1 Section for Construction Materials, Department of Civil Engineering, Technical University of Denmark2 Department of Civil Engineering, Technical University of Denmark
Supplementary cementitious materials (SCMs), such as slag and fly ash, are increasingly used as a substitute for Portland cement in the interests of improvement of engineering properties and sustainability of concrete. According to studies improvement of engineering properties can be explained by a refinement of the pore structure in the paste. Among others, chemical and physical properties of SCMs and curing conditions (moisture and temperature) affect the microstructure of hydrated pastes. This thesis comprises investigations of pore structure of pastes with and without slag and fly ash, providing a valuable knowledge of SCMs, which is relevant for performance based design of concrete structures. In addition, the thesis provides guidelines for porosity investigations with focus on the applicability of the methods and sources of error. Pore structure was here determined by several methods, mercury intrusion porosimetry (MIP), low temperature calorimetry (LTC), scanning electron microscopy (SEM) and water desorption. These methods provide information on various pore characteristics (total porosity, pore threshold, pore size distribution etc.) in different size ranges and do therefore to a large extent supplement each other. Cement pastes (w/b=0.4) with and without slag and fly ash cured at two moisture (sealed and saturated) and temperature (20 and 55ºC) conditions were used to investigate the combined impact of SCMs addition and curing on the pore structure of pastes cured up to two years. Also, the porosity measurements were compared with porosity modelling. The results of this thesis showed that slag caused a refinement of pores illustrated by a decreased threshold pore size for a given volume for pastes cured saturated at 20ºC for 28 days or more. For the investigated fly ash pastes such impact was not observed. Both, temperature and moisture affected porosity parameters (volume of pores, pore threshold and pore size distribution) of all pastes especially after 28 days of curing. A beneficial effect of high slag addition for pastes cured at high temperature was observed where a reduction of the pore volume and threshold pore size were found when comparing with plain cement paste at the same curing conditions. The porosity methods MIP, LTC and SEM have been shown to be suitable to characterise pore parameters of the pastes. MIP is a simple and fast method which covers a large range of pore sizes. As complementary information, LTC can be used as it is not required to dry the sample. SEM is relevant as it is a direct method, however different magnifications are required to analyse a large range of pores and a proper sample preparation (impregnation and polishing). Modelling of the capillary porosity based on assumptions of degree of reaction and product densities gave for plain cement pastes results comparable to MIP data.