Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.
During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.
We encourage researchers to utilize the full-text search on this page to navigate our oral histories or to use our catalog to locate oral history interviews by keyword.
Please contact [email protected] with any feedback.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
In footnotes or endnotes please cite AIP interviews like this:
Interview of Brandon Sorbom by David Zierler on May 24, 25 & 26, 2021,
Niels Bohr Library & Archives, American Institute of Physics,
College Park, MD USA,
www.aip.org/history-programs/niels-bohr-library/oral-histories/47258
For multiple citations, "AIP" is the preferred abbreviation for the location.
In this interview, Brandon Sorbom, Chief Science Officer at Commonwealth Fusion Systems, discusses the development of his company and interest in nuclear fusion. Sorbom speaks about his time as an undergraduate student at Loyola Marymount University where he majored in Electrical Engineering and Physics and how he discovered his interest in fusion during this time. He describes how his interest in nuclear fusion led him to pursue graduate school at MIT. He details his time as a graduate student working at the MIT Plasma Science and Fusion Center, as well as his experience working with his advisor Dennis Whyte. Sorbom discusses how he first became involved in the development of SPARC, whose goal is to generate net energy from fusion, during his time at MIT. He details the variety of investors for his company and the roles he and his cofounders take on within CSF. Sorbom explains CSF’s current project of demonstrating that superconducting magnets at high fields can be used in fusion. Lastly, Sorbom discusses how fusion energy will likely become the dominant form of energy in the future and how it can help combat climate change.
OK, this is David Zierler, Oral Historian for the American Institute of Physics. It is May 24, 2021. I'm so happy to be here with Dr. Brandon Sorbom. Brandon, it's great to see you. Thank you for joining me today.
Yeah, thank you. It's my pleasure.
To start, would you please tell me your title and institutional affiliation?
I am the Chief Science Officer at Commonwealth Fusion Systems.
Now, does Commonwealth have any official or unofficial connections with the MIT Energy Initiative?
It does, yeah. So right now, I believe, we are still the largest non-governmental sponsor of research at MIT. So a company, for example, like IBM can come in and do a sponsored research agreement with MIT. We have a sponsored research agreement type of thing set up through MITEI and MITEI has a project structure set up where generally, energy companies come in and do sponsored research. And so, we were a spin-out from MIT, Commonwealth Fusion Systems. All the founders were from MIT, and we still have very close ties to the school. And in order to even start the company, we actually realized that to do something like Fusion, you had to be born at scale. And this was the way to do it.
We said, "We already have our home institution that we know and love, and we know all the people there," who, by the way, when we started the company, were months away from all being laid off. So we were able to save the lab from going under and at the same time, utilize all these resources that were sitting there to hit the ground running on Fusion and not have to start as three people in a garage.
To give a sense of the overall structure of CFS at this point, as Chief Scientific Officer, who reports to you, who do you report to, and who are some of the outside people that are important as advisors or overseers of CFS?
I report to Bob Mumgaard, who's our CEO and one of the cofounders of the company, along with myself. And as for people who report to me, right now, I have a team of something like 14 direct reports. I lead the R&D team and the team that's in charge of getting us our superconducting tape and a team looking at the scoping of ARC, which is our power plant concept.
Just to give a snapshot in time right now, what's going on? What's important at CFS? What are you dealing with in late May, 2021?
One of the big things that we're gearing up for that we're really excited about is, the first phase of our project was to demonstrate that superconducting magnets at really high fields can be used in fusion. To show that we can build these powerful superconducting magnets. And I'll take a step back from that. Tokamaks have been around for a really long time, since the 50s. And as we built larger and larger tokamaks, people figured out that the performance scales a certain way. It scales with size and field. But it happens to scale with field to the power of four. So basically, if you increase the field by a factor of two, you can build a device that's roughly a factor of 16 smaller than if you hadn't increased the field.
And so, that's a huge lever that you have on performance. And that was our hypothesis going in, that these high field magnets can fundamentally change the way that people do fusion. Because instead of having to build a device that's the size of ITER, which is the very large tokamak being built in France right now, you could think about building a break-even device that's the size of SPARC, which is the device that we're going to build in Devens, Massachusetts. And so, we know the physics because of the 50 years of tokamak physics, plus the heritage at MIT of doing high field copper magnet tokamaks. You can't use copper magnets for a power plant because of the resistive losses that you get in them.
But you can use a superconducting magnet for a power plant because you don't have resistive losses. And the trick, up until now, was that superconducting magnets out of low temperature superconductors could only achieve up to a certain field, and you couldn't break this field barrier, if you will, on building these magnets. But now, with high temperature superconductors, there's the possibility of building a very high field magnet. Which is still a very large engineering challenge.
So that gets me to where we are now, which is the culmination of three years’ worth of effort to build the prototype magnet for SPARC, that will unambiguously demonstrate that this magnet technology works. We can engineer it, we can manufacture it, we have the supply chain set up for it, it works from a technical sense and hits the performance that we said it was going to hit. That's coming up in the next couple of months. And so, there's a lot of preparations and things that are coming to a head right now. Everybody is getting together for this big test, where we get the magnet cold and turn it on for the first time.
For me, as a historian of physics, I'm well-positioned to know, and I'm sure you do as well, that the joke with fusion energy is that it's been just around the corner for the past 40 years. Which of course, begs the question, why is this different?
That's a great question. It's one that we're pretty used to answering by now. The reason it's different this time is that the size of the device that you can build fundamentally changes how fast you can iterate. If you think of it from a startup mentality, where you basically want to have lots and lots of iterations, the size that we're building things at this point already has led us to a place where we can build things really quickly, and iterate, and learn. So even though this particular magnet is the first of its kind, the largest magnet in the world, built out of superconductors, at least, and we haven't turned it on yet, it's not like we just put everything together the first time and prayed that it would turn on.
There was a whole bunch of R&D that led up to this point, where we tested sub-scale things, and it got bigger, and bigger. And even this magnet, which is the culmination of three years of research, only weighs about ten tons. Which may sound like a lot, but MIT has cranes that are inside of buildings that can move things that are ten tons. That's a couple cars. And it's something you can iterate on really quickly. It's not like you're going to need to build 100 tokamaks to iterate, but SPARC is at a size where you can build it fast, and there's an existence proof that SPARC is of the size of many tokamaks which have already been built. The world collectively has built about 170 of them. And there's about a couple dozen of them that are SPARC size, that have been built already, by governments or academia that have turned on and worked.
Obviously, again, not with high temperature superconducting magnets. That's the new piece of this. But at that scale and size. For example, the ITER magnets are enormous. Everything requires specialized tooling to move around. At the ITER site, they have some of the largest cranes in the world that were brought there to move things. Even shipping components to the site is a really big challenge. And so, to your question about how we're getting to fusion faster and how we're going to break the cycle of fusion always being 40 years away, it's by the small size.
To refine that question and zoom out a little bit, in what ways have there been advances, both on the theoretical side and the side of instrumentation, experimentation, even computation, that make now this time finally right for fusion energy?
I would say that from the physics side, just the fact that we were able to design this machine from first principles to start with is a reflection of the fact that we have 50 years of plasma physics, specifically tokamak plasma physics, that we've been able to boil down to first principles, both first principles understanding and generalized rules of thumb that we can use to even do the thought experiment of, "Can you make this design work?" On the experimental side, we have empirical curves that such as one that has the size of the device on the Y axis, the magnetic field on the X axis, and then there are contours of what we call constant Q. Q is the plasma physics gain, so fusion power out of the plasma over heating power in. So obviously, you need a Q of 1 to break even, and then higher than that for a power plant.
And it's a very simple-looking plot. It's just contours on an X, Y plot. But the reason that we can draw the contours where they are on that plot is because there have been 50 years of physics that have been done that allow us to draw those curves and even do the thought experiment of, "What if we went to a higher field?" There’s a quote attributed to Richard Feynman to the effect of, "If you can't explain your concept to a high school or undergrad physics class, it's not fully baked yet." And that's kind of why I feel like we're sort of at the point now with tokamaks where all of this physics and experimentation has gotten us to the point where, obviously, it's still a very complicated subject, there are a lot of intricacies, and people spend their entire careers going into the details of the really cool physics behind it, but at the end of the day, it's at a point where you can boil it down, explain it to people on a simple curve, and say, "Here's a contour of gain." Everybody understands what break-even is. And you can see very clearly that higher field equals smaller device.
On that point exactly, I'll share with you that our research audience at the Niels Bohr Library includes a lot of high school students, who will definitely flock to this interview as an example of all of the cool stuff you can do if you get a PhD in science. And so, for that crowd, can you explain the connection between plasma physics and fusion energy? What's the connection there?
Sure. So the connection there is that, in order to have fusion occur in a way that you can make power from, you need to have a plasma. You can make fusion happen by, for example, using an accelerator and shooting a beam into a target. From the physics that we know today, you can't use that process to generate net energy. You can make fusion happen, but you're always driving it and putting way more energy in than the small amount of energy you'd get out. And so, I think the answer to that question is, if you look up at night and look at all the stars, all of them are made out of plasma, and they all run on fusion.
And a plasma is a very, very hot gas. It's an ionized hot gas. And by hot, I mean that the particles are moving very fast. The only way to get the positives to come together and get close enough to fuse, since positives want to repel, is to have them be hot enough that, by definition, the state of matter that you're in is a plasma.
I wonder if we can talk a little bit about the research culture and the mission of CFS a little more broadly. So in your work, obviously, there's a basic science component to it, just understanding how nature works. There's an applied mission: "What are we going to do with fusion energy once we have it?" And then, of course, there's a public policy component because we need, very badly, energy solutions, and fusion energy is right at the top of that list. So either as a pie chart or in the overall way of thinking about these things, how do you see all of these issues play together at CFS?
Well, first of all, it's an interesting question because I think the pie chart is fluid. It's not a static pie chart. It kind of flexes depending on where we are. And so, if you take a snapshot of that pie chart today, I would say it's very much on the applied side with the large superconducting magnet we have built over the last year and are testing this summer. That is a hardcore engineering project. There's also a team that's doing the physics design of SPARC, which is coming up to another big milestone that we have around the same time that this magnet is going to be turning on. And then, Kristen and the Movement Building team are out laying the seeds to get the public policy in place so that once we show that fusion's viable, it's not like we're starting from the ground up, trying to convince people that it's a good thing and it's not, like you said, 40 years away.
Basically, by the time we have the demonstration, it's not just the demonstration that's proving that fusion works, it's that it's almost confirmatory, if that makes sense. It'd be really great if our demonstration wasn't people being like, "Oh my gosh, it actually works?" and instead saying, "Oh, OK. We were pretty sure it was going to work, but they demonstrated it for the first time." So right now, I'd say a large chunk of the pie is towards applied. But that's going to expand and contract. As we get closer to building our plant, I think the policy side is obviously going to expand quite a bit. Actually, after we've built SPARC, the basic physics side is going to expand quite a bit. SPARC's main mission is to achieve break-even, but it's going to be a device that we're going to learn a whole bunch of physics from, physics that will inform how we refine designs to build ARC, which is the power plant.
And so, the way that we've staged it is, our mission is to get to fusion as fast as possible. So in some sense, as part of the way that we've structured the designs and how everything works, we're actually doing things in parallel. So we're already designing ARC, even though we don't have the answers from SPARC yet. We have another really nice, simple plot on this that kind of shows extrapolation out to ARC. If you think of error bars, if there was a line on this plot, and there were a broad range of error bars, SPARC helps us collapse the error bars. So we know basically what shape ARC is going to look like, and we can start designing the big pieces of ARC. And as soon as we build SPARC, we can add the fine details into what will become ARC.
As you say, there are certainly new physics to be learned. And on that point, what stands out at you right now that's not yet understood, and how will the successful outcome of this project get us to understanding this new physics?
So one of the core philosophies of our entire plan is built around not having to discover any new physics. We'd like to. We would definitely like to benefit from it if we do. But you could say that the physics design is actually pretty conservative. Some people in the plasma physics community have criticized SPARC as too boring and conservative. And our response was, "Well, SPARC isn't going to put electricity on the grid, but it is producing as much heat as a power plant would. And if you have a power plant, you would really like that power plant to be boring. You don't necessarily want a power plant to be exciting and shut down on you all the time." It's not that there's a safety risk, but for example if you had a solar panel that turned off randomly, it'd be very exciting, but it wouldn't be a great solar panel.
So we've designed SPARC somewhat conservatively on the physics end, and we've made up for that by saying, "With these magnets, we have a big engineering challenge ahead of us." But we're a lot more confident that engineering problems can be engineered around and solved by clever people as opposed to banking on finding a law of nature that you might or might not be able to get around. If you're doing exploratory physics, and you're hoping for one answer, and you get another answer you can't break the laws of nature. But engineering-wise, you can engineer structures and things like that that can be clever to give you more leeway on the physics side. And that's been our philosophy. But that being said, the physics that we hope to discover on SPARC are around the performance of the plasma.
Without getting into the nitty-gritty details, a lot of the physics we hope to learn indicate that there are regimes in the plasma that will operate very well at high fields, that the plasma might actually perform better. Again, we're not taking credit for that right now because we want this to be boring. We don't want to be surprised in the wrong direction. But we are setting ourselves up to be able to take advantage if we do find that these regimes that we think exist are very viable at high field. If that's the case, ARC will be even better than we think it will be right now.
In the broader world of fusion energy, where is CFS situated? In other words, who are the competitors out there? Who are the collaborators? Is there a race to get there first?
I think everybody wants to get there. I'd say there's a race to get there as soon as possible. The faster that somebody somewhere on earth gets to the solution, the faster humanity wins. I, personally, am a little biased, but I think CFS has the best shot of getting there the fastest. But in general, there are a lot of other fusion companies out there, which we view as a really good thing. We don't really see it as cutthroat competition, trying to force our competitors out of business. We see it as a good thing and an affirming sign that more and more companies are popping up out there, trying different ideas and populating the ecosystem. So I would say it's more of a collaborative environment than a competitive environment with the other companies. And there are over 20 different companies. We actually have a Fusion Industry Association. And you asked what the landscape looks like.
So we're the only company that is building a conventional aspect ratio tokamak. There's one other company in the UK that is building what's called a spherical tokamak, which is sort of in the same family as the conventional aspect ratio tokamak that the world has built 170 of. People have built about a dozen spherical tokamaks. And so, the physics is a little bit different, but there are a lot of similarities between the two, a general shape of things. The fact that they use a toroidal array of magnets, just like we do. So there's that company, there's a company called Tri Alpha, which is a linear device, out in California. It's a cylindrical device that also uses magnets and uses a completely different configuration from a tokamak. Another one of the larger companies is called General Fusion out in Canada that has sort of a piston design, where they use a liquid lead liner, and they compress a plasma inside of this liquid lead liner.
And then, it releases energy and pushes the pistons out. And there are actually surprisingly few companies that are building the exact same type of fusion concept. There's a whole zoology of plasma confinement topics. But I would say the reason that we're focusing on the one that we are is that, by and large, that's the one that the global academic community has looked at and essentially endorsed as the best shot. But until we found this new high temperature superconducting material, it was, unfortunately, also one of the longer shots in terms of timeline because these tokamak machines were so big. We knew how they worked, and we could build them, but they were so big that the time between building machines was taking longer and longer.
And so, all the founders of the company were from the academic community and worked on ITER (the large tokamak being built in France by the world academic fusion community). And we saw this opportunity with high temperature superconductors and said, "Hey, we think we could actually shrink this, and fundamentally change how fast it could go through iteration cycles, and build these devices a lot faster, but use the same physics that the most of the world's fusion community is using."
Given the fact that humanity wins with a successful outcome, that makes it not just a national but a global priority. Why would this effort be the domain of relatively small private companies and not something that the full might of the DOE can accomplish at a place like Argonne National Laboratory? Not asking you to speak on behalf of DOE, but what are you doing that DOE can't or won't?
Well, I think there's a very complicated answer to that question. So the broad brush answer to that is, right now, I think the DOE is very much tied up in ITER. No judgment associated with that, but ITER is a very large project that takes a lot of resources. And it really takes sort of the full focus of DOE at this point. And so, DOE is still very much a part of ITER. But I actually think it's not necessarily a bad thing that the private companies are trying–this is sort of why public and private partnerships exist. You have a company that has a different risk profile than, say, the government does and can try one of these ideas like, "Hey, let's build a super high field magnet out of HTS."
And the risk profile for a company is such that when we were first developing the technology, we definitely designed experiments in such a way that was different than my experience, coming out of grad school, where you kind of wanted everything to be perfect the first time. SpaceX, for example, was able to blow up three rockets before they finally launched one. Not that it was a good thing that they blew up three rockets. I'm sure they didn't celebrate when the rockets exploded. But the nature of that being a private enterprise is such that it wasn't killing the project. In a public project, if you do something like that, you potentially risk the entire project. Everybody's very cautious, and generally things are slower. There are definitely times where that's a good approach, but there are also times where you'd like to do the SpaceX approach, go really hard really fast, and try things out, even if you risk failures early on.
And I think it's great that we now have astronauts on the International Space Station that got there using a rocket that was developed by private industry in collaboration with NASA. The reason that SpaceX has the Falcon 9 is because of a program called COTS, where NASA did a cost share with SpaceX and said, "We have a vested interest in being able to use American rockets to send astronauts to the Space Station. And so, we'll put some money in if you're also willing to put some skin in the game and put your money in." And it worked really well.
So we have modeled a program after the COTS program that Brandon just mentioned, and it actually passed in an omnibus bill that passed Congress and was signed by the President in December. And it essentially would set up a similar program to COTS for commercial fusion. And that passed, signed by the President. And now, we need the funding for it. So that's kind of the next step, looking for the appropriation to fill in that project. And I would just add that there is kind of a global race among governments to get commercial fusion up and happening. And we've seen that with China, who's kind of thrown hundreds of millions of dollars into their program, as well as the UK. Recently, we saw an investment from the UK government for about $250 million. So there is a global race in addition to what's happening at the industry level. [Ed. note: Cullen is Head of Public Affairs and Communications at CFS and joined Sorbom for the interview.]
Thank you for that, Kristen. Brandon, as a corollary to why not government, it also begs the question, why not major industry? As impressive as what CFS is doing, it's a relatively small outfit. Why are we not seeing this happening at Tesla, or Google Labs, or IBM? Why does it have to be at a smaller scale? What's the obvious advantage there?
My personal belief on that is that there's sometimes a really big benefit to having a small team that can move really fast. That you don't necessarily always want to have a huge team. Organizational inertia can happen not just in the government but large corporations as well. And so, I think there's a time and place to have a very small, dedicated team that's laser-focused on one thing. Instead of being a small fish in a much larger pond, you have an entire organization that has one mission, and that mission is to get to fusion. And it's interesting that some of the investors of the company are actually very large companies. Interestingly enough, and this is publicly disclosed, two of our big investors are actually oil and gas companies in Europe. So there's Eni, which was our first investor in our series A round, and recently, Equinor, a Norwegian oil company, has also joined us.
And both of those companies have said, both to us in private as well as publicly, that they see themselves as wanting to be energy companies. And if they could make energy in a way that didn't harm the planet and didn't produce carbon, they would switch to it immediately. And they're investing in a lot of renewable technology. The CEO of Equinor publicly came out and said that they were going to be completely carbon neutral by 2050. It’s pretty amazing that a major oil company basically said, "We're not going to be an oil company anymore." But instead of starting their own division, they've invested in us and have a stake in us because I think they also see the value of having a very laser-focused, mission-driven organization that they can contribute to, invest in, and have a stake in and access to, but have that be its own thing.
The politics of that statement obviously makes a lot of sense. Nobody wants to be contributing to the problem. But obviously, it also indicates, from a business viability point of view, that this is actually something that's going to come to fruition, that the technology is going to be there to provide the energy.
Yeah. And we thought really long and hard when we first started. We said, "Do we want to take money from an oil company?" And I think we want to because after we talked to them, the investors that signed up with us have really pure intentions. They would like to transition into not being oil companies in the future. But also, the other big advantage of having them as investors is that they've built big infrastructure projects. And so, having them on not just as a financial additive to the company but as advisors. We actually have had employees from the oil company be seconded to our company.
We had somebody who was a project manager in a large project at one of the oil companies who came and worked with us for an entire year. She had a sabbatical from her job in Europe, flew out to the US, and worked with us for a year. And we got a lot of value out of working with her and getting the experience that she had, working on these large projects. So I think it's been really good to have. And we have a wide range of investors, too. In general, our philosophy when we were looking for investors was not just, "Find whoever can give money to us," but, "Find who can contribute more than just money." Whether it's people who are really knowledgeable about the energy landscape, or in the oil companies' cases, companies that have built large projects, in the venture capital communities, people who may not be experts in our technology but are experts in funding companies that build high technology things.
A lot of the VC firms that are investors in us invest in hard tech (also known as tough tech), where it's hardware development, not just software development. Where you actually have people building things you can hold in your hand. And so, it's been a really cool thing, from my perspective, to see all of the input we've gotten from our different investors. We've gotten help in choosing our site from some of our investors who are experts in real estate. And so, it's been really cool actively working with them, and not just having sort of silent partners that put money in. Having them help contribute to getting fusion on the grid.
Because so much of CFS is still provisional in terms of projecting confidence in terms of what it can do with regard to energy fusion, how much are you relying on computation to simulate what it will do before the whole project actually gets up and running in the real world?
Again, with the analogy of narrowing the error bars, we know the general shape of how things are going to do. There's no computational modeling we're doing from first principles where it's like, "We have no idea how this system is going to respond, but we're going to try and model it." But there is a lot of what I would call detailed computational work going into two areas. One is, there's a lot of collaborations we have with the academic community. It's one of the nice things about being closely linked with MIT, and being from the academic community ourselves, that we've always been of the opinion that, even though we're a private company, and there are some things that have to be IP protected and all of that, we're a very open company. We actually published the entire physics basis of SPARC for open access in a large plasma physics journal so that anybody who wanted could read our plasma physics basis.
And I think that combined with us having our heritage in the academic community means that we have a lot of close collaborations with people in that community who do a lot of these advanced simulations. And so, the machine does not rely on any of these advanced simulations working or not, but it's certainly very helpful to have access to these resources to help us narrow down exactly what we think we're going to see. And I think it's really cool for our collaborators, too, in four years, once we've built SPARC, to see how close the simulation results were to reality. The other area that we do use a lot of computational resources is actually in the engineering.
There are a lot of really cool computational tools, multi-physics modeling for the superconducting magnets, for the structural, mechanical, and electrical properties of these things, as well as for the tokamak itself, doing structural analysis. The tools that we have now, we've actually hired some people, for example, from the auto industry who worked on crash test simulations. Crash tests are really expensive to perform. And so, the auto industry has put a lot of money into making very, very accurate simulations of exactly how one of these tests go, so that you can do a bunch of simulations, and maybe instead of doing ten tests, you only have to do two tests. You always want to do a real test. You never want to do something solely on a computer. But we have people with that these skillsets who are used to modeling very advanced engineering problems.
I wonder if your embrace of open publishing makes CFS immune from industrial espionage.
[chuckles] There's definitely intellectual property in how we make the magnets. That's an industrial process that we've developed and patented. But as far as the plasma physics, I don't think there's any way to espionage our physics bases. We've published it. We intentionally published it, so that people could look at it and critique it. Because we said, "If we're wrong, we want people to say so."
You mentioned some partners domestically, and you mentioned ITER. Globally, who are the most important, either, countries, companies, or individuals that you're working with to see all of this research come to fruition?
Oh, man, it's a long list. I'm worried that I would unintentionally leave somebody off because the list is so long. But I guess I can speak to my experience, some of the places I've collaborated with. Again, this is not a complete list. But when I started, actually, one of the places that we collaborated with quite a bit was New Zealand. Turns out that there's a center of excellence for Superconductivity in Wellington, New Zealand. We still have extensive collaborations with them. Back when I was able to travel, I was lucky enough to visit there several times.
There's another group in Switzerland that we actually use to do critical testing on the sub-scale prototypes of our magnet cables. They have a facility that was actually built for ITER, that basically did stress testing. They poked, prodded, and beat up the ITER cables to make sure they would survive. And we were able to adapt their facility to test our cables as well. And we're still collaborating with them to this day. On the plasma physics side, I would say there are several groups all over Europe as well as the US. I'm not going to do it justice by getting more detailed than that. But there are many countries in Europe we have collaborations with, along with all the national labs and universities in the US. A lot of people on the team have either worked at those facilities, or people in grad school spent a year or semester abroad working at one of these facilities. There are a bunch of tokamaks in Europe.
I wonder if you could help me square the circle between your embrace of sharing this information, sharing the research, and the fact that you have investors where there's a proprietary issue, and they want to see a profit at the end of this. How do those two approaches work together?
The really tough engineering challenges are something we are patenting and developing the processes of how to make these magnets. So I think that's one piece of IP that we have that is very valuable to the investors. In order to get investment money, you have to have a product at the end. And the product is going to be some piece of a fusion power plant. We're probably not going to build the steam turbine that you hook up to it, but we are going to build a substantial portion, if not all of the fusion part of the plant. And a lot of the technology development that goes into that is for the actual nuts and bolts of how to build a particular type of magnet configuration are things that we can patent. And that way, we can make a return for the investors who've put up the money to get this off the ground in the first place.
A patent only protects the intellectual property. It doesn't necessarily guarantee that the idea will be monetized. Obviously, your investors believe in it, otherwise they wouldn't have invested in it. But who will be the end-use consumers who ultimately will make this technology profitable?
I think that's something that we're still developing. Power development, even just within the US, is an insanely complicated field, how power is produced and how people make money off of electricity generation. But I think at the end of the day, the way this is going to be profitable is that it's going to be a source of energy that's cheap to produce and can be sold at a rate that's competitive with current sources of electricity. At the end of the day, that's how this whole thing makes money. The details of which parts we end up building, and owning, and controlling are still up in the air.
Your answer suggests, though, that your thinking very much in terms of infrastructure and not necessarily particular industries. Meaning that the end users are going to be municipal governments that control or regulate the grid as opposed to particular industries looking for a competitive edge over other people in that industry. Or not necessarily.
It could be one or the other. We've talked to people in different industries who would definitely love to have a fusion power plant powering their process. I think even in the last year, things have accelerated a lot, where companies broadly are very much more aware of their carbon footprint. Compared to ten years ago, when it was a curiosity if a company was thinking about being very environmentally conscious and their carbon footprint. Now, everybody is thinking of that. So I think the answer is, hopefully both. The goal is to get as much electricity production made with fusion as possible.
And this would be an on-site proposition? So that, for example, when GM starts rolling out of all of its EV line, and they say, "We're doing this, not fossil fuel cars anymore," what a great PR and environmental coup if they could also say, "We're powering our factories with this technology." Is that to suggest that what you're creating is something that can be transported on a tractor trailer and set up on-site?
I wish that were the case. It is more of an infrastructure build. So you would actually have to build something. It's not like you're going to have a portable generator that you could wheel around with you. That being said, our goal is to have the technology small and modular enough that setting it up wouldn't require an enormous, multi-year sort of civil works project, and you could actually bring all the components in on palets and shipping crates and put the pieces together. If you look at how a modern gas turbine plant comes online nowadays, it's very modularized, and everything is on skids.
You drive in with a bunch of semis, you unload everything with some cranes, and you hook it up. And you can set up one of those plants in under a year, compared to an old coal plant, which can be a five-, six-year project to set up these big civil works and huge buildings. So unfortunately, you can't just roll up in the truck and plug it in. But we don't want it to be the same scale as building a dam or something like that. Ideally, this is something that you'd be able to set up pretty quickly.
Meaning that if GM wants this to happen, it would encourage the state of Michigan to get the infrastructure up and running to make it happen.
Yeah, or GM could even, if they wanted to, say, "We'd like to have one of these plants to be dedicated to our facility." And again, power development world is extremely complicated, and these things are never that simple. But in principle, you could have what they call an “off-taker” in the industry. If you find an off-taker who's willing to pay a certain price for the power, you can set up a contract before you've even built the plant so that we would come in and build a plant, and then we agree to sell X percent of the electricity of that plant to GM to build their electric fleet. I think that's definitely in the cards.
A political question. As you've seen the transition from the Trump to the Biden administration, where the Trump administration, of course, was not particularly supportive of clean energy initiatives, and of course, the Biden administration very much is, that at least is the media narrative. But I wonder from your vantage point, either in the way that investors talk about the overall political situation and how likely it is that this is really the way of the future, are you seeing these changes in leadership in Washington filter down to the things that you're doing in your day-to-day?
Honestly, I feel like fusion is somewhat apolitical. I realize that's a little bit of a cop out. But it really is an issue that, I think, whatever side of the political spectrum you're on, regardless of what your politics around climate change are, I feel like everybody can get behind fusion. Even if, say, you didn't believe in climate change at all, the idea that somebody could make a power plant that runs on basically water, you filter the fuel out of a little bit of water, it's just a good business proposition. Even if you didn't care about climate change at all. And so, I feel like, personally, I haven't seen it become a political issue at the ground level. And I think that's really a good thing. Ultimately, the goal of this technology is, really, to broadly benefit everybody. And I think the less political it can be, the better. I would love people on both sides to champion it and love it bipartisan-ly, and it becomes an issue that everybody lifts up. So from my perspective, it doesn't have a partisan lean one way or another.
Last question for today, a question we've all been dealing with, how has both the science and the administration, the business of running CFS, been affected one way or another by the pandemic and remote work?
That's a great question because during the pandemic, we grew the company from about 40 people to about 160 now. That’s pretty fast growth even outside of a pandemic. And so, it's definitely been an interesting process to hire and bring on so many people to the team while we've been having to work remotely. I should point out that that's just on the company CFS side. At MIT, there's always been a large team that's been there. But on the CFS side, it's grown quite substantially. We were lucky, if you can call it that, that the pandemic hit when it did in our design cycle because we were able to kind of shuffle things around to allow us to be smart about how we sequence the work that had to be hardware versus design work.
And so, when everybody got sent home, we were able to shift things around so that people could do a lot of work on their computers with the understanding that, as people got back into the office, there would be a push towards hardware work. And we were lucky enough to be able to sequence things that it didn't really affect our overall timeline. It affected, on kind of a week-to-week, month-to-month level, how we structured things. But as far as affecting getting to fusion on the grid, we got really lucky not to be impacted too much by COVID, I would say.
And now that we're starting to peek toward a post-COVID reality, vaccinations, low positivity rates, things like that, are people starting to go back in? Is there a hybrid approach where there's a new normal, where not everybody defaults to going in every day?
I think we're still figuring that out. It's not unique to us. All businesses are figuring out what this new world looks like. But definitely, there are a lot of people who have been coming back in. I have to say, personally, even though I don't necessarily touch hardware much anymore, I definitely work with it a lot, and it is really nice to be back in the office, just the energy that you get from a bunch of really bright people all working in the same space, building really cool stuff. It's something that's very hard to replicate on a Zoom screen, as hard as we try.
There's just no substitute.
Yeah.
[End Session 1]
[Begin Session 2]
OK, this is David Zierler, Oral Historian for the American Institute of Physics. It is May 25, 2021. I'm delighted to back with Dr. Brandon Sorbom. Brandon, it's great to see you again.
It's great to see you, too.
All right, Brandon, today, we are going to do some oral history and take it all the way back to the beginning, following yesterday's terrific discussion, where you provided a tour de horizon of fusion energy specifically and CFS as well. So let's go back all the way to your origins and start with your parents. Tell me a little bit about them and where they're from.
Wow, OK. So I grew up with my mom. She grew up and still lives in Torrance, California, which is a city in LA County. And so, that's also where I grew up. I was born there and raised in California, so I was a little bit of a beach bum growing up. I went to elementary, middle, and high school out there. Ended up going to college at Loyola Marymount University.
How many generations back does your mom go in California?
Her parents were there, and they were immigrants from Europe. So my grandpa on my mom's side was Swedish, and my grandma was Belgian. And so, both of them came out after World War II and ended up coming out to San Francisco, and then moving down to the LA area. And that's where they stayed. I guess they liked the warm weather and the beach.
What was your mom's profession?
My mom was a teacher, actually. Still is. And so, before I was born, she taught grade school. And when I was born, she'd always had a private piano practice. So she continued doing that. Previously, she'd taught music and math in elementary and middle school. After I was born, she continued teaching piano. And then, when my two brothers and I got a little older, she went back to teaching music at school.
Did you have a relationship with your father? Or did you grow up with a father figure at all in your life?
Yeah, so my father left when I was around 10. So I have memories of him, but I actually don't know that much about his family history. I did have a pretty strong father figure, though. Because my dad left relatively early when I was a kid, my mom's brother (my uncle) was a really strong father figure to me. And I think I got a lot of my love of engineering from him. He was an electrical engineer. He also started his own company. Until three years ago, when we started CFS, I never thought that I would end up starting a company. But it's kind of funny, looking back, that I did have a very strong role model in my life who did start his own company.
I was wondering about that, if there was any early influence that might've given you some guidance.
There must have been. But it's funny because I never remember thinking, "I'm going to start a company." But it must've had an influence on me because I think it's guided me to where I am now.
Tell me a little bit more about your uncle. What was his business?
So he and his partner ran a computer business. They started, I believe, in the 80s. And they did kind of everything. It's hard to nail it down. Normally, now, you say a computer business, it would be very specialized, somebody who repairs computers or runs servers. But they really did everything. His first job was at GE. And he was a field engineer at GE. He worked on a whole bunch of systems, and I think he carried that sort of mentality to his business, where he and his partner actually wrote an operating system in the 80s that, for a while, became standard for a certain subset of business computers. They also repaired hardware. They did a lot of hardware work on the really big computers and mainframes.
Later on, when it became a thing, they also ran web hosting and email hosting services. So they kind of ran the gamut of the spectrum. But I think, for me, something that was really formative was that he would take me to his job sites. One thing they would do is go on site and repair people's systems, especially systems that were sort of legacy systems that a lot of people didn't know how to fix. And so, he got to go to some really interesting places because of it. University labs, some heavy industry that he would take me to. I remember once, one of his clients was a well casing maker. They took raw plates of steel and turned it into wells. It was this big factory in downtown LA that had all these computer control systems. I think I got to see a wide range of different engineering disciplines. I remember also going with him to the Mount Wilson Observatory and watching him work on their computer systems.
I wonder if that planted a seed in you early on, specifically thinking about science from an applied perspective, how this could help people, how it could benefit society.
I think so. And I think a lot of that also came from my mom. It was kind of a service-oriented culture. My mom was a teacher, and so teaching was a big thing. And music, specifically, was a big thing. And so I grew up in a culture of teaching and of giving back to society. Both my mom and my uncle went to my alma mater, Loyola Marymount, which is a liberal arts university. And I think both of them took very seriously, as I did, that you could have a technical degree, but also have more of a liberal arts education, kind of a very holistic view of everything. I think that definitely influenced both of them and played into the choice of, ultimately, me wanting to go into fusion. Actually, a little bit of a tangent, but I just remembered this. My middle name, as well as my grandfather's, is Nils. You said your audience was the Niels Bohr Library.
That's great. What a connection. Did you go to public schools throughout?
No, I actually went to a private Catholic school for elementary school, and then for high school, I went to a public school. So a little bit of a mix.
Was that religious influence important for your mom?
Yeah, I would say it was a lot more important for my mom than for me honestly. [laugh] But it was a good school. I think I got a good education there.
Was the church a big part of your childhood?
Yes, and no. I was never very religious, but the musical and community aspects of church were big in my childhood. So I was in a lot of musical groups. I was in choir for a little bit, and later on I was in a band that played at the church, stuff like that. So I was always involved kind of in the music community.
And then, what were the considerations, switching into public school later on?
I think I wanted to branch out. There was a private school, but the private school also cost a lot of money. And by that time, my father had left, and my mother was supporting both of my brothers and I by herself. I was the one who made the decision. I think my mom wanted me to go to the private school. But partially for financial considerations, and partially just because I wanted to experience something new, I decided to go to the public school. My mom, I think grudgingly at first, supported me. But the public school that I went to was great.
And this was the Torrance School District?
Yeah, I went to West Torrance High School.
Was it a strong curriculum in math and science?
Yeah, I would say so. I think that was definitely an advantage of going to a larger school. I guess that was another consideration going from the private to public school, the private school was quite a lot smaller. The public school had a lot of AP classes and things like that. So I was able to kind of jump into the more advanced math and science classes. And I had a lot of really strong teachers in both math and science.
Did you have laboratory opportunities in high school? Could you tinker around?
A little bit, yeah. I took high school chemistry, I took high school biology, high school physics. We had classrooms that had sinks, tabletops that were somewhat resistant to people beating on them, spilling chemicals on them, and such. There's nothing I remember that was crazy, but we were able to tinker. I remember in physics class, for example, we built little trebuchets.
Were energy issues on your radar at all, even in high school?
Honestly, not really. In the background, of course. I was in high school from 2001 to 2005. So I remember being aware of it, but it was never really at the forefront for me. Going through high school, I knew I wanted to go to college. I wanted to do engineering and music in college. I knew I wanted to do that. That was sort of about as far as my horizon went. I didn't really know what I wanted to do after that. I just knew I wanted to go to college.
Between grades, financial constraints, and geographical considerations, what kind of schools did you apply to? What was in reach? Who gave you advice?
So I only applied to the one school. And I was very fortunate. So I did have good grades. I was, I think, one of the top GPAs, and I was very fortunate to also win a scholarship that was an interesting, sort of idiosyncratic scholarship by a gentleman named Mr. Granoff, who owned a large tract of land that was later purchased by LAX. And so, he became fabulously wealthy when the airport bought his land. And he established a scholarship, and every year, he would give a full-ride scholarship to one student from each of the four public high schools in Torrance. And the only stipulation was that the school had to be in California. But it wasn't a fixed dollar amount, it was, "I will pay all of your expenses through college, as long as you go to a place in California."
Wow, that's pretty good. What was your sense of how you distinguished yourself? It could've have just been good grades. Lots of kids get good grades.
I was definitely very involved in the music scene in high school. I was a band geek, I was in the pit, drum line, and marching band. I played and taught piano. That was part of my mom's influence. I should mention that my grandma was also a music teacher. So I'm sort of a third-generation piano teacher, even though I don't teach anymore. So in high school, I had several elementary- to middle school-aged students that I taught piano to, kind of overflow from my mom's teaching schedule. So I was very involved in the music scene. Basically, every single band opportunity at school, I was part of. And my sophomore year, when the school established a swim team, I joined that and eventually became captain. So I was relatively involved in sports as well.
You mentioned earlier that you specifically wanted that broader liberal arts education. So that makes me think that even though a place like Caltech was in Southern California, you did not want that experience. You did not want that more narrowly focused technical education as an undergraduate.
Yeah, I definitely thought about it. The considerations to LMU were partially that I heard lots of stories from both my mom and my uncle of how much they loved it there. I did look at other schools, but in the end, even though it was a terrible strategy to only apply to one school, it worked. I decided, "I want to go to this school, and I have a scholarship, so I'm only going to apply to this school." I figured since I had a scholarship, I would probably get in. And I did. Although again, probably not a great strategy to emulate.
And what was the game plan? A dual major in engineering and music?
The original game plan, as a starry-eyed freshman, I was like, "I'm going to major in everything." And so, the original plan was to major in electrical engineering, which was what my uncle had done, and music, which was what my mom had done, and sort of combine those two. And I did that for my freshman year. Also, my freshman year, I ended up joining the rowing team. The school has a D1 rowing team. So I ended up walking onto that, a little bit on a lark. I joined the rowing team, and that, plus music, plus engineering was a little bit too much. And I felt myself starting to burn out sophomore year. So, with a heavy heart, I dropped the music major. I ended up taking piano lessons for a long time. That was the one thing I kept on doing a while. There was a great piano teacher that I had a really good relationship with there. But I dropped the music major halfway through my sophomore year and focused on electrical engineering.
How much physics was there in the electrical engineering curriculum?
It's interesting you ask. There was a decent amount of physics, and my junior year in undergrad, I'd been injured in rowing. I injured my hip. And I had what's called a “redshirt” year. So if you have to sit out a year for NCAA eligibility, you can basically take a fifth year and play your sport for the fifth year. And I was really into the rowing team, and I really wanted to take a fifth year. And there were just enough physics classes that I'd gotten in electrical engineering that I said, "This is really cool." And I was on track to finish my EE degree in four years, and I needed something to do for my fifth year. I couldn't just row. And I thought, "This is a great opportunity to take more physics classes." So I ended up taking a fifth year, doing an engineering physics double major. So effectively, because I'd done all of the engineering part of the engineering physics, I just loaded all of the physics courses into my fifth year and pretty much exclusively studied physics my last year in school.
How much specifically did you study nuclear physics?
Almost none at all. In the general physics courses and our quantum physics courses, of course, we talked about nuclear. But there was no specific nuclear physics course that I took. So in the summer between my senior and super-senior year is when I actually started thinking really hard about my future. And Loyola Marymount has always been big into aerospace because El Segundo is right next door, all these big aerospace companies are there. SpaceX, which started in El Segundo, was there. They might've moved to Hawthorne by then. But still close by. And so, that was kind of the new hot thing that people were doing, graduating out of electrical or mechanical engineering and going there.
And so, I thought about that. But then, I also started to think of energy. And that's when I kind of just started to do a lot of reading. And there was more and more momentum, during college, towards clean energy. And I thought, "This is a really good thing for the world." And I kind of made a decision that summer, between my senior and super-senior years. I was kind of between aerospace and energy. I ended up wanting to go to energy. And then, I did a lot of reading, and within energy, I found fusion in the back of a physics textbook. Last chapter, it said, "Here are some miscellaneous topics that you might think are cool." And I discovered fusion. And I was hooked. I read it, and I said, "Wow, how have I never dove into this before?" So I started reading a lot more about fusion.
And that summer, actually, I ended up stumbling upon this internet hobbyist community called Fusor.net, which is an entire community devoted to building these little devices called Farnsworth fusors. Instead of using magnetic fields to confine the plasma, in a fusor you use electric fields to confine and shape a plasma. And while there are a few people who are still trying to make those into an energy device, it's sort of been first principles proven that unless you discover some crazy new physics, that's not going to work. But the advantage of those devices is that somebody with a couple thousand dollars, or who's really good at scavenging spare parts, can build a device that actually makes fusion reactions occur. Not a lot, and not anywhere near break-even. But you can actually do fusion in your garage with one of these devices.
A lightbulb went off for you.
Yeah. And so, I said, "Hey, this is really cool. You can do this for a couple thousand dollars." You have to take safety precautions, you have to make sure that you shield against the x-rays that the machine produces. There's high-voltage hazards you have to be respectful of. But I figured, "Hey, I'm an electrical engineer. I can talk to people in the physics department. We can be safe about this." And I ended up convincing one of the physics professors to give me a little lab space they had over the summer as kind of a trial period, which then went into the next year. And I was able to get a couple thousand dollars from them. And I also scavenged a bunch of parts off of an SEM that the school was throwing out. I managed to cobble together, with the help of the lab manager there, my own fusor.
Were there any professors as an undergraduate who you would consider mentors, who exerted a particularly strong intellectual influence on you?
Within electrical engineering, there was a professor, Dr. Marino who was a mentor to me. She was the big influence and I went to her a lot for advice. There was another professor named Dr. Page, who was the EE professor who was known for kind of being the guy who was tough but built up great engineers. And he was a big influence on me. And on the physics side, my mentors there who really encouraged me to go into fusion, and really, the whole physics department, but especially Dr. Bulman, who was the head of the physics department, who gave me the lab space. There were two professors, Dr. Sanny and Dr. Berube, who encouraged me to actually do a self-study in plasma physics. They set it up with the administration. Before I did the fusor project, I worked a summer with them in astrophysics. They studied data from the GOES satellites. We studied astrophysical data, looking at plasmas in the magnetosphere. And so, they encouraged me to do plasma physics research. And the school wasn't big enough to have a dedicated course to plasma physics. But both of them knew it and said that they would guide me along in my self study. Finally, he wasn’t a professor, but our physics lab manager, Anatol Hoemke, was a mentor and was instrumental in getting the fusor project I mentioned above working.
I wonder if LMU was sort of the best of both worlds in the sense that it was small enough so you could develop real relationships with professors, but big enough that the professors were engaged in research, they were not just teaching professors.
Absolutely. I definitely had very close relationships with a lot of my professors. And I'm leaving out quite a few. I can't name all of them. But I would say it was more of a fluke if I didn't go to office hours than if I did go to office hours regularly for all of my classes, just because the classes were small enough. My electrical engineering graduating class was, I think, nine or ten people. And my engineering physics class was three people. The pure physics class was a bit larger. But it was relatively small class sizes. I think there were 100, 150 engineers total.
As you're coming into academic maturity, are you in touch with your uncle? Is he enjoying seeing this process?
Yeah, we exchange long email threads. He always has questions about the technology. He loves hearing about all the updates that we're doing with superconductivity. Every time I go back home, we have long discussions about what I’m doing out on the East Coast.
You mentioned one fork in the road in terms of career considerations, aerospace and energy. And obviously, you chose energy. But another fork in the road is industry or graduate school. Did you think, at some point, about not pursuing the PhD, that you had entrepreneurial inclinations, and you might want to jump into industry right after undergrad?
Well, when I was still thinking about aerospace versus energy, and even before I branched off into fusion in energy, I did think, in both aerospace and energy, about going straight from undergrad into industry. I never thought about starting my own company. For energy, my uncle had gone to GE, and he had lots of stories about the cool stuff that he'd done there. Of course, GE has a huge power division. And so, that was a consideration that I was thinking about. And then, on the aerospace side, because we were so close to the aerospace companies, we had a lot of impromptu field trips where we went and talked to engineers from those companies. So I did think about going into industry. But when I got into fusion, I realized at that point, I think the only fusion company that existed then was Tri Alpha. And Tri Alpha's a lot more open now, but back then, they were a lot more secretive.
And so, there was virtually no online presence with them. The commercial fusion landscape as a whole was nonexistent and very opaque. So, really, the only way that I saw to get into fusion was to go into grad school. The academic route was the only way, and it was sort of the obvious path to me. If I wanted to continue with fusion, then that sort of de facto meant that I would be going to grad school.
This fortuitous hip injury that gave you the bandwidth to read up on nuclear fusion, I guess this is a bit of a yin and yang question, but to what extent do you need to get smart about nuclear fission before understanding nuclear fusion, since obviously, the fission side of things is much more developed and has different applications?
So the actual physics behind it is sort of taught together. When you look at it from the most basic sense, which is quantum mechanics, the two subjects were taught as sort of a pair. Fission is breaking apart an atom on a nuclear level, and fusion is putting two atoms together. And in fission, you need a really big atom to split, and in fusion, you need very small atoms to go together. But it's really just two sides of the periodic table that you're looking at the same quantum processes. But past that, I would say, the engineering considerations are very, very different for fission and fusion. So I actually didn't study very much about fission outside of my own personal interest in it because I thought it was interesting. Because in fission, you're dealing with the chain reaction. You're always trying to moderate this chain reaction and keep temperatures under control.
Fission is a lot of thermo, and fluids, and engineering around that, whereas fusion is a lot more electromagnetics-focused. So when you start to get into the applied realms, they're actually very different. Even at MIT, in the nuclear science and engineering department, they've changed the curriculum since I was in grad school, but when I was a first-year grad student, everybody took the quantum course. But then, the fusion students would take an advanced electromagnetics course. And then, I believe the fission students would take an advanced thermo course. So at the applied level is where things split, just because the engineering systems of fission and fusion are so different.
Obviously, you're too young to remember Three Mile Island or Chernobyl, but I wonder if Fukushima registered with you when you thought, "Maybe there's a better way."
Oh, absolutely. Actually, it's interesting you bring that up because Fukushima happened my second year in grad school. And actually, MIT, specifically the nuclear science and engineering department, was actually very much a public presence. There was a lot of misinformation swirling around. And I don't know exactly how it started, but the department actually helped set up a blog that was run by grad students, answering questions from concerned Japanese citizens who wanted to know what was going on and maybe didn't trust the news that was coming out of the government. And MIT was seen as this source of truth. And so, the grad students stepped up. I was only a very, very small part of that because I wasn't on the fission side. But a lot of my really good friends in school contributed in a very big way to answering questions.
And even though I was mostly watching that from the sidelines, I thought that was a really interesting and cool way to see outreach from grad students. And people really reaching out and trying to do good in the world by informing these people, who were terrified of these things, that radiation from Fukushima is scary because you can't see it, and really calming people down, and informing them of what the actual risks and hazards were. But yeah, I was already on the fusion track, but I was like, "There has to be a better way to do energy than this." Because even though, in principle, you can make lots of systems to try and make fission safe, and there are lots of people with ideas for Gen 4 advanced reactors--a couple friends from grad school even started companies to make advanced fission reactors--you're always dealing with this chain reaction. You're always having to manage it, and there's always this risk of the chain reaction to do something like Chernobyl or Fukushima.
That's just baked into the process, no matter what safeguards you put in place.
Exactly. It's baked into the process. And so, in fusion, it's baked into the process that it's inherently safe. And that's why fusion is much harder to achieve. Fusion is so safe because it's very difficult to keep all the atoms together long enough to fuse. And if you shut down your confinement system for a second, the whole system just goes poof and disappears. Like we talked about yesterday from a power plant perspective, you obviously don't want your power plant going down intermittently. So you want to make sure that confinement works. But from a safety side, if your confinement fails, your plant just shuts itself down. The default state of fusion is off, whereas I would sort of say the default state of fission is always on. You can't really turn it off.
How much materials science education did you receive as an undergraduate? Not just knowing how to put stuff together, but understanding, particularly with new technology, that you'll need a mastery of new materials and understanding how they work. To what extent was that part of your curriculum and education?
I would say, outside of the basic courses that we had in engineering, where all the engineers got sort of a basic mechanical engineering course that did have a materials component to it, and all the physics students did have their basic physics that talked about materials science. But I never had a materials-specific course in undergrad. It was really grad school when I started taking my more focused, very heavy materials science courses.
As you explain it, you had to go to graduate school, essentially, because that was the only path forward in nuclear fusion. But the motivations for going to graduate school, did you accept the possibility that that would put you on an academic, scholarly, professorial path? Or was the idea you're going for the PhD because you did have this entrepreneurial instinct, you did want to go into industry? It could be yes and yes.
Well, it's funny, I did consider trying to go into industry. There were starting to be some fusion companies when I was in graduate school. But really, the paths I was looking at were more sort of industrial plasma physics positions. There are a lot of plasma physics used in other fields, like the semiconductor industry, for example. Obviously, that wouldn't have been fusion. But in order to stay directly in fusion, up until we had the idea to start our own effort, it wasn't really on my radar that there was a commercial fusion route. It was really more being a professor or becoming a research scientist at a school or a national lab.
Because, for example, MIT has staff research scientists, career scientists that don't necessarily teach classes. Their main job is all research. And so, those were, really, the paths. And the reason I laughed is because I had a booklet of all the different paths I could take. And I remember, every summer, I would update it. And being an entrepreneur was never one of the paths in that book. I had all these things I could be. Not even in fusion. I could become a consultant, there were some people who would get into finance. I never really considered that, but I put it down as an option. But I never even had entrepreneurship as an option, which is kind of funny.
So it sounds like, at the very beginning, the plan is simply basic research, understanding the science behind nuclear fusion. Once you have that, see where it takes you.
Yeah. And I was working at the MIT Plasma Science and Fusion Center, which was very much involved in ITER and the international efforts around ITER. And so, for the first few years of grad school, my path was going to be, if I worked on fusion, it would be tokamaks, and it would be in an academic or national lab setting, and it would be in support of ITER.
Now, was MIT the be-all and end-all for nuclear fusion? Were there other programs you considered?
Yeah, there were. I knew, especially with my experience building the fusor in undergrad, I wanted to go that a place where I could work on a machine with my hands. I wanted to go somewhere they let the graduate students touch the hardware. And so, there are only a couple actual schools that have large fusion experiments in the United States. So there's MIT, Princeton, Columbia has some experiments, Wisconsin, Washington, and UCLA used to have a very large fusion experiment that they shut down. They now have basic plasma physics experiments. I looked at them. And then, also, even though it's not on campus, UCSD is very close to a tokamak called DIIID, which is actually about the size that SPARC is going to be. Of course, with copper magnets.
So I looked at all of those schools. And the first time I actually applied, I applied to five of those schools and was rejected from all of them. So I had saved up a little money through piano teaching and working construction a couple summers when I was in undergrad. I went out to visit a few of the schools, and the appeal of MIT was, when I took a tour of the labs, there were graduate students directly working in the lab. There were people climbing on Alcator C-Mod, which is their tokamak, and installing parts. I said, "This is the place for me. These are my people." And so, I figured, "I didn't get into any of the schools I wanted to.
If I'm going to try one more year to get into a school, I might as well try to get into my top choice." So I flew out to MIT and paid to take sort of an extension course there. Some businesses will send people to school to take a single course. You can pay to register and be in a course. So I did that. It was kind of the last remaining money in my bank account. Lived pretty close to the ground for that year. But I just knocked on a bunch of doors and said, "Hey, can I work as a technician in your lab and just work on stuff? I'll work for free. Just give me something to do." And my advisor, or who later became my advisor, Dennis Whyte, said, "Sure, I'll hire you as a technician." And actually, there were also two research scientists in the nuclear science and engineering department at an accelerator down in the basement of one of the buildings.
And so, they and Dennis paid me to be sort of a contractor technician and help repair a couple of the accelerators. They said, "We need an electrical engineer. You have electrical engineering training. So let's see how it goes." I did that, talked to a whole bunch of people, and then reapplied. And of course, being on the ground, and having access, and talking to all these people, I knew what all of the specific things I could do to make my application stand out and be better were. And I was able to get letters of recommendation because I took the course. And so, I asked the professor I took the course from, "Could you write me a letter of rec if you thought I did well in the class?" So I reapplied, and I was able to get in. And I was there for six more years getting my PhD.
Now, besides the attraction of the research culture, it's great to see graduate students doing the actual experiments, being hands-on, I wonder, was part of it administrative at all in the sense that the program is nuclear science and engineering, so it's really the perfect combination of the things that you're interested in? That administratively, that academic track just made a lot of sense to you?
Yeah, and I think that was what led to that culture, too. Because some of the other programs I looked at were very much more science-focused, and because they were so science-focused, there wasn't really an expectation or an opportunity for the grad students in those programs to build things. They would do experiments, but they would design an experiment, then the staff engineers and technicians would implement that and build it on the device. Because the students weren't engineering students. They were, ultimately, science students.
And so, I think you're exactly right. This combination of nuclear science and engineering that we had at MIT, combined with the fact that the experiments that we had weren't just part of one department–the Plasma Science and Fusion Center at MIT is its own lab that has multiple departments. There are people from electrical engineering, mechanical engineering, aero/astro, and nuclear science and engineering that all work together in this lab doing fusion. And physics, of course. But there's a mix of a whole bunch of different types of disciplines that contribute to this lab, which made it a very hands-on place.
Now, this lucky break that you got with Dennis, where he let you work in his lab, what was his research at that point?
That was before he was the head of the Center. So he had just gotten tenure as a professor in nuclear science and engineering, and his focus of research was on something called plasma material interactions. And so, it's basically what happens when this super hot state of matter touches a material. And first of all, it turns out that there's a very intuitive misperception that when a plasma touches a surface, like in Spider Man 2 or something, it's this hot thing that will melt through walls. But in reality, plasmas are these very delicate things. Even though they're super hot, 100 million degrees, there's so little stuff in there, that you could blow a puff of air onto plasma, and it would extinguish it because the air is a million times denser than the plasma.
And so, it's almost like if you dropped a drop of boiling water on an iceberg. The boiling water wouldn't melt through the iceberg. It would freeze immediately because the iceberg has so much more mass. There's so much more cold stuff than hot stuff. And so, for the plasma, it's the same way. Even though the plasma's 100 million degrees, when it touches a material surface, which is even denser than air, it will cool down immediately. But in the very thin layers of that interaction, there are some really interesting materials science things that go on. And you do actually have, at the nano or micro level, these effects that happen when the plasma touches the wall. The wall will actually spit things out into the plasma.
And so, if you start with a plasma that is a purely deuterium plasma, and the plasma touches the wall, you can scrape off particles in the wall that go into the plasma, and then there are interactions that happen. So it's a very rich area of research, plasma material interactions, because it combines a whole bunch of different fields. There's materials science, there's plasma physics, electromagnetics, all these things that combine. That was Dennis's field of research. So the reason that we had particle accelerators is, one of the things we would do for experiments–there were two ways that we could use the beams. One of them is that, in our lab, we had some small plasma devices, where you make a plasma, and you intentionally shoot a beam of plasma into a target. And then, you use a beam from a particle accelerator to actually interrogate that material.
So you shoot a beam from a particle accelerator into a material, and you measure the reactions that come off of it. And from that, you can infer characteristics about how the surface is changing. And same in the tokamak. So the experiment that I ended up working on, instead of building a little plasma device at the accelerator, you actually take a tiny accelerator to the tokamak, and in between pulses of the tokamak, you would actually shine the accelerator beam inside and scan the material surfaces in the tokamak, and then read out what happened. You do a pulse of plasma, the machine turns off, then you scan the beam and see how the surface changed.
To give a sense, if at all, how interdisciplinary this project was, would this be on the radar of theorists like Ernie Moniz or experimentalists like Bolek Wyslouch? Would they be interested in what was going on? Or was this a different world, essentially?
They've both been to the lab, but they're very much in the nuclear science realm. And plasma physics, at least at MIT, was kind of its own beast. And so, I'd say there were collaborations, but they didn't necessarily actively work on C-Mod, which was the big fusion device. One of the big ones. There were a few other experiments we had while I was in grad school.
I'm not sure the origin story of the MIT Energy Initiative, but does it go back to your earliest days in graduate school?
I feel like I should know exactly when it was started…but I don’t.
But the more important answer is, it wasn't something that was immediately obvious to you. You didn't see that as part of your research agenda, at least in the beginning.
At least in the beginning, no. It was started by Ernie Moniz, and I do remember it was one of these things that was more and more gaining attention. I would see advertisements for it on campus, and the graduate students working with it would have events, and things like that. But I would say the MIT Energy Initiative got really big around the time we were starting to very seriously think about our plant design as well. And I'm probably a little bit biased because that was when we also started interfacing with them a lot. But it was around the same time, I would say, that the MIT Energy Initiative became a much more public, outward-facing thing from the school. I remember in graduate school, there were protests and sit-ins at the president's office, that the school needed to do more to combat climate change. And the MIT Energy Initiative already existed, but they ramped up their efforts quite a bit during that period as sort of a response, I think, in part to a lot of the student activism at the time.
To go back to Dennis, were his research inclinations more on the applied side or the basic science side?
I would say, really, it was a split. Of experiment versus theory, Dennis is definitely an experimentalist. But as far as science versus applied, I would say it was definitely a split. He's very much a scientist in that our experiments were diagnostics looking at understanding sort of basic physical phenomena. But at the same time, Dennis also was very applied because, for example, he taught the class that the ARC paper came out of. He would teach a design class every two years that was a very, very applied engineering design class of, "Take this new novel piece of technology and apply it to tokamaks."
And on that applied side, what were some of the societal issues this research might have solutions for?
Really, almost all of it went back to power production. Different ways of looking at fusion power plants. There have been lots of design studies that have happened over the years of people thinking about what a fusion power plant could look like. But as we learn new things, those designs evolve. And you could start with an idea that was maybe a little bit too out there for the rest of the academic community to really take seriously enough to write a paper about, but it was great for a graduate class because that's the place where you explore new ideas that may or may not pan out. And that's how you get a really great idea like the use of HTS. People had talked about it before, but the idea of, in a big way, using that to actually make a device much, much smaller was in the heritage of the Center already because we were always about compact, high field fusion with copper magnets. It was almost a natural extension to say, "OK, with a copper magnet, you can't make a power plant. But now, there's actually this material you could use to make a power plant. What would an instantiation of that look like?"
Last question for today, obviously it's a big leap to go from Dennis saying, "Yeah, sure, you can be a tech in my lab," to accepting you as his student. What was that process like? Obviously, there's that intellectual connection between your interests and what his research is. But there also needs to be that interpersonal connection, that click that the two of you will work well together. How did that process go about?
One of the things that I really loved about having Dennis as an advisor is, I felt like he was the perfect blend of involved enough to give guidance and mentorship, but also gave all of his students enough space to actually grow with our own ideas. It was not a lab where you were handed a project like, "Here's a list of instructions. Carry out these tasks and come back to me." It was more of, "Here's a problem statement. You figure out how to solve this problem. Come back to me with questions." Not just throwing you out in the wind, but, "You own this problem, and it's your job to figure out how to solve it. I will help guide you through that process." And so, I really enjoyed that part of the relationship with Dennis because it really gave a lot of freedom to come up with things and try new things. Sometimes, you would fail because you'd try a new thing, and it turned out not to work. But that's how you learned.
So Dennis really appreciated the teaching value of failing. You learn a lot, even from things not working.
Yeah. And just like with my uncle, I wasn't explicitly thinking about entrepreneurism, but it was almost like my graduate program was a little bit entrepreneurial because like I said, there was a problem statement, but you, as the student, had to kind of figure out what the research plan was around it. And there was support, but at the end of the day, it was your thing that you were trying to put together. You'd have to give committee presentations, which felt a lot like what our pitch presentations are like now in front of investors. You have to go with a case of why you should go in a certain direction. And looking back, it was a very entrepreneurial way of doing research.
That's a perfect place to cut it for today.
[End Session 2]
[Begin Session 3]
This is David Zierler, Oral Historian for the American Institute of Physics. It is May 26, 2021. It's great to be back once again with Dr. Brandon Sorbom. Brandon, good to see you again.
Good to see you, too.
What we're going to do today is set the intellectual stage for what ultimately would be your thesis research. And so, I wonder if we can put it on a spectrum with regard to the intersection of your interests and where they fit in overall with Dennis's lab. So perhaps, on one end of the spectrum, you have your own ideas, they're very well-formulated, and you're trying to figure out how to make the research at the lab work for those goals, and on the other end of the spectrum, all of those goals and all of the things that you're interested in are already baked into what the lab is doing, and it's a matter of carving out your space and how you're going to further the mission of the lab. So I wonder if you might situate us intellectually in how your research thesis developed along that spectrum.
Sure. It was actually an interesting progression. So I think I mentioned the other day that the initial thing I came to MIT to work on before I was even a student was a particle accelerator. We were using a type of accelerator called a radio frequency quadrupole. So I was working with two graduate students, Harold Barnard and Zach Hartwig. Zach is actually an MIT professor, and he's one of the cofounders of CFS. And Harold is a research scientist at Lawrence Berkeley, still working on accelerators, on the Advanced Light Source there. But both of them were using this accelerator for their thesis research on Alcator C-Mod, which was the fusion device at MIT. And the plan was for me to pick up where they had left off and continue their work. And so, that was great, I was really excited to do that.
But then, in 2012, everything fell apart. That's when all the funding for the lab, which had been receiving about $20 million a year from the DOE, ran out. And sort of overnight and unexpectedly, all of the funding just gone. Sent the lab into an existential crisis. And so, everybody was sort of scrambling to figure out how we were going to ramp down. Normally, funding is kind of ramped down gradually so people can plan. It's not just gone one day unexpectedly. But the funding had been chopped off, so we went into a mode of, "How do we get the money back?" And it wasn't even really about the research at that point, it was about, "How do we just continue existing as an organization?" And so, that's a whole other story. We managed to get the money back enough to basically ramp things down as they normally are over the next four years, which culminated in 2016 in our record-setting fusion performance. Basically, the very last shot we did on it, we turned everything up to 11.
So in 2012, it was very uncertain what was going to happen with Alcator C-Mod, which was, of course, where my thesis research would've been headed towards. So I was a little bit in limbo. And for the next, I would say, year and a half, I ended up just doing development work to say, "OK, maybe we can adapt this accelerator to put it on another tokamak somewhere else," but that's really complicated because you'd have to take it apart, ship the whole thing, and put it back together. It was already a pretty finicky device. The accelerator we were working on was a free accelerator that was donated us through another random sequence of events I think from a national lab. And it never really quite worked properly. It was a prototype device, and there were always little things breaking on it that we had to fix.
Anyways, long story short, over the next year and a half, we pretty much determined that with C-Mod being shut down and this accelerator not really working great that that probably wasn't a viable way for me to do my experiment. Because we didn't want to spend a lot of time fixing this machine not to have a tokamak to use it on. So in this time, I was figuring out what I was going to do and how I was going to pivot. I had a bunch of ideas, we had the class where the ARC design came out of also in 2012. It was an exciting year. And so, I knew I wanted to do something that had to do with HTS and ARC. And at that point, I had a couple of years to experience working on particle accelerators.
And so, I knew how to operate them. There was actually another one that I fixed up in the basement of one of our lab buildings that nobody was really using. It was this really great tool that we didn't have much of a use for at the time. And so, I ended up basically coming up with, "How can I somehow combine ARC," which at that point wasn't part of my thesis at all, just a side project I was doing, "with the experimental work I'm doing with particle accelerators and come up with a thesis topic?" So after a lot of going back and forth, I realized that a really big unanswered question was how long these superconducting tapes, these pieces of HTS survive under the instantaneous irradiation that's coming out from when a fusion device is on. When a DTfusion plasma is hot, it's making neutrons that fly out, and the neutrons are easy to contain inside of a building with some concrete. They're not hazardous to people and it’s easy to shield against them.
But they do damage materials, and the HTS is very sensitive to the neutrons. But interestingly enough, the HTS had actually not really been tested to failure at our conditions. It had been tested to a certain amount, where we knew it would still work, and maybe the performance was starting to degrade, but it was still working very well. But we didn't really know what the absolute limits of the material were. And the reason for that was that it turns out to be a very difficult, time-consuming, and expensive exercise to irradiate these tapes inside of a fission reactor. You can approximate the neutron radiation by doing experiments in research fission reactors but that's a big, complicated undertaking.
And so, there's a field of work for materials in general, not just HTS, where people use particle accelerator beams to simulate the effect of neutron damage in the materials. And the idea is that a neutron isn't a charged particle, but when it has a collision, and it makes a cascade–if you imagine a neutron zipping through a material, it's not hitting anything, and then it hits an atom, the atom that it hits is charged, and that will create a cascade. If you get the conditions of the material right, you can say, "OK, here's the equivalent proton dose, for example, that I send into a material to approximate the neutron dose." And protons are charged and have different types of collisions with particles, but you can look and make things sort of similar to approximate what neutron damage would be doing with a beam of protons. And getting a beam of protons, although still not an easy feat, is much easier than dealing with a fission reactor.
And luckily, the accelerator that I had prepared in the basement happened to be able to run protons. So I did some homework, and I said, "Hey, nobody's really done an experiment with HTS in the actual operating conditions that it's in," essentially, while it's cold. People had irradiated HTS while it was warm with both neutrons and protons, but it turns out that the kinetics of how a material is damaged actually changes quite a bit with temperature. Things in general get really weird when you get down to almost absolute zero. And same is true for radiation damage. And so, it wasn't really a great analogue to say, "We've irradiated these things warm. That should be how they behave when we irradiate them cold. And again, the reason nobody had done that is because it's a lot harder to irradiate something that's close to absolute zero.
So that was really the meat of my thesis, setting up a test chamber–I'd fixed the accelerator already, so I was able to include that in my body of work for the thesis. I built a test chamber that allowed me to irradiate a piece of HTS with protons and change the temperature either up or down. Because I wanted to also see what the variation was with temperature under a whole bunch of conditions. And so, I irradiated somewhere around 50 different samples. And then, I actually took the samples with me, before we had our own world-class characterization equipment we do now at CFS, to New Zealand, where I mentioned we have collaborators. I flew there with my samples and was able to test these samples in their machine over a couple of months. And that was really what formed the basis of my thesis research.
And the punchline was, turns out that due to the change in kinetics, the radiation damage is actually slower when you irradiate it while it's cold. Which is a really good thing for us because in a fusion device, the only time that the magnets would see neutrons is when there was a plasma, which means the magnets are on and cold. So effectively, that showed that the HTS would last longer in a fusion device.
As I'm hearing it, this sounds more like a story of scrambling and serendipity and not necessarily thinking grand thoughts about a clean fusion energy future.
[laugh] Yeah, there was definitely quite a bit of scrambling and serendipity. It sometimes feels, at the beginning of the project, there was a lot of that. Obviously, the team, myself included, worked extremely hard to get to where we are today, but there are also a lot of things that happened to work out. All the pieces kind of came together at the right moment in time to enable us to start what we've done. But the mission of a clean fusion energy future was always at front and center of my thinking.
Administratively, how does nuclear science and engineering and Plasma Science and Fusion Center at MIT work together and fit for your research as a graduate student?
I was a student in the Department of Nuclear Science and Engineering, working at the lab, which was the Plasma Science and Fusion Center, along with students from many other disciplines. And so, almost all of the professors on my thesis committee were from nuclear science and engineering. My advisor, Dennis, and my committee chair, who actually turned out to be Zach because he was a professor by that time, were in nuclear science and engineering. And they were basically the arbiters of whether I did enough, got enough data, and my methods were sound. And all of the administrative logistics and everything were done through the policies of nuclear science and engineering, but I did my physical work at the Center.
What was Dennis's style like as a graduate advisor as you're going through all this? Is he hands-on, are you in contact with him? Or is it more you're on your own, and you go to him when you have a problem?
I wouldn't characterize it as I was on my own, but I really enjoyed being very autonomous while still having his full support. So it was a little bit on my own by choice. He was always there when I needed him. I would say there were alternating periods where sometimes I would talk to him very often over the period of a week, and we'd been brainstorming and collaborating, then sometimes, I would go off for a month and do my thing. And he was really good about not micromanaging and trusting that I was able to be getting my work done.
To go back to the binary between applications and basic science, for your thesis research, how much of it was sort of figuring out what all of this stuff did, and how much of it was about what you could apply, knowing how these things were working?
I would say it was about half and half. I talked a lot about the experimental portion, where I was building things and doing experiments. And then, there was, of course, the analysis of the data. But in addition of the analysis, there was a modeling component to my thesis, where, as part of looking at how you can simulate best the neutron damage with protons, I did a lot of what's called molecular dynamics modeling, where in a supercomputer, you create a lattice of atoms to simulate your HTS, and you put all of the forces between the atoms that are holding the lattice in place, so you set up this big grid, and then into that grid, you fire a particle and let the system react.
So you put a bunch of boundary conditions on the system, and you put in the physics, the interactions between the different particles with each other when the system is at rest. And then, you perturb it, and you watch how that system evolves. And you need to run it on a supercomputer because, especially with electromagnetic forces, all the particles interact with each other to a certain extent. And so, you're having to account for all these things when you perturb the system. I was actually really lucky, MIT and the Nuclear Science and Engineering Department has a relationship with Idaho National Lab, which had a supercomputing cluster that they happened to be phasing out with a newer, bigger, badder one.
So there was a sort of second-generation cluster I was able to get time on that not a lot of people were using because it was the second-generation one, and I was able to spend several months on that cluster, running my simulations, which was really fun. I'd never really used a super high-performance computer before then. It was pretty cool to be able to say I was running a simulation with 5,000 processors.
To go back to an earlier comment you made about there being a working assumption that ultimately, this thesis research might lead to some affiliation with ITER or working in a national lab, was that sort of your going assumption right through the defense of your research? Or earlier in the process, you saw a different path?
Oh, yeah, so we started thinking about SPARC in 2015. I was still in grad school at that point. We started thinking about SPARC in 2015, and then 2015 to 2016, there was a group of four of us. So Bob, who's our CEO, Dan, who's our CTO, and then Zach, who's a cofounder and professor at MIT. The four of us were in various stages of our academic career. Dan was a research scientist at that point, Bob was a post-doc, and Zach was an assistant professor. The four of us put together a group that we called the SPARC Underground. It was kind of an homage to the Pluto Underground, which was a group of students that came together to put together the proposal to send the probe to Pluto. That was a concept that was originally shot down.
So there was a group of students that, in their spare time, did all the work to show that it was actually very feasible to send a probe to Pluto. And they eventually succeeded in getting the proposal funded. So as a nod to that, we called ourselves the SPARC Underground, and we met on nights and weekends. And we put together the first design of SPARC over nightly beers and a Slack channel to record everything. And we worked for about a year, not telling anybody, even within the Center. PSFC is pretty open to new ideas. But at that time, it was still a pretty radical idea. ARC was out there as a thing, but we were actually saying, "OK, what would it look like if we actually decided to build this thing?" And that was a pretty radical idea, to take the jump from a paper study to a design that we actually want to build.
And so, we put about a year's worth of work into it and then pitched the idea to Dennis and Martin. Martin Greenwald was the Assistant Director of the Center. And they both really liked it. And then, we sort of slowly socialized it with the rest of the Center. And it was a little bit of a shock to some people to actually think about whether we would build this thing or not. But eventually, the whole Center got behind this. And we said, "Well, now we need to get the money to actually build this thing." And at that point, we were still thinking that even though SPARC is effectively a power plant making heat, it was not a power plant making electricity. The main goal of SPARC is to generate net energy from fusion, not necessarily electricity. You want to do it in the smallest, minimum viable product.
Going back to our SpaceX analogy, their first product was not Falcon Heavy or Starship. Their first product was the Merlin engine that they built. And then, they built the Falcon 1 with one of the engines and the Falcon 9 with nine of the engines, the Falcon Heavy with 27 of those engines, and so on. So SPARC is either the Merlin or Falcon 1, where it's really the minimum viable proof of concept. And so, because of that, we looked out in science funding, and there was a lot of philanthropy out there. And we estimated that the cost of SPARC would be somewhere around $500 million, which is nothing to sneeze at, but ITER is projected to cost, depending on who you ask, between $30 and $60 billion by the time it's done. So SPARC was a fraction of that cost for a machine that would be able to fulfill a very, very similar mission.
So we looked out in the landscape, and Bob, at that point, was something called a Translational Fellow at MIT, part of a program looking at the bridge between academia and business, getting funding and stuff like that, and we actually found that there was a lot of philanthropy going into big science projects, especially around climate change. The Schmidt Oceanic Institute is a multi-hundred-million-dollar institute that is sending boats out to do climate change research. And even not related to climate change, there are telescopes. The 30-meter Telescope, despite all the problems they've had, it's been impressive that they were able to raise hundreds of millions of dollars of philanthropy money for a telescope to advance basic science.
So our going assumption going into this was, "Go find your favorite billionaire philanthropist, and convince them to put up the money for this." And luckily, through MIT, we had a good network. We weren't necessarily able to go straight to Bill Gates of Jeff Bezos, but we did actually end up talking to them and several other people like them through a series of connections leading back to MIT. And the response that we got every single time was, "This is a great idea." People were really stoked about fusion, which we thought was great. By that time, climate change was getting more and more to the forefront of people's consciousness. And so, we got in front of these people who were all very wealthy, and they were excited about the project, but they also said, "We're not going to give you money to do this because this is going to be a business opportunity. A telescope is great. We love supporting Big Science. But a telescope is never going to make anybody any money. It's to further knowledge. But if this works, it'll be really great for humanity, and it's also going to make a lot of money because you're going to be selling a lot of electricity. So we're not going to give you money, but we might invest in a company."
So we all took a step back at that point and said, "Wow." We'd played around with the idea of starting a company but hadn't really pursued it seriously until we got this feedback. And then, after getting a lot of feedback and data from going out and talking to people who had money, we said, "Well, I guess we're going to have to make a company." So that happened around the middle of 2016, beginning of 2017. And then, we went through the process of starting to put CFS together, what it would look like, what the strategy would be. And we talked the other day about how we didn't want to just start a company and do what Uber did at Carnegie Mellon, where Uber sort of ripped out the entire computer science department at Carnegie Mellon. Which did not lead to great relations with the school or the academic community. We wanted to have a symbiotic relationship with academia, and we also didn't want to start from square one again. We said, "All of us are coming from this great lab at MIT, and because the lab's funding ran out, it's uncertain what the future of it is."
So in addition to coming up with the idea of the company, we said, "How can we set up an agreement with MIT that will allow us to keep this lab going and use it to catapult the company, so we don't have to spend many years building up infrastructure, and knowledge, and people we need, and we can start from day one out of the gate with 200,000 square feet of lab space that has ten-ton bridge cranes, lots of power, lots of chilled water, and the full might of MIT backing it up? And comes with a team of really talented scientists, engineers, and technicians?" So that was all happening while I was still in grad school. So I was helping out with this, but I was also frantically trying to finish my thesis. It was great because Bob, Dan, Zach, and I were all in the same hallway at the time.
And I would get a knock on my door from one of them (usually Bob) every day asking, "How many pages have you written in your thesis? When are you going to graduate?" So there was definitely a bit of pressure for me to graduate. But Bob and the others didn't want me to just phone it in for my thesis. Everybody wanted me to have a really rigorous defense. And so, they were good about being supportive but also pushing just the right amount to make sure I was going as fast as I could. So I defended my thesis in August of 2017, and at that point, it was just before we incorporated the company but didn't have any money. We got the first money for the company in June of 2018.
In the interim, Dennis hired me on as a post-doc in his lab to continue some of the work I had done with the HTS to make sure it was in a state where it could be handed off to somebody else, but also to focus my work on ARC and do the next iteration on what ARC was going to look like. And my second “job” in my spare time was to continue to set up the mechanisms to allow us to eventually raise money in June of 2018, when we officially raised our first round of money.
How was the process put together to arrive at $500 million? Why not $100 million? Why not a billion? And when CFS was up and running, how close to the mark were you in terms of what this actually would cost?
We looked at it a bunch of different ways. We weren't starting in a vacuum because 170 tokamaks had been built. And like I mentioned the other day, several tokamaks that ended up being the same size as SPARC have been built already. And obviously, the magnet technology is new and different, so we were extrapolating a little bit more with that. But a lot of the other subsystems, we could go and talk to people who had bought parts and kind of put together an estimate for what we thought things were going to cost. And I would say, at the end of the day, we were pretty close. We're still putting together the actual itemized bill of materials for SPARC, but for the bill of materials, I'd say we'll be within 10% of our initial estimate.
And it's interesting, there are a lot of things you can cost by weight almost. If you take the volume of something, you can apply the scale factor on how much something weighs and, based on how advanced it is–like for cars, you can put a multiple factor on the raw material it takes to make a car, and the overall price works out pretty well. And like I said, for some of the magnets and other novel technology, we had to extrapolate a little bit because the magnets themselves, in that particular form, hadn't been built before. But the really strong magnet structures that we needed had been built before. We were actually able to use a lot of the learning from ITER, learn from their lessons.
And that was a great success. The mission of ITER was to get all of this experience so the world could use the experience to continue building tokamaks that were net energy producing. And it's in the charter of ITER to share their data with the world. That's part of their mission. And so, I would consider that a really big success that we were able to use data from them and other tokamaks to say, "This is how we're going to build our device." And because it's so much smaller, it cost a lot less.
As you were ramping up CFS, what were some of the key intellectual property considerations you had to get right exactly at the beginning? Between your funders, MIT, and all of the scientists with the ideas right in the middle, how did you work out all of these considerations?
That was a big piece of the contract that we worked out with MIT, the IP considerations. We were able to put together a contract in place with MIT that said, "We're going to guarantee to fund research for a certain amount of time, and in return, MIT will own all of the IP. It'll be kept within MIT. But we will have pre-paid for an exclusive license to use that IP." And so, from our investors' point of view, that's basically the same thing as us owning the IP because we have an exclusive license to it, and we've paid for the license already. That's been pre-negotiated.
And in addition, in order for us to merge the two teams together and not have worries about silos and people doing work in separate camps, we said that all of the intellectual property we would develop as part of this agreement was also pre-negotiated. And that way, the teams could work as a joint team. We didn't have to care about what the IP ownership was. It would all be under the same umbrella. The investors were cool with that, and we were able to get to a place where the investors liked it, MIT liked it, and we liked it.
Between the key founders, what was the division of labor? Who was responsible for what at the beginning?
Oh, man, that's a hard question. All of us were kind of responsible for everything. At the very beginning, it's pretty hard to pin down a division of labor. Bob and Dennis were the ones who spearheaded going through the networks and trying to get funding. But all of us were in the funding meetings. Somebody would come to us and ask questions, everybody would be in the room, and everybody would answer questions based on their discipline. As far as technology, because my thesis had been in HTS, I ended up becoming the HTS expert on the team. Martin actually has a plasma physics law named after him, and so he's sort of the go-to for plasma questions for investors to authoritatively answer questions about plasma physics. And then, the rest of us answered questions about the magnet engineering and our ideas for that, or our ideas for the tokamak engineering, things like that.
What were some of the expectations, formal or informal, from the investors regarding time scale of when CFS would turn a profit?
I think informally, as soon as possible is kind of what people would like. But with the caveat that as soon as possible, everybody realizes, is probably ten years from now. I should point out, we do intend on selling magnets as well as fusion power plants. There's a division within our company where we said, "We're developing this new technology, and it would be a shame, from a business sense, to not monetize this technology earlier. And it would also be a shame, just from a technology sense, that we're the experts now in building these magnets, and there are a lot of people who would really like reliable high field magnets. And it would be a shame not to get that technology out into the world and wait for tokamaks to happen before getting that tech out."
So the investors agreed and said it would make sense to create a division within the company–the main company is focused on fusion, but because the magnet technology is such a core piece of that, we have people working on magnets for other applications. So putting that aside, we do want to make revenue from magnets. The revenue from fusion, it's always been understood with the investors, will come when ARC is done, which is in the early 2030s. So a big piece of us fundraising was, we didn't want to just take money from anybody who would throw money at us because there were plenty of people who would throw money at us and expect a return in a couple of years.
And we said, "No, that's not the nature of this effort. This is hardware, and hardware is hard. It's not an app that we can throw together in six months and have a beta out for people to start making money on." The investors had to understand that. So it was actually interesting going through the diligence process with the investors because not only were they interviewing us, we were interviewing them to see if they really meant it when they said, "We're in this for the long haul, and we understand that this is a technology that is going to require some development." In the end, of course, the payoff is going to be enormous. But we had to get comfortable with people that we knew would be in it for the long haul.
Obviously, the ideal investor, as you say, is somebody who recognizes this is not going to happen overnight. What about considerations of having just a few mega donors versus a lot of smaller mid-size donors? What were the options there?
So we looked at that, too. Our philosophy was that we wanted to make sure that in this whole effort, it was never controlled by somebody with a purely business interest. Obviously, we need to make money, and all of us are aligned on that. But we wanted to make sure that somebody who really, deeply understood the technology was always at the helm while it was being developed because we didn't want to make decisions that were not sound engineering decisions. And so, even though Bob is the CEO, he also got his PhD in plasma physics from MIT and is still involved in a lot of our technical meetings, he's always at reviews, and is an extremely intelligent guy. And it's really great that he's the CEO and in charge of the leadership of the company because that basically ensures that there's direction that is always technically sound.
So putting that aside, where the founders will have enough to have control, we also wanted to make sure that there was no single investor that could push us in a way that we thought was scientifically unsound or was a short term gain type of thing. And so, we did build a coalition of investors. Mid-size is a relative term because our Series A was $115 million, so some of our mid-size investors would be considered very large investors for a typical company that raises a couple million in Series A. We had about a dozen investors of various scales, from a couple million dollars to $50 million in our first Series A round. And so, we wanted to have a mix so there wasn't one investor who would dominate the board meetings, but also, like we talked about yesterday, we'd have a wealth of experience from the investors and people who've built big projects.
Breakthrough Energy is one of our investors, which is interesting. Going back to talking about Bill Gates and Jeff Bezos, at the same time CFS was germinating around 2015, the Breakthrough Energy fund, Breakthrough Energy Ventures, was also germinating, which was their idea that there would be a venture capital fund that focused on tough tech projects that would be long time horizon because they're hardware, but be very impactful for the world, and specifically climate change. Same thing with The Engine. That was spun out of MIT. That's another tough tech venture capital firm that is really focused on long-term returns as opposed to very short-term software plays.
At The Engine, all of the things people are building are hardware. And The Engine actually also provided space. We still have some lab space at The Engine. They've been a great home. Before we moved into our own offices, we started at The Engine's incubator offices. Our corporate headquarters was there. They've been really great. And we also have these large strategics on board as investors. We now have Temasek, which is a sovereign wealth fund, so that's another really interesting perspective that we get from a country, Singapore, that really wants to modernize and make all of its energy clean and, being an island nation, they have a lot of very specific concerns about how not to be reliant on shipping lanes, and natural gas, and stuff like that.
There's a hedge fund invested in us, they have a really interesting perspective of the world. A whole variety of investors. A lot of the early investors of SpaceX are also invested in us. And we got a lot of experience through that because they were deeply involved in the formation of SpaceX. So they were able to pass on a lot of lessons learned there. So a lot of really diverse investors.
What did you learn about gaming out a return on investment ten years out into the future? In other words, what amount of forecasting would go into figuring out how competitive fusion energy would be in 2033 with regard to where solar will be, where wind will be, where coal-powered plants will be? How much planning went into the idea of how viable and competitive fusion energy will be as part of an overall energy infrastructure?
Quite a lot, actually. That was, and still is, one of my main responsibilities, to run the team that puts together those models and predictions. And so, I'd say on the demand side and the landscape side, we've done a lot of research into what we think the prices of electricity will be from all different sources. So for example, on the public side, there's NREL, which is the National Renewable Energy Laboratory, which has a wealth of information out there. They release a report every year called the Annual Technology Baseline. It's great, it's this big spreadsheet they release with a whole bunch of data that projects out to 2050 under a range of assumptions, what all the different technologies would cost.
Unfortunately, fusion isn't in the NREL ATB yet, hopefully it will be in there soon. But what we did was create our own models and tried to benchmark them against NREL's. We said, "OK, take the capital cost of the plant. What do we estimate the capital cost of our product's going to be?" Used the same financial assumptions or tweaked the financial assumptions to say, "Maybe because it's a first-of-its-kind technology, that will affect our fraction of equity and depth that's used to finance it," and things like that. And so, we put together what we call a techno-economics model that allowed us to tweak both the design of what the fusion power plant would look like and to treat the financial assumptions to say, "If we talk to people out in the world, and you say, 'We think you'd be able to get this interest rate on your loan,'" put that into our model and project out what we think the cost of electricity will be.
Unfortunately, I don't get to do a lot of the actual modeling work anymore. It was really fun in the early days. I have a very talented team of people doing that now who have turned it into a really great tool that not only takes single cases, but looks at probability distributions, Monte Carlo modeling. We have a whole range of different assumptions and a distribution of what we think the probability is of each of these parameters. "Now, do a Monte Carlo model to pick points from all of these probability distributions and see what the curve of the cost of electricity looks like." So we've actually done quite a lot of work into it.
Obviously, it would be great to have subsidies, just like wind, solar, and other clean technologies, but by itself, without subsidies, we think that fusion would be competitive with other electricity-generating technologies. People generally quote cost per watt, and our models right now tell us we are looking in the range of $3 to $4 per watt, which is definitely competitive with the base load power, and it's pretty competitive with a lot of renewables, too.
So the name of the game, obviously, is all renewables together to get us off carbon-based energy as fast as possible. We all know some of the structural limitations of solar, wind, fission energy. What are some of the structural limitations that fusion energy faces when it ramps up to be a significant part of the energy infrastructure?
I would say one of the biggest advantages is that there are very few structural limitations. The limitations around fusion are whether you can build a medium-sized construction project somewhere in the world. Like we talked about before, unfortunately, we're not going to be able to have something so portable that you can ship the entire power plant on a truck. But you can ship all of the pieces on trucks and put them together on a site. And so, from a raw materials standpoint, the great thing about fusion is, at the end of the day, the main fuel that it uses is deuterium, which is heavy water, one part per 6,000 in all water that exists on earth.
And because it's a reaction inside the nucleus, there's so much energy density in this fuel that we have literally a billion years' worth of fuel contained in all of the water on earth that you'd be able to easily get. It's kind of funny, in our techno-economic models, we break out where all the costs come from, and we had to add many extra decimal places to do the fuel cost because the fuel cost is essentially negligible. Obviously, the plant costs money to build because there's steel, and concrete, and the superconducting tape. But as far as fuel inputs, it's basically nothing. So it's actually really great because you can get the fuel from water. It's not resource-constrained. Any nation that has water will be able to make the fuel for this.
And that's fresh water? Or ocean water counts as well?
Ocean water counts as well.
That's good because the last thing we need is to drain potable water for energy infrastructure.
And it's such a small amount, too. If you were to extract the deuterium from a swimming pool's worth of water, you end up with a water bottle worth of heavy water, and that's enough for an entire person's energy use for their entire life. And the rest of the water that would come from a swimming pool, even if it was potable water, you would basically extract a water bottle’s worth from a swimming pool, and the rest of that water doesn't have any deuterium in it at all, and it's still drinkable. So the energy density is really amazing. There are so many wins you get from it.
So if I can dare you to think really big thoughts, why not fusion energy just taking over everything? Why should there even be solar and wind if fusion energy can do these things?
That's a really great question. Like you said before, I think climate change is such a pressing concern right now that we really need to throw everything at the problem. In order to reach the climate goals that we want to hit, I think it doesn't make sense to put all of your eggs in any one basket. Eventually, I do believe that if you look 100 years out, it's very probable almost everything will run on fusion. But I think in the near term, in all of our lifetimes, there really does need to be a mix. And I think it's great that there is a mix, that there are these other supplemental sources. And fusion is really synergistic with a lot of these technologies because fusion can be used as baseload, and you can actually ramp the power up and down on these plants.
And so, you could see a situation where you could have an area that, instead of having to build two fusion plants to start, you could build one fusion plant and supplement with renewables, and the two balance each other. And you have completely clean energy generation. And then, eventually, you replace the renewables with another fusion power plant.
I think another important thing with fusion, which I think is a reason that potentially fusion will be sort of ubiquitous, is that the first step is replacing all of our carbon-emitting generation with clean generation, but the second step with fusion is that it's so power dense, and because it's base load, and you can use the power any time of day that you want, you can actually think about solving other problems with energy that you wouldn't normally have been able to think of solving. You mentioned clean potable water. That's a great example. Because right now, most people don't use energy desalination because it's a very energy-intensive, expensive process, but…
But you can use fusion energy to desalinate water.
Exactly. And so, there are a lot of things like that. You can use the heat as industrial process heat, as opposed to burning something that emits carbon to generate industrial process heat. You could use electricity to do that, which currently isn't used because it's a little less efficient. But if you were to build a fusion plant right next to the industrial process, you could use the raw heat coming out of the fusion plant, or you could use the electricity because it would be cheap enough that you could use it to make heat again on the other side. So there are a lot of really cool things you can think about doing with fusion. Even after you've achieved the climate change goals, you can actually use fusion to make the world a better place in lots of other ways, which is what really excites me.
I'll ask two last questions to round out this fantastic conversation. A theme of your journey has been that you've been a beneficiary of having excellent mentors in your life, scientific mentors, from your uncle to your graduate advisor. Obviously, you're not a professor, but what opportunities do you have, as somebody who's becoming senior in the field and a leader in the field, both within your company and in terms of outreach to people who are in their teens, 20s, who realize that climate change is the overriding threat, and they need to be involved and need to understand that there are solutions in their future? What opportunities do you have in mentorship in that regard?
That's a great question. Within the company, obviously, I have a team of people, and we've been hiring a lot of people straight out of school. So it's great to have a cohort here that I can help to mentor. And we have a lot of interns as well. And externally, I think it's really cool that MIT and the PSFC have a heritage of outreach. Even in grad school, becoming involved in outreach was encouraged for grad students–and I loved giving tours of the fusion device. We would have high school students come in all the time, and we'd give tours of this device. Unfortunately, in the last year with Covid, it hasn't really been possible, so we've had to get creative to do outreach.
Outreach is always so much better when he can come and see the hardware. There's no substitute for getting in a room with a whole bunch of cool shiny things. But we've been able to do several outreach talks to various audiences. Every January, MIT has this thing called independent activities period (IAP), and we always have seminars that are free and open to the public, where we talk about the science, the technology, where we think it's going. And yeah, I've had the opportunity to give several talks to various audiences to kind of spread the word. That's one of my favorite parts about my job.
Last question. Chronologically, if we want to establish bookends between your hip injury and when you started thinking about fusion energy all the way toward when CFS turns a profit, when it starts to actually do the things that you want it to do, where are we? How close are we to fruition of the dream?
I think late 2020s, early 2030s is when we're going to start turning a profit. Even before that, our plan is to have SPARC online, producing fusion heat, more energy out than in, by 2025, 2026.
Who's that first customer going to be? If you had to guess.
That's a really tough question. We're still trying to figure that out right now. One option is that we could sell electricity either in the market or to an off-taker in the United States. But I think it's still such an open question. There are a lot of great places. One of our investors, Temasek, is invested in us because they want fusion energy in Singapore. And so, I could also see us having the first fusion plant in Singapore. Our Italian investors from Eni, which is an oil company, may want to build a plant in Italy. And so, it's really pretty wide open, which is pretty exciting. Obviously, the first plant is going to have to be built somewhere, but there are a lot of really good options.
I want to thank you for spending this time with me. It's been so interesting and exciting to hear your journey. And obviously, on behalf of everybody, I wish you success because this really needs to happen. Let's make a deal, let's check back in in five years and see where things are at that point.
It's a deal. I would love to.
Terrific. Thank you so much, Brandon.