Evaluating the composite behavior developed through bond in the steel-concrete interface for future hot-rolled asymmetric steel I-beams

Abstract

Conventional steel-concrete composite floor systems consist of hot-rolled steel beams with metal decking on the top flange that supports a concrete deck slab. An alternative to achieve shallower floor depths is to utilize stay-in-place formwork, either precast concrete panels or steel deep decking, placed on the bottom flange with a cast-in-place concrete slab. For ease of construction, an asymmetric section is needed for vertical placement of the precast panels or deep decking. However, there are no hot-rolled asymmetric steel I-beams (termed A-shapes) readily available in the United States. The purpose of this research is to evaluate the composite flexural behavior of a shallow-depth floor system for future large-scale production of A-shapes. To achieve minimal floor depths and increase the efficiency of steel building construction, shear studs (generally used in conventional composite floor systems) are not utilized to transfer longitudinal interface forces. The top flange and web of the steel member are encased in concrete, which will allow partial composite flexural behavior due to the formation of concrete-steel bond. The research presented herein includes eight experimental beam tests to understand the flexural strength and stiffness developed through concrete-steel bond shear. The eight tests performed well and achieved 74 % to 83 % of the full composite flexural strength before the bond started to slip, although only minimally. Following the initial slip, the shallow-depth beams were unloaded and reloaded to evaluate the robustness and ductility of the composite cross-section. The beams proved to be highly ductile and robust as they reached 77 % to 91 % of the full composite strength upon reloading due to reengaging of the bond after the occurrence of initial slip. The composite flexural stiffness of the beams was well intact under the service loading, indicating that the transformed moment of inertia of the cross-section can be utilized for serviceability analysis. A bond shear strength of 0.69 MPa (100 psi) was established since it undercuts most of the experiments with a bond perimeter assumed to be above the elastic neutral axis of the composite section. The partial composite strength was further evaluated utilizing three numerical methods: (1) linear interpolation between steel yield and full composite, (2) partial plastic stress distribution, and (3) strain compatibility. All methods predicted reasonable results but the method of linear interpolation was the most conservative one in relation to the bending moment that causes the initiation of bond slip.