Contents lists available at ScienceDirect Engineering Fracture Mechanics journal homepage: www.elsevier.com/locate/engfracmech Review of unloading tests of dynamic rock failure in compression Hongyu Wang a , Arcady Dyskin a, ⁎ , Phil Dight b , Elena Pasternak c , Ariel Hsieh b a Department of Civil, Environmental and Mining Engineering, School of Engineering, University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia b Australian Centre for Geomechanics, University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia c Department of Mechanical Engineering, School of Engineering, University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia ARTICLE INFO Keywords: True triaxial unloading test Shear stress End friction effect Fracture growth Fracture shape ABSTRACT Dynamic rock failure at the walls of an opening, so-called rock burst or strain burst is one of the major types of rock failure capable of causing human and financial loss. Initially the rock at the place of the future excavation is in a polyaxial stress state. When excavating an opening, the rock elements at the excavation boundary are overloaded in the tangential (with respect to excavation wall) direction and unloaded in the radial direction. This situation can be reproduced by the modified true triaxial test in which only one surface of prismatic rock sample is unloaded. This paper reviews the tests under true triaxial unloading condition and the corresponding failure modes and types (static or dynamic). In these tests the violent (dynamic) ejection of rock frag- ments and arc-shape fractures near the free face are often observed. However these tests are affected by friction between the sample and the loading platens. We analyse the end friction effect and show that unlike the conventional end friction effect produced in uniaxial and biaxial loading tests, the true triaxial unloading test induces additional shear stress needed to prevent sliding of the rock sample from the loading platens. This additional shear stress is shown to considerably affect the geometry of fractures. 1. Introduction The surface failure at the excavation surface can be nonviolent or violent, i.e. spalling (slabbing) and skin (strain) rockburst, respectively. According to the first analysis by Fairhurst and Cook [1], spalling is the failure process involves extensional splitting/ cracking during the excavation of deep tunnels under high-stress conditions. In the course of skin rockburst or strainburst, fragments of rock, usually in the form of thin plates with very sharp edges, are violently ejected locally from the rock surface [2,3]. Spalling and strainburst are more likely to occur in more massive rock types than in significantly jointed and fractured rock masses [4–7]. Spalling and strainburst represent a significant hazard and a danger to the safety of personnel and infrastructure in civil engineering ex- cavations and deep underground mines. Understanding of the rock failure behaviour near the excavation boundary is critical for the design of a tunnelling and under- ground mining. Excavation of an opening leads to the in-situ stress redistribution. The stress concentration near the excavation surface manifests itself in the reduction of the radial stress component, which is zero at the surface and a considerable increase of the tangential stress component, which may lead to slabbing and strainburst [2,6,8]. Many researchers have studied the rock failure behaviour near the excavation boundary by conducting uniaxial compression tests [9–11], biaxial compression tests [12–14] and surface instability tests [15–18]. https://doi.org/10.1016/j.engfracmech.2018.12.022 Received 29 August 2018; Received in revised form 27 November 2018; Accepted 18 December 2018 ⁎ Corresponding author. E-mail address: arcady.dyskin@uwa.edu.au (A. Dyskin). Engineering Fracture Mechanics xxx (xxxx) xxx–xxx 0013-7944/ © 2018 Elsevier Ltd. All rights reserved. Please cite this article as: Wang, H., Engineering Fracture Mechanics, https://doi.org/10.1016/j.engfracmech.2018.12.022