Carbonation of concrete bridge structures in three South African localities

• Main Author: Yam, Wood Kuen
• Other Authors: Alexander, Mark Gavin
• Format: Dissertation
• Language: English
• Published: University of Cape Town 2014
• Subjects: Civil Engineering
• Online Access: http://hdl.handle.net/11427/7427

Includes bibliographical references (leaves 220-227). === The rate of carbonation for the localities of the Cape Peninsula, Durban (i.e. Durban - KwaZulu Natal South Coast) and Johannesburg (i.e. the motorway system and between Heidelberg Road and Geldenhuis interchanges on the N3 freeway) were studied in order to derive carbonation prediction models for each of these localities. The derivation of the prediction models was based on field carbonation data measured from approximately 30 in-service bridges in each locality. One of the uses of the derived models was to allow the preparation of maintenance plans so as to avoid carbonation-induced corrosion for structures in these localities. Since the rate of carbonation depends strongly on material and environmental factors, the carbonation data from each locality were analysed separately on the grounds that these localities have different climatic conditions. The data within each locality represent different material and exposure conditions, and the data were therefore grouped according to the concrete strength grade (as a measure of concrete quality) and exposure conditions, prior to statistical analysis. Based on the method of least squares, as well as integration of the understanding of the process of carbonation and knowledge of climatic conditions of each locality, carbonation prediction models for a variety of concretes for each locality were derived. Results show that bridge structures in the Johannesburg locality have the highest carbonation rate due to the relatively dry environment throughout the year. Bridges in Durban locality exhibit a lower carbonation rate than Johannesburg bridges, but higher than Cape Peninsula bridges owing to shorter rainfall duration and higher temperature. In addition, the carbonation rates of both exposed and sheltered elements with similar concrete strength grades for bridges in Durban are very similar, i.e. exposure condition has little influence on carbonation rate for these elements. The same is true for bridges in the Johannesburg locality. It is surmised that short precipitation times and high relative humidity in Durban locality make the near surface moisture content of exposed and sheltered elements very similar. Likewise, it is surmised that short rainfall duration and low relative humidity in Johannesburg locality result in essentially the same near surface moisture content of concrete elements throughout the exposure time. The data in Durban locality show that old concretes have a slower carbonation rate than modem concretes with the same concrete strength grade. This is likely due to the changes in cement properties over the years, related to the need for fast track development for modem structures. This finding indicates that the prediction models are not suitable for carbonation predictions for future structures (produced by modem cements) as the rates of carbonation will be different. Oxygen Permeability Index (OPI) was investigated in an attempt to predict the rate of carbonation. According to the philosophy and testing procedures for OPI, it is considered that early age OPI may be superior to concrete strength grade for carbonation predictions because of better characterisation of the permeability of (cover) concrete. However, due to the lack of early age OPI information for the data, using OPI as a carbonation prediction tool was not entirely successful. Further research in this regard is worthwhile.