Effect of temperature rise on the microstructure of cementitious materials: A study through X-ray computed microtomography (µ-CT)

•µ-CT was used to evaluate the alterations of mortars under elevated temperatures.•The µ-CT results were corroborated by the TG/DTG tests.•There is variation in pore structure when the test specimen is heated.•Temperature effects over the microstructure of a mortar are expressive from 600 °C. The microstructure of a mortar that was submitted to elevated temperatures undergoes many degradation processes in terms of the morphology of crystalline phases, fissuration and consequently, of its porous structure. As a result in macrostructural scale, many of its physical characteristics and mechanical properties may be affected. This article has the objective of evaluating the effects of temperature rise on the microstructure of cementitious materials, considering the changes on porosity and deterioration of the cement mass hydration products. To do so, mortars with 1:3 trace and (20 × 40) mm dimension were submitted to elevated temperatures of 150 °C, 300 °C, 600 °C e 900 °C, being compared to mortars maintained on ambient temperatures, i.e., 23.7 °C. The cooling regime was not controlled. For the microstructural evaluation, the samples were tested through Computed microtomography (µ-CT), thermogravimetry and derived thermogravimetry (TG/DTG) to verify the alteration on the initial porosity and to identify the peaks on the mass loss, respectively. The samples showed that the pores size distribution is altered depending on the temperature rise, however, it is not possible to confirm that these pores increase every time. Furthermore, the fissuration inducted by the thermic incompatibility between the paste and aggregates may have contributed for the increase on the porosity values, specially from 600 °C. The highest mass loss peak registered by the thermogravimetric analysis occurred on 418 °C and can be associated to the CH decomposition. It is concluded that the behavior of the mortars submitted to elevated temperatures is not linear, given the most expressive results occurring from 600 °C.