Can Nuclear Energy Scale? Tony Roulstone on SMRs, Cost, and Repeatability

Nuclear energy is having another moment. But enthusiasm alone will not create a nuclear renaissance.

The bigger question is whether the industry can finally learn to deliver complex projects predictably, repeatedly, on time, and at a cost investors and utilities can accept.

In this episode of Green Giants: Titans of Renewable Energy, host Wes Ashworth, President of Lee Group Search, sits down with Tony Roulstone, Lecturer in Nuclear Engineering at the University of Cambridge and former Group Managing Director of Rolls-Royce Nuclear Engineering.

Tony brings decades of experience spanning nuclear energy, aerospace, complex engineering programs, business transformation, and nuclear education. That perspective leads to a fascinating argument: the biggest breakthroughs nuclear needs may have less to do with reactor physics and more to do with how reactors are manufactured, financed, constructed, and repeated.

The conversation begins with a Rolls-Royce aerospace failure that left Boeing 747s waiting for engines. The technology problem ultimately exposed a deeper organizational problem. Tony explains why that lesson still matters for nuclear projects today.

From there, Wes and Tony examine one of nuclear energy's fundamental challenges. Building a large reactor can resemble manufacturing an aircraft in a field. Thousands of people, intricate supply chains, enormous documentation requirements, specialized inspections, and long construction timelines make repetition and productivity incredibly difficult.

That is where small modular reactors, or SMRs, could change the equation.

Tony explains why smaller does not automatically mean cheaper and why the real economic opportunity behind SMRs depends on three principles:

• Standardization: Stop redesigning the product for every customer and site.
• Modularization: Move significantly more construction into controlled factory environments.
• Production learning: Build repeatedly enough for teams, suppliers, and processes to improve from one unit to the next.

The discussion also explores advanced nuclear reactors, including high-temperature gas reactors, liquid-metal systems, and molten salt concepts, along with what new reactor technologies could mean for safety and industrial applications.

Tony and Wes also confront nuclear's most persistent public challenge: trust. They examine lessons from Chernobyl, the importance of organizational safety culture, and why modern nuclear safety cannot be understood solely through the industry's historical accidents.

Finally, they tackle the false choice between nuclear energy and renewable energy. Deep decarbonization will require an entire energy system, including electricity, industrial heat, transportation, storage, renewables, and firm low-carbon generation. The real question is not which technology wins. It is how these resources work together.

In this episode:

• Why technically brilliant projects can still fail
• What aerospace can teach nuclear about systems engineering
• Why large nuclear construction struggles with productivity
• The real economics behind small modular reactors
• Why SMRs require fleet deployment and repeatable production
• Standardization, modularization, and manufacturing learning
• Advanced nuclear reactor technologies and their potential
• Nuclear safety, Chernobyl, and public confidence
• Private capital and the growing nuclear ecosystem
• Why nuclear and renewables should not be treated as opposing camps
• What nuclear must accomplish by 2035 to prove it can truly scale

The ultimate test is simple: nuclear must become predictable.

If the industry can turn extraordinary engineering projects into repeatable industrial products that arrive when promised and near the cost promised, nuclear could play a much larger role in the global energy transition.

Links:
Tony Roulstone on LinkedIn

Wes Ashworth: https://www.linkedin.com/in/weslgs/

• Email: wes@leegroupsearch.com
• https://leegroupsearch.com/green-giants-podcast/
• https://leegroupsearch.com/

Transcript

Wes Ashworth (00:25)

Welcome back to Green Giants, Titans of Renewable Energy. Our guest today is Tony Roulstone Tony has spent his career inside some of the most complex engineering environments in the world. He has worked across UK Atomic Energy Authority, Rolls-Royce's nuclear and aerospace businesses, major engineering program recovery, corporate transformation, consulting, and nuclear education at the University of Cambridge. And before we begin, you don't need a background in nuclear engineering to follow this conversation. We're going to explore three things.

First, why brilliant technology can still fail when the organization and delivery system around it break down. Second, whether small modular reactors can make nuclear more affordable by turning nuclear from a one-off construction project into a repeatable industrial product. And third, what would need to change in financing, manufacturing, workforce, safety, and public trust for nuclear to scale? At the center of all three is one question.

Can nuclear become an industry that repeats success instead of rebuilding the entire system for every new project? With that, Tony, welcome to the show.

Tony Roulstone (01:25)

Thank you very much.

Wes Ashworth (01:26)

It's a pleasure to have you. I know this is a very hot topic currently. And your career does give us that sort of rare view across engineering, aerospace, nuclear energy, and business transformation. And I want to begin with what those environments taught you about the difference between invention and execution. And so, what separates a clever technology from one people can really depend on?

Tony Roulstone (01:47)

it is an exciting time in nuclear. We thought we had identified a nuclear renaissance 15 years ago, and it wasn't there.

But there is one, there's lot going on now. So it's great to able to talk about these things, both from a perspective of what's happening in the US and what's happening in Europe. I spent a dozen years in the Army in a very practical work. I was trained to fix helicopters and I sent me to a tank regiment and it was about fixing stuff very hands on. And when I came out of the Army, I joined the Atomic Energy Authority and then Rolls-Royce. And I learned that actually

engineering is very much more about theory and modeling and then practice and testing. And the thing that I think you're asking about is one could have all sorts of bright ideas, there's lots of bright ideas around the moment. What distinguishes a bright idea from something that's going to last and work? And I think it is this combination of modeling and testing. It's really at the heart of what Rolls-Royce does is its engineering. It has very good engineers who get in the computers and model stuff, but they always go

there and test beds and destroy engines and look at them and look at the effect of it. It's a combination of these two things which makes something that's reliable and can be dependable and then you can then franchise it out and get lots of people to do it. These two things need to go hand in hand.

Wes Ashworth (03:02)

with and

we'll dig into those two things a little bit more as we go, but you know, it's that idea, like proving something can work is very different from proving that people can really depend on it every day and kind of go into that execution. And you brought up Rolls-Royce. I wanna talk about an example, not necessarily nuclear, but that I think connect and will tie it in here. But at Rolls-Royce, a failure to deliver the FADEC control system left engines unavailable and Boeing 747 sitting in Seattle.

tell us a little about like what had gone wrong and what did you discover like once you got inside the problem?

Tony Roulstone (03:32)

So that was very interesting for me. I'd been running the electrical control unit in the nuclear business and we'd struggled with some of the same issues that they were struggling with, which was that nuclear is a sort of mechanical engineering physics area. Aerospace is full of physicists and mechanical engineers, but actually the microprocessor and control systems and software are bringing wholly new challenges. And what we found when we across aerospace was We couldn't make the black boxes which would control these modern engines. Therefore we couldn't deliver the engines and because we couldn't deliver the engines there were 747s, were 747-400s in those days sitting on the tarmac in Seattle with concrete blocks instead of engines destined for BA and Qantas. And BA and Qantas were not at all amused by this thing and Boeing were livid about it and they wanted something done.

You can focus on the black box, but the real issue was that the organization had grown up with lot of people who were mechanical engineers into governors and gearboxes and oil systems and such like. And actually this thing had come out of left field. Sort of electric's been about contacts and sparks and it was then about microprocessors and software and complexity and the FADEC system, the full authority digital engine control on the 747, which is now

routine but then was brand-spanking new is the biggest electronic system or 747 other than the flight control so here was us in Rolls Royce in Derby designing the biggest outside the flight control system for Boeing and feeding into this thing and what there were growing pains and struggles and we're working with a specialist supplier in Birmingham and between the two organizations we won't make a very good shot of it

So what we needed to do was some immediate actions on trying to fix the problems that were there and then try to find much more productive ways of forward and I think partly my experience of coming from nuclear where we've done more of the systems engineering but also looking at the organization with several hundred engineers and in there were some very good systems engineers but they're quite low down in the organization. The organization wasn't focused around with systems so when one or two people decided to move on we reorganized

the

whole thing to bring systems really to the surface and give them an overview and bring the systems practices and the people who knew about it knew about software and systems we brought them up the chain and we've trained all the old skills but this enabled the organization to run much more much better and to face the challenges that Boeing set us and Airbus set us. I remember having to go and tap the boards at Boeing and with and at Airbus to explain that we were going to do what

they wanted us to do and it was going to work out but you had to give us a little bit of space and time.

Wes Ashworth (06:13)

it's a great

example and just that journey and kind of learning through that. And I'll tie this back to nuclear just to make that direct connection. So what did that experience teach you that still applies to nuclear projects today?

Tony Roulstone (06:24)

There's a lot of know-how in the industry, but the industry moves along and the focus of these things moves. If you go back to the old days of nuclear, it was ruled by physicists. That was the really sexy bit about nuclear and everybody who was at the top was a physicist. That's not necessarily the case now. Mean, software and systems is very important and our job is important. Materials is an absolutely crucial thing which the physicists might not understand. And so it's understanding that you need the range

of technologies and you need to listen to them all and you need to move with where the focus of industry will get onto economics and business experience in due course. But the nuclear industry has been nominated by very bright, very capable mathematicians, physicists and theoretical engineers and probably they're not the most important thing that we require, at least with the existing technologies now. Need people who can craft these

and work with supply chains and get things done in a completely different manner to be effective and I think probably we're going to talk about that.

Wes Ashworth (07:24)

I think, you know, looking at this sort of aerospace example gives us this useful lens for nuclear. And, you know, the reactor may be technically sound, but as you're saying, like the system responsible for delivering it,

that could remain, you know, fragmented, temporary, and inefficient. And so that maybe is even more valuable. So and I know you've made this comparison to sort of large nuclear construction to say like building a 747 in a plowed field.

tell us about that. Like, why is this such an inefficient way to build?

In terms of how it's been done.

Tony Roulstone (07:53)

I only spent quite a short time in aerospace, but this sort of aerospace ideas and analogies I often go back to, the experience there was important. It was actually brought to me when I went to a nuclear conference when someone said, well, we've done some analysis of the Westinghouse fall of PWR and AGRs and actually they've got the same amount of connections and valves and pieces and rivets, they're welds, as a 747. This thing on the ground,

It's

just as complicated as a 747. How do they make 747s? Well, have a sort of production line in Seattle, the same at Airbus, they move them along one month at a time and people do all the work. It's systematized and they've got the paperwork and they've got the material flowing to the station they're going to do it. It's not quite like a car factory.

where

they just move, you know, along two miles an hour and the cars go through in six hours. It takes several months, but the same idea of systematization, getting the tools and equipment coming to the people, coming to the work is important. When you look on a nuclear power station, I'll just give some facts about we're building two very big reactors in the West Country at a place called Hinkley Point in the UK. They are some of the biggest reactors in the world.

But currently there are 13,000 people on site. They started building in 2019. And when you look at a picture of it or you see a video of it, you see these forests of tower cranes and you see the concrete being poured and you see that the caves.

that people are trying to build around the reactor and then they're going to put piping into it and then instrumentation and wiring into it and it all has to be separated. The complexity of building these things makes it pretty odd that we're doing such a complex, high quality, paper-bound, inspected system.

in a field, which it is. It was a field before. We've dug all of the ground, put some concrete down. It's a field. The people who have been doing it have been brought from all sorts of subcontractors. Because they take so long, these things, it's not as if you've done one last year and we're going to do the same job this year. You've probably not done one. We haven't built a reactor in the UK for 30 years. It's the first time. It's the first time for people. That's not how to build complex systems. But it's probably the

way you can build these very big reactors such as EPR in Europe or AP1000 in States. So it's not surprising that these things take a long time. The site is very unproductive because you can't get access. People have to crawl over, almost crawl over each other, or get clearance to access the site. They have to get the right material and paperwork and inspectors to go to every place and then it's signing off and then some other inspector to have a look at it before the next

check

I can get into it. It is just the wrong way of building something complicated so I use the analogy of building a 747 in a field.

Wes Ashworth (10:58)

it's such a powerful image, that immediately kind of clicks when you think about it 'cause nobody would deliberately choose to manufacture aircraft that way. And

Tony Roulstone (11:06)

It sounds like I'm moaning about the people who do it. The people who make these things are doing almost superhuman things. They make things and they do not stop until it's high quality. They do not stop until all the paperwork is right and they don't stop. It just takes a long time and you find that organizations who have done it before recently are much better at it. So in the 1980s the French built

30 3 Loop Westinghouse PWRs and 24 4 Loop Westinghouse PWRs, very similar but slightly bigger. The crews built them in four or five years and they moved from, they built two on a site and then another two on a site or they moved between sites. And so they built up a know-how about how to do this extremely difficult task and manage the paperwork and the flow of materials because they, in their bodies, they learned the practice of

this stuff and so you need extremely experienced and skilled people to do this thing and you need a program so instead of a production line at Boeing we had a production line of them and they moved between the people being between sites but then they were in the same sort of thing on the same sort of reactor down the road and they got very good at it. Of course more recently the people who got very good at it are Koreans. They took an American design

and

of course the combustion engineering system 80 and they adapted it and they've called it OPG 1000 and they built one of these a year generally on very similar sites so you can see sites in Korea with four or six reactors on the site very similar designs and the people who built them and this is now not the nuclear physicists this is the welders it's the guys bending the reinforcing guard and putting the concrete down. It's the planners, the detailed planners, it's the schedulers, it's the getting the materials, it's the guys who do the inspection and the guys who keep the paperwork tidy. All those things work. I know I visited China ten years ago and talked to people at CNNC and CGN, the big builders over there, and it was very funny. We worked on our project there.

on the other people's project and then they can work on our project. So they were all working on each other's projects and I thought it was very odd but actually it was really important because they realised the skill, although the designs were not big difference but there a difference between the ones built in the north which were Chinese design and the ones in the south which were French design but the practice of working on site and getting these things done to a high standard and getting the paperwork organised, they'd learnt the best way is take the guys that had done it before

and get him to do it again and then we'll move him to get and get him to do it again which under their centralized control they can move people around but this business about repeatability if you're to have to build these complex devices on sites is really important.

Wes Ashworth (13:55)

I appreciate you pointing that out as well, too. I think some of the sort of lost knowledge and experience and then and some of the things that are happening there with that sort of shared execution, it just creates a different kind of knowledge and that combination of ideas. And you know, I think those hardest skills often live in teams, routines, judgment, and experience rather than in and in manuals. So important to get that right. And so I think we've now successfully sort of

Painted the picture of current state, some of the opportunities and how it's been done before. And now kind of moving into some of the new ideas, which I'm sure people have heard, but I want to dig a bit more into, which is really first into like SMRs, which are a hyper buzzword right now and a lot of stuff going out there. But you know, they're SMRs are presented as a way out of the one-off mega project model. And

the real promise is not simply a smaller reactor, it's a different

production, financing, and delivery system, which we started to kind of get into there. But SMRs are presented as the answer to maybe that broken model. Why is a smaller reactor not necessarily automatically a cheaper reactor? Let's start there and then we'll dig a little bit further into it.

Tony Roulstone (14:53)

so we started looking at SMRs about a dozen years ago and at the time when we talked to people they just said you're mad.

And they said, you're mad because everybody in the nuclear industry believed in this immutable law of the lower costs of larger scale, the scaling law. And they enacted this throughout the nuclear industry. So if you look at American designs, in 1970s, there were three or 400 megawatts size. This idea, it's just in came from coal power, where it definitely did work. It worked in Europe and in the US, where they built smaller coal plants.

bigger ones and bigger ones, so they got economies of scale that the cost didn't go linearly with the size of the output. For nuclear, they absolutely believed in it, and every time they had a cost problem, they just designed a bigger reactor. So was 300, 400 in the early 70s. It was 600 or 800 by about the end of the 1970s, then 900 in the early 1980s, and then 1,200 somewhat later in the 90s.

and nowadays there are designs which are, well the EPR is 1700 megawatts and if you talk to any designer in one of these big design houses they say absolutely anything small is going to cost more and we have made them larger in order to get economies of scale and if we haven't yielded them we can't see the benefits it's because of other effects and this is the problem.

There's

some neat way of calculating this thing. But one of the problems is when I looked up some statisticians who done some work in the US actually, then elsewhere, and they said, when we look at the economies of scale in the cost of the 100 reactors we build in the States, we can't find it. We just can't find it. It doesn't seem to be there. The smartest thing that was said by some of these statisticians was there may be economies of

scale but larger reactors take longer to build so instead taking six years to build, in fact eight years to build and because you have this standing army of administrators and overheads any cost savings get consumed by the longer duration and therefore there may be cost savings associated with making them larger but they take longer to build and so the bill at the end is no less. So this is always

been saying that if you reverse this argument for small reactors, say, it's going to cost more. You build them just the same, they're going to cost more. And they will also say that...

you sort of need the same minimum number of people to operate a reactor. In fact, think it's actually legally required in the US at the moment. There's so many operators and so many security staff and so many health physics people, whether it's right or wrong, per reactor, and therefore it's going to cost you an operating cost. But if we just stick to the capital cost, which is the biggest element in nuclear, how do you square the circle? How do you get

the

benefits. The idea that we've been doing research in is by using a smaller size of reactor we can take work away from sites because the bits you're doing are smaller and you can put it into a place where you can have a higher productivity to get better tools. You can have the systems that do all the documentation and bring the inspectors at the right time in place and also you can have the flow of goods and

tools towards the people, not the people who going go and find the goods and tools. And therefore, a factory manufacture of as much as you can is really important. But you have to design it has to be designed at outset for what we call design for modular manufacture and site assembly. The work we did

basically says that if you build a small reactor the traditional way using

on

site with concrete and rebar and wiring coming there, actually they will cost more than large reactor. So anybody who wants to build a small reactor in a traditional way has got a big hill to climb. It's only if you think about the strategies, there were three strategies that we identified. One is a radical idea of standardization. Again, the nuclear industry doesn't really standardize things. They like to tinker and change the design. They either change the design or they change the prime contractor who does his own detail.

design

about how he's going to build it or he picks his own turbine or he picks his own pumps or he picks his own supply chain and so unfortunately the best example of this is in the US where there's 101 reactors built in the 1970s and 80s and apart from a couple of Diablo Clannan they're pretty well all different even ones on the same site even provided by the same reactor vendor that built at a different time they're not the same and so this idea of radical standardization

which is absolutely routine in other high-tech businesses. Don't customize every one to every customer or every site. It's too expensive and it doesn't lead to reliability and such like. So standardization. Standardization is also important for the other two steps. So if you don't do standardization, you can fit step two and three. Step two is modularization. Taking the design and make it into chunks and making those chunks so you can make it in factories.

reasonable amount of stuff made in factories at the moment, the vessels and the turbines and the control system and the fuel and such like, but this is only probably 25% of the cost of the unit. So you've got to really dig in and say what else can we do? How can we modulise? And certain things are quite easy to modulise. So there's things like some of the water treatment and water conditioning systems. You put all the bits together on a little frame, make it in a factory and put to site. You just have to connect it up.

nice and easy. There are more complicated things to do with a lot of the steel work, which you could make into sections and you could take it to the site and weld it up on site and bolt it or do something. And then there's more complicated things because the thing that's really time consuming is all this concrete and rebar. And concrete and rebar in nuclear, unfortunately, is very complicated and also has the high quality standards because people want to be able to demonstrate they can work after an earthquake.

earthquake

and therefore instead of rebar being the reinforcing bars being half an inch in size, they're two inches in size and instead of being spaced four inches apart, they're

three

inches apart with an inch between. You've got to pour concrete in between these things. And by the way, they also have to be tolerant so they can only be a little bit out of whack in terms of space. There's a story which I thought was apocryphal, but I've had it confirmed so many times I don't think it is. At Hinkley Point, because the safety inspectors inspected the rebar, big cages where they had the foundations, they said there were 70,000 anomalies where the rebar wasn't spaced correctly.

each one of these things and each one of these things that it had to computer analysis which costs several thousand pounds of each one each condition and if you multiply that out it's about half a billion pounds

the irony of this thing is when you reanalyse these things for the tolerances, they didn't finally have failed. So actually it was a design fault that said, you're to have to these spaced like this. But there was a tolerance which was acceptable. If they'd done that, save half a billion. Anyhow, that's the second stage. The third stage is when you have stuff in a controlled environment, something appears in every other industry, it's production learning. So this is embodied know-how, learning on better practice.

as you go along. If you have people who do things...

quite often, know, a month or once every couple of months they say, well we did this like before, we knew exactly the same, but actually we got this stuff in first week, it'll take us a shorter time, or we get that paperwork and got this chapter appear, we can do it in less time. And it's a well proven thing, every other industry does it. When you look at the data for nuclear industries, absolutely no evidence of learning. And it's probably not because there's no learning, it's because the time between these events is so long that people forget, as well as this learning,

There's a forgetting idea. If you don't do things for a couple of months, the guy moves off or he goes sick or he concentrates on something. I've always promoted or he goes on a training course. And so there's collective forgetting. And if you have a year between things, any learning is lost, is forgotten. And so you do need a flow of this work, which is similar and same people doing it to get learning. But these three ideas, standardization, breaking to

bits of modularization and production learning, which means you have to have a series of these things going forward. It's just not possible with these very large reactors. So large reactors cost 10 or 15 billion dollars a unit. It takes a lot of effort and people putting together to find that money to build a couple. The couple had been built, finished in Vogel in Georgia. That project's been going on for 20 years. It's a big effect. But if you have smaller reactors,

you say a couple of billion.

You can raise that sort of money as long as you're confident you're going to deliver it on time, it's going to work. Instead of taking 10 years, it takes, say, three or four years to build. But the financiers like this, that you can then, instead of building one big reactor, you can build half a dozen small reactors and you can get this standardization, production learning effect. And you can make it more product-like. I'm not saying that you push the whole reactor out of a factory.

but you can make it more product-like and this is what the object is.

Wes Ashworth (24:12)

it's such

a useful explanation. And you've made such a great case for SMRs and sort of like why it works. And I think the counter arguments too, where people might go to, you know, the upfront cost of the capital expense initially, and then

why that, you know, really makes sense when you look at the long-term sort of holistic view of what that will mean when it where it rolls out. So it makes a lot of sense of why SMRs are getting so much attention and why this is starting to happen.

Tony Roulstone (24:36)

I use SMRs to mean light water based. So all the reactors in the world, apart from the ones in the UK, are based on light water, boiling water or pressurized water reactors. We know how they work. 18,000 reactor years of operation in the West and China and Japan, which is all very good, and another 18,000 in submarines as it happens. So we know how light water reactors work. So the innovation in this space is not in the

reactor technology there might be some fiddling about. Fundamentally it's about how you change the production approach for this and the success of this is about not the reactor physics, not the thermal hydraulics, this is about quality and investment and factories and organisations such like. So these skills of business management and more in the engineering space are very important. I just wanted to talk about why I'm so optimistic about this.

possible happening. There are three or four, well half a dozen designs going around, but there's three or four which have attracted quite a lot of orders and projects and this is absolutely key. You can't build something of this scale and complexity unless you've got the infrastructure that goes with it and train people. Unless you've got an order book, again Rolls-Royce, we didn't launch an engine unless we had an order for 500. How are you going to pay for the development?

costs and how you're going to pay for all the setup and such like and nuclear things tend to built one-off one-off this is pointless so you've got to have a decent order book and GE have got an order for four in Canada they're building there's probably others going to be building sex, caturand and CVA they've got orders in Poland and think Slovakia and they probably get others so I think they will get to the very close to the

minimum order level to build a proper supply chain and organize studios and roles are in a similar position they've got order for three in the UK from the UK government but they'll probably be six or twelve they got roles for six in the Czech Republic three in Sweden and there's opportunities in other countries in Holland and elsewhere and so this minimum order level which is absolutely fundamental and completely unknown in nuclear to have this sort of thing to work at is possible

And there are other people out there, Holtech are a little bit behind, but they may well get to the same position. They've got some orders to do with data centers and New Scale, who were leading, are now getting some traction with some of the big companies in Entra that want to build, I don't know, a dozen of these things down in TVA. And other people in the States are saying, well, these are the reactors out there. These are the things we're going to build, so let's build a lot of them. And this will enable them to

start

facing the industrial problems rather than technical problems.

Wes Ashworth (27:11)

Great context there. I appreciate you sharing that as well and getting into that. Before we move from SMR, is any other kind of key things you would want to share the audience should know?

Tony Roulstone (27:20)

The World Nuclear Association does a tracker of how many SMR projects there are around. This includes light water-based ones and also these ones we're going to move on about with novel coolants, gas, molten salt and liquid metals, so-called advanced reactors. And they say that when I last looked, 127 small reactor projects that have been done seriously around the world. They're in every continent.

major

countries, there's lots in the US, there's lots in Western Europe, there in China, there in Japan and Korea, there's some in the Middle East, there only people who haven't gotten here, Australia, because they've got a moratorium on nuclear in Australia, but they're fighting that at the moment. So there's a lot going on, and it's going on across the board, and this is why I say there's the nuclear renaissance going on, because not all these 127 will succeed, but there's so many of them, and a lot of them are

funded privately. The old style was you wait for the government to give you money and if the government doesn't you money you sit on your hands till they give you some money. And then the vendors used to spend money but most of the vendors because it's been such poor industry.

revenues for the last few years. They just haven't got balance sheets to do anything. The new factor is private equity coming in and saying, well, you can have a billion for this, or can have 10 billion for that, or 5 billion for this, and we'll give you this, and we'll give you this later. I mean, there's a good example. Just recently, there's a small company called Valar making a microreactor, and they've just gone and done an IPO and raised a billion dollars. And this is a company of, I can't remember, 50 guys down

somewhere in and they've got a design. They went through the first stage of President Trump's criticality exercise and used this leverage to raise real money to build some reactors.

Wes Ashworth (29:01)

I think great context and helps us really understand like what is really happening in the world of SMRs and other reactors

as well, too. And as a part of that, I know we sort of touch on, but just a lot of talk about SMRs, also, you know, coming up more now, like advanced reactors. You're starting to hear more about that. And,

you know, some promises of more radical changes, you know, in fuels and coolants and pressure, temperature, safety. Creates opportunity, also creates another maybe layer of execution risk, but

with advanced reactors again promising something more radical,

from your perspective, like which advanced reactor idea has the strongest maybe practical case?

Tony Roulstone (29:33)

we'll be contentious, but I'll give you some sort of general figures. So these things are characterized, best characterized by their different types of coolants. So you've got high temperature gas cool reactors, gas, microcarbon, dioxide, helium. And we know a lot about that in the UK because we had that advanced gas, we still have advanced gas cool reactors. But these go to higher temperatures and

They actually go back to the 1960s when the first Dragon reactor in the UK and then there were other ones in Germany and then in South Africa. These high temperature gas cool reactors, much studied in the States but not built in the States until recently, what they offer is more...

efficiency because of these higher temperatures. They have, either because of the experience in the prototypes that have going around, or the ADRs in the UK, they've got a lot of operating experience. And one of their key things is they use this new fuel, tricer fuel, which is a ceramic fuel. Instead of putting the ceramic fuel in a metal tube, which is the usual approach in reactors, they put it in a coating

of silicon carbide and high density graphite and more silicon carbide and such like. These one millimeter fuel pellet, the pimples, surrounded by these surrounding layers retain the fission products. Because now all ceramic, it's almost impossible to melt, instead of metals which melt.

6, 7, 800 centigrade. We're talking about 3,000 centigrade to melt these ceramics. And so this makes them inherently very safe.

and when you add this to the fact that they have rather low power density that you can have what might be called an incident, that is you have a power change but you don't do anything about it, actual reactor shuts itself down. It's very tolerant to operating outside the normal envelope and the Chinese who built the reactors called HTRPM

They've demonstrated this just recently. It's interesting, they're actually Western technology which was put into a project in South Africa called Pebble Bed and that ran out of money and then the Chinese bought this Pebble Bed technology from them and they built the reactor and they've shared all the information, they show it works. And this technology now has been taken up by X-Energy who are projects in Texas with Dow to build some smallish ones about 100 megawatt electrical. Slightly

less and a number of these things and Dow want both power out of them and because they have very high temperatures perhaps seven or eight hundred centigrade they want to use some of the process heat for some of their petrochemicals. So that's the first type of advanced reactor and then changing coolants

you can use, it's odd to people, you can use liquid metals. A traditional one to be used has been sodium or sodium potassium mix. Sodium melts at about 75 centigrade and it's liquid above that until it boils at about 800 or 900 centigrade. So you've got a very big range in which it's a liquid and it's got very high conductivity, high thermal capacity, it takes a lot of energy where you can have very dense core

and

Their advantage is because you don't slow the neutrons down, you can breed new fuel from old fuel. Using these fast reactors, you can breed

Fissile fuel used in these high-energy neutrons and they were popular and they were researched around the world for this feature of breeding more fuel. And the third and there's various people in the US, Oklo are very noisy proponents of this.

Bill Gates has funded a project called Natrium using EBR technology and a bit of Prism reactor from GE. And he's building that. That's further ahead. It's got a ticket from the NRC. They're actually building and doing real work at KEMERA in Wisconsin. That's a real project, whereas Okla is a bit further behind. And there are other people in Europe.

just mention that other people say rather than have sodium as a liquid metal there are other ideas of liquid lead and this has been pursued in Sweden and in France.

The third type is the most novel. There is quite a lot of experience because of all these countries in liquid metal. But I have to say this because there's very little experience in the next one, but lot of enthusiasm around molten salts. So you take something like table salt and it does melt at about 350 centigrade. And once it's molten, it conducts heat. It operates pretty like water. It's not very viscous. It flows around. It takes heat away.

It's quite good. Most people don't use salt, is sodium chloride. They use sodium fluoride, partly because it's got better physics characteristics. You may mix it with other things, lithium and beryllium, but the same idea. They are, they freeze about 300 centigrade in their liquid up to very high temperatures. And they operate in a similar way to molten liquid.

metal ones, but they don't have the problem for liquid metal. Sodium, if you've done experiments at school, you mix them in air, they burn into, they burst into flames, and you add water, they produce hydrogen, it bursts into flames. So the sodium-cooled fast reactors have this thing you have to protect against, but the molten salt ones don't.

it's a

open space. There people who have fast systems, that's with high energy neutrons, people who have moderated neutrons, have people who have fluoride salts, you have fluoride salts, have with beryllium or without beryllium, and you have ones which have fuel dissolved in the salt and the other ones have fuel in tubes or in trisobar particles. So actually we don't know what the right answer with molten salt is, but everybody is trying everything, I think is the answer.

you

Wes Ashworth (35:07)

absolutely.

A good overview of just the different technologies that are happening, some of the advancements and things like that. And you brought up a couple of them kind of tie back to safety. And I do want to bring this up.

I think obviously new technology may improve the safety case, but public confidence is still shaped by nuclear's history.

So what's the clearest answer you can give to someone who associates every reactor with a Chernobyl, you know, a meltdown type of

event? How do you clear that up?

Tony Roulstone (35:31)

So I teach at the University of Cambridge and one of the things we make sure all students do is they do a course in nuclear safety and part of the process is go through all the terrible accidents and we do it early on and they're all very glum about it but I think it's instructive to face the issues and to learn from them.

In the case of Chernobyl, is the one that's got the most profile, there aren't that many actually, so it doesn't take long to go through, but they are very instructive each one. Chernobyl, everybody probably seen the film and such like a terrible thing. But two really significant matters there. One is the reactor wasn't stable under all conditions. What I mean is if you have the reactor operating and it's somehow disturbed, it should go back to its previous state.

very odd characteristics that in certain configurations when he did the obvious thing like push the control rods in to put the use the energy it's actually the energy went up not by no way it was a known problem by the Russians they did know this occurred and they tried to have rules to avoid it but this led to the second issue which was the way that safety was managed and the organization and management of it was not very broad

and very questionable. So at the time at Chernobyl they were doing an experiment to do with...

getting energy of a run-down reactor to keep running the feed pumps to keep providing power even though it's going to be switched off. So the people came to do this experiment, the people who operated here, and they didn't seem to know what they were doing here. And between the two of them, they blew the reactor up because this lot disturbed it. And these people, it was actually in a very odd configuration towards the end of its life and of its fuel life with all the rods drawn out, probably beyond the limits of what

allowed so they should have never been drawn out there. Were rules against this and they had this disturbance and the reaction they put them in and the reactivity went up and the whole thing went up. So there was a fundamental design fault of not having a stable reactor. You want something stable in all characteristics. Again getting back to aircraft,

you want to be stable all the time. There's a perturbance, it goes back to normal. If something goes back to normal, it actually in a robust, benign way, and the pilot has got something to do about it. They didn't have that, so it's a first strike. And the second strike, which is absolutely fundamental in our safety assessment processes in the West, and the second one is this rather...

unthinking and disorganized management approach which seemed to exist in Russia.

After Chernobyl, the socialists around the world got together and they didn't just jump on Russia. They said, we all have to learn from this. And they formed a thing called WANO, the World Association of Nuclear Operators, to share experience and to raise the standard of nuclear operators everywhere. And they've had a very positive effect. And it operates in China, it operates in Russia, it operates in the USA, it happens in the UK.

everywhere. Well, Bangladesh is just starting up. It'll operate in Bangladesh. You know, this is, this is a human aspect of it. It just copies what happens with pilots where pilots report problems and they get to the base of it and then share these out with no-tams in the aviation world. And the same idea with Wano to share the best experience and the bad experience, then how to avoid this and push it aside. And I think

there was something really important to learn in terms of sharing and best practice and having high standards in operation and the mixing of normal operations with test things and such like and You know if you have those things together you have a very safe system and if

you know if you go

to the stats, which most people don't go to the stats. You have independent people who look to the safety of all sorts of energy generation industries and nuclear is right at the top. It's accidents. Have got lot of attention, which is right, but it's not a poor record and if you do things right, the record should be zero.

Wes Ashworth (39:28)

Absolutely. Great additional context. I appreciate going through that. I think it's always great to clear up those misconceptions and kind of give people a realistic idea of yes, we acknowledge the history, but kind of the current state of where things are today and

how safe

it is and reliable it is today. I got couple more questions as we start to wrap up here. Obviously, you know, this being a renewable energy focused podcast, we always bring different energy parts of that equation into the mix And you know, there's this argument of sort of nuclear versus renewable energy, which I disagree with. I would love to hear that from your perspective. Like, what should we be thinking about instead?

Tony Roulstone (40:02)

well, we see that over here as well,

So one of the things I did a few years ago was because if you want to make the business case for nuclear

you've got to understand the system it's going to fit into. So you've got to understand the energy system and what the alternatives are. And the big new effect was climate change, means in the future all energy systems are going to have to either have none or account for or minimize the carbon emissions.

In our future system, everywhere we'll have at least 50% renewables. It is just where to go, it will be the way to go. Some places it will be much higher than that, some places it will be a bit lower than that. But pretty well everywhere I think we'll have 50%. If you go above 50%, the storage issue becomes more acute. But if you actually look at the climate change...

challenge in most countries. It's not just about electricity, it's about heating, it's about industrial, it's about transport, every damn thing.

Unless we use every tool in our box, we will not get the carbon emissions of modern Western societies and the developing countries who want to copy us down. We will need higher temperature sources of energy, which nuclear are good at. You need all types of energy to get to the volume. It's not an either or. It is both.

Wes Ashworth (41:18)

that I appreciate you sharing that insight so much. I love having different thoughts and perspectives on the show and kind of covering that. And I've often said it's an and, not an or, we need all of it. And you've kind of brilliantly put that together as well, too. And I agree. So I think we've got to get rid of this kind of us versus them type of mentality and coming together and understanding that it's all needed and it will all be

Necessary, you know, to use all those tools and toolboxes you just shared there. So I

love that a lot. I like the perspective. Closing question: if we're thinking about this, long term, thinking all things nuclear, you know, by 2035 and really beyond that, what would convince you that nuclear has really just finally really learned how to scale? This has happened. We've gone into this you know,

and it really does take hold and happen at large scale. Like what would convince you? What would need to happen?

Tony Roulstone (42:09)

there's lot of enthusiasm, a lot of ideas, a lot of project people work at it, and something good will come out of that. The question is, how will we know when it's good? It's all about delivering what it says on the tin.

nuclear does deliver the power but at what cost and what delay. We have to learn to deliver if we say the target is to build this reactor in three years or two years or five years we deliver it in three years or whatever it was. If we say it's going to cost X, you know after the event saying it was unfortunate it was a bit more complicated we thought it's 3X, we've got to do X plus or minus 10 percent because if you want people to use a lot of it has to be financeable

It has to be attractive. Has to be arrived when the utility wanted Utilities can't wait the financiers hate ruining their business case quite being late so doing stuff on the time scales that you say and The cost you say is the sea change and we're looking for which is why I focused on production engineering I don't really mind whether it's advanced reactors or like water reactors. They have to meet the mark of providing that

power which nuclear does at the moment it does it very reliably but it has to do it not at any cost and not many years late. It's not about what I promised to do it in three years but it took me six years. I promised to do it three years, I did it in three year one month well that's okay we're late off the month but this becomes routine and this is what happens out of all successful industries they have to become predictable and regular and the target by 2035

is to have it to be not a big deal when you order one because you expect it to come on a particular time scale and people know how to do it and they build it so it gets done and they move on to do something else and it becomes unacceptable. It's lots of explaining why exceptional circumstances and why it was very difficult. You've got to make it simple, you've got to make it repeatable, you've got to make it on time budget.

Wes Ashworth (44:01)

Simple, repeatable on time, on budget. Great wisdom to live by there. And I guess final thought, like what gives you hope that it's headed that direction, that this is the path and that it will happen?

Tony Roulstone (44:13)

I think that the number of people are trying. The youth, we get students come to us, but you also talk to these people in these start ups. They're very enthusiastic and they have a sense that they're to do something which is good for humanity and they have lots of good ideas. They need to go through the safety and testing regime and some will fail, unfortunately, but some will succeed. And I think the idea that it has to be a production process, not a one off process.

be pretty widespread. These things brought together, enthusiasm of new people, the private equity money and a production engineering approach, I think will get us to a place where it becomes just another industry which is successfully doing what people asked it to do.

Wes Ashworth (44:54)

Great, hopeful points to end on there. I think a great way to wrap up our conversation. Tony. Thank you so much for joining us and helping really make such a complex subject much easier to understand. To everyone listening out there, thank you for spending this time with us. If this conversation changed the way you think about nuclear energy, share the episode with someone who would find it valuable. Please subscribe to Green Giants wherever you listen. Leave us a review, it helps more people discover the show and supports the work we are doing.

You can also check the show notes for important information about Tony and some links and topics we discussed and additional resources there. Thank you for listening again, and we will see you next time on Green Giants Titans of Renewable Energy.