A technique for three-dimensional characterisation of asperity deformation on the surface of sheared rock joints

Abstract

1. IntroductionThe surfaces of rough joints consist of asperities, which play anessential role in the mechanical and hydraulic behaviour of thejoints. Numerous laboratory investigations [1–4] have describedthe asperity damage that can occur as a result of attrition (over-riding of asperities) and/or asperity breakage caused by over-stressing upon shearing. As a consequence, the degraded materialsform gouge in the joint interface that affects the shape of thesurface and subsequent response of the rock joints [5–8]. In orderto improve our understanding of the mechanical and hydraulicbehaviour of rock joints, the asperity damage and gouge thataccumulates on surfaces must be visualised and appropriatelyquantified. The characterisation of asperity damage evolution andgouge accumulation on the joint surfaces is important for a varietyof fields including applications in mining, tunnelling, petroleumengineering, rock slope engineering and earth sciences.Asperity deformation has been directly characterised by asses-sing the surface morphology of a joint before and after shearing[9–15]. Ladanyi and Archambault [9] proposed a simple character-isation of asperity damage by measuring the ratio of the damagedasperity area over the nominal area of the joint surface. Later,Homand et al. [10] and Yang et al. [11] modified Ladanyi andArchambault's method for characterising irregular joint surfaceasperity damage by comparing the entire area of joint before andafter shearing. The deformation of joint surface asperities has beencharacterised by comparing a theoretical variogram model of jointsurface before and after shearing [12]. Two dimensional (2D) imageanalyses on sheared joint surfaces were used by Riss et al. [13] andGentier et al. [14] to characterise the asperity deformation associatedwith shear direction. Asperity def ormation has been directly charac-terised by evaluating the critical apparent dip angle from a 3Ddigitised joint surface [15]. In contrast, asperity deformation has beenindirectly characterised by assessing the variation of the dilation angle[1,5,6,16]. Also, Barton [17] indirectly described asperity deformationthrough the concept of roughness (friction) mobilisation. Morerecently, asperity damage has also been studied on the basis ofnumerical modelling [18,19].As discussed by Pereira and de Freitas [3], Plesha [5], Leong andRandolph [6], Olsson and Barton [7], and Zhao et al. [8], the depthof asperity damage and thickness of accumulated gouge on thesurface of a joint could affect the mechanical behaviour and thehydraulic behaviour of a joint upon shearing. However, as dis-cussed in Table 1, the previous direct methods of characterisationwere mainly focused on asperity damage, while only a limitedamount of attention was given to the accumulation and distribu-tion of gouge on sheared joint surfaces. Moreover, most of thesedirect methods only characterise the asperity deformation featuresin two dimensions (i.e., the area of damage) and not in threedimensions to capture the depth of asperity damage and thicknessof accumulated gouge. Therefore, these parameters on the shearedsurfaces need to be characterised in order to enhance our under-standing of their influence on shear behaviour.In addition, most previous studies have been conducted using aconstant normal load (CNL) boundary condition. There have beenonly a limited number of studies conducted based on the constantnormal stiffness (CNS) boundary condition [6,20–22]. In the CNSapproach the normal stress varies during shearing when the jointsundergo dilation or compression under a constant normal stiffnessthat represents the surrounding rock mass. The CNS boundaryContents lists available at ScienceDirectjournal homepage:www.elsevier.com/locate/ijrmms