While swelling of clays (as well as its counterpart, shrinkage) is a problem in many parts of the U.S. (middle and western parts) and in many other countries around the world (e.g., Scandinavian countries, Canada and South Africa), there is currently no reliable method for the quantitative evaluation needed in geotechnical design. The empirical methods based on plastic limits and water contents have failed in a large number of cases. Expansive soils are normally rich in clay minerals such as montmorillonite. Swelling occurs when sufficient amount of water fills the inter-particle and intra-particle spaces. While the quest for the science of swelling is as old as modern soil mechanics, the past research has led to more questions than answers. For example, the relative contributions of double-layer repulsion and forces due to hydration of clay mineral surface (the structural forces) to the overall swelling have become controversial after some of the recent studies. Without resolving this fundamental issue, a rational correlation for use in practice cannot be established. With the recent advances in computer technology, the molecular dynamics (MD) and the discrete element analysis methods (DEM), it is now possible to unequivocally resolve this issue, and to develop a predictive model for clay swelling based on the actual physics. Introduced by Alder and Wainwright in the 1950s, the principle behind MD remains almost the same, but the methods of accounting for the interactions between atoms have become very sophisticated over the years. MD has been used to study the behavior of liquids, gases, electrolytes, micelles, colloids, crystal structures, sorption in porous media, metals and proteins, just to name a few. In this project, MD will be used to quantitatively calculate the contributions of double-layer repulsion and structural forces to the overall swelling force, and develop equations for quantifying them as a function of system variables. DEM will then be used to bridge the micro and macro scales. In DEM, a numerical specimen comprising a collection of clay particles is assembled and loaded just as a real specimen is loaded in a laboratory testing setup. The accuracy of the output results depends on the accuracy with which the interparticle forces are quantified. The PI and his students have pursued this technique for the past two decades and developed methodologies for quantifying such forces as the mechanical forces, double-layer repulsive forces and van der Waals attractive forces. Based on the proposed MD study, a methodology will be established for quantifying the structural force and used to calculate the structural force in the DEM study. A series of laboratory swelling experiments are planned for verification purposes. Based on the proposed study, simple, scientific methods based on measurable basic characteristics of soil such as the composition, cation exchange capacity and specific surface will be developed for use in geotechnical engineering practice. Several items are planned for the broader impact purposes. The most important among these is the plan to develop a demonstration module based on expansion of montmorillonite and computer simulations, and to use this module in courses at Johns Hopkins University, selected high schools in Baltimore, and at the Maryland Science Center to educate on the adverse effects of expansive clays and how the results from advanced research are used to mitigate them. The research team will consist of a recent doctoral degree graduate, one graduate student, several undergraduate students, a few high school students and the PI.
