Experimental and numerical study of chloride induced corrosion in reinforced concrete

Abstract

Chloride-induced corrosion is considered as one of the major concern for durability of reinforced concrete (RC) structures. Especially vulnerable, are structures located in coastal marine environment or highways and garages treated with de-icing salts during winter seasons (Tuutti, 1993; Cairns, 1998). Consequences of chloride-induced corrosion of steel reinforcement have negative effects on structural behavior and involve several aspects related to the life cycle of the structure, such as serviceability, safety and structural performance. Direct and indirect costs of maintenance and repair are relatively high and constitute nowadays a huge economic exertion. Therefore, challenging task is to develop and improve a numerical tool, which can realistically predict corrosion processes and the related mechanism of deterioration in RC structures, supporting the service life prediction of damaged and undamaged structures.The main objective of the present work is to validate the recently developed 3D Chemo-Hygro-Thermo-Mechanical model for concrete by means of an extensive experimental program, which includes tests under natural and laboratory controlled conditions. The comparison between numerical and experimental results is significantly important in order to quantitatively calibrate the parameters responsible for the computation of corrosion rate and distribution of rust in pores and cracks. To better describe the problem, hysteretic moisture behaviour is coupled with the transport of corrosion products in cracks and a relationship between diffusivity of corrosion compounds and crack width is proposed. It has been demonstrated experimentally that the production of corrosion compounds are strongly dependent on the environmental conditions and presence of chemical reactants. Therefore, type of corrosion products and corresponding distribution are in the present work investigated on concrete thin section by means of microscope analysis and Raman spectroscopy. Computation of corrosion rate and related corrosion induced damage is directly related to the assumed position of anode and cathode on the reinforcement surface and currently there is no algorithm which can predict the combination between anode and cathode surfaces that results to the highest corrosion induced damage. To investigate this influence, the expression for maximum entropy production, deduced from irreversible thermodynamics, is formulated. The entropy is considered produced by dissipative processes, which are in this special case the flow of ions through the electrolyte, the anodic and cathodic polarization and the diffusion oxygen process. Through several numerical examples, in which the size and position of anodic and cathodic surfaces are varied, is demonstrated that maximum entropy leads to maximum corrosion induced damage.