Small Nuclear Reactors

There has of late been a lot of promotion of the idea of Small Modular Reactors (SMRs) of a few tens or hundreds of megawatts, which it is claimed will be cheaper than conventional gigawatt scaled plants since they can benefit from economies of mass production in factories. Much has been promised for SMRs, including the delivery of power at £40-60/MWh, but there is still some way to go before any project actually goes ahead and we can see if the promises hold up in practice.   

In the past, the nuclear industry had tried to improve the economics of nuclear plants by going for larger plants, without too much success: the on-site construction costs have escalated. However, it is not clear if small plants will have any more success. Scaling down does not necessarily reduce complexity, and, given the need to ensure safety, it is that which may drive costs most. Nevertheless, developers are trying their luck, with many proposals for devices emerging around the world. 

In fact, few are actually new. Most are basically variants of ideas proposed, and in some cases tested, many decades ago, but mostly then abandoned. The most developed of the new retro- wave, the NuScale reactor, is however a scaled down version of an idea that did get followed up, the standard and widely-used pressurised water reactor. Given that its basic PWR technology was familiar, it is perhaps not surprising that this design has achieved regulatory clearance relatively rapidly. That’s just design acceptance, NuScale still have apply for construction permission, but they are expected to do that by 2022. So we may see some prototype tests in due course, and, possibly later this decade, if all goes well, some commercial projects. Rolls Royce are also promoting a mini-PWR design, which, it is claimed, will be ready for grid use by 2030.    

Some of the other SMR proposals are less developed and may take more time to get to that stage. But it is claimed that one of the more novel design, the Natrium fast reactor system, proposed by Terrapower and backed by Bill Gates, will be on line this decade. Given that this makes use of liquid sodium and molten salt heat storage, that is quite a claim. So is the idea, also being pursued by Terrapower, that reactors can be run with molten salt fluoride as both reactant medium and reaction coolant. It has even been claimed that reactors like this, with suitable fast spectrum neutron fluxes, can burn nuclear waste. That has yet to be proven. But certainly, if they are to use thorium as a fuel, they will need an input of plutonium or some other neutron source, since thorium is not fissile, so in that sense they do recycle something.  Though we still have to have uranium reactors, possibly fast breeders, to make the plutonium.

As can be seen, there are many possible problems ahead for SMRs. Perhaps the central one is safety. Working with high radiation fluxes in small confined spaces is not easy. Even in the case of molten salt systems, which avoid the need for control rods, the super-hot corrosive fluids have to be pumped around for heat and waste extraction. It may not be easy to design compact systems that can do this reliably long-term.

The safety issue interacts with the other key issues for SMRs- location. If they are going to be economically viable, some say that SMRs will have to be run in Combined Heat and Power ‘Cogen’ mode, supplying heat for local used, as well as power for the grid. That implies that they will have to sited in or near large heat loads i.e. in or near urban areas. Will local residents be keen to have mini-nuclear plants near by?  That issue is already being discussed in the USA, with some urban resistance emerging.  

A key issue in that context is that it has been argued that since they allegedly will be safer, SMRs will not need to have such large evacuation zones as is the norm for standard reactors, most of which are sited in relatively remote area. Indeed, unless that requirement was changed, operation in cities could be impossible- they could not easily be evacuated fast if there was an accident, or perhaps a security threat. On the basis of this view, SMRs will only ever be relevant for remote sites, and of course there are plenty of such locations where local power generation might be welcome, although arguably, renewable sources might be easier, safer and cheaper to use. Indeed, that might be said of all locations.

The debate over safety, security and location will continue to unfold, with folksy mini-nuke designs emerging for remote rural locations, but concerns also growing over the many unknowns, not least the costs and market potential. There are SMR programmes in the US and UK and elsewhere, but there are big doubts about whether there would be a viable market for this technology. That is despite the fact that there is some dual use/expertise overlap between civil and military nuclear, and, more specifically, that mini reactors are used for submarine propulsion. While that may be one reason why companies like Rolls Royce are pushing for SMRs, on its own military submarine use is a relatively small market.

There is no shortage of promotional enthusiasm for SMRs for a variety of reasons, including, it is claimed, defence-related, and some arguably extravagant claims on comparative  investment costs have been made. However, there have also been some strong critiques and gloomy prognoses.  At best, they say, SMRs may have a role to play in some remote locations and, as with nuclear generally, perhaps for heat production and hydrogen production, for example for industrial purposes. It has also been claimed that SMRs could produce synthetic aircraft fuel as substitute for kerosene, although ‘at around about twice the price’. 

That all seem to be a long shot, with many unknowns, and in terms of energy supply of whatever type, renewables may have the edge in most contexts. However, it is just conceivable that SMRs could be used to back up renewables. Some types of SMR may be able to run more flexibly than can large conventional reactors, so that they could play a role in balancing variable renewables.  That is still very uncertain, in operational and cost terms, and there are many other arguably simpler, safer and cheaper options for grid balancing.  Though, evidently keen to try their luck, a UK developer has talked of using  NuScale units in a hybrid wind-SMR system.

So what’s the bottom line?  For the moment, although being pushed in the US and UK and elsewhere, SMRs are some way off, with very mixed prospects.  But technology can move fast, and although there will no doubt be local resistance, and they may not pop up near you for a while, we may yet see fission-based SMRs emerge for some remote applications within in a decade or two. Can the same be said for fusion? Some very optimistically are talking about the arrival soon of mini fusion! That seems unlikely, and my guess is that, if fusion SMRs are ever possible, their main use will be off-planet. Same possibly for most fission SMRs! Back on this planet, we’ve got plenty of renewables to get on with, and in that context, arguably, small nuclear, of whatever sort, does not really offer anything different from big nuclear. Just another costly distraction from getting on with renewables…