Acta Polytechnica CTU Proceedings doi:10.14311/APP.2019.22.0139 Acta Polytechnica CTU Proceedings 22:139–144, 2019 © Czech Technical University in Prague, 2019 available online at http://ojs.cvut.cz/ojs/index.php/app EXPERIMENTAL STUDIES OF ACCELERATED CHLORIDE TRANSPORT IN CONCRETE Vojtěch Zacharda∗, Jiří Němeček Czech Technical University in Prague, Faculty of Civil Engineering, Thákurova 7, 166 29 Prague 6, Czech Republic ∗ corresponding author: vojtech.zacharda@fsv.cvut.cz Abstract. This contribution deals with the efficiency of electromigration of chlorides used as a repair method for reinforced concrete structures. Experimental studies of accelerated chloride transport tests were performed on samples of concrete without chlorides and with admixed sodium chloride during concreting. Two concrete types from Portland cement characterized with normal and low compressive strengths were studied. The electromigration was applied to penetrate chlorides into the chloride-free sample and for extraction of chlorides from the sample. The effectiveness of the chloride extraction process for rehabilitation of reinforced concrete in terms of lowering the chloride concentration in different concrete types and surface concentration was observed. Electrical extraction was found to be effective for lowering of initial chloride concentration by 15-20% after 24 hours. The decrease in surface concentrations was found in the range of 40-50%. The extraction process was found to be feasible and effective for both concrete types. Keywords: Concrete, electromigration, chlorides, extraction. 1. Introduction Reinforced concrete structure is exposed to a variety of environmental processes during its lifetime that can cause degradation. The penetration of chlorides into the concrete from deicing salts or in sea splash areas are the most common deterioration processes. Chloride ions diffuse through concrete towards steel reinforcement and cause its corrosion. After reaching a critical value of chloride concentration on the rebar surface the corrosion of the steel starts [1–3]. The corrosion is accompanied by an expansion of the oxidic products causing high pressures, concrete cracking and finally spalling of a cover layer. Therefore, the resistance of concrete to penetration of chlorides is one of the very important properties for steel-reinforced concretes. Since the natural diffusion of ions [4, 5] is a slow process taking years further complicated by concrete aging and chloride binding [6] testing of concrete chloride permeability is often done in an accelerated manner using an electric field [7–9]. Both types of tests need to be properly interpreted and diffusion characteristics of concrete can be derived from the tests [9, 10]. Electrochemical extraction [7, 8] is one of the solu- tions that can be used to repair reinforced concrete structure already attacked by chlorides. Besides chlo- rides, other particles with the charge, such as corrosion inhibitors or nanoparticles, can also be transported via electric field [8, 11–13]. The convective flux of ions caused by the electromotive force is much larger that the diffusive flux and leads to radical shortening of the time needed for extraction of chlorides from the structure. In the reality, the electric field is re- alized between the steel reinforcement and the outer surface electrode (usually stainless steel mesh) placed on the concrete surface. In the lab scale, the tests are performed on concrete samples placed in between the mesh electrodes submerged in a suitable electrolyte. A standard test involves rapid chloride penetration [14] which categorizes samples into levels of vulnerability to the chloride attack. More precise measurements with chloride profile assessment will be presented in this paper. 2. Materials Two types of concrete mixtures were prepared for this study with low compressive strength (labeled as L) and with normal strength (labeled as N). Each type was prepared without chlorides and with 1% of cement weight NaCl addition to the mixture prepared from ordinary Portland cement CEM I-42,5R, sand and natural crushed aggregate. The mixtures composition is shown in the Table 1. The concrete was mixed in a 50 l laboratory mixer for 10 minutes. The samples were cylindrical with height of 200 mm and diameter of 100 mm. After casting they were vibrated for ap- proximately 30 seconds. As prevention before water evaporation the specimens were covered with a foil. After 2 days after casting they were unmoulded and stored in water. The resulting 28-days compressive strengths and density of the L samples were 24.3 MPa and 2216 kg/m3 and 50.5 MPa and 2340 kg/m3 for N samples, respectively. For chloride penetration tests the specimens were cut to slices with the thickness of 50 mm. In this study the lower strength concrete represents a weak or damaged material that is worth to be re- paired. The different content of cement also play a role 139 http://dx.doi.org/10.14311/APP.2019.22.0139 http://ojs.cvut.cz/ojs/index.php/app Vojtěch Zacharda, Jiří Němeček Acta Polytechnica CTU Proceedings Type of concrete CEM I 42,5R Sand 0/4 Aggregate 4/8 Aggregate 8/16 Water w/c NaCl [kg] [kg] [kg] [kg] [kg] [kg] N 436.4 872.7 290.9 581.8 186 0.43 0 N-1% 436.4 872.7 290.9 581.8 186 0.43 4.36 L 261.8 1150.0 290.9 581.8 210.2 0.8 0 L-1% 261.8 1150.0 290.9 581.8 210.2 0.8 2.62 Table 1. Concrete mixtures composition, mass per 1 m3. Figure 1. An electromigration setup used for accel- erated chloride penetration/extraction tests. in chloride binding that is higher in higher strength concrete. 3. Methods In order to test the effectiveness of the electromigra- tion process two types of accelerated tests were per- formed in an migration chamber. First, the chloride penetration test was made on the concrete samples without chlorides (N and L series) and second, the accelerated chloride extraction test was performed on the samples with 1% NaCl per cement addition (N-1% and L-1% series). The migration chamber composed of two containers with electrolyte solutions. In case of chloride penetration tests, a 3% NaCl solution was used in the compartment with positive electrode and 0.3% NaOH solution in the compartment with neg- ative electrode. In case of chloride extraction, both compartments were filed with 0.3% NaOH solution and chlorides driven into one of them. A DC power source with constant voltage of 20 V was connected to stainless steel mesh electrodes submerged in the electrolytes. The analyzed samples were inserted be- tween the electrodes and the sample sealed (Fig. 1). The accelerated chloride penetration/extraction run for 24 hours. The scheme of tests is shown in Fig. 2. Before the tests of samples with 1% chloride ad- dition, the original total chloride concentration was analyzed in 20 mm depth steps by drilling of the pow- der. Subsequently, the drilled holes were sealed in the samples and samples analyzed further. After the tests, the concentration was analyzed from powder collected in 5 mm depth steps. Always, the powder was mixed with extraction liquid based on potassium dichromate and acetic acid. After at least 24 hours, the solution was analyzed with Cl ion selective electrode and the chloride profile constructed. 4. Results and discussion 4.1. Penetration of chlorides The results of individual accelerated chloride penetra- tion tests were used for calculation of average chloride concentration profile from 3-4 samples as shown in Fig. 3 for L samples and in Fig. 4 for N samples. The profile exhibits a gradual reduction of chlorides from surface of samples, which was exposed to the 3% NaCl solution. The shape of the profile is different in comparison with traditional diffusion tests. Here, the chlorides are driven by an electromotive force which is dominant and the profile is a “convective” one with the “middle bulge” caused by the convective force. It can be seen in the figures that the surface concen- tration after 24 hours exposure is higher by 50% for N samples while the extent of the penetration, i.e. the penetration depth, is larger for L samples. Substantial chloride concentration appears within 40 mm of the L sample while only to 10 mm in the case of N sample. The difference in the surface concentration between L and N samples is related to the amount of cement in the mixture and the ability of the concrete to bind chlorides. The binding is higher for N samples which has twice as cement compared to L samples. The density of the N samples is however higher compared to L samples. Thus the permeability of the N samples is lower and the penetration depth is substantially lower compared to L samples. 4.2. Extraction of chlorides Original and after-the-extraction chloride concentra- tion profiles are shown in Figs. 5 and 6 for L and N samples, respectively. Chlorides are extracted from the sample to the right compartment. The extraction causes decrease in concentration for both sample types. Most significantly, the concentration drops down in the surface region (left part of the profile in Figs. 5 and 6). The average decrease of chloride concentration was about 15% on L samples and 20% on N samples. The decrease in the surface region was more the 50% for L samples and about 40% for N samples. 140 vol. 22/2019 Experimental Studies of Accelerated Chloride Transport in Concrete (a). (b). Figure 2. Scheme of accelerated chloride a) penetration test and b) extraction test. Figure 3. Concentration profile in L samples. The profile is in agreement with the convective transport [10] which leads the chloride ions from left of the samples to the right compartment of the migration chamber. Note again, that the transport lasted for 24 hours with voltage of 20 V. Further acceleration can be reached by using higher voltages or using larger times. Increasing voltage is, however, impractical due to safety reasons and due to Joule effect causing heating of the specimen. Increasing of the exposure time is, on the other hand, possible and higher efficiency leading to further lowering of the concentration would take place. 5. Conclusions The presented paper provides results of accelerated chloride penetration and extraction tests performed on two typical concrete mixtures with different composi- tions with normal and low compressive strengths. The principle of electromigration was applied to penetrate chlorides into the chloride-free sample and to extract chlorides from the sample with some initial Cl con- centration. The chloride profiles with concentrations that can be developed after long diffusion tests were gained here in 24 hours while applying safe voltage of 20 V. The chloride concentrations were assessed from concrete core drills using an extraction liquid and ion selective electrode. Typical convective profiles with steep concentration change and different surface con- centrations were obtained. The different permeability of the samples caused differences in both surface con- centrations as well as in penetration depths. The surface concentration was twice as high in N samples compared to L samples. The penetration depth was three to four times larges in L samples compared to N samples. The first phenomenon can be attributed to a larger binding capacity of N samples given by 141 Vojtěch Zacharda, Jiří Němeček Acta Polytechnica CTU Proceedings Figure 4. Concentration profile in N samples. Figure 5. Concentration profile of L-1% samples before and after extraction. 142 vol. 22/2019 Experimental Studies of Accelerated Chloride Transport in Concrete Figure 6. Concentration profile of N-1% samples before and after extraction. the twice amount of cement in comparison to L sam- ples. Larger penetration depth in L samples is related to their increased permeability given by their lower density and higher porosity. Electrical extraction was found to be an effective tool for lowering of initial chloride concentration. 15- 20% average concentration decrease was reached after 24 hours. The decrease in surface concentrations was even higher (40-50%) which corresponds to the con- vection of ions out of the samples. The extraction process was found to be feasible and effective for both concrete types. Acknowledgements Financial support of the Czech Science Foundation (project 16-11879S) and the Grant Agency of the Czech Techni- cal University in Prague (SGS18/114/OHK1/2T/11) is gratefully acknowledged. References [1] U. Angst, Ø. Vennesland. Critical chloride content in reinforced concrete – State of the art. In Alexander, et al. (eds.), Concrete Repair, Rehabilitation and Retrofitting II. Taylor and Francis Group, London, 2009. [2] J. Cabrera. Deterioration of concrete due to reinforcement steel corrosion. Cement and Concrete Composites 18(1):47–59, 1996. doi:10.1016/0958-9465(95)00043-7. [3] K. Tuutti. Corrosion of steel in concrete. Ph.D. thesis, Swedish Cement and Concrete Research Institute, Stockholm, 1982. [4] C. Page, N. Short, A. El Tarras. Diffusion of chloride ions in hardened cement pastes. Cement and Concrete Research 11(3):395 – 406, 1981. doi:10.1016/0008-8846(81)90111-3. [5] Y. Xi, Z. P. Bažant. Modeling chloride penetration in saturated concrete. Journal of Materials in Civil Engineering 11(1):58–65, 1999. doi:10.1061/(ASCE)0899-1561(1999)11:1(58). [6] Q. Yuan, C. Shi, G. De Schutter, et al. Chloride binding of cement-based materials subjected to external chloride environment – a review. Construction and Building Materials 23:1–13, 2009. doi:10.1016/j.conbuildmat.2008.02.004. [7] M. Sánchez, C. Alonso. Electrochemical chloride removal in reinforced concrete structures: Improvement of effectiveness by simultaneous migration of calcium nitrite. Construction and Building Materials 25:873– 878, 2011. doi:10.1016/j.conbuildmat.2010.06.099. [8] H. E. Cardenas, L. J. Struble. Electrokinetic nanoparticle treatment of hardened cement paste for reduction of permeability. Journal of Materials in Civil Engineering 18(4), 2006. doi:10.1061/(ASCE)0899-1561(2006)18:4(554). [9] P. Hlaváček, J. Němeček. Accelerated chloride migration tests in concrete. In Proceedings of 22nd International Conference Engineering Mechanics, pp. 213–216. 2016. [10] J. Němeček, J. Kruis, T. Koudelka, T. Krejčí. Simulation of chloride migration in reinforced concrete. Applied Mathematics and Computation 319:575–585, 2018. doi:10.1016/j.amc.2017.07.029. [11] J. Němeček, Y. P. Xi. Nanoparticle injection into concrete using electromigration. In Special Concrete and Composites 2014, vol. 1054 of Advanced Materials Research, pp. 6–10. Trans Tech Publications, 2014. doi:10.4028/www.scientific.net/AMR.1054.6. [12] J. Němeček, L. Li, Y. Xi. Electrokinetic nanoparticle injection for remediating leaks in oil well cement. Construction and Building Materials 15:63–72, 2017. doi:10.1016/j.conbuildmat.2017.08.152. 143 http://dx.doi.org/10.1016/0958-9465(95)00043-7 http://dx.doi.org/10.1016/0008-8846(81)90111-3 http://dx.doi.org/10.1061/(ASCE)0899-1561(1999)11:1(58) http://dx.doi.org/10.1016/j.conbuildmat.2008.02.004 http://dx.doi.org/10.1016/j.conbuildmat.2010.06.099 http://dx.doi.org/10.1061/(ASCE)0899-1561(2006)18:4(554) http://dx.doi.org/10.1016/j.amc.2017.07.029 http://dx.doi.org/10.4028/www.scientific.net/AMR.1054.6 http://dx.doi.org/10.1016/j.conbuildmat.2017.08.152 Vojtěch Zacharda, Jiří Němeček Acta Polytechnica CTU Proceedings [13] J. Němeček, R. Šulc, J. Němečková, J. Kruis. Nanoparticles in concrete: Application in fresh and hardened state. In Special Concrete and Composites 2017, vol. 760 of Key Engineering Materials, pp. 3–9. Trans Tech Publications, 2018. doi:10.4028/www.scientific.net/KEM.760.3. [14] ASTM C1202-19 - Standard Test Method for Electrical Indication of Concrete’s Ability to Resist Chloride Ion Penetration. Standard, American Society for Testing and Materials, West Conshohocken. doi:10.1520/C1202-19. 144 http://dx.doi.org/10.4028/www.scientific.net/KEM.760.3 http://dx.doi.org/10.1520/C1202-19 Acta Polytechnica CTU Proceedings 22:139–144, 2019 1 Introduction 2 Materials 3 Methods 4 Results and discussion 4.1 Penetration of chlorides 4.2 Extraction of chlorides 5 Conclusions Acknowledgements References