Abstract: ObjectiveConventional carbon steel tube-concrete columns (CFST) often exhibit compromised structural performance in corrosive environments, resulting in reduced load-bearing capacity and ductility. This study addresses these limitations by developing ultra-high-performance concrete (UHPC) -filled duplex stainless steel tube (UFSST) columns, which significantly enhance corrosion resistance, load capacity, and deformation behavior.MethodsTwelve rectangular short-column specimens were subjected to axial compression tests to evaluate the performance of UFSST columns. Experimental parameters included three UHPC strength grades (89~164 MPa) and three duplex stainless steel tube thicknesses, with yield strengths ranging from 307 to 807 MPa. Failure modes, load-displacement (N‒Δ) curves, and steel tube strain behavior were examined to assess: 1) axial compressive capacity; 2) interaction effects between duplex stainless steel and UHPC, and 3) the evolution of confinement mechanisms. Results were benchmarked against five international design codes (BS EN 1994‒1‒1, ANSI/AISC 360‒16, ACI 318, T/CECS 952—2021, and CECS 159—2004) through a comparative analysis of the ratio of the test value to the calculated value (Ntest/Ncal).Results and DiscussionsRectangular UFSST stub columns demonstrated good deformation capabilities under axial compression loads, with failure modes categorized into two types based on the ξ coefficient index: primarily waist drumming buckling failure for ξ=2.52 and shear failure for ξ⩽ 1.62. A ξ⩾ 1.62 should be employed in UFSST stub columns to utilize the properties of duplex stainless steel. The N‒Δ curves exhibit a three-stage pattern, namely the elastic stage, elastoplastic stage, and degradation stage. During the elastic stage, the duplex stainless steel tube exerts no restraint on the UHPC. The confinement effect of duplex stainless steel tubes on concrete generally emerges during the elastoplastic stage, as the value of the lateral-to-longitudinal strain ratio (v) increases from approximately 0.3 to 0.8, contributing to Nu of the CFST and UFSST stub columns. Replacing ordinary concrete with UHPC can increase the ultimate bearing capacity of specimens by up to 31%. The confinement effect of the duplex stainless steel tube is also the main reason why the load development of the specimen remains relatively stable as axial displacement increases after reaching the ultimate bearing capacity. The UFSST stub column demonstrates superior residual bearing capacity and a more stable degradation stage of N‒Δ curves compared to its CFSST counterparts, primarily due to the bridging effect of steel fibers in UHPC. The utilization of UHPC significantly improves the ultimate bearing capacity of stub columns while simultaneously satisfying the ductility requirement. However, enhancing the strength grade of UHPC has minimal impact on the axial compressive bearing capacity of UFSST rectangular short column specimens, whereas increasing the wall thickness of duplex stainless steel tubes can further enhance the bearing capacity and effectively reduce the occurrence of local buckling in duplex stainless steel tubes. Investigation into the strength index and concrete contribution ratio indicated that the enhancement effect of duplex stainless steel tubes on UHPC compressive strength is not as noticeable as it is on ordinary concrete. However, UFSST stub columns perform better than CFSST stub columns in terms of concrete contribution ratio. Experimental results were compared to calculated results of bearing capacity design formulas in current codes (European code BS EN 1994‒1‒1, American standards ANSI/AISC 360‒16 and ACI 318, Chinese codes T/CECS 952—2021 and CECS 159—2004). T/CECS 952—2021, based on a unified theory, considers the confinement effect of steel tubes on concrete and introduces ξ into the calculation formula. However, it does not account for the non-uniform constraint effect of rectangular steel tubes on core concrete. The calculated average value of Ntest/Ncal is 0.85, which tends toward danger and cannot be directly used for calculating the axial compressive ultimate bearing capacity of rectangular UFSST short columns. In contrast, other codes adopt the superposition theory, neglecting the constraint effect of steel tubes on concrete, often resulting in predicted values of Ntest/Ncal greater than 1, thus underestimating actual values. CECS 159—2004 yields an average Ntest/Ncal value of 1.17, while BS EN 1994‒1‒1 yields 1.06. ANSI/AISC 360‒16 and ACI 318 yield an average Ntest/Ncal value of 1.12. Therefore, among the current codes, the BS EN 1994‒1‒1 formula yields results closest to the experimental values. However, considering the confinement effect of duplex stainless steel tubes on UHPC, the ultimate bearing capacity of UFSST short columns is more realistic when calculating. Therefore, based on the formula in T/CECS 952—2021, the influence of the non-uniform constraint effect of rectangular sections on the ultimate bearing capacity of UFSST stub columns was further considered, and adjustments to the formula were made. The calculated results of the modified formula are in good agreement with experimental results, with an average Ntest/Ncal value of 1.00 and a standard deviation of 0.07. This implies that the revised formula can be utilized to predict the ultimate bearing capacity of the ultimate axial compression of UFSST short columns. It effectively accommodates a wide range of concrete strengths, from 89 to 164 MPa, as well as steel tube yield strengths ranging from 307 to 807 MPa. However, it is important to note that the formula does not account for certain influential factors, such as stress redistribution following concrete damage and the strain-hardening behavior of duplex stainless steel. Therefore, it is essential to conduct a comprehensive reliability analysis in future studies to assess the model’s reliability and robustness.ConclusionsUFSST columns demonstrate superior axial performance compared to conventional CFST, with 31% higher ultimate capacity and enhanced post-peak ductility through UHPC fiber bridging. The confinement coefficient ξ critically governs failure modes, requiring ξ ⩾ 1.62 for optimal material utilization. Current design codes exhibit either dangerous underestimation (unified theory) or unconservative overestimation (superposition theory) of UFSST capacity. The proposed modified formula addresses rectangular section non-uniform confinement effects, achieving less than 7% prediction error across a wide range of material parameters. Practical applications should consider additional factors, including stress redistribution after concrete cracking and duplex stainless steel strain hardening, which require further reliability analysis.
Processing Request
No Comments.