In the political thriller Occupied, Norway declares a halt to oil and gas in favour of thorium-fuelled nuclear energy, alarming countries from the European Union (EU) to Russia. This isn’t so far-fetched: nuclear could help slash emissions in the datacentre and beyond, according to Daniel Bizo, research director at Uptime Institute.
“Nuclear power offers low-carbon dispatchable generation, which can complement renewable energy in displacing fossil fuels, as well as providing stability as the share of intermittent generation like wind or solar grows,” he tells Computer Weekly.
“The backdrop to this is that we are not on track with global greenhouse gas emissions reductions to meet 2030 goals, which makes it likely that longer-term goals for 2050 will also be missed by a wide margin.”
Bizo points out that the rising use of renewables is making for mismatched supply and demand, especially as total demand grows. Another factor is the increasing electrification of industries and transport, which is also important for cutting emissions.
Meanwhile, the carbon intensity of power generation in France, with nuclear comprising 66% of its power mix (versus 16% in the UK), is well below other major developed economies. The only better-performing grids are where hydro and geothermal energy dominate, such as Norway and Iceland. Even Sweden, with ample hydro power, uses nuclear for stable baseload capacity.
“The moment electricity purchases enter corporate sustainability reporting it starts mattering. And as large energy buyers, datacentre operators send signals to the energy market and regulators through their purchases,” notes Bizo.
Without support from energy consumers, the valuation of nuclear initiatives can be affected, or even hasten the closure of nuclear plants rather than extensions of their licences, he says.
Shutting down existing plants hinders sustainable development, not least because building new infrastructure of all kinds is highly carbon-intensive. One danger is that datacentre operators pursue renewables at the expense of nuclear – even when nuclear is already present in the energy mix – sometimes as a response to public opinion, he adds.
Turning waste into energy
Max Schulze, founder of the Sustainable Digital Infrastructure Alliance (SDIA), suggests datacentres should be early customers and supporters, and outline what they are willing to pay for smaller modular reactors (SMRs), defined as supplying 60-300MW each, that can work as on-site generation.
“An industry-wide commitment to nuclear energy, as well as such a commitment to pay X dollars per megawatt hour (MWh), could accelerate commercialisation of SMR and other technologies and set a clear boundary condition around price,” Schulze suggests.
“Importantly, smaller reactors are better – they’re less risky. Then, next-generation reactors can actually burn old nuclear waste from previous reactors. So if we focus on those aspects, nuclear can be a step forward.”
A Rolls-Royce-led consortium is developing SMRs, each about the size of two 105x68m football pitches. These can be ordered off a production line and installed like any other piece of kit. Rolls-Royce says each could supply 220MW-440MW – equivalent to around 150 onshore wind turbines – for a similar price per megawatt as a large-scale reactor.
US startup Oklo is developing a 1.5MW nuclear microreactor with the goal of processing spent high-assay, low-enriched uranium (HALEU) nuclear fuel from Idaho National Laboratory (INL) by 2024. Other companies are also reportedly developing <10MW reactors.
Nuclear physicist Stefano Buono, chief executive of nuclear tech developer Newcleo, says even existing fission technologies are becoming safer and more sustainable, through the development of new ways of processing spent fuel. Instead of burying plutonium and other waste products, Newcleo is working on ways to use nuclear waste as fuel in modern Generation IV reactors.
Generation IV reactors like Newcleo’s have been recognised by the EU as sustainable energy generation sources. When thorium is used as fuel for fission, for example, the waste can be reprocessed and is cleaner.
Buono says Newcleo’s reactors, cooled with liquid lead, can not only use all the uranium mined for the “old” nuclear industry, but can be made smaller and for a lower cost.
“We will drastically reduce environmental impact also on the fuel sourcing side. Our technology will induce a very significant reduction in both volume and lifetime of existing waste by transforming this into energy,” says Buono.
Mixing it up
While wind, wave, geo and solar are great for the energy mix, their power-generating capacity is intermittent. One of nuclear energy’s strengths is it can provide baseload power when the wind doesn’t blow and the sun doesn’t shine. Nuclear power works alongside these other sources of renewable energy as part of a diversified energy mix, says Buono.
Buono and Schulze agree that nuclear fusion energy, on the other hand, is decades off commercialisation. Strides are being made, however, with considerable advantages in train.
In February, EUROfusion scientists and engineers at the Joint European Torus (JET) facility, currently at the UK’s Culham Centre for Fusion Energy in Oxfordshire, announced the generation of 59MJ (megajoules) of sustained fusion energy.
During the experiment, JET averaged a fusion power of around 11MW (or MJ) per second, suggesting future potential for safe, efficient, low-carbon energy supply.
Tony Donné, professor, CEO and programme manager at EUROfusion, says the fuel JET used for fusion is deuterium (2H) and tritium (3H) – both forms of hydrogen. “Tritium is a proton with two neutrons. It is a beta emitter and the half-life is very short, only about 12.5 years,” he says.
That compares with the slow radioactive decay of the uranium that fuels conventional fission reactors. Of the various isotopes used, uranium-238 has a half-life of 4.5 billion years and uranium-235 a half-life of 704 million years.
“In our fusion reaction, the product is helium and a neutron. Helium is non-toxic and non-radioactive, and the neutron is also not radioactive. But it has a lot of energy, which is converted into heat,” Donné explains.
Only a small amount of fuel, sourced from fission, is needed to fuel the fusion reaction, heated to a plasma, typically at 150,000,000ºC in a vacuum chamber. If something happens, such as a breach of the reactor, cold air enters from outside and the reaction stops. There’s no equivalent to the nuclear meltdown situation seen in fission reactors, Donné says.
“You convert mass into energy, but you need much less. For instance, when we produced that 59MJ of fusion energy, we only needed 170 micrograms [μg] of fuel.”
With coal or gas, you’d need four kilograms of coal or one kilogram of natural gas. The largest issue, says Donné, is that the emitted neutron must be absorbed by a barrier that then becomes radioactive, which means they need materials that “lose” radioactivity quickly. The project is currently working with steel-vanadium alloys that become almost non-radioactive in 50-100 years.
“About 100 years after you stop a reactor or after you change the blankets, you can use those materials again,” he says. “You do always have to be careful: with radioactive 3H, you should not inhale it. That said, if it’s in the air, after just 100m it’s very dilute.”
Nor does radioactive 3H accumulate in human organs, like caesium or radon does – the biological half-life is 10 days. The safety issues are orders of magnitude less than in fission, he says.
Complex carbon equation
Donné says that, because commercial fusion power is likely over 10 years away, a diverse energy mix that includes current nuclear facilities is needed, especially because the full costs of renewables developments, including embodied carbon, can be higher than thought.
Germany once had a mix including 30% nuclear, but the phase-out has thrown it back to relying on coal, even with 40% of supply from renewables such as wind and solar. This means that overall emissions from electricity generation have not decreased.