16/01/18 – The Prospects for Small Modular Reactors

Speakers: Alan Woods, Rolls-Royce, Director, Nuclear Strategy and Development; David Powell, Vice President, GE Hitachi; Tyrone Sparkes, Director of UK Operations, NuScale Power

16 January 2018 – meeting notes

All-Party Parliamentary Group on Energy Costs

The Prospects for Small Modular Reactors

Chair: Lord Haworth

Chair’s Opening remarks

Good evening everybody. I’d like to extend a warm welcome to you all to this, the 39th meeting of the All-Party Parliamentary Group on Energy.

I am Lord Haworth, a Labour peer and vice-Chair of this Group. Our meeting this evening is to discuss small modular nuclear reactors.

Nuclear power is the energy technology that has never quite lived up to its promise. While most technologies are expected to become cheaper as they mature and achieve larger scale, from a Western perspective nuclear seems to have done the opposite.

In some parts of the world large nuclear plants appear to be built reasonably quickly and cheaply, for example in Russia and China. However, in Europe this appears to be impossible.

There are many reasons why this has been the case:

  • The hostility of the public and politicians alike
  • Planning processes that can take decades
  • Loss of the skills base
  • The seemingly insatiable desire of regulators in each national jurisdiction to impose their own demands on plant design
  • Ever increasing safety requirements and demands for perfection in manufacturing

all of which create enormously complex designs and project programmes and increase commercial risk.

Many of the latest designs promised to address these problems with passive safety, but the recent experience of developing and building new nuclear in both Europe and the USA gives little cause for optimism.

We are here this evening to discuss the prospects for Small Modular Reactors as a solution to the nuclear conundrum.

And we have three excellent speakers who can take us through the issues. I have asked them each for no more than 7 minutes opening remarks after which we will go on to our normal Q&A.

Alan Woods, Rolls-Royce

I’m pleased to be here tonight to speak on the challenge of why small modular reactors (SMRs) can address the challenge that the nuclear industry has got today, which in the west is fundamentally one of cost.

As installations have got larger and larger and the capital costs have become overwhelming, so has the complexity of such projects e.g. Hinckley Point has taken three governments. We believe it’s an unsustainable position and we need to reverse the trend to go larger, and we think SMRs can be the solution to bringing projects back to being more manageable.

The key thing is that SMRs must be commercially investable, and we’ve designed our power stations to be so, that is, financially investable without the need for government intervention.

How do we reverse the trend? The economies of scale argument, we think, is somewhat flawed. It’s based on the fact that the operational costs of a nuclear power plant do not scale as power drops, but actually the biggest sensitivity in the construction of a power plant in the cost of electricity terms is in the financing cost, which must be brought down. There are three aspects to the financing cost: the build up of the capital, the time that the capital is tied up, and the programme risks, and we need to address all three areas.

We have designed our SMR for each of the stages of the life cycle so that we can reduce those three things. We’ve not introduced technology and innovation for its own sake, but for where it has advantages and benefits in these areas.

We have designed it for manufacturing, meaning that we have taken activity off the site, where it is expensive and at the mercy of weather conditions, and put it into factories by using the modular approach. Modules are designed to be road-transportable. We limit the expense that goes into the facilities for the modules, because the bigger and heavier the module, the higher the cost of the facility, so we keep within size limits that don’t put too much cost burden on the supply chain.

We are building modules which bring in commonality in terms of manufacturing processes, using machinery to manufacture them which facilitates a production line approach. Doing that we can significantly reduce the cost of plant and machinery which goes into the power station.

Equally important is that all our modules are common across all power stations, wherever we build it. We’re not in a situation where every site is different depending on site characteristics.  Our plant is designed to have a very small footprint, which allows us to mount it on a seismic raft which insulates all the plant and machinery from the conditions on which it sits, so no matter where the plant is sited we can have common modules

In minimising construction costs we have to look at the risk profile of construction, and that is driven by the timing so it’s a principle that we minimise the time taken to construct. The modularisation approach helps that as modules can be manufactured and installed in a two year period and the site where we have seismic raft built is another two year period.

The construction aspect is also de-risked because it’s a much lighter plant, the depth of piling is less and there’s a smaller footprint, so it’s cheaper.

We can break up the way we finance, from the lower risk above-ground aspects to the higher-risk below-ground ones. That means that we can potentially attract a different level of financing cost – we don’t have to attribute the highest-risk financing cost to the whole power station, and that’s a really important aspect.

We design for operation. Utility customers aren’t interested in the technology: they’re interested in a plant that will deliver electricity at an affordable rate that can be guaranteed. They don’t want risk built into the operation. So we design for reduced maintenance periods and for certainty of operation, which is where we’re introducing digital aspects to improve operational efficiency and reduce maintenance periods.

Finally, we have to design for decommissioning. The modular construction helps this because deconstruction is just the reverse of construction. When we say modular, we mean that everything above the seismic raft is modular, including the civil structures, so all of the civil structures are also manufactured in factories which enables us to considerably reduce the overall construction and deconstruction time.

In summary, our plant is designed around being economically viable and commercially investable. That means that our approach is aimed at reducing not only the capital cost but most importantly, the financing cost and ease of financing for our utility customers. We do not want this project to rely on governments for construction.

David Powell, GE Hitachi

GE and Hitachi have worked together on nuclear for over 60 years now, and so have quite an enviable track record in nuclear reactors. Before I talk about the work we’re doing on the small modular reactors, I just want to say that we believe in both big and small reactors, and we see both types are necessary in the battle to combat climate change.

As a little bit of background, GE has been active in the UK for well over 100 years, with 18000 employees based here. We are heavily engaged in Anglesey on the Horizon nuclear project, which would involve building two of our big reactors of 1350 MW size. They’ve been through a generic design assessment process recently, which was completed by the UK Regulator. The design if based on experience from Japan where the time taken was 39 months from first concrete to full outout. This demonstrates that nuclear, and big reactors, can be built to that timeframe of just over three years. We’ve seen similar projects going on around the world, showing a track record of building those big ones.

The way we have approached the small modular market is that for a long time we didn’t  invest anything into small reactors because we had a next generation technology that was focussed around how you reduce the liabilities of nuclear. This is called our Prism reactor, which is a fast-reactor technology aimed at the specific application of taking irradiated nuclear fuel from conventional reactors, and recycling that fuel and generating electricity.

That type of reactor can burn up those high level waste components that you have in that used nuclear fuel, and end up with a waste that only needs 300 years storage, so avoiding that long-term storage issue in the future.

So for years, developing that technology was the focus of GE Hitachi. Then we approached our customers and asked them what they wanted from nuclear reactors in the future, and we were told, lower cost generation and lower capital investment. As Alan said earlier, if you can get the capital cost down then the cost of generation comes down too.

So over the last 18 months or so we’ve been developing SMRs, based on our Boiling Water Reactor (BWR) technology. They use the same safety systems as our big reactors, but our target is to get to a capital cost of around $2000 per KW, which translates to around $40 per MW hour.

Our design approach has been to use the same safety system but to think differently about how we do it, by simplifying the reactor design (going from 1300/1400 MW down to about 300 MW) and eliminating the unnecessary systems.

So that’s really where we are, focussed around our small reactors for two particular applications: 1) lowering cost of nuclear and 2) reducing liability costs, and then we have our big reactors for when we have countries like the UK where there are very good grid systems. We’re focussing around using a proven modular build system that we’ve used successfully in Japan to build our big reactors, and adapting it to suit smaller reactor designs.

Tyrone Sparkes, NuScale

We’re a leading US-based nuclear technology developer and we’re at an advanced stage of bringing an SMR to market. In the US we’ve submitted the design and it’s been under review for about a year and approvals are starting to come through. Earlier this year the safety aspects were approved, and we’ve got strong backing from the US Dept. of Energy. We’re also the only US developer to have a planned project that we’re working on, together with other planned projects for 2026-2030.

In terms of the attributes of SMRs, all that has been said before applies to NuScale, e.g. the modular aspects which bring the cost down, construction safer, quality has improved and you have more control over your variables.

The NuScale approach was to start with a blank sheet and look at the transport options for the modules. That gave us the outer dimensions of the power module and we then reverse engineered and reached a thermal output of 160 MW which produces 50 MW out.

Our standard design takes 12 of these 50 MW units, sticks them in a building and that results in an overall output of 600 MW. The benefits of doing this are twofold: 1. Financial – after the first one is in, you start to generate revenue while you are building the rest. You can add modules bit by bit which allows customers to manage cash flow, and 2. with small modules like this we also have phenomenal load following capability – you can not only vary the output on a reactor, but you actually have 50 MW of dispatchable load at your disposal, which is a great opportunity for grid operators as it gives them a lot of flexibility.

Here in the UK we see great opportunities, as there is a strong nuclear heritage here, there is the manufacturing capability, a good regulator, and a strong grid. All of these make the UK a really attractive market for us to get involved in.

Questions and Comments

Damir Ahmovic, Alpha Energy: (to Alan Woods) What is the life cycle of an SMR, and what is the strike price required for pay back?

Alan Woods: In our design, 60 years and just short of £60 per MW hour based on a 50/50 debt to equity ratio. You should note that about 30% is financing costs, so the more we reduce the timeframe of the construction, the lower cost will be. That’s amortised over 30 years.

Chris Harris, Npower: I get that this needs to work without finance but you still have got the long-term energy price risk, and neither government nor supply companies have been well-placed to take on that risk. What’s your economic model for energy price risk long-term contract?

Alan Woods: As consultants we lean heavily on our customers for their views on pricing, and our current forecast (£60 per MW hour at 2016 rates) is seen as hugely competitive by them on the basis that nuclear does not suffer the same cost volatility as some other forms, such as gas or other fossil fuels. There’s certainty there in terms of the generating cost, which provides them with comfort and confidence.

That was the first thing. The second point is that, in order to overcome the significant requirement for capital (nowhere near the large reactors, but still significant) some form of contract for difference from the Government is needed, at least for the early years. This is different from an equity or cash injection from the Government. Also, for the consumer, £60 per MW hour represents much better value that £100 per MW hour, which it was for other nuclear.

David Powell: Our challenge from our utility customers is how to get the cost down to where gas is, so that we are competing in that marketplace, with low-carbon electricity. This was driven mostly by our US customers and the current situation there is that there is very little nuclear investment going on, so when we ask “What would help you get there?” that was the target they gave to us.

As Alan mentioned, you have to go through a first-of-a-kind cost. In any country you have to go through a regulatory approach. So with our advanced boiling water reactor, we’ve licensed that in four countries now, and this has taken 4 or 5 years each time. So, whichever country we go into first to build our SMR, we will have to go through that licensing process. That will give us a really good indication of what the regulator thinks of it and what design changes might be needed. On the basis of that we would create some kind of mechanism to guarantee price of construction, but the first one is always the difficult one to build. After two or three you know what the costs will be.

Alan Woods: The critical point is certainly in construction

Tyrone Sparkes: NuScale did a costing, extracted from a 3D model, and we came out with a cost (when converted from dollars) of just under £60 per MW hour. But that’s at first-of-a-kind: one of the benefits is when you repeat you get better, and there should be up to a 10% reduction after 12 or so.

The other thing, in order to get these future savings you have to have an order book to build multiple stations.

Jack Simmonds, Fuellers and SONE: What sort of approaches have you had so far?

Alan Woods: We’ve had interest from utilities wishing to build in the UK and also interest from countries where affordability is an issue: where they wish to build nuclear power for energy security reasons but don’t have money and so are probably reliant on the Russians or the Chinese. We’ve had interest from countries where there is existing nuclear as well as those wishing to introduce it, in some cases where they don’t have the grid infrastructure to support a single large reactor and so are more interested in SMRs. We’ve also had approaches from countries who wish to take a participatory approach in both manufacturing and investment.

David Powell: Taking our two SMRs that there are different applications, for our BSP which is our Boiling Water SMR, we’ve had a lot of interest from US utilities. It’s a relatively recent development for us (just over 12 months) so we haven’t gone fully international on this, although we’ve talked to the UK Government about it and we’ve started to get a lot more interest in the UK from both a build point of view and a supply chain point of view. Because it’s small, it could be mostly built in the UK if we decide to build it here. We’re in a great position with Hitachi investing in the Horizon project over two sites in Anglesey and Oldbury, and those are the sites where we could build our BSPs in the future. Understandably, all the focus is going on fairly big reactors at the moment.

Our other type of reactor is our PRISM reactor (which uses spent fuel). By the time processing finishes there will be a UK stockpile of 140 tonnes of plutonium at Sellafield, with no clear strategy for what to do with it. We’ve been talking to the Nuclear Decommissioning Authority for the past six years about this as it’s a huge UK asset, if the Government decides it wants to go down that path.  We think if we build 12 Prism reactors as SMRs, each about 300 MW,  you could generate 15% of energy for 50 years plus, and they would all be built in the UK.

Ashatosh Shastri: Prism is a fast reactor? Is the UK prepared for that kind of closed cycle processing?

David Powell: The UK has been reprocessing for the last 30 or 40 years, but that will end in 2/3 years. Hinckley and Horizon waste won’t be reprocessed. Prism represents the closing of the cycle.

Ashatosh Shastri: Is Prism at the same level as its international competitors?

David Powell: Yes, based on a US programme over 30-40 years developing a small experimental fast breeder reactor number 2, which demonstrates that this can work. It’s a walk away reactor and Prism is the same concept. There will be no obstacles to getting the licence but the Government needs to commit to going down that path.

Tyrone Sparkes: At NuScale we have our first client project in the US that we’re working towards. The company is in the mid-west and originally, coal-fired power stations were the norm. The company looked at alternatives, for example gas, although that would mean bringing gas in. In the end they assessed our design and looked at the pricing and agreed to go with us. They’re one of multiple western states who are interested and keen to see how first one goes so they can then all follow suit

One of the things that the US Government has recognised is that there are a lot of first mover risks, and developers and investors are obviously very cautious. The Government acknowledges that they are going to have to support the development of SMRs and we are receiving $217M from them: but our client is also receiving $31M from the Government for site licensing activities.

David Rose, Fuellers:  What could go wrong? Rolls-Royce have been building for 60 years but what do you say to the sceptics?

David Powell: We’ve had 60 years of learning experience. If you look at our track record of building reactors, with the first four that were built in Japan with the advanced boiling water reactor type being built in that 39/40 month timeframe, this demonstrates that you can achieve a consistency with building if you take a modular approach.

The issue we tackled was construction risk and removing as much as possible of that risk by going modular helps to bring the price down.

Safety-wise we’ve learned a lot from Chernobyl, Fukishima, Three Mile Island, etc. We are currently going through very rigorous licensing processes but there is now a huge amount of experience out there. The big issue is public acceptance, and maintaining it through independent regulation, is going to be key.

The construction risk will always be there, but we need to build the first one and move on from there.

Alan Woods: Scepticism is higher in the UK than elsewhere. Look where there’s been successful fleets built and you can plot the trajectory of the cost coming down. The French, the Japanese and the Russians have done it. The obvious point is that we can say start building in volume and you can bring the cost down. We’ve run some polling and a lot of the scepticism is around cost. The other issue in the UK is we don’t have a national utility and therefore we have to start from a capital perspective. We have to remove those risks that are inherent in large reactor designs by modularisation, and reduce other risks associated with the construction time-frame by moving activity into factories and taking it off-site. We can do a lot to address the cost and schedule risks that hamper one-off builds which is what leads to scepticism.

If we take the view that we stick with incredibly highly capital intensive large one-offs, then scepticism will prevail.

Alice Barrs, RWE: An increasing concern of Government is making best use of water, so can you explain whether SMRs contribute to meeting that objective in any way?

Tyrone Sparkes: Thermal output will dictate the amount of water used for cooling, so smaller plants will use less water and therefore be less wasteful, but the ratio is still more or less the same.  Smaller also means that there are other cooling options, e.g. dry cooling, although with that method you take a knock of 7% because you use some power for fans.

David Lewis: You’ve talked about N of a kind. Is there a number of projects built where you think you might get to a point where other governments and regulators in other countries accept that they don’t need to go through the process again?

David Powell: Each regulator will take a nationalist approach. In our experience they have a different approach in how they go through their own regulatory processes. For example in the US they are very prescriptive whereas they UK isn’t: they’ll say, “tell me why it is safe?, and why you can’t get it any safer?” which is challenging. It would be nice to get to a harmonised regulatory approach.

Alan Woods: There is an analogy with the aerospace industry, where they have a harmonised regulatory body. The IAEA is pushing for more harmony and more commonality but it’s a long way from achieving that. There’ll always be different technology types and different vendors and new vendors.

Tyrone Sparkes: In addition to the IAEA, Governments are starting to realise that harmonisation needs to be a priority if they want to see successful deployments. It’s a significant cost and schedule component, and one of the major risks. I think what I can see in the future is that if certain aspects have been reviewed by other regulators then some overlap could be done and that would be beneficial.

Wade Allison, Oxford SONE: As well as history of different kinds of reactors, we’ve ended up with out of sight, out of mind, with reactors dotted around the coast. Now in this new generation personally I think it’s regrettable that Hinckley and Wylfa are still out on the edge. I hope new smaller reactors can be close to population centres and industry where the waste heat could be used. Local reactors associated with a town could be identified as “our” power station.

Alan Woods: I agree, SMRs could be sited differently. We have existing licensed sites

and broadly have supportive populations already, so that’s where we will go first. But it’s absolutely feasible and makes absolute sense to start using the residual heat, and that would also drive down the cost.

David Powell: I agree, and if you look at Wylfa – the local support is tremendous. In North Wales the Government is extremely supportive of developing that whole area, and Bangor University is very keen to get involved in the training aspects. When those reactors come online in the mid 2020s, the current teenagers could be nuclear operators, which is why local communities see it as important.

Wade Allison, Oxford SONE: Employment opportunities are quite rapidly changing.

Alan Woods: With SMRs you don’t have to build a new town to house the construction workers because they’re fabricated in factories. The employment opportunities are much more about steady long-term jobs and careers: not construction workers but skilled nuclear operators. With SMRs we avoid that influx of workers over the 6 or 7 years construction period.

Wade Allison, Oxford SONE: Can the outage period be more evenly spread?

Alan Woods: Yes, the advantage of multiple small units is that you can plan your outage period so that you get a constant flow of demand from that resource. With our SMRs, we’re aiming to reduce the outage period significantly anyway.

Tyrone Sparkes: You wouldn’t take all 12 through the cycle at the same time. We don’t have a refuelling team. Every two months refuelling happens as part the regular maintenance cycle.

The meeting closed at 7 pm