Fracture and Cohesive Parameter Identification of Refractories by Digital Image Correlation Up to 1200°C

Abstract

The mechanical and fracture properties of refractory ceramics are determined by means of an inverse identification procedure between experimental data and numerical simulations. An experimental set-up is proposed to perform Wedge Splitting Tests (WST) at elevated temperature with Digital Image Correlation (DIC) to assess the crack propagation. The ceramic Young's Modulus, fracture energy and strength are determined by indirect confrontation to Finite Element simulations of crack propagation in WST specimens employing Cohesive Zone Modeling (CZM). The variations of the force and crack length are used to set an inverse problem for estimating the material parameters for various temperatures. The method, illustrated through the analysis of an industrial refractory ceramic from 25 • C to 1200 • C, combines experimental and numerical approaches to understand and optimize the fracture behaviour of refractories in application.Refractory ceramics are widely employed in parts that are used in high temperature environments such as, e.g., glass furnaces, steel ladles or blast furnaces. These parts are subjected to severe thermal gradients that may provoke the initiation and propagation of cracks. The presence of such cracks is critical since they may weaken the whole structure and locally enhance the corrosion effects, leading to a lifetime reduction of the furnaces.A major challenge thus consists in predicting the initiation and propagation of such cracks. To this end, modeling the temperature distribution within the part as well as providing realistic description of failure can help in the definition of a robust design. These models require some essential parameters such as the refractory fracture properties (critical energy release rate and strength) and also their variations from room temperature to really high temperature (up to 1500 • C).

UO2 micron scale specimen fracture: Parameter identification and influence of porosities

“…In addition to the strength and fracture toughness, required as inputs, a third parameter intervines in CZM: The traction-opening displacement profile. This parameter mainly has an influence on dynamic phenomena such as crack initiation [28,44,45], crack branching or unstable propagation [28,46], whereas similar results are obtained whatever the chosen profile in case of stable rectilinear crack propagation [45,46] provided the characteristic length is small with respect to the specimen dimensions.…”

On the effect of sintering temperature on the fracture energy of an Alumina-Mullite-Zirconia castable at 600 °C

“…A first case in which the Crack Mouth Opening Displacement (CMOD) was measured from images of MgO-C samples in WSTs up to 1500°C was reported by Brochen et al [41]. Moreover, to the authors' best knowledge, only one work reported WSTs analyzed with full-field measurements via DIC at a temperature of 1200°C) [42] on a commercial refractory for the calibration of a cohesive zone model. Pan et al [43] studied through WSTs the influence of different sintering temperatures on the fracture energy evaluated at room temperature.…”

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

“…The same models may also be used to study interface cohesion in such heterogeneous materials [14,15,16]. More recently, several examples are found about its usage to simulate initiation and propagation of macrocracks [17,18,19,20,21,22].…”

“…More recently, inverse algorithms were used to find the best fit of the loading curve [18,19]. Vargas et al [20] used measured CMODs to prescribe boundary conditions and used the force measurements in the cost function of a Finite Element Model Updating (FEMU) procedure, while Doitrand et al [21] also used force measurement and crack lengths estimates to find the best fit through manual optimization.…”

“…The authors also calculated displacement residuals between experimental and numerical results, as well as gray level residuals directly in the images using the kinematics from the Finite Element (FE) simulation. Doitrand et al [21] considered strain fields assessed via DIC to look for crack branches on the surface, to estimate the crack length using a user-defined threshold, and to define a tortuous crack path with cohesive elements. The authors also tested a set of different parameters to check which one would best fit the experimental force and crack lengths.…”