*5.2. Thermal Analysis Results*

The amounts of gypsum formed during the sulfate exposure period were quantified by TGA. To avoid an overestimation of the quantities of gypsum formed (due to the continuous dehydration of the amorphous silica gel in the temperature range of ~90–150 ◦C), the

methodology, originally developed by Kim and Olek (2012) [2] for the estimation of the calcium hydroxide, was implemented. The results of the quantitative evaluation of the gypsum formed during the exposure test are presented in the graphs shown in Figure 7.

**Figure 7.** Amount of gypsum formed during exposure to (**a**) sodium, (**b**) magnesium and (**c**) aluminum sulfate solutions.

As it can be seen from these graphs, exposure to the sodium sulfate solution did not result in the formation of any significant amounts of gypsum compared to other sulfate solutions. Specifically, at the end of the exposure period, only a small amount (about 2.5 g of gypsum per 100 g of paste) was observed in cement #2 paste samples. In contrast, exposure to magnesium sulfate solution resulted in the formation of much more significant amounts of gypsum in all paste samples, except in the case of carbonated wollastonite, i.e., cement #1 samples (only about 0.5 g/100 g of paste was formed in this case). By the end of the 120 day exposure period, ~24 g of gypsum per 100 g of paste formed in cement #3 samples, and ~4–6 g of gypsum per 100 g of paste formed in cement #2 and #3 samples. In the case of the aluminum sulfate solution, about 30–32% (wt.) of gypsum formed in all paste samples within the first 12 h of exposure. This was due to the high acidity of the solution (pH ≈ 3–4). Consequently, this resulted in the quick decomposition ("digestion") of the calcium carbonate phase, pH stabilizing at around 6–7, binding of released calcium ions by sulfate ions and precipitation of gypsum. That amount of gypsum remained more-or-less constant throughout the entire exposure period.
