Rising costs and re-evaluated risks make nuclear power a poor choice, even for developing nations that can afford it, says Jos Goldemberg.
There are 440 nuclear reactors operating around the world, providing about 14 per cent of the world's electricity supply. Most were installed 3040 years ago, when the relative cost of producing nuclear energy made it an attractive option.
After 1985, lower petroleum costs, combined with concerns over nuclear safety (raised by the nuclear accidents at Three Mile Island, United States, in 1979 and at Chernobyl, Ukraine, in 1986), stagnated worldwide expansion of nuclear-generated electricity.
But more recently, concerns over greenhouse gas emissions helped spark a nuclear energy 'renaissance', stimulated by government subsidies. Unlike thermoelectric generation using coal or other fossil fuels, nuclear-generated electricity contributes little on a life-cycle basis to emissions, and could help solve global warming problems.
Now, the most recent nuclear disaster at Fukushima in March 2011 has again dampened enthusiasm. Countries are pausing to re-evaluate nuclear power and ask whether it will truly put them on the right track for sustainable energy.
It is still too early to evaluate fully what the Fukushima accident means for the future of nuclear energy. But several OECD countries (Belgium, Germany, Italy, Japan and Switzerland, among others) have already decided to phase out existing nuclear reactors at the end of their useful life and have cancelled plans for new ones.
Before Fukushima, the International Atomic Energy Agency (IAEA) predicted that nuclear plants would add 360 gigawatts of generating capacity by 2035, or the equivalent of over 200 new reactors; it is now reckoning on half as many.
This is due partly to diminishing public acceptance of nuclear energy in many countries, but also to the increased costs of nuclear security improvements and of insurance premiums for accident-related damages.
The estimated probability of major nuclear accidents, which was considered very small in the past, has increased significantly. The pre-Fukushima estimate for the probability of a major nuclear accident was roughly 1 in 100,000 for the 440 reactors in operation over the next 2025 years.
But the likelihood of core melt and containment failure had been underestimated: the accidents in Chernobyl and Fukushima amount to catastrophic meltdown in four nuclear reactors over the past few decades, more than originally assumed.
A simple calculation shows that in reality, the probability of any of the currently operating nuclear reactor having a major accident over the next 2025 years is 1 in 5000. This means that another major nuclear accident can be expected to occur once every 20 years. Based on the earlier estimate, we were expecting one accident over a 100-year period.
Only 6 per cent of the worldwide capacity for nuclear power is in developing countries: in China, India, Brazil, South Africa, Mexico, Argentina and Pakistan. But by the end of 2008 more than 50 developing countries had approached the IAEA with interest in installing their first nuclear plant.
Of these, it is unlikely that countries with a GDP smaller than US$50 billion would be able to purchase a nuclear reactor costing at least a few billion dollars. Countries would also need electricity grids with a minimum capacity of approximately 10 gigawatts to accommodate a large nuclear reactor.
Eliminating the countries that do not meet these criteria leaves 16 serious candidates for purchasing large nuclear reactors: Algeria, Belarus, Chile, Egypt, Greece, Indonesia, Kazakhstan, Kenya, Malaysia, Philippines, Poland, Saudi Arabia, Thailand, Turkey, United Arab Emirates, and Venezuela.
But a close examination of their other potential energy resources, such as oil, gas, biomass or hydropower, indicates nuclear is not the best option for generating the electricity they need.
In all of them, the cost of nuclear-generated electricity is significantly higher than other options, although estimates vary depending on the availability of gas or hydroelectric sites.
In Brazil, for example, the cost of nuclear energy is at least 50 per cent higher than other options. In Iran, gas is abundant and thus a more cost-effective option.
In economic terms, nuclear energy should be a 'last resort option' for supplying electricity.
So what could motivate developing countries to pursue the nuclear option? Reducing greenhouse gas emissions is not a priority for them, as they are exempted by the Kyoto Protocol only industrialised countries are committed to emissions targets.
The main attraction of the nuclear option seems to be the 'status' and prestige associated with mastering nuclear technologies.
In developing countries, nuclear technology has often been viewed as a passport to the first world and to the bureaucratic self aggrandizement of the nuclear establishment.
And since there is no clear distinction between the technologies needed for peaceful uses of nuclear energy and manufacturing nuclear weapons, there are also concerns that new nuclear reactors increase the danger of nuclear weapons proliferation.
Whatever countries' true motivation, under current conditions, and if the question is how to secure energy supplies for future generations in the developing world, nuclear power is not part of the answer.
Jos Goldemberg is a physicist and Professor Emeritus of the University of So Paulo, Brazil. He has served as Brazil's Secretary of State for Science and Technology and Minister of State for Education.