The Importance of Establishing an Operational Approach for the Selection of Materials in the Reactors of the Future

Abstract

Nuclear scientists and engineers should consider adopting a more operational approach for the purpose of selecting their future materials. For each type of nuclear power generating reactor, for each coolant (water, helium or liquid metal), the next generation of specialists and decision-makers will need to choose and optimise the iron or nickel alloys, steels, ODS (oxide dispersed strengthened steels) and ceramics that are going to be used. It may well be considered that either each reactor type has its own, specific materials, or, in a complementary manner, that the efforts for improvements should be shared. At high temperatures, as found on fuel-cladding liners, heat exchangers or even tubes or tube liners, different types of steels and alloys may be envisaged. It is considered that austenitic steels provide a better creep resistance at high temperature but they must be stabilized by nickel, thereby becoming more expensive. Ferrite steels could be better as far as swelling, mechanical strength and thermal behaviour are concerned. To withstand corrosion, chromium or aluminium, ODS steels could turn out to be good solutions, if they can comply with stringent criteria. Concerning heat exchangers, choices must be made between iron and nickel alloys, according to proposed operating conditions. In the case of sodium-cooled rapid neutron reactors (RNRs), ferritic-martensitic alloys with 9%–12% chromium or chromium ODS steels could prove suitable, especially if we judge by their specific mechanical behaviour, up to at least 700°C. Nevertheless, behaviour of these steels — with respect to ageing, anisotropy, radiation induced segregation, radiation induced precipitation, reduction of activation products and welding — needs be better understood and qualified. Sodium heat exchanger materials should be carefully chosen since they have to withstand corrosion arising from the primary flow and also from the secondary or tertiary flow (either sodium or molten salts, gas or water); therefore, experimental loops are necessary to gain improved understanding and assessment of the designs envisioned. One way to improve alloys is through thermal, mechanical treatments or by surface treatments. A better way could, however, be to improve the nanostructure and mesostructure of the materials chosen at the drawing-board stage, for instance by nano-size cluster dispersion and grain size controls; experimental tests, microscope and spectroscope observations, multi-scale modelling and thermodynamics computing could also help calibrate and implement these improvements. Large, experimental databases and codes will be the keystone to defining more operational knowledge bases that will then allow us to determine terms of reference for the new materials. Failing this, time will be running out — within the next twenty years — to design and develop nuclear prototypes consistent with the criteria laid down for “Generation IV” reactors.