This page describes the DDIF2 stress potential which describes a damaging law used in CEA’ fuel performance codes. See [1–3].

The DDIF2 behaviour is used to describe the brittle nature of nuclear fuel ceramics and is usually coupled with a description of the viscoplasticity of those ceramics (See for example [6]).

Internally the DDIF2 stress potential is derived from the Hooke stress potential, so the definition of the elastic properties follows the same rules. See this page for details.

Local coordinate

This description is currently limited to initially isotropic behaviours, but the damage is described in three orthogonal directions. Those directions are currently fixed with respect to the global system. For \(2D\) and \(3D\) modelling hypotheses, those directions are determined by a material property, which external name is AngularCoordinate, giving the angular coordinate in a cylindrical system.

Material properties associated with damage

The description of damage is based on the following material properties:

In each case, a material property must be given as a value or as an external MFront file.

Fracture energies

Following Hillerborg approach (see [7]), softening slopes can be related to fracture energies by the mesh size. Thus, rather than the softening slopes, the user can provide the fracture energies through one the fracture_energy or fracture_energies options. In this case, an array of three material properties, which external name is ElementSize, is automatically declared.

External pressure effect

The effect of external pressure on the crack surface can be taken into account using the option handle_pressure_on_crack_surface. If this option is true, an external state variable called pr, which external name is PressureOnCrackSurface, is automatically declared.

Resolution algorithm

By default, the DDIF2 damage behaviour is treated using an algorithm based on statuses. In each damage directions, the damage state is kept constant during the Newton iterations. Once converged, the consistency of the damage state with the solution found is tested. If the state is inconsistent, the iterations are restarted with a new state.

The implicit scheme is divided in two steps by default. In the first step, the time step is set to zero before the prediction stage. This is meant to filter all viscoplastic flows and the convergence is thus performed only on the damage state (unless another rate-independent mechanism is considered). Once converged on the damage state, the time step is reset to its original value and the implicit resolution is restarted.

The DDIF2 stress potential (and thus the DDIF2 brick) has two options to change this behaviour:

References

1.
Michel, Bruno, Sercombe, Jérôme, Thouvenin, Gilles and Chatelet, Rémy. 3D fuel cracking modelling in pellet cladding mechanical interaction. Engineering Fracture Mechanics. July 2008. Vol. 75, no. 11, p. 3581–3598. DOI 10.1016/j.engfracmech.2006.12.014. Available from: http://www.sciencedirect.com/science/article/pii/S0013794406004759
2.
Michel, Bruno, Helfer, Thomas, Ramière, Isabelle and Esnoul, Coralie. 3D continuum damage approach for simulation of crack initiation and growth in ceramic materials. Key Engineering Materials. 2016. Vol. 713. DOI 10.4028/www.scientific.net/KEM.713.155.
3.
Michel, B., Helfer, T., Ramière, I. and Esnoul, C. A new numerical methodology for simulation of unstable crack growth in time independent brittle materials. Engineering Fracture Mechanics. 12 August 2017. DOI 10.1016/j.engfracmech.2017.08.009. Available from: http://www.sciencedirect.com/science/article/pii/S001379441630412X
4.
Monerie, Yann and Gatt, Jean-Marie. Overall viscoplastic behavior of non-irradiated porous nuclear ceramics. Mechanics of Materials. July 2006. Vol. 38, no. 7, p. 608–619. DOI 10.1016/j.mechmat.2005.11.004. Available from: http://www.sciencedirect.com/science/article/pii/S0167663605001882
5.
Salvo, Maxime, Sercombe, Jérôme, Ménard, Jean-Claude, Julien, Jérôme, Helfer, Thomas and Désoyer, Thierry. Experimental characterization and modelling of UO2 behavior at high temperatures and high strain rates. Journal of Nuclear Materials. January 2015. Vol. 456, p. 54–67. DOI 10.1016/j.jnucmat.2014.09.024. Available from: http://www.sciencedirect.com/science/article/pii/S002231151400614X
6.
Salvo, Maxime, Sercombe, Jérôme, Helfer, Thomas, Sornay, Philippe and Désoyer, Thierry. Experimental characterization and modeling of UO2 grain boundary cracking at high temperatures and high strain rates. Journal of Nuclear Materials. May 2015. Vol. 460, p. 184–199. DOI 10.1016/j.jnucmat.2015.02.018. Available from: http://www.sciencedirect.com/science/article/pii/S0022311515001130
7.
Hillerborg, A., Modéer, M. and Perterson, P.-E. Analysis of crack formation and crack growth in concrete by means of fracture mechanics and finite elements. Cement and Concrete Research. 1976. Vol. 6, p. 779–782.