When you turn it off, you don't actually turn it off. This is one of those things that's kind of wild, but when you shut down a nuclear reactor, you're actually still producing heat. But you have to again, you have to be able to have a system that can cool and remove that heat, and that's typically been the biggest challenge in the headline events that you talked about. You can accommodate and achieve that using natural phenomena as well.
Welcome to the Emergent Podcast, where we interview product and startup leaders from emerging technology companies about their work in the trenches to bring the hardest and most impactful technologies to life. Today's guest is Jacob Dewitte. Jacob is the co-founder and CEO of Oklo, a Santa Clara based company developing advanced nuclear reactors that went public through Sam Altman's SPAC earlier this year. Jacob has experience working on a variety of advanced reactor designs, including sodium fast reactors, molten salt reactors, and next generation pressurized water reactors, as well as the nuclear fuel cycle and analysis. Jacob also worked with experimental irradiation testing facilities, and he's originally from Albuquerque, New Mexico. He earned his BS in nuclear engineering from the University of Florida, and his Masters of Science and PhD in nuclear engineering, both at MIT. In this episode, we discussed the importance of nuclear energy in our planet's fight to produce enough energy for mankind for the next billion years. We dive deep into how Oklo's Aurora reactors work, what went wrong in historical meltdowns, and how Oklo has engineered their nuclear reactors to avoid these pitfalls. We discuss how to retain investor confidence while building a startup that takes a decade plus to commercialize, recycling of radioactive nuclear waste, which spoiler can power the planet long into the future, AI data centers and their new demands on the grid, and what it looks like to press the industry forward in one of the toughest regulatory environments in existence. We talk about Oklo's decision to go public via SPAC, competition in the nuclear industry, and where the future is taking this exciting technology. Now a word from our sponsor. This episode is brought to you by FlyKit. FlyKit is the first full solution to jet lag. This might seem hard to believe, but I've flown from Los Angeles to Tokyo and back with virtually no jet lag using FlyKit. Based on research with Navy Seals and fighter pilots, the FlyKit team discovered that long flights cause inflammation in addition to disrupting your circadian rhythm, and they figured out how to solve both these challenges using a customized program for when to eat, sleep, and use their patented supplements and blue light blocking glasses. It's simple to follow and you don't have to change anything before your travel day. With FlyKit, more than 90% of people can sleep well in Europe, Asia, or anywhere else in the world on night one. Visit flykit.com and use coupon code FLYWITHEMERGENT to get 15% off FlyKit. I am pumped to be discussing nuclear energy with you. I'm so excited about the fact that nuclear energy is producing millions of times more output per atom than combustion, and I think as you're aware, a long time ago in my career I did some of this work as an intern at Lawrence Livermore but on the weapons side, and then worked in nonproliferation. So I'm really stoked to have you today, Jacob. So welcome to the podcast.
Thanks for having me, excited to be here. It's kind of fun too because my background was starting actually in the weapon side as my first kind of gig, I guess, was an internship at Sandia National Lab. So eager to jump in.
I love it. We'll have to at some point have a classified version of the podcast where we can talk about nuclear weapons technologies. That would be an interesting way.
Yeah, right. I don't know how that would work.
Well, let's set the stage at a high level for people who are unfamiliar with the relative benefits and risks of nuclear technology. Can you explain why Oklo's mission is so important to the planet?
Yeah, I mean this gets to what you said in the intro, right, which is energy density and the fact that you can unlock all that energy without having emissions that come out from it that are climate contributing. So you don't have greenhouse gas emissions. It gives you two huge benefits that kind of come from that. Obviously there's the sustainability piece from an emissions perspective and from a condo perspective, but there's also this phenomenal benefit of just a general resource scalability and sustainability perspective. One thing that sort of follows on the incredible energy density of fission, and just to put some fun numbers on it, when you split an atom, you're typically releasing about 50 million times more energy than a typical combustion oxidation reaction. That's insane, right? And that means you need a lot less material obviously to get energy from it. But to extrapolate that, one of the interesting things that is also nice about fission is it makes heat, and a fairly, I'll call it this and this is a little bit reductionist, but a fairly easy to use form of heat. Sounds funny, I know, but what that means is at the end of the day, what a nuclear reactor generally is is a water boiler in some form. And at the end of the day, you're making heat that's being transferred through media to go boil water. And the heat from fission in fuel is ultimately, sorry, heat energy from fission is mostly manifest as heat in the fuel that is then typically conducted out of the fuel and then convectively carried away by the coolant. All of that translates into a system that can be engineered to harness all of that incredible energy density in a very materially efficient way. And what that means is not only is the fuel very energy dense, but actually the amount of materials you need to make each megawatt hour of energy from a fission power plant is actually the fewest of all energy sources. That's kind of an awkward way of saying nuclear fission requires a much smaller bill of materials to make each megawatt hour of energy it produces compared to all other energy sources out there, which is a pretty fantastic advantage to have because that gives you massive scalability benefits and also sustainability benefits, and also should obviously translate to significant economic benefits. So all around, it's a pretty promising approach to give you really scalable, abundant energy and do so in a sustainable way. And then you kind of put all that together, then the other number I like to add in is from a true scalability perspective, because the energy density is so high and the materials needed, the heavy metals you use as fuel, these systems are actually relatively abundant. There's enough energy content in those materials that with advanced technologies like fast reactors like what we're doing and advanced recycling techniques that allow you to keep reusing the unused parts of fuel, you can actually power the entire planet's energy needs for billions of years, effectively the durable lifetime of the planet or the inhabitable lifetime of the planet. So this is a terminal energy climate solution that we know works. True definition of sustainable, right? That's what sustainable is actually supposed to mean in this context.
Interesting. Is there a theoretical maximum sort of velocity of power production given that you're ultimately still just boiling water, but you're just boiling it with a much better source because the water boils at 212 and that's it, right? So how does that work?
Yeah, so the way this all translates is the laws of thermodynamics dictate how much energy we can actually convert from one form to the other, dealing with all the various loss factors that have to account for what thermodynamics dictates to us. In other words, the heat you make in a reactor, the actual energy made in the reactor that takes the form of heat, so the joules if you will, the megawatt hours in thermal terms, you can only convert so much of that into electric energy, and that's going to be dictated by just the actual thermodynamic limits and efficiencies of the turbines or the technologies used to convert that electricity. There are promising alternatives to that that might be more efficient at these temperatures in converting the energy from heat energy into electric energy, but steam systems are just the best. They're the most mature technology, they just work at a better advantage state now. But there might be things that come in the future that have some more efficiency around them. That said, the technology is just very immature. But to get to that point, how that actually works is you build up a system that you're running the steam at higher operating temperature, so you're running at a higher temperature, higher pressure, and that means it's a pressurized system on the power conversion side, just like you see in gas boilers and solar thermal boilers and coal boilers. It's very similar. Each one is obviously tailored to the different considerations there, but in a nuclear system, you're going to be operating in a range that can be on the lower end to the higher end depending on the technology type, and then you're just heating up that water with the energy that's transferred to it, probably by another coolant, sometimes directly from the fuel, to then boil it, make the steam, to then spin a turbine. So I'd say that the nice thing about this is this is a deeply mature technology that can scale rapidly. We're definitely not limited or bottlenecked by turbine availability, if that makes sense. And in terms of physical maneuvering of power production, there are thermal time constants in a system across the board. Those are actually features though in general, because having really fast responses to the grid is important, but spinning turbines provide a lot of shock absorption to the grid system. So to kind of pull the thread you said about velocity, that's actually a huge benefit about having spinning turbo machinery in terms of being a shock absorber to the grid with all the voltage dynamics that happen. The grid is an incredibly complex system. It's wild to think about how well it operates considering all the challenges and constraints it has. But these are important assets in terms of that, not just the power they produce but also the reactive power provided by having all this spinning inertia in a big turbine.
Interesting, interesting. So you know, a lot of founders with really big ideas like this, they're paralyzed to act because they're not sure how they're going to build something that in the case of Oklo requires a decade plus and a massive amount of capital before it can be put into practice. How have you navigated showing intermediate progress, earning revenue, all these things before your Aurora reactor is fully available on the open market?
Yeah, it's a great question. I think this is one of those things where if you have a big view, big vision, big idea, you kind of have to think about what it is you want to get to and then work backwards from there to some degree about the likely path to get there, and then iterate to go back forward and say, okay, if I come to this conclusion of the idea, I'm executing on this, how do I make sure I'm getting closer and closer towards the goal where I want to get to? And be flexible. I think that's one of the hardest things, because it's always going to be radically different. What you envision at the earliest days before you start something to when you start something, the actual roadmap and the path is pretty much always going to be different than what you thought. I think for me, for this side of things, the first principles are just so heavily on your side. The physics are just so heavily on your side. Better energy source, more scalable, known mature technology. And in a world where despite some of the headwinds, I remember back before we started the company in 2013 or around that time, we go to the MIT Energy Conference and you hear one panel talking about how energy demand is flat and going to be flat or maybe declining, but then you hear the very next panel, or even a panel contemporaneously, maybe that was the problem, it was contemporaneous, was, oh my gosh, we're going to need so much more energy if we're going to drive electrification and all these things from transit to home heating, cooking, all this stuff. And you're like, yeah, okay, I'm going to bet on that latter, because rarely, I don't think humans have ever gone using less energy. So if you have a better source of doing it with a technology that works despite some perhaps optical things, I think those are actually ideal considerations around starting a company when it's somewhat contrarian in that sense. Some things are definitely harder, but I think in some of the ways, some of the macros are easier. But that's because you have the inertia behind you, the physics, the first principles on your side. So I think that was what we set out for. We definitely had a long-term vision and had sweeping iterations about what we thought was going to be the case, and then just dove in with the most meaningful incremental step possible. And I think that's the thing that's hard: if you chase everything, it's just too hard. If you go after kind of a niche angle of how you get a toe in the water, so to speak, that's a great way to start, but be flexible, because that might not be the best answer and you kind of have to iterate from there. But if you at least do that, that's how you take that first step, because sometimes taking the first step can be the hardest if you feel like that first step is super far. But if you kind of just stumble into it because you're leaning into it pretty hard, that makes that first step a lot easier.
Yeah, yeah. But help me decompose what was the first step for you guys?
Yeah, I mean what we thought, sort of the thesis for what we're doing at Oklo, largely came from watching and being part of an industry. First of all, my background, I grew up in New Mexico, fell in love with the technology at a young age, obviously got the chance to get started on it from a kind of unique lens, actually unique angle from the weapon side to start, but that gave me a good platform to springboard into the science, the physics, the engineering systems of it with a focus on the power generation side across a lot of different facets of the industry. So I spent time in the national labs and some of the industry industrial players on the fuel cycle to the reactor designers and academia. The thing I got into though is thinking, okay, new technologies are going to be what drive forward, because to me it was always like, man, nuclear is just inevitable, it has too many reasons to not be. And I thought what ushered in the new waves, because we have all these plants built, it's amazing, but the new waves of growth that I thought were inevitable were going to be driven by new technologies. What I quickly learned though was that actually that's maybe part of the equation. The real challenges though I thought were around the stagnation of the industry at large, and there's a lot of drivers that made that happen. I think there's a couple things that you can point to. But we saw, hey, we haven't innovated on the business model side in nuclear, maybe ever, or at least in decades. We were still stuck in the same model of designing a power plant to partial completion. This is what a company like ours would boilerplate look like: you design a reactor and a power plant somewhere between 50 and 80% completion, take those incomplete blueprints, go ask people to pay you a couple hundred million to then take it over, do the rest of the design, work site prep, build it themselves. That's massively burdensome and puts a lot of friction on the customer, so it makes buying new nuclear systems really hard. So our view is, why don't we make it easier there? And then the other part was couple that with a couple other factors. We found that there's a lot of interest if we actually made it easier to buy the power by just designing, learning, printing the systems and selling the power through power purchase agreements or similar vehicles. Then the other part was to then start at a smaller size so that we could actually be economically more attractive from a total capital needs side. If you start a company needing billions of dollars to go from start to first product, that's a really hard thing to solve for. If you can incrementalize...
...up at 15 megawatts as an initial entry point, and that allows us to reduce the total capital requirements. But there's also benefits from an organizational perspective. If you have to build one massively hard thing, hard tech companies tend to do is they weigh over hire. I think some of that's afforded by sometimes too much easy and abundant capital early on, a lack of cultural perspective that keeps you focused on cost discipline, or a view that you have to have staffing to meet certain goals to move quickly. When I would argue that's actually you can really invert a lot of those paradigms, but it does mean you have to be very deliberate and intentional. But then that positions you to have much more capital efficiency. And being smaller means it's easier to take a smaller bite. It allows you to more efficiently tackle that opportunity with a team that can understand the whole system across multiple individuals, each understanding the whole system. That's worth a lot from systems engineering and systems organization and integration perspectives, from team building. If you have a system so complex you need a lot of people to just be able to have expertise, multiple people in a room to have expertise on the whole system, that's a hard thing to do. You can definitely do it, but that's a really hard thing to build a company from scratch to be able to do.
But like, what did you... this all makes sense, but you're in year seven, you're in year eight, you're in year nine, and you're briefing the investors on the progress that you've made, but you don't yet have a full Aurora plant pushing power out today. What did you show? Because you've talked about staying on the critical path, but what did you end up showing? What are, to the extent that you're willing to discuss any of those milestones, what are one or two of those interesting milestones on that complex journey?
I think this gets back to being intentional about how you tackle risk. I think we weren't taking on technical risk, and that's kind of the next pillar here was business model, size, and then the technology choice. One thing nice about the nuclear side is we don't have to reinvent the wheel. We could, but we don't have to. Maybe I should say it differently: we don't have to invent a new wheel, we already have wheels that work. We're building up a mature technology that has a long history of operational excellence behind it. A lot of the learnings came before, thanks to government research and development investments spanning decades before. That's a fantastic platform to be able to build off of. As a result, that takes a lot of the elements around certain R&D elements off the table and allows you to then focus on different milestones. Now interestingly, if you want to hack VC, if that's your goal, and I don't think that's aligned to actually building a business in the best way, then you might want to do something that's very new and different so that you can point to all this kind of hardware testing and development. But you're going to be driving that in a manner that ultimately is deferring product development. And arguably, especially in a space like nuclear, I would argue that a lot of the really good ideas have already been tested and explored to a pretty high degree. There's been some technological progress in certain areas that maybe weren't looking, brushing off some of those ideas and looking at them again, but I think most of that work's already been done. So starting fresh and new and trying to do something really novel, to then build a little test stand, non-nuclear, but a little test stand and show some phenomena, and then build a little bigger one and just march on that, that's definitely an easy way to tell investors that you're making progress. What might be lost in that though is you're taking this really scenic long path as opposed to taking a direct path to market and being efficient on that.
So for us, what that looked like was okay, we built this framework and then it's just a matter of execution. We knew what we have to do: you have to build a customer order book. That was a key thing we wanted to show as validation of the business model and also market discovery and exploration, make sure you find some people really love what you're doing. Then on top of that, use that to better understand the unit economics, which is something that too often gets lost in early startups, but really understand that this does work, this does refine out. And then make the right progress in terms of supply chain development and regulatory development in the right phases. Since there's so much regulatory overhang in the space, there's a lot of opportunity to drive into that very early. And then you're able to lay out a roadmap and a plan, communicate that, and then execute against it. So you have all these milestones, you set the goals, set the stage with that, you can then execute against and then iterate upon as you find that maybe some of those weren't the best or the right ones, some are more important in different orders perhaps. So I think that's what we tried to set the stage to do: tackle these things in the right staging, the right order, but make sure people understand that this is a long game. We see a path to be very aggressive on the time and perhaps do some things in crazy fast ways, but also make sure people understand there's risk that might compound in these different areas. So if we thought we could produce a plant in year X, it might slip by one or two years based on the progress through these points. It doesn't change the overall value of what we're trying to do. By focusing on many, and that was another big thing here that I think we tried to focus on, this isn't about building one plant. If you make it all about that, you're actually in a bad spot to succeed because everything gets fixated on what happens with that one plant. Our view is more like, hey, this has been built before, we're just modernizing it, and we're trying to pump as many out the door as quickly as we can. So it's really about the volume and the time it takes to get to volume, and walk our way into that. That's what we really wanted to focus on. So I think the first steps were really leaning into talking to customers and laying out a path to actually take this thing to market that included finance and regulatory milestones, and then start executing on those and then dynamically iterate.
Yeah, so you pointed to a lot of historical trust points, your own experience, and you said this is the path, these are the things that are actually not risky. Let us focus on these parts that are very difficult but not technology risky from the standpoint of hey, no one has ever done a thing like this before, but maybe bringing the pieces together like this is the thing that no one's ever done before. That's kind of the insight there for how to stay on strategy without making these detours.
Yeah, I mean I think it's like, hey, if we can build more... I mean what we saw in this space right is we have a technology set we know works. We obviously have to optimize the design for our use case with a bunch of known things, and we got to simulate, model, and test those things, and we don't have to do a lot of the testing, get even done. And then you bring that forward and make sure that then that's tied with the actual path to market, because that's the key thing: you have to integrate all those pieces and just constantly sweep through those share updates. I think the hardest thing about a journey like this is a lot of things become qualitative, they're not exactly quantitative. And that's where there are some interesting challenges there. But that's where qualitatively speaking, optics and this can be the detour that presents itself. So you could go build a test stand that proves out something that doesn't actually prove out anything. It's like, oh look, I can heat up and the heat gets removed. To be a little reductionist, it's like, okay, what's the point? What did that really show you? In some cases, you can definitely do things that are meaningful on that, other times it's honestly mostly just a dog and pony show. And that could be a big distraction. It can also be useful, and I totally respect that. Let's be real about what it takes to actually build a business in the space: you do have to be mindful about optical considerations and playing to the market. But I do think sometimes there's a tendency to overhack the venture capital perspective and frankly n accordingly, and that just ends up becoming a distraction at some point.
Yeah, I know. It's great to hear a founder owning the narrative even in the face of the VCs. Okay, so let's get safety out of the way. I kind of want to nerd out here a little bit on meltdowns. So the general public obviously has not been able to completely shake the nuclear accidents at Three Mile Island, awareness of Chernobyl and Fukushima and other countries. And Oklo is trying to dramatically increase the number of reactors in our backyard, so to speak. And I saw on your site there were some interesting experiments that had been run in the 60s where the reactor sort of cooled itself down naturally. I was hoping that you could kind of unpack how this technology works for us, what you guys are leveraging, what happens in a normal meltdown, why do those circumstances not even exist in an Aurora reactor, and how have you guys hacked this challenge?
No, I think that's a great question, and I think this is one of the key things that's also pretty underappreciated about what you can do on the nuclear technology side. When I talk about physics being on your side, it's not just because it's materially dense and all these other things, it's actually because these are systems that can be tightly integrated from a multiphysics perspective and actually be inherently safe systems. They're self-controlling, self-cooling systems. And you can do that with a number of types of reactors, because really at the end of the day, when you talk about reactor safety, you're talking about managing and containing radionuclides that are produced that could present a potential threat if they got out or were released in a way and dispersed. The ways in which you typically are going to best do that is make sure you maintain integrity of the things that are containing those things, whether it be the cladding, whether it be the fuel itself, a number of different barriers you typically engineer and number those in so you have what's known as defense in depth in place. And then what you can then do to maintain that is want to make sure the system is controlled, and it's great if you can be self-controlling. So if something happens, you produce too much power, obviously you can overheat, that can be challenging, so you want to have a system that will shut itself down naturally. It's a really nice thing to be able to do. And then the other thing with nuclear is when you turn it off, you don't actually turn it off. This is one of those things that's kind of wild: when you shut down a nuclear reactor, you're actually still producing heat. I like the term afterheat, the industry often calls it decay heat, but basically the byproducts of fission itself are decaying away radioactively and they're releasing a lot of energy, much of which gets manifest as heat, so it has to be removed. Now that heat makes up a small percentage of the total amount of power produced at a reactor at full power, but it needs to be managed. It does decay, which is great, but you have to be able to have a system that can cool and remove that heat, and that's typically been the biggest challenge in the headline events that you talked about. But you can accommodate and achieve that using natural phenomena as well. So to dig into that, the specific technology we're working on, a liquid sodium fast reactor, has shown that you can actually do this. Most notably at the EBR-II in Idaho that you talked about, in April 1986, they put this reactor through these incredible challenging situations that would be catastrophic for most of today's reactors by and large, and it survived them without so much as a scratch. And the reason was because they had a system that was able to control itself naturally. As the reactor heated up, the fuel thermally expanded, and that caused more neutrons to leak out of the fuel, get absorbed in the stuff around the fuel, and have the reactor shut itself down. And on top of that, it also had an interesting ability to just be able to cool itself naturally. In other words, the heat was naturally removed from the fuel by the natural circulation of the coolant, which then ultimately rejected its heat to air naturally. That was also naturally flowing. What I mean by that is the coolant that flows between the fuel pins, that's convectively removing heat, as it heats up it thermally expands, that causes it to then rise, cool, cooler coolant behind it, so you're convectively able to remove heat that way, and you're able to flow on natural circulation because that hotter coolant rises. And in so doing, it's able to transfer its heat through the vessel basically into just an air cooled system that's also operating on natural convective phenomena where air heats up along the surface, expands, gets less dense, and then buoyancy causes it to rise and pull cool air behind it. You have this kind of open infinite loop almost. That means you have a system that's able to be self-stabilizing and self-cooling. And in 1986, what they showed was they had the reactor at full power, they turned off the pumps, which meant you have a significant mismatch between heat generation and heat removal, which would have caused the reactor to significantly overheat. It did heat up a bit, but shut itself down naturally, and then all those natural phenomena kicked in and cooled itself off. They went to lunch, they came back, turned the reactor back on, and then locked the control rods out again. That's the things that you would normally put into the reactor to shut it down. So you're basically handicapping yourself pretty considerably. And then they turned off power to the plant, basically what happened at Fukushima but way, way worse because they had shut down for an hour before they lost power, this had full power and they lost power, and the reactor went through the same phenomenology, shut itself down naturally and cooled itself off without an issue. That's phenomenal.
That gives you incredible safety characteristics, but also translates economically in beneficial ways because it allows you to simplify the design and reduce cost accordingly. That's a pretty attractive thing to be able to do. And you could do that with a number of types of reactors, so I think that's a pretty important thing. That's kind of what people call in the industry passive and inherent safety. It's a pretty big focus for new reactor designs because of those benefits. But I also think there's something that's broadly becoming more aware for folks. Getting back to what you said at the beginning, you're also benefiting just from public awareness. It turns out, and this is something that sometimes gets provocative, but Three Mile Island really wasn't that bad of a situation. All those safety systems that were there, well most of them actually worked pretty well and contained things. The public didn't receive any unnecessarily large or dangerous doses of radiation. Obviously the hype, the media headlines, the narrative, the active disinformation and misinformation campaigns by anti-nuclear special interest groups that often had backing and support from competitive alternatives, those found footing back in that time. It wasn't to their benefit when the internet came along. So we live in the information age, people can actually access information and see through the misinformation and disinformation campaigns that anti-nuclear folks have sewn across the public's mindset about the dangers of nuclear. What we've seen many, many times over and over again is as people begin to realize, wow, actually this is amazing technology, and even without these amazing advanced passive safety features that allow you to achieve phenomenal safety characteristics and maybe achieve them at lower cost which is great for deployment and scalability, our current plants are phenomenally safe, one of the safest sources of energy generation there is, just full stop. And the data supports it. So I think that alone is actually causing a significant amount of reset. Just people being able to go online and search this stuff and find out about it, and then people become pretty enthusiastic. So I think you've seen a massive sea change there just because of the availability of information that corrects the false narratives that have existed about nuclear safety in the past.
This episode is brought to you by Chaser. Chaser is a task management solution completely embedded in Slack. When teammates have requests for each other, it's no longer necessary to involve a project manager, add the issues to a task tracker, and remember to follow up. Right from within any Slack channel or message, with a few simple keystrokes, you can identify a message as a task for another person or yourself, and Chaser automatically tracks it and reminds the assignee of the work until it's done. We started using Chaser at First Principles to create stronger clarity on those important but less formal tasks that must happen to get from start to finish on a project, but that often don't arise until the middle of a discussion. We now spend less time context switching and we're not dropping the ball on each other. Plus, there's great transparency into who does what in order to bring the team to success. Get a 14-day free trial at www.trychaser.com and use the code FP50 to get 50% off your first 3 months.
Yeah, I think there's also, forgive me for this sort of ludicrous comparison, but I think that for a long time people didn't have a conception of what benefit they would directly derive from nuclear power other than maybe I pay a few extra or a few fewer cents on my energy bill if I have nuclear power in my backyard. Now people can see ChatGPT, they can see Claude, they can see Gemini, they can see these models impacting their lives, and so there's almost this pull that's like, well we know we're going to need more energy for that, so we're willing to maybe reconsider some of the technologies that from a public opinion perspective... you know, it's not the public making the decisions in a direct fashion, but I think there's something to that.
Let me put a little bit deeper on this. So I came across this term, I think this is what you were explaining. Tell me if this is wrong: were you talking about the negative temperature coefficient of reactivity? Is that the concept that you were describing? And is the idea basically that as the reaction gets too hot, the efficiency at which neutrons bombard each other to continue the chain reaction basically becomes lower, and so the reaction is cooling itself down? Am I getting that right or not quite?
Not quite. What it really is, is as the reactor heats up, you get the fuel pushing itself further apart from itself. In other words, fuel atoms become more distant from each other, which means the likelihood of neutrons actually interacting with fuel atoms becomes lower, which then means you're just having less fission events, which means you're producing less neutrons and less power, and as a result you get this cascading benefit of shutting the reactor down that way.
Yeah, yeah, okay. If that's the case, then what keeps the reaction going when you're not introducing a loss of power, not stopping the control rods, not removing coolant or somehow impeding coolant? What keeps the reaction moving?
Yeah, so what that really looks like is you design it to be operationally stable at the steady state conditions you expect it to operate at, so that you're able to be operating, producing power. The reactor is considered critical, and what that term really refers to is it's a steady state of producing basically as many neutrons from one generation to the next, so you're not getting growth or decline in neutron population, and that translates to steady state of power production because neutron population is dependent on power production, because power is manifest by fissions and fission is what releases the neutrons. So we split an atom, not only do you release that energy but you release two or three neutrons pretty much every time. So the number of neutrons produced is dependent on the number of fission events happening, which directly translates to the amount of power being produced. And so as the system is operating at a steady state, you're at this level of producing x amount of power. This is one of the really cool things about what you can do with this, because it's load following. Meaning if you change how much power you're pulling out of the system, like electric power, you can actually then just have the reactor naturally stabilize. Because if you're removing less than what's being produced, or you reduce the amount of heat you're pulling off the system through boiling water and spinning a turbine and converting that to electricity, the reactor will heat up because you're pulling less power, and that will cause the fuel to expand and then re-equilibrate at a lower power. And the inverse is true too: if you're operating in a steady state and you end up pulling more power out of the reactor, you actually end up causing the fuel in the reactor to cool down a little bit, and then as a result produce more power as it re-equilibrates into that new regime. So it's a very elegant system that gives you this natural feedback phenomenon. One of the other extrapolations is then, okay, well if you have an event that happens where the reactor goes through one of these events you talked about and then it shuts itself down naturally because something happened and overheats, eventually that's going to cool down enough because the decay heat decays away, and so the system's actually going to cool back down to the point where the reactor could restart again, just all be at a very low level of power because the power will ultimately be dependent on what you can remove through the natural phenomena. And that's true. And then what would happen there if nobody intervened or shut the reactor down and addressed whatever situation it was through the control rods, it would just naturally sort of turn on and turn off and turn on and turn off until it ultimately depleted its fuel and just coast down. So just a long coast down effectively is what would happen. So anyway, that's kind of an extrapolation, but does that answer the question?
Yeah, yeah, absolutely. And so have other companies been building reactors that follow this design pattern? I know we haven't had a lot of new nuclear reactors other than the one in Georgia. Are they doing this?
I mean the degree of what has been demonstrated and what was demonstrated with this EBR-II reactor, and also it had a cousin plant in Washington state called FFTF that demonstrated similar phenomena, that's pretty unrivaled. But you can achieve similar capabilities with water cooled systems or even with gas cooled systems. Slight different physics of what drives things, like it's not just the fuel being spaced further apart, it's also changing the ratio of fuel to other materials that slow neutrons down. We don't have that in a fast reactor, it's a very tightly coupled system. I know, kind of in the technical weeds, but this trend line of having these, putting physics on your side, things like gravity, thermal conduction, thermal convection, those have been employed even in those reactors we built in Georgia, and they're being employed with some of these designs like we're bringing forward and some others are building out. This is definitely the new era. I sometimes like to think of nuclear technology having gone through three macro design evolutions that spanned decades, or design epics. Maybe one epoch was the initial one, just making it work, that was kind of the 40s into the 50s. Then I think from the 50s until really probably the 70s, 80s, it was about, to me and this is just my view, but it was about maximizing power density out of the fuel. It's one of the amazing things about nuclear: you can generate a ton of power in a small volume. And so that became pretty exciting for people to do. But the problem with that was the implications it had on the rest of the system, because when you maximize power density, it becomes a lot harder to achieve all those amazing inherent and passive safety features, and you end up having to bolt on all these systems, backup systems, auxiliary systems to cool, to keep the system cooled, and all these other... almost too greedy. Yeah, it becomes complex at the system level, increasingly complex. And I think where we started in the 80s was recognizing actually it's really great if we sort of change the equation of maximizing fuel power density at the fuel level and actually maximize power density at the plant level. That affords you these naturally simplified features.
Or design characteristics so that you're able to just know how much power you can get out of a system that is able to cool itself naturally by airflow around the vessel or whatever sort of design you're pursuing. And then that induces, you know, that lines itself also typically towards smaller reactor designs, which then gives you benefits though because now you also get rid of all of these backup and auxiliary systems because you don't need them. And that comes with great benefits, right? And it's that piece that I think sometimes people forget about because it kind of gets lost in the academic origination of this. And that motivation kind of was manifest more in the 80s. Now we're seeing the fruits of that labor though, and I think it's yeah, so we're not the only ones designing a system that can do stuff like this. In fact, pretty much everything is being explored has quite similar flavors to this. We just love that we're building on a technology that actually demonstrated it with a real machine, with a real reactor, and that's pretty unique. It reminds me a little bit of the, the founder was on a podcast recently talking about how the biggest limiting factor for them is actually trust. They could roll out in more cities, but they want to build trust. And I think in an interesting way, by having a generation of reactors where we're not running the reactors at 11 all the time, instead at eight or nine, there's a chance to actually build that trust with a high volume of more modular reactors, show people this actually works, show them that it solves their needs.
Okay, so let me ask you this question. So because you guys are running fast fission reactors, you are in a wider range of fuel isotopes in the first place. Does this also mean that you're producing more radioactive byproducts at the end of your reactions? And does this also mean that it's more complicated to build your fuel rods in a safe fashion in the first place? Can you talk a little bit about that side of things?
