Influence of integral bridge abutment stiffness on backfill pressure build-up caused by seasonal temperature changes

Abstract

Although the construction of integral abutment bridges (IAB) is simplified as the members are all cast monolithically with each other, the soil-structure interaction between the abutments and the retained soil significantly complicates their design. More complex thermal movement must be taken into account as opposed to the simplified methods recommended by existing design codes for jointed structures. The thermal expansion and contraction of the integral bridge’s deck forces the abutment into and away from the retained soil which causes changes to the earth pressure distribution along the height of the abutment over time. For longer integral bridges, these changes are more pronounced and have greater influence on the design of the bridge. The density of the sand increases with time due to the cyclic loading caused by the thermal action of the IAB and therefore results in a stiffer backfill against the abutments. The earth pressure build-up will occur for each thermal cycle and is highly dependent on the nature and properties of the soil and the structural stiffness of the deck and abutments of the integral bridge. This development in earth pressure over time due to cyclic loading is known as the strain hardening/soil ratcheting effect of soils and is more severe for longer bridges and for greater numbers of applied thermal cycles. Based on current literature, it is postulated that the retained soils can reach Rankine passive pressures during the bridge’s design life. Large settlements can also be expected at the face of the abutments due to the corresponding volume reduction caused by the beforementioned densification of the sand that is continuously taking place. It has been identified that sands generate the greatest rate of earth pressure development for the same amount of applied thermal cycles compared to finer soils such as silts and clays which have significantly more resilient behaviour and displayed significantly less increases in pressure over time. There is considerable uncertainty regarding the behaviour of the abutments of integral bridges due to cyclic temperature changes as the entire system (i.e., the bridge structure and retained soil) is highly dependent and influenced by each other. International design codes recommend length restrictions for the different types of integral bridges and are heavily governed by the amount of settlement at the abutment faces. The main objective of the study was to identify the effects of integral bridge abutment stiffness during soil ratcheting caused by temperature change. More specifically, seasonal temperature changes were investigated as the bridge deck is expected to expand the most during these cycles compared to daily cycles. The earth pressure evolution of a coarse silica sand, mechanical behaviour of the sand particles (displacements and particle flow), resiliency of the retained fill, and the overall structural behaviour (deflection, curvature, and bending moments) were evaluated experimentally by varying the stiffness of the substructure of several model integral abutment specimens that were constructed from reinforced concrete. Current literature has only investigated the soil-structure interaction of scaled abutments in the geotechnical centrifuge that were made from steel or aluminium. It should be noted that there does exist few studies that have incorporated larger scale models, and as such this study aimed to demonstrate more realistic IAB behaviour based on the constructed concrete specimens. After testing of multiple specimens (each having different relative stiffness to the bridge deck) it was found that the degree of restraint exerted by the retained sand after 120 seasonal cycles (recommended design life of integral bridges by the British Standards) increased for greater abutment stiffnesses. It was found that the shape and magnitude of the pressure distributions as well as the structural behaviour demonstrated significant dependence on the stiffnesses of the abutment and retained sand. Based on the experimental data, the optimum choice of stiffness for the abutment specimens was evaluated.