This thesis investigates the seismic performance of circular-sectioned steel-encased reinforced concrete bridge piles.

Constitutive models which account for lateral interaction of tube and concrete under monotonic loading were developed and calibrated against test data. Steel-encased reinforced concrete members, with casing diameter to thickness ratios in the range of 34 to 214, were tested under longitudinal and cyclic lateral load. Sound performance was displayed under the simulated seismic attack. Despite the formation at relatively low ductility levels of local buckles in the tube; strength, ductility and energy-dissipating characteristics were found to be equal or superior to those of conventionally designed ductile reinforced concrete members. Good agreement was obtained between the experimental results and predictions based on moment-curvature analyses. Steel-encased reinforced concrete members with casing circumferential discontinuities in the plastic hinge zones also behaved in a ductile fashion.

Soil-pile interaction under monotonic and cyclic lateral load was investigated by conducting tests on small-scale (diameter = 115mm) piles. Non-linear lateral load-deflection responses were obtained and ultimate loads were not reached despite imposed lateral displacements of up to 2.6 pile diameters. Soil lateral pressures of up to 15 times passive pressure were obtained and a large reserve of hysteretic damping was available in the soil. For the pile which formed a plastic hinge at depth in the ground, pile plasticity was well-developed over a considerable length. Thus it is recommended that piles which form plastic hinges in the ground be designed for full ductility. Due to lateral interaction between piles and pile head rotation under the overturning effect of lateral load, twin-capped piles were found to have only 1.5 times the lateral load-carrying capacity of similar single piles with free heads. A method was described in which the differential equation governing pile-soil interaction under lateral load is solved using a finite difference approach to allow for non-linear pile and soil behaviour and P-∆ effect. This method was used to calibrate simple tri-linear soil models, and subsequently the performance of a prototype pile under large lateral displacement was theoretically analysed.