Science in Society Archive

Deconstructing the Nuclear Power Myths

Peter Bunyard disposes of the argument for nuclear power: it is highly uneconomical, and the saving on greenhouse gas emissions negligible, if any, compared to a gas-fired electricity generating plant

Limitations due to the quality of uranium ore

A critical point about the practicability of nuclear power to provide clean energy under global warming is the quality and grade of the uranium ore. The quality of uranium ore varies inversely with their availability on a logarithmic scale. The ores used at present, such as the carnotite ores in the United States have an uranium content of up to 0.2 per cent, and vast quantities of overlying rocks and subsoil have to be shifted to get to the 96,000 tonnes of uranium-containing rock and shale that will provide the fresh fuel for a one gigawatt reactor [1].

In addition, most of the ore is left behind as tailings with considerable quantities of radioactivity from thorium-230, a daughter product of the radioactive decay of uranium. Thorium has a half-life of 77 000 years and decays into radium-226, which decays into the gas radon-222. All are potent carcinogens.

Fresh fuel for one reactor contains about 10 curies of radioactivity (27 curies equal 1012 becquerels, each of the latter being one radiation event per second.) The tailings corresponding to that contain 67 curies of radioactive material, much of it exposed to weathering and rain run-off. Radon gas has been found 1 000 miles from the mine tailings from where it originated. Uranium extraction has resulted in more than 6 billion tonnes of radioactive tailings, with significant impact on human health [2].

Once the fuel is used in a reactor, it becomes highly radioactive primarily because of fission products and the generation of the ‘transuranics’ such as neptunium and americium. At discharge from the reactor, a tonne of irradiated fuel from a PWR (pressurized water reactor such as in use at Sizewell) will contain more than 177 million curies of radioactive substances, some admittedly short-lived, but all the more potent in the short term. Ten years later, the radioactivity has died away to about 405 000 curies and 100 years on to 42 000 curies, therefore still 600 times more radioactive than the original material from which the fuel was derived [3].

Today’s reactors, totalling 350 GW and providing about 3 per cent of the total energy used in the world, consume 60 000 tonnes of equivalent natural uranium, prior to enrichment. At that rate, economically recoverable reserves of uranium — about 10 million tonnes — would last less than 100 years. A worldwide nuclear programme of 1 000 nuclear reactors would consume the uranium within 50 years, and if all the world’s electricity, currently 60 exajoules (1018Joules) were generated by nuclear reactors, the uranium would last three years [4]. The prospect that the amount of economically recoverable uranium would limit a worldwide nuclear power programme was certainly appreciated by the United Kingdom Atomic Energy in its advocacy for the fast breeder reactor, which theoretically could increase the quantity of energy to be derived from uranium by a factor of 70 through converting non-fissile uranium-238 into plutonium-239.

In the Authority’s journal [5], Donaldson, D.M., and Betteridge, G.E. stated that, “for a nuclear contribution that expands continuously to about 50 per cent of demand, uranium resources are only adequate for about 45 years.”

The earth’s crust and oceans contain millions upon millions of tonnes of uranium. The average in the crust is 0.0004 per cent and in seawater 2 000 times more dilute. One identified resource, the Tennessee shales in the United States, have uranium concentrations of between 10 and 100 parts per million, therefore between 0.1 and 0.01 per cent. Such low grade ore has little effective energy content as measured by the amount of electricity per unit mass of mined ore [6].

Below 50 parts per million, the energy extracted is no better than mining coal, assuming that the uranium is used in a once-through fuel cycle, and is not reprocessed, but is dumped in some long-term repository. Apart from the self-evident dangers of dissolving spent fuel in acid and keeping the bulk of radioactive waste in stainless steel tanks until a final disposal is found, reprocessing offers very little if at all in terms of energy gained through the extraction and re-use of uranium and plutonium in mixed oxide fuel (MOX) [7].

To date, nuclear power has been built and subsidised through the use of fossil fuels, which have provided the energy for mining, extraction, enrichment and construction. Hence, nuclear power cannot be considered to be free of greenhouse gas emissions. Use of the next grade down could lead to a greenhouse gas inventory every bit as bad as for a gas-fired electricity generation plant, and considerably worse than for a gas-fired co-generation plant, in which both electricity and end-use heating are produced.

As Jan Willem Storm van Leeuwen and Philip Smith point out in their document [6], the cumulative energy produced by a nuclear plant compared with the energy expenditure shows a relatively small net gain over the course of 100 years, which incorporates the time needed to get a handle on the costs of final disposal of the radioactive waste, including the radioactively contaminated structural materials of the reactor. Poor grade uranium will result in a net deficit of energy. Hence a massive worldwide nuclear programme, based on the use of poor grade uranium ores, will add cumulatively to energy demands, rather than resolving them.

Gas-fired plants better than nuclear plants

On that basis, comparisons between the carbon dioxide emissions resulting from the full once-through cycle of a nuclear plant and an equivalently sized gas-burning plant, indicates that with the poorer uranium ores, below 0.02 per cent, the gas-fired plant comes out better, with lower overall carbon dioxide emissions. Indeed, the efficiency of a combined-cycle gas plant can now achieve efficiencies of 56 per cent, more than double that achieved for nuclear power. With gas, the costs of electricity generation have therefore reduced in real terms.

If that gas-fired plant were to be used in co-generation, with the simultaneous production of electricity and useful heat, it would win hands-down for all but the best uranium ores, such as are in use today.

Quite apart from the relative paucity of good uranium ores, if the world were to embark simultaneously on the construction of nuclear plants to replace all coal-fired power plants, that would require one gigawatt-sized (electrical) nuclear reactor to be built every two and a half days for 38 years. Total nuclear capacity, according to Worldwatch’s 1989 State of the World, [8] would be 18 times greater than today, at an annual cost of $144 billion (1989 money).

In his 1990 report for Greenpeace [9] William Keepin came up with similar numbers in terms of requirements but at a more pessimistic annual cost. He pointed out that 5 000 nuclear plants would be needed to displace the 9.4 TW of coal equivalent estimated to be necessary in electricity generation in the world by 2025. Again he figured on the need to begin construction on a new plant every couple of days, assuming a favourable six-year completion time. On the basis of highly optimistic assumptions concerning capital costs and plant reliability, total electricity generation costs (1990 money) would average $525 billion per year.

Nuclear power has an appalling record for long drawn-out construction times. The last reactor to come on line in the United States took 23 years to complete. Fifteen years has been the average time taken in many Eastern European countries using USSR technology. In France, the average time taken for construction to operation is 8 years.

We must also not neglect the considerable and proportionately increasing impact of other greenhouse gases to global warming. The use of nuclear power, even to its best advantage, would not make a jot of difference to the emissions of both methane and nitrous oxide since they are primarily derived from agriculture and in particular from deforestation in the tropics.

France — a test case

There are other costs in running nuclear power plants. Even the nuclear industry now admits that the generation of electricity that originates from nuclear power is not wholly free of greenhouse gas emissions. France provides a useful background to review the efficiency of power generation and consumer preference. In 1999, France generated 375 TWh from its nuclear stations. EdF (Electricité de France) estimates that the cost in CO2 emissions of operating its nuclear plants amounts to 6 g CO2 per kWh [10].

France’s electricity board provides an estimate that includes construction, removing the spent fuel, reprocessing and the storage of wastes. On that basis the total CO2 emissions per year from the operation of its nuclear plants amounts to 2.25 million tonnes. That estimate does not include the mining and preparation of the fuel and hence is not dependent on the quality of the ore.

On the other hand, the Öko-Institute of Germany, taking the full fuel cycle costs into account, comes up with an average figure that is nearly 6 times higher — 35 g/kWh — compared with EdF’s, in which case the total CO2 emissions would amount to 13.125 million tonnes of CO2 equivalent [11].

In 1990, France emitted 144 million tonnes of CO2 equivalent. Therefore, nuclear power’s contribution to the total emissions amounted to 1.6 percent on EdF’s estimates and 9.1 percent, according to the Öko-Institute, both numbers being significant and far from trivial. Nevertheless, banking on the naivete of the public, the nuclear industry exaggerates the advantages of nuclear power in terms of avoided greenhouse gas emissions by comparing its relatively low emissions compared to a coal-fired plant of the same generating size. On that basis, nuclear power comes out 300 times better than coal [12].

As Mycle Schneider, director of WISE (World Information on Safe Energy)-Paris, points out, those seemingly low percentages of carbon dioxide emission from nuclear power plants hide an elemental truth, that the use of nuclear power in France has to be augmented, because of consumer preference, by the use in the home of natural gas-based heating systems, both for hot water and space-heating. For home-heating purposes electricity from whatever source is an expensive and inefficient option, and basically the public, let alone industry, prefers to turn away from it.

In an average French household, aside from transport, two-thirds of the energy consumed is for heating and just one-third for electricity. Consequently, if we are going to make any comparisons as to the carbon-economy of nuclear power versus fossil-fuel systems, we should do so only by taking the end-use preferences into account.

  • First, the differences of any one system lie in its efficiency to provide end-use energy whether for heating or electricity
  • Nuclear power stations are built away from population centres
  • They are relatively inefficient from a thermodynamic point of view, losing as much as two-thirds of the energy produced as heat to the immediate environment (a body of water or cooling tower).
  • The one-third remainder of electricity must be transmitted into a central grid system, where the losses can amount to as much as 10 per cent
  • The net result is that about one quarter of the energy originally released gets to the consumer.

If the consumer were to obtain both electricity and heating from a single co-generation system; the efficiency returns can amount to as much as 90 per cent of the original energy and, therefore, some three to four times better than if nuclear generated electricity were to be the sole source of energy in the home.

A proper evaluation of greenhouse gas emissions therefore demands that the method of production gets taken into account when estimating the total release of greenhouse gases. Both coal and fuel oil used in a co-generation plant are still inferior by a factor of two to a nuclear power/natural gas combination in terms of greenhouse emissions. But that figure is already far-removed from the 300 times advantage so heralded by the nuclear industry and its supporters.

Meanwhile, a natural gas co-generation system is level-pegging with the nuclear power/natural gas combination again in terms of emissions, while being far cheaper to the consumer simply because of the three fold better efficiency in delivering end-use energy. And what about a co-generation system based on biogas? The Öko-Institute estimates that it emits seven times less greenhouse gases in providing end-use energy compared to a nuclear power/natural gas combination [11].

Although concern over the consequences of accidents, such as at Chernobyl or Three Mile Island impinges on the issue, the high, uneconomic cost of nuclear power, more than any other factor, has brought about the industry’s failure to make its mark as a major source of energy in the world.  Increasingly too, local ‘embedded’ generation, such as from a wind farm, or a co-generation plant, is becoming an important competitor against the notion of single large power plants attached to a central grid. In a world ever more competitive in terms of reducing cost, an inefficient, high capital cost nuclear power plant is increasingly an anachronism.

If nuclear power were the answer to a cheap source of energy, why has there been a massive turning away from nuclear power since the 1970s? In the United States, where nuclear technology originated, all civilian nuclear reactors were ordered in the ten-year period between 1963 and 1973, all with huge subsidies from the federal government, including so-called turn-key contracts. No new ones have been ordered since 1973, six years before the accident at Three Mile Island, and a string of cancellations in the 1970s and 80s plus permanent shutdowns meant that total electricity generated by nuclear power went down rather than up. In 1989, the cancellations and shutdowns exceeded those coming on stream by a considerable margin, 4 GW compared to 10.4 GW.

Article first published 11/07/05


Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a tripod?