Porosity of geopolymers using complementary techniques of image analysis and physical adsorption of gases

Abstract

Previous research on geopolymers has not fully established their porosity and its influence on the matrix structure, as well as its relevance to mechanical and durability properties, supporting the potential of this material as a sustainable alternative to traditional construction materials. In this study, three geopolymer mortar (GM) mixtures were prepared: the first was obtained with fly ash (FA) without mechanical grinding (GM_FA), the second with FA that required crushing and sieving through a #200 sieve (GM_FA_200), and the third was a GM with FA that required crushing and sieving through a #325 sieve (GM_FA_325). The main objective was to evaluate the porosity of the geopolymeric paste and the interfacial transition zone (ITZ) between the aggregate and the geopolymerization products. Due to the susceptibility of this area to develop higher porosity, which leads to reduced mechanical properties and durability, it has become a significant focus of investigation in materials such as concrete and mortar. These analyses were carried out using physical adsorption of gases (PAG), and a methodology for image analysis of GM microporosity was implemented using micrographs obtained from a scanning electron microscope (SEM) and processed with the NI Vision Assistant 8.6 software (VA). The results from both image analysis and physical adsorption demonstrated that the GM_FA_325 matrix exhibited 19% less porosity compared to the GM_FA matrix. The results confirmed that GMs are predominantly mesoporous. It was observed that GM_FA_325 has the lowest total porosity, resulting in a denser and more compact microstructure, which is a key factor in its mechanical performance and potential applications as an eco-friendly construction material for coatings and precast elements such as blocks, panels, and similar products. In addition, image analysis using VA is highlighted as an efficient, cost-effective, and complementary technique to PAG, enabling robust results and resource optimization.

This research was funded by Fondo Sectorial de Investigación para la Educación SEPCONACYT,grant number A1-S-15401 for which we are grateful of the current Secretaría de Ciencia, Humanidades, Tecnología e Innovación (SECIHTI) for its support and its doctoral and postdoctoral scholarship program.

Peer Reviewed

Postprint (published version)