Analysis of reactivity measurements in hector on single plutonium-uranium oxide clusters

A series of reactivity perturbation measurements has been carried out in the HECTOR reactor on single plutonium-uranium oxide clusters. These clusters are 14 in. long, with a geometry similar to that used in the Windscale A.G.R., and contain up to 2% plutonium with a Pu-240 content from 3% to 26%. These plutonium contents cover the range encountered both in burn-up and in possible plutonium enrichment of an A.G.R. A method of analysing these measurements has been developed, which uses as a starting point reaction rates produced by any suitable single-cell lattice code. A comparison between the measured and calculated reactivity perturbations for a given single plutonium-bearing cluster then provides a measure of the accuracy of k M calculated by that code for a complete lattice containing that particular fuel. The experimental results have been analysed in this report using reaction rates obtained from the ARGOSY 4 code. Good agreement is found within the errors between calculated and measured reactivity perturbations for all samples. These errors, in terms of k∞ for each sample lattice, are typically ± 0.5% for samples containing 0.25% and 0.4% plutonium, and ± 0.9% for samples containing 2% plutonium. The effect of changing the Pu-240 content of a sample is predicted to within 5% of the change. Temperature coefficients over the range 22 deg. C to 427 deg. C are correctly predicted within 1-2 mN/deg. C. These results are found to be independent, within wide limits, of the reactor environment lattice in which the reactivity perturbation measurements are made. It is possible, therefore, to carry out useful measurements on plutonium-bearing fuel in a lattice loaded with uranium fuel, with a considerable saving in the amount of the plutonium-bearing fuel required. (author)