Francesco Sciortino, Co-Founder & CEO of Proxima Fusion: Building the critical power infrastructure for clean energy

0:00Francesco Sciortino:Hello and welcome back to the Scaling Europe show. I'm Sid Johnson. Today I've got an amazing guest on the show, Francesco. Francesco is a co-founder and CEO of Proxima Fusion. Proxima Fusion, now the most well-funded fusion company in Europe. You've just raised, I think, 411 million euros at a 2.4 billion euro valuation. Amazing news. How are you doing? I'm doing great. Interesting week. I bet. I mean, it's phenomenal news. No small amount of funding. It's a remarkable thing that you've built. So thank you for taking the time. Thank you. It's great to be able to also tell the story of how the round happened and what it's been built with that money.

0:36Francesco Sciortino:Amazing. We're going to get into all of it. But before we do, I want to go a bit into the science because everyone knows about fusion, everyone talks about fusion, but I don't think there is a depth of understanding as to what it is and why you've taken the bet that you're taking. So can you just explain in very high level terms, what is fusion? It's the physical process that happens inside every star. It's the process by which some light atoms can join up and when they do under high energy in high energy conditions part of their mass can be transformed into energy fundamentally the Einstein formula e equals mc squared is what's at play here so we're transforming some of the mass m into e energy and that is similar conceptually to what's happening in fission so in the kind of nuclear that everyone is familiar with from nuclear power plants that are on the grid in many countries.

1:29But we're doing the opposite in some sense. Instead of looking at heavy ions, atoms, and splitting them, instead we're taking the lightest atoms and joining them. But fundamentally, it's still mass going into energy. The benefit of trying to do fusion rather than fission is that it's intrinsically safe. There is no so-called chain reaction, so the reaction cannot run away. So safety is intrinsic. and also we're talking about the most abundant element on the universe in the universe so heavy forms of hydrogen and not simple to do but if we get this right we get a clean safe effectively inexhaustible source of energy for humanity and why haven't we been able to do it yet it's bloody hard trying to do fusion for many decades we've known how to blow up things with fusion for a while that's what an h-bomb does that's not terribly useful in the context of making energy in trying to adapt that immense source of power of the stars to something that reliably safely makes energy in obviously a way that is completely different than what a bomb would do we've learned that it's not easy you're fundamentally trying to use this enormous source of power of the universe in a confined space.

2:49And you can try to do that in many ways. There is a zoo of concepts for fusion, going from the very pulsed systems with lots of energy being deposited on a small pellet for a very short amount of time. One way to do that is with lasers or with large capacitor banks of current that discharge for just a short amount of time. And then you pulse, repeat. At the opposite end of the spectrum, there is the QI stellarator, that Proxima Fusion is pursuing. And that's a system that is stable, reliable, does not pulse. It just, like a microwave oven, you turn it on, and if you've designed it right, you built it right, it just works.

3:27And this is what we're trying to work towards, fundamentally a baseload, but most importantly, dispatchable source of energy. And this involves a spectacularly multidisciplinary kind of challenge. So you're looking at the extremes of materials from a thermal, a mechanical, an electrical, a neutron, perspective, you take any challenge, any material in that device, and you find that it's pretty complicated, which is why you need a certain kind of engineering team to tackle this.

3:55Francesco Sciortino:And you've described it as more of an engineering problem than, I guess, a physics problem, I think I heard you say. Going back in time to 2022, 2023, you had spent a lot of time studying fusion, and I think you were quite close to Tokamaks, which is like another school of thought within this approach to Fusion. The approach that you're taking is building a stellarator. Can you just kind of explain the difference between those two and what changed in 2022, 2023 that gave you conviction that this was the right approach? That's exactly right. So I actually started in a domain called ZPinch in the UK when I was doing my undergrad at Imperial and that was my first love in Fusion.

4:33It's what guided me towards saying, this is the challenge of my life. And then I came to Tokamaks, which are the most conventional way of doing magnetic confinement fusion. So like in a stellarator, in a tokamak, you have a donut-shaped device, and this donut is creating a magnetic bottle, effectively. You're trying to hold this hot ionized star stuff in a cage, and that cage is not physical, it's not made of materials, it's actually made of magnetic fields. So basically, you're levitating hot stuff at 100, 150 million degrees, so very hot in a empty space in a donut-shaped device so i started doing that while spending one year in lausanne at a pfl and then i kept going in in at mit working on true tokamaks in the u.s and then i joined the max planck society in in munich and was one of the european coordinators for tokamaks until as you said 2022 in 2022 a bunch of things happened the first trigger for me and my co-founders was the realization that some work has had been happening that showed that the last big physics drawback of stellarators simply didn't exist anymore so the confinement of higher energy particles was just fine in stellarators it was still not clear whether you could achieve all the physics metrics and the engineering metrics at once but we did know that you could achieve them one by one just fine.

5:56And also in 2022, W7X, the most advanced accelerator on the planet, which happens to be in northern Germany within the Max Planck Society, in 2022, it went full power, full build out, and it did exactly what it was supposed to do, which is a very rare thing to do in the world of science. That was totally on the scientific side. The challenge there was, it's a big project, but fundamentally it was about validating the science. And then we started Proxima saying, okay, it's not like all science has been discovered. That's a bit of a funny statement. It's that we have a physics basis. We have a clear baseline and we can rely on it.

6:35And so we can build an engineering company. So Proxima focuses on engineering.

6:39Francesco Sciortino:And so at that point in time, you went out to start the company. It's like a really old school traditional venture bet in the sense that it's like all or nothing. Either this works and you can build this amazing thing, but it's very capital intensive. there's lots of things that could go wrong. What was the investor appetite like that within Europe to back you? And you'd obviously put together this phenomenal team of people who really knew the science and the industry. But what was the reaction to investors that you spoke to in the earliest days? It varied. There were folks that were asking for a business plan and that was not the focus in the early days.

7:15The right question for a deep tech founder at the early stages is what is the problem that you're trying to solve? and do you have a solution? Tell me about feasibility before you tell me about whether the time is this big or that big. It's big. That's the point. And the smart investors do get that and they understand that investing in a SaaS company is different than investing in a dip tech company. Asking a fusion founder how big the market is is not a helpful conversation. It's philosophically interesting. And at some point, you want to think about it. And at some point, you do need to have a business plan and think about all the financial elaborate thinking that you need along the journey, but there are different stages.

7:56In the past three years, we've really seen a change though. I mean, fusion has become more of a discussion for venture capital. I would say a lot of people have been a little late and are not yet fully familiar and happy with the fact that fusion is a topic for venture capital in the 2020s, not in the 2030s. Once we start in power plants, that's not a space for venture capital. That's a space for a very different kind of capital, typically. So the conversation has become more elaborate. More people ask better questions. There are technical risks in what we're doing. Otherwise, we would have built a power plant already.

8:33And so asking the right questions, but understanding what is a scary mountain that you cannot climb versus what is something that if you build the right kind of team in the right ecosystem, thinking about supply chains in the right way, with the right public research partners, then you might just have a shot at that massive price at the end of the road.

8:52Francesco Sciortino:And I guess you've gone from being this crazy, crazy bet three or four years ago at a time when Fusion was quite unpopular, even deep tech itself was quite unpopular, or certainly not as mainstream it is now across European venture, to now raising this huge round with some amazing investors. Can you talk a bit about what has changed, what you've achieved maybe in the last three years that has meant more investors have been able to understand it and get on board and make it seem, you know, you've kind of closed that gap from very, very low likelihood to slightly less, you know, higher chances of success.

9:26Francesco Sciortino:So I'd love to hear about like what has been achieved? What have you done to kind of move the needle? Yeah, there are things both internally and externally. On the internal side, I would say what Proxima is selling is speed in a domain where people don't expect speed. So the growth of the team, now almost 200 people, the locations in Munich, working with the Max Planck near Zurich with the Paul Scherrer Institute, near Oxford with the UK Atomic Energy Authority, that creation of speed by leveraging also ecosystems, the creation of relations with public institutions, also on the financing side.

9:59So our agreement with the Bavarian government came all very fast. So selling speed is fundamentally the key thing that a startup slash scale up can give and that you won't get from a major corporate at the end of the day. On the internal side, that then reflected on prototypes, on getting to the lab, on iterating in the lab and showing that that same space in a thousand meters squared of floor, every month it looks different. That's the kind of stuff that we've delivered over the past three years. On the external side, there have been some really big changes as well. In 2022, at the end of the, I think in December 2022, the National Ignition Facility in California showed net energy gain on a physics level from a laser fusion system.

10:44And that, from my perspective, was a spectacular physics achievement, not very relevant from an energy perspective. It's a system that does not look at all like a power plant. but it's sort of normalized also from a denture perspective the fact that it's okay to invest in fusion because clearly things are happening. More relevant to the commercial side in my view there have been companies on the magnetic confinement fusion side that have shown magnet technology that is scaling that is achieving performance that is of the kind that you need for a commercial system and these devices, Tokamaks hostelerators are not that far away.

11:24they're a generation away from a first-of-a-kind fusion power plant and that triggers every day more people to think okay it's not 30 years away anymore and was there a key unlock that enabled

11:39Francesco Sciortino:that agreement with the Bavarian government I mean this is I think it's like the only public-private partnership that we've seen in fusion so it really was like a historic agreement to come out can you touch a bit on like what what gave them conviction how that partnership came about and what it means for you going forward? Yeah, I think one day if everything goes well, we'll write a book about it because all might have happened and it is a sort of new playbook. And I think a playbook that Europe can use again and again. And it requires strong, smart institutions. You know, we have a bit of a habit of complaining about public institutions.

12:13In this case, we found a state government in Germany that was ambitious, that understands that there is risk involved because we haven't built a power plant quite yet but they understand the construction of an ecosystem and i think this is what this is the work that they've been doing in munich for two decades at this time and that is now seeing deep tech picking up speed in in munich again and again so the bavarian government put this 400 million euros of non-dilutive capital knowing that fundamentally they can't lose so much because clearly we have a legit player involved There is a national laboratory with 1 ,200 people working on magnetic confinement fusion.

12:52That's the Max Planck Institute for Plasma Physics. There is an ability that was already suggested, let's say, of raising private capital. At that time, we had raised 200 million. And we basically said to them, look, you go one step, we'll go the next step. And then we have an open advocacy for the German federal government of doubling down together with the Bavarian government. and then there is the European Union and then there is work that we're doing with individual states within the EU but also in the UK, also in Switzerland. So we think of ourselves as thoroughly European, not just about a small bubble but connecting with those public institutions, understanding that you need to have a policy team that isn't, you know, not a tech company one might think is just engineers.

13:39It's a lot more than just engineers. Without the engineering quality, you don't go very far. The product must work but in creating these relationships you need a lot more and so that that agreement came up

13:50Francesco Sciortino:i think it was announced february this year how how impactful was that to the kind of fundraising conversations that you were having at the time it was the trigger we had that agreement signed with the bevan government with the max blank and with the energy utility called rwe we signed on the 26th of february and the day after we started fundraising it's amazing three months afterwards we signed the term sheet. We did not wait just because we wanted the three-month mark. It just happened to be exactly at the three-month mark. And, you know, three months for a round like this is objectively pretty fast.

14:25And it just meant that people were ready until the very last minute. We were trying to make sure that we didn't just take some money, but we got the investors that we wanted. So, you know, we had the RWE investing. Obviously, the story here has a lot to do with RWE and the site, the ex-Fission Power Plant site in Gundramingen that we're working with RWE to reuse. But there are other players that maybe were less obvious. It was very exciting that Google stepped in as a strategy partner. Then we had KFW helping, supporting with some of our existing investors. And we have some family offices from Europe.

15:06One spectacular thing is that with this timing, we managed to actually keep the round fundamentally very European. And Proxima sees itself as a company that is bound to Europe now and forever. And having a European capital base at the earliest stages was very important to us. Moving forward, we want the best capital from wherever it comes. We want the best partners to scale and build a global company. But in this round, having it with such a strong European base was fantastic for us.

15:33Francesco Sciortino:I mean, it's amazing. It's almost unheard of it rounds of this scale and size to be so dominated by european investors normally rounds of this size you kind of become reliant on most of the capital to come from predominantly the us when you look forward to capital requirements how does this change you know you've got i guess like still 10 years perhaps until you build the kind of the first you or you've got the commercial plant up and running you mentioned that type of capital changes a lot as you scale and grow. Right. What do you need to achieve in the next five years to unlock this new type of capital to then enable you to build the things that you need to build?

16:11There is a key hardware milestone ahead of us. We call that the Stellarator Model Coil or SMC. That's a full scale, full power magnet of the kind of high temperature superconducting technology that we've been developing. This is in manufacturing as we speak. So it's not a theoretical thing. We have built many prototypes along the way over the past two years or so. And now that magnet is really coming to life. The moment you see that magnet, Seb, you will have many, many questions as to whether this stellarator, the net energy gain alpha, can be made. The answer will be, oh yeah, it can be made.

16:48It is very complicated, but it can be made because I see that magnet. So that magnet is the key technical de-risking that says, yes, you can get a QI accelerator. So a donut with magnets that is fundamentally steady state, so it operates continuously as long as you want it, and that is intrinsically stable. So it doesn't have the disruptions, the instabilities that a tokamak inevitably has. So that SMC is coming by the end of 2027. And then we jump and we go ahead with the Alpha build out for which we have a site. We have a conceptual design. Half of the company works already on Alpha. So we're not waiting.

17:30We're moving in parallel.

17:31Francesco Sciortino:Amazing. As aggressive as we can. In parallel, we're also working in Gundramingen with RWB on preparing the site for that first-of-a-kind fusion power plant. If we don't do these things in parallel, there's zero way that you can have a power plant by even the late 2030s. So you've got SMC, which is the kind of, it sounds like the final hardware barrier. Like if you can overcome this, if you can build this, you can then get to Alpha, you can then get to Stellaris, but at the same time, you're essentially doing all three at once. That's right. In different ways, the key priority for the company is the SMC.

18:08It's the mountain that we're trying to move. if we move that mountain there is a next mountain which is alpha that is enabled by the first one and if you move if you get to the top of alpha then you got Stellaris they are in this sequence but that doesn't mean that you you you can only go to the next mountain before you after you finished the previous one some are necessary and you released i think it was the Stellaris concept design was it last summer that's right it came up it came out in february 2025 it was It was created in mid-2024. So it took us only about a year from foundation of the company.

18:44We did our pre-seed and we said to our investors, over two years, we're going to develop a team of 30 people and we're going to get to a thing like this, this coherent view of what a power plant concept looks like. And then it actually only took a year. And that was because of the people that we managed to hire fundamentally. And then it took six months and more of peer review process. So we decided to publish it. so anyone can google Stellaris and they see what that looks like obviously we didn't give away anything we didn't want to give away so no our colleagues overseas or in the east can't quite get the sweet sauce for what we're doing but it was important to build the reputation to say

19:23Francesco Sciortino:this is now possible and what was the reaction externally like the Wall Street Journal called it the quoting Dennis White of MIT the most important thing infusion in the previous decade so it was warm let's say the realistic answer to our stellaris paper was impressive didn't know that that was possible it's still bloody complicated so great that you but it's possible will it be something you can simplify and that you can build because the objective here is not to show you know this is not a science project so the objective is not to do a one-off it's also not to think of stellarators as something that you put into a factory and you just make stellarators in the tens of thousands.

20:04This is not an auto industry kind of large scale manufacturing. This is gigawatt scale power plants. But you do need them to be commercially compelling if you want to see the scaling that we want. So when we talk about 50 reactors by 2050, that requires them being very, very compelling, lowering the energy cost of the entire energy system. It doesn't need to be as cheap as photovoltaics, but it needs to complement photovoltaics such that the overall system becomes cheaper so it is about cost and schedule at the end that we have to deliver and what about supply

20:38Francesco Sciortino:chain because there's i think there's like supply chain risk on some of the magnets you know this these high temperature superconducting or hts magnets right which is like one of the key ingredients in the stellarators how much is that a concern in terms of the supply chain risk there a lot of work on supply chain so this is you know an ecosystem of technology with this multidisciplinary it's it gives you a bit of a headache the moment you how you do this um good news is that so much of that supply chain is european for us and even if you were to try to build this machine in boston you would still end up with european supply chains the one exception as of today is that high temperature superconducting tape so this material that we buy at the moment the strongest suppliers for that are in Japan.

21:25China is rising very fast. And there are a number of players across Europe and the US that are trying to scale up, and we intend to help them. But that's still the domain where Europe needs to get up and run. The fact that Japan is so strong at this topic is great. We have very important supplier partnership relationships with those key players. We want to make sure that this becomes a sovereign, clean, safe, cheap enough energy source. So of course, that does mean that we need Europe to punch and show that we can actually make the vast majority of this machine in Europe. And that's totally possible.

22:02Francesco Sciortino:And it's an amazing story because there's such a push to sovereignty, but it feels like your supply chain is predominantly European. The company is predominantly European. The cap table is predominantly European. At the moment, it feels like a very rare example of where Europe has actually built something that is really sovereign and can really replace non-sovereign energy alternatives. There are cases where we really feel like we have a shot. It's amazing. And it feels like you combine that with the public support, the government state support that you've got as well. It's a really phenomenal story, even though you're only three or four years in.

22:36It's truly phenomenal that we have public institutions that actually are smart, that are actually listening. They have lots of complexity because that's part of the democratic process. You can't just think that this bunch of founders look great and they, I cut you a billion check. That's just not, it shouldn't work like that. So there is clearly a lot to show in order to be able to build this public private picture. But the momentum is clear. Most importantly, the technical concept needs to work. Because in Fusion, there is a lot of noise around things that feel cheap, but too bad. They don't actually work.

23:14So you need to make it work before you make it cheap. The other way around doesn't really promise that much success. But yeah, all the stars seem to be aligned for us to build a European fusion champion. And there will be an American fusion champion and there will be a Chinese fusion champion. The question is, can we hold our foot here, build that European sovereignty picture and then export stellarators?

23:36Francesco Sciortino:Yeah, that'd be amazing. Well, Francesca, thank you so much for joining me. I think what you and the team have achieved in such a short space of time is amazing. And you've still got a long journey ahead, but it's going to be amazing to see how you progress in the next 18 months to get the SMC over the line and then Alpha and Stellaris and everything. So thank you so much and best of luck with it all. Thanks so much.