Calibration of cohesive parameters for a castable refractory using 4D tomographic data and realistic crack path from in-situ wedge splitting test. Issue 2 (February 2023)

• Record Type: Journal Article
• Authors: Vargas, R.; Canto, R.B.; Smaniotto, B.; Hild, F.
• Is Part Of: Journal of the European Ceramic Society. Volume 43:Issue 2(2023)
• Page Start: 676
• Page End: 691
• Publication Date: 2023-02
• Subjects: 3D crack path -- Cohesive Zone Model (CZM) -- Digital Volume Correlation (DVC) -- Finite Element Model Updating (FEMU) -- Parameter sensitivity -- Refractory castable -- Wedge Splitting Test (WST); Ceramic materials -- Periodicals; Composite materials -- Periodicals; Matériaux céramiques -- Périodiques; Composites -- Périodiques; Ceramic materials; Composite materials; Periodicals; Electronic journals; 666.05
• DOI: 10.1016/j.jeurceramsoc.2022.09.040
• Languages: English
• ISSNs: 0955-2219

Abstract

Abstract: Crack propagation in an alumina castable refractory with mullite-zirconia aggregates was investigated in-situ using a wedge splitting test performed inside a laboratory tomograph. Four-dimensional ( i.e., 3D space and time) data from digital volume correlation were used to investigate the influence of a realistic crack path on the simulation of the fracture process. A cohesive law was chosen, since toughening mechanisms were present, and calibrated via finite element model updating. When a straight crack path was assumed instead of the experimental crack path, a 10% higher fracture energy and a 35% higher cohesive strength were calibrated. Although the force alone could be used in the minimized cost function, the kinematic information gives valuable insight into the trustworthiness of the geometrical hypotheses assumed in the finite element model. Such framework can be applied to study nonlinear fracture processes for different materials with complex toughening mechanisms such as crack deflection or branching. Highlights: A CZM is calibrated via FEMU using force and DVC data from an in-situ wedge splitting test. The calibration results in differences between experimental and simulated data less than ten times the uncertainty level. 10% higher fracture energy and 35% higher cohesive strength if a straight crack path is assumed instead of the experimental surface.