Developing clean, cheap and abundant energy is about as ambitious as startup dreams come. But spend 30 minutes with Francesco Sciortino, CEO and cofounder of Proxima Fusion, and you might just believe it can happen in your lifetime.
Proxima, which has raised over €27m from VC firms Plural, Redalpine, Visionaries Tomorrow and UVC since it was founded in early 2023, is developing nuclear fusion reaction technology that could usher in a new energy era.
It’s an area that Europe has already invested hugely in — and could become a world-leader in, Sciortino tells the Sifted Podcast. “Fusion [offers an] opportunity of building a more-than $100bn company, and we just have to get into it,” he says. “There aren't that many opportunities like that.”
The company, which spun out from the Max Planck Institute for Plasma Physics, employs about 50 people and has attracted engineering talent from the likes of Harvard, MIT, Stanford, Tesla and Google X.
Read on for a lightly-edited, shortened version of our conversation — covering the engineering milestones Proxima needs to hit, when; and how its technology actually works — or listen to the full episode of Startup Europe — the Sifted Podcast here.
What is fusion power?
Fusion is the ultimate source of clean energy. It's effectively the source of energy of the stars; when you look up in the sky at night, it’s what you see shining in every star. It's a fundamental process that happens when light nuclei hit each other; if you basically take light atoms and you smash them really hard, there is a probability that they will fuse, hence the word fusion. And when there is such a fusion reaction, some of the mass is changed into energy. You might know the famous Einstein formula E = mc2. Basically, we're transforming mass into pure energy.
What we are after really is creating a star in a laboratory; if we manage to do that, instead of just receiving sunlight and trying to convert it into electricity in this indirect form that we are familiar with, we could actually access the source of energy itself. That would be an enormous change for humanity; if we are able to enter the fusion era that's nothing short of a new civilisation.
The problem with fusion is that it's fantastic in so many ways — it's clean, it's abundant, it's safe — but it's very hard. We've been thinking about fusion for 100 years, active fusion research has been ongoing for 60 or 70 years, and now we are approaching a very different phase, where there are more and more private companies chasing this.
You’re developing a reactor with something called stellarator technology. What is that, and what advantages does it have over other ways of producing this energy?
A stellarator is basically a doughnut with magnets that surround the doughnut, which create a magnetic field cage. The word stellarator stands for ‘stella’, or star; it's a very catchy name for a machine that creates a star on Earth. By creating this magnetic cage in a helically winding magnetic field geometry, we can confine hot, ionised matter — which is called a plasma — and that's basically what the stars are made of. But this matter that we're trying to confine needs to be brought to 100m-150m degrees — that's about 10 times [hotter than] the centre of the Sun.
This has been done, across a number of fusion approaches. But at Proxima Fusion we believe that stellarators are really the best approach for bringing this plasma to the right temperature, 150m degrees, at as high of a density as possible, and keep it there in a stable form for as long as we want. We're chasing a source of energy that can produce electricity in continuous operation — a substitute for the coal power plants, the oil-based energy production that we are so used to — we can really change it all.
How long might it be until this kind of technology is powering the podcast booths at Slush?
We are just a few years away from seeing net energy production in a commercially relevant system. But going from that to energy powering the Sifted podcast will take, I believe, more than 10 years.
Now every fusion company has to reach a milestone of demonstrating net energy production — and it's already happening. We are a couple of years away from a company in the United States called Commonwealth Fusion Systems showing a tokamak producing net energy for 10 seconds or thereof. Then the next stage is showing an approach that can take that and extend it over a longer time — so that will mean net energy production from the plasma. The next thing we have to do is show we are producing more energy than the energy that goes into the full engineering system. But the real milestone that we're after is making cheap energy, and that's yet the next level. That’s not 10 years away — it’s an evolutionary step for humanity, so it takes the time that it takes.
[Fusion] may be the biggest industry of the next century
The interesting thing is that we're now able to privately fund this, because the intellectual property that we are producing across so many concepts is extremely valuable. We are setting the basis for what will be maybe the biggest industry of the next century, or maybe for as long as humanity has. We’re living in that transition phase.
If we zoom out, what does your roadmap look like? And how do you develop that technology while also hiring people, making sure you have enough money, actually running the company?
It's a long journey, so it's important to focus on the milestones. At every milestone you get to unlock the next bit. For us, the key technology component that needs to be demonstrated is a weirdly shaped, high temperature superconducting magnet — a very complicated machine, which is a few metres wide and 20 tesla, which is an enormous magnetic field. We're aiming to demonstrate that in 2027. If we manage to do that, we think we have unlocked the path to net energy stellarators, and we aim to build that net energy steady state device in 2031. If we manage to get that, then the road is open to go and build the first of a kind fusion power plant.
We tend to think about the roadmap the other way around. In 2050 the objective of all fusion companies is to be relevant to the climate challenge, and if you want to be relevant in 2050 you have to talk about scale. Today, the world consumes 17 terawatts of energy, so to satisfy our ambition, we need to talk about the terawatt kind of scale. A terawatt is 1,000 gigawatts. If you have a gigawatt power plant, you need 10k of them to make 10 terawatts — so you need to think about 1,000s of power plants. The systems that we're talking about are gigawatt scale. If during the 2040s we are able to scale impressively fast like France scaled fission in the 1980s, then we need to get to a first of a kind fusion power plant within the 2030s.
So the milestones for us are: the magnet in 2027; the net energy steady state demonstrator in 2031; and a first of a kind as soon as possible thereafter.
How did you get investors behind this vision? Are they very tied to those milestones? What I’m interested in is how you have enough brain space to focus on the actual technology and hitting those milestones, when you also have to worry about all the other stuff, like, have I got enough money to pay people?
The vast majority of what we have to do is technology. Fundamentally, we are an R&D company. The key point is we are producing enormous value at the moment — in the know-how, in the patents that we're producing — at the basis of this enormous industry that we are seeing being created right now. The investors, of course, need to see the milestones, but it’s important to distinguish between a company that focuses on doing science versus a company that focuses on engineering. We are an engineering company. We think of what we're doing with stellarators as being fundamentally an engineering challenge, and you can put engineering on a roadmap, whereas you can't put scientific discoveries on a roadmap. To accelerate scientific discoveries all you can do is get really smart people and tell them, please have a really great idea.
We have an engineering project, and so part of what we get judged on is quality of execution, and quality of execution becomes much more reasonable as an ambition when you bring together some best-in-class individuals. We've done a really excellent job so far. I've also been quite lucky in finding incredible humans to join us, and that gives us hope that we can keep showing that progress at the right pace. Really, it's a game of outpacing everyone else.
How have you managed to attract these amazing people from places like Tesla and Google? How do you convince them to take the risk on a very early-stage startup?
It's the mission — and showing that they can join a group of people, each of whom looks like them, or better. Everyone wants to learn, everyone wants to be surrounded by individuals that they can be inspired by. There will be people that will write books about how engineering is being redefined in Proxima Fusion — and there aren't that many challenges that offer that much of a transformational opportunity. People are at Proxima to do the best work of their lives.
How much of an impact do you think Trump’s reelection will have… Do you think it might in some senses be a good thing for companies like yours, because lots of Europeans might think ‘Oh, we need to stand on our own two feet a little bit more, because we don’t know how it’s going to play out over there’?
I struggle a little bit to call it a good thing, but I do think that it's a wake up call for much of Europe. Over the last few months, people have been talking more about European competitiveness. The Draghi Report didn't really say much that people didn't know already, but somehow it resonated. We talk more and more about Europe having a sluggish economy; Germany, especially, really needs to shake itself up.
It's really a cultural problem that we have to fix in Europe. We need to get back to being competitive in mindset. We need people to get into the startup space and want to build things and not look for too much comfort. Competitiveness does mean getting out of the slow, big technology giants that have driven our economies for the last 100, 150 years. Fusion is that opportunity of building a more-than $100bn company, and we just have to get into it. There aren't that many opportunities like that. And the one thing that most people don't know is that Europe has invested enormously on this topic. We have spent twice as much public funding on magnetic confinement fusion compared to the US. If we don't step up and actually take what has been funded with taxpayers’ money and really step up our ambition, and say ‘We're not going to be second on this one, there is zero excuse to be second’. We really have to find those Europeans who have gone to the US, drag them back into Europe, tell them, ‘Come build the thing here, because the opportunity is here. You won't get a second chance like this’.”
Another well-rehearsed topic is how to get more of that amazing innovation, that's often pretty well funded in European universities, out and into companies. What made you want to build a company rather than stay a researcher? And what do you think we need to see more of to have more people do that?
I think it's a cultural investment that we need to make. We need to encourage people to look for simplicity more than anything else, to get out there and do things that matter, rather than more startups that have an interesting economic ambition, but don't really change society.
One of my cofounders, Lucio [Milanese] and I did our undergrad at Imperial College in London, a very entrepreneurial place. Then we spent one year at EPFL in Switzerland, another very entrepreneurial place. And then both of us did our PhDs at MIT, which is an entrepreneurial bubble in many ways. And then when I joined the Max Planck Society — one of the most renowned institutions for scientific quality, integrity — that entrepreneurial spirit suddenly was really missing. For us, there was just a very clear vacuum. We had a very clear idea — the idea of building what are called QI [quasi-isodynamic] stellarators. Nobody was jumping into it. There was a feeling that venture capital funding wouldn't be interested in this. And we were saying, ‘No, we know for a fact that this can be pulled off’. But somebody had to leave their position and just get it done.
"People who want to get into startups shouldn't tiptoe into them
Maybe that's one generalisation that I may afford to make: people who want to get into startups shouldn't tiptoe into them. You should leave your job and get it done. You will only know if it was a good idea or not based on whether it succeeds or not, but if you tiptoe into it, you already know the answer: it's very unlikely that you will succeed.
So how do we get more startups out of institutes like Max Planck? I think the effort of sending out their technology scouts is good, but those scouts need to educate people to think that stability is not that important at the end of the day. If you're a hyper educated individual with skill sets, you will get a job, you'll be fine. If you're doing the best thing that you could be doing, you'll be fine.
What's one lesson you've learnt, or piece of advice that you would give another researcher or engineer currently at a university or a lab who's discovered an amazing thing — other than, ‘just quit your job and go and do it’?
I wouldn’t say that every researcher should necessarily make a startup. There is a point in public research, and it's also important to distinguish what is science and what is on an engineering commercial path. So, if you're in love with astrophysics, please don't create a company to try to do that as a private venture. That doesn't necessarily make sense.
First, find mentors to understand the landscape. If you're doing a truly deeptech moonshot, and somebody asks you on your first day to give them the business plan, ignore them. You don't need a business plan on day one; don't let yourself be bogged down by writing too many documents. Focus on building the team, which includes mentors. Get out there, get initial financial support; bootstrapping is not a bad idea, but if you're trying to build a hardware company, get some money, because you can't just do this with four people — no matter how smart they might be.
