Thorium Reactors Are About To Get Very Serious

TITLE: Thorium Reactors Are About to Get VERY Serious CHANNEL: AtomicBlender DATE: 2026-03-28 ---TRANSCRIPT--- If additionally for some whatever reason an operator sits there and they don’t like what they see, they have one action and that is to stop it and the the reactor will go subcritical within seconds. We believe that the reactor could run without an operator. Whatever the regulator wants, that’s just he adds to the price of the electricity. We will still continue to have the R&D here in Copenhagen, but I think we will relocate part of the company elsewhere outside of Europe. likely we would have to look at the world and see who needs a lot of energy. Copenhagen Atomics says it can make nuclear simpler, cheaper, and scalable with a thorium molten salt reactor. So, I came to Denmark to see what they’re actually building and how close that vision is to becoming a reality. What we want to be really good at in combing and tons is making a reactor unit that can work with molten salt and thorium and be very efficient. So low cost is the most important selling point for us. Low cost energy output. What we are trying to do is basically change the way nuclear energy is priced. I mean, if we look at lightwater reactors is it it typically cost, you know, at least $5 billion for a small module reactor up to Hingley Point C is like $60 billion or something like that. It’s always many billions of dollars uh for those reactors and and basically it’s only something that governments can afford. Uh but I I think now with our type of reactor and our type of fuel, we actually have a chance to change this so that it becomes something that is in the same price range as gas turbines or or wind turbines. And that is complete game changer for the nuclear industry. Uh and sort of our business model. I think you should go and look at it on our website. You can you can see the price. I think we’re one of the only nuclear companies that actually put the price on our website and it’s $50 million for the reactor unit to get started and then it’s $2 million per year for everything, the fuel, the maintenance and and the replacement of components. So that’s the hardware and the fuel and then of course if you are the customer you you need to build the plant and get the license from your government and nobody knows exactly what that cost. We have some ideas we can discuss afterwards. But but that itself, you know, what you what you would pay coping atomics is $9 per megawatt hour uh of converted into electricity. Uh so it’s it’s um that’s super cheap. I mean, if you look at uh some of the other power plants today, most of them cost more than $100 per megawatt hour for the entire plant, but what we deliver only cost $9 per megawatt hour. So there’s lots of room for paying for your license and your financing cost and your uh buying the piece of land. So I have to say that it’s not electricity that comes out of our reactor unit. It’s heat and then somebody else the people who own the power plant they have to convert that heat into electricity. And we also provide the fuel because we’re very good at making uh very very pure fuel. uh meaning that we can um avoid corrosion and we can use less expensive material when constructing the reactor and uh so they they kind of go together and we offer that as a package at very very low cost. I knew already many years before that that energy is sort of a most important component to build prosperity in a society. So if we can make energy at very low cost and make lots of it, we can bring about lots of prosperity to people all over the world. and and then I stumbled upon this thorium that I’ve never heard about uh before and I realized that the the amount of mining that you would need to do to make a a thorium reactor uh would be a 100 times less than the amount of mining we do today with existing uh supply chains for uranium reactors. And I thought okay that’s a big deal like 100 times less mining that’s going to have a big impact on the price. Later I learned what the impact would be. Um but I sort of along the way I also learned that uh molen salt reactors are just a lot more smarter because they they can remove the fishing products while the reactors operating and this means that the you will get less spent fuel and you can maybe even help solve the spent fuel problem from all the existing reactors. Solving the spent fuel problem is is it’s technically not that difficult but it’s uh politically and uh from a regulatory point of view difficult. This facility is like uh 11,000 square meters or roughly 100,000 square ft. Uh and we will see most of it when we walk walk around. So uh let’s let’s go.

Let’s go. This is the famed onion core. So walk us through what exactly is going on here because this is a quite unique shape in terms of nuclear power. We have a monsole reactor which is heavy water moderated uh and the moderator the heavy water is at 80° and ambient pressure. So it’s not pressurized like a light water reactor. And that allows us to have thin vessel walled. What we have is a channel of molten salt flowing here in this gap. And then you have to imagine that this uh concentric shells they continue all the way around uh and they are then enveloping a central cylinder of heavy water uh and an outer heavy water here. And these two uh layers basically encapsulate the fuel salt uh and moderate the reaction and then instead of having control rods we use the adjusting uh of the heavy water level to um compensate for reactivity gain or losses. So basically this is fully filled with heavy water and this one has a variable amount of heavy water. Um and then we have a gap in between and because we have 6 to 700° salt on one and 80° water on the other we have a layer of graphite filled insulation in between so that we have a minimal heat loss of conduction through the insulation but we then have an additional heat loss of moderation of neutrons and stopping of gamas uh which then additionally heats the water. So we need to circulate the water to keep it cold. Uh and then in this geometry in if any of the pumps stops they’re placed below the core. So the salts and the water would drain back to respective tanks where they’re separated and the fuel salt drains to tanks where you can have decay removal. Uh, and the whole purpose of this core is to make a compact reactor that can fit inside of the shape roughly of a 20- foot shipping container so that you can factory produce and assemble them and ship them out. Uh, but when you make a small core, you get more neutron leakage because you have more surface area per volume. Uh, and so to not have a huge neutron leakage with a small core, we have a blanket which is a lithium thorium salt that circulates on the outside. And that’s basically because with this core geometry roughly half of the neutrons are used to drive the chain reaction and the other half more or less leaks and gets captured in the blanket where the forum decays to protonium and then uranium to produce new file fuel and this way we can make a breeder reactor in a very small form factor. The salts they pump from the bottom h and the water they pump from the top. Um so the salts are filled and pumped out from the top and the water drain down and uh the water then goes to pipes back to the uh water tanks where there’s just an orifice there. So if the pump stops there’s nothing restricting the water flow to drain back into the tanks. Thorium can’t form chain reactions like uranium 235. But if thorium is placed close enough to other nuclear reactions with uranium 235 or plutonium 239, which are both fizzle materials that release neutrons when they split, then thorium 232 can absorb a neutron and become thorium 233, which then quickly decays into protactinium 233 and eventually into uranium 233. And because uranium 233 is fizzle, it can split and release neutrons that sustain a chain reaction. So, we can put that newly created uranium 233 back into the fuel cell and continue the reaction for literally hundreds of years. Instead of just consuming uranium fuel, we’re using thorium to make more. In this geometry, we pump the salt and the heavy water to make a critical assembly. And in notably, it’s only in this geometry that it goes critical. And uh we then pump the salt and the water through the core to remove heat. If we didn’t pump the salt, it would still be critical, but it would barely produce any heat, only enough to heat up the materials and reach equilibrium. So to produce power, we have to remove heat. And then the reactor will uh respond to that amount of heat removal by producing an equivalent amount of power. And so we we pump the salts through the core from the bottom up and then back out and through heat exchangers, which you see some examples of here, which we use plate heat exchangers for. And then we have a secondary coolant salt that basically cools the fuel salt and removes the heat from the fuel salt. And that coolant salt has an tertiary coolant salt that then further removes the heat from the first coolant salt. And so by having these uh sequential circuits of salts with lower and lower temperature, we can remove the heat from the reactor while it’s running. So why do you have then three levels of heat removal? Why not have one or two instead? Why go to three? So because our fuel is liquid and the primary heat transfer medium, it’s also very radioactive because it has a lot of fishing products and actonites in it. So if you pump this salt directly to a steam boiler, you can imagine that if there was some sort of accident, you’re basically like disposing or spilling like huge inventory of fish products. So for this reason, we have a coolant salt that interacts with that fuel salt so that you take the heat from the salt within the reactor and then you have an non-raacttor salt that can move that heat to a boiler. But then in case that the first heat exchanger fails and the fuel salt starts to mix with the coolant salt, we have a tertiary coolant salt circuit to basically have multiple redundant barriers from the fuel to the outside environment. So here we have a bunch of different components um that we use in our systems. So in general for the systems we have to pump molten salt around. So we get salt from a supplier like this where it’s a powder. Um and these contain impurities such as um moisture and oxide species. So we have to melt those salts and get it into a ingot form where we have a solid chunk of salt where we’ve removed all those impurities which becomes important to eliminate or minimize corrosion. So that’s where we start and then now we have to pump that salt around where in a reactor that would contain our uranium or thorium or transuranics. Um and so we have to pump that around in a system to remove the heat. Uh and so we started by saying if we want to build a reactor, we first have to be able to build the components that runs a reactor and show that they work reliably because if you run a test reactor or even a commercial reactor and it something breaks that you could have fixed in a non-commercial setting, it’s much cheaper to do it there before you go to a actual running reactor. So we built systems to run these tests and these are then different components that we need to run those systems. And you guys have designed and built all of these yourselves. Yeah. So we do the design ourselves and then some of the parts we get manufactured by sub suppliers but we do the final assembly ourselves and the testing. Okay. And I understand it that there’s you know this is maybe pump generation one, this is pump generation two. Can you talk to us a little bit about you know what you were trying to do in generation one, how that went and then why you decided to move and what changes you made to go to generation two? What we’re trying to do in the first pump was to make a pump that uh could uh work for a 1 megawatt test reactor. So the current pump size is enough to a produce a few megawatts of thermal power. Um and then we wanted a pump which doesn’t have any dynamic seals. So we wanted to do what’s called a canned pump where instead of having a dynamic seal, you have the whole motor and uh and pump assembly in one unit but then separated by a can. And that allows the whole assembly to be compact and inside the furnace but importantly without any uh dynamic seals. So it’s can basically be fully welded and sealed as a component. So the first generation was just to show that we can actually make a pump operate inside a molar environment using design principles that we would want to use in a nuclear reactor even though it’s not important for non-raactive salts. Then we wanted to show that this type of design was feasible. Um, and that entailed designing a stator that can operate at 700°. So, the windings in this coil is made by thick copper wires with ceramic spacers. And we do this because then you have low voltage and high current. And that means that the breakdown voltage you don’t reach to spark inside the gap. With this design, we’ve now have pumps that has operated for 2 years. So, it’s reliable enough for a test reactor. But for commercial reactors, we want to run for five years at a time without maintenance and with high reliability. So instead of using bearings that are lubricated by the salt, we use electromagnetic bearings, which is the next generation here where you have the same design of high temperature coils that can run at 700° and you have the whole pump with the rotor and the pump uh assembly in one piece inside the furnace. But additionally, we then have other coils here that uh actuate the rotor. We here have um uh radial bearings that can pull up, down, and sideways. And then we have an actual coil that um controls the uh actual direction. So what we’re trying to do here is levitate the rotor inside the assembly. And that means that when we can only pull, we have to pull from opposite direction and always balance the forces with the position of the rotor. uh and this is done many times every second. Uh but what you end up with is that you have a levitating rotor that is pumping the salt without touching the sidewalls of the stationary parts. And by doing that you don’t have any u wearing surfaces. Uh and this is called in general pollins uh active magnetic bearings or AM uh pumps. Um and this has been demonstrated in other industries like deep sea uh pumps and compressors and many other industries to be able to run for five or even 10 years or more. So this is how we want to make a reactor that doesn’t need any maintenance because the pump is designed to last the lifetime of the reactor unit. Uh if anything goes wrong we always just shut off power and drain all the liquids to their respective tanks. So there’s not a lot of conditional if this happens then you need to do this or if this happens then you need to do some other thing. In all failure cases the outcome is the same is to shut it down. So the operation is extremely simple in terms of outcome there. There’s no way for the operator to I guess direct the control system on what to do. So uh you know what happens in the case when an operator sees something and maybe the system isn’t responding as expected. uh normally in like a lightwater reactor you have a control room an operator sees that maybe a parameter is out of line and they take an action to correct that if the automated system does not automatically do that. So here you’re you’re not even permitting that to happen. So how does how does that work? Just help me understand. So basically the the procedure that an operator would follow that procedure is the software we write here but because it’s a simple reactor there’s only a few procedures and um if this system fails to act on any of those let’s say something goes outside the limit and it fails to shut the system down then there’s a backup shutdown system that will shut it down instead with a high reliability. If additionally for some whatever reason an operator sits there and they don’t like what they see, they have one action and that is to stop it. But importantly that operator in the safety case is not required to meet dose target. So it’s sort of a voluntary thing that there can be an operator who can choose to stop a reactor but it’s not important in the safety case. So then like for call operator training or operator certification um what does the operator need to know and understand about the reactor? Because if you look at again a lightwater reactor, operators go through many years of training to understand all of the systems, the subsystems, how the reactor behaves. They have to understand reactor theory. Um, and what it sounds like here is they have maybe some parameters that they’re looking at, but their only action, and it’s a nonsafety credited action, is to shut down the reactor. Yeah. So they have no they have no method of being able to uh intervene or adjust if they wanted to. The reactor is running entirely autonomously. Yeah. Correct. Okay. Interesting. Uh and it’s it’s seen from the perspective of our reactors 100 megawatt and we want to scale to tens of terowatt of power. So it’s hundreds of thousands of reactors and it doesn’t make sense to go out and train and have operators for all those reactors that are trying to do a job better than a computer can do. But of course the computer has to have high reliability and the reactor has to have uh high reliability as an overall. But part of this is based on the fact that uh if anything goes wrong we always just shut off power and drain all the liquids to their respective tanks. So there is not a lot of conditional if this happens then you need to do this or if this happens then you need to do some other thing. In all failure cases the outcome is the same is to shut it down. So the operation is extremely simple in terms of outcome. So first of all uh our base model is that we don’t want any humans in the reactor building because we have many reactors next to each other. So a whole line of reactors and most of them will be operating while we swap out one of them and you know move the fuel from the old unit to the new unit. So that will happen while the others are operating and we want to do all of that with remote operation. So basically uh remote control forklifts, remote control cranes. So there will not be any humans in the reactor building while all of this happened. So, so we can still help the the site operator or the site owner replace fuel from one unit to the next. Uh, the next thing is that in traditional lightwater reactors, there’s a lot of work that goes on when you have to swap out the fuel assemblies and put in new ones. Uh, with a molto reactor, it’s much easier because it’s it’s liquid. So, it’s basically you just uh unload it into some tanks and let it freeze and then you melt it again with electricity and pump it back in. So, it’s sort of a I mean, it’s all uh just a couple of presses on buttons. It’s a it’s not it’s not manual work. It’s all completely automated. Yeah. I I think automation is probably something that especially if you’re scaling to, you know, like you said, uh 20 or 30 reactors at a single site. You’re going to have to start automating some things. Uh but you’re still going to have some amount of staff on site. If we look at uh Kendu reactors, they also have this automated uh uh fuel replacement and basically it’s kind of like that just for molten salt which is done of course a little bit differently but but it’s all automated. And then at the site there’s going to be people who operate the steam turbines and the steam generators. Uh but those are not under our control anyway. they the the operator of the power plant they they decide what brand of steam turbine they want what brand of steam generator they want and how they want to control those h and uh the special thing about a molden so reactor is that it’s completely load following so it it it only generates the amount of heat that you remove so actually it’s the the pumps that feed the steam generator that determines how much heat is generated inside the mold reactor so in any case it’s It’s not a nuclear operator. It’s a it’s the guy who operates a steam turbine who decide how much energy we take out of the molto reactor. And then if they go below a certain level, then it shuts down. Or if they try to go above uh sort of 100% of the maximum power, then it’ll also shut down. So they have to stay within that operational range. So there’s really nothing they can do to the reactor. So we don’t need sort of nuclear engineers to run the reactors because what what should they do? Yeah. So do you have an idea of like what does the staffing or the uh configuration of like you know is it a single operator per unit or is it one operator can oversee many units or how does that work? It’s basically up to the regulator. Uh we know that there’s some of the micro reactor companies that want to make their reactors uh operator free. So basically no operator. Uh we have not yet seen if those micro reactor companies are successful in licensing that. So of course we we look at that. Uh we also expect that some countries will sort of want to have a 100 people there looking at the temperatures or flow rates or whatever. Uh and it doesn’t really affect us. You know whatever the the the regulator wants uh the power plant owner will have to put there. And it doesn’t change how our reactor is operated or how it’s built. Uh but of course if you have 100 people they’re going to be quite bored. But uh but if that’s the requirement from the regulator that that just adds to the price of the electricity and that’s what they want. Uh there’s only one button. There’s a stop button. That’s the only button that they can touch. And if they see some data like temperatures or flow rates or whatever they don’t like, they can press a stop button and the the reactor will go subcritical within seconds. So it’s it’s super easy operation. And of course they can press this stop button every day. then the power line operator will probably get angry uh when the reactor uh the automated system detects that there’s something out of line some parameter maybe we’re getting too hot or we’re getting too cold uh or there’s too much power or not enough power too many neutrons too much radiation you have a leak what happens when that is detected and how does that system work yeah so for normal operation of the water level since the reactor is autonomous we have a P loop that looks at the temperatures coming in and out of the core and looks at the neutron level, specifically the change in neutron level and then it adjusts the heavil uh to reach a certain target or a certain average fuel salt temperature. And if we have some sort of transient that could be caused by a blockage of a pipe or damage of a vessel, then we will either see it on the temperatures of the salts. But more quickly, we can actually see it if it’s a power transient on the neutron level or the change of the neutron level. So the first derivative and when there that derivative becomes larger than a certain value you um actuate a shutdown or trigger a shutdown and that stops the pump which allows all the salts and heavy water to drain to their respective tanks. Okay. And where are those tanks located? So here in this simple mockup which is the first prototype we built, we have the salt tanks here underneath the core and then we have the heavy water tanks over here uh to the side and here there’s a thin wall but in a actual reactor that is a 2 m thick wall of neutron and gamma shielding that separates the high dose environment from the lowd do environment. And only in the core do you have a critical assembly. As soon as you shut the pumps down, even when the water level drops just a few centimeter, the reactor becomes subcritical. I want to talk a little bit about safety and and licensing approaches. So, uh, one of the things that we look at in light water reactors face is what can go wrong and then how does the system respond and ultimately it seems here that the response of the system is to go through a shutdown where everything is drained out. Um, one of the scenarios that can happen and is considered in lightwater reactors is things like blockages like you had mentioned. Um, inadvertent actuation of systems is another. So in in this type of reactor design, one of the things I was curious about was what happens if let’s say you know you’re controlling the reactivity level with the level of the heavy water. Let’s say the system malfunctions and provides excess heavy water. Now your activity goes up. Uh so how does the system respond? Can you kind of walk through what is the process of what’s detected, how it’s detected and how the system responds? Yeah. So if we take a sort of simple example of a failure in the inner water pump speed and that it over speeds. So it starts adding more water to the core and the heavy water level increases in the core increases reactivity. then the fuel salt temperature will be the first to go up and the outlet temperature of the core will basically rise and the normal temperature is 700 degrees and when you go a few t degrees above that you reach a trigger limit where you would shut the reactor down. So a normal transient like this is fairly small and in many cases it wouldn’t even trigger a shutdown the system would just respond but if you have a persistent maintained uh over speed then you would reach this upper temperature let’s say 720° And then you would trigger a shutdown which cuts the power to the pump. So even if the controller of the pump is malfunctioning, if there’s no electricity to the pump, it shuts down. Okay. Um the sensors that are doing those temperature sensors and neutron radiation sensors, are those all safety related sensors then? Because if you’re crediting them as being able to detect an anomaly and shut down, I would imagine they would have to be safety related. Yeah. So we have two SIP systems. One system is the electronics and software that we develop ourself which is not safety not safety credited. Um and then we have another system which is our safety credited shutdown system that is provided by an external supplier. Uh and so even if our own system fails the other supply systems comes with a high reliability in order to ensure that we actually meet these shutdowns and meet the ultimate dose criteria. Okay. Um another safety and and licensing question. So other things that we tend to consider are blockages. So you have the fuel salt circulating. Let’s say that you know for whatever reason maybe there’s excess excess corrosion something happens and you have a blockage of the outlet of the fuel salt. Uh how does the system then respond to that? Because originally you know you were saying that this when the system shuts down we just drain everything out. Is it still possible to drain that out if you have a blockage at the bottom of the fuel salt exit? I guess you’re pumping to the top, but I mean, if you still had a blockage at the bottom, it would not be able to to drain down. Or am I misunderstanding? Yeah. So, if you had a blockage at the top of the core, it would still be able to drain down, but you could uh postulate some combination of blockages where there’s pockets of salt that wouldn’t fully drain. So, what happens then if you have, let’s say, fuel salt that is is stuck within the core. It didn’t fully drain down. So uh in a sort of severe accident and you have like a pocket of stock salt you have decay heat in the salt. So even when you drain the salt the reactor goes subcritical. So if one circuit fails to drain another circuit would not be blocked. And since we actually have multiple circuits so either the inner or outer heavy water or the fuel all three of those needs to be continuously pumped or maintained in the core in order to be critical. So if just one circuit blocks, the other ones would still drain and you would still go subcritical, but you still have a possibility of damaging the core if it’s the fuel cell circuit because you have decay heat. And so if you’re not removing that decay heat because it’s stuck in a pipe, then you could possibly damage or make a hole in that structure or a leak of some kind. So we have a list of different scenarios that could happen from mild to severe. And all these different scenarios we envelope by the worst thing that we could imagine to happen and that we postulate even if it’s not necessarily feasible and that is that the whole core disintegrate at full power at the last day of operation and all the heavy water and all the salt mixes and spills into the container and that it starts boiling the water and produce steam. And then we have our shielding structure and then outside of that a building and that that building can contain that radiation and that steam pressure and still be below the dose limits even in that scenario. And by taking this extreme scenario and showing that we can meet those criteria, we’re then saying that all the other things that we could come up that was a lesser consequence than these uh they are all within this envelope. And so we basically only need to license based on the most extreme thing that could happen. Getting this past the regulator is going to be a challenge. But before we get to that, if you’re liking this kind of content where we deep dive on nuclear technology companies, then let me know by subscribing to the channel. Not only does that let me know what kind of content to make more of, but we can continue to do so without having to rely on sponsors. That means you get more direct access, real conversations, and fewer interruptions. So, if you haven’t already, go ahead and subscribe now. It’s the best way to stay up toate and see more of what’s going on in nuclear on the ground now. And now, back to Copenhagen Atomics. What these machines are here for is uh testing all the components. So before we can sort of be allowed to start a test reactor, the authorities want to know, you know, how does it work? What are the if we increase the temperature by 50 or 100°, what happens? Will it fail or like how will it fail? And if we lower the temperature, will it freeze or block or something like that? So we need to create a lot of test data in order to provide that information to the authorities to let them know what are the boundaries and where this operates safely. And of course it not it’s not only temperature, it’s also corrosive like how corrosive the salt is. It’s the mix of salt. It’s the the how fast the pump is spinning. Uh how how quickly you can do thermal cycling and before something breaks. What are the pressures that you can use in the system? All kinds of data and and we also need to test check how our sensors work in all those different conditions. Are the sensors working correctly and so on and and also in long term like if we run this for many many years how will things degrade and maybe something starts working differently or sensors give wrong data or whatever might happen the versions that of these test loops that you have now this is your current generation but you’ve had many generations before so why is that if we change everything at one time it it will be difficult to know you know why things stop working so we try to only change a few things in every time we make a new generation. Uh and in those design meetings, we all the engineers agree about what should be included in in the next version and what has to be postponed. So why don’t we go take a look and see what’s inside one of these things. Uh first of all, there’s a sort of a cold section where things are room temperature where we have the electronics and gas systems, sensors and pump drivers and whatnot. And then there’s another section over here uh which is the the furnace and it has the insulation around the furnace and everything inside the furnace would be between 6 and 700° when we run the test depending on what it is we want to test. And inside the furnace we have a tank down here where we have the salt and right now it’s all the salt is down there and it’s frozen. And then we have this this is the main loop. So the the pipe where there’s the main flow and then there are some other flows. There’s a filter here where we have a side flow and we have a flow sensor and another flow sensor there and we have some pressure sensors and then over there in the background we have the pump that drives the salt around. Uh and essentially this is a very basic loop where we just pump the salt around and around for thousands of hours and check that uh how the pumps and valves and filters and everything performs. But so like in here you have many different components. You have the pump, you have a filter uh and you have some piping. So what kind of things would you test in something like this and would you rearrange it? Uh is it you know modular and configurable depending on what you want to do? So this one I would say is mostly for testing filters and pumps. I mean that this is sort of the main purpose of this particular one. But of course if there’s a heat exchanger it’s probably the heat exchanger that is the main target. Some of them we do thermal cycling. Some of them we do uh run at very high temperature to see what it you know how does the higher temperature influence the durability of all the different components. One of the main issues is that if we build 10 that are supposedly exactly the same, we don’t get the same results always. Also, if we run a test for say 500 hours and we do that 10 times in a row, we don’t get the same results. And of course, that’s irritating. uh we need to get to a full understanding of why the results are not exactly the same. This one is running with the Fleck uh again a molten salt that has similar properties as the uranium and thorium salt that we eventually will use but it’s a less expensive. It doesn’t require any permits from the authorities. Uh, of course we need to follow sort of work safety guidelines with because it’s chemicals, but other than that, it doesn’t require any permits. It it cannot go critical, but we can still now then we have the right chemistry, we have the right density, so we can do a lot of the testing that we’re not able to do in this one. And uh, and of course, uh, every time we build these, we learn a lot about how we construct it and what to change. And that’s really great that we can do that. I mean there’s not a lot of other reactor companies that can just go in with an angular grinder and change things sort of and then run again the next week. And you we talked about this earlier this how quickly can we do iterations and as long as we don’t work with radioactive materials we can do iterations really fast here. This one we saw before was only for water testing testing the flow and pumps of that. And now you’re at basically you’re operating full temperatures here in this one. And so that’s why there’s a lot of this insulation uh on the outside and you have these large tanks. So what do you hope to learn that you would be able to use then in the next generation? Is it sensor controls? Is it uh pumping? Like what what is the most valuable information that you’re getting out of this version of the reactor? It’s a lot of these uh flow rates, draining rates, uh rates, um the temperature of different like pipe segments depending on how we do things because we in the reactor core we both have water at sort of 80° C and we have the salt at 600° C or 650 depending on where it is and and of course the there’s a certain transport of heat from the salt into the water and we we have to understand those depending on the flow rates and so on. uh so there’s a lot of learnings about sort of mechanical uh workings of the machine uh and uh also chemistry uh related to corrosion and other things. So it’s it’s mostly mechanical chemistry uh and also just construction wise how do we construct this in a way where it’s easy to repair or disassemble. This first part is the cold section where we have room temperature and then through the wall down there we can crawl kind of crawl into the hot section where the onion core you can kind of see part of the onion core all those um strips are heaters. So actually this is one of the things I want to highlight with the operation systems we have in our reactor. So many other systems we we build in the world they they have sort of single point of failure. So as soon as one little bitty thing fails then the system fails and then in nuclear we try to triple that or whatever. So we make triple systems so that if one of them fail we still have two that are working. Uh but in coming atomics we actually did it a little bit differently. So uh we have many computers. So we have more than 40 computers and we can lose a number of those. Let’s say you know onethird of all of those computers can die and the system still works because we have many temperature sensors and many pressure sensors and they are connected to different computers. So we when you do the statistics uh I mean it it’s um even if some of the computers or some of the sensors die we still have enough sensors to continue to operate and uh and this make it uh very resilient because we have so many of everything. The electronics is built with redundancy. So if one or a few modules or even a dozen breaks, the reactor can still continue to operate. Uh but if enough components break that you can’t continue to operate, it shuts itself down. And then instead of going in and replacing some of these components, we scrap the whole reactor and put in a new reactor. So the reactors are factionally built with the you know tamperproof seals and you you’re not allowed to change anything and even IE inspectors or similar can come and see that you’re building the reactor according to specification and what was installed on the site was not tampered with. Um and part of the reason we have this scheme is also because we want to remove fishing products online inside the reactor and transfer uranium from the blanket to the fuel which is sort of a mini hack inside of a container. So you can imagine from a safeguards perspective in order to make this commercially viable it has to be extremely simple and and and secure and that’s why the whole thing is completely sealed so that you don’t have operator access because every time anytime you would have that kind of operator action you would also have to make sure that no one tampered or altered with that system. Right now the salts that you guys are running you’re running fleck in some of the loops you’re running anything any other salts. We’ve used both fluoride and chloride salts. So basically equivalent salts to fleak but chlorides. Okay. And ultimately you’re what salt are you planning on using in the reactor? Yeah. So the for the fuel salt it’s lithium fluoride, uranium tetraflloride and for the blanket salt it’s lithium fluoride and thorium tetraflloride. And then the lithium is enriched lithium 7. When you make the change to the new salts that are closer to the fuel or with the fuel in it, how does that then impact the work that you’ve already done and you have, you know, tens of thousands of hours of testing? If you go to a new salt, does that mean you’re starting over or how does it how does the change affect that? To some extent, so in terms of the hardware, pumps and sensors and valves, there’s not a big change, but specifically in corrosion, of course, there are differences. Additional factors that we also don’t test today is having high burn up where you get fishing products in the salt and those will also contribute to the corrosion. So how do you get to that point then? Because once you turn on the reactor and you start generating power and a lot of neutrons and things like that, the effect on corrosion is going to be different. If we look at like a lightwater reactor, especially on the fuel rods, the hottest part uh you have a lot of oxidation uh and hydrogen increase along the zirk uh cladding of the fuel rods. Um and that’s just in a water sometimes water with boron and a few other things to control the chemistry, some lithium to control the chemistry and the pH. Uh but that’s a relatively I would say maybe simpler uh chemistry environment compared to what you guys are doing. So when you guys go and you turn on the reactor and you start producing neutrons and fision products, uh that’s in my maybe uh naive view a very different environment from running a purified salt. So how do you either prepare for that or account for that or like how do you then make the leap to we went from this nice purified salt to we now have you know 50 elements floating around inside. Yeah. So the first thing is that we will also purify the salts that goes into a test reactor. So the initial impurities are very low. Uh and the other factor is that we’re starting with a test reactor at 1 megawatt for 1 month. So the amount of birdup is very low. Uh so basically the fision product concentration is below one part per million. Uh so we don’t expect to see any uh additional corrosion in a test reactor that we wouldn’t see in a loop. Um and then we plan to incrementally increase the power and duration of test reactors. So we’re not going from 1 megawatt for one month to 100 megawatts for 5 years. So uh we’ll have through those test reactors we’ll have a good chance to study phenomena not just corrosion but many other things which is part of the reason for doing test reactors. uh and through that process we’ll be able to see more about how bad is it or is it something we have to do an additional work and if we for some reason needed to go beyond using stainless steel it’s it’s possible but there’s not a reason to do it before it becomes strictly necessary but from the MSRE that ran for 5 years in the 60s there is a lot of data on what happens when you have fishing products in the salt and they have a very similar salt to what we have and they didn’t see beyond tourum and brittle they didn’t see any adver increased corrosion. Actually, some of the data pointed towards the corrosion rate decreasing once the reactor was running, which was vaguely attributed to noble metals plating out onto the structures. So, there is even some expectation that corrosion might be less severe in a reactor. In the past, there’s been a lot of salts that have used fly uh that has this additional burillium in there, but you guys have elected not to do that. So, why is there no burillium? What what’s the advantage of adding burillium and why have you not gone that direction? Yeah, so for the Oakidge MSRE project, they were trying to make a molten salt breeder where they had a fuel salt which had pure uranium 233 or more or less pure. So it was only a small abundance of a few% of uranium tetllororide. And the whole purpose here is that we need something a dilutant to lower the melting point of the uranium and thorium tetllororide because otherwise the melting point is above a thousand degrees CC and we don’t have materials we can use at those temperatures. So they used a mixture of lithium and burillium to lower the melting point of the salt. In our case we use um 5% rich uranium. So there’s a lot of uranium 238 in there. So we need a higher concentration of uranium in the salt. And if we then use a otactic between lithium fluoride and uranium tetralloride we get a melting point of 490 degrees CC without any burillium. So basically because we have a much higher proportion of uranium in the salt we don’t need the burillium to lower the melting point. Uh and the added benefit is that burillium is can be toxic to some people that have burillium toxosis. So like an allergic style reaction to burillium. And since we don’t have to work with that, that also becomes a benefit, especially now when we’re developing and testing. So you have all these components in the supply chain that you have to bring in. You have the fuel salt, you have the fabrication itself of the things that you’re you’re building here. What is the cost of the reactor, the cost of the fuel, and how do you control the different costs? Like what are the the main impacts on the cost of what you’re actually building and then delivering? So, the reactor unit itself, basically the reactor core, the pumps and the heat exchangers and some pipes and valves and tanks and simple things and some electronics and sensors. All of that is roughly $6 million. Uh, and then you look at the fuel and you say, “Okay, we need two and a half tons of uranium. It’s 5% of rich uranium.” So, you can go on the web and look up what does that cost? And and then we need a special version because it’s for molden salt. So we need to convert whatever these companies that sell uranium, it needs to be converted into UF4 or u uranium uh tetraflloride. Um and then it needs to be mixed with lithium fluoride where the lithium is enriched. So it’s not standard lithium fluoride. It’s special enriched lithium which is also expensive. So, so the total cost of all that fuel, both the uranium fuel and the thorium fuel salt is at least $20 million, maybe quite a bit more in the beginning. And that there’s also a economics of scale. So those suppliers who supply those uh things, they of course ask, you know, do you want us to build a supply chain or or a whole factory for you? You can you guarantee that you’re going to buy this from us the next 10 years? And that’s really difficult for us to guarantee that we can buy such and such many tons in a row. So in the beginning we will have to pay the the the most of the cost of setting up that supply chain. So that’s we call that R&D cost but you can sort of you can yeah is it R&D cost or is it scaling cost or what is it? But uh but so in the beginning the fuel is going to be expensive much more expensive than the uh the reactor. And I’ll just give you sort of a ballpile ballpark number. Maybe $30 million for the fuel and the heavy water and all of that for for one reactor unit, but that price is going to come down over the years. You’re also designing for mass manufacturability. And you’ve said that you’re targeting ultimately to hit one reactor per day out of a factory. I know that’s probably a number of years away, but if you’re designing for that one reactor a day, how do you actually make that happen? One per day sounds kind of not not crazy, but it it is very very ambitious. The amount of components that go into our reactor design is less less than the amount of components that go into a modern car. In a car factory, you get the steering wheel and the and the wheels and the brakes and, you know, all the different components you get from suppliers and they send them all over to the car factory and then you assemble it in an assembly line. And we’re going to do the same. We already today we buy from suppliers and we know the price of the stuff you saw the reactor. So, we know the price of each component we put into it. So, we have the bill of material and we know how many days it takes for those suppliers to make those components. And that can easily be scaled up. That’s the that’s the least difficult thing in all of this. I mean they but sometimes investors ask us you know how are you going to license one reactor every day and that’s where the whole thing breaks down because of course today it takes like five or 10 years to license a reactor. So if you if you have to license one every day you have to have 3,000 licensed processes going. That doesn’t work. That’s why eventually we will have to go to the site the type license uh model. So this one, this was obviously the first one that we built that’s uh only ever run on water. So we uh and we did it fairly quickly just as a proof of concept. Um and it was uh it was almost all half done, right? Like the drawing was half done and we went, “Yeah, let’s just build it.” Right? And then you find the problems along the way and we sorted out quite a few issues with this one. And almost before this one was finished, well actually before this one was complete, we were building the next one. So we try and just keep pushing forward, you know, keep reiterating fast. Okay. Can is there anything in here that you’d like to point out about either the fabrication or the design or what was a challenge? Yeah, I guess you can probably see the core is the challenge, I suppose. Um, and that design’s changed a few times, so it’s different again in the the hot version that’s behind me there. Um, we knew that this one would only ever run on water. So, we tried to make it in a different way. Um, which, yeah, it works fine. It’s running water now. But then once we went to salt, we uh we changed it again. So, I think that was probably the biggest um or the most difficult thing to fabricate. So, why was the the core and this onion core? Why was this difficult for fabrication? Uh, it’s it’s literally the shape of it. So, and it is Yeah, you’ve seen the the breakdown of it. It’s multiple layers together. So you have to be able to assemble the first layer, then the next one, then the next one, and then there’s installation in between. So it was a bit of a learning process about how to do that. We have some really highly skilled fabricators that come from very different backgrounds like sort of architects, goldsmith, that kind of background. So they have a very different, you know, good good views on how to make things. But as I said, it continues to change and I think it’ll um the next iterations will be different again the way that we do it. Okay. Um last question then. So what does success look like for you at Copenhagen Atomics? Like what is your ultimate dream of if you could, you know, flip a switch and it would be magic. What does a a magic successful outcome look like? Of course, is turning that reactor on, you know, seeing it make power. So that’s that’s the end goal. Why would you go for an over moderated versus an undermoderated type of approach here? Yeah. So with the the neutron diffusion length, there is also some kind of a half-life associated with it, you can sort of say. So we need more moderation than just the mean path because we’re not having like a uh stacked uh tubes or stacked fuel pins. Um and we want the neutrons to be very well moderated when they reach the blanket. So that we need a thinner blanket to capture the neutrons in thorium. We could have a thinner layer of outer heavy water and then a thinner thicker blanket. But in terms of cost, the blanket is more expensive than the heavy water. So we actually want to optimize to have as little blanket and so as possible while still having a very low neutron leakage. And in our case, we have around 2% of neutron leakage. So quite well for a reactor that’s basically a little bit more than 2 mters in diameter. And additionally I would add to that also in terms of enrichment level which is basically a four factor that I should have mentioned before. Uh in terms of enrichment level we also want to use as low enrichment as possible and we’re using 4.95% rich so just below uh 5%. Um and if we used an higher enrichment level we could use a smaller more compact core and with less heavy water but in terms of cost it’s not actually beneficial both because Halo or higher enrichment is more expensive but it’s also less available. Yeah, we uh have some fisions happening here. We moderate those neutrons and then we go over into the blanket region. What’s happening in the blanket region? Yeah, so in the blanket we have lithium 7 fluoride and thorium tetraflloride and uh there when the neutrons are streaming out most of them are captured by the thorium where they then subsequently decay to protectinium and then to uranium 233. And in some reactor uh systems from the orchids from the 60s, they looked at separating protitinium because when protinium is in the blanket, they can then further capture additional neutrons and become non-fistal isotopes uranium 244 um which is not beneficial. But in our case, we actually have a quite large amount of blanket salt. So we can actually allow that protitanium to stay in the blanket and decay turanium 233 before we extract it and transfer it into the fuel salt. But in general we have around 3 cubic meters or a little bit less of blanket salt that is circulating in the outside and it then receives some heating from the capture of neutron and gamas. Um, and that salt is then cooled by a heat exchanger, same as the fuel salt and it’s circulated around. And then we have an online system where we can transfer your thorium or sorry, we have a a system running online where we can separate uranium from the thorium blanket salt and transfer it into the uh uh fuel salt with the uranium. You have that system developed already. All right. we have a proposed method and we have a a IP around how to transfer that. Okay. So I mean at that point then you’re also going to have um uranium 233s existing in the blanket. So you will slightly fishisioning and you will be slightly fishisioning. So that’s my next point is that you will have a small amount of fishision occurring after a while in the blanket region as well. So not just in the fuel channel. Uh do you know about how much that’s going to be in the fuel channel like the fuel channel versus the blanket channel if it’s in some sort of equilibrium state? Uh yeah but um let’s see in what units like it’s um well below 1% of the total fishing rate is happening in the blanket. Okay. So it’s very low compared to what’s happening in in the salt. So when you’re producing the protectinium in here, um it is a neutron poison, but one of the things is about concerns around proliferation and the idea that if you have a liquid fuel or a liquid blanket like this, it becomes relatively easy compared to a solid fuel in terms of being able to separate out materials because it’s a liquid and you can, you know, perform the right chemical processes in order to pull out protectinium directly. and then you would have theoretically a pure stream of uranium 233. So how do you prevent bad actors from doing that sort of thing? Yeah. So the first uh sort of uh proliferating or the say the first um aspect that helps the reactor in a proliferation aspect is that you have very high activity. So you have um uranium 233 but you also have some uranium 232 which has a daughter um with a high gamma emitter and you also have some fishision products. So the salt of this reactor for better or worse is highly radioactive after you run it. So it’s not something you can just handle with bare hands. Um but additionally this whole uh reactor container is uh contained inside of a shielding structure that is a welded steel structure for the entire 5year operation of the unit. So you only have very periodic access to the reactor and there’s no streams of fuel or blanket salt that are accessible from the outside. So that means that you only have access to the salt or the material when you open the reactor every roughly 5 years. It’s a little bit similar to a light water reactor where you have access to the fuel every one and a half years. And um it’s true that with any neutron source whether it’s fusion fision or espalation source you can create file material and in our reactor it’s no different we will create file material that could be misused but the way we protect it from a proliferation standpoint is by having the whole structure completely sealed for the majority of the operation and not having that accessible and then you can do inspections when the material is accessible and that is also why it’s important for us to have the removal efficient products and the transfer of uranium from blankets to the fuel happening online inside the reactor so that it’s not some stream that is coming from the side that you can access from like a worker’s perspective. And the main way you could misuse this is if you are able to change the reactor design, if you’re able to um uh modify a reactor to do some nefarious purpose. But the same is true for basically any nuclear reactor. If you just lightly roast the fuel in any lightwater reactor, you have weapons grade plutonium. So the challenges are more or less the same as any other nuclear technology. I think the difference between a lightwater reactor where the fuel is solid in the the pellets is then you know you have to physically take it out of the reactor, cut it open, you have gases that will come out. It all has to be done in in hot cell work. Whereas in a molten salt reactor, because it’s already liquid, at least in theory, you could then specifically target the protectinium. And I know you mentioned that uranium 232 is in there. Uh but if you’re targeting only protectinium, that is theoretically possible to then pull out as a specific uh proliferation stream. So you seem to be protecting that primarily by sealing off the reactor and preventing access. So I mean at that point is it maybe uh beyond your responsibility as the designer to ensure that people cannot have access to it. Um the example I’ll give is India and Pakistan starting with the can do reactor technology uh then use that for production of plutonium that eventually leads into a weapons program. In that case I mean it’s a deliberate attempt on part behalf of those governments of the direction they wanted to go. So does Copenhagen Atomics then see that okay we’re going to make sure that you know this is being delivered in places where proliferation is either not a concern or it’s very low risk because once a country has this reactor I mean theoretically they can take it and they can modify it and do what they want with it for their own needs. So how do you how do you kind of address and balance that? For one uh we need export approval whenever we send a reactor to another country. Um so of course there there’s from a governmental perspective some kind of restrictions on who you can sell to and where. Um and beyond that the reactors would still be under IE inspection. So uh when the reactor is sealed and when it’s accessed inspectors can come and see. So the regular sort of safeguards prevention and then uh copatomics will actually be the ones replacing and you can say operating the reactors. So it’s not that we’re selling them and then letting the customers figure out what to do with them. So there’s a level of control there as well, but we always have to then hold this up against the cost of creating like a clandestine or covert project from a nation state versus the cost of just doing uranium enrichment. Because if we make it uh the burden on showing safeguards too high for the companies, then we’re basically overpaying for insurance compared to companies or countries or covert actors just making an enrichment program. And that cost is probably going to decrease over time uh as well as the IQ to execute it. So I think the industry as a whole has a problem in terms of on a long-term timeline. How can you actually ensure safeguards of these material when it becomes incredibly cheap and simple to do enrichment? Molten salt reactors, I think it could be argued, are maybe even half a step or a step behind since there is no large near commercial scale type of reactor. So if molten salt reactors are in somewhat of a similar situation, how do you get ahead to the point that even with some like the Russians that they have decades of experience and they’re still struggling? How do you get to the point where yes, we can hit, you know, 80% plus? We already do this with our loops, these small units where we have mold salt and a pump and we circulate the salt and we we fight every week to improve the capacity factor of those and and we have sort of this idea that before we can get 95% capacity in these systems, there’s I mean there’s no chance that we could put a commercial reactor online. So we we we have to sort of we fight in an area where there’s no uh where we don’t have to slow down because things are radioactive. We’re actually able to move really really fast because we work with non-raactive materials. It’s still chemistry. So there’s of course some uh work safety uh things we have to comply to, but we don’t need to apply for license every time we change something. So we can change some things every day. And we do that and and the these guys out there fighting every day to to reach higher and higher capacity factor. And I think this is where we will win because where other people built one react unit every five or 10 years, we improve our stuff every day. Mhm. And do you think that then with a smaller team, if you’re iterating faster with a smaller team, that this is something that would be able to take on uh, you know, programs like the Russians or the Chinese where, you know, they have, sorry to say, like essentially unlimited resources. They could dump, you know, billions of dollars into these programs if they wanted to. And you guys are working on a much more uh you’re on a startup budget, which is just the reality of the situation. So how do you beyond just iterating more quickly, how do you then get to the point where okay yes we can now scale up and we can make something that will compete with uh these large state enterprises. Yeah. So so we we looked at that in the beginning like already before we started the company and we said okay light water reactors that’s super easy technology. Almost every lightwater reactor we have ever turned on kind of worked. I mean maybe they had not perfect capacity factor but they they worked right away and there’s been many different designs and almost all of them worked and I I had the good fortune to visit some of the first ones the the Hanford B reactor and the X10 reactor in Oakidge some of the really early reactors and when you go there and look at they were built in the 40s and they were so simple you’re like I could build this myself in my own garage I mean that’s how simple it is and you’re like okay so why why is it that governments French government for example, that spent billions on the Phoenix and super Phoenix. Why is it they cannot make it work? And and you sort of have to think what is it that they did wrong? And obviously I think fast reactors are just more difficult than lightwater thermal uh reactors. So that so that’s one thing. The fast spectrum is more difficult. The sodium is also more difficult. it can catch fire and and because of that people want all kinds of safety uh um things around the reactor and all that safety slow things down and make it more expensive. But we look there there’s been built more than 20 fast reactors in the past all by very uh rich governments and none of them became commercial. Uh, of course the Russian one is close to getting commercial maybe, but it’s still not there and like you said, they have worked on it for 50 years with large budgets and we knew we cannot compete with that. So, we have to do it differently and and I think we found somewhat of the right model with building all this stuff that we you saw today. We build full scale reactors here. I know very few startup companies that have already two full scale uh reactor units and are building the third. Even though we’re not allowed to turn it on yet, we can still learn a lot from the units we’re building. And I believe by the time we actually turn our first test reactor on, all these mechanical issues with mechanical things, pumps, flow rates, chemistry, all of that will be solved uh to a very high capacity factor. Not 100% but more than 90. And this means that we’re very confident that the first reactors are going to work. Uh because also today’s simulations are much better than they were in the past. And may I don’t know maybe you know that also it’s it’s more difficult to simulate fast reactors than it is to simulate thermal reactors. So I believe these simulation tools we have are adequate for actually knowing that what we built will work once we turn on the chain reaction. So where are you guys in terms of like being able to raise capital, being able to then allocate it wisely? Uh being here in Denmark, being here in Europe, it’s maybe a bit different than the uh what I’ll call the excitement that’s happening in the US. Uh how are you guys navigating that and how are you managing? We’re not seeing that help from the European uh countries. H there is a little bit of funding from European governments but essentially uh most of the funding funding here has to come from private investors. Uh we have been able to raise uh funding so uh we will soon announce sort of an a total investment of hund00 million. Uh but of course some of our you could say colleagues in the US was able to get government funding of more than that. Uh so it is easier for them. Europe is maybe not the best place to be. We will still continue to have the R&D here in Copenhagen, but I think we will uh relocate part of the company elsewhere outside of Europe because the likelihood that Europe sort of does what the US did in the last two years in the next 5 years is uh not high. It might happen but uh the the probability is not high. Uh so likely we would have to sort of um look at the world and see who needs a lot of energy. We also still need to see I mean so both India and China has embraced thorium energy or thorium based reactors and and China is currently operating a mold reactor with some thorium in it. So I think that’s great but in the US we still need to see the NRC and the DOE and so on support mold reactors. There there are a few mol reactor startups in the US but uh I would have hoped that they had gotten more support. Uh we’re supporting them but also a support from the US government. Uh I find it highly unlikely that we will make our mass manufacturing of reactors inside of Europe. It’s simply too expensive. If you look at something as simple as steel, I mean steel is three times lower price in China than it is in Europe. I mean that by itself tells you that there’s trouble. Uh also I think today China produces 60 or 65% of all steel in the world. And the reason they do that is because they have cheap energy. They also produce something like 70% of all aluminum in the world. uh and uh and they are growing their economy. They it’s not like they’re producing less and less steel. They are producing more and more and the steel factories in Europe are closing down. Uh and uh and I think what US is trying to do is create a lot of additional energy but not for steel and aluminum but for data centers. So I think what the US is trying to do is sort of get into sort of a the new technology and and I’m wondering you know how is China and India going to do on AI over the next 12 to 24 months like what should we watch out for from Copenhagen Atomics while we’re still Copenhagen Atomics? What what what are the main things that are coming up soon? Yeah. So, so definitely we, as I said, we we’re right now closing an investment round of $100 million and that will allow us to set up our tonscale manufacturing of lithium. So, that’s an important cornerstone. Uh, right now there’s no no company anywhere in the western hemisphere that are making highly enriched uh lithium at tonscale. They can make it like small grams or maybe half a kilogram, but nobody can make it in tons. So, we will be able to do that within the next 24 months. Uh, we will also set up the fuel production. So the the molten salt production of uranium and thorium salts I mean we already have that running in tons scale for um fleac. So it’s not very different to change over to uranium. The only thing is we need we need some uh approvals to do that here and if we cannot do it here we can do it somewhere else because the technology is not the problem. It’s a it’s a you know approval and licensing problem and then we will build the the next version or finish building the next version of our uh test reactor the one with the cocoon. So actually the cocoon is like a giant uh box of steel uh encapsulating the reactors to limit the amount of radiation coming out but also protect the reactor from natural disasters and airplane crashes and whatnot. So so it’s an important part of the reactor um design and we will build one of those here in the next 24 months and test that as well. So you will see that we get very close to having the the full supply chain for uranium and thorium and lithium. Uh and then we will also try to find a supply of heavy water. That’s the next actually an issue. Uh right now there’s not enough supply of heavy water in the world. But we will work on that and um and let’s see if we can solve that in the next 24 u months as well. I will be really happy. So one of my primary focus is on the reactor design and documentation for the licensing work. So even though this is just 1 megawatt and one month, there’s still a lot of documentation that has to be done in order to get a reactor or an experiment like this approved and we’re trying to do it in a way where it doesn’t uh end up in a situation where the paperwork weighs more than the reactor. Um so that’s one of my primary focus is just that we get to that point uh in a sort of reasonable amount of documentation. Copenhagen Atomics is planning to build and ship a 1 megawatt test reactor to the Paul Sher Institute in Switzerland. There they’ll get a first look at how the reactor behaves under real operating conditions with real neutrons flying around. After a decade of work, it’ll be the first criticality test of their design. Who is leading the licensing efforts? Uh because Switzerland is acting as the regulator. Is PSI leading the effort or you guys or how is that working? So we have a collaboration agreement with them where they’re the LE. So they will be applying for the actual license and we’re supporting them with documentation and then we’re uh manufacturing the actual equipment that will be running in the test reactor. Okay, cool. Anything else I should ask you about that we didn’t talk about? Um yeah, I think it’s important to note that the reason we’re going through all this effort with developing a molten salt reactor and a thorium molten salt reactor that has the potential to become a breeder reactor is because we see this as a vital technology in order to be able to scale nuclear to meet global energy demand. And if it wasn’t because you could make this much cheaper or it could scale. There’s other technologies that we know work. But this technology basically needs to be developed because it has a much bigger potential to scale nuclear than many of the other alternatives. When I left after my visit, they gave me this, a metal ball a bit smaller than a golf ball. This represents the entire amount of thorium I’d consume for all the energy I’d ever use in my lifetime. Not just for electricity, but transportation, food, heating, and everything else, all in the palm of my hand. If instead I had to power my entire life using only natural gas, I would need about 21 million cubic feet of it, and it would cost nearly half a million, which is not sustainable. Copenhagen Atomic still has a way to go to prove that this vision can work at a commercial scale. There are real technical, regulatory, and operational challenges ahead. But after spending time with the founders and the team and seeing what they’re actually building, it’s clear to me that this is not just another reactor concept living in PowerPoint or 3D renderings. They’re testing real hardware and building up the kind of practical experience that very few people in this space will ever get. The technology is promising and whether they ultimately succeed or not, Copenhagen Atomics is already doing something that matters. Getting us closer to abundant, cheap, clean energy with thorium molten salt reactors.