Critical heat flux experiments were performed with an alternate high and low heat flux profile in an internally heated annulus. The heated length was 84 inches (213 cm) with a chopped wave heat flux profile over the last 24 inches (61 cm) having a maximum-to-average heat flux ratio of 1.26. Three test sections were employed: one with an axially uniform heat flux profile as a base case and two with 60 inch (152 cm) uniform and 24 inch (61 cm) alternating high and low heat flux sections. The third test section had a 2.15 inch (5.46 cm) section with a peak-to-average heat flux ratio of 2.19 (hot patch) superimposed at the exit end of the alternating high and low heat flux profile.

The steady state gas release, swelling and densification model previously developed for oxide fuels has been modified to accommodate the slow transients in temperature, temperature gradient, fission rate and pressure that are encountered in normal reactor operation. The gas release predictions made by the model were then compared to gas release data on LMFBR-EBRII fuels obtained by Dutt and Baker and reported by Meyer, Beyer, and Voglewede. Good agreement between the model and the data was found. A comparison between the model and three other sets of gas release data is also shown, again with good agreement.

A set of critical experiments, comprising thirteen homogeneous Pu-H/sub 2/O assemblies and twelve UO/sub 2/-PuO/sub 2/ lattices, was analyzed with ENDF/B-IV data and the RCPO1 Monte Carlo program, which modeled the experiments explicitly. Some major data sensitivities were also evaluated. For the Pu-H/sub 2/O assemblies, calculated K/sub eff/ averaged 1.011. The large (2.7%) scatter of K/sub eff/ values for these assemblies was attributed mostly to uncertainties in physical specifications since no clear trends of K/sub eff/ were evident and data sensitivities were insignificant. The UO/sub 2/-PuO/sub 2/ lattices showed just one trend of K/sub eff/, which indicated an overprediction of U238 capture consistent with that observed for uranium-H/sub 2/O experiments. There was however a approx. 1% discrepancy in calculated K/sub eff/ between the two sets of UO/sub 2/-PuO/sub 2/ lattices studied.

The previously reported swelling and gas release model for oxide fuels has been modified to predict fission gas bahavior during fast temperature transients. Under steady state or slowly varying conditions it has been assumed in the previous model that the pressure caused by the fission gas within the gas bubbles is in equilibrium with the surface tension of the bubbles. During a fast transient, however, net vacancy migration to the bubbles may be insufficient to maintain this equilibrium. In order to ascertain the net vacancy flow, it is necessary to model the point defect behaviour in the fuel. This model is reported. Knowing the net flow of vacancies to the bubble, the bubble size, the diffusivity can be determined and the long range migration of the gas out of the fuel can be calculated. The model has also been modified to allow release of all the gas on the grain boundaries during a fast temperature transient.