Optimisation of the operating conditions of a MC-ICP-MS for the isotopic analysis of gadolinium in spent nuclear fuel using experimental designs

Abstract

Besides its natural occurrence, gadolinium is one of the many elements found as a fission product of uranium in spent nuclear fuel. Because of its high neutron absorption cross section for some isotopes, gadolinium acts as a powerful neutron absorber. Thus, isotopic analysis of gadolinium and its concentration in comparison with uranium is of prime interest in understanding and forecasting the behaviour of the nuclear fuel during fission reactions, notably through neutron code calculations. In this work, a multiple collector ICP-MS has been used for the analysis of gadolinium in different samples of spent nuclear fuel. The aim of this work is double: (i) to obtain optimum and stable conditions for isotopic analysis of gadolinium in spent nuclear fuels, (ii) to understand how the different analytical parameters influence the signal. Ten factors have been studied in statistically designed experimentation. It is the first time that an approach based on statistical designs has been chosen to optimise the analytical conditions of MC-ICP-MS, assessing the effect of all parameters of influence on the instrument response in terms of signal sensitivity and mass bias stability. The methodology is detailed. Results obtained on natural gadolinium are discussed in terms of sensitivity and mass bias stability. The robustness of the optimum conditions resulting from statistical designs has been assessed. Using these optimum conditions, we present the first isotopic analysis of gadolinium in spent nuclear fuel by MC-ICP-MS. The accuracy obtained is better than 0.5% for isotopic composition of gadolinium in the spent nuclear fuel (excepted for the 157Gd (10%) due to a very low intensity for this isotope). The uncertainty obtained is better than 0.1% for the ratios Gd/U.