Seismic performance of elliptical FRP-concrete-steel tubular columns under combined axial load and reversed lateral load

By combining FRP composites with both steel and concrete, double-skin tubular columns (i.e., DSTCs) with a hollow core are a novel structural member to be used as bridge piers or other columns in harsh environments. Up to now, DSTCs with circular or square sections (i.e., C-DSTCs or S-DSTCs) have been investigated extensively, while DSTCs with an elliptical section (i.e., E-DSTCs) are rarely studied. E-DSTCs are composed of two concentric elliptical tubes (i.e., the external FRP tube and the internal steel tube) and a sandwiched concrete layer. E-DSTCs can be designed with two distinct bending stiffness with respect to the two axes of symmetry. Similar to C-DSTCs, E-DSTCs have an FRP tube with a continuously curved circumference, thus effectively confining the concrete layer. To extend the existing research, this paper tested six large E-DSTCs under combined axial load and reversed lateral load to investigate the effect of elliptical aspect ratio. The testing parameters also included the direction of reversed lateral load and the FRP thickness. Experimental results showed that, E-DSTCs performed with rounded hysteresis loops and remarkable ductility even for the specimen with a large aspect ratio (i.e., for the elliptical aspect ratio from 1.0 to 2.0, the ductility index increased monotonically from 6.50 to 7.75); a thicker FRP tube led to marginally higher peak load but noticeably better ductility (i.e., the specimen with 2.10 mm FRP had an average peak load of 77.65 kN and a ductility index of 7.14, while the corresponding specimen with 1.05 mm had an average peak load of 77.35 kN and a ductility index of 6.29); as expected, the specimen bending around its strong axis had much superior ductility than the companion specimen bending around its weak axis. A numerical column model was established based on OpenSees to simulate the hysteretic behavior of E-DSTCs, and a parametric study was conducted to assess the influences of FRP thickness, concrete strength, axial load ratio, steel tube thickness, void ratio as well as elliptical aspect ratio. In practical engineering design, it was recommended that: (a) the elliptical aspect ratio as well as other key parameters (e.g., the FRP thickness, the steel tube thickness, the area void ratio) should be optimized based on specific loading conditions; (b) to make more effective usage of FRP composites, the FRP tube thickness should be optimized for different height regions of the column according to the moment gradient caused by the reversed lateral load.