Reappraisal of soil extrusion for geomechanical characterisation

Abstract

Reverse extrusion has been promoted over the past four decades for remoulded undrained shear strength (s u) measurement and consistency-limit determinations. The technique employs a uniaxial-compression test machine to set the die travelling against the confined soil specimen (billet) at a constant displacement rate (v), with the soil extruded through the die orifice under a steady-state extrusion pressure (p e). This paper presents the first comprehensive literature review of this topic, including a reassessment of extensive data sets presented for many hundreds of fine-grained soils covering an extremely wide plasticity range. Specifically, the paper critically examines (a) the soil billet’s assumed undrained condition and hence the constancy of the p e/s u ratio value, which is central to s u determinations using this approach; (b) the dependence of the steady-state p e value on both the billet’s area reduction ratio (R) and v; (c) the role of soil remoulding toughness; and (d) the pitfalls of various data analysis and interpolation techniques employed. The author concludes that, depending on mineralogy and gradation, localised billet consolidation may occur for the slow displacement rates employed and high p e values required, particularly for stiffer soils, such that the present extrusion approach is generally not recommended for s u measurement or Atterberg-limit determinations.