Eccentric Axial Load Capacity and Behavior of Composite Columns using 800MPa Steel and 100MPa Concrete

Abstract

The use of steel-concrete composite columns has been popular concurrently with the increase in the construction of high-rise buildings and long-span structures, and for the improvement of structural safety, space efficiency, and cost effectiveness, and for the reduction of lifting weight, the utilization of high-strength materials in composite columns is expected to grow more. The use of a high-strength steel exceeding 600MPa (yield strength) may cause difficulty in the design of composite columns. Since the yield strain of the high-strength steel is greater than the ultimate compressive strain of concrete (approximately 0.003 for the ordinary concrete subjected to short-term loads), the steel section may not develop its full plastic strength due to the early crushing of the concrete. As a result, the plastic stress distribution method of Eurocode 4 is expected to overestimate the load-carrying capacity of the composite sections using the high-strength steel exceeding 600MPa. On the other hand, the strain-compatibility method of ACI 318 seems to be appropriate in the strength prediction of the sections. However, if the concrete is well-confined by transverse re-bars and/or steel sections, the ACI 318 method ignoring the confinement effect could be too conservative. In the research, therefore, to investigate the mechanical properties of the composite sections using high-strength steel, to derive requirements for maximum use of the high-strength steel strength, and to evaluate and improve the applicability of current design codes to the composite sections, experimental and analytical studies were performed for the concrete-encased steel composite columns using 800MPa grade steel and 100MPa (or 200MPa) grade concrete. Prior to the experimental and analytical studies, solutions for the maximum use of steel strength and an optimal encased section were developed by preliminary study, and the test parameters were determined based on the results from the preliminary study. In the eccentric axial load test programs, recognizing the possibility of early crushing of the high-strength concrete, the effect of the ultimate strain of laterally confined concrete on the peak strength and ductility of six fully-encased specimens was investigated, the steel section was located outside the concrete in two partially-encased specimens in order to cause a larger strain demand on the high-strength steel, 200MPa grade ultra high-strength concrete was used in a fully-encased specimen to investigate the effect of the higher ultimate compressive strain (more than 0.004) of concrete, and the developed optimal encased section, a prefabricated steel cage with lattice, was used in a test specimen to examine the effect of outer placement of the steel section and the improved confinement effect due to the steel angles at corners. For comparisons, a slender fully-encased specimen and two rectangular concrete-filled specimens were also tested. The confinement effect increased/ maintained the strength of the confined concrete after concrete cover spalling (the first peak load, or existing design strength), and as a result the fully-encased specimens regained their strengths up to the second peak load, in company with the propagation of steel yielding (especially, in well-confined specimens, the second peak load was greater than the first peak load). In the partially-encased specimens, the compressive steel flange yielded before concrete crushing, but the effect on strength was limited due to early local buckling of the steel and lower confinement. The ultra high-strength concrete material did not show its higher peak strain in the specimen due to the size effect, weakness plane by reinforcements, and small size of coarse aggregates. The load-carrying capacity of the prefabricated steel cage specimen increased greately up to the maximum strength with a minor degradation in stiffness, even after the cover spalling. To analyze the contribution of structural components, and to evaluate the applicability of current design codes to the composite sections using high-strength materials, the test results were compared with the nonlinear numerical anlaysis, which was developed using MATLAB, and the strength predictions by the design codes. Since the high-strength steel did not show its full plastic strength along the section, Eurocode 4 and AIJ-SRC produced unsafe predictions. On the other hand, the ACI 318 method ignoring the confinement effect underestimated the maximum strength. For more exact prediction, which leads to more economical design, a modification of the ACI 318 method was proposed considering the confinement effect. The modification gave a better prediction. In the case of the concrete-filled specimens, Eurocode 4 and AIJ-SRC were applicable. Design methods in current design codes conservatively estimated the effective flexural stiffness of the test specimens, but overestimated the stiffness of the concrete-filled specimens with large eccentricity. Meanwhile, to examine the effect of the time-dependent deformation of concrete on the behavior of the composite columns using high-strength steel and high-strength concrete, concentric and eccentric long-term load tests were performed for three fully-encased specimens and a rectangular concrete-filled specimen, followed by ultimate strength tests. The creep and shrinkage of concrete caused relaxation in the axial load of the concrete, which was gradually transferred to the steel and re-bars, and as a result the steel approched its yield strain at ultimate load. However, the long-term effect was not detrimental for the performance (strength and stiffness) of the specimens, since high-strength concrete, higher steel ratio, and high-strength steel used in the specimens. To analyze the long-term behavior, a time-dependent analysis procedure was developed and applied to the test specimens, and the effective flexural stiffness of the test specimens were compared with the ACI 318 predictions. Based on findings from the research, design considerations were summarized and the direction for revision of KBC 2009 design provisions was briefly proposed.