Micro-macro mechanics of damage and healing in rocks

Abstract

This paper presents the state of the art of the theory of rock damage and healing mechanics, with a particular emphasis on the strategies available to relate the micro-scale of crystals, cracks and pores to the scale of a Representative Elementary Volume (REV). We focus on mechanical degradation and recovery of stiffness and strength. Damage and healing models formulated in the author's group are used as examples to illustrate and compare the reviewed micromacro approaches, which include fabric enrichment, micromechanical formulations and homogenization schemes. This manuscript was written for doctoral students or researchers relatively new to the field of damage mechanics of geomaterials. Equations are provided to explain how to formulate a thermodynamically consistent model from scratch. Reviewing damage and healing modeling strategies led to the following conclusions: (i) The framework of hyperplasticity, which does not require any postulate on the existence or uniqueness of yield functions and which automatically ensures thermodynamic consistency, was never applied to Continuum Damage Mechanics (CDM). There may be an avenue to improve state-of-the-art damage and healing models in a similar framework of "hyperdamage mechanics". (ii) In damage softening models, the mesh dependence of the width of the damage localization zone is currently alleviated by non-local regularization. Perhaps the next step is to couple micro-macro damage and healing models at the REV scale to discrete fracture mechanics at a larger scale to understand how damage and healing localization occurs. (iii) There may be an opportunity to use fabric-enriched models to capture the effect of microstructure organization on both mechanical properties and permeability. (iv) Coupling chemomechanical damage and healing processes across the scales would be useful to model the competition between damagre and healing whenever both can occur at the same temperature and pressure conditions. (v) Many challenges still exist to implement healing models in the Finite Element Method, especially in regards to the mapping of net damage.

Coupled Brittle and Viscous Micromechanisms Produce Semibrittle Flow, Grain‐Boundary Sliding, and Anelasticity in Salt‐Rock

“…Additionally, as water is almost always present in natural salt rocks (Roedder, 1984), elastic deformation may be dependent on loading rate with pronounced hysteresis at low strain rates in between stress cycles. Numerical modeling of salt caverns should consider the time‐dependent behavior of salt and the consequent damage and healing processes, which will affect the mechanical behavior and sealing capability of salt caverns (Arson, 2020; Shen & Arson, 2019b).…”

A Time-Dependent Directional Damage Theory for Brittle Rocks Considering the Kinetics of Microcrack Growth

“…( 17) follows naturally from the microcrack distribution density function ρ(𝜔, a) under the noninteracting homogenization. Here, a keydifference from the generally employed microcracking-induced damage definitions is that the distribution density function ρ(𝜔, a) is incorporated in its general form through functional analysis-based formulation rather than using scalar (Kachanov 1982;Fanella and Krajcinovic 1988) or tensorial (Arson 2020) approximations. Substituting Eq.…”

Predicting the Non-Deterministic Response of a Micro-Scale Mechanical Model Using Generative Adversarial Networks

“…In numerical modelling, identifying the cases that need a micro-scale description is important to advance towards more efficient and accurate multiscale models [ 2 ]. Some examples of materials with an important micro-structural contribution to the bulk response are listed: masonry structures [ 3 , 4 ], coal reservoirs for methane production [ 5 , 6 , 7 ], shale rocks [ 8 , 9 ] as well as other rocks [ 10 ] and brittle porous materials governing seismic events [ 11 ].…”