14.4.5 Role of chemical admixtures
Chemical admixtures are normally used to reduce the limitations of cement hydration, with examples being: water reducer, superplasticizer, retarder, accelerator, shrinkage preventer, segregation reducer, and heat evolution reducer. Interactions between chemical admixtures and cement phases are very important and need to be examined carefully in order to obtain the desired properties and the most durable construction materials. In general, chemical admixtures affect cement particles differently depending on the type of cementitious materials and type of admixture and content. Surface adsorption takes place when organic admixtures meet cement particles due to electrostatic forces between charged particles and ionic groups of admixture molecule (e.g., SO3−, COO−) (Jolicoeur and Simard, 1998). Organic admixtures, e.g., lignosulfonates contains hydrophobic, polar and ionic groups acting very usefully for changing surface chemistry of cement particles. Due to very fast setting of CSA cement thus hindering the workability, chemical admixtures such as retarder are required in order to modify the rheology of cement, mortar and concrete slurries in maintaining high early strength as desired by the construction design. For example, organic chemical admixtures such as carboxylic retarders can adhere to the precipitated hydration products, thus hampering further growth on their surfaces. Zajac et al. (2016) studied the effect of retarders on the early hydration of CSAs. This paper studied the use of different retarders (sodium gluconate, tartrate, and borax) and investigated the developed hydration products. Used CSA clinker consisted of 24.8 wt% of C4A3Ŝ, 52.4 wt% of C2S, 6.6 wt% of C4AF, 2.1 wt% of C2F, 1.2 wt% of CA, 1.9 wt% of CaCO3, 0.5 wt% of K2SO4, 1.5 wt% of C2KŜ, and 2.2 wt% of anhydrite. Two cement samples were prepared as ground clinker, and ground clinker with 10 wt% of anhydrite. To study the hydration and pore solution, a w/c ratio of 2.0 and 2 wt% of each retarder were used. For microstructural analysis, the w/c ratio was 0.5. Without retarder, both samples showed very fast dissolution rate with a huge release of heat during the induction period. CSA cement with the presence of additional anhydrite showed the highest heat release. However, the cumulative heat releases of samples with the presences of sodium gluconate and borax at early hydration (<10 h) were the largest and smallest, respectively, for ground clinker without additional anhydrite. For ground clinkers with additional anhydrite, and tartrate and sodium gluconate, their cumulative heat releases were the highest and about the same values. The cumulative heat release of samples with the presence of gluconate at 100 hours was the smallest. In 30 minuets, 5 wt% of ettringite (AFt) can be formed in samples without additional anhydrite and retarder and the formations were retarded to form 5.7, 4.6, and 0 wt% and of AFt at 8 hours, with the use of tartrate, gluconate and borax, respectively. Thus, borax was the most powerful retarder, according to the heat of hydration experiment. The amount of AFt found at 168 hours from the sample with the presence of gluconate was smallest. Cement clinker with additional anhydrite contained 4.9 wt% of AFt in 30 minutes. When tartrate, gluconate, and borax were used, AFt formations were 3.8, 3.9, and 0.0wt%, respectively at 8 hours. It was suggested that tartrate and gluconate retarded the dissolution of C4A3Ŝ and precipitation of AFt via the adsorption of negatively charged tartrate and gluconate onto AFt surfaces, thus inhibiting its growth while borax acted as pH reducer to retard the dissolution of C4A3Ŝ phase.
Chemical admixture such as superplasticizers are used to reduce the water to cement ratio and to control the setting time, while maintaining the flow ability of cement pastes. SP molecules can stick to cement particle surfaces that change the cement surface charges in order to repel each other. Ma et al. (2014) studied the compatibility between polycarboxylate (PC) superplasticizer and belite-rich SAC, by investigating the setting time and hydration properties. The used cement consisted of 25.4 wt% of C4A3Ŝ, 56.2 wt% of C2S, 6.6 wt% of C3A, and 11.8 wt% of C4AF, which was made of raw meal containing limestone, FA, and FGD–gypsum and which was sintered at 1320°C. Cement clinker was mixed with 10 wt% of gypsum, with varying amounts of PC from 0.025%–0.25% (42.83% of active phase). The water to solid ratio was 0.26. Heats of hydration during the induction period for PC-added samples was not significantly different, but total heat releases of samples without PC and with 0.25% PC were relatively different in showing the retardation of hydration. With less than 0.075% PC, total heat releases were not significantly different. Due to the negatively charged PC, it could effectively adsorb onto positively charged cement particles, prolonging initial setting times from 18 minutes (without PC) to 38 minutes (with 0.25% PC). The compressive strengths at 28 days curing of samples without PC and with 0.25% PC were 61.4 MPa and 66.8 MPa, respectively. The maximum strength was at 0.075% PC addition, relating to the largest amount of AFt formation. In addition. Furthermore, the most advantageous effect of PC was a modification of particle–particle interaction, changing from macropores to micropores over time. Another work investigating the addition of polycarboxylic superplasticizer (25 wt% of active matter) in the range of 0%–0.4% in high SAC was reported by Garcia-Mate et al. (2012). The used CSA cement consisted of 72.3 wt% of C4A3Ŝ, 14.5 wt% of C2S, 6.8 wt% of calcium titanate, and 2.5 wt% of C4AF, 1.6 wt% of MgO, 1.4 wt% of C2MS2, and 0.9 wt% of calcium sulfate (where M stands for MgO). Gypsum additions were observed at 10, 20, and 30 wt%. Cement pastes were prepared using w/c ratios of 0.4 to 0.5 with the higher content of gypsum. Water to cement ratios of 0.5 and 0.6 were varied, in case of mortar preparation. The rheology of cement pastes showed shear thinning behavior for every amount of gypsum addition (without PC) and became Newtonian fluids when PC was added. At only 0.1 wt%, viscosity was reduced significantly. At 0.4 wt% addition, the viscosity became overdeflocculated showing an increase in the viscosity. The optimum PC content for 10, 20, and 30 wt% of gypsum was 0.2%, 0.15%, and 0.15%, respectively. An increase in gypsum content did not affect the rheology of cement pastes. The ettringite amount was the largest (47.6%) at 7 days in 30% gypsum added sample without PC addition while AFt amounts of 40.6% and 39.4 for samples with w/c ratios of 0.4 and 0.5, respectively at the same content of PC (0.15%) could be obtained. In samples with 10 and 20 wt% gypsum additions, AFt amounts were lower than that with 30% addition at 7 days curing. Without PC, cement paste with 30 wt% of gypsum showed the lowest open porosity, whereas the lowest open porosity could be obtained in paste with 10 wt% of gypsum with the presence of PC. Cement mortars were prepared in order to investigate strength development. At 7 days, mortars with 10 wt% of gypsum and w/c ratio of 0.5 attained the highest compressive strength (50 MPa). The higher w/c ratio degraded the strength in all cases. Significantly, mortar without PC obtained slightly higher strength than that with PC. In addition, compressive strength of CSA mortar was comparable to that of OPC mortar (50.6 MPa) at 7 days curing. The difference in strength development of those two research papers (Garcia-Mate et al., 2012; Ma et al., 2014) was caused by the water to cement ratio: the lower the w/c ratio, the higher compressive strength—although containing a lower content of ye’elimite phase. In addition, the active phase of polycarboxylate superplasticizer significantly played a role in the different hydration properties.