Jeff Waksman is the Principal Deputy Assistant Secretary of the Army for Installations, Energy, and Environment. During this 38-minute discussion, Waksman explained why it’s so difficult to get reactors built, the need for intensive focus on materials science, and why the military will be the first to get a SMR built. Follow Waksman on X. Full transcript is below. (Recorded on February 27, 2026.)
Transcript
Robert Bryce: Hi everyone. Welcome back. I’m delighted today to have a short conversation with Dr. Jeff Waksman. He is the Principal Deputy Assistant Secretary of the Army for Installations, Energy and Environment. And I ran into some people at the Texas Nuclear Summit last October who said, well, if you’re interested in what’s going on with nuclear in the federal government, you have to talk to Jeff Waksman. So here we are. Dr. Waksman, thanks for making time to talk to me today.
Jeff Waksman: Oh, it’s great to be here. Real pleasure to talk to you.
Robert Bryce: Okay, so good. You have a long title. Tell me, Deputy Principal Assistant. No, Principal Deputy Assistant... you’re in charge of energy for all Army facilities, is that the punchline here? Am I reading this correctly?
Jeff Waksman: Well, the main thing of my title is it means when I go to conferences, they have to use smaller font to fit my title on the badge. But really, the way to visualize it, because even most soldiers don’t know the structure of the service, the Army has a secretary. Then there’s an undersecretary who’s the deputy. And then there are five assistant secretaries. We break it up into five assistant secretariats, and within that I’m the principal deputy, the number two for one of the assistant secretariats. Our assistant secretariat oversees installations, energy and environment, also things like permitting, infrastructure, et cetera. So we do everything from building the border wall to overseeing energy to cleaning up PFAS. It’s a pretty broad remit that we have, but within that, energy is really what my passion is and particularly nuclear energy, so when it comes to that, those are really my babies.
Robert Bryce: Sure. And by way of background, Dr. Waksman, you have a PhD in physics from the University of Wisconsin, expert on fusion. I think as we talked about before, that was your doctoral work. So let’s talk reactors. This is really why I wanted to connect with you. You’re overseeing Project Pele. The project is being designed, I think, by BWXT, if I’m right. It aims to provide one to five megawatts of capacity at remote military bases. I’ll cut to the punchline: how soon could we see this reactor up and running?
(00:02:15) Jeff Waksman: So Project Pele, which I started back in the fall of 2018, is what I did for six and a half years before I took this job, and that reactor is under construction right now. That design is effectively done. You’ve got to be careful about that because the thing with any first‑of‑a‑kind hard tech is, your final design versus what you exactly build is always slightly different, so you always do another review of what you actually build as opposed to what you design. But the reactor is as designed as it could be. It’s under construction now. All the fuel is fabricated, and in fact we had a ceremony a few months ago when all the final fuel showed up at Idaho. The current intent is to have that reactor fully fabricated by some point in 2027, and then once it shows up in Idaho, realistically it’ll probably be six to nine months before it actually turns and reaches a megawatt. So sometime in early to mid‑2028, but we very much intend and have all expectations that this will turn on in this administration. It remains the furthest‑along microreactor in terms of an entire system, and it’s important to differentiate that because I think there’s been some confusion about the criticality tests which are happening this year in DOE, which are great and important milestones, but are just criticality. Pele is an entire system. It’s going to make the electricity, it’s going to be hooked up to a microgrid, everything, and that will be in Idaho National Lab. At Idaho National Lab, that’s right. So the fuel has already shown up. The reactor is currently being fabricated in a few different places, but it’s being led by BWXT in Lynchburg, Virginia, and when that’s completed, it will be shipped to Idaho and then it will be fueled in Idaho.
Robert Bryce: And for people who are listening, BWXT is publicly traded. It is the former Babcock and Wilcox, a very old‑line industrial firm. Babcock and Wilcox—I’m not showing off here—built the boilers that Edison used on Pearl Street. So this is a company that’s been around for a while. And correct me if I’m wrong, Jeff, this is the company that’s been building nuclear reactors for the U.S. Navy for decades.
Jeff Waksman: Yeah, in fact we make the fuel in the facility at Lynchburg in which the fuel for every reactor that the Navy has ever had has been built in that building. They have all come from that site in Lynchburg, Virginia. So if you actually go there, Admiral Rickover chose that site specifically. He wanted to build the plant there because it’s got a rail spur and it’s on the James River; it’s a convenient location to do big, heavy things.
Robert Bryce: Right. So then once that is fabricated, will it move by—because there was a lot of interest in the movement of the Valar Atomics modules by C‑17 cargo plane—are you going to, well, I don’t know whether you want to tell me or not, are you planning to truck them? Are you going to fly them? Or is that even discussed on rail spur, or how are you going to get it out to INL?
(00:05:01) Jeff Waksman: So the intent is to fly that reactor out as well. But the most important thing that we’ve been very clear—and this has come down from the leadership of the Pentagon—is that reactor has to turn on in this administration, and so if we get any sense at any point that trying to do the flight adds extra risk, then we will cut it off. But the intent now, we believe we have plenty of time to fly it out there. But flying it does create extra risks. You’re always afraid of breaking something, you have to do extra inspections once it gets there, so we will run all that down and a final decision will get made collectively in the Pentagon. The final decision lies with the program manager at SCO, but it’ll be a collective decision.
Robert Bryce: And in looking at Pele, the project and the work, I’ve just looked back at it: it started in 2022, and the target prototype delivery at INL was apparently in mid‑2024. Am I right on the timeline slipping here? I mean, the timeline slipped—is that history correct?
Jeff Waksman: Well, we started to design the program in 2019, and at that point we were hoping to turn on in 2024. By the time we awarded the full contract in 2022, we knew 2024 was no longer realistic. We were hoping for 2025. It is a challenge, and you know it’s one of the things that we’ve tried to impose—and to take it back a step, if you ever read anything about Admiral Rickover, he comes off as this real cynical jerk, negative, whatever, but then you realize once you try to actually build a reactor, it is very difficult. Anyone who’s read the paper‑reactor memo knows. This is just the nature of it: nuclear reactors are complicated, especially if you want to handle the whole balance of plant and everything that’s involved in it. Schedules slip to the right. But I give the comparison to SpaceX and nuclear. Every time Elon announces a schedule, he always ends up being two years late, but he usually delivers. And so the most important thing in Pele has been that there have been a number of government advanced‑nuclear‑reactor programs over the last 40 years; none of them got to hardware. The point of Pele was, above all else we had to actually deliver real hardware. This could not end in press releases, and that is what this is doing. Pele is going to get built, it is going to turn on, and it has already had major impacts because when the president signed his executive orders last year, one of the things they asked for was a reformulating of the Department of Energy regulatory requirements. Well, who did they turn to? The folks who did Pele, because we’d actually regulated a reactor under DOE licensing, and our team was able to come and say, hey, this is what worked, this is what didn’t work, here’s how you can reduce paperwork. And so they’ve come out with a new set of rules, and the reason why these crit experiments were able to go so fast this year under DOE licensing has been all that learning under Pele. You just have to build something and then get better at the next one and better at the next one. That’s the nature of hard‑tech development.
(00:08:03) Robert Bryce: So, as you talk through this, I want to emphasize this point. This will be the first microreactor—and micro, I guess in this case I’m thinking under five or 10 megawatts, I think is micro. This will be the first microreactor that actually gets to the hardware stage and actually starts producing power in the U.S. in modern history. I guess we can go back a long time, but am I right there?
Jeff Waksman: Yeah, so the Army had some microreactors back in the 1960s, but the last one of those turned off in the mid‑70s. So yeah, it’ll be the first one in 50 years.
Robert Bryce: So let me jump ahead because I think I met some—well, I did meet some people with BWXT in October here in Austin at the Texas Nuclear Summit. Why gas‑cooled? Why is this—there’s a school of thought, obviously, in the nuclear‑reactor world: go with what’s tried and true, go with light water; we know we have hundreds of operating years, maybe thousands of operating years, of experience with light water. Why is the military, why is the Army going with a gas‑cooled reactor with Pele?
Jeff Waksman: Well, while Pele is gas‑cooled, we’re actually not requiring gas‑cooled, and under Janus we may be considering other options. But the only other realistic option is heat pipes. The reason not to do water—I’m sorry, Janus, forgive me. Janus is the broader program that Pele is under?
Robert Bryce: Right, thanks for clarifying.
Jeff Waksman: No, Janus is a new program that we announced in October. So let me bring that back for podcast listeners. Pele, we started that seven years ago; that’s under construction now. When the president signed his executive orders in May of 2025, one of the things he said is that the Army is going to have the lead for installation nuclear power across the Department of Defense, and the Army is going to have to start a program of record and deliver reactors to power military bases under Army licensing. So we spent about four and a half months formulating what that would look like, and in October I stood on stage with the Secretary of the Army and the Secretary of Energy and we announced the Janus program. That program is currently under source selection, but it is going to be a well‑funded program, and that program is going to select multiple vendors and we will hopefully build multiple reactors in the coming future. We’re in source selection right now so I can’t talk about specific companies—we haven’t decided on specific companies at this point—but those don’t have to be gas‑cooled. They can really be anything. However, realistically it’s got to be gas‑cooled or heat pipes.
Robert Bryce: The reason you don’t want to do light water is…
Jeff Waksman: …that light water makes a lot of sense for the Navy—you’re in the ocean—but because most commercial reactors are light‑water reactors, they have to be along a large body of water, either along a big river or along the ocean, something like that. Well, we need to have the flexibility to not have to be there. The whole idea of microreactors is that you can move them other places. We’ll put them in remote locations, places in the Arctic, for instance in Alaska, potentially. So in order to do that, you have to pull yourself away from light water. Now, in theory, you could do molten salt, but generally in my opinion things like molten salt are going to be difficult for transportable reactors; those make more sense for bigger, non‑transportable reactors. And so generally what we’ve been seeing from all the microreactor companies coming to us is that they’re gas‑cooled or heat‑pipe designs.
Robert Bryce: To summarize that, if I could, Jeff—and this was the folks at BWXT—they very quickly said, well, weight is the key attribute here, and size, that you can make a reactor that’s lighter and more compact than one that uses water. Is that fair?
Jeff Waksman: Yeah, there are a lot of unique challenges to making a reactor small enough to fit inside conex boxes. It’s just different from when you have tons of space to build a gigawatt plant; you can put in all the concrete and whatever. You have to fundamentally think differently about your neutron shielding versus gamma shielding. It’s a completely different paradigm when it’s small, and it just turns out to be pretty convenient. And the other attribute is higher temperature, which—as the folks at BWXT said—gives you higher‑temperature output that you can use for multiple different processes if you want to in a remote location: desalination, other things. So size, weight, and temperature are the key attributes for gas.
Robert Bryce: Yeah, absolutely. And I know this is something that you’ve written about before. I mean, I’ve read a couple of your books and you talk about how we focus on the electricity in the trying‑to‑get‑off‑fossil‑fuels conversation, but people forget that most of what fossil fuels are used for is not electricity. We have a lot of other applications that need energy, and some of those are high‑temperature heat. And for the military that’s largely going to be heating in the winter, but there are other applications when it comes down to processing, doing different chemical and industrial processes; there are different things that we’re increasingly doing where you need high temperatures, and you can get that with fossil fuels or you can get that with these advanced reactors that get upwards of 900 degrees, 1000 degrees Celsius.
(00:13:19) Robert Bryce: So let’s talk about TRISO. You mentioned that the first load of fuel for the Pele reactor, which is TRISO—Tri‑structural Isotropic, I forgot the acronym there; you can fill me in.
Jeff Waksman: Isotropic.
Robert Bryce: Okay, thank you. So why is the Pele reactor using TRISO? And the second part of this is, isn’t that a problem? Because TRISO is hard to fabricate. I mean, this is a very difficult fuel to fabricate. A lot of the SMR companies are planning to use TRISO, but again back to the supply chain and talking about light water versus newer chemistries, this is one of the big issues. It’s not just how you cool it, it’s the fuel that you put into it. So in this case, why TRISO?
Jeff Waksman: So TRISO is definitely significantly more expensive to make than traditional fuel; you’re right, this is the Cadillac of fuel. But the logic is that fuel is generally a pretty small percentage of the cost of a nuclear power plant, and the intent is that yes, we’re going to spend more on the fuel, but in exchange we can spend less on a lot of other things. We have far fewer redundant safety systems because the fuel itself is its own containment. The fact that it can get so hot without melting just makes the system a lot safer and we can regulate it differently. So the idea is that you can hopefully bring down your costs, because what has really killed these big‑reactor projects like Vogtle is not making the fuel and making the core; it’s the concrete building they built around it. That’s what is really killing them from a financial perspective, so that is in large part why people are pursuing TRISO. Another benefit for us is the safety aspect. Obviously the first priority for nuclear is safety, and that’s always everyone’s top priority. But beyond that, if you were to ask the Navy, “Beyond safety, what’s your number‑two priority?” their number‑two priority is performance. They need extraordinary performance and they’re willing to pay extra and to have very highly trained personnel at that reactor at all times to make sure it runs. Well, after safety our second priority is ease and safety of operation and not needing to have so many highly trained people, to be able to have something that has much higher margins for failure so that we don’t need a bunch of Naval Reactors‑trained people operating these things. So Naval Reactors has looked at TRISO and said, this is insane, your uranium density in the core is so low, why would you do this? And the reason is because we want to have a core that is what we call walk‑away safe: if all of our operators were to keel over with a heart attack and there’s no one at this reactor and it starts to go wrong, the physics of the core itself prevents a meltdown. That is one of the advantages of TRISO, and because it has that safety capability, that is part of what allows us to have less redundant safety and drive down costs. So yes, we will have less energy per unit volume than a naval reactor, but that’s because we’re not trying to do what Naval Reactors does; it’s a very different problem we’re trying to solve.
Robert Bryce: Right. So when you say less energy per unit volume, you’re talking about power density per square meter, power density per cubic centimeter.
Jeff Waksman: That’s right. We’re significantly lower in watts per cubic centimeter than a naval reactor—way lower.
Robert Bryce: But you may have higher power density in watts per kilogram then. Is that fair?
Jeff Waksman: Yeah, potentially. We’re trying to see, and these are meant to be commercial reactors; these are going to be commercially owned and operated reactors. We want to allow these private companies to be creative and to come up with their own solutions, and more than anything we’re trying to have these costs be low. We need these to be affordable for the Army; they don’t need to be 10 cents a kilowatt‑hour, but long term they need to be something like 25 cents a kilowatt‑hour for this to make sense for the Army. Now, we recognize that it’s going to take a while to get there, and so the money that we are going to be spending over the next few years is to help these companies get to those sorts of prices. We recognize no one’s building a first‑of‑a‑kind microreactor and selling it for 12 cents a kilowatt‑hour. We get it. But in the same way that when NASA got SpaceX going, they had to put several hundred million dollars in to help get SpaceX going; once they got there, SpaceX was able to launch things way cheaper than NASA could. And that’s our hope: that we will get these companies over and through their challenges to the point that they get to nth‑of‑a‑kind, and these can be affordable so that any installation can purchase them in the future.
(00:17:55) Robert Bryce: So let me follow up on that. I like that because the nth‑of‑a‑kind is always the hurdle, the technological valley of death, as the term goes. The first‑of‑a‑kind is going to be expensive, and then as you build more, we know from project learning that costs come down. But you said something important there at the end: so that these commercial entities can build them on military installations. As I hear that, it reinforces for me that—as I look at this, and I’ve been pro‑nuclear, we’ve spoken briefly before, I’ve been saying the same thing for 15 years—it appears increasingly that the DOD could be the lead horse, the bell cow, when it comes to new deployments of nuclear in the U.S. It seems to me that the pathways that are laid out seem easier for the DOD, or Department of War, than perhaps any other route. Am I right there?
Jeff Waksman: Yeah, this has been one of my arguments for a military microreactor program, because there’s a recognition, as you’re saying, that we need to build a bunch of reactors. It’s the only way we’re going to get good at it. We can sit around waiting for a bunch of AP1000s to get built, and the administration is working very hard to try to make that happen, but it’s going to be very hard to start building five AP1000s a year; that’s going to be very difficult to pull off. But when you talk about microreactors and the size of these systems and what they’re going to cost, this is something that the Department of Defense can afford to do, can actually put in a few hundred million dollars and make this happen and build enough systems that we potentially get good enough. And as we’ve seen with the utilities, all the utilities want nuclear power but no one wants to be first. They all want someone else to go first; they all want the third one. So we will take that risk. Any company that’s bidding on Janus has to give us a minimum of two reactors, and realistically we’re hoping for more. What that exact number is—give us a few months and we’ll announce—but we want to get to multiple reactors, and that way the commercial sector is not buying the first one or the second one, but maybe they’re buying the sixth one and we’ve already done a lot of that de‑risking for them. I think that’s one of the things we can do uniquely as the Department of Defense. And also we have more reason to pay for this, because microreactors actually have operational value to the military where an AP1000 doesn’t. I can’t plug a military base into an AP1000; that’s not useful to me.
Robert Bryce: But your advantages in this, if I could just run through them thinking out loud: one, the obvious one, you don’t have to deal with the NRC. You bypass the NRC; you have your own licensing process; you have the land; you have the customer; and you have the supply chain in terms of the fuel as well. Those seem like four key advantages you have. Are there other ones that I’m missing?
Jeff Waksman: Well, also the military through its history is good at getting stuff done and bringing new technology forward—among other things, nuclear power. Nuclear power came out of the U.S. Navy. But there are a lot of other technologies. An example I like to give, because it blows people’s minds, is that the U.S. Army over the last 30 years has had a number of major breakthroughs in breast‑cancer research, and you go, why would the Army research breast cancer? Well, because you give the Army a mission and they will accomplish the mission. It is a mission‑oriented organization and, more than any other part of the government, the military is good at just getting stuff done. The program‑management capabilities, the contracting capabilities, just the ability to manage is unique. There are many examples of the military leading a technology, and I think that is an advantage we have over an organization like the Department of Energy or other civilian agencies that just don’t necessarily have that same drive that the military does.
(00:21:49) Robert Bryce: And as you say that, I think it’s important to remember—and this is an important but largely forgotten bit of history—that the Army operated the Sturgis, the MH‑1A, in the Panama Canal Zone—I looked it up again—from 1968 to 1976. It was moored in Gatun Lake; I’ve been there. It was a 10‑megawatt reactor on an old Liberty ship. It was a power ship before power ships were cool, and power ships are really cool now. But this is an important part of that history: this is 50 years ago, but still what’s old is new again in some ways. I like how you explain that the Army can lead on this again. And I guess, if we look back, we should have said, why did we let that program die? But hindsight’s always 20/20. Have you thought about that project at all?
Jeff Waksman: Yeah, absolutely. The Army had eight reactors that were built largely in the 1960s. The Army Reactor Office still exists, in fact, and it has now been moved underneath my organization. It’s obviously a lot smaller than it used to be. But the reason they stopped back then is that prior to the last 20 years or so, logistical challenges were about fuel to the front; when we were fighting in Korea or Vietnam, it was about whether we could get fuel to the front. What has changed now, what is dramatically different about energy, is that we are now at risk at home: we could have attacks on our electric grid, take down the U.S. electric grid, and we will be on our backs. We operate so much of what we do here at home, so having energy resiliency here is new; this was not a problem back in the 70s. So in the 70s they said, hey, we have budget cuts, it’s Vietnam, gas is cheaper than nuclear, so kill the nuclear program; gas is cheaper. But now gas is not enough. We know that, as we saw with the Colonial Pipeline a few years ago. Our adversaries with very simple cyberattacks can shut down our ability to move fossil fuels around the country. Having a source of power that does not need refueling for years at a time is a complete game‑changer from a resiliency perspective.
(00:23:54) Robert Bryce: So how many reactors then might the Department of Defense—the Department of War—how many might they build? In the U.S. alone, I’m just trying to imagine these numbers: 200 military bases in the U.S. that in theory could become grid‑independent or grid‑adjacent. You could have grid power, but you could effectively be self‑sustaining. So the question again: how many of these reactors might the department build, or buy, or commercialize?
Jeff Waksman: Theoretically, the Army did a study a few years ago about, with all the money in the world, how many different reactor sites would potentially make sense, and they came up with a number of a couple hundred. Now, I want to be clear, the Army’s not committing to building that many at this time. In fact, I can’t talk about specific numbers because the president’s budget’s not out yet and we’re in that time. But what I can say is that we are going to take a minimum of three companies, and they’re going to be delivering a minimum of two reactors each. So at least six, and it should be more than that. We have already released a list of nine Army installations that we are seriously considering for these. We did a down‑select on a number of categories and we came up with the top nine and we have released that. Eight of those sites are in the contiguous 48 states and there’s also one up in Alaska. It doesn’t mean we’re going to put reactors at all nine sites, but we’re going to down‑select from there because we’re trying to find the right balance based on the companies we select: which sites work best for them. We also need to make sure that we can work with a utility, because some of these companies have relationships with one utility or another. It’s a bit of an awkward seventh‑grade dance; we’ve got to match up the boys and the girls. But we’ll do that in the next few months.
Robert Bryce: I live in Austin, Texas. I think Fort Hood is just north of here by about 60 miles. I would assume, of those eight installations, am I right to think Fort Hood is one of them?
Jeff Waksman: It is. It is, because it’s one of the biggest Army bases in the world, and it has a significant power load. So I’m just making an assumption there, but yes, that would be on the list.
Robert Bryce: Because it’s one of the biggest Army bases in the world, right? And it has a significant power load. So I’m just making an assumption there, but yeah, I assume that would be on the list.
(00:26:04) Robert Bryce: Let me switch gears here just a minute, Jeff. We talked before briefly, some weeks ago, about the challenges in the nuclear sector. One of the things that you brought up was the importance of engineering codes that are promulgated by the American Society of Mechanical Engineers. Why are these ASME codes important, and why are they important for the nuclear sector?
Jeff Waksman: Sure, so you brought up how Babcock and Wilcox has been making boilers going back to the 19th century. These engineering codes actually originally came out of boilers. Back in the 19th century, when boilers started to become a thing, they started to explode on people. They started exploding and killing people, which is a bad outcome. And so then you had Hartford Insurance, Hartford Boiler Company, and insurance for boilers became a thing. A lot of people died, and they finally realized, okay, we have to collect a bunch of data and figure out how these different materials will behave under different conditions, and they built these codes. You started then having code stamps—people would stamp the boiler saying, this has been approved under this code. This is now applied to all sorts of engineering. I used to work at NASA; this is how you deal with rockets. Similarly with nuclear, there are nuclear stamps to say that we understand that this graphite or this steel or whatever is how it’s going to behave under these conditions and that we have confidence it’s not going to break over the lifetime of the system for these reasons. And this is based on large amounts of data.
So what is the data that we have in the nuclear world? It’s largely light‑water reactors; it’s what we’ve been operating, under very specific conditions. You mentioned before how these advanced reactors can operate at higher temperatures. Well, if you’re operating at higher temperatures than reactors have traditionally operated, you don’t have all that data. You don’t have all that history. The engineering codes are often very sparse in terms of data under those conditions. What we’ve discovered is that every microreactor program, when it gets far enough—whether it was Pele or other programs like MARVEL and DRACO that have gotten underway—they’ve all had this problem. They realize that their materials can’t be validated under the codes because they’re under these new conditions, and you just have to do testing. You have to take this material, heat it up, irradiate it, whatever you have to do to validate and basically create new codes. The Pele program is leading to a change in the ASME code for material because of something that we learned.
(00:28:32) Jeff Waksman: This also gets into what we’re trying to do under Janus, which is: let’s say we have four different companies and they’re using four slightly different flavors of graphite. That’s not very useful if I’ve got to validate four different types of graphite at these new conditions. If we can just get everyone to compress to the same thing, validate it for everybody, it’ll be cheaper and faster for everybody. So this is part of what we’re trying to do. It’s not just about choosing the design you like best. This is about creating an ecosystem for these advanced nuclear reactors to actually happen.
Robert Bryce: I like how you describe that—an ecosystem. So it’s not just that you’ve got the gas‑cooled reactor; let’s say plate steel or some kind of steel you’re using in that, you’re going to need an ASME code for that—for temperature, for pressure, and for radiation. But if you’re building a salt‑cooled reactor or something, you’re going to need the same kind of—well, is it right to call it a certification? It’s called a code, but it’s effectively a type of certification then. Am I apprehending this right?
Jeff Waksman: Yeah. So, you know, traditionally in nuclear they call it an “N‑stamp.” You put it on and say, hey, this is nuclear‑grade. One of the things that microreactor companies are arguing is, maybe less of the reactor needs to be nuclear‑grade. Maybe we can get away with commercial‑grade for more of the system, and that’s a conversation we’re having with the regulator. But regardless, you have to satisfy the regulator that your materials, whatever they happen to be, are going to be able to handle the environment that you’re saying they’re going to have to handle. That’s just a very complex technical conversation that goes on between engineers. We lived it every day on Pele; I have scars from this. Our team is coming in eyes open. Some of these startup companies kind of have this idea of, well, Elon Musk got to just blow up rockets in the desert, so maybe we can just screw around with reactors and blow them up in the desert. That’s not really going to fly in nuclear. We really have to have confidence that you’re not going to have a failure or a meltdown before we get started. These codes are what drive how we have confidence that things won’t actually melt down in the real world.
Robert Bryce: And to just finish that point, you said on behalf of the regulator, the regulator has to be confident. So here we have the regulator being the Department of the Army for the reactors that you’re going to deploy. But the same process is going to apply for anybody that’s going through the NRC with a different type of reactor as well. So different, but you may have one ASME code that will satisfy both. Nevertheless, it’s going to have to go through that rigorous process, which is time‑consuming.
Jeff Waksman: Yeah, and so the idea is, the NRC, the Department of Energy, and the Department of War all have the authority to regulate nuclear reactors, but the standards are the same, right? It’s not going to be such a situation that DOE says this graphite’s good, but NRC doesn’t. To use TRISO as an example, the TRISO flavor that we use has been validated by the Department of Energy. DOE did a whole bunch of research and said, under these conditions, this is how TRISO behaves; as long as you stay within this range, you’re okay. NRC has that data. So any company that comes to the NRC with TRISO, if they’re staying within the range of what DOE validated, NRC goes thumbs up, you’re good to go. If you’re trying to push beyond that—if you want to do some more advanced TRISO variant that’s outside the specs of what DOE did—NRC is going to ask for more data. Smart companies just leverage the stuff that’s already been validated rather than spend a lot of time and money trying to validate something new.
(00:32:01) Robert Bryce: So I know you have a hard stop here in just a few minutes, but let’s briefly talk about materials science. I don’t know if it was in your interview with Emmet Penney, my friend Emmet, or whether it was something we briefly discussed, but you said that the materials‑science part of this—and I think the ASME codes would fit under this—is the single biggest challenge that the nuclear industry has. Can you briefly tell me why that is? Am I right to think the ASME codes are part of that? And more broadly, what you’re thinking about when you talk about materials science?
Jeff Waksman: Yeah, this is the hard part of a reactor. Any nuclear‑engineering undergraduate student knows how to do basic neutronics. Anyone can design a basic reactor core to operate. If your teacher says, hey, I need a reactor that operates for this many years at this power, anyone can design that; there are basic neutronics codes that tell you how that works. That’s the easy part. The hard part is, how do you make sure that when you have these thin strips in your heat exchanger, or you have your power‑conversion system over here, that when you scram it a couple of times you’re not going to break something that’s going to take that reactor down for a year? That’s really the hard part, and that is what has been validated in commercial systems. These old PWRs, they know that they can run for years and years; they can scram as many times as they want and they’ll be able to withstand it. As you start to compress to these microreactors, that’s really the challenge. You start to make stuff thinner and thinner—how do you make sure it doesn’t shatter when you scram a few times? This is a challenge that every microreactor that I’m aware of has had, and it’s all materials science. Materials science is not sexy. One of the things I’ve come to appreciate is you have these huge amounts of data that these codes are based on, and you go, where did all this data come from? It tends to be that some post‑grad in the 1950s just sat there and did 10,000 experiments, wrote all the data down in his notebook, and that gets injected into the codes. Tons and tons of that data were collected back in the day.
So I’ve had people say to me, hey, maybe AI can just figure this out. But AI can’t figure out what we don’t have data for. Materials don’t always behave linearly. Once you have the data and you have the codes, then yes, AI can optimize your reactor. But that materials science has to be done. Someone has to put that material in a test, heat it up or put it under fluence or something to get the data. Once you have the data, then yes, you can use all the computer models and AI to optimize your system. But without the materials science, you have nothing and your reactor is not going to work.
Robert Bryce: We talked about the different regulatory regimes. I’m going to skip over the political support because I do see—and I think you agree—that we’ve seen a change on Capitol Hill with more bipartisan support for nuclear. Particularly given where the Department of War is on this and leading in this resurgence and renewed interest in nuclear, that’s very positive. So I’m going to put you on the spot here.
(00:35:02) Let’s assume blue sky everything for me: that the nuclear sector, both the commercial sector and what you’re doing within the Department of War, succeeds. How many gigawatts of new nuclear capacity will be online in the U.S. by 2035? Not refurbished reactors, not Palisades, not Three Mile Island—give me a number for the number of gigawatts of new nuclear capacity that will be online in the U.S. by 2035.
Jeff Waksman: I am going to say 20 gigawatts. I think that can be done, but the only way that happens is—it’s not going to come from microreactors. It’s not going to be that we’re going to build 10,000 microreactors. This administration is working very hard to make an AP1000 deal happen. They’re trying to get 10 AP1000s built, and if they could close that deal, that’s most of your way to 20 gigawatts. So I have optimism that they will get that done, and that’s how I think that happens. The reality is microreactors are great for learning how to build reactors; they’re great for military operational needs. But if you want to power the country with gigawatts and gigawatts, that’s got to be bigger systems. The learning that we do in microreactors is going to help us know how to build bigger reactors. We saw that when I worked with X‑energy on the Pele program, where a lot of knowledge flowed back and forth between the Xe‑100 program and their microreactor, because it’s the same materials, a lot of the same things. But it’s going to have to be these bigger reactors; that’s going to be the bulk of it. I’m an optimist, so I’m going to say 20 gigawatts.
Robert Bryce: Well, I like that number. I think 10; I’ve settled kind of on 10 because I think that’s doable and I’m trying to be sober. So last question, because I know you need to run and I’m glad we were able to make this happen. You’ve spent a lot of time in this arena. You’ve worked hard. I know you’re dedicated to it. What gives you hope?
Jeff Waksman: What gives me hope is that we have an incredibly smart group of people. Because Pele has been the fastest‑moving program and has had a lot of money, we’ve been able to attract the best talent from the labs. We’ve been able to say, hey, we need the world’s leading expert on this—oh, that guy, he’s at Oak Ridge. All right, when you need him, you can have him. So we just have an incredible set of people. We also have incredible support. I’m a Principal Deputy Assistant Secretary of the Army, whatever that means, but the Secretary of the Army and the Chief of Staff of the Army support this program. Every time I run into the Chief of Staff of the Army, he always asks me, “What’s going on with Janus?” And we know that this is supported over at the Office of the Secretary of War as well. So there is tremendous support up and down the government. We’re going to have a very significant budget that’s going to get released by the president in a couple of months, and we have this incredible team. So if anyone is ever going to build microreactors, it’s going to be now. This kind of combination has never come together and will never come together again. If this is ever going to work, it’s going to be now.
Robert Bryce: I like that, and that is hopeful. I appreciate the work you’re doing. It’s inspiring, and I’m glad to see that—my quibbles and my criticisms of the Trump administration aside—on this front I think they’re doing exactly the right thing, and I’m glad you’re leading the charge. So my thanks to Dr. Jeff Waksman. He’s the Principal Deputy Assistant Secretary of the Army for Installations, Energy and the Environment.
Thanks to you, Jeff, and thanks to all of you for tuning in. See you.
As always, please hit that like button. Thanks.
