Civil Engineering Departmental Seminar - Building a Better Mousetrap – The Continuing Quest for More Reliable Models of Clay Behavior

Building a Better Mousetrap – The Continuing Quest for More Reliable Models of Clay Behavior

Extensive prior research has shown that advances in modeling complex aspects of clay behavior, from non-linear deformation and anisotropic shear properties, can enable more reliable predictions in practical geotechnical applications (from foundations to tunnels).  Much of this work has been achieved using the prevailing paradigm of elasto-plasticity and conceptual framework of critical state soil behavior for clay sediments.  This lecture compares and contrasts recent research to advance predictive capabilities in modeling the rate dependent behavior of clays and in simulating the behavior of residual soils.

There are many (often contradictory) observations of the viscous properties of sedimentary clays, and there is a long-standing dilemma regarding the coupling of creep and consolidation processes.  We have addressed these issues through a formulation that introduces a physically-based evolution law that attributes the macroscopic viscoplastic strain rate to an internal variable associated with the prior strain rate history. By varying the rate-sensitivity parameter the proposed model provides a unified framework that can describe a wide range of observed time-effects in 1-D compression and shear behavior.

Residual soils are formed through in-situ physical and chemical weathering processes.  Our experimental investigations of Old Alluvium led to a conceptual model of its complex microstructure and degradation under mechanical stress. Using heuristic reasoning we have described these transitions using Cation Exchange Capacity as an internal state variable and are able to develop a consistent model to explain transitions of engineering properties within the weathered soil profile.