Size effect predictions by fracture models for a refractory ceramic

Abstract

Refractory ceramics used as kiln furniture are designed to operate at elevated temperatures with a high thermal shock resistance. In practice, however, the material fails due to thermal fatigue after a limited number of cycles. To predict this failure behaviour, it is generally not possible to use linear elastic fracture mechanics due to the fact that the coarse grained, porous material shows a dissipative mechanical behaviour. Differently sized specimens are tested to determine the size effect associated with this material. Four different finite element models widely used for concrete modelling are investigated on their ability to describe this nonlinear failure behaviour, in particular the associated size effect phenomenon. The results of the initially promising Nonlocal Continuum Damage Mechanics (CDM) model indicate that this approach cannot properly describe the observed size effect. The other three models (Adapted Local CDM, Fictitious Crack and Smeared Crack), however, give comparable results with a good description of the observed size effect phenomenon.

Fictitious Crack Modeling of the Fracture of a Refractory Ceramic: Fitting through Experimental Measurements of the Displacement Field around the Fracture Process Zone

“…That is, in successive steps the position of the crack tip is shifted one node simulating a propagating crack. The numerical fitting is presented in figures 1 to 3, together with the experimental values and the numerical fitting obtained by van Gils et al [10,11] through numerical fitting.…”

Calibration of cohesive parameters for a castable refractory using 4D tomographic data and realistic crack path from in-situ wedge splitting test

“…A modified version of the previous CZM to better account for mode-mixed debonding [15] was calibrated via an inverse procedure considering the segmentation of the sample and the evaluation of the Young's modulus in different regions of the sample. Van Gils et al [16] studied deviations from linear elastic fracture mechanics with cohesive laws in refractories, and Hein et al [17] the influence of changing CZM parameters to simulate layered refractories under thermal shock. Both references utilized parameters with no experimental exploration.…”

On the calibration of cohesive parameters for refractories from notch opening displacements in wedge splitting tests

“…The latter ones lead to the formation of fracture process zones (FPZs) where cracks have some cohesion between the newly created surfaces. Linear elastic fracture mechanics (LEFM) was applied for structures with very small defects made of brittle materials, although severe deviations may arise in the presence of a developed FPZ, especially in small specimens [7]. Therefore, this work focuses on cohesive zone models (CZMs) suitable to be calibrated under usual laboratory characterization of materials, aiming to provide first estimates of parameters needed prior to simulating in-service applications.…”

“…The use of traction-separation laws to describe the FPZ [8,9,10] has been successfully applied to refractory materials with different approaches. For instance, CZMs were used to study thermal shocks [11] and to check size effects and deviations from LEFM [7]. Later on, they were utilized for analyzing damage during thermal cycles [12] and fracture [13] in various refractories.…”

“…In few of the afore-cited references, the material parameters were estimated without experimental exploration [7,16]. Some of them briefly comment on parameter calibration without giving many details [12,13,14].…”