Chloride penetration in concrete: evaluation of rapid conductivity methods and development of a service life prediction model for marine conditions

Surana, Saarthak

Chloride penetration in concrete: evaluation of rapid conductivity methods and development of a service life prediction model for marine conditions

Thesis / Dissertation

2025

Publisher

University of Cape Town

Department

Department of Civil Engineering

Faculty

Faculty of Engineering and the Built Environment

Abstract

Chloride-induced corrosion is the predominant cause of loss of service life in reinforced concrete structures subjected to marine conditions. To protect reinforcing steel from the corrosive effects of external chloride ions present in these environments, a proper combination of concrete quality and cover depth is needed. Short-term electrical conductivity methods are rapid alternatives to the time-consuming diffusion tests for the characterisation and quality control of concrete mixtures. To estimate the long-term behaviour of a concrete mixture in marine conditions, however, and propose a proper combination of adequate cover depth and concrete quality, a service life prediction model is needed. In this study, a service life prediction model was developed to predict the rate of chloride ingress in marine environments using conductivity measurements to estimate diffusivity parameters. A comparison between the chloride conductivity test and the Wenner resistivity test was also performed to evaluate the relative impact that pore solution conductivity and unsaturated conditions may have on the proper quantification of chloride penetration resistance of concrete. Diffusivity and conductivity measurements were carried out on various concrete mixtures made with different cementitious binder combinations including fly ash and slag, at different w/b ratios, and cured under wet conditions or plastic wrapping. The conductivity results of this study showed that the use of the conductivity measurements to assess the ionic transport resistance without regard to the pore solution chemistry can lead to misinterpretations. This effect was especially noticeable in fly ash concretes. These findings address the effect of pore solution on the Wenner resistivity results for practical applications, which has not been explicitly and extensively addressed in the literature. Similarly, the studies on the effect of unsaturation on the Wenner resistivity mostly concern with the effect of surface drying on the Wenner resistivity. In this study, the effect of rewetting on the Wenner resistivity was evaluated on concrete specimens that underwent air drying as part of the curing regime as is common in practical curing regimes before the testing age of 28 days. Despite the use of prolonged specimen rewetting, the moisture gradients in concrete affected the evaluation of the curing efficacy on the cover concrete by the Wenner method. The chloride conductivity test showed more robustness against both these important issues. This study illustrates the constraints of the Wenner resistivity method for compliance testing—an indispensable part of the performance-based quality control process for durability. Notwithstanding this, the Wenner method was found to be very useful in the characterisation of the time-dependence of the instantaneous chloride penetrability of concrete under saturated conditions. A promising application of this finding is in characterising the time-dependence of the chloride diffusivity for service life modelling of reinforced concrete exposed to marine conditions. On the relationship between the apparent chloride diffusivity and chloride conductivity index (CCI), it was found that the time dependent variation of microstructural penetrability could explain the time dependent reduction in apparent diffusivity. By accounting the time-dependence of CCI, a general binder-independent correlation was obtained between the apparent diffusivity and CCI. This finding suggests that the differences in the apparent diffusion coefficients in different SCM-blended concretes could be mainly attributed to the continued microstructural development of concrete with time. This is a significant departure from the existing understanding that the chloride resistance of SCMs is to be ascribed to a large extent to their superior binding ability. Instead, the results and analysis of this study strongly suggest that the main differentiating factor is the long-term ageing-dependent reduction in microstructural penetrability. This is a significant improvement of the current understanding of chloride diffusion behaviour in concrete and one of the crucial novel contributions of this study. Using the laboratory-exposure results of the present study and the field-exposure results of previous studies, a simple user-friendly service life prediction model is developed. A multi-factor model framework based on the error function solution to Fick's 2nd law is used to derive the apparent diffusion coefficient as a function of the test method, curing conditions, and the exposure environment. The general binder-independent correlation between the apparent diffusion coefficient and the time- averaged CCI is used to describe the apparent diffusion coefficient based on early-age CCI and the ageing coefficient of CCI. This approach minimises the need for diffusion testing to characterise the material's resistance to chloride diffusion and its development under moist marine conditions. The quantification of the effect of exposure conditions on the apparent diffusion coefficient was done in terms of a simple empirical factor that could be used to quanitify the severity of exposure conditions pervasive at relevant marine sites. Environmental characterisation parameters were derived for selected South African marine sites using field data from previous studies. The proposed model represents a simplified approach to service life prediction that integtrates design and quality control aspects aimed at providing more effective tools to enable construction of durable reinforced concrete structures.

Keywords

Chloride-induced corrosion

Wenner method

Reference:

Collections

PhD / Doctoral

Full item page