Abstract

In this work, a computationally efficient law of a nuclear fuel property obtained from molecular dynamics calculations is implemented in a fuel performance code, bridging the gap between atomic and pellet scales. The Cooper-Rushton-Grimes potential is employed through molecular dynamics simulations to compute the heat capacity of stoichiometric mixed-oxide fuels U1-yPuyO2 from 1000 K to the melting temperature, and over the whole range of plutonium content from pure UO2 (y=0) to pure PuO2 (y=1). The heat capacity is found to exhibit a peak at the temperature TB = 0.84 Tm, with Tm the melting temperature, for all the compounds with a significant effect of the plutonium content solely at high temperatures (T > 1800 K), i.e. around the peak. This peak is related to the so-called Bredig transition known to occur around 0.8 Tm. An analytical law of heat capacity Cp(T,y) has been established from our molecular dynamics data, and is valid from 700 K to the melting temperature and for the entire range of plutonium content. Concerning UO2, a good agreement is found between our calculations and the most recent experimental measurements. The law we propose in the present study has been implemented in GERMINAL V2, the fuel performance code developed by CEA to predict the in-pile behavior of mixed-oxide fuel in sodium-cooled fast reactors, and tested through the simulations of transient power test operated in the CABRI reactor. The results show that our law yields a lower margin to melting in the case of Reactivity-Initiated Accident, but throughout power ramp transient tests, the results obtained with our new law are very consistent with the reference heat capacity law currently used in GERMINAL.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call