Microstructure and compositions of the hydration products

Backscattered electron and X-ray images obtained in the SEM (021,U17,S68) show that the microstructures of hardened pastes of cement or CjS with pfa broadly resemble those of pure Portland cements, though, as would be expected, there is considerably less CH. Reaction rims may be seen around the pfa particles in sufficiently old pastes. 

The hydrated material has been analysed by X-ray microanalysis and analytical electron microscopy. In a 3-day old paste, that formed in situ from alite or belite did not differ significantly in composition from the corresponding product in pure Portland cement pastes (H4). but at later ages Ca/ Si is lower and Al/Ca higher (R25,R26,T44,U 17,U 18,R42). Ca/Si is typically about 1.55, but the value decreases with age and ratio of pfa to clinker. Uchikawa (U20,U17) reported a value of 1.01 for a 4-year-old paste with 40% replacement of cement by pfa. Several of the studies (R25,T44,U20,U 17) showed that the C-S-H was higher in alkalis if pfa was present, but one cannot tell to what extent potassium or sodium apparently present in the C-S-H has been deposited from the pore solution on drying. For material close to the pfa particles in a 10-year-old mortar. Sato and Furuhashi (S92) found a Ca/Si ratio of 1.1-1.2. 

TMS evidence shows that silicate anion condensation occurs more rapidly in CjS-pfa pastes than in pure CjS pastes. Mohan and Taylor (M80) found that at 1 year, about 60% of the silicon (excluding that in unreacted pfa) was present as polymer in the former case and 40% in the latter. Uchikawa and Furuta (U9) reported data showing ratios of silicon present as polymer to that present as dimer of 0.56 for a pure CjS paste 180 days old and of 0.80 

---