Contribution of the Politecnico di Milano to the FUMEX-III Project

In nuclear reactors, irradiation continuously alter the thermal, mechanical, and chemical properties of nuclear fuels (Olander 1976). To assure the safe and economic operation of the nuclear fuel rods in all the operation conditions, there is a need for fuel characterization and optimization through an integrated theoretical, experimental, and computational approach. The aim of computational fuel modelling is to predict the changes in properties and evaluate the thermo-mechanical behaviour of the fuel rods during the life in the reactor. For this purpose, increasingly complex fuel performance codes are developed, which include physical models of the processes taking place in the fuel rods during irradiation (Aybar and Ortego 2005). In this framework, international benchmark exercises on fuel modelling are of high importance for the development of fuel performance codes, since they provide the possibility for cross-comparison and complementary validation of a large number of codes involved. Three such exercises were organized during the last 3 decades: D-COM in the mid 80's (Misfeldt 1983), and the Coordinated Research Projects (CRPs) FUMEX-I (1993-1996) (Chantoin et al. 1997) and FUMEX-II (2002- 2006) (Killeen et al. 2007, IAEA 2011). In extending the previous CRPs on the subject of improving the predictive capabilities of fuel performance codes for extended burn-up and transient conditions, the focus of the CRP FUMEX-III (2008-2012) is on the topics of fission gas release, pellet-cladding interaction (PCI) and dimensional changes (Killeen et al. 2009). The TRANSURANUS fuel performance code (Lassmann 1992, Lassmann 2001) is presently available at the Politecnico di Milano (POLIMI). Based on the assumption of axial-symmetric cylindrical rod and the superposition of a one-dimensional radial and axial description (11/2D approach), the mechanical-mathematical framework of TRANSURANUS allows to analyze, at reasonable computer cost, the integral fuel rod during a complicated, long power history. TRANSURANUS is applied for design as well as for licensing of nuclear fuel, and is therefore used by research centres, universities, industrial partners and nuclear safety authorities. Moreover, TRANSURANUS is featured by a flexible structure into which physical models can easily be incorporated. A review of the validation of the code is given in (Van Uffelen et al. 2007). In line with the specific research objectives of FUMEX-III, a primary interest of POLIMI within research on computational fuel modelling lies in the analysis of the behaviour of LWR-UO2 fuel rods during both normal reactor operation and transients. According to the original POLIMI proposal (Luzzi 2008), the main topic of interest of POLIMI in the frame of FUMEX-III was the modelling of PCI, with the aim of predicting the PCI failure thresholds by means of the TRANSURANUS code. However, subsequent assessments of the prediction capability of TRANSURANUS (Pastore et al. 2009a, Pastore et al. 2009b) pointed out that the incorporation of new physical models in the code, with the aim of improving the description of the integral fuel rod behaviour under power transient and pellet-cladding mechanical interaction (PCMI) conditions, is a prerequisite for an accurate modelling of PCI. In particular, developments are needed of the modelling of the fuel swelling due to fission gas build-up (fission gas swelling). In the current version of the TRANSURANUS code, the fission gas swelling rate is described by means of an empirical correlation and neglected under PCMI conditions (due to the lack of a description of the process dependence on the fuel stress state). Moreover, some room for improvement was noticed for the treatment of fission gas release (FGR), which is physically coupled with the fission gas swelling. On this basis, a new physics-based model of fission gas swelling and release for the TRANSURANUS code has been recently developed in the frame of a collaboration between the POLIMI and the ITU (European Commission, Joint Research Centre, Institute for Transuranium Elements, Karlsruhe, Germany). The model calculates the fission gas swelling and release through a physical description of the underlying microscopic processes, consistently considering the coupling between the two processes as well as their dependence on the fuel stress state. The developed model has been firstly implemented as stand-alone version, namely, a computer program has been set up, which receives the fuel fabrication data, temperature, hydrostatic stress and specific power as input and performs the model calculations for a single point in the fuel. The application of the stand-alone version to the analysis of power-ramped AGR-UO2 fuel has allowed a first verification of the model through comparison with experimental data of local fission gas swelling. Subsequently, the model has been incorporated in the TRANSURANUS code and applied to the integral analysis of LWR-UO2 fuel rods under both normal operation and transient reactor conditions, allowing a first assessment of the predictions against experimental data of FGR. This report gives an account of the present state of development, implementation and validation in the TRANSURANUS code of the new model of fission gas swelling and release, with focus on the application in the frame of FUMEX-III. A description of the model is given and the stand-alone version calculations are discussed. The first results obtained by applying the model within the TRANSURANUS code are presented, including those of some priority cases of FUMEX-III. Despite the encouraging results, the work is still continuing. The conclusions and perspectives are outlined in the last section