Global warming is rooted in one of the most fundamental ideas of Newtonian physics: there is no action without a reaction. Put simply, we cannot continue to pump carbon dioxide and other pollutants produced from the burning of fossil fuels into our environment without suffering the consequences. Environmental scientists have been highlighting this problem for some time, but only now are governments giving the issue the attention that it deserves. Man-made climate change is one of the greatest threats our planet faces, and is already estimated to be responsible for over 160,000 deaths worldwide each year resulting from heatwaves, flooding and crop damage.

Yet in tackling global warming we face a dilemma. Fossil fuels provide at least 85% of our total energy needs, from the electricity that powers our homes to the production of manufactured goods and our food supply. Renewable energy sources, such as those utilizing the Sun, the wind and the waves, can help reduce our dependence on fossil fuels, but their unreliable nature and often low output means that they can only provide a small part of the energy solution. Indeed, most forms of renewable energy have a significant environmental impact of their own – for example by disfiguring the landscape or by endangering wildlife. They also require fossil-fuel power stations to be on standby for when output is low, for example when wind turbines are not generating in still conditions.

Fortunately, there is another option to tackle our looming energy crisis: nuclear power. At the atomic level, the thermal energy released in a fission event is 200 MeV, compared with only a few electron-volts produced each time a hydrocarbon molecule is broken by burning carbon-based fuels. As a result, a single nuclear reactor fuel pellet just 1 cm long can produce the same amount of electricity as 1.5 tonnes of coal. Furthermore, nuclear power produces tiny amounts of waste, as opposed to the vast volumes of pollutants pumped unchecked into the environment by the burning of fossil fuels. Although nuclear waste is much more toxic than these pollutants, it can at least be completely contained.

Nuclear power came to the fore in the late 1950s and 1960s, with the building of many nuclear power stations around the world. However, the environmental hazards associated with nuclear waste have always been an argument against nuclear power. Combined with the Chernobyl accident in 1986 and market forces in the energy sector, the nuclear industry went into decline in the 1980s and 1990s. But the tide now appears to be turning. In May, for example, the UK government signalled its intention to build a new fleet of nuclear power stations across the country, and several other countries, including China, Finland, France, India and Russia, have announced or even begun work on building new reactors.

It is not just the urgent need to combat climate change that is fuelling this nuclear revival. Economic arguments based on spiralling gas and oil prices, plus strategic interests in ensuring individual countries have a stable energy supply, are also major factors. In fact, so strong are these economic and strategic arguments that it now seems impossible to see a realistic solution to our energy needs without nuclear power playing a significant role once again. And where there is nuclear power, there are physicists.

A history of design

Nuclear reactors are powered by the energy released in nuclear fission. This process involves firing neutrons into uranium-235 nuclei, which convert into uranium- 236 nuclei with enough excess energy to become distorted and split into two heavy fission fragments plus two or three additional neutrons per fission event. The small mass difference between these final products and the initial neutron and uranium-235 nucleus is released as energy through Einstein’s famous equation.

Most of this energy ends up as the kinetic energy of the fission products, which generate a lot of heat by colliding with surrounding atoms. This heat is carried away by a coolant such as carbon dioxide or water (which forms the primary coolant circuit) and is used to heat boilers in a secondary circuit that produces steam to drive a turbine and generator – just as in a power station based on fossil fuels. Of the neutrons released, some will escape from the reactor while others are absorbed, but about half will split further uranium nuclei, triggering a chain reaction. To keep this process under control most reactors require a moderator – usually made of graphite or water because their light atoms are good at absorbing the kinetic energy of the neutrons.

The world’s first commercial nuclear power station opened in the UK in 1956 at the Sellafield site on the Cumbrian coast, and it ran for almost half a century before closing in 2003. The four Calder Hall reactors were of the Magnox type, which means they used a magnesium “no-oxidation” alloy to encase the uranium fuel rods. As well as retaining volatile fission products, such as caesium and strontium, this Magnox cladding has a low neutron-absorption cross-section and therefore reduces “parasitic absorption” of neutrons. Made of graphite and containing holes both for the fuel rods and to allow the cooling gas to flow, the moderator slows the neutrons by elastic scattering such that their kineticenergy distribution becomes comparable to that of a gas in thermal equilibrium with the graphite. Since at these energies neutrons have a much higher probability of interacting with atoms, Magnox reactors can use fuel containing naturally occurring levels of uranium-235 (about 0.7%), avoiding the need – and expense – for the uranium to be further “enriched”.

By the early 1970s the UK had 11 Magnox nuclear power stations (containing a total of 26 individual reactors) either fully operational or in various stages of construction or planning. It had also exported the Magnox design – since termed “Generation I” – to Japan and Italy, which each have one plant. In a bid to increase the ratio of electrical to thermal power output, however, the then Central Electricity Generating Board introduced the advanced gas-cooled reactor (AGR) concept – now referred to as a “Generation II” design. First opened in the mid-1970s, all seven AGR stations (14 reactors) in the UK are still operational.

The moderator (graphite) and coolant (carbon dioxide) are the same in both the Magnox and AGR designs. However, AGRs have much higher thermal efficiencies by operating at a temperature of 600 °C as opposed to about 370 °C in a Magnox unit. Since at high temperatures uranium undergoes a crystalline phase change that makes it expand, potentially weakening the cladding, AGRs use uranium oxide as their fuel. And as Magnox becomes soft and may even ignite in air at AGR temperatures, stainless steel is used as the cladding instead. Since stainless steel absorbs more neutrons than Magnox, AGRs require uranium with a uranium-235 content of a few per cent, the extra cost of which is recovered through increased energy output of the fuel.

The UK also carried out research into “fast reactor” designs until the early 1990s, for example at the Dounreay site in northern Scotland. These reactors have no moderator and the neutrons released in a fission event therefore retain their large kinetic energies. As a result, fast reactors can convert depleted uranium (i.e. uranium with almost all of its uranium-235 removed) into plutonium, which can also be used as a nuclear fuel. Since for every plutonium atom destroyed through fission at least one more is created in spent fuel, the fast reactor – or breeder reactor – creates more fissile material than it consumes, thereby potentially increasing nuclear fuel reserves enormously.

Since the energetic neutrons in a fast reactor have a lower probability of interacting with another nucleus, however, the reactors require more dense fissionable material and materials that can survive very large neutron fluxes. As a result, fast reactors are more complex and expensive than Magnox reactors or AGRs, partly because they require an additional cooling circuit, and the design was never used commercially.

Light-water reactors

Elsewhere in the world, France initially followed the UK’s lead by building reactors similar to the Magnox design during the 1960s. Meanwhile, the US realized that the most economical reactors are those that are collectively referred to as light-water reactors (LWRs). These are simpler to build and to operate than Magnox reactors or AGRs, and they also benefit from economies of scale. The fuel, for instance, has been improved through the joint efforts of many countries so that now it can sustain higher useful energy outputs than AGR fuel, which was developed by the UK alone.

LWRs use ordinary water as a moderator and as a coolant, running on uranium-oxide fuel enriched with up to 5% uranium-235 and contained in a zirconium alloy cladding. LWRs come in two basic types: the pressurized water reactor (PWR) and the boiling water reactor (BWR). PWRs maintain the water in the primary coolant as a liquid and raise steam in a secondary circuit that operates at a lower pressure (see “Power from the nucleus” figure). In contrast, BWRs use a single, two-phase water–steam pressure circuit in which the steam from the core directly drives the turbine. The advantage of this design is that it does not require a secondary coolant circuit and the associated heat exchangers, pipes, valves and pumps. However, this advantage tends to be offset by increased complexity in other aspects, notably maintenance and decommissioning because the steam travelling to the turbines is radioactive and hence contaminates them.

Many of the advantages of LWRs stem from their very compact reactor cores, which are possible because water is the most effective of all commonly used moderators at slowing down fission neutrons. This makes LWRs more economical and much easier to build and operate than Magnox and AGR plants (although the latter do not require such high levels of uranium enrichment). For example, the pressure vessel in which the reactor is contained plus all the surrounding structures are small enough to be built in a factory and transported to site, whereas Magnox and AGR pressure vessels are so big that on-site construction is required.

Although the UK had designed AGRs to be competitive with LWRs, which they almost were, the design ended up being slightly more expensive to build and operate than LWRs. Combined with their slightly worse operational performance, the competition from LWRs was too much in the end – a bit like Boeing versus the small UK aircraft manufacturers. The UK recognized this by deciding to follow its AGRs with PWRs, and the construction of the UK’s first and only PWR – Sizewell B on the Suffolk coast – began in 1988. Indeed, of the 436 reactors currently in operation worldwide 357 are LWRs of which 264 are PWRs, and it is the latter that are predominantly being built today.

New builds

Today many countries are grappling with the problem of how to meet their energy demands while producing less carbon dioxide, and the UK is no exception. When former UK Prime Minister Tony Blair came to power in 1997 – two years after Sizewell B came online – he decided to “park” the issue of nuclear power. But it now seems clear that the UK government has accepted that the only way to meet its ambitious targets for reducing carbon-dioxide emissions is to at least maintain the currently 18% contribution that nuclear energy makes to its “energy mix”.

Building of a nuclear power station does not take place over night. When and if the UK government decides to go ahead with a new fleet of nuclear plants (a decision that is currently under consultation and which will be finally taken in October), it then needs to decide what technology to opt for and who will build and operate the plants. These latter choices are left to market forces, based on whichever consortiums of reactor vendors and owners come forward, and are then subject to stringent safety and environmental criteria. In total, it takes about 10 years before a new nuclear plant can be hooked up to the national grid.

The two mostly likely candidates for new-build reactors in the UK are, like Sizewell B, PWRs: the Areva EPR (European pressurized water reactor) and the Westinghouse AP-1000 (AP stands for “advanced passive” and the 1000 denotes the 1000 MW of electrical power that such units can produce). Other design possibilities are the Advanced Boiling Water Reactor (ABWR), which is basically an optimized version of the BWR, and the Advanced Candu Reactor (ACR), which is based on the very successful Canadian Candu reactors. These units are similar to PWRs but use heavy water (D₂O) as a moderator. Heavy water captures almost none of the neutrons, but as it contains deuterium it is very good at slowing them down. This means that more fission neutrons are available, allowing ACRs to operate with very low-enriched fuel.

The common feature of all these “Generation III” designs is that they are simple to operate: they require less intervention, less fuel and are easier to maintain than previous designs. They also have advanced, passive safety features that rely on physical forces such as gravity and convection, with little or no need for mechanical devices such as pumps. However, campaign groups such as CND and Greenpeace have effectively ignored such features and instead have concentrated on raising concerns about the nuclear waste that a new fleet of nuclear power stations would produce.

While it is certainly true that more nuclear power stations will mean more nuclear waste, the volume of waste generated per kilowatt-hour output will be much less in the new designs than in the older ones. For instance, a fleet of 10 new gigawatt-capacity LWRs would deliver about twice the amount of electricity over their 60 year lifetime as the current fleet yet would produce only about an extra 10% of high-level radioactive waste over the same period under reasonable assumptions. Furthermore, these new reactors could allow us to utilize reserves of civil plutonium by using “mixed oxide” fuel made of uranium and plutonium oxides.

The AP-1000, EPR, ACR and BWR designs all use the same fuel, pressure vessels, steam generators and other key components as today’s operational Generation I and Generation II reactors. New stations based on these designs could therefore be built immediately. Indeed, an EPR is already being built in Finland (see “Generation III” figure), with one in France to follow, while China has ordered several AP-1000s. Perhaps 20 years from now, however, we may be ready to build what are known as Generation IV reactor designs.

Generation IV

In the late 1990s the US Department of Energy selected six Generation IV designs from a shortlist of more than 100 concepts to “broaden the opportunities for the use of nuclear energy”. Three of these designs are fast reactors, which have a sustainable fuel cycle in which plutonium-239 is produced from uranium-238 neutron-capture reactions and could therefore operate for many hundreds of years with existing uranium reserves. The three fast-reactor designs differ mainly in the choice of coolant: namely liquid sodium, liquid lead and helium gas, some of which are better heat conductors, while some are more problematic if they leak.

Another Generation IV design is the supercritical water reactor, in which water in its supercritical phase is used as a coolant. Water in this state (i.e. where there is no distinction between a liquid and a gas) has a very high specific heat capacity, enabling a higher thermal efficiency than with existing LWRs.

There is also the very high-temperature reactor (VHTR), which is related to current HTR reactor designs such as the pebble-bed technology being pursued by South Africa (see Physics World July 2002 pp42– 43, print version only). These reactors typically use graphite moderators and gas coolants, and hold the prospect of high thermal efficiencies. Furthermore, VHTRs are incredibly safe because the radioactive content of the fuel is contained even if the reactor reaches temperatures in excess of 1500 °C (i.e. 500 °C more than the normal operating temperature).

Perhaps the most exciting aspect of the VHTR design, however, is that it can produce hydrogen via electrolysis in water or thermochemical reactions and thus play a role in a future hydrogen economy. Generating hydrogen is an extremely energy-intensive process, requiring either large amounts of electricity or heat – both of which are plentiful in the VHTR design with virtually no carbon-dioxide emissions. The production of hydrogen does not compromise the performance of the reactor, although it does reduce the electricity output. Using fossil fuels to create hydrogen, on the other hand, is not environmentally justifiable.

The final Generation IV design – called the molten salt reactor – is the most radical. Here the fuel is in the form of a uranium salt that circulates in the coolant so that any loss of coolant would shut down the chain reaction. How this works in practice has not yet been formally decided, as research into the molten-salt design – and all the other Generation IV designs, in fact – is at a very early stage. It is unlikely that all six designs will succeed in a real commercial setting. Some will eventually be discarded as some reactors prove more viable than others. Optimistically, nuclear fusion will start to come along at a similar time and add a brand new dimension to nuclear power.

A nuclear renaissance

The nuclear power industry in Europe (with the exception of France) and the US has stagnated since the mid-1980s, with few new plants having been commissioned. This is partly due to the efforts of antinuclear groups and also the Chernobyl accident in 1986, but market forces have played a role too. In the UK, for instance, competition from natural gas, the deregulation of the energy market and uncertain government support made it difficult for new nuclear plants to secure the necessary private investment. In other countries, competition from cheap coal undermined the case for new nuclear plants, while in both the UK and US the successful extension of the operating lifetimes of existing nuclear plants has, ironically, hampered the building of new ones.

Today, however, we are entering a renaissance in nuclear power. Although not the complete solution to climate change in itself, nuclear power can help slow down global warming and provide a reliable supply of electricity as part of a diverse energy mix. And in a reversal of fortunes, the recent rises in gas and oil prices have meant that nuclear power plants are now the most economic energy option in many countries. Given that oil and gas reserves are beginning to run out, the UK and other governments need to follow the lead set by China, France, South Korea and Japan in pursuing a new nuclear programme.

We can therefore expect to see a new fleet of UK reactors, possibly a mix of EPR and AP-1000 designs, coming on-line in the next 10 to 15 years that will run perhaps until 2080. Before that date we may also see Generation IV reactors, some of which may also produce hydrogen. By then, nuclear power in conjunction with renewables would have helped the UK reduce its carbon-dioxide emissions to a more sustainable level. In contrast, by simply replacing the current nuclear fleet with renewables that supply the same proportion of energy (currently about 19% of the total in the UK) we will make no in-roads into reducing our emissions at all.

With prospects in the nuclear power industry appearing much brighter than they were even five years ago, physicists are likely to find themselves in increasing demand. The nuclear field is of worldwide importance and is one of the few areas where physicists can actually use their skills outside academia. Knowledge of the materials science and heat transfer inside a reactor can be just as important as knowing about nuclear reactions and neutron physics, topics that are taught on postgraduate masters courses (see Physics World April 2006 pp42–43, print version only).

Ironically, the recent successful ruling brought by Greenpeace earlier this year against the UK government’s energy-review process also appears to have signalled a revival of the UK’s nuclear-energy programme, since the government responded with a steely determination to repeat the review and insisted that nuclear power is required. Unfortunately, the protests by environmental groups against nuclear power stations have merely led to more fossil-fuel plants being built. The setting up of the National Nuclear Lab in the UK is further evidence of a nuclear revival, and it seems that the tide is turning and we should now welcome a new dawn for nuclear power.

Box: Keeping a nuclear reactor under control

Contrary to what some antinuclear groups would have you believe, nuclear reactors are not unstable contraptions ready to run out of control at any moment. Sound physical principles are used to ensure the safety of any properly constructed reactor. For example, in a water-moderated reactor the neutrons released in fission are slowed down by collisions with hydrogen and oxygen nuclei (the vast bulk is done by just the hydrogen), making them easier to capture. If for some reason the number of reactions increases, however, the additional heat output will cause the moderator to expand – thereby reducing the reaction rate and preventing the system from running out of control. A similar feedback mechanism called Doppler broadening is provided by the increased absorption of neutrons in the reactor materials as they heat up, and the goal is to design a reactor in which several such mechanisms combine to produce a stable system.

An example of a poorly designed system was the Russian RBMK reactor. In 1986 one such reactor at Chernobyl was responsible for the worst nuclear disaster in history. These reactors used graphite to moderate the neutrons and water to cool the system, which are normally good choices. However, an unfortunate combination of the two made the RBMK extremely dangerous: most of the moderation was provided by the graphite while the water mostly acted as an absorber. As the water heated up it boiled off and so the density of the absorber was reduced. This led to more reactions, which boiled off even more of the absorber, triggering an unstable feedback loop.

Another contributing factor in the accident related to the neutrons emitted by the fission products. The two or three neutrons released in a fission event are known as prompt neutrons, since they are emitted either immediately at the point of splitting or rapidly “boiled off” from the excited fission products. However, these fission products themselves sometimes eject neutrons following beta decays. Although accounting for less than 1% compared with the number of prompt neutrons, these “delayed neutrons” – which can go on to initiate further fissions – ensure that in a normal reactor the power levels in the reactor change very slowly and safely. But at Chernobyl the number of neutrons increased very rapidly and unsafely due to the prompt neutrons alone, causing the reactor to go from 10% of full power to 100 times full power in three seconds. This fault was only present in RBMK reactors, which means that “another Chernobyl” could not happen.

Box: Dealing with waste

The common perception is that nuclear reactors generate large amounts of radioactive waste, that this waste is difficult to manage safely, and that it is somehow different from other toxic waste generated by industry. The reality is different: the volume of waste is relatively low, especially so for the high-level waste comprising mostly radioactive fission products and transuranics (which amount to only a few cubic metres from a 1 GW PWR per year). While this high-level waste represents a significant hazard, half a century of experience shows that it can be managed very safely. As for the much larger volumes of lower-level radioactivity waste, its hazard potential ranges from very low to practically negligible (the average extra daily dose to a worker at the repository for low-level waste near Sellafield, for instance, is of the order of that from eating one Brazil nut a day!). Unfortunately the UK, for instance, has failed to make much progress towards building final geological repositories for its higher-level radioactive waste, having only decided late last year that this is the best way to deal with such material. This is a political rather than technical challenge, and one that is largely independent of whether new nuclear power stations are built or not. Irrespective of whether you are in favour of nuclear power or against it, the existing waste has to be disposed of eventually and will still exist at almost the same magnitude even if we build no new nuclear power stations.

At a Glance: Nuclear power

• The first commercial nuclear power station opened in the UK in 1956 and today there are over 400 reactors in operation worldwide
• Most of these plants are light-water reactors, in which water is used both to cool the reactor (thereby extracting energy to drive turbines) and to moderate the neutrons released during fission
• Producing huge amounts of energy without any greenhouse gases, nuclear power can play an important role in combating global warming
• Although hampered by the image problem of radioactive waste, nuclear power is once again back on the agenda of several countries, including the UK
• The next generation of nuclear reactors will be safer and more economical than existing designs and will also produce less waste

More about: Nuclear power

J J Duderstadt and L J Hamilton 1976 Nuclear Reactor Analysis (Wiley, New York)
T Goddard 2006 A future for nuclear power Physics World April pp15–17
K S Krane 1987 Introductory Nuclear Physics (Wiley, New York)
W J Nuttall 2005 Nuclear Renaissance: Technologies and Policies for the Future of Nuclear Power (IOP Publishing, Bristol)
W M Stacey 2001 Nuclear Reactor Physics (Wiley, New York)