Elsevier

Cement and Concrete Research

Volume 29, Issue 9, September 1999, Pages 1475-1485
Cement and Concrete Research

Papers
Effect of initial curing on chloride diffusion in concrete repair materials

https://doi.org/10.1016/S0008-8846(99)00130-1Get rights and content

Abstract

This paper presents the results of an experimental investigation on the effect of initial curing on chloride diffusion in three generic repair materials and a concrete mix of similar grade. Two field conditions of initial curing were simulated as follows: (1) exposure to a chloride environment after 24 h of casting to simulate repair situations, for example in the tidal zone of marine structures where repairs cannot be protected from tidal exposure for long periods; and (2) 28 days of curing at 20°C, 55% relative humidity to simulate field conditions where the repair patches are exposed to the ambient environment without any prolonged protected curing. For comparison purposes identical specimens were cured in water at 20°C for 28 days before exposure to the chloride environment. Chloride concentration profiles in the materials were determined after 28, 90, and 180 days of exposure. Regression analysis of the experimental data was carried out applying Fick's second law of diffusion to determine the diffusion coefficients (DC) and chloride concentration on the surface (C0). Based on the results obtained, an analytical expression for the long-term prediction of chloride concentration in repair materials is given. The results show a high rate of chloride diffusion in the cementitious repair materials containing polymer latex additives relative to plain concrete. The effect of inadequate initial curing on chloride penetration (e.g., exposure to chlorides after 24 h of casting or 28-day air curing) is more pronounced in concrete mixes than in generic repair materials.

Introduction

Chloride-induced corrosion of reinforcement is one of the major causes of deterioration in reinforced concrete structures 1, 2. Structural members damaged by corrosion are often reinstated by applying repair patches. Repairs, however, are successful in the long term only if the causes of the original damage have been understood and appropriate repair materials are applied to resist future deterioration. The repair material should have suitable properties required to form an effective physical and chemical barrier against the penetration of deleterious substances, such as chlorides and carbon dioxide. Initial curing to which a repair is subjected prior to exposure to a chloride environment plays an important role in determining the chloride penetration rates into the repair material [3]. In practical situations, it is often impossible to provide ideal conditions of initial curing (high humidity, normal temperature over a long period of time, e.g., 28 days) to repair patches. The effectiveness of curing compounds is usually uncertain; repair patches are frequently applied to structures located in hot, arid climates and indeed in some situations (e.g., repairs to marine structures in the tidal zone) repair patches are exposed to seawater within a few hours of application. Consequently, a knowledge of the chloride penetration characteristics of inadequately cured repair materials is important for the effective design and selection of concrete repair solutions. Concrete repair formulations often contain special constituents (e.g., polymer latex modifiers) that can significantly influence the effects of initial curing with respect to chloride penetration. These characteristics need to be quantified.

Drying of cementitious materials due to poor curing, particularly at the surface, leads to restricted hydration in the surface layers and thus to higher porosity and permeability [4]. This can result in rapid chloride penetration. On the other hand, an initial period of moist curing that is followed by rapid drying can lead to cracking of the concrete surface [5]. These cracks form major flow paths for chloride penetration. In cracked and uncracked sections alike, chloride ions diffuse through capillaries and voids within the matrix and along cracks in the aggregate particles and along the interface of coarse aggregates [6]. A study on precracked marine concrete concluded that chloride concentrations in the vicinity of cracks are not significantly affected at crack widths less than 0.2 mm. Crack widths greater than 0.5 mm, however, lead to marked increase in chloride concentration in their vicinity 7, 8.

Section snippets

Materials and mixes

Three generic repair materials (A, B, and C) that are produced commercially were used together with a plain concrete mix of similar grade. These materials were supplied as single component systems, ready for on-site mixing with the addition of potable water.

Properties of the repair materials

The basic properties of the materials under standard water curing (at 20°C) are given in Table 1[9]. Repair material A and concrete have similar fresh densities but the 28-day strength of material A is much higher. Repair material B has a low density, low strength, and high permeability. Material C has a lower density than concrete but the strength is slightly higher. The permeability coefficient is lowest for materials A and C, followed by the control concrete mix. Material B has by far the

Conclusions

The following conclusions are based on the test results reported in this paper.

  • 1.

    A simple correlation between permeability and chloride penetration does not exist for the different repair materials and control concrete. This is because different generic repair materials are comprised of chemically different constituents, such as polymers and admixtures, which control shrinkage and other conditions.

  • 2.

    Repair materials that incorporate a polymer admixture show a much lower rate of reduction in

Acknowledgements

This paper presents some of the results of an EC supported BRITE/EURAM research project BREU P3091 “Assessment of performance and optimal strategies for inspection and maintenance of concrete structures using reliability based expert systems.”

References (17)

There are more references available in the full text version of this article.

Cited by (109)

View all citing articles on Scopus
View full text