A review of the state of the art in modelling progressive mechanical breakdown and associated fluid flow in intact heterogeneous rocks

The purpose of this review is to present techniques, advances, problems, and new developments in modelling the progressive mechanical breakdown of, and associated fluid flow in, intact heterogeneous rock. In general, the theoretical approach to this physical process can be classified into the three categories of discrete models based on fracture mechanics, the continuum damage mechanics approach, and statistical approaches. This categorisation forms the skeleton of this article.Recognising that intact rock contains ubiquitous cracks and flaws between grains and particles of various shapes, fracture mechanics has been widely used to study the mechanical breakdown process in terms of the growth of these discrete defects. Two types of fracture mechanics models, namely open crack and sliding crack models, are used to simulate the progressive microfracturing of rock upon loading, and the application of these to modelling mechanical breakdown is discussed in Section 2.As an alternative to the explicit treatment of cracking given by fracture mechanics, continuum damage mechanics (CDM) takes a phenomenological route that considers the averaged effect of microstructural changes, and has found widespread use in simulating macroscopic stress–strain responses. Through the introduction of damage variables as state operators that quantify change in mechanical properties such as stiffness and strength with respect to damage, CDM models are capable of reproducing realistic hydro-mechanical responses during rock disintegration. This method is discussed in Section 3.Recognising that intact rock is never strictly an isotropic and homogeneous material, statistical approaches use, in general terms, statistical distributions as a means of describing the variation of material properties in natural rock. These approaches have enjoyed a great deal of attention in past decades. By employing different numerical schemes, three major statistical models have been developed: continuum-based damage mechanics models, particle models, and network/lattice models. The merits and drawbacks of these models are discussed in Section 4.Coupled deformation and pore fluid diffusion can be important in the process of progressive breakdown. This poromechanical effect involves the interaction between the solid constituents of, and the interstitial fluids in, heterogeneous rocks under those circumstances when mechanical perturbation occurs sufficiently rapidly that induced pore pressure changes cannot fully dissipate. The mechanisms and analytical approaches, and their development relevant to progressive mechanical breakdown and fluid-flow modelling, are outlined in Section 5.Finally, based on this review, remarks are made summarising the current progress of, and fundamental problems with, these developments. Ideas for further improvements towards more comprehensive and robust techniques are suggested.