A new type of small modular nuclear reactor has been developed which is a form of molten salt reactor. The main radioactive products are stable immobile salts instead of high pressure gases, its inventor Dr Ian Scott tells Tildy Bayar
Artist’s rendition of a Stable Salt Reactor installation
Credit: Moltex Energy
Small modular nuclear reactors, or SMRs, are a hot topic these days, with multiple companies working on designs and a number of countries, from China to the UK to Iran, considering how SMRs might be used in a low-carbon energy mix. In addition to generating power, SMRs can be used to produce high-temperature process heat for industry and district heating purposes.
But which type of SMR will win in the end, in terms of economics, safety and deployability? The main types of SMRs being developed around the world include light water reactors (LWR), fast neutron reactors (FNR), graphite-moderated high temperature reactors (HTR) and molten salt reactors (MSR). Of the latter, a number of companies are currently working on developing a molten salt-based SMR.
In a 2015 study, consultancy Energy Process Developments (EPD) evaluated molten salt reactor (MSR) technologies in terms of their potential for implementation in the UK. The study evaluated six MSR firms which “claim they are ready now with proposals for the next step to implementation, namely engineering design to prepare the safety case and to proceed to design and build”. In comparing MSR technologies with industry-standard solid-fuelled technology, EPD found that “the outcomes reflect favourably on MSR technology.” In addition, EPD concluded that, among the technologies viewed as “credible”, one design “has emerged as most suitable for UK implementation”.
This design is the Stable Salt Reactor (SSR) from UK-based Moltex Energy, which also features in the World Nuclear Association’s list (updated in September) of 18 small (25 MW and up) modular reactors “at earlier design stages”. And Moltex is one of 33 companies shortlisted in the UK’s design competition to build the first commercial SMRs by the end of the 2020s.
The main radioactive products of Moltex’s Stable Salt Reactor (SSR) are stable immobile salts instead of high pressure gases. While this technology has been around since the 1950s, Dr Ian Scott, Moltex CTO and inventor of the SSR, says it is “not an exaggeration to say that [the SSR concept] is the first actually new nuclear technology for at least three to four decades”.
Scott believes that the SSR design fully addresses the central concerns of today’s nuclear sector: safety, cost and waste disposal.
Safety and licensing
One area where the SSR can beat conventional SMRs, Scott believes, is safety. While safety regulations have not yet been established for SMRs, in March the International Atomic Energy Agency (IAEA) hosted the world’s first SMR workshop for regulators, with a view to “establish[ing] a set of clear and pragmatic requirements for safety and licensing”.
“Though smaller, the safety and security measures for this next generation of nuclear power reactors are no different from the international obligations that present-day reactors are subject to,” said Stewart Magruder, a senior nuclear safety officer at the IAEA. And Greg Rzentkowski, IAEA Director of the Division of Nuclear Installation Safety, said the safety measures are “likely to include the development of an overarching safety objective and a guidance document on establishing relevant requirements in accordance with the facility type and size.”
Scott explains the safety advantage of the SSR in terms of its fuel: “The fuel system in all conventional reactors is a solid. In most cases it’s a solid based on uranium oxide. It’s ceramic, and has the properties of a dinner plate – like very strong china. It’s strong, it’s stable, but it has one really major drawback: when fission takes place in it, uranium atoms are split in two and produce two fission products, both of which are half the weight of the original uranium.
“Some products are gases under the conditions of the reactor,” he continues. “Two of those gases are cesium and iodine; these gases are extremely radioactive and extremely dangerous if absorbed in the human body. At Chernobyl, at Fukushima, the hazardous radioactivity that was widely released was cesium and iodine. At Chernobyl, most of the cesium and iodine in the core left the reactor site and spread across western Europe, while the plutonium and strontium stayed where they were as solids or liquids. That’s the problem with uranium oxide fuel: you build up gases to incredibly high pressures, they’re held in so long as the pellet doesn’t overheat, but if the fuel pellet heats above a certain temperature long before it melts down, these gases are released explosively. And, if the fuel containment fails, you have a real catastrophe. That’s the Achilles’ heel of solid fuel.
“The molten salt fuel is uranium chloride. The beauty of it is that the cesium and iodine are not gases, but form salts themselves, and these salts don’t evaporate. So the really dangerous fission products are chemically locked in the fuel and don’t have a large gaseous potential for release.”
In addition to the risk of a nuclear accident, EPD notes that all civilian nuclear power plants are now treated as being at risk from a terrorist attack. But the consultancy says MSRs’ small footprint allows the reactors to be placed in holes in the ground, offering better protection from both air attacks and ground-based explosives, as well as ensuring that any leaking container will be isolated. And, in the event of an attack, EPD says “these reactors are not inevitably going to spread radioactive material into the atmosphere”. Furthermore, it says the MSR offers a low risk of its fuel being weaponized.
In a 2015 report on licensing and regulation for SMRs, the World Nuclear Association (WNA) noted that the technology has the potential to significantly reduce costs within the sector, leading to vastly expanded market potential. Because SMRs could be mass-produced in factories and installed modularly, the level of construction quality and efficiency could be boosted. This increased construction efficiency, as well as increased safety and smaller size, could mean easier financing for projects, while economies achieved through serial production could reduce costs further. Thus, SMRs could expand the market range of nuclear power to include countries and regions that can’t currently afford it.
Scott, though, is sceptical about the cost benefits of conventional SMRs. “I see nothing at all to convince me that a small modular PWR will cost less per kW than does a large PWR,” he says, citing a 2014 SMR feasibility study from the UK’s National Nuclear Laboratory, which considered four smaller PWR designs in terms of both capital cost and LCOE.
The study found the SMRs to be competitive with large-scale nuclear power on a first-of-a-kind basis and, given the use of modular construction and factory production, “conceivably more competitive” on a next-of-a-kind basis. “Large-scale reactors require significant up-front capital investment and long costly construction; by contrast SMRs will require less capital investment before producing returns and have the potential for quicker construction,” the study concluded.
Of the full-sized PWR designs, Scott says “the best is the AP1000 from Westinghouse – it has an elegant design, and is also the most compact.” But he notes that, “for a given power, the SSR is one-fifth the volume of the AP1000. That has an enormous impact on the cost – the amount of concrete and steel drops dramatically.”
“Individually, [SMRs] would be a lot cheaper [than large reactors] because they would also be a lot smaller,” he adds. “But in terms of economics, that doesn’t really help you very much. The hope is that the economics of large numbers [of SMRs produced] will offset the loss of economy of scale. There is a fair chance that it will, and that it will lead to a cost comparable to current PWRs. But the technology is not economic without subsidy.”
Scott also believes that safety concerns incur plant operational costs that early stage reactor designs fail to take into account, and that cost estimates for many nuclear projects are thus too low. The capital cost involved in building nuclear reactors has steadily risen due to post-Fukushima safety concerns, which require increasing engineering complexity to address. This results in having to build very large conventional reactors in order to gain the economies of scale necessary for commercial viability.
“The way the cost estimate is done at the early stages of projects,” Scott says, “is through taking account of nuclear construction factors. You take the cost of the materials needed to build something and multiply it by the construction factor: how much more does it cost to build it than to buy the raw materials,” he says. “For a bridge, a tunnel, an automotive production line, the construction factor would be 2 to 2.5. In nuclear, for anything which is associated with the safety system – which is pretty much everything – the construction factor is 10. That is the construction factor used in costing the SSR. While modular construction may reduce these costs, we gave no credit for that in our cost estimates.” He also notes that, in building the SSR, Moltex would be able to use “a lot of the existing nuclear supply chain”.
“The construction cost [in the case of the SSR] of a 1 GW unit would be £1.3 billion ($2 billion),” says Scott. “The biggest additional cost is the cost of the regulatory process,” which he estimates at “in the region of $100 million”.
One surprising aspect of the cost reduction equation for the SSR is the steam turbine. “PWRs and so on produce quite low-temperature heat,” Scott explains, “around 300˚C, whereas if you look at the steam turbine in a CCGT or coal plant, they all produce steam at around 600˚C. A higher temperature equals higher efficiency – but what I didn’t know until we got our quotations back from suppliers was that it also substantially reduces the capital costs of the steam system.
“Nobody [in the power sector] uses large, very inefficient low-temperature steam turbines, except the nuclear industry which has to. Steam turbines in the conventional nuclear industry are much bigger, and are also only produced in very small numbers. They’re almost bespoke; in Hinkley they are bespoke. That’s one reason they’re so expensive.”
And there is the issue of plant security, which also adds cost. According to Moltex, no conventional SMR design is likely to be economically viable as a stand-alone reactor, because the fixed cost of the necessary security for any nuclear reactor will be unaffordable for a single small reactor. This means that multiple reactors would need to be located at a single site to save on security costs, which would add to the operational complexities. The SSR design, Scott says, while it is modular, can be used to construct single reactors ranging from 300 MW to 1200 MW, achieving both the economies of modular construction and the economies of scale of large reactors.
Dealing with waste
Whether nuclear waste materials should be stored or reprocessed is a heavily debated issue from both the safety and cost angles. In a 2003 report, Harvard University’s Belfer Centre for Science and International Affairs concluded that reprocessing and recycling plutonium in existing LWRs “will be more expensive than direct disposal of spent fuel until the uranium price reaches over $360/kg, a price that is not likely to be seen for many decades, if then”, and could lead to as much as an 80 per cent increase in the costs associated with spent fuel management.
Nuclear waste disposal is another area where Scott feels the SSR is more competitive than conventional SMRs. “Reprocessing nuclear fuel as we do at Sellafield [the UK’s nuclear fuel reprocessing plant] and at various places in France is a hideously expensive process,” he says, and is “absolutely not economic. Producing fuel for conventional reactors [including SMRs] from that reprocessed plutonium is also not economic. It’s happening because the government wants to do it, not because the market says it is sensible.
The Stable Salt Reactor
Credit: Moltex Energy
“The economically sensible thing to do with spent fuel is to bury it in a very deep hole,” he adds, “but even that is expensive.” He cites an estimate from the UK’s (now defunct) Department of Energy and Climate Change (DECC) which put the cost of burying a single fuel assembly in a deep geological repository at £350,000.
However, the SSR “burns up all the long-lived components of spent fuel,” he notes. “We’ll still have a radioactive waste stream, but it only stays dangerous for 300 years rather than 300,000 years. The amount of fission product waste is essentially the same as conventional [SMRs], but the volume of waste from this reactor is probably one-twentieth of the volume of waste from current [large] reactors.”
And he says that, because the SSR’s molten salt fuel doesn’t require high-purity plutonium – “The plutonium we’d use can be only 50% pure” – reprocessing existing spent fuel to make molten salt fuel could be comparable in cost to, or even cheaper than, mining and enriching uranium and would “put fuel reprocessing on a commercial footing for the very first time”.
A successful SSR rollout, Scott believes, would see “one-third to one-half of global electricity production coming from [SSRs], a 20- to 30-fold increase in our nuclear fleet, with most of the remainder from renewables. This would give us a completely carbon-free electric power system. But it would produce a lot more radioactive waste, which is why the ability to process to get rid of the really dangerous stuff is more important.
“We wouldn’t store our initial spent fuel for very long. In general, it takes 10 to 20 years” before reprocessing can begin, but with the SSR, “you just have to wait a year”. In terms of storage, Scott says the process for “the waste we’re still producing would be similar to what we use now: a process called vitrification, which turns [the liquid waste] into a big lump of resistant glass.” This material, he says, can be stored in secure warehouses or in shallow underground repositories, and can – after a few hundred years – be treated as lower-level waste.
Fuel breeding, or generating more fuel during the nuclear fission process than is consumed, is another area where the SSR could offer benefits. According to the WNA, current global uranium use stands at around 66,000 tonnes per year, with a measured resource of 5.9 Mt. With costs around 1.5 times current spot prices and with use only in conventional reactors, the WNA estimates that this resource is sufficient for around 90 years.
While Moltex’s first reactor design is not configured to breed new fuel from thorium, which is more abundant than uranium, Scott says the basic technology is suitable for this process, which would “sustain energy production for thousands of years”.
Potential for energy storage
Another advantage Scott points to is the SSR’s potential to be used with heat storage. He notes that, for concentrated solar power (CSP) technology, the only “glimmer of hope of making that technology economic is the fact that they can store the heat in molten salt heat stores. So they can sell the electricity when it makes the most money for them, which might make them economically viable, but they’re not there yet. However, the technology they use to store the heat, which is a molten salt heat store system, will work with nuclear energy provided the nuclear reactor has an output temp of around 600˚C.
“This is unusable with a PWR or a boiling water reactor, or any of the current generation,” he says, as their output temperatures are too low. But “some MSRs, including ours, do have output temperatures that high. That gives us the ability to use an energy storage system. We can store heat output for as much as eight hours, produce no electricity at all, then for eight hours we can use both the reactor’s output and the stored heat to produce 2 GW on a 1 GW output.”
And this aspect of the SSR is “very, very compatible with renewable energy,” he says. “You can double the power output when demand is high and renewables aren’t working, selling your power when it’s worth a lot more.
“With molten salt heat storage on megawatt-scale reactors, the LCOE goes up by approximately $10/MWh, from $45 to about $55. The cost per kW of storage capacity from, say, lithium-ion batteries – even the [35 GWh Tesla] Gigafactory – is many times higher than that.”
He notes that “the costs I’m giving you are pessimistic because they’re based on real power stations, with capacities in the 100 MW-300 MW region. On a 1 GW scale there are very big economies of scale to be had in molten salt heat storage.”
SMRs in the UK and Canada
In the UK, the Energy Technologies Institute (ETI) has recently published a report titled Preparing for the deployment of a UK small modular reactor by 2030. The report found that “a credible integrated schedule demonstrates the potential for a UK SMR operating by 2030” provided the government takes steps to introduce a risk-reducing policy framework.
In March, the UK announced a £250 million investment in nuclear R&D, as well as a competition which aims to “identify the best value SMR design for the UK,” according to DECC. The new Department for Business, Energy and Industrial Strategy (BEIS) is carrying on the competition and aims to develop an SMR roadmap.
Wherever it is in the world, Scott says, conventional nuclear power simply “cannot compete on price” and its global market penetration is doomed to be limited. He adds that “the cost in the UK is probably as bad as it gets,” which is why nuclear power in the UK “is only being built if it’s got major subsidy”.
He says the LCOE for the SSR is “a third of coal, gas and conventional nuclear: $45/MWh [for the SSR] vs $140 for coal, $125 for gas, $145 for nuclear. If we assume that we can build an SSR in China for the same ratio of cost of a conventional reactor in China to the UK, then we are cheaper than coal in China.
“With sensible cost assumptions, mostly driven by our capital cost, we can compete everywhere with everything, with the possible exception of renewables.”
According to Scott, the UK and Canada “are probably the best places in the world to license a new reactor technology” due to the “technology blindness” of the regulatory process, unlike the US where “the rules are written for PWRs”. Moltex’s development process is currently focused on building an SSR in Canada.
“We are pursuing development in both Canada and the UK at once,” he says, “but the reality is that the UK nuclear regulator is seriously under-resourced for the work it’s having to do. The Canadian regulator has about 1200 employees, so they’re open for business.
“We’ve also had very strong encouragement from Canadian nuclear laboratories who said ‘We’d love to build your prototype on our site’, so they’re offering a site, easy access to the regulator and a generally positive North American can-do attitude.
“However, the UK wants to develop its nuclear technology base, the government has put £1.4 billion aside to support that, and we would be delighted if we got support for it in the UK.”