AbstractThe limitations on the performance of breeding blankets in a fusion power plant are evaluated. The breeding blankets will be key components of a plant and their limitations with regard to power density, thermal efficiency and lifetime could determine to a large degree the attractiveness of a power plant. The performance of two rather well known blanket concepts under development in the frame of the European Blanket Programme is assessed and their limitations are compared with more advanced (and more speculative) concepts. An important issue is the question of which material (structure, breeder, multiplier, coatings) will limit the performance and what improvement would be possible with a ‘better’ structural material. This evaluation is based on the premise that the performance of the power plant will be limited by the blankets (including first wall) and not by other components, e.g. divertors, or the plasma itself. However, the justness of this premise remains to be seen. It is shown that the different blanket concepts cover a large range of allowable power densities and achievable thermal efficiencies, and it is concluded that there is a high incentive to go for better performance in spite of possibly higher blanket cost. However, such high performance blankets are usually based on materials and technologies not yet developed and there is a rather high risk that the development could fail. Therefore, it is explained that a part of the development effort should be devoted to concepts where the materials and technologies are more or less in hand in order to ensure that blankets for a DEMO reactor can be developed and tested in a given time frame.