Ep 456: Paul Wilson - Chair, Dept. of Nuclear Engineering & Engineering Physics, University of Wisconsin
Show notes
Hannah McGrath [00:00:58] Hi everyone, and welcome back to Titans of Nuclear. I'm Hannah McGrath, and our guest today is Paul Wilson, who is the founding principal of the Computational Nuclear Engineering Research Group and Grainger Professor of Nuclear Engineering and Chair of the Department of Nuclear Engineering and Engineering Physics at the University of Wisconsin-Madison. Which is a mouthful, but we're going to dive into a lot of your research today, Paul. So, thank you so much for joining us. And we'll also be talking a little bit about your role in bringing a new generation of engineers and scientists into the field of nuclear engineering.
Hannah McGrath [00:01:33] So, it feels only appropriate to start with a little bit of a background intro. So, if you could tell us a little bit about yourself. Where'd you grow up? Where'd you go to school? How did you get into nuclear engineering in the first place?
Paul Wilson [00:01:47] Sure. Yeah, nice to be here, Hannah. It's fun to chat with you and share some of the things we're going to talk about here. I grew up in western Canada, near Edmonton, in the province of Alberta. For American listeners, I like to sort of compare Alberta to being like the Texas of Canada. And whatever you think about Texas, you can probably apply to Alberta, and I'll let you make your own conclusions about how you feel about that. But oil-man ranchers, and a lot of the same sorts of sentiments.
Paul Wilson [00:02:15] I went to university, then, at University of Toronto, in an engineering science program, and specialized in the nuclear energy option in that program. I was one of only about six students at that time in that option. And in fact, the option has faded away since then, about three years after I finished. Then after that, I came to graduate school at the University of Wisconsin. And after a few years here at University of Wisconsin, I was super interested in accelerator transmutation of waste. And that was a topic that was on the decline in the US and on the rise in Europe. So, I went to my advisors here and I said, "I'd really like to do this research. Can you help me find some way to do that in Europe?"
Paul Wilson [00:02:59] In hindsight, I'm shocked that they were so supportive because they had already invested a bunch of time in me and they were happy to help me move along. And so, they did help me find a spot in Karlsruhe, in Germany. So, I went to what is now KIT; it was then the Forschungszentrum Karlsruhe. It was a federal research center at the time. It's now part of the university system. And I did my PhD there.
Paul Wilson [00:03:23] It turns out I did a very similar topic there that I might have done if I'd stayed at Wisconsin, but I didn't do that accelerator transmutation waste after all. But then, I had a great few years living in Germany, came back to UW, and then a couple of years later got hired as a faculty member. So, I've been a professor here for about 23 years, and I do research in computational methods and teach courses in nuclear engineering, often related to some of the computational and policy-driven things.
Hannah McGrath [00:03:53] Wow, cool. Can you just really quickly break down what is accelerator transmutation of waste for our listeners?
Paul Wilson [00:04:00] Well, this is a concept that's still around. And the idea at the time was that we would use large particle accelerators to produce a whole bunch of neutrons that would then be able to take the waste, radioactive waste, spent nuclear fuel, and try and convert it into something more benign so the disposal pathways would be easier.
Paul Wilson [00:04:22] Various groups are still interested in these ideas. Most nuclear engineers think the best way to do this is actually in a reactor rather than with a big accelerator. And then, there are all kinds of subtle technical arguments about why one is better than the other and so on and so forth. I'm kind of on the fence now about which would be the best way to do it. It all requires a whole bunch of reprocessing, and that's probably the biggest obstacle, rather than the system where the transmutation would actually happen.
Hannah McGrath [00:04:49] And given the current system of storing waste for long term in really carefully sealed containers, is this better? Is this something that's appropriate for a future where there's more nuclear energy and waste being produced? What's the goal of that?
Paul Wilson [00:05:04] Right. I mean, it is definitely looking at a future where you have concerns about the long-term management of used nuclear fuel. Part of my research, actually, is in modeling the nuclear fuel cycle, and tools that model the transition from one type of fuel cycle to another. And I got into this because I feel that in a society with a large share of electricity coming from nuclear energy, we will ultimately need to explore reprocessing, closing the fuel cycle, doing something different with used nuclear fuel.
Paul Wilson [00:05:40] There's a spectrum of opinions on this within the nuclear engineering community. There's one camp of people who believe that it will never be economic and it will always come with undue risks and we should never do reprocessing. There's another camp of people who think we should have always been doing reprocessing and it's too bad we aren't doing it today. And I kind of fall in the middle where I don't think the drivers for reprocessing are really there.
Paul Wilson [00:06:02] If you look at the economic and political drivers, it really doesn't make sense to do it today. But I do believe there will be a time in the future when it does. And so, one of the things that motivates part of my research is trying to understand what will the conditions be when it is time to do reprocessing and how will we know? How will we be able to see it coming early enough so that we can be ready for it and not be caught on the back foot, suddenly trying to figure out what to do?
Hannah McGrath [00:06:30] Well, I know this is an ongoing area of work, but do you have any initial ideas on how we might know when it's time?
Paul Wilson [00:06:36] Well, I think there's going to be three main drivers, and I think these are well-understood. One is going to be the cost of uranium, the availability and the cost of uranium. And in the history of nuclear energy, this was always imagined to be the primary driver. That running out of uranium was going to be a reason we need to close the fuel cycle. Now, that looks to be a couple hundred years away, and maybe longer. Humans are amazingly resourceful at finding more minerals when we start to need them. And so, I would say we've got at least 200 years and probably more if we start looking.
Paul Wilson [00:07:06] The other one is waste management. And depending on the strategy you use and the kind of repository you build and your philosophy around ultimate disposal, the benefits of reprocessing are fuzzy. There's a lot of work that I got connected to and involved with about 10 or 15 years ago that looked at this idea of how much benefit there was in a geologic repository from reprocessing. And really, to get a major factor, you had to have a pretty dramatically different fuel cycle. And then, it comes with economic costs that are unclear.
Paul Wilson [00:07:37] And then, the last one is going to be nonproliferation. And this one's a little more complicated, and it's unclear how if and when this would ever be a driver. There was a program in the 2000s called the Global Nuclear Energy Partnership, or GNEP for short. And this imagined a world in which there was a need for a different kind of fuel cycle to support a growing international nuclear energy enterprise, and a sentiment that there were certain parts of the world where we were more comfortable with some of those fuel cycle operations happening than others. And so, how could we set up an international fuel cycle and create the kinds of security and safeguards that would allow the global community to feel comfortable with a different kind of fuel cycle emerging? And so, that could end up being a driver.
Paul Wilson [00:08:25] I like to characterize that one as saying that there could be a global benefit to reprocessing, a global reduction of proliferation risk, by accepting a modest increase in some specific countries where we think that it can be well-managed.
Hannah McGrath [00:08:41] Very cool. Well, that opens a lot of doors to talking about your current research, so let's jump right in. Can I ask a little bit about that fuel cycle reprocessing simulation work that you're doing? I think Cyclus is what it's called?
Paul Wilson [00:08:56] That's right. One of the main products of my research group is a tool called Cyclus. It was developed in about the 2010 timeframe, a little bit later. The vision was to create a tool that was very flexible to lots of different ideas about nuclear fuel cycle modeling. Up until that point, there were a number of tools that were being developed within the US DOE complex. Most of them were designed to answer specific questions. And then when the question changed, it was a major lift to take that tool and sort of reconfigure it to ask a different question. So, that's why one of the main design features of Cyclus was to really aim for flexibility and make it much easier to plug in different ideas and different concepts. And the other thing we wanted to do was to make it open source and accessible.
Paul Wilson [00:09:44] To date, there's been a handful of institutions that have used it in different capacities, probably six or seven different organizations. Both universities and National Labs have contributed to it and participated. We have a new project just starting up right now primarily focusing on the front end of the fuel cycle and adding more realism so that we can actually have a tool that's useful for exploring some of the questions of today, like sanctions of uranium from certain countries, like the supply constraints of certain services in the front of the fuel cycle, and potential disruptions.
Paul Wilson [00:10:19] Everything I do in my group is software development and methods development, and so it was also a fun project to really try out some new ideas from a software side as well.
Hannah McGrath [00:10:30] Cool, cool. And can you maybe walk us through what it looks like to simulate a fuel cycle? What are the inputs? How does the software work at a high level and what is the output and how you might use that to make a decision?
Paul Wilson [00:10:46] Right. So, part of that flexibility that we built in is that the Cyclus itself just acts like a big broker of material moving between facilities. And the user and developer get to design what the facilities actually do. All they really need to be able to do the work in Cyclus is be able to request material of a certain type and offer material of a certain type, and Cyclus puts all those together; it's like a big commodity market.
Paul Wilson [00:11:16] Those are the things they're required to do. And then possibly, they can do other things. So if it's a reactor facility, it also will simulate what happens in a reactor. And it will turn fresh fuel into spent fuel. If it's a reprocessing facility, it will turn spent fuel into multiple streams of different nuclides coming out, different elements coming out the other side. If it's an enrichment facility, it will turn natural uranium into enriched uranium. And so, people can develop those facilities with as much fidelity and as much detail as they want.
Paul Wilson [00:11:45] And then, a whole fuel cycle model would start with a source, maybe a mine, and you would follow the nuclear material as it went through from facility to facility over time and explore what happens as we add more reactors to the mix. For example, do we need to add more fuel fabrication plants? And when do those need to be added? And how does that affect the flow of material?
Paul Wilson [00:12:09] Probably the main place that these were designed for is looking at, as I said earlier, moving from today's light-water reactor fuel cycle to some future advanced fuel cycle. And there, there are a lot of questions about how quickly you can do that. If you're relying on reprocessing light-water reactor fuel to provide the plutonium and the fissile material for your fast reactors, then you can only start fast reactors about as fast as you can reprocess the fuel. And so, you've got these bottlenecks that are built into the system. And different bottlenecks can lead to different outcomes as to the accumulation of certain kinds of material as a function of time and so on. So, that's kind of what the whole thing is about.
Paul Wilson [00:12:51] One of the real advantages of Cyclus that comes with that flexibility is we can really easily add new technologies into the mix without having to retool the entire infrastructure.
Hannah McGrath [00:13:03] Cool, cool. Okay, so it's about really understanding the network of all the different players in a fuel cycle and how they all balance together at an ecosystem level, maybe to use a nature analogy.
Paul Wilson [00:13:15] Exactly.
Hannah McGrath [00:13:16] Cool, neat. And in terms of the different products... If we talk about the end stage, the reprocessing, how exactly is it that you end up with fuel for a Gen IV reactor, a fast breeder reactor, from an existing light-water reactor? Can you talk a little bit about that?
Paul Wilson [00:13:37] Sure. So, in the real world as it's being done today, the country in the Western world that does most of the reprocessing is France. And it's a chemical engineering process, largely. And so, you take the used fuel, you dissolve it into an acid, and then you use various technologies to separate different chemical elements from each other. In the process that's been used historically in the US in the weapons complex and a similar process being used in France today called the PUREX process, they largely focused on extracting of plutonium. And so, in the end you basically take the used fuel, put it in, and what you get out is a stream that's mostly uranium with a little bit of whatever comes along with it, a stream that's mostly plutonium, and a stream that's everything else. And in France, they use that plutonium and mix it with uranium to put it back into light-water reactors.
Paul Wilson [00:14:27] If you were to move to a fast reactor system, you would maybe mix that plutonium with natural uranium and put that into a fast reactor, probably with a blanket of natural uranium where you would breed new plutonium and you would have material sort of going in a cycle, coming into that reactor as plutonium, partly, and it's partly natural uranium. And the stuff that came in as natural uranium would have bred some plutonium. That would be reprocessed to make the plutonium for the driver, and so on and so forth.
Hannah McGrath [00:14:55] Gotcha. Is it the potential, then, to have a fully circular fuel cycle? Like, a completely closed loop?
Paul Wilson [00:15:02] Yeah, there are many different ways to design them. And a big study done in the early 2010s or so called the Evaluation and Screening Study looked at and classified nuclear fuel cycles into 40 main categories. And it turns out that you can capture every possible idea in these 40 main categories. And some of them involve fully circular fuel cycles, where the only thing going in at the front is natural uranium, and the only thing coming out at the back are fission products that need to be stored or disposed of for orders of magnitude of hundreds of years, maybe a thousand years.
Hannah McGrath [00:15:38] Right. Cool. And are all 40 of these in Cyclus?
Paul Wilson [00:15:44] I mean, Cyclus is able to model most of them, I think. I talked about how it's easy to make new modules for facilities. Some of the facilities may need to incorporate some degree of fidelity and modeling that we don't currently provide. We've never had the funding and the opportunity to try and do all 40 of them, but we have modeled two or three... Well, I shouldn't say that. That study was done to evaluate which ones were deemed the most promising, and six or eight of them rose to the top. Cyclus has been used to model at least half of those.
Hannah McGrath [00:16:22] Gotcha, cool. All right, I'm seeing how the fuel cycle analysis and the nuclear security nonproliferation policy side of the research in your lab are tying together. But I want to talk a little bit about the radiation transport in complex geometries and understand how that goes with the other two or maybe stands on its own.
Paul Wilson [00:16:45] Sure. I mean, it is largely separate. I can weave a narrative and we can come back to that. But the network is really driven by performing radiation transport calculations on complex geometries, as you just said. And really, what started that off was work for ITER, the big fusion experiment in the South of France. As they were finalizing engineering drawings and wanted to understand what the radiation environment would be around the device, they needed to find a way to build models to perform radiation transport calculations. And historically, in the late '90s, early 2000s, they had humans reading engineering drawings and manually creating the input for a tool like MCNP, which maybe people have heard of, the Monte Carlo radiation tool that's sort of the gold standard out of Los Alamos.
Paul Wilson [00:17:35] It was like one or two person-years of effort to take an engineering model and create an input file that MCNP could use to perform radiation transport. And as you can imagine, as you're finalizing this engineering design and you're tweaking things and making changes, now you've got to figure out how to close the loop. And so every time you make a change, you can't afford two more person-years of effort to update that model. And the models were so complicated, it wasn't easy to figure out how you make changes.
Paul Wilson [00:18:00] And so, this really spurred an international effort, and three or four different groups responded to automate the way that we take CAD models that were built by engineers for really complicated design work and be able to directly incorporate them into Monte Carlo radiation transport. Most of the solutions around the world took an approach to translate the CAD models into the language of those radiation transport tools. There was one called MCAM, in China, which became SuperMC. There's one called McCAD, in Germany, which still is sort of around. There's a newer one that was developed in Spain called GEOUNED.
Paul Wilson [00:18:39] But our solution we call DAGMC took a very different approach and said, "We're going to use a different representation of the geometry. Rather than converting it to this form that the Monte Carlo codes already know, we're going to make a little software library and we're going to use a form that's much more convenient. And we're going to allow those Monte Carlo codes to directly access our software library to figure out the answers it needs, where that complex geometry is."
Paul Wilson [00:19:07] We've been doing this for... I don't know, 16, 18 years. And we've incorporated this library with MCNP and offered that up to different people. We've incorporated it in Geant4, which is international open source code. We've incorporated it as a demonstration into Tripoli, which is the French code, although that now is about 14 years out of date, that implementation. But most excitingly, it's being incorporated as what we call a first-class part of OpenMC, which is a relatively-newer open source Monte Carlo tool that's available in the United States and largely driven by Argonne National Labs. It's seeing a lot more use now because it's available off the shelf in OpenMC. And users can just download it and run it and they don't have to do a lot of work to make it work.
Hannah McGrath [00:19:57] Cool. What's the advantage of the approach you took having the Monte Carlo codes access that software library as opposed to translating the CAD models in? Is it easier on your computer?
Paul Wilson [00:20:10] Right. There are a couple of advantages. It sometimes can be a little harder on the computer. Computationally, it may or may not be much better, but we think it actually saves human time more so than it saves computer time. Gotcha. So, one of them is that in certain CAD models, engineers can design things to have shapes that simply can't be represented in the native representation of these Monte Carlo codes. They're generally limited to spheres and cylinders and cones and planes and things like that. And if you've got some fancy shape... Think of the aerodynamic shape of an automobile. You can't represent that as spheres and cylinders; you have to make a lot of approximations. And often, humans need to spend some time and use their engineering judgment to decide what's the right kind of approximation to make that work. So, that's one advantage.
Paul Wilson [00:21:00] Another advantage is if you want to do multiple kinds of physics simulations on that same model, then you're going to have different ways that you represent that model. For finite element analysis, you're going to build some sort of mesh, probably. And it's going to be derived from this geometric model. And if you have to convert that geometric model to be a slightly different shape, then you're going to need to have a mesh derived from that slightly different geometric model if you want to be consistent. And so, it adds a little complexity in making all those things fit together.
Paul Wilson [00:21:28] Since we developed this tool, another approach that has become more common is just to generate a mesh of the entire model and use that to do the radiation transport calculation. And that's a perfectly good approach, and that can also be useful in connecting with finite element analysis. But what we have found is if you don't need to resolve either variations in your model or variations in your solution within certain regions of the problem, then that mesh can be cumbersome and can be overkill. And so, what we found in our approach is some regions can be meshed, but some regions can just rely on our surface representation. And then, we can sort of marry the best of both worlds.
Hannah McGrath [00:22:15] Cool, cool. Can you give an example of an area that you might need to mesh versus one that you could just leave as a surface? Is there a physical property that makes one need that increased resolution?
Paul Wilson [00:22:26] In many of the problems that we do that are not reactor problems, like a shielding problem for a fusion power plant, then you may care about the damage to a given layer of your system but not care about the radiation damage to other parts. And so, you can represent many of the regions from the first wall and the blanket as whole cells. And they're homogenized, so they have homogeneous material compositions. And then, it's the layer at the back, maybe the magnets or the vacuum vessel, where you need to know what the 3D distribution of the radiation damage is. So, you only have to mesh that layer, and the other layers can be represented as whole objects.
Hannah McGrath [00:23:09] Cool, cool. And I skipped over this a little bit, but I think maybe for listeners who are interested but not entirely familiar with the Monte Carlo method, could you explain a little bit how that code is applied to radiation transport?
Paul Wilson [00:23:23] Sure. I like to think of it as sort of the most intuitive way to do radiation transport. I like to tell students, "You're riding on the back of a neutron. And it gets emitted from some source, and then you are using random sampling, random numbers, to track what that neutron does. So, it has some probability of traveling a certain distance, and you roll the dice and you figure out how far it travels. And when it reaches that point, it's going to have a collision. Then, you roll the dice and figure out what kind of collision happens and which way it scatters off and which direction it goes and how much its energy changes. And you just keep doing that over and over again."
Paul Wilson [00:23:59] For a given neutron, you do that until it leaves your system, and you do that for millions or billions of neutrons. And then from that, you can take the average behavior of each of those and figure out where the neutrons go and what impacts they have, either causing radiation damage, activation, heating, things like that.
Hannah McGrath [00:24:19] Cool, cool. I can imagine that for a nuclear reactor with many, many tens of millions and billions of neutrons, that must be, computationally, very intensive.
Paul Wilson [00:24:31] Right. If you want to simulate a nuclear reactor and you care about more than just whether it's critical or not, then you probably are going to have some mesh in which you're trying to figure out where the power distribution is or things like that. And as that mesh gets more and more fine, you need to simulate more and more of these neutrons to get a good answer. And so, often that leads to highly-parallelized simulations where you may run a billion particles on each of 100 processors so you get 100 billion total particles and get enough detail for your solution.
Hannah McGrath [00:25:07] Right. Okay, so the Monte Carlo as applied here is an approximation of the system. It's something so complex that you couldn't actually sit down and do the exact math on the direction of every single neutron in the system and know for sure where they're going to go, but this is very close.
Paul Wilson [00:25:27] Right. I mean, there are two main approaches to doing radiation transport. And historically, we have used approaches that formulated as a big mathematical equation. But there, you have to make approximations about how you represent the whole system in both space and energy and direction. And those approximations can be tolerable for certain kinds of systems and can be challenging for other kinds of systems.
Paul Wilson [00:25:55] Monte Carlo radiation transport doesn't have to make approximations in space, energy, or angle. But as a tradeoff, you've got to run many, many, many histories to get a statistical analysis that is reliable. And so, in one case you're trading off making a really big problem because you have to divide up in space, energy, and angle in really fine ways to get the right physics. And you're trading that off with many fewer approximations in space, energy, and angle, but many, many more particles that you have to simulate.
Hannah McGrath [00:26:28] Gotcha, gotcha. Very cool. I know you hinted a little bit earlier at adding some realism at the front of your fuel cycle analysis tool. What are you working on right now? What do you expect to be working on soon? What are you excited about?
Paul Wilson [00:26:45] Sure. I mean, I'm really excited about the fuel cycle work that I hinted at. There are a few things that we want to add in at the fundamental level. One of them is a more realistic representation of costs. When we developed Cyclus, as a proof of principle... Maybe I'll say a little bit more about a key part of Cyclus... At every point in time, each facility that needs new material issues a request and says, "I would like to have some quantity of material that has some composition." And all the facilities that are able to provide that kind of material will offer up what they have; they'll make a bid. And the quantity that they have available and the composition they have may or may not exactly match the request.
Paul Wilson [00:27:28] But the consumer receives all of those requests and they get to use their own calculations of the physics and of the preference to decide which of those offers is the best match for them, if any. And they assign a preference to that, then the whole system collects all these preferences and resolves it and ideally tries to get everyone their most preferred material. So, that's sort of in a nutshell what's happening over and over and over again.
Paul Wilson [00:27:59] We would like to convert those preferences into a cost. We would like to say that rather than this generic notion of a preference, every facility is going to have a real or perceived cost to accepting a certain offer. And then, we want the whole system to minimize the cost. And so, rather than maximizing preference, we're going to minimize cost.
Paul Wilson [00:28:19] And so, the first step to that is adding places for the user to define elements of the cost of doing actual things and ways to accumulate costs. If you're an enrichment facility, you're going to have costs of the feed that come into your system, plus you're going to have costs of the process you're providing, and you're going to have to add those together to offer some cost to your customers. And so, figuring out how to get that notion of cost in there. That's one feature we want to add.
Paul Wilson [00:28:46] Another one which builds on that is the concept of long-range contracts, long-term contracts. Right now in Cyclus, every time step, let's say typically every month, this whole commodity transaction takes place and consumers choose their best choice based on what's available that month. So, that's a good approximation of a spot market, but it doesn't really incorporate the realism that exists with long-term contracts.
Paul Wilson [00:29:14] Typically, if you're a nuclear utility, you've got all kinds of long-term contracts, for uranium that you're buying, for enrichment services that you're buying. And so, we want to have a way to build in the idea that we form long-term contracts, and potentially put in things like penalties so people can choose to break long-term contracts paying some penalty because they perceive that the alternative that they're going to switch to is going to be cheaper, even accounting for that penalty. So, obviously we need costs to be realistic before we can do that, and then incorporate long-term contracts.
Paul Wilson [00:29:47] And then, the final piece I'll say that sort of has been in from the beginning is the notion of sanctions and tariffs and things like that. So from its original design, Cyclus had this hierarchy of facilities being operated by institutions which represent companies, operating entities. And those institutions operate within a region, which represents some geopolitical region. And so, this notion of cost and preference can be influenced by whether or not the two facilities that are trying to trade material are in the same region or not.
Paul Wilson [00:30:24] If we look at the headlines of today, with growing interest in restricting uranium access from Russia, then the question is can we simulate that and see what the impact would be of either an all-out sanction or some sort of tariff or some sort of financial mechanism that will perturb the flow of material and where it's coming from, where it's going?
Hannah McGrath [00:30:43] Right. And I imagine that shifting the rest of the model from a preference-based distribution towards everything having a cost might make that easier?
Paul Wilson [00:30:52] Yeah, it'll make that easier. I think one of the challenges is figuring out how to translate preference into cost in some cases. If a consumer gets two offers of material, neither of which is a perfect match physically, both of which have a real cost associated with it, the consumer is going to have to add some sort of virtual cost to represent why they prefer one of those to the other. So, that's going to be one of the research questions, how do we bring that key part to bear?
Hannah McGrath [00:31:25] Right, right. Assigning cost of preference, one of those minor economic questions that haven't been troubling the whole field since it began. That's really cool, though.
Hannah McGrath [00:31:35] I would love to take a step back. You are the chair of this program. I imagine that beyond your own research, you are also noticing what other people are working on. And we'd love to know how you've seen the field of nuclear engineering, and maybe nuclear science, more broadly, depending on how far your scope is, how it's evolved in the past couple of decades. What are people focused on now that they weren't before, or something that they used to be focused on and aren't anymore?
Paul Wilson [00:32:06] It's a great question. I mean, I think there are lots of little changes. I think at the very high level, the general trajectories don't look very different. The broad area that I think has been going on throughout my career is research in advanced reactor concepts.
Paul Wilson [00:32:22] Back at the beginning, in the 2000s, I was involved with the Gen IV roadmap effort. Before that, the main advanced reactor idea that was on people's minds was sodium fast reactors. Since Gen IV, there's been sort of a bigger portfolio of concepts that have been of interest, and there's a lot of R&D that goes into understanding how we can advance those designs into something we might actually want to build.
Paul Wilson [00:32:44] A big part of that, and has always been true, is in materials. Nuclear materials is a hot area of research. I think it always will be. I like to tell people in any engineering field, materials are often our biggest constraint. And so, if you're a material scientist and you can make a better material, then you can probably unleash all kinds of opportunity, both financially and technically. And so, a lot of interesting things are going on in nuclear materials.
Paul Wilson [00:33:09] Certain events cause particular changes in direction. I think if you look at Fukushima, it precipitated this whole research area in what's called accident tolerant fuels. Really, how do we design fuel in different ways to be less sensitive to accident conditions? Some interesting work has gone on here, some patents have come out of here in that area.
Paul Wilson [00:33:34] When we look at advanced reactors, the one thing that has changed dramatically is a focus on smaller reactor concepts and broader applications of nuclear energy. The sodium fast reactors of the early 2000s were all multiple hundreds of megawatts, 300, 600, whatever the case may be. And now, we see a lot more interest in small modular reactors. Some of which could be light-water reactors, like Last Energy, like NuScale, and some of which are more advanced concepts that look at different coolants.
Paul Wilson [00:34:07] And with that, I think the most exciting thing that really is coming around is alternate applications of nuclear energy beyond electric utility customers. In this country, I now believe that those alternate applications are maybe going to be what drive nuclear forward. Electric utilities are kind of conservative, a very economically competitive business to be in. It's difficult for electric utilities to take risks on brand new technologies.
Paul Wilson [00:34:40] I think if we look at what X-energy and Dow Chemical are doing... I think if that's successful... And cross our fingers... If that's successful, that combined heat and power application... There are many potential customers in our economy that have potentially deeper pockets than electric utilities and a very different risk tolerance profile. And we could see things like the industrial applications of nuclear energy and dedicated customers like data centers and things like that be the places that really, separate from the utility business, the electric utility business, allow some of these new ideas to be demonstrated. And then, following a handful of those demonstrations, utility customers may be more interested in following suit and adding them to their portfolio. So, that's a pathway that seems increasingly likely to me, and we'll see if that proves true.
Hannah McGrath [00:35:43] Yeah, I see a little bit of an analogy to some other renewable energy sources where you have some private group. And maybe with solar, it's people putting it on rooftops that take a while to test and prove the technology, and then the utilities are ready to scale it up and make it a significant part of their portfolio. So, that's really good.
Paul Wilson [00:36:00] Yeah, I think especially with microreactors. One of the other things that I think is true for microscale reactors is that they become a technology that can be considered on a community scale. Large light-water reactors have been technologies that require large corporate operators, a lot of regulatory infrastructure. And individual communities in the United States have ways to weigh in on that decision making process through all of our public intervention processes, but really aren't in control of those decisions. It's the utilities and public utility commissions and state governments and things like that that really have a say. And it creates, historically, a societal dynamic that puts nuclear energy in this special category of things that seem remote and distant from people.
Paul Wilson [00:36:51] Microreactors that are on the scale of tens of megawatts... Here near Madison, we've got tens of megawatt-scale solar facilities going in with great community involvement, investment, and engagement. And so, I think it's possible for communities to actually choose to adopt technologies at that scale and be in the driver's seat as to whether or not they want or don't want those technologies. And I think it really changes the societal relationship with nuclear energy.
Hannah McGrath [00:37:21] Cool, cool. Kind of a community nuclear program, it would be.
Paul Wilson [00:37:26] Right.
Hannah McGrath [00:37:26] Really cool. Great. Well, I wanted to take some time as we get further into our conversation to talk a little bit about your work getting young people interested in nuclear and getting more nuclear scientists, more nuclear engineers. So, have you gotten the sense that young people are more interested in nuclear energy now at a societal level? And do you think that has trickled into people actually choosing nuclear science for their education and for their career and profession?
Paul Wilson [00:37:57] Yeah, I mean, I think it's been pretty static for the last decade or so, but things are starting to look up, even just in the last couple of years. What we saw in our program was a big jump last year in enrollment. It's hard to explain exactly, it's hard to do the experiments to know exactly what's causing that. But my best assessment is that the national narrative really changed in the last couple of years. Up until 2020, let's say, we've saw a decade where the headlines were about plant closures, nuclear power plants being closed down around the country for various reasons. There wasn't necessarily that many of them in the grand scheme of things, but that's all that people were seeing about nuclear energy, and it led to a lot of... As high school students are making career choices, it doesn't seem like a wise choice. Even though if they'd asked me, I could have told them that even with those plants closing down, there are going to be plenty of career opportunities for them for the rest of their lives.
Paul Wilson [00:38:51] But in the last couple of years, we've seen a real shift in the narrative for a long list of reasons. But I think probably the top of them is a growing concern for climate change. And what that has led to is really a strong bipartisan support for nuclear energy at the national level. And that bipartisan support really, I think, trickles into the psyche of the population in a way that it didn't before. And that has then come to fruition in a number of major pieces of legislation where nuclear energy is getting substantive support as part of the legislative process. And so, I think that is really what is changing, and then the message getting out that nuclear energy is an important part of a climate change solution.
Paul Wilson [00:39:36] It's shocks me, still, sometimes to read surveys, including some just from this year, that some fraction of the population misunderstands and believes that nuclear energy contributes to climate change. And so, we'll have to keep working on that narrative. But the people who do understand that difference really, I think, recognize that one of the biggest ways to make a difference in the future energy-climate relationship is by expanding nuclear energy.
Paul Wilson [00:40:07] Our tagline for our department is "Saving this planet and exploring the rest." And so, we really use that a lot when we're talking to public audiences, high school students, prospective students. We're in the season right now of meeting with admitted students who've been admitted to come to college, and I like to point out to them that in their lifetimes, over half of all the low-emission electricity that was generated in their lifetime has been from nuclear energy. And so, as much as they may be excited about other alternative energy forms, the track record is that nuclear has really been carrying the load.
Hannah McGrath [00:40:45] Absolutely. And it's crazy to see things like Palisades reopening. That's something completely new.
Paul Wilson [00:40:52] Right, yeah.
Hannah McGrath [00:40:54] As a young person, that's exciting for me to see, and I imagine that it's just going to snowball even further.
Paul Wilson [00:41:01] And a state like California choosing to keep Diablo Canyon open. I think the story behind that really hits home pretty strongly in a state which has a reputation for being way out in front of our energy system, banning natural gas, heavy on wind and solar. And coming to the realization that given the constraints they've created in other parts of their energy system, they can't do without nuclear energy. It's just a necessity for their system to survive.
Hannah McGrath [00:41:32] Can you talk a little bit about your involvement with Young Generation in Nuclear and how that came to be? In addition to the broader societal trends of climate change... We're keeping our nuclear plants open. People are seeing a different side to nuclear energy. What does it look like to actually put effort into getting people more interested and more supportive of nuclear energy? What was your role in that?
Paul Wilson [00:41:57] Sure. So, I was in on the ground floor of NAYGN, North American Young Generation In Nuclear. When I was still a graduate student, I was active as a young member in the American Nuclear Society. And at that time, in the late '90s, sort of ahead of us was in the European Nuclear Society, they created this Young Generation Network.
Paul Wilson [00:42:18] Some of our ANS leaders had been over in Europe and seen the empowerment of these young people and the role it was having and came back and said, "We need to find some young people to empower in the United States and take the charge." In its early form, that resulted in sending a small group of us, at that time, younger folks, to go to the UN climate change meetings. So COP 4 in Buenos Aires in 1988 and COP 5 in Bonn, Germany, and a few since then. And really, what we did there was work with our European Young Generation counterparts to put on programming and change the view of nuclear energy to help all the delegates and all the participants recognize that there were young people who saw this as a technology of the future. And it wasn't just a dying technology, it was one that was sort of vibrant and we were excited about.
Paul Wilson [00:43:10] And that led in 1999 to the foundation of NAYGN as a new organization. There were about seven of us involved in sort of the founding group to kick that off. And the vision there was really to create an organization that embraced this vibrancy, the idea that this was a vibrant industry. There were lots of exciting things going on. And to be inclusive of a much broader set of people than a typical nuclear society had been.
Paul Wilson [00:43:42] And I'm still a very active member in ANS. I think each of our professional societies have a lot of importance in our professional careers as engineers and other disciplines. But NAYGN was a way to bring all of these different professions together even beyond engineering. Finance people, HR people who worked in the nuclear industry to really bring that youthful energy together and find a way to help support this thriving industry.
Paul Wilson [00:44:09] And so, from the beginning, NAYGN's mission was to really provide a place for young people to gather and connect across disciplines while also being active members in their own professional societies, whichever made sense to them. We never imagined it would be as successful as it is today. It's the 25th anniversary of NAYGN this year. I know they're planning a big celebration at their conference in June. And there are thousands of members and a lot of interesting things that NAYGN is doing. It's really settled in to be a premier place for young people to gather around their interests in their own industry.
Hannah McGrath [00:44:50] Cool. Well, I know we're coming to the end of our time, so I'll wrap up with an advice question. From your experience in your lab and NAYGN and this youth engagement that you've done in your career, could you offer some advice to, let's say, maybe someone who's just starting high school and thinks to themself, "I want to do nuclear engineering. I want to work in a lab, and I want to be publishing papers and conducting research in a department like yours." How do they position themself for success?
Paul Wilson [00:45:24] Oh, that's a big question.
Hannah McGrath [00:45:27] An easy last question.
Paul Wilson [00:45:29] Yeah, just do everything right. I mean, I think if you go back to the high school level, obviously, there are sort of the obvious questions about engaging all the math and physics you can to be prepared academically. I think one of the big things I would say to students at that stage is really think about whether you've learned how to learn and the way that you approach taking on new knowledge.
Paul Wilson [00:45:54] I think another thing that's sort of a passion of mine is it's never too early to start understanding how this technology fits in a societal way. I think nuclear energy is in a special place where all of its applications really come with sort of societal conversations. There's a lot of different kinds of engineering you can do and you can get into industries where everyone just takes it for granted and nobody really has much conversation about it, but nuclear energy isn't really in that place. So, the sooner people understand how members of their family, people they grew up with who are not nuclear engineers, what they think about it and what's important to them... And then, try and incorporate that into how you approach nuclear engineering. How can that steer what you think are the priorities... If you're going into research, what are the priorities for improving the nuclear energy enterprise in a way that's going to have a positive societal impact? You kind of have to know what all those other stakeholders think about it to really to make the right choices.
Hannah McGrath [00:46:59] Right. And I'll add, as someone who got interested in nuclear energy even after finishing undergrad, that I've found the field to be really welcoming of people who want to make a switch and dive in later. And I wonder if you agree with that?
Paul Wilson [00:47:15] Yeah, I think that's true. I mean, it's a small field. I guess it could go either way. But fortunately, what I've found is, as a small field, it is very welcoming. We sort of market that to our prospective students. We sort of point out to them that up here at the University of Wisconsin, we have this close-knit community within the discipline. I tell incoming first-year students that if they choose nuclear engineering, I and most of the other faculty are likely to know their name and say "Hi" to them in the hallway by the time that they graduate. And that's not true of other engineering disciplines where they have 1,500 students in a major. And so, I think that's a real point of pride for our program and for the relationships we can build with students.
Hannah McGrath [00:48:00] Very cool. Well, I think we've come to the end of our discussion time for today. But Paul, thank you so much for joining us on Titans of Nuclear.
Paul Wilson [00:48:08] Thanks for having me.