Predicting the chloride penetration of fly ash concrete in seawater

In this study, a model for predicting chloride penetration in fly ash concrete under long-term exposure in a marine environment is developed. The empirical model was based on 2-, 3-, 4-, and 5-year investigation of concretes in a marine site. Regression analysis of the data was carried out by applying Fick's second law of diffusion to generate an empirical formula for predicting chloride concentration in concrete. The model uses the water to binder (W/B) ratio, fly ash content, distance from the concrete surface, and exposure time. Model validation revealed that the predicted chloride concentration levels were within a ±25% error margin (R² = 0.91 ? 0.99) in the samples used to develop the model. The model was also verified using data from previous laboratory and field studies. Most predicted chloride concentration levels were within a ±30% margin of error from field samples. The model also predicted the strong effect of fly ash and W/B ratio on reducing chloride diffusion in concrete. Results clearly indicated that a high volume fly ash replacement (up to 50% by weight of binder) and a low W/B ratio will yield good chloride resistance in concrete under long-term exposure in a marine environment.