LUXEMBOURGāS energy minister described it as . Austria declared it would . And climate activist Greta Thunberg railed against .
Their anger was directed at a draft regulation from the European Commission, released earlier this year, that would designate nuclear power as a āgreenā source of electricity and thereby make nuclear projects eligible for favourable financial terms. France, Europeās leading atomic state with and more planned, was unperturbed.
Advertisement
Such are the fault lines. āThe debate over whether we need nuclear power is very polarised,ā says at the University of British Columbia in Vancouver.
For many, it is too dangerous and expensive. Others say that nuclear power is a reliable source of clean energy ā indispensable if we want to meet increasingly ambitious climate goals. In its 2021 report, the Intergovernmental Panel on Climate Change (IPCC) included nuclear-generated electricity in all four of its carbon emissions by 2030. Add in fresh concerns over energy security, and it is no surprise new life has been breathed into an old question: do we really need nuclear energy?
In the fog of claim and counterclaim, it can be hard to know what to think. Arguments over nuclear seem to generate a lot more heat than light. However, now more than ever we need answers to key questions: Is nuclear prohibitively costly? Can we build it quickly enough? Is the waste problem a deal-breaker? And, ultimately, is there a better way to meet demand for carbon-free electricity when renewable energy sources fall short?
Nuclear has been unpopular for decades, and existing anxieties were stoked anew when a tsunami hit the Fukushima plant in Japan in 2011. The country suspended all 50 of its reactors and several other nations delayed or cancelled plans to build new ones. Germany decided to phase out nuclear power all together.
That seemed to make economic sense too, at the time. Gas was cheap and the cost of renewables was tumbling fast. It also meant no more radioactive waste (see āThe waste problemā). The conflict over nuclearās value in the fight against climate change looked settled, as demonstrated by the fact that the number of active reactors has hardly changed since the 1980s and looks set to drop (see āWhat if nuclear fades away?ā).
In 2022, though, things look a bit different. True, Germany is set to follow through on phasing out nuclear even as Europe struggles with energy shortages resulting from Russiaās invasion of Ukraine. But France and Finland will start producing electricity at new, more advanced reactors, the first of their kind in the West. The US is building two new reactors in Georgia. And China, the worldās biggest carbon emitter, is committing over the next 15 years.
Ģż
Ģż
The IPCC would approve. The International Energy Agency (IEA), too, insists that nuclear must be part of the solution to the climate crisis. In its , the IEA said that by 2050 āalmost 90 per cent of electricity generation comes from renewable sources, with wind and solar together accounting for almost 70 per centā and .
The main reason the IPCC and the IEA consider nuclear indispensable is that the most abundant sources of renewable energy are variable: the sun sets, the wind drops and the seasons vary every year. āYou only get energy from them 20 to 30 per cent of the time, and not at the right time,ā says , director of the Center for Advanced Nuclear Energy Systems at the Massachusetts Institute of Technology (MIT). To get around that, you need a means of storing wind and solar energy when it is in excess to fill any gaps, which makes the renewables package much more expensive than advertised. āYou canāt just look at the cost of a single kilowatt-hour; you have to look at the cost of the whole system,ā says Buongiorno. āThat aggregate cost is much higher than if you have a little bit of dispatchable low-carbon capacity.ā
A ādispatchableā supply is one that can be ramped up and down as required or at least fed into the grid at will. We can no longer rely on coal or gas for this, and many experts argue that nuclear is the most sensible alternative. āAll the serious analyses ā from MIT, Stanford and the IPCC for example ā show that the path to deep decarbonisation, with the least cost and shortest time, has renewables, a little bit of storage and quite a bit of dispatchable carbon-free capacity, like nuclear,ā says Buongiorno.
Some would question the description of nuclear as ācarbon-freeā. In construction, nuclear plants require vast amounts of concrete, production of which has a hefty carbon footprint. Despite this, a emissions per unit of energy generated as equivalent to wind and lower than solar power. āI can see why the IPCC put nuclear power in a category of low-carbon,ā says at the UK policy research institute Chatham House.
Not everyone agrees. When you take into account the 10 to 21-year time lag between starting work on a nuclear project and a reactor becoming operational, the numbers change, according to at Stanford University in California. We need another supply of energy for all that time, he argues, and its carbon footprint should be pinned on nuclear. āThe carbon dioxide emissions of nuclear are nine to 37 times those of wind power per unit of electricity generated,ā he says.
There are other wrinkles, too. One is that investment in nuclear seems to reduce commitment to curbing carbon emissions. A of 123 countries over 25 years showed that ālarger-scale national nuclear attachments do not tend to associate with significantly lower carbon emissions, while renewables doā. France was 4 percentage points behind its target for renewable generation in 2020, while almost all other European Union countries were way ahead of theirs. Germanyās nuclear phase-out induced investment in wind and solar, and the country was 1 per cent ahead of its 2020 target, despite using fossil fuels to plug its āenergy gapā as reactors shut.
Where critics say nuclear canāt possibly compete with renewables is in how quickly newly commissioned installations can begin to deliver. āIt takes too long to site and build nuclear reactors, especially compared to solar and wind installations,ā says , who researches technology and society at MIT.
That is because building new nuclear plants is hard. Franceās Flamanville-3 reactor is a decade behind schedule. Finlandās Olkiluoto-3 turned on partially this year, but it was meant to be finished in 2009. Since 2011, ground has been broken on the construction of 57 new nuclear reactors globally. By 2021, just 15 had started to deliver electricity.
Despite these problems, Buongiorno is āsomewhat optimisticā about new nuclear. āThere are reasons to believe that the next wave of construction might be a little bit better,ā he says ā namely a series of studies into what went wrong with previous projects. Moreover, countries such as South Korea, China, India and Russia routinely deliver nuclear power plants on time and within budget. And some claim that new reactor technologies will make a difference to the time lag problem (see āNuclear but nimbleā).
We now know that integrating the designers and builders from the start, for instance, makes an enormous difference to the delay and cost escalation down the line. These insights, says Buongiorno, will get us new nuclear in time to make an impact on carbon emissions. āThings are certainly tight, but theyāre tight for all technologies,ā he says. āIt has been very hard to scale up renewables as well.ā
The trouble with that argument is that we can scale up renewables faster than nuclear, says Froggatt. āThe next decade is crucial to meet global greenhouse gas targets,ā he says. There needs to be a 45 per cent reduction in emissions by 2030 to keep global warming below 1.5Ā°C above pre-industrial levels, but āif a reactor is not currently under planning, it will not be completed by 2030ā, he says. āSo nuclear power will only, at best, very marginally increase its contribution to the global energy mix. Therefore, accelerating global emissions reduction by 2030 canāt be with nuclear power.ā
Others are thinking longer-term, though. They say that acting now to establish nuclear in the energy mix by 2050, will go a long way to ensuring that we can sustain a low-carbon energy sector through the rest of this century.
Whatever the deadline, investing in new nuclear will require deep pockets. When construction started on Flamanville-3, it was expected to cost ā¬3.3 billion. The latest estimate is ā¬12.4 billion. The price tag for the two new US reactors has risen from $14 billion to $29 billion. But maybe Buongiornoās optimism about the industry learning from the past will be validated. Projections do show that the cost of nuclear-generated electricity is falling: according to the IEA, it is expected to drop by 7 per cent between 2020 and 2050. But the cost of renewables is expected to fall faster: onshore wind by 14 per cent, offshore wind by 66 per cent and solar by 58 per cent. In some markets, investors in nuclear could struggle to get their money back.
We have to be careful with all such figures, warns Ramana. Energy cost forecasts are inherently unreliable. But even if we assume that cost comparisons will make it hard to justify using anything other than a minimal amount of nuclear power in the future, proponents say that is all we need.
Do we even need that though? These days, some experts argue that the intermittency problem of renewables can be solved without nuclear. āBaseload, in the sense of needing something that generates 24/7, is outdated,ā says , an energy policy researcher at University College London. āAs we get more renewables, weāre going to need more flexibility, whether from managing demand, from storage, or from connecting to neighbouring countries.ā
Grid operators already pay large companies to reduce their electricity use at times of peak demand, and there are moves to do the same in domestic settings: we could pay people to store energy in electric vehicle batteries and draw on it when the vehicles arenāt in use. Industrial-scale battery storage is also coming online. This tends to be a short-term solution, but there are longer-term options such as pumped hydropower, which uses excess energy to pump water uphill, ready to run through power turbines. And then there are ācontinental supergridsā ā sprawling energy networks that shunt power thousands of kilometres ā which would mean you can always get electricity, wherever it is generated.
It remains to be seen if such innovations can pick up the slack. And herein lies the problem when it comes to making your mind up on nuclear: so much depends on things we donāt know for sure. On the one hand, why bother with it, given its drawbacks, if we can meet our net-zero targets with nothing but renewables? āI tend to believe we can do 100 per cent renewables,ā says Ramana. On the other hand, why put all your eggs in one basket? āItās not an either-or situation: weāve got to do them both,ā says Buongiorno. Even among those with all the facts at their disposal, those fault lines remain. Add in the fact that nuclear weapons make things even more complicated (see āThe weapons connectionā), and it is clear why nothing is clear. Whatever call we make, some people will say it is wrong. The only thing we can do is be sure that we donāt make the call too late.
THE WEAPONS CONNECTION
Russiaās invasion of Ukraine reminds us that we live in a nuclear-armed world. Nuclear weapons are a part of the global calculus of peace, and they have to be maintained and positioned ready for use to serve their purpose as a deterrent. All of which requires nuclear know-how, often maintained through civil nuclear-power programmes.
Most governments keep quiet about this interplay, but US president Barack Obamaās former energy secretary Ernest Moniz has made the link explicit. In 2017, Monizās think tank, the Energy Futures Initiative, issued a report entitled . It said that āa strong domestic supply chain is needed to provide for nuclear Navy requirements. This supply chain has an inherent and very strong overlap with the commercial nuclear energy sector and has a strong presence in states with commercial nuclear power plants.ā Franceās president Emmanuel Macron put it even more succinctly in 2020:
If the pool of experts gets small, civil and military nuclear can even end up competing. Rolls-Royce, the UK governmentās choice for its new small modular reactor programme is also the UKās only naval nuclear propulsion supplier. In a 2006 to the UK parliamentās Defence Committee, it complained that the depletion of civil nuclear skills had āreduced the support network available to the military programmesā. The firm said building new civil nuclear power plants would put āconsiderable pressure on the UKās remaining [nuclear] skills capacityā.
So the decision about whether to invest in civil nuclear power isnāt as simple as ādoes it provide marketable electricity?ā. However you feel about nuclear weapons, a dearth of relevant expertise isnāt conducive to a good nightās sleep.
NUCLEAR BUT NIMBLE
While many countries persist with large and difficult-to-build full-scale nuclear reactors, there is now a less cumbersome option. Small modular reactors (SMRs) are entirely or partially assembled in factories and brought to site on the back of a lorry. The idea is that although they have a relatively low energy output, SMRs are faster and cheaper to build. āOverall, the scope of these projects is going to be more manageable than traditional multi-billiondollar, 10-year projects,ā says , director of the Center for Advanced Nuclear Energy Systems at the Massachusetts Institute of Technology.
In which case, SMRs could change the calculus regarding the cost of nuclear energy (see main story). The trouble is that no commercial SMRs are up and running. āThe ānew reactorsā donāt exist yet, so we have no idea what their costs, time requirements, waste risks, meltdown risks and weapons proliferation risks will be,ā says Mark Z. Jacobson at Stanford University in California.
There are efforts to get SMRs off the drawing board. In the UK, Rolls-Royce is leading a consortium to produce one based on the āpressurised water reactorā (PWR) concept originally designed for military aircraft carriers and submarines, and subsequently used in a large proportion of the worldās nuclear power plants. Meanwhile, US start-up NuScale, which is also using a modified PWR design, has received regulatory approval.
There are plenty of other SMR designs in development, too. US businessmen Bill Gates and Warren Buffett are putting money into TerraPower, for instance, whose design involves a sodium-cooled reactor. Pending regulatory approval, the aim is to begin construction of the first plant in Wyoming, with energy being delivered in 2030.
But hereās the rub: the end of the decade is the finishing post for most SMR projects. This is too late for helping with 2030 emissions targets. And the finishing post tends to move, says M. V. Ramana at the University of British Columbia in Vancouver. āAt this point, NuScale is about 15 years behind its promised schedule. It started in 2008, they were hoping to generate power by 2015 or 2016. Now they are talking about 2029 or 2030.ā The upshot is that it is still too early to say if SMRs will make a substantial difference to arguments over nuclear power.
THE WASTE PROBLEM
Radioactive contamination from accidents and waste from spent nuclear reactors are often cited as key reasons to shun nuclear energy. If there are other routes to net-zero emissions, the argument goes, why saddle future generations with these toxic legacies.
That problem isnāt going away. There are no neat solutions, even if Finland has begun construction of an underground repository for waste that hailed as a . But letās be honest: waste storage and contamination have yet to stop the world turning. The climate crisis requires urgent solutions and nuclear could be part of that, so doesnāt it make sense to set aside the waste issue?
It isnāt as if renewables are guilt-free. Wind turbine blades last for about 20 years, but they canāt all be recycled at the end of their life. At the moment, . When solar panels reach the end of their lives, they are arguably worse, because they contain toxic chemicals such as cadmium and lead. The International Renewable Energy Agency has predicted that the planet will be littered with up to .
Whatās more, modern nuclear reactors should create less of a waste problem than before. At Hinkley Point in the UK, new reactors will use 17 per cent less uranium and produce almost a third less long-lived radioactive trash than todayās water-cooled reactors. The small modular reactors coming off the drawing boards will also produce less waste than other reactors.
This is something to celebrate, surely? Not according to Mark Z. Jacobson at Stanford University in California. āThose that produce less waste will result in greater weapons proliferation risk since waste is reduced by refining uranium to a greater degree, and that results in closer to weapons-grade uranium,ā he says. As ever with nuclear energy, it is more complicated than it first appears.
Ā鶹“«Ć½ audio
You can now listen to many articles ā look for the headphones icon in our app newscientist.com/app
