A model for predicting coolant activity behaviour for fuel-failure monitoring analysis

A CANDU fuel element becomes defective when the Zircaloy-4 sheath is breached, allowing high pressure heavy water (D2O) coolant to enter the fuel-to-sheath gap, thereby creating a direct path for fission products (mainly volatile species of iodine and noble gases) and fuel debris to escape into the primary heat transport system (PHTS). In addition, the entry of D 2O coolant into the fuel-to-sheath gap may cause the UO2 fuel to oxidize, which in turn can augment the rate of fission product release into the PHTS. The release of fission products and fuel debris into the PHTS will elevate circuit contamination levels, consequently increasing radiation exposure to station personnel during maintenance tasks. Moreover, the continued operation of a defective fuel element may diminish its thermal performance due to fuel oxidation effects. It is therefore desirable to discharge defective fuel as soon as possible. Hence, a better understanding of defective fuel behaviour is required in order to develop an improved methodology for fuel-failure monitoring and PHTS coolant activity prediction.;The model has been implemented as the STAR (Steady-state and Transient Activity Release) stand-alone code written in the C++ programming language using a custom developed finite-difference variable-mesh (FDVM) numerical solution to the mass transport equations in the UO2 fuel grain (matrix). For numerical benchmarking and model validation purposes, the stand-alone STAR C++ code has also been implemented using the COMSOL Multiphysics finite-element commercial software package. The STAR C++ code has been benchmarked against analytical solutions to the release-to-birth rate ratio R/B of both short and long-lived fission product species, as well as the specific analytical solution to the 1291 coolant activity concentration. The STAR model has been validated against the in-reactor X-2 experiments conducted with defective fuel elements containing natural and artificial failures at the Chalk River Laboratories (CRL). In addition, the STAR model has been validated against two documented defect occurrences at a commercial nuclear generating station (NGS).;A mathematical model has been developed to predict the release of volatile fission products from operating defective nuclear fuel elements. The fission product activity in both the fuel-to-sheath gap and PHTS coolant as a function of time can be predicted during all reactor operations including steady-state operation as well as reactor shutdown, startup, and bundle-shifting manoeuvres. In addition, an improved ability to predict the PHTS coolant activity of the 135Xe isotope in commercial reactors is discussed. Moreover, a method to approximate both the burnup and the amount of the tramp uranium deposits in-core, as well as the tramp uranium fission rate is proposed.