Implications for critical state theory from isotropic compression of sand

Abstract

The paper steps back to one of the original propositions of critical state soil mechanics, the coupling of isotropic compression behaviour as the hardening law for a general yield surface first suggested by Drucker and coworkers in 1957, and examines the implications of such isotropic compression in the case of clean quartz sand. Two broadly different concepts are examined: (a) that normal compression only truly occurs at high stress and is associated with grain crushing and (b) that there are an infinity of normal compression loci (NCLs) regardless of stress level. Isotropic testing of reconstituted samples, for which the critical state locus had been independently determined using undrained shear testing of loose samples, shows that concept (a) is erroneous. Perhaps surprisingly, this does not turn out to be a limitation for critical state theory. An infinity of NCLs does not conflict with a generalized critical state view of soil, and abandoning the conventional λ−κ isotropic compression model is essential for the ideas of critical state theory to represent real soils. Generalized critical state theory simply requires adopting the state parameter ψ as an independent variable from the overconsolidation ratio.