Interaction between model bored piles and swelling London Clay

The aim of this research is to investigate the interaction between model bored piles and swelling clays. Pile foundations are commonly used in expansive clays to isolate buildings from subsidence and heave. The research is unique in that it is the first time that the effects of swelling London Clay on bored piles have been studied using model tests.
London Clay was obtained from BRE’s pile test site in Kent and tested to determine its properties. Instrumented model piles and swelling apparatus were designed. Tubular steel piles were grouted into holes drilled into compacted desiccated clay, within a special Rowe cell. Water was introduced and changes in the clay and forces and strains in the anchored pile were measured. A long-term model testing programme including two swelling and two pile tests was undertaken.

The test results show the nature of the interaction between the model piles and swelling clay. Three stages of heave were shown by ‘free-field’ swelling clay, including a ‘tertiary’ phase, similar to a ‘wetting wedge’ hinged above the clay and opening at three different rates. The model pile affected the swelling clay, restricted heave to a distance of about 2 pile diameters, and produced ‘bulbs’ of swelling pressure. Average shear stresses along the pile/clay interface reached a maximum when the average heave displacement was about 0.25 mm, in about the same time as that for primary swelling.
A conceptual model for the interaction between piles and swelling clay was developed, consisting of an interactive loop of: inverted ‘wetting cones’ governed by ‘heave restriction cones’ around the pile; swelling pressure and heave mobilisation of clay strength in a ‘simple shear zone’; and pile/clay interface shear stress reversal by relative displacement.
The validity of the model tests is discussed and implications for full-scale pile design are given. Two methods gave the best predictions of uplift shear stress in the model tests: the effective stress method using final swelling pressures assuming K0 = Kp, or desiccated swelling pressures assuming K0 = 1, with shear through the clay and  = ; and the total stress method using  = 0.6 applied to undesiccated cu, or  = 0.3 applied to desiccated cu, which is lower than commonly recommended and could lead to more economic pile design for swelling clays.