Theme: In light of the increasing debate on global energy issues, it is important to quantify and analyse the role of nuclear power, so as not to fall back on self-complacent mirages and to avoid repeating past errors. Nuclear reactivation is often proposed in response to rising oil prices and climate change. However, its limitations –both qualitative and quantitative– are frequently overlooked, and its costs and disadvantages compared to other options are not taken into consideration. A detailed study of the nuclear option raises serious questions –some unanswerable– regarding its viability and appropriateness both at the national and international levels.
Summary: Reactivating the nuclear option would be a sensible and worthy one only if it were chosen globally and on a very large scale. However, it is not viable to imagine a network of nuclear power plants covering the world demand for electricity in the coming decades, for various reasons, and a stripped-down nuclear programme would have little impact on oil prices: it would not significantly reduce CO2, would very likely come up against a scarcity of fissile uranium, would monopolise energy investments (entailing great financial risk), would generate huge quantities of long-term waste, would be a security risk and would increase the risk of nuclear proliferation. The first part of this analysis quantifies what would be involved in replacing fossil fuels with nuclear energy for electrical generation by 2030, showing it to be an entirely unviable option. The second part will analyse other less ambitious proposals by MIT and the World Nuclear Association, determining what they would mean in terms of reducing emissions and reducing fossil fuel consumption. We also consider their economics and the role of the state, particularly in Spain.
Analysis: In a debate of this kind, a time frame must first be established so as not to be seduced by technological illusions that are impossible to materialise in time to deal with the problems to be faced. The year 2030 is taken a point of reference. This is the year generally used by international bodies and it is a good one for considering the long-term investment decisions involved in energy infrastructure. Furthermore, the Spanish situation must be considered from a global perspective, since the decisions to be made must necessarily take into account the determinants and strategies evolving at the international level.
Energy Challenges for the 21st century
From the energy perspective, there are four fundamental challenges to be
faced in the first half of this century:
(1) The more than likely beginning of a decline in conventional oil production, followed a decade or two later by a similar decline in natural gas production.
(2) The sharp increase in global energy demand, due above all to the emergence of important developing economies, such as China and India, and to the need to improve living conditions in third world countries.
(3) The high concentration of final oil reserves in geo-strategically unstable areas, and the resulting competition for access to these resources.
(4) The need to continue to reduce greenhouse gas emissions to levels that will not destabilise the world’s climate.
This is the context in which a new campaign to construct nuclear power plants has been proposed. Proponents of this idea say that it would significantly attenuate CO2 emissions and would help moderate the increase in fossil fuel prices by providing clean energy to replace them. This would reduce dependence on oil-producing countries and would contribute to geo-strategic stability, while providing the electricity that developing countries need and freeing up the fossil fuel that is essential for them to industrialise. Based on these premises, proponents affirm that the disadvantages presented by nuclear energy –in terms of waste, security, proliferation and costs– would be fewer than the advantages and that, in any case, any problems could be dealt with acceptably and manageably. However, a detailed, quantitative analysis shows that this idea is simply too good to be true.
The International Energy Scenario
The reference scenario of the International Energy Agency (IEA) contained in the World Energy Outlook 2004 (WEO2004)[1] projects that, at the current rate, world economic growth –the determining factor in energy consumption– will progress at 3.2% annually in 2002-30, while the Chinese economy will grow at a rate of 5% a year. The world population will rise from 6.2 billion people to more than eight billion in the same period, with 80% of people living in developing countries. The raw energy needed to feed this economic and demographic growth will increase by 1.7% a year, reaching 16.5 billion tons of oil equivalent (a 60% cumulative increase), with two thirds of the increase in developing countries. Energy intensity (energy/GDP) will increase by 1.5% annually in practically the whole world by the end of the period.
Fossil fuels will continue to dominate, accounting for more than 80% of the energy mix, with oil remaining pre-eminent, due to significant increases in exports from the Middle East. Nuclear energy will lose ground, increasing only slightly (0.4% annually), while natural gas will be the fuel that shows the most growth (2.3% annually). The result of all this is that CO2 emissions will reach 38 billion tons a year in 2030 –a 62% increase over 2002 levels–. Of this increase in emissions, 37% will be due to oil, 33% to coal and the remaining 30% to natural gas. Of all emissions in 2030, oil will generate 39%, coal 36% and gas 24%.
Demand for electricity will double in 2002-30, from 16% of all energy consumed in 2002 to 20% of the total in 2030. Nearly half the natural gas consumed will be used to generate electricity, while nearly 40% of today’s nuclear power plants will be closed during these years. Global electrical consumption will rise by 2.5% a year, from 16,074 TWh in 2002 to 31,657 TWh in 2030, requiring increased generation of 4,800 GWe, including new installations and the replacement of outdated infrastructures. Of this increase, 2,000 GWe will be in OECD countries and 2,800 GWe in developing countries. Obviously, nuclear energy could play a much larger role in this increased generation than the IEA is giving it.
This reference scenario would be viable only if there were continuous growth in oil and natural gas production. It also ignores the possible consequences of such consumption on the world’s climate or implies that some way can be devised to immobilise a significant part of the CO2 generated. Oil production would have to rise from 77 mb/d in 2002 to 121 mb/d in 2030, and natural gas from 2.6 trillion m3/year in 2002 to 4.9 trillion m3/year in 2030, thereby increasing emissions by 62% over 2002 levels, as was mentioned earlier. This is, therefore, a scenario of questionable viability, given the great increase in fossil fuels and unpredictable climatic consequences involved. For this reason, there is increasing agreement on the need to find alternatives to the growing consumption of fossil fuels.
This need is increased by predictions that world oil production is nearing its peak. Although it is true that the AEI says in its WEO2005[2] that it does not expect it to peak before 2030 –if the necessary investments are made– the discrepancy between its forecasts for average prices over the next 25 years (US$35-37/barrel in 2004 dollars) and observed market behaviour in the past two years (hitting over US$70/barrel) makes it seem likely that others are right and that the peak is indeed close. Consider, for example, Figure 1, in which Repsol-YPF forecasts that both oil and natural gas production will peak well before 2030:
Figure 1. Annual world oil and gas production, 1930-2050
Source: 67th EAGE Conference&Exhibition, Madrid, 13-16/VI/2005.[3]
It therefore seems likely that either reduced supply or climatic factors will make it necessary to gradually replace oil and natural gas for other, cleaner fuels in the period in question. In principle, nuclear energy could play a part in this process, for example by generating hydrogen, as is often suggested. This is not the place to join the debate on the most likely date for peak oil production[4], but rather to analyse how nuclear energy could contribute to the reference scenario described above and to replacing fossil fuels, at least for electrical production.
In other words, would it bepossible to maintain the viability of the AEI’s macroeconomic scenario by significantly increasing nuclear capacity so as to greatly decrease fossil fuel consumption and CO2 emissions? The main objective of this paper is to answer this question.
The AEI itself, in its WEO2004, describes an alternative scenario in which global raw energy demand would be 10% lower than the benchmark level, being covered by a significant increase in nuclear energy (14%) and renewable energies (30%), which would make it possible to reduce fossil fuel consumption by 14% in terms of the benchmark figures. As a result of this different energy mix, emissions would rise by 39% over 2002 levels. However, in this scenario, oil consumption continues to increase significantly and the report itself indicates that emissions ‘will not drop sufficiently to guarantee the stabilization of atmospheric concentrations’ of greenhouse gases. Therefore, a 14% increase in nuclear energy would be insufficient to reach the above mentioned goals, leading us to consider something more ambitious –if we heed those who propose re-launching nuclear development to put the brakes on climate change and rising oil prices–.
The French Model: Electricity from Nuclear or Renewable Sources
One of the first options to consider would be to follow the French model and gradually increase the number of reactors to produce a good deal of the world’s electricity by 2030 or perhaps a little later. This would take the pressure off fossil fuels and, in principle, would not require technical innovations of any kind. Electricity would be produced emission-free, based either on nuclear or renewable sources. This would save enormous amounts of natural gas and coal, as well as considerable oil, thus reducing emissions and perhaps putting downward pressure on fossil fuel prices (or at least keeping them steady), while making non-renewable fuel available for a longer period.
Let us consider the implications of this. According to the International Energy Outlook 2005 (IEO2005)[5] by the Energy Information Administration (DOE/EIA), in a business-as-usual scenario, installed electrical capacity would evolve as follows:
Table 1. Installed electrical capacity (GWe)
| 2002 | 2025 | 2030 | |
| Coal | 987 | 1,403 | 1,511 |
| Natural gas and oil | 1,207 | 2,560 | 3,011 |
| Nuclear | 361 | 422 | 437 |
| Renewable sources | 763 | 1,110 | 1,201 |
| Total | 3,318 | 5,495 | 6,160 |
Source: International Energy Outlook 2005, DOE/IEA. Figures for 2030 have been extrapolated.
According to these figures, in order to replace the projected capacity (based on fossil fuels) with nuclear-generated electricity, it would be necessary, before 2030, to build more than 4,500 reactors with 1GWe capacity to replace 1,511 GWe of coal and 3,011 GWe of gas and oil, plus 146 reactors to replace the current ones and another 72 to cover new construction already planned. In total, 4,740 new 1GWe reactors would have to be built and put in operation every two days for the next 25 years. An optimistic estimate of construction times (five years) would mean having 950 teams of technical specialists, workers and machinery simultaneously working full time. This is hard to imagine, despite talk of standardising designs. In the previous period of nuclear construction (1963-88) only 423 reactors were built, at a rate of 17 per year.
Leaving aside the logistic (and financial) difficulties involved in a nuclear construction programme on this scale, let us calculate the amount of fuel necessary to feed that many reactors, since this could be another limiting factor. Without doubt, we would see the construction mainly of third-generation thermal neutron reactors with a semi-open fuel cycle (fed with MOX uranium enriched with some plutonium). At best, the fourth generation of fast neutron reactors with closed fuel cycles (which are expected to multiply output by 60, due to the massive use of plutonium) is not scheduled to be operative before 2030. Therefore, uranium will continue to be the main nuclear fuel in the coming decades. The thorium option does not seem possible in this time frame either, especially when, as Anne Lauvergeon, president of the Areva group said at the latest World Nuclear Association Annual Symposium,[6] “to stimulate a new era of growth, evolved reactor designs are more attractive than revolutionary ones [and]… the new designs should not be based on over-innovative technical solutions” because “investors have a relatively high perception of risk when it comes to new nuclear projects”. Everything indicates, therefore, that traditional plants would be used with light water reactors, though perhaps with certain enhancements in terms of output and security.
A simple calculation suffices to show how an extension of the French model would collide with a scarcity of uranium. This is old news, given the serious doubts that already exist regarding the availability of uranium even to feed a few more reactors than now exist. In 2004, 365 GWe of nuclear capacity consumed about 67 kt of uranium (approximately 180 tons of uranium per GWe per year), of which 36 kt came from currently operating mines, while the rest came from recycled nuclear weapons and other secondary sources (that is, from prior production). Supply forecasts for the reactors currently in operation (plus foreseeable growth) put uranium mining production at 50 kt per year in 2015, with a significant shortfall developing in 2010, by which time Russia’s nuclear weapons will have been dismantled and their uranium will have been consumed, as the following table shows:
Table 2. World uranium supply and demand (tons of uranium)
| 2004 | 2005 | 2006 | 2007 | 2010 | 2015 | |
| Demand | 66,658 | 68,400 | 69,600 | 70,800 | 74,800 | 79,400 |
| Supply | ||||||
| Military uranium | 10,700 | 10,600 | 10,700 | 11,100 | 12,400 | |
| DOE stocks | 385 | 1,192 | 1192 | 1,192 | 2,154 | 2,346 |
| Commercial stocks | 7,876 | 7,000 | 7,000 | |||
| Russian stocks | 2,900 | 3,500 | 3,800 | 3,900 | ||
| MOX | 2,500 | 2,500 | 2,600 | 2,800 | 3,000 | 3,600 |
| Reprocessed uranium | 1,500 | 1,500 | 1,700 | 1,700 | 2,000 | 2,000 |
| Reuse of tailings | 4,250 | 3,650 | 3,300 | 3,000 | 1,500 | |
| Mining production | 36,263 | 36,575 | 36,094 | 42,286 | 48,014 | 50,321 |
| Total | 66,374 | 66,517 | 66,386 | 65,978 | 69,068 | 58,267 |
| Shortfall | -284 | -1,883 | -3,214 | -4,822 | -5,732 | -21,133 |
Source: Moukhtar Dzhakishev, World Nuclear Association Annual Symposium, London, 2004.[7]
If we assume linear growth from the current 365 GWe to 4,959 GWe in 2030, uranium demand would be around 400 kt in 2015 and 700 kt in 2030. This means multiplying by eight today’s estimates of production capacity in 2015, and multiplying by fifteen for 2030. Considering that annual production has never been above 68 kt, that today’s production capacity stands at near 45 kt, and that known mines have limited potential, the only possible way to satisfy this increase in demand is to discover and exploit significant new reserves. To help understand the magnitude of the gap, Figure 2 shows how past production pales before the needs of this scenario.
Figure 2. Uranium production versus demand, 1947-2030
Source: historical data and Jeff Combs, World Nuclear Association Annual Symposium, London, 2004.[8]
Also, according to Tim Gitzel, vice-president of Areva (see Challenging or Easy? Natural Uranium Availability to Fuel a Nuclear Renaissance),[9] there is an “uncompressible” 20-year lag between the first signs of scarcity in the market and the development of sufficient supply. If so, this would rule out any significant increase in the number of nuclear reactors before 2030.
Let us suppose, however, for argument’s sake, that it were possible to achieve a production capacity of 700 kt/year by 2030. In the context of this analysis, two questions are raised: first, the CO2 emissions that would be generated in this phase of the nuclear cycle. Given the amount of uranium necessary, it would almost certainly be necessary to make use of hard rock deposits and low concentrations (ore < 0.02%) or phosphate mines (0.003% ore). Producing 700 kt of uranium from minerals with 0.03% of ore, for example, means extracting and processing 2.3 billion tons of mineral a year (about 50% of all metallic minerals now extracted from the Earth). Presumably, this would be done with fossil fuels, calling into question one of the main arguments used in favour of the nuclear option. In fact, some studies[10] suggest that mines with less than 0.02% ore produce more CO2 in the mining and enrichment of uranium than is later avoided, compared with an equivalent amount of energy generated with natural gas.
Secondly, there is the issue of the total amount of uranium that can be produced at an acceptable cost in terms of money and energy. Supposing that the useful life of the new EPR reactors is about 60 years, a total volume of 45 million tons of uranium would be necessary to feed the 4,959 reactors during their entire life cycle (even with a gradually increasing use of MOX fuel). Natural sources of uranium can be classified according to cost and degree of certainty, as follows:
Figure 3. Natural uranium sources
Source: Red Book, NEA/OCDE-IAEA, 2003.[11]
It is known, therefore, that 3.2 million tons have been located and calibrated (RAR = Reasonably Assured Resources) at various prices levels, while another 1.4 million tons may be inferred from deposits with similar characteristics. To these 4.6 million tons, another 2.3 million “hypothetical” tons could be added (based on indirect evidence), along with 7.5 million “speculative” tons (believed likely to exist in unexplored areas with promising geology). This makes a total of about 14.4 million tons of conventional resources and perhaps another 15-25 million tons in non-conventional deposits (phosphates, etc.), the cost of which is completely unknown. Not even a combination of all the resources catalogued by the NEA/OECD –conventional, non-conventional, located, hypothetical and speculative– would cover the consumption forecast for the entire life cycle of the reactors necessary to satisfy world electrical demand in the coming decades. There are some who think that this problem could be solved with the immense quantity of uranium diluted in the sea, but the concentration of marine uranium is 3 mg per ton of seawater; a simple calculation indicates that the energy needed to pump and extract this uranium would be greater than the amount of energy that would later be obtained from conventional reactors. Other extraction techniques based on absorbents could be viable in the long term, with costs estimated at between US$200s/kgU and US$1,000/kgU, compared with US$34/kgU today. However, all this is little more than speculation.
Therefore, it does not appear viable to feed such a large number of reactors during their entire useful life; to do so in these conditions would be reckless and the financial world would be very unlikely to accept it. This is the main reason why France, Japan and Russia decided early in the nuclear era to develop fast neutron reactors that generate and consume plutonium from non-fissile uranium, such as the French Superphoenix and the Japanese Monju –both recently shut down due to poor functioning–. It is also the main reason why the fourth generation of reactors has been proposed– reactors which, as has been pointed out, are not expected to materialize before 2030.
Figure 4. Uranium utilisation, 2000-2100E
Source: A Technology Roadmap for Generation IV Nuclear Energy Systems, DOE, 2002.[12]
As Figure 4 shows, the fourth generation project requires the introduction of fast neutron reactors around 2030 to avoid a foreseeable scarcity of uranium resources resulting from simply maintaining today’s reactors and building a small number of new ones. We can therefore rule out this technology for the next 25 years.
However, and again for argument’s sake, let us suppose that the necessary uranium was somehow obtained, either from mines or from the sea, and that it were possible to extract it without producing more CO2 than would be saved by using the uranium as fuel –and this is supposing a great deal–. The problem of radioactive waste remains. The 4,959 reactors operating in 2030 would produce 86 kt of irradiated fuel each year; in 25 years, more than a million tons of highly active radioactive waste would have been accumulated for storage. Considering that the projected capacity of the Yucca Mountain geologic storage site is 70 kt, it would be necessary to construct such a site nearly every year, or else several dozen sites a year of the kind planned in Finland (the 4-kt Eurajoki site), distributed around the world. This is clearly impossible. YuccaMountain has been under study for fifteen years and there is still not a single geological storage site in operation there. It is not scheduled to start operations until 2012 and, when it does, it will not have the capacity to store the waste already generated in the United States alone by that date. The waste problem is often addressed by the argument that by applying separation and transmutation techniques, the volume of wastes could be reduced and their half life could be shortened; however, once again, this is theoretical futurism which, in practice, would only partially solve the problem –undoubtedly at the cost of creating new problems– and even if it were possible, it would require much long times frames and extremely high costs. For a detailed analysis of the problems facing transmutation as a strategy for waste management, see The Nuclear Alchemy Gamble, by Hisham Zerriffi and Annie Makhijani.[13]
To top it off there is the problem of nuclear proliferation and security against possible terrorist attacks. Since, as we have mentioned, more than two thirds of the electricity to be generated is needed in non-nuclear countries –some with unstable political systems or in some form of confrontation with nuclear powers– there is a clear risk of contributing to widespread nuclear proliferation. There is broad consensus that the construction of nuclear reactors should not be extended unless this risk (arising from the commercial trade in nuclear fuel), can be reduced to acceptable levels; but this would mean restricting enrichment and reprocessing facilities to only a few countries. The Non-proliferation Treaty would have to be modified –not an easy task– and furthermore, a majority of states would have to accept dependence on a handful of others to supply their fuel, which would very likely put unacceptable limits on their sovereignty. The attitude of Iran and North Korea is an example of what could be expected of other countries.
Regarding security in the face of terrorism, there are numerous flanks to protect, starting with the pools where spent fuel is now kept and the temporary repositories where, after the pools have been filled, waste is sent to await its final destination. According to a recent report by the US National Academy of Sciences,[14] an attack on these pools could produce radioactive emanations similar to those caused by the Chernobyl accident. It has also recently come to light that a classified French government report[15] warns that the new EPR reactors are not equipped to withstand a 9/11-style air attack, which has paralyzed the work of the commission in charge of public debate on these reactors. To all this we can add that these new nuclear facilities would be scattered all around the world, requiring constant movement of radioactive materials that would be a very attractive target for spectacular terrorist actions. Clearly, security against terrorism is an added obstacle to a massive project of the kind we are considering.
Therefore, without even beginning to make economic assessments or considering the issue of competitiveness, it does not appear that extending the French model to other countries is a viable option, as Juan Velarde Fuertes and others suggest.[16] But even if it were, would it be advisable from the perspective of operational security? Not counting the reactors in the former Soviet Union, experience suggests that the rate of accidents with damage to the reactor core has been 10-4 accidents/reactor-year. The new generation of reactors may be able to reduce this to 10-5 accidents/reactor-year, as an MIT study[17] suggests. At that rate, if 4,959 reactors operate for sixty years, a serious accident would occur approximately every twenty years. Such a risk would surely be unacceptable to society, nor would it likely be acceptable to the financial community, since another accident of the Three Mile Island or Chernobyl type would probably paralyze the entire programme.
Scaling down this scenario, the French Institute of Nuclear Sciences (INSTN) has drafted a nuclear-intensive energy scenario for 2030 which projects 85% of energy needs being covered by nuclear energy in OECD countries, 50% in the former East Bloc countries and 30% in the rest of the world.[18] This scenario implies the construction of 3,000 GWe of nuclear power in twenty years, which means approximately one reactor every three days and uranium consumption which, the authors say, would exhaust estimated reserves during the useful life of these plants. This construction programme would have to stop in 2025 due to lack of uranium and it could continue only if by then fast neutron reactors had been developed that were capable of using plutonium and thorium as fuel –the goal of the Fourth Generation project–. This scenario, however, is subject to the same considerations as those indicated above, corroborating the difficulties that would face any massive nuclear programme.
Conclusions: In summary, we can conclude that increasing nuclear capacity to replace the fossil fuels now used to generate electricity is not viable with the nuclear technology foreseeable in the coming decades. Nuclear energy is not even viable to produce the electricity we consume directly, much less extending its use to hydrogen production as a possible replacement for oil. A third of the raw energy is used for transportation and comes almost exclusively from oil derivatives. Using nuclear energy to substitute even a part of this oil for hydrogen (produced either by electrolysis or through the thermo-chemical decomposition of water) is an absurd proposition, since it would mean replacing fossil fuels with electric energy for transportation, while continuing to use fossil fuels to generate the electricity we consume directly. Producing hydrogen for transportation by nuclear means would make sense only when all the oil and natural gas used to generate electricity has been replaced. Those who advocate nuclear energy and point to the hydrogen mirage would do well to go over their figures and realize that before reaching that hypothetical and distant oasis, we would have to cross a desert of difficulties in the present and in the more immediate future.
There are no grounds to claim that nuclear energy is a real alternative to the consumption of fossil fuels and that it is the solution to the problem of climate change, at least in the next few decades. Perhaps this is why those who propose a nuclear renaissance rarely give specific figures and why their allusions to these two very real problems must be interpreted simply as a smokescreen to save an industry in trouble, rather than a serious attempt to deal with the energy/climate change dilemma.
In the second part of this analysis, we study two proposals –one by MIT and other by the World Nuclear Association– which do in fact quantify a possible programme for nuclear construction. We assess the contribution that these proposals could make to reducing emissions and fossil fuel consumption. We also consider their economic and political implications in terms of their impact on the current trend toward gradually liberalizing the energy sector, with particular attention to how this could affect Spain.
Marcel Coderch Collell
PhD, Telecommunications Engineer at MIT, Secretary of the Asociación para el Estudio de los Recursos Energéticos (AEREN)
[1] http://www.iea.org/textbase/publications/free_new_Desc.asp?PUBS_ID=1266
[2] http://www.iea.org/textbase/publications/free_new_Desc.asp?PUBS_ID=1540
[3] http://www.eage.nl/conferences/index2.phtml?confid=17
[4] The author’s opinions on this subject can be found at:
El fin del petróleo barato, at http://www.fp-es.org/oct_nov_2004/story_5_19.asp
[5]http://www.eia.doe.gov/oiaf/ieo/
[6]http://www.world-nuclear.org/sym/2005/pdf/Lauvergeon.pdf
[7]http://world-nuclear.org/sym/2004/dzhakishev.htm
[8]http://world-nuclear.org/sym/2004/combs.htm
[9]http://world-nuclear.org/sym/2005/pdf/Gitzel.pdf
[10] For example, http://world-nuclear.org/sym/2005/pdf/Gitzel.pdf
[11]Uranium 2003: Resources, Production and Demand, OECD Publishing, 2004.
[12]http://gif.inel.gov/roadmap/pdfs/gen_iv_roadmap.pdf
[13]http://www.ieer.org/reports/transm/report.pdf
[14]http://darwin.nap.edu/execsumm_pdf/11263
[15]EPR : Document “Confidentiel-défense”
[16] “El ejemplo energético francés”, ABC, 8/IX/2004.
[17] http://web.mit.edu/nuclearpower/
[18] H. Nifenecker, D. Heuer, J.M. Loiseaux, O. Meplan, A. Nuttin, S. David and J.M. Martin, “Scenarios with an Intensive Contribution of Nuclear Energy to the World Energy Supply”, International Journal of Global Energy Issues (IJGEI), vol. 19, no. 1, 2003.