Fission-product release from CANDU fuel at 1650degC: the HCE4 experiment

The HCE4 (Hot-Cell Experiment 4) experiment was performed to study fission-product releases (FPR) from CANDU fuel at 1650degC. Three sets of tests were conducted to assess the effects of fuel element length, atmosphere, and heating rate on fission-product release. The fuel samples were from the Darlington reactor (burnup of ∼8.8MWd/kgU, maximum linear power of 42 kW/m) and an irradiation in the NRU reactor (burnup of 6.3MWd/kgU, maximum linear power of 64kW/m). The first set of tests was performed to assess the effect of fuel element length on fission-product release from clad samples of CANDU fuel at 1650degC in a 76mol% steam and 0.5mol%H2 (balance Ar) environment. For most fission products released under the test conditions (Xe, Kr, Cs and I), no significant differences were found between the releases from clad fuel samples of different lengths (20 and 100 mm). The fractional Te-129m release seems to have been higher from one of the shorter fuel samples, but the uncertainty in this result is large due to poor counting statistics. The difference in the Te-129m releases may arise from its affinity for unoxidized Zircaloy and local differences in the extent of clad oxidation. These results indicate that it may not be necessary to account specifically for any effect of fuel element length on releases of volatile fission products from the fuel in CANDU safety analysis. The possible exceptions to this conclusion are those species that are known to have a specific chemical interaction with the cladding that may be affected by steam ingress into the fuel element, e.g., Te and Sb. The second set of tests was performed to assess the effect of environment (inert, steam and air) on fission-product release from clad samples of CANDU fuel at 1650degC. The release of the volatile fission products (Kr, Xe, Cs and I) from the clad fuel samples were low (<10%) in the Ar/2%H2 environment, and high (75-100%) in the steam and air environments. The Te-129m release in the inert atmosphere test was negligible. The Te-129m releases in the oxidizing atmospheres varied significantly (30-80%) and did not appear to depend on the nature of the oxidant (steam or air). Some Ru release (∼15%) was observed beginning about 3600 s after the onset of rapid Cs release in the air environment. No Ru release was observed in the inert or steam atmosphere tests. No statistically significant releases of Ba, Eu, La, Nb, Nd, Pr, Y and Zr were observed in these tests. The third set of tests was performed to assess the effect of heating rate on fission-product release from full-pellet-cross-section fuel fragment samples in an inert (Ar/2%H2) environment. Good results were obtained on releases of volatile fission products (Cs, I) in these tests, but the noble gas (Kr, Xe) release data are not as reliable because of experimental difficulties. There appeared to be little effect of heating rates between 0.2 and 6degC/s on FPR from fuel samples heated to ∼1650degC in an inert (Ar/2%H2) atmosphere. Heating rate effects are thought to be more significant for fission products located on grain boundaries and the fuel used in the HCE4 experiment had a relatively small grain-boundary inventory. The data from the HCE4 fission-product release experiment appear to be suitable for use in the validation of the SOURCE fission-product release code. (J.P.N.)