Lighting the Way: Toward a Sustainable Energy Future

Search this Publication

  • 3.2 Nuclear power

    Nuclear power supplies approximately 16 percent of today’s global electricity demand and, along with hydropower, accounts for the largest share of power generation from non-carbon energy sources. More than two dozen reactors are now under construction or will be refurbished over the next few years in Canada, China, several European Union countries, India, Iran, Pakistan, Russia, and South Africa. The world’s existing base of nuclear capacity includes some 443 reactor units with a combined capacity of about 365 gigawatts (Figure 3.4). The great majority of these units (nearly 80 percent) are more than 15 years old.

    While total nuclear electricity output is likely to grow modestly within this decade, reflecting the addition of new capacity now planned or under construction, the overall nuclear contribution is expected to plateau thereafter and even decline slightly over the next two decades as more plants retire than are added worldwide and as growth in nuclear plant output falls behind growth in overall electricity demand. As a result, the most recent IEA reference case forecast (Figure 3.5) indicates that nuclear power’s share of global electricity production will fall to just 12 percent by 2030. The IEA estimate of total nuclear output for 2030 is just under 3,000 terawatt-hours, only slightly more than the 2,500 terawatt-hours produced by the industry in 2002. These projections are roughly consistent with projections released by the International Atomic Energy Agency (IAEA) in 2004 that show the nuclear contribution falling to 13–14 percent of global electricity production in 2030 under high- and low-growth assumptions.












    Figure 3.4 Existing and planned/proposed nuclear reactors in the world

    Source: International Nuclear Safety Center, Argonne National Laboratory

    Current expectations of flat or declining nuclear output reflect an assumption that high upfront capital cost32 and other obstacles will continue to disadvantage nuclear power relative to other options for new electric-generating capacity, particularly compared to conventional, pulverized-coal power plants.

    Current interest in reversing this trend and in supporting an expanded role for nuclear power is driven largely by climate change considerations and by concern that the other non-carbon options alone—including energy efficiency, renewable energy, and advanced fossil technologies with carbon sequestration—will not be adequate to reconcile burgeoning global energy demands (especially growing demand for electricity) with the need for greenhouse gas mitigation. On the one hand, nuclear technology offers important advantages: it can provide a reliable, large-scale source of baseload electric-generating capacity;33 it does not produce emissions of greenhouse gases or conventional air pollutants; and supplies of nuclear fuel, in the form of uranium ore, are relatively abundant worldwide.34 In addition,















    Figure 3.5 Projected world incremental electricity generation by fuel type

    Note: 1 terawatt-hour (TWh) equals 3.6 petajoules.

    Source: IEA, 2006

    the potential exists to use nuclear power for high-temperature hydrogen production, which would enable the technology to serve a wider array of energy needs besides electricity production. Plans for ‘hybrid’ reactors that would co-produce hydrogen and electricity have been proposed.

    Other factors that are likely to continue motivating some governments to support nuclear power include energy-security concerns, especially in light of recent volatility in world oil markets and the perception that development of an indigenous nuclear capability offers a route to technological advancement while conferring a certain ‘elite’ status among the world’s industrialized powers. Finally, efforts to build a domestic nuclear industry can provide useful ambiguity for governments that wish to leave open the possibility of developing nuclear weapons. Associated equipment (like hot laboratories), operator training, and experience with health and safety issues are some obvious examples of the potential carry-over from nuclear power technology to nuclear weapons capability that is latent in any civilian nuclear power program.

    But nuclear power also suffers from several difficult and well-known problems that are likely to continue to constrain future investments in this technology. Chief hurdles for primary investors include high upfront capital cost, siting and licensing difficulties, public opposition, and uncertainties regarding future liabilities for waste disposal and plant decommissioning.  In addition to—and inextricably intertwined with—these issues, many experts agree that concerns about reactor safety, waste disposal, and nuclear weapons proliferation must be resolved if nuclear technology is to play a prominent role in the transition to a sustainable global energy mix. A further obstacle in many parts of the world relates to the need for significant amounts of capital and considerable institutional capacity and technical expertise to successfully build and safely operate nuclear power plants. Some of these issues could be resolved by the successful development of nuclear fusion (as opposed to fission) technology, but this is a long-term prospect. Even if nuclear fusion ultimately proves feasible, the technology is unlikely to be available until mid-century or later.

    In sum, nuclear power plants are much more complicated than fossil-fuel power plants, and the consequences of accidents are far greater. In fact, potential dependency on other countries for technological expertise or nuclear fuel may discourage some governments from developing nuclear capacity, even as a desire for technology status or energy security may motivate others in the opposite direction. Brazil’s decision in the 1970s not to pursue a relationship with Germany that would have led to a major expansion of Brazil’s nuclear power capability was driven by these types of considerations.

    Current, near-term plans to expand nuclear-generating capacity are largely centered in Asia with India, China, and Japan leading the way in terms of numbers of new plants proposed or under construction at present. Increasingly, these countries and others are interested in developing and building their own reactor designs. Figure 3.6 shows the regional
















    Figure 3.6 Regional distribution of global nuclear capacity in the IAEA’s high projection

    Source: IAEA, 2004; McDonald, 2004.

    breakdown of new nuclear capacity in the 2004 IAEA high-growth projections for 2030. According to this figure, the largest increase in nuclear capacity (in terms of net gigawatts added) will occur in the Far East, while the strongest growth in percentage terms will occur in the Middle East and South Asia. Net capacity also increases, albeit less dramatically, in Eastern and Western Europe, but stays essentially flat in North America.

    Most of the new plants expected to come on line in the next few years incorporate substantial modifications and improvements on existing reactor designs, including safety features that simplify cooling requirements in the event of an accident. These designs are therefore expected (though not yet demonstrated) to provide more reliable safety performance at lower overall cost.35 Efforts are already underway to develop a third generation of nuclear reactor designs that would be ‘passively safe,’ whereby the chance of a core meltdown would be (nearly) impossible, even in the event of a total loss of operation of the reactor control systems. The fourth-generation reactors could, in addition to incorporating passive safety features, achieve further improvements in cost and performance while also reducing waste disposal requirements by minimizing fuel throughput and/or recycling spent fuel.


    Box 3.1 Four generations of nuclear reactors

    The first nuclear power plants to be developed, many small, are now called Generation I (GEN I) reactors. Perhaps the only GEN I reactors still in operation are six small (under 250 megawattselectricity) gas-cooled plants in the United Kingdom. All others have been shut down.

    Most reactors operating today are Generation II reactors. Designed in the late 1960s and 1970s, they are of two main types, either pressurized water reactor (PWR) or boiling water reactor (BWR). GEN II reactors have achieved very high operational reliability, mainly through continuous improvement of their operations.

    Generation III reactors were designed in the 1990s, and geared to lower costs and standardized designs. They have been built in the last few years in France and Japan. More recent designs are labeled GEN III+ reactors and are likely to be constructed in the coming years. Typical examples are the advanced boiling water reactor (ABWR) in Japan, the new PWR in Korea,the evolutionary power reactor (EPR) in France, and the economic simplified boiling water reactor (ESBWR) and the AP-1000 (advanced passive) in the United States .

    The GEN III+ light water reactor (LWR) are based on proven technology but with significant improvements and, in the case of the AP-1000 and ESBWR, with passive emergency cooling systems to replace the conventional powerdriven systems. The 2004 World Energy Assessment specifically mentions the pebble-bed modular reactor (PBMR) as a design concept that is being revisited because of the potential for a high degree of inherent safety and the opportunity to operate on a proliferation-resistant denatured uranium/thorium fuel cycle. The PBMR is also considered a GEN III+ reactor. GEN III+ systems probably will be the type used in the next expansion of nuclear power (UNDP, UNDESA, and WEC, 2004).

    None of the Generation IV ‘advanced reactors’ have been built and none are close to being under construction. GEN IV is widely recognized as an R&D program for reactors with advanced features well beyond the GEN III+ LWR. GEN IV reactors are being prepared for the future, starting in 2035 to 2040. Whereas previous reactor types progressed in an evolutionary manner, GEN IV reactor designs attempt to significantly shift the nature of nuclear energy, either by incorporating high-temperature, high-efficiency concepts, or by proposing solutions that significantly increase the sustainability of nuclear energy (reduced wastes; increased usage of natural resources).

    Six reactor types are being studied by a group of ten countries: the very high temperature reactor, which uses gas cooling, can reach very high thermodynamic efficiency and might be able to support production of hydrogen; the supercritical water reactor, which also allows for higher efficiency and reduces the production of waste; three fast neutron reactors, cooled either by gas (gas fast reactor), lead (lead fast reactor), or sodium (sodium fast reactor), which make use of closed fuel cycles; and the molten salt reactor. The very high temperature and gas fast reactors can both use pebble-type fuel.

    Future nuclear systems, such as those that are studied in the GEN IV program and the Advanced Fuel Cycle initiative are all aimed at making nuclear energy more sustainable, either by increasing system efficiency or by using closed fuel cycles where nuclear waste is either partially or totally recycled. Another objective for these systems is to reduce both capital and operational costs. Significant scientific and technical challenges must be resolved before these systems are ready for deployment:

    – high temperature high fluence materials (i.e., materials not crippled by ultra- high neutron fluxes);
    – fuels that can contain high quantities of minor Actinides need to be demonstrated;
    – novel technologies for transporting heat and generating electricity with smaller footprints than the current steam cycles;
    – separation technologies that offer high proliferation resistance and produce minimal wastes;
    – more compact designs that reduce capital costs.

    To achieve these ambitious objectives, a three-pronged research strategy is being implemented in the United States:

    (1)The role of the basic sciences is being enhanced. Current empirical research tools need to be phased out and replaced by modern techniques.
    (2) The role of simulation and modeling will become central, when current generation software—largely developed in the 1980s— is replaced by high performance tools. One can expect that certain key difficulties, for example the development of advanced fuels, can be solved more efficiently once these tools are in place
    (3) The design process itself will be simplified and streamlined.


    In 2002, ten nations and the European Union formed the Generation IV International Forum (GIF) to promote international collaboration in developing a fourth generation of nuclear plants.36 After more than two years of study, each participating nation agreed to take the lead in exploring at least one of several different reactor types for potential deployment by 2030. The reactor types identified by the GIF as most promising include the very high temperature gas reactor, the super-critical water reactor, the lead-cooled fast reactor, the sodium-cooled fast reactor, the gas-cooled fast reactor, and the molten salt reactor. In addition, other potential reactor designs have been studied or developed in recent years, including designs for smaller, modular and even transportable types of reactors, as well as designs that are geared toward the production of hydrogen

    At this point, none of the proposed fourth-generation reactor designs have been built, though a number of countries are pursuing active research and development efforts and have adopted policies aimed at facilitating the construction of new plants. Even while many of the new designs offer important advantages over older generations of reactors—at least on paper—the industry’s longer-term outlook remains uncertain. The remainder of this section provides further detail about the specific challenges that now confront nuclear power and reviews current prospects for addressing these challenges with further improvements in reactor design and nuclear technology.

    Challenges facing nuclear power
    Nuclear fusion remains a distant alternative to fission technologies at present. In nuclear fusion, energy is produced by the fusion of deuterium and tritium, two isotopes of hydrogen, to form helium and a neutron. Effectively unlimited quantities of the primary fuels, deuterium and lithium (from which tritium is produced), are easily available. Due to low fuel inventory, a runaway reaction or meltdown of a fusion system is not possible. Radioactive waste from fusion decays in 100 years to activity levels similar to that from coal. The proliferation risk from fusion is minimal since any fertile materials would be easily detectable.

    Investigations of possible commercial development of fusion energy include inertial fusion and various forms of magetic confinement of high-temperature plasma. Current research is focused on magnetic confinement in toroidal (doughnut-shaped) geometries and on laser-induced inertial confinement. Laboratory experiments in tokomaks—machines that produce a toroidal magnetic field for confining a plasma—have produced 10 megawatts of heat from fusion for about one second. The ITER project (ITER means ‘the way’ in Latin), a collaboration of China, Europe, India, Japan, Russia, South Korea and the United States, is planned to produce 500 megawatts of fusion heat for over 400 seconds. In parallel with ITER, research is planned to target higher power and continuous operation and to develop advanced materials and components that can withstand high neutron fluxes. Some ITER partners anticipate demonstration fusion power plants about 2035 and commercialization starting about 2050.


    While operating costs for many existing nuclear power plants are quite low, the current upfront capital cost of constructing a new plant is higher than the cost of conventional new fossil fuel-fired electricity-generating technologies.37 Cost reductions could help to improve nuclear energy’s competitiveness in terms of real, levelized cost in cents-per-kilowatt-hour, relative to other options (Table 3.3).38 Projections of future cost for nuclear power are, of course, highly uncertain, especially in the case of advanced reactor designs that have yet to be built or operated anywhere in the world. In some countries, moreover, cost uncertainty is likely to be compounded by the potential for delays and difficulties in siting, permitting, and construction. For all of these reasons, private financial markets in many parts of the world will tend to assign a substantial risk premium to new nuclear investments for some time to come.











    Note: Gas costs reflect real, levelized acquisition costs per thousand cubic feet (MCF) over the economic life of the project. CCGT refers to combined cycle gas turbine; kWeh refers to kilowatt-electricity hour. Figures use 2002 US$.

    Source: MIT, 2005.

    Obviously, a number of developments could change the relative cost picture for nuclear power. Further technology improvements, greater public acceptance and regulatory certainty, and progress in addressing the waste disposal issue would produce lower cost estimates and, perhaps more importantly, alter current perceptions of investment risk.39 Successful development of simplified, standardized reactor designs that would expedite licensing and construction, in particular, could greatly improve the industry’s prospects. Nuclear power would also be more competitive in the presence of a binding carbon constraint and/or if fossil fuel prices rise. Whether a carbon constraint would by itself produce a significant shift toward nuclear power would, of course, depend on the magnitude of the price signals and on the cost of other non- or low-carbon alternatives, including renewable energy sources, coal with carbon capture and sequestration, and highly efficient natural gas technologies. Without the presence of a carbon cap or tax on carbon and/or active government intervention in the form of risk-sharing and/or financial subsidies, most experts conclude that the private sector is unlikely to make substantial near-term investments in nuclear technology and other non-or low-carbon alternatives —especially in the context of increasingly competitive and deregulated energy markets.

    An IEA analysis of nuclear economics shows that various OECD governments already subsidize the nuclear industry by providing fuel-supply services, waste disposal, fuel reprocessing, and R&D funding. Many governments also limit the liability of plant owners in the event of an accident and help with remediation. A recent case in point is the U.S. Energy Policy Act of 2005, which contains substantial subsidies and tax incentives for a new generation of nuclear power plants. Whether these incentives will prove sufficient to spur a new round of nuclear power plant construction in the United States is not yet known; in the meantime, immediate prospects for further expansion of nuclear energy capacity are likely to remain concentrated in the rapidly growing economies of Asia, notably in China and India.


    Accidents at Three Mile Island in 1979 and Chernobyl in 1986, as well as accidents at fuel-cycle facilities in Japan, Russia, and the United States have had a long-lasting effect on public perceptions of nuclear power and illustrate some of the safety, environmental, and health risks inherent in the use of this technology. While a completely risk-free nuclear plant design, like virtually all human endeavors, is highly unlikely, the role of nuclear energy has to be assessed in a more complete risk-benefit analysis that weighs all factors, including the environmental impacts of different energy options, their energy security risks and benefits, and the likelihood of future technology improvements.

    A related challenge is training the skilled personnel needed to construct and safely operate nuclear facilities, including not only existing light water reactors but also safer GEN III reactors. The challenge of developing adequate skills and expertise is more significant in the case of GEN IV reactors, which are (a) very different from GEN III reactors,40 (b) present more difficult safety and proliferation issues, and (c) require considerable expertise to design, construct, and operate.

    In recent years, of course, the threat of terrorism has added a new and potentially more difficult dimension to long-standing concerns about the safe and secure operation of nuclear facilities and the transport of nuclear materials. While the safety record of the light-water reactors that dominate the world’s existing nuclear power base has generally been very good, Chernobyl remains ‘a powerful symbol of how serious and long-lived the consequences of a nuclear accident can be,’ however low the probability of such accidents might be (Porritt, 2006). In response to potential terrorist threats, many countries have implemented additional security measures at existing nuclear power plants; going forward, innovative reactor designs—possibly including facilities that can be built underground or have otherwise been reinforced and equipped with passive safety features to withstand outside attacks and internal sabotage—may help to alleviate public concerns about the particular vulnerabilities associated with nuclear facilities.

    One of the selling points of a new generation of pebble-bed reactors is that they can be built underground.

    Disposing of high-level radioactive spent fuel for the millennia-scale period of time that nuclear waste could present a risk to public safety and human health is another problem that has long plagued the industry and that has yet to be fully resolved in any country with an active commercial nuclear energy program. While long-term disposal in stable geologic repositories is technically feasible, no country has yet completed and begun operating such a repository. (At present, Finland is closest to implementing this solution). Without a consensus on long-term waste storage, various interim strategies have emerged. These include storing spent fuel temporarily at power plant sites, for example using the dry cask method; or, in some countries, reprocessing or recycling the spent fuel to remove the fission products and separate the uranium and plutonium for re-use in reactor fuel. Reprocessing reduces the quantity of waste by more than an order of magnitude and has the potential of reducing the storage time by several orders of magnitude; but even after reprocessing, hundreds of years of safe storage are required. Reprocessing also raises significant proliferation concerns since it generates quantities of plutonium—the essential ingredient in nuclear weapons—that must be safeguarded to prevent theft or diversion for weapons-related purposes.

    In fact, proliferation risks are a substantial concern for all current ‘closed fuel-cycle’ reactor designs, especially for the so-called ‘breeder’ reactor, which requires reprocessing of spent fuel to separate and recycle weapons-usable plutonium. An interdisciplinary study of nuclear power by MIT (2003), which analyzed the waste management implications of both once-through and closed fuel cycles, concluded that no ‘convincing case can be made on the basis of waste management considerations alone that the benefits of partitioning and transmutation will outweigh the attendant safety, environmental, and security risks and economic costs.’ Other experts disagree and are more optimistic that the security, safety, and environmental concerns associated with closed fuel cycles are technically resolvable. They point out that fast neutron reactors would extend uranium supplies by 100-fold and allow for the use of thorium, while reducing the quantity of waste to be handled. Based on these advantages, they argue that concerted research efforts should be undertaken to see whether such reactors can be part of this century’s energy solutions.

    Given that uranium is relatively abundant and inexpensive at present and given that the waste reduction benefits of spent fuel reprocessing do not appear to outweigh the downsides in terms of proliferation risks, once-through fuel cycles are likely to remain the safer option for at least the next few decades although research that may lead to technical solutions could change that. The latest reactor designs tend to require less fuel per kilowatt-hour generated; a higher ‘burn-up rate’ in turn reduces the quantity of waste left to be managed at the end of the fuel cycle. This is true of newer pebble-bed designs, though it is also the case that the fuel pellets used in these designs require much higher uranium enrichment.

    Meanwhile, seemingly irreducible political stresses continue to inhibit solutions to the problem of nuclear waste disposal all over the world. Half a century ago, the nuclear industry imposed on itself a standard of waste management that some experts believe has turned out to be unrealizable. The industry agreed that it would manage nuclear wastes in such a way that there would be no discernible impact on later generations for a period that was often in the range of 10,000 years. With the understanding of geology gained since, this task might have become easier. In fact it has become harder. There seems to be little prospect that the original objective can be met within this generation, though perhaps it can be met one or two generations from now.

    With this realization, a consensus is beginning to emerge among experts that the objective of waste storage should shift from irretrievable storage to retrievable storage. In other words, wastes would be stored with the expectation that they will require further handling in a few decades. In the United States and elsewhere attention has recently focused on ‘dry-cask’ storage technology that could keep nuclear wastes thermally secure for time periods on the order of a half-century. A shift in nuclear waste-management objectives, while increasingly under discussion in expert circles, has not however been widely proposed to the general public and would require changes in the legal framework governing waste management in the United States. The latter could present a major near-term hurdle in the United States and elsewhere.

    Other countries, meanwhile, have continued to focus on spent-fuel reprocessing and long-term geological storage as primary strategies for waste management. In 2006, France, for example, adopted legislation that (a)formally declares deep geological disposal as the ‘reference solution’ for high-level and long-lived radioactive wastes, (b) sets 2015 as the target date for licensing a repository, and (c) sets 2025 as the target date for opening a long-term repository.41 Meanwhile, some experts have suggested that if countries could reach consensus on establishing international facilities to provide spent-fuel reprocessing and uranium enrichment services in a highly secure and transparent environment, this option could be very helpful in addressing both proliferation and waste management concerns. Until this or other long-term solutions can be found, however, the waste issue is likely to continue to present a significant and perhaps intractable obstacle to the significant expansion of commercial nuclear power capacity worldwide.

    Nuclear proliferation and public acceptance
    The development and use of nuclear technology for commercial energy production has long generated concern that associated materials or expertise could be diverted to non-peaceful purposes. To date, no operating civilian nuclear program has been directly linked to the development of nuclear weapons, but the risk exists that commercial nuclear energy programs could be used to as ‘cover’ for illicit weapons-related activity or as a source (voluntarily or involuntarily) for the highly enriched uranium or plutonium needed to construct nuclear weapons. Both in India and North Korea, reactors nominally intended for civilian research were used to produce plutonium for weapons. Proliferation concerns apply most strongly to the uranium enrichment and spent-fuel reprocessing elements of a civilian nuclear energy program. As the American Physical Society has pointed out, ‘nuclear reactors themselves are not the primary proliferation risk; the principal concern is that countries with the intent to proliferate can covertly use the associated enrichment or reprocessing plants to produce the essential material for a nuclear explosive’ (APS, 2005: i).

    The existing international regime for managing proliferation risks is widely viewed as inadequate and would be further stretched by a significant expansion of nuclear power to many more countries with widely varying security circumstances. Here again, it matters which technology is being deployed: the risks presented by GEN III reactors are very different and likely to be more manageable from those that would be presented by the international deployment of fast neutron systems. Given the devastating impact even a single nuclear weapon linked to a civilian nuclear energy program could have, current international safeguards will clearly need to be strengthened. Efforts to develop proliferation-resistant technologies, especially for fuel enrichment and reprocessing, also merit high priority. Increased international collaboration is needed to explore options for addressing enrichment and reprocessing needs in ways that minimize public safety and proliferation risks. In particular, it has been suggested that stronger multi-lateral arrangements—including facilities that would enrich and reprocess fuel for use by multiple countries under multinational supervision, perhaps in combination with international supply guarantees—could help to address proliferation concerns.

    In some countries, public acceptance is likely to continue to present a significant challenge for nuclear power, though locating future capacity additions at existing plants may help to alleviate siting difficulties to a significant degree. Public perceptions are likely to change over time, of course, and may become significantly more accepting of nuclear energy as concern over climate change grows and as countries and communities become familiar with nuclear energy systems. However, even if the climate of opinion around nuclear energy already shows signs of shifting, it remains the case that the public is likely to be extremely unforgiving of any accident or attack involving civilian nuclear energy systems. A single incident anywhere would cast a pall over nuclear power everywhere. A substantial increase in both the number of plants operating worldwide and the amount of fuel being transported and handled for enrichment, reprocessing, or waste disposal inevitably heightens the risk that something, someday, will go wrong, even if the probability of any single event is extremely low. As a result, some experts have estimated that a further order-of-magnitude increase in reactor safety, along with substantial international progress to address current proliferation concerns, will be required to maintain public acceptance in the face of a greatly expanded worldwide nuclear energy program. In the meantime, it seems clear that the fundamental challenges for nuclear power are as much—and perhaps more—political and social as they are technological or scientific.

    In summary: Nuclear power
    Based on the foregoing discussion, no certain conclusion regarding the future role of nuclear energy emerges, except that a global renaissance of commercial nuclear power is unlikely to materialize over the next few decades without substantial support from governments; effective efforts to promote international collaboration (especially to address safety, waste, and proliferation concerns); changes in public perception; and the imposition of greenhouse gas constraints that would make low- or non-carbon energy technologies more cost-competitive with their currently cheaper fossil-fuel counterparts.42 In the case of nuclear power it is fair to say that understanding of the technology and of the potential developments that could mitigate some of the concerns reviewed above—both among the public and among policymakers—is dated. A transparent and scientifically driven re-examination of the issues surrounding nuclear power and their potential solutions is needed.