Elasto-visco-plastic constitutive models of geomaterials

Using the generic term "geomaterials" to cover the main civil engineering materials - soils, rocks and concretes - we aim to provide engineers with all the information they need to make the right choice between the various laws or phenomenological-type behavior models that modern industrial codes make available, within the framework of calculations using the "finite element" or "finite difference" methods. And this choice is never trivial, between simplistic models that fail to take into account the major aspects of the behavior the engineer wants to describe, and complex models that require costly and time-consuming determination of the model's many mechanical parameters. Our starting point is a phenomenological analysis leading to the main visco-elasto-plastic models, which we will classify exhaustively. The general elasto-plastic formalism will be established and examples of behavior laws used in industrial calculation codes will be given. The problem of cyclic loading, which is important in practice, will be addressed. The limits of phenomenological models will be clarified. In addition, the fundamental question for engineers of the failure of these materials has been the subject of very significant advances over the last 20 years, and we will present a failure criterion detecting the different modes of ruin that can be observed in situ in a wide variety of forms (mudflows, liquefaction under earthquakes, landslides by mass translation or rotation, distributed cracking or localized fracturing of rocks, etc.). Major conclusions will be drawn for practical civil engineering.

Furthermore, in-situ geomaterials are most often in an unsaturated state and can therefore be considered as three-phase materials: solid granular skeleton, interstitial liquid (water, oil...) and interstitial gas (air, water vapor, natural gas...). While geomaterial behavior models have hitherto only taken into account dry or saturated states, it has long been established that their mechanical behavior can vary drastically with their degree of saturation. A clay, for example, will go from being a cohesionless clay slurry to a very cohesive soft stone, simply by varying the amount of water present in the material. A better understanding of soil-water-air couplings now makes it possible both to provide a framework for formulating this coupled behavior and, for the first time, to make realistic numerical models available to engineers.