Effectiveness of Silicon Dioxide (SiO2) Nanoparticles on the Internal Structure and Compressive Strength of the Calcium Oxide (CaO) Plaster at Different Heat Treatment Temperatures

In the last few decades, research into nanomaterials' role in improving building materials' performance has grown and produced excellent results. As a chemical element, SiO 2 is observed in many building materials, such as ceramics, glass, sand, cement, pastes, and binders. In the building materials category, standard lime-based pastes for roofs and screeds (LRP) are used as building insulation materials (BIMS). LRP is a complicated multilayer system composed of various binders and ingredients. This study investigated the effect of various percentages of nanosilica (NS) replacement from 0 to 2.5% by the weight of the LRP. The specimens were demolded after 48 h, and 75% of the total samples were separated and treated for 7 h at different heat temperatures (T) of 40, 60, and 80 °C. After heat treatment, all samples were cured at room temperature (T = 25 °C) for 3, 7, and 28 days. The XRF and EDX analysis indicated that the LRP consisted of SiO 2 and CaO, which primarily transformed into calcium silicate and calcium carbonate throughout the hydration processes. The increase in compressive strength indicated that the addition of NS played an important role in forming calcium silicate during hydration and reduced pore structures. In addition, the highest compressive strength results were obtained at T = 40 °C. Thermogravimetric (TGA) analysis showed that the sample with 2.5% NS had a weight loss of 2.3 times less than the sample without NS content. Based on the compressive strength and acceptable flowability, the optimal percentages of NS content replaced with LRP at the water/binder ratios (w/b) of 60% and 75% were obtained to be 1.5% and 2.5%, respectively. In terms of modeling criteria, the full quadratic model was the most accurate model to predict the compressive strength as a function of w/b, T, curing age (t), and NS (%), of LRP, with a higher coefficient of determination (R 2 ) and lower root mean square error (RMSE).