On the Residual Strength of Rocks and Rockmasses

Abstract

The design and construction of structures in rock depend heavily on knowledge of rock strength and deformation characteristics (Cai et al. 2007). A commonly adopted method for assessing these characteristics is to utilize standard laboratory tests, including triaxial compression tests (ASTM 2015). Since the development of stiff servo-controlled testing machines and procedures in the 1960s and 1970s (see, for example, Cook 1965; Rummel and Fairhurst 1970; Wawersik and Brace 1971; Hudson et al. 1971), such tests have been capable of maintaining stability beyond peak strength and recording post-peak material parameters (Brady and Brown 1985). Generally speaking, the stress–strain behaviour of rocks as observed in compression tests can be separated into four phases: (1) pre-yield, where behaviour is approximately elastic; (2) post-yield but pre-peak, where short term frictional strengthening effects lead to an artificial strain-hardening behaviour; (3) post-peak weakening, where significant cohesion loss during fracture growth and shear leads to an overall weakening of the rock material; and (4) the residual state, where the strength and deformation characteristics become approximately constant with further straining (Martin 1997; Diederichs 2003). Material properties (e.g., strength, stiffness) do not change significantly during the first (pre-yield) and last (residual) phases of deformation, whereas the middle phases between the onset of yield and the attainment of residual strength are associated with changing material parameters as the rock undergoes progressive damage processes. The focus of this study is on the final phase of rock deformation, and in particular on the residual strength characteristics of rocks and rockmasses. The residual strength of rock is relevant in several engineering contexts, such as the assessment of stability of small and large-scale excavations or the collapse/flow properties of rock in mining operations (Labrie 2017; Peng et al. 2017). For instance, the stress–strain response of the rockmass surrounding a tunnel will vary significantly depending on its residual strength, and this should be accounted for in the design of appropriate support and reinforcement (Alejano et al. 2009). It has long been known that residual strength can be determined from triaxial tests, and several authors have suggested that residual strength is a material property which depends on the type of rock considered (Hobbs 1966; Jaeger 1969; Kovari and Tisa 1975). Although no universally adopted definition of residual strength exists, in this study, we adopt the approach commonly applied in laboratory testing studies (e.g., Jaeger 1969; Kovari and Tisa 1975; Martin 1997; Walton et al. 2015) of estimating the residual strength as the stress level at which the post-peak stress becomes approximately constant following the initial loss of strength; examples are presented in a subsequent section of this paper. Few studies concerned with the estimation of the residual strength of rock and rockmasses are available in the literature. The main aim of this study is to present a first analysis of a large database of peak and residual strength data of different rocks. Furthermore, variability in strength characteristics across different rock types is interpreted, and multiple approaches to estimate residual strength are assessed.