Mini nuclear reactors are already losing their glow

Nascent technologies don’t come more hyped than small modular nuclear reactors. These pint-sized nukes, which are expected to be operational in the UK by the mid-2030s, have won fans from the public sector and tech giants such as Amazon and Microsoft. Yet the glowing case for SMRs appears to be fading already.

The technology has attracted a lot of interest, and with good reason. Even the largest SMR designs, with a production capacity of 350 megawatts, are some 10 times smaller than the UK’s Hinkley Point C plant. That’s one reason why upfront investment costs are lower and build times shorter — adding to the attractions of nuclear as part of the cleaner global energy system.

If all the stars align — which is to say that government policy is supportive, industry delivers and the technology works — there could be a global fleet of 1,000 SMRs with aggregate capacity of 120 gigawatts by 2050, the International Energy Agency forecasts. That is one-tenth of the combined overall nuclear capacity pledged by 30 countries in that year. Governments are pressing forward. In the UK, engineering group Rolls-Royce last month won approval to site its maiden SMRs off the coast of Wales.

Yet despite the promise, many SMR developers listed in the US have had a horrid time of late. NuScale has shed more than half its market value over the past six months. Oklo has experienced a similar loss of value in the past three months and Nano Nuclear Energy is not far behind.

As a fledgling technology, SMRs still have a lot to prove. One issue is that, while smaller nukes require smaller cheque books, their cost per unit of capacity is higher than those of their massive conventional cousins. A single standard has yet to emerge: the Nuclear Energy Agency counts nearly 130 SMR technologies globally.

On top of that, they share many of the same challenges as their larger peers. Three-quarters of the world’s uranium hails from a small handful of countries, and enrichment is dominated by Russia and China. Several studies suggest SMRs will produce higher levels of nuclear waste than traditional plants. One 2022 survey found the volume of waste could be double or as much as 30 times bigger, due in part to the use of chemically reactive fuels and coolants. Security, too, remains paramount.

SMRs are one part of the solution to combine growing demand for electricity and decarbonisation. But with so many options under development, clients are inclined to play the field. Tennessee Valley Authority, the US’s biggest power provider, has signed agreements with a clutch of SMRs.

It’s worth remembering too that even the projections of energy demand that SMRs are designed to address may shift. More efficient chips such as Google’s TPUs could kibosh the exponential growth charts illustrating data centre demand. That serves as another reminder: capital-intensive industries can get sidelined by fast-moving technologies.

[email protected]

Letter in response to this column:
Nuclear (con)fusion / From Tom Greatrex, Chief Executive, Nuclear Industry Association, London WC2, UK