Towards a more durable concrete repair system by the development of chloride entrapping agents for repair mortars and innovative surface treatments of the concrete substrate

The general objective of this PhD thesis is to improve the durability of concrete repair systems. Therefore, two major subject matters were considered including the adhesion between concrete substrate and the repair mortar as well as the durability of the repair mortar when shrinkage and chloride ingress are concerned. Removal of degraded concrete during repair activities might have a detrimental impact on the bond strength between the concrete substrate and the repair mortar. The first objective of this thesis was to investigate how concrete removal techniques affect substrates of different compositions, and as a result, the bonding with repair mortar. Different aggressive concrete removal techniques such as jack-hammering (JH) and water jetting (WJ) were compared and the experimental results showed that JH resulted in the highest detrimental impact on the substrate and so the bond strength between the concrete substrate and the repair mortar. Although there are many research studies on this topic, the findings of this section added valuable knowledge to the already published results. It was suggested to consider the removal depth as a criterion to select and compare removal techniques in laboratory studies. The other novelty of this section was to study the effect of concrete substrate properties such as mix composition. It was found that the failure modes and bonding were different from one mix composition to another. Failure modes, surface roughness, and bonding strength, apart from the removal technique, did depend on the characteristics of the mixture such as aggregate size and shape. The bond strength of samples with large aggregate size (i.e. crushed stone concrete mixtures) was more detrimentally affected by the aggressive removal techniques than for other mixtures. The influence of substrate surface roughness on the bond strength was shown to be dependent on the state of stress applied during the bond test, and therefore, on the applied test method. The second goal of this phase of the thesis was to restore the lost bond caused by the JH. To this end, colloidal nano-silica (CNS) and bacterially induced CaCO3 were employed in singular and combined strategies. Finally, a novel solution was introduced to restore the lost bond (measured by the pull-off tests) induced by the aggressive removal technique. The singular application of CNS resulted in the highest adhesion between the JH substrate and the overlay. Some pores were observed in the Scanning Electron Microscopy (SEM) images of JH samples. However, a dense microstructure was observed at the interphase zone of the CNStreated substrates. Moreover, Energy Dispersive X-ray (EDX) analysis showed to be a promising method to ease the investigation of the microstructure of the interlayer in layered composites. Chloride ingress is widely known as a detrimental factor that affects the durability of repaired concrete structures by the corrosion of steel reinforcements in concrete. Therefore, in this phase of the thesis, the focus was on the reduction of chloride ingress into the repair mortar by applying layered double hydroxides (LDHs) or by developing and adding chloride entrapping superabsorbent polymers (SAPs). This should be mentioned that the incipient anode phenomenon which is also very important in repair systems was not studied in this research. The addition of LDHs to the paste samples resulted in the reduction of the free chloride ions in the system. This was assessed by the performance of pore expression experiments. However, when the chloride diffusion results were concerned, one could observe a difference between the bound chloride levels depending on the NaCl concentration. The higher the exposure solution concentration (165 g/l versus 50 g/l), the higher the bound chloride levels were for the samples with 10% LDH compared to those of the control samples. A novel method (incorporation of Ag+) was employed to enable SAPs (Alg-MA) to bind chloride ions. The experimental results were evidence of the reduction in the depth of penetration of free chloride ions into the system when the inhouse developed Alg-MA incorporated with Ag+ (Alg-MA-Ag) were used. This was determined by the chloride diffusion test. The pore expression tests also showed the reduction in the chloride content with the incorporation of Alg-MA-Ag. The bound Cl- ions on the surface of the Alg- MA-Ag were confirmed by SEM-EDX analysis. The last phase of the thesis was related to the shrinkage of repair mortars. In this phase, further tailoring of Alg-MA took place to prevent autogenous and total shrinkage of mixtures. Polyacrylic acid-grafted Alg-MA (P-Alg- MA) was successfully synthesized, incorporated with Ag+ (P-Alg-MA-Ag) and the effects on the repair mortar were compared with the effects caused by a commercially available superabsorbent polymer, SNF. The in-house developed P-Alg-MA-Ag could mitigate the autogenous shrinkage and in some cases reduced the total shrinkage of the mixtures unlike the SNFcontaining and Alg-MA-Ag samples. The reduction in the total shrinkage of mixtures was mainly observed when P-Alg-MA was incorporated with Ag+ and also no additional water was added to the mixtures. In addition, CT-scan images of samples proved the presence of Ag+ in the SAPs at older ages and the reaction between Ag+ and chloride ions after exposure to the NaCl solution. This is important as it shows that Ag+ remains in the SAPs till samples are exposed to the NaCl solution and later can react with the Cl- ions. In addition, a comprehensive investigation was performed on the influence of incorporation of Ag+ in Alg-MA and P-Alg-MA on their swelling properties and also on the hydration and microstructure of repair mixtures. Similar to Alg-MA, the incorporation of Ag+ in P-Alg-MA by immersion in AgNO3 with a concentration of 0.05 M was reported to be optimal. A significant reduction was observed in the swelling capacity of both SAPs in Milli-Q® H2O. Nevertheless, the swelling capacity was not affected in cement filtrate solution after the incorporation with Ag+. Moreover, the lost swelling capacity could be regained with the reaction between Ag+ and chloride ions in the Milli-Q® H2O. However, this was not the case in the cement filtrate solutions. Different isothermal calorimetry results compared to those of Alg-MA-Ag were achieved for P-Alg-MA-Ag. The results showed the impact of incorporation with Ag+ on the hydration with shifting the main peak to a later time when Alg-MA-Ag was used by 0.8 wt% (by the mass of binder). Therefore, 0.4 wt% was considered as the maximum amount of Alg-MA-Ag in the system for further analysis. No impact of incorporation with Ag+ on the hydration was observed for the PAlg- MA. Different results with Alg-MA-(Ag) were also observed when MIP tests were performed. The MIP results showed an increase in the porosity of mixtures with P-Alg-MA-(Ag). When the results of SNF-containing samples were considered, one could notice a significant influence of SNF SAPs on the porosity, mechanical strength, autogenous shrinkage and total shrinkage of mixtures. Among which only autogenous shrinkage was mitigated and the effect on the remaining parameters was negative. In the end, this thesis could successfully introduce a novel method to restore the lost bond strength (measured by the pull-off tests) caused by an aggressive removal technique such as JH and also develop SAPs to bind chloride ions and therefore limit the depth of penetration of such ions into the repair mortars. Moreover, the autogenous and total shrinkage of the repair mortar could be reduced by the incorporation of the in-house developed SAPs (P-Alg-MA-Ag)