Journal of Advanced Concrete Technology Vol. 3, No. 2, 253-265, June 2005 / Copyright © 2005 Japan Concrete Institute 253 Scientific paper Simulation of Coupled Corrosive Product Formation, Migration into Crack and Propagation in Reinforced Concrete Sections Kukrit Toongoenthong 1 and Koichi Maekawa 2 Received 13 September 2004, accepted 22 December 2004 Abstract This paper aims to numerically simulate corrosion induced cracking, its propagation over sections of reinforced con- crete members and the penetration of corrosive gel product into crack gaps. A coupled steel core and surrounding corro- sion product are mechanically represented by a fictitious growing composite, with which the corrosive cracking initia- tion and subsequent propagation are simulated by 2D nonlinear crack analysis. The injection of corrosive gels into evolving cracks is substantiated in cases where corrosive cracks stably propagate such as large covers and/or compara- tively small diameters of steel, and the coupled system of gel formation, migration and crack propagation is newly pre- sented. The simulation scheme was verified through RC sections subjected to accelerated corrosion by electric current with regard to crack patterns and critical corrosion rates when cracks reach the outer surface of members. 1. Introduction Steel corrosion in reinforced concrete structures has been recognized as one of the major threats for build- ings and infrastructures in service. The safety and ser- viceability of corroded RC structures has been widely addressed by practicing engineers. In some durability designs, initiation of corrosion is chosen as a limit state of durability performance with regard to carbonation and chloride penetration. This is due to the current diffi- culty to predict the lifetime of mechanical damage of the concrete cover after corrosion initiation. Since the 1980s, theoretical proposals and experimental investiga- tions on steel corrosion in concrete have been intensely carried out and reported in the literatures (e.g., Bazant 1979; Browne 1980; Tuutti 1982; Ravindrarajah and Ong 1987). Here, not only experimental investigation but also a reliable numerical approach for predicting critical corrosion mass loss for concrete cover cracking is indispensable as an important tool for rational dura- bility assessment. Morikawa et al. (1988) and Tsunomoto et al. (1990) conducted the analysis of critical weight loss for cover cracking due to reinforcement corrosion. In their analy- ses, the target specimens have smaller covers compared to the reinforcing bar diameters (C/D ratio 1.2 to 2.8). The analytical result of weight loss for cracking was close to the experimental one at the high corrosion rate. However, in the case of long-term corrosion, the analy- sis underestimated the critical weight loss compared to the reality. Molina et al. (1993) proposed the numerical model and verified it against their own experimental results. In their analysis, not only the critical mass loss but also the numerically estimated crack width was computed, and the analytical results showed overestima- tion of corrosion weight loss of steel. The analysis of corroded RC specimens having a relatively large C/D ratio was conducted by Tsutsumi et al. (1998). Their model was verified by RC specimens having C/D of 2.31, 3.85 and 5.26. It was found that the weight loss was much underestimated, especially in the case of large C/D ratios. Pantazopoulou and Papoulia (2001) investi- gated simple analytical modelling for cover cracking. The verification of the model was made using many past experimental results. While a qualitative tendency was seen, it was quantitatively found that the computed weight loss was greatly underestimated compared to the experimental results. In their conclusions, an important comment was made on a more reliable approach and the need for experimental evidence on the penetration of corrosion products into cracks. As a matter of fact, this point raised by Pantazopoulou (2001) is the main target of this paper. The non-linear analysis with multi-mechanics of cor- rosive substances and structural concrete has been pro- posed for simulation of post-corrosion cracked members (Toongoenthong and Maekawa 2004). Here, a core of non-corroded steel and its surrounding rust substances are mechanically represented by a comparable growing material, and the corrosion degree is dynamically asso- ciated with the numerical simulator based on thermo- dynamic multi-chemo physics (Maekawa et al. 1999), or non-destructive testing and inspection data in reality. This approach has been verified chiefly in terms of structural safety assessment of RC members after the formation of section-penetrating cracks caused by cor- rosive volumetric expansion of steel (Toongoenthong and Maekawa 2004). However, it has not been dis- cussed nor verified on predicted duration from the ini- 1 Design Engineer, Civil Engineering Division, Taisei Corporation, Japan. E-mail: [email protected] 2 Professor, Department of Civil Engineering, The University of Tokyo, Japan.
Simulation of Coupled Corrosive Product Formation, Migration into Crack and Propagation in Reinforced Concrete Sections
May 23, 2023
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Related Documents
