Performance of RC Columns Affected by ASR. II: Experiments and Assessment

This paper describes the structural performance and analytical methodology for large-scale column specimens with a lap splice and concentric axial loading affected by varying levels of alkali-silica reaction (ASR) under displacement-controlled monotonic loading. The lap-splice length in the specimens is typical of the Texas DOT practice for the bar size used and is conservative by code standards. The specimens with varying degrees of ASR showed no evidence of bond deterioration within the splice; had similar initial stiffness and behavior up to the first cracking; had a 25–35% increase in postcracking stiffness up to yielding; had a 5–15% increase in yield strength; and showed no overall detrimental effects on the structural response when compared with control specimens without ASR deterioration. This improved behavior can be explained by the resulting volumetric expansion of the concrete because of the ASR that engaged the longitudinal and transverse reinforcement—this is believed to have resulted in increased axial loading and better confinement of the core concrete. Analytically, a proposed step-by-step postcracking stiffness model for axially loaded members provided the best fit with the force-displacement experimental data. In addition, for specimens exhibiting ASR, increasing the axial load on the column provided a better match with the measured experimental response.