Practical bond model for corroded RC bridges

Abstract

Corrosion of steel reinforcement is a common cause of deterioration in reinforced concrete bridges and many existing bridges are damaged to varying degrees. The rate of deterioration of the bridge stock has been shown to increase due to climate change. Unsympathetically, the demand for load-carrying capacity is however often increased with time. Therefore there is an increasing need for reliable methods to assess the load-carrying capacity and remaining service-life of existing infrastructure. A simple model for the assessment of Anchorage in corroded Reinforced Concrete structures (ARC) has previously been developed. It was originally based on fib Model Code 1990 and has been verified with experiments and three-dimensional nonlinear finite element (3D NLFE) analyses for both accelerated and natural corrosion as well as for different degrees of corrosion. The model was applied when assessing two road bridges in Sweden. The investigation demonstrated great cost savings but also areas for improvement, in particular regarding (a) applicability to practical cases and (b) incorporation of uncertainties in the assessment. The primary focal point of this paper is to present an overview of the development of the ARC model together with recent verifications against a large bond test database as well as foreseen future developments. It was found that the ARC model represents the physical behaviour reasonably well, and gives conservative values of bond strength compared to the bond tests database. In future works, among others, uncertainties of the input variables will be incorporated by means of probabilistic modelling, making way for implementation of the ARC model into semi-probabilistic safety concepts by extraction of modification  factors. Overall, with more accurate and reliable assessment methods for corroded RC structures, environmental and economic savings are imminent as more of the potential of existing structures can be realized.