In recent years, the pessimism of the 1990s has given way to some tendencies towards reinstating nuclear power among the priorities of energy strategies in a number of large countries, such as China, India, Iran, and Russia. The National Energy Policy of the USA also is quite symptomatic in this respect. Nevertheless, whatever the motives for nuclear power revival may be, the primacy of non-proliferation will remain an invariable priority of international politics. If large-scale nuclear power is to be considered as a realistic option, there is no escape from the conclusion that the foundation of the industry should be formed by fast reactors. Down the line, successful solution of the problem of controlled thermonuclear fusion may only add to nuclear's capabilities to meet ever-increasing global energy demands.
Beyond energy and proliferation concerns, the issues of nuclear plant safety and radioactive waste disposal are important to consider.
On the waste front, the nuclear engineering expertise built up throughout the years has helped find very efficient ways of radioactive waste disposal. These include various methods of sealing it off from the environment and burying it in carefully chosen geological formations. It is always a problem, however, to demonstrate safety of any storage facility - let alone a spent fuel repository - for a geologically meaningful span of time. This points to the need to develop a fuel cycle that does not add to waste problems, but minimizes them.
A nuclear electricity system based on fast reactors and a closed fuel cycle would make it possible to achieve what has been called "radiation-equivalent management" of nuclear materials. This management involves a process known as "transmutation" of minor actinides and fission products that is being developed as an alternative strategy for reducing and managing long-lived radioactive waste. With a closed fuel cycle for fast reactors, for example, the total activity of nuclear waste would approximate that of mined ore in no more than 150 to 200 years. This is certain to influence public perceptions of waste management.
Regarding plant safety, I cannot but acknowledge impressive achievements in the safety improvement of existing nuclear plants, through the use of probabilistic safety assessments and other measures. However, if we pursue the right innovative nuclear technologies, reactors can be developed that present no chance of severe accidents by virtue of their design, physics and materials. The advantages of such facilities may prove decisive in the public choice.
Such reactors have been referred to recently as "natural safety facilities". They would rely for their safety on laws of nature, rather than on additional engineered safety barriers and extra personnel. For instance, fast reactors can be designed so that their physics would exclude the possibility of serious accidents such as occurred at Chernobyl in 1986 or at Three Mile Island in 1979. (The differences are illustrated in Figure 2.)