Contents lists available at ScienceDirect International Journal of Rock Mechanics & Mining Sciences journal homepage: www.elsevier.com/locate/ijrmms Technical Note The shape effect on the morphology of the fracture surface induced by the Brazilian test Arash Khosravi a, ⁎ , Richard Simon a , Patrice Rivard b a Department of Civil, Geological and Mining Engineering, École Polytechnique de Montréal, Montréal, QC, Canada H3T 1J4 b Department of Civil Engineering, University of Sherbrooke, QC, Canada J1K 2R1 ARTICLE INFO Keywords: Roughness Brazilian Failure mode Length-to-diameter ratio Loading direction 1. Introduction Fractures are one of the most important features affecting the mechanical behaviour of rock masses. The degree of fracturing in a rock mass can be influenced by several factors: in situ stress, boundary conditions and geological history. Fractures are the main factor that must be taken into consideration in the control of thermo-hydro- mechanical behaviour. Two important physical characteristics of fracture that are used in rock mass classification systems are: the aperture and the roughness. Many studies have been conducted on these two components to show their relation with the mechanical behaviour. 1–3 In applied rock mechanics, the shear strength of discontinuities is the most important parameter for stability analyses. To estimate the shear strength, the surface roughness can be taken into account to model the shear behaviour of fractures. 4,5 Fracture roughness has been quantitatively characterized based on morphological approaches, such as empirical, 6 statistical 7,8 and fractal approach. 9–11 Each of these approaches uses parameters that quantify the fracture topography. Among them, empirical approaches are the most commonly used in rock mechanics. The Joint Roughness Coefficient (JRC) – as an empirical approach – is a widely used roughness parameter. 12 Since JRC is a subjective parameter that is estimated by comparing the roughness profile with ten standard roughness profiles, 13 relationships have been proposed between the JRC value and various roughness parameters used in statistical approaches. These relations quantitatively express the roughness parameters. 8,14–19 Brady and Brown 20 showed that fractures occur in three conditions or modes: shear, tensile or a combination of both. Mode I defines the fracture created in tension, while Mode II and III are the fractures induced by both tensile and shear stresses (Mode II as in-plane shear and Mode III as out of plane tear). These fracture modes can be created in laboratory using direct or indirect tensile, uniaxial and triaxial compression tests. The fracture surface induced by an indirect tension test (Brazilian test) is created in Mode I. Fig. 1 shows the roughness variation of these fracture modes on Gabbro samples. There are many experimental and numerical studies for the stress and strain distribu- tion in the samples split by Brazilian tests. 13,21–27 More recently, several investigations have been conducted on the cracking patterns of foliated and pre-notched samples with different arrays. 28–33 Stirling et al. 34 performed Brazilian tests based on ASTM, ISRM and Flattened loading methods and identified the location of crack initiation and fracture patterns. They showed that the onset location of cracking is clearly at the center; however, the crack propagation could not be captured by Digital Image Correlation (DIC) in these tests. Apart from the ultimate stress applied on samples submitted to Brazilian tests of different L/D ratio, the strain variation along the loading direction was also investigated by Stirling et al. 34 Accordingly, the degree of the deformation perpendicular to the loading axis increases across the diameter as L/D ratio decreases. This may be due to the confinement induced by the length of the specimen. The results of this study showed that the effect of the strain distribution in terms of the L/D ratio is more tangible on the fracture surface along the loading direction. To date however, little work has been done on the relationship between the fracture roughness and loading modes. Seredin et al. 35 http://dx.doi.org/10.1016/j.ijrmms.2017.01.007 Received 14 February 2016; Received in revised form 8 January 2017; Accepted 9 January 2017 ⁎ Corresponding author. E-mail address: akhosravi53@gmail.com (A. Khosravi). International Journal of Rock Mechanics & Mining Sciences 93 (2017) 201–209 1365-1609/ © 2017 Elsevier Ltd. All rights reserved. MARK