and content, pore size distribution, and aggregate type and content. In general, the binder type has significant influence on shrinkage and cracking potential of concrete. Sounthararajan and Sivakumar (2013) reported that concrete made with binary blend of cement and FA had higher resistance to drying shrinkage. For a given workability, a mixture containing FA has lower water demand, which can contribute to lower drying shrinkage of concrete (Tangtermsirikul, 1995). Lee et al. (2006) reported that for the given w/cm, autogenous shrinkage of concrete made with SL increases with an increase in the replacement level of SL from 0 to 50%. On the other hand, partial substitution of cement by SL can result in lower drying shrinkage (the term “drying shrinkage” refers to “total shrinkage” which includes shrinkage induced by both autogenous and drying shrinkages) due to the denser microstructure and lower capillary porosity (Li and Yao, 2001). Li et al. (2010) investigated the autogenous shrinkage and the pore structure of the cement paste with ternary blends of FA and SF, or FA and SL. Their results indicated that the use of FA can reduce autogenous shrinkage, but using SF can increase autogenous shrinkage, which can be due to the higher surface tension in capillary pores. El-Chabib and Syed (2012) reported that replacing up to 70% of cement by SL decreased the free drying shrinkage of concrete by more than 45% compared to that of the control mixture. They found that the use of Class F FA is more effective than Class C FA in reducing
drying shrinkage at the same replacement level. For example, SCC made with 60% Class F FA resulted in 20% lower free drying shrinkage compared to the similar mixture containing 60% Class C FA. This can be due to the relatively high pozzolanic reaction of Class F FA which can produce denser microstructure and reduce water evaporation. Güneyisi et al. (2010) studied the drying shrinkage of SCC incorporating various binary and ternary SCMs, including FA, SL, MK, and SF at a w/cm of 0.32 and 0.44. In the case of a binary system, the replacement of cement with FA, SL or MK is shown to significantly reduce the free drying shrinkage of SCC at w/cm of 0.32. For example, the drying shrinkage of the concrete mixtures with 5% and 15% MK were 22% and 26% lower than that of the control concrete, respectively. The partial replacement of cement with either 60% FA or 60% SL resulted in 26% and 15% reduction in drying shrinkage, respectively. However, the binary use of SF from 10% to 15% increased the total shrinkage compared to the control mixture made with 100% OPC. The negative effect of SF on the shrinkage of SCCs was relatively eliminated with the ternary use of SCMs. The lowest shrinkage of 300 µstrain was observed for the mixture proportioned with 45% FA and 15% MK compared to the 480 µstrain for the control mixture.
The effect of SCMs on cracking resistance of concrete under restrained shrinkage is not well documented. A delay in the early-age hydration and rate of strength development of the concrete may lower cracking potential due to the increase in the early-age creep and higher stress relaxation. Replacement of cement with Class F FA is typically more effective in delaying the strength gain compared to Class C FA or SL replacement. Hwang and Khayat (2010) evaluated the cracking potential of SCC made with two w/cm values of 0.35 and 0.42, and three blended binder types, including FA, SL, and SF. The type of binder is shown to have considerable influence on shrinkage cracking potential. For a given w/cm, SCC made with the B1 binder (70% OPC + 25% Class F FA + 5% SF) exhibited 130% to 380% greater resistance to restrained shrinkage cracking than similar SCC made with the B2 (70% OPC + 25% SL + 5% SF) or B3 (quaternary blended cement containing Class F FA, SL, and SF) binders. Ray et al. (2012) investigated the effect of three different SCM combinations (B1: 30% SL + 5% SF, B2: 20% FA + 5% SF, and B3: 10% MK) and three different w/cm (0.4, 0.35, and 0.3) on shrinkage cracking potential of HPC targeted for bridge deck. The mixture made
with 10% MK exhibited 10% lower drying shrinkage after 90 days of drying compared to mixtures made with other SCM combinations. In the case of w/cm of 0.40, the mixture with binder type of B1 developed elapsed time to cracking of 22 days compared to 20 days for mixtures made with B2 or B3 binder type. Li et al. (1999) indicated that the use of SF as a partial replacement of cement not only increases the cracking tendency, but also increases the crack width in the restrained shrinkage ring test. Bucher et al. (2008) investigated the autogenous, free drying, and restrained shrinkages of mortars made with three cements containing 0%, 5%, and 10% limestone. The autogenous shrinkage during the first 3 days was highest for mortars without limestone (215 μstrain) and the lowest for mortars with 10% limestone (185 μstrain). The amount of drying shrinkage also decreased with increasing limestone content. Restrained mortar samples produced with cement without limestone exhibited elapsed time to cracking of 87 hours. The presence of limestone increased the time to cracking slightly, but all samples cracked after 96 hours. Based on the obtained results, mortars made with PLC exhibited slightly lower shrinkage and a lower tendency to shrinkage cracking compared with similar mortars produced with OPC.