The first fast reactor, Clementine (25kW(th)), was built in 1946 at Los Alamos as part of the weapons programme. Five years later a 1.4 MW(th) Experimental Breeder Reactor (EBR1) located at the Idaho National Reactor Testing Station became the first reactor in the world to generate electricity. This reactor was fuelled by a mixture of metallic plutonium and enriched uranium in a 20 cm diameter core. A ‘blanket’ of natural uranium surrounded the core and the whole was cooled by liquid metal — a mixture of sodium and potassium (NaK) that was liquid at room temperature. The 1950s also saw other liquid metal-cooled fast reactors operated in the US, Russia and UK. Throughout this period it was assumed that a rapid expansion of nuclear power in the decades to come would create a uranium shortage. One way around this perceived problem was to increase the utilisation of uranium or thorium by using fast neutrons to fission the even numbered isotopes of uranium and plutonium and, more especially, to ‘breed’ fissile material from the more abundant fertile material. Demonstration of breeding was provided by EBR1.

In the event, nuclear power did not expand as quickly as expected (total nuclear capacity today is an order of magnitude lower than the projections of the 1970s), uranium turned out to be more abundant than originally thought and difficulties were experienced in commercialising fast reactors which, hitherto, had existed only as small scale demonstrations. As a result, US research on fast reactors virtually ceased after 19857 and the UK programme was abandoned in 1988.8 France continued to operate the fast reactors Phenix and Superphenix until 2009 and 1998 respectively. Four countries (Russia, India, Japan and China) now have operating fast reactors which are expected to be an essential component of the complementary set of future reactor types known under the heading of Generation IV (Chapter 13).