Enhancing the service life of concrete exposed to chloride attack

  • Zhengtian Song

    Student thesis: Doctoral ThesisDoctor of Philosophy


    This study was carried out to enhance the durability of concrete exposed to chloride attack by means of two approaches: (i) to design sustainable concrete compositions and (ii) to apply suitable maintenance techniques consisting of electrochemical chloride extraction (ECE) and surface coating (SC) treatments. Subsequently, the study was divided into two separate phases for this study.

    The first phase was to assess the role of the cement combinations on enhancing chloride resistance of concrete. The cement combinations composed of, Portland cement (PC)/fly ash (FA) and PC/ground granulated blast furnace slag (GGBS) in binary mixes, and with silica fume (SF), metakaolin (MK) and limestone (LS) as addition to form ternary concretes. The w/c ratio of concrete was selected in the range of 0.35 to 0.45 for the requirement of high performance. To improve the degree of accuracy, four contemporary chloride test methods were used in this work to measure the chloride penetration of concrete, which are: (i) Method A-rapid chloride permeability test to ASTM C1202; (ii) Method B-chloride migration test to NT Build 492; (iii) Method C-Multi-Regime test to UNE 83987:2009; and (iv) Method D-chloride diffusion test to CEN/TS 12390-11. The variability of different methods was evaluated and the correlation between them discussed. In addition, the microstructure and chloride binding capacity of relevant cement paste were investigated in order to provide more information for assessing the improvement on chloride resistance of different binary/ternary mixes.

    The results indicated that the chloride resistance of concrete was benefited significantly when blended with additional cement combinations. The concrete blended with FA or FA/LS showed no contribution on improving chloride resistance before the age of 28 days, but significant improvement up to age of 90 days. Compared to 100PC concrete at the equal w/c ratio, generally, it demonstrated 3-7 times decrease of the chloride durability indicators in the binary and ternary concrete. When comparison was conducted at equal strength grades (45, 55, and 65 N/mm2), the improvement on the chloride durability was more significant in the binary and ternary concrete. However, for individual binary or ternary concrete, the decrease in w/c ratio from 0.45 to 0.35 did not cause significant expected reduction in the chloride diffusion coefficient, especially when w/c ratio went down below 0.4. This was mainly attributed to the reduction of chloride binding capacity (loss of accessible binding sites to chlorides) of concrete caused by decreasing w/c ratio, although which still can refine the pore structure of concrete.

    The maintenance approach is necessary and inevitable when the chloride durability of designed concrete is insufficient to satisfy the designed working life. The second phase investigated the application of suitable maintenance approach involving ECE and SC treatment on the chloride contaminated concrete prior to the corrosion of embedded steel. The selected specimens were 100PC, 70PC30FA and 50PC50GGBS concrete with w/c ratio of 0.35, 0.45 and 0.55. Chloride was driven into 28-day concrete under the migration test rather than mixing at time of casting concrete. Two cycles of ECE process were applied on concrete, and then the pore structure characteristics and carbonation resistance of concrete were measured. Those specimens, including reference concrete, desalinated concrete and desalinated concrete coated with coating materials (Silane and RheoFIT®790) were placed into the artificial spraying salt solution of 1.0 mole/litre concentration for 30 days. The chloride content profile of sprayed specimens were obtained, and used to calculate the chloride diffusion coefficient based on three different calculation methods.

    The observations stated that approximately 65% of penetrated chloride was extracted during two cycles of ECE treatment for 100PC and 70PC30FA concrete, and about 57% for the 50PC50GGBS concrete. The ECE treatment increased the porosity of all mixes due to movement of ions and dissolution of solid phase in concrete, while much more significant in 50PC50GGBS concrete than 100PC and 70PC30FA concrete. It also reduced significantly the carbonation resistance of concrete at early exposure time (before 6 weeks), but insignificantly up to 10 weeks exposure period. Additionally, it showed significant increase of chloride diffusion coefficient in the desalinated 50PC50GGBS concrete compared to the non-desalinated concrete. However, the influence of chloride diffusion coefficient was not significant in the desalinated 100PC and 70PC30FA concrete. Coating materials performed effectively on preventing chloride ingress from a spraying environment according to the comparison of chloride profile between coated and uncoated specimens, except for the RheoFIT®790 coated on 100PC concrete. Based on the current calculation methods for the chloride profile, the calculated chloride diffusion coefficient of coated concrete also demonstrated significant reduction compared to those uncoated concrete. Finally, the predicted service life of concrete after ECE and SC treatment demonstrated long extended life for 70PC30FA concrete, where only ECE treatment can increase total service life by 47%-66%. Both ECE and SC (Silane with 50 years working life) treatments will extend total service life to maximum 1.9-2.5 times, and minimum 1.7-2.0 times. The extended service life is highly related to the working life of coating materials. However, for 50PC50GGBS concrete (with increasing chloride diffusion coefficient caused by ECE treatment), only 3%-4% increase of the total service life was found due to the ECE treatment. Both ECE and SC (Silane with 50 years working life) treatments can offer extension of service life to maximum of 1.3-2.2 times, and minimum of 1.2-1.3 times.

    Date of Award2014
    Original languageEnglish
    SupervisorRod Jones (Supervisor)

    Cite this