ISSN 0862-5468 (Print), ISSN 1804-5847 (online) 


Mamen Belgacem 1, Benali Farouk 2, Boutrid Abdelaziz 1, 3 , Sahli Mohamed 4, Hamidouche Mohamed 5, Fantozzi Gilbert 6
1 Department of Civil Engineering, Abbè
2 Laboratory of Non-Metallic Materials, Université Ferhat Abbas Sétif 1, Ferhat Abbas University, Sétif 19000, Algeria
3 Mineral Processing and Environmental Laboratory, Department of Mines, Badji Mokhtar University, Annaba 23000, Algeria
4 EMTO-ST Institute, CNRS/UFC/ENSMM/UTBM, Department of Applied Mechanics, Université Bourgogne Franche-Comté, 25000 Besancon, France
5 Emergent Materials Research Unit, Université Ferhat Abbas Sétif 1, Sétif 19000, Algeria
6 University of Lyon, INSA-Lyon, MATEIS CNRS-UMR5510, 69621 Villeurbanne, France

Keywords: Silica-alumina refractory concrete, high temperatures, cracking propagation, nonlocal finite element model

This paper describes an experimental characterisation and a non-local finite element analysis on the influence of the testing temperature on the mechanical properties and cracking propagation in refractory cement bricks. Therefore, isothermal four-point bending and uniaxial compression tests have been carried out at different testing temperatures (25, 500, 800, and 1000 °C) to determine the stress-strain response for each independent testing temperature. Based on this response, material constants are identified using the inverse estimation method. Then, they are introduced in a non-local finite element model using CAST3M software. The experimental results indicate that with an increase in the testing temperature, the thermomechanical behaviour of the refractory concrete shows a critical temperature of 800 °C, for which the compression and tensile strengths are the largest. Their values are respectively around 28 and 9 MPa. The present numerical simulation results indicate two types of crack propagation; continuous crack failures when the temperature varies between 25 and 800 °C and multi-identified cracks producing a localised damage zone at 1000 °C. Notably, the sample tested at 1000 °C requires a deflection of 0.2 mm to achieve 0.3 (30 % damaged). In contrast, the damage variable achieves 1.0 (100 % in damage) for the sample tested at 25 °C with the same imposed displacement (0.2 mm). Finally, the enhanced non-local damage model produces a realistic simulation of the experimental failure mechanisms, proving the validity of the implementation method.

doi: 10.13168/cs.2021.0031
Licence Creative Commons © 2015 - 2021
Institute of Rock Structure and Mechanics of the CAS & University of Chemistry and Technology, Prague
Webmaster | Journal Contact