In early November news broke that NuScale faced a setback. The Utah Associated Municipal Power Systems (UAMPS) pulled the plug on an already scaled down six-unit plant. This brings up the question of the business case of small modular reactors. Is the hype justified, or an empty shell about to burst?
I’m going to make three arguments for the use of SMR’s. But first I’d like to make clear that I’m not against large water reactors, or ‘traditional’ nuclear power plants. Many proponents play a zero sum game, pretending LWR’s are somehow outdated/dangerous/costly/etc. in order to make room for their preferred solution. I find this very harmful behaviour.
Often touted advantages of SMR’s aren’t actually unique to them at all. For example and as a short aside, they’re supposedly going to be built in a factory, which allows for costs coming down. A clear advantage over large units, right? Not really. Back in 1981, France was in the middle of their Messmer Plan and Framatome was building these by the dozens. Rosatom today employs a very similar ‘assembly line’ method. We can, no, we need this level of fabrication again. The French led nuclear alliance is currently working on legislative change in the EU, but the logical next step (in the coming decades) is to ramp up production of EPR’s across the continent. They can be assembled again in a factory hall like this.
We need all solutions towards the goal of zero emissions. That is, all nuclear, but also solar and wind. I previously explained that the latter are very resource intensive and, therefore, very wasteful. They’re also very expensive compared to nuclear. This recent overview by Bank of America makes the same point. I stand by those facts, but solar and wind can lower our fossil fuel needs a lot faster. They can lower them up to a point, corresponding to their capacity factor, but we haven’t reached that yet. So yes, we can lower our fossil fuel need, currently providing about 82% of all our energy, by maybe 40% for continuous energy needs like electricity, and perhaps more if we employ them to make synthfuels. This would come at enormous cost, in money, resources and human labour, but it’ll have to do for now.
In the longer term, it is obvious we’ll become a fully nuclear species. It’s simply vastly more efficient and preferable: in land use, in resource use, in waste created, in fuel needed, etc. But it has a major problem, at least in the West, which is that we’ve been building too few units over the last generation to make headway. Indeed, it gave rise to the myth that nuclear energy has a ‘negative learning curve’ meaning, supposedly, that over time it has only become more expensive.
This is the first area where SMR’s actually come in. The main problem is that there’s a lack of confidence in the technology and it shows in several areas, like high financing costs and lack of experience. The lack of experience can actually be solved easily, by just building units. Hinkley Point C reactor 2 already shows being built 25% faster than reactor 1. So much for the ‘negative learning curve’.
But financing costs remain an issue. So much so that I was told in no uncertain terms by the main defender of Vogtle, Tim Echols, that he expects no AP1000 units will ever be built in the US. SMR’s can help. They might produce electricity at higher costs than LWR’s, but their upfront capital costs are going to be lower. Think low single digit billions of dollars, not dozens. That makes a difference in that they become more reachable for smaller investors or are seen as a smaller risk for big investors.
Once a few are built in the 2030s, costs come down, they deliver reliable energy, trust grows. Public confidence in the technology is growing too, which is only set to grow further once more units get built. This then sets the stage for building larger LWR’s again. Think AP1000, APR1400, EPR. They are mature designs and experience has been gained. If we keep that experience alive, then we could easily expand from there. I call this the temporary argument in favor for SMR’s: it being a facilitator for LWR’s to get built again in the 2040’s and later.
Then there’s the application argument. Many SMR designs are in reality scaled down LWR’s, which operate at a temperature of at most 300 ºC. The use for this is limited, like district heating. More interesting however are going to be fast breeder reactors – like sodium and lead-cooled reactors – and thermal breeders – like molten salt reactors. What makes them interesting, is their higher operating temperature, at 500 ºC to 800 ºC. This can be applied as process heat for a wide variety of industrial processes. It’s very feasible that an industrial area builds one or more SMR’s of this type just for their own use in process heat. This could indeed bring great strides in the deep decarbonisation of industry.
Last but not least, there’s the spatial argument. Most of humanity lives in cities nowadays, but many don’t, or live in smaller towns. A big LWR wouldn’t make much sense in those areas, so an SMR could help a lot in decarbonising these.
SMR’s have a place in our future. Not against old school large reactors, but side by side, fighting for a cleaner world.
This commentary was shared for free. If you want to support me, please consider becoming a patron.
