Axial compressive behaviour of composite steel elements incorporating rubberised alkali-activated concrete

Abstract

This study presents an experimental and numerical investigation into the axial compressive behaviour of steel tubes infilled with rubberised alkali-activated concrete. An experimental programme involving circular and square concrete filled steel tubes with different length-to-diameter or length-to-width ratios and concrete infill mix designs with varying rubber contents, of up to 60% crumb rubber replacement of natural aggregates, is firstly described. A detailed account of the experimental results, including the axial capacity, stiffness, toughness, ductility, stress-strain response, and failure patterns, is given. The numerical study is performed in ABAQUS/CAE and the concrete compressive behaviour is modelled using the Concrete Damaged Plasticity model with a modified function for the compressive behaviour. The numerical results are validated against the experimental results, and a parametric study involving 315 finite element models is carried out to cover a wide range of concrete and steel material properties and different steel tube dimensions. The results show that an increase in rubber content in the concrete infill leads to a reduction in the axial capacity; however, this reduction is lower than that observed for unconfined specimens. The results also illustrate an increase in ductility with higher rubber content, which is mainly noticeable for members with circular sections as compared to those with square sections. The experimental and numerical results are used to examine the axial capacity prediction approaches in Eurocode 4 and AISC 360, with particular focus on assessing the confinement effects. It is shown that codified prediction equations for square concrete filled steel tubes give reasonably accurate results. Both codes, however, result in poor predictions for circular concrete filled steel tubes, with Eurocode 4 leading to unsafe predictions while AISC 360 gives overly conservative estimates. Modifications to Eurocode 4 and AISC 360 axial capacity approaches for circular tubes are proposed and are shown to offer significant improvement in terms of safety and accuracy of design.