Constitutive Modeling of Inherent Sand Fabric Anisotropy and Numerical Applications within Critical State Soil Mechanics

University Of California-Davis, Davis CA

Investigators

Abstract

PI: Yannis Defalias Institution: University of California at Davis Title: "Constitutive Modeling of Inherent Sand Fabric Anisotropy and Numerical Applications Within Critical State Soil Mechanics" It is well known that natural soil deposits are inherently anisotropic due to their deposition and its effect on the ensuing fabric. In particular, for sandy soils the fabric induced inherent anisotropy may have dramatic effects on the mechanical response of the soil to subsequent loading in the range of plastic deformations. Otherwise identical sand samples (same sand, same void ratio, etc.) may behave in totally different ways if prepared by different methods; or, if they are identically loaded in different directions, as if they were samples of very different densities. The constitutive modeling of such inherent sand fabric anisotropy and its effect in the numerical analysis of geo-structures is the main objective of this research award. In order to address the main objective, a sand constitutive modeling platform will be required. This platform is based on an existing sand constitutive model (Manzari and Dafalias, 1997; Li and Dafalias, 2000) within the framework of Critical State Soil Mechanics (CSSM), which despite its simplicity has been proved capable of modeling the sand response under a large variation of densities, confining pressures and monotonic or cyclic loading conditions with the same set of model constants, but, with one exception. It cannot account for the effect of inherent initial anisotropy due to fabric formation. And such effect is often truly drastic, as shown experimentally. The incorporation of anisotropy in the theory of plasticity has a well understood theoretical framework in terms of the so-called joint isotropic invariants, but if used in its utmost generality, it will produce extremely complicated constitutive models which are not feasible tools for practical applications and analysis. In order to avoid such complications, the inherent anisotropy effect will be introduced in the chosen constitutive platform by a very simple method. The main theoretical tool, motivated by micromechanical observations on particles and voids orientation distribution, will be the introduction of a scalar-valued anisotropy parameter A, which will effect some, but not all, of the basic features of the platform constitutive model in order to achieve the simulation of anisotropic behavior. A number of different avenues will be investigated, and preliminary results show that this is a very promising approach, despite its simplicity. In the process the fundamental premise of CSSM on the uniqueness of the Critical State Line in the void ratio-confining pressure space is questioned, and instead, its dependence on fabric via the anisotropic parameter A is proposed. Once the inherent anisotropy has been successfully modeled, the second important step will be to evaluate its effect at the level of the response of a geotechnical structure subjected to different loadings and analyzed by numerical implementation of the proposed constitutive model. It is expected that the effect inherent anisotropy has on the dilative and contractive characteristics of sand, will have a dramatic effect on the overall geo-structural response under monotonic and cyclic loading. Besides the plausible impact of the research findings on the specific area of soil constitutive modeling and analysis, some broader impact characteristics will emerge. It is of great importance to increase our ability to use simulations of natural phenomena in the modern era of computations, in order to address societal needs for optimum and safe design of infrastructures. Such phenomena are not only geo-mechanical, but the latter are integrated in a larger network of other systems. Thus, the poor performance of a simulative component such as a geo-mechanical simulation, is detrimental to the overall performance of the larger system of phenomena simulations (e.g. in seismic design). The provision of a reliable soil modeling approach that accounts for the important issue of fabric effect, is in fact a contribution to better overall performance of this broader system of simulations of natural phenomena. Another broader impact the successful completion of this award will have, is educational at large scale. Despite the advanced character of the research, the central idea of the dependence of the critical state line on a scalar-valued measure of fabric anisotropy has an inherent simplicity that can easily be conveyed to undergraduate students of geotechnical engineering and constitute a standard addition to textbooks of soil mechanics everywhere. Finally, its incorporation in many different modeling techniques the geotechnical profession uses (and not only the modeling platform used in this research) will be a simple matter once the main findings are published and distributed.