogeneity. The Lateral Strain Response (LSR) method demonstrates the least coefficient of variation, indicating less subjectivity. Further research and validation is recommended to improve confidence in the LSR method using parallel measurement techniques. INTRODUCTION In rock engineering, the crack initiation (CI) stress threshold refers to the stress level where new microcracks start to develop within the rock matrix. In practice, it has been demonstrated that the CI stress level can be used to estimate the strength of rock against spalling and rockburtst type failures in deep underground excavations (Martin, 1997; Diederichs, 2007). Further, the CI stress also provides an indication of the stress conditions where breakout can occur in boreholes and petroleum wellbores. The CI stress is usually recognised to be around 30 to 60% of the peak value measured during an Unconfined Compressive Strength (UCS) testing (Brace et al., 1966, Bieniawski 1967, Holcomb & Costin, 1986; Cai et al., 2004; Nicksiar & Martin 2013, Wen et al., 2018) There is a wide range of methods that are available to estimate the CI stress which includes several strain-based methods (SBMs), acoustic emissions (Keshtgar & Modarres, 2013), electrical resistivity (Lataste et al., 2003), digital image correlation (Dong & Pan, 2017), optical diffraction patterns (Li et al., 2002), laser speckle interferometry (Ennos, 1975), and ultrasonic probing (Mi et al., 2006). The strain-based methods are most common due to their low cost and simplicity (Mutaz et al., 2020). With these methods, the CI stress is interpreted from a characteristic change in the stress-strain response that is measured during compression loading.