Nuclear data requirements for thermal reactor design and operation

The accuracy required in calculations of the properties of thermal reactors, in all stages from initial assessment to final operation, is a matter of judgement. The factors to be considered in making this judgement are outlined, and it is suggested that a common basis of accuracy requirements for reactor properties be used. Even if this common basis for properties, such as reactivity, is assumed, there are many ways of meeting the desired objectives for a particular reactor system, ranging from large-scale experiments on a range of lattices to a more theoretical approach relying on sophisticated methods of calculation and accurate nuclear data. Nuclear data requirements should be determined on the basis of the latter approach. A variety of thermal reactors must be considered in arriving at the required accuracy of nuclear data, because materials differ from one reactor system to another, and even for the most common fissile isotopes, such as 235U and 239Pu, the concentrations and neutron spectrum will vary over a wide range. Finally, there are many ways in which the errors in nuclear data may be compounded, but the accuracy requirements at different neutron energies may be arranged to be consistent with the known difficulties of cross-section measurements at different energies. The nuclear data needs for thermal reactors are on the way to being met. The status of available data is briefly reviewed in the light of required accuracies, and outstanding requirements are considered, bearing in mind the contribution which integral measurements can make. (author)