Abstract

The paper presents the dependencies assessment that occur between physico-mechanical properties, and the parameters describing the geometry of thermal cracks of modified cement pastes. The subject of the research is cement paste modified with metakaolinite, which has been subjected to the influence of an elevated temperature. The pozzolanic additive was used as a substitute for 10% of the cement’s mass. Four series of cement pastes were analyzed, which differed from each other with the class of Portland cement used and the metakaolinite content. Within each series, samples were made with 3 water/binder ratios equal to 0.4, 0.5, and 0.6, respectively. As part of the earlier research, the basic physical and mechanical features of the cement matrix were determined, such as: compressive strength, tensile strength at bending, and apparent density. The tests were carried out in accordance with EN standards on reference and thermally loaded samples. The elevated temperature load caused cracks on the surface of the cement matrix, which created a characteristic network of cracks referred to as the cluster cracks – the thermal cracks – the map cracking. The computer image analysis tools were used to quantify the cracks’ structure. The samples’ surfaces were scanned and the following parameters were determined: the cluster average area, the cluster average perimeter, and the crack average width. The aim of the study conducted was to determine whether on the basis of the measurement of the geometrical characteristics of thermal cracks it is possible to estimate with a good accuracy selected physico-mechanical properties of modified cement pastes. The measurement of geometrical features of material’s cracks is a non-destructive and non-invasive method, in contrast to tests aimed at determining, in particular, the mechanical properties. The statistical analysis tool, i.e., the least squares method (LSM) was used to define dependencies that occur between material properties and the geometry of thermal cracks. The quality of matching the calculated functional dependencies to the empirical data was evaluated using three diagnostic statistics: the determination coefficient, the standard error of estimation, and the coefficient of random variation. The results obtained indicated the existence of very strong correlations between the compressive strength, apparent density, and the cracks’ geometrical parameters. This allows an accurate estimation of these two material characteristics based on the analysis of the cement paste surface cracks, which has been subjected to elevated temperatures. The analyzes carried out are also of great practical importance because the results obtained can be used to assess the degree of degradation of a cementitious material that has been damaged due to a thermal interaction.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call