 Okay, so hello everyone. Thank you all for joining us today in another sessions of fusion breakthroughs for those who missed the previous sessions. I'll put the links on the chat and on the comments box, depending on whether you're watching us live or on YouTube. But before we begin today's session, let me congratulate congratulate the team at the USDA Lawrence Livermore National Lab for the medical achievement announced yesterday, scientific energy gain in a fusion experiment that was extraordinary. So now, many thanks to our speakers today for their availability. Dr. Katsumi, executive director of the project, Dr. Siwuyon director of case star research center and Dr. David Gates chiefs, technology officer at Princeton accelerators incorporated. I'm Matteo Barbarino from the International Atomic Energy Agency. Today's sessions features recent results achieved at the large helical device or LHD in Japan at case star in South Korea and studies on simple dipole arrays to simplify the design and production of accelerators. So recently at the National Institute for Fusion Science in Japan, physics experiments on plasma turbulence and instability have provided important insights for developing control methods in the LHD. At the Korea Institute of Fusion Energy, several studies published in 2022 reported great results. And some experiments at case star were able to produce a plasma fusion regime that satisfies power plants performance requirements, including high temperature above 100 million Kelvin and sufficient control of instabilities to ensure steady state operation on the order of tens of seconds. In addition, recent recent progress has been made in designing permanent magnets as an approach for achieving desired magnetic fields with simpler coil geometry in accelerators. We're about to hear about these three topics. The format will be seconds of three talks, 30 minutes each. Tap your questions and comments into the chat box and we're going to go through your questions during the 30 minutes Q&A at the end. So now without further ado, please welcome Dr. Katsumida, Executive Director of the LHD project at the National Institute for Fusion Science in Japan. Thank you very much. So let me start the presentation. Can you see my slides? Not yet. Not yet. Sorry. It's coming. Yes. Thank you. Okay. Thank you very much for the introduction. Today I will present the recent LHD experiment. And the title of my talk is academic research on the turbulence and black instability of isotope mixture plasma in large helical device. So this biograph summarized the recent change of the strategy of LHD research. And from last year, we are focusing the experiment on the basic plasma physics issue for solving the future program of burning plasma, as well as the current issue in toroidal plasma. For example, the current issue of toroidal plasma is one of the important issue is the turbulent transport study and also the transport improvement. And for this issue, we are encouraging the international collaboration. And today I will give a few example of a very, very good international collaboration result. For the future issue of the burning plasma, today I will pick up three issues. The one of the issue is the reduction of the heat load. The question is how we reduce the adiabata heat load because the higher heat load of the diabata plate would be the problem because we cannot frequently replace the adiabata plate. And the other issue is for alpha channeling or how we can heat the ion with alpha particles. So that is the second issue. And the third issue is how we can exhaust the helium ash. So the basic plasma physics issues for this future program is one is the turbulent spreading and that found the turbulent spreading is a useful tool to reduce the adiabata heat load. And for the ion heating, recently we found a wave particle interaction through an inverse and a dumping at an event. So we can evaluate how much is the energy transfer from the energetic particle to bulk ion without collision. And some of the topics is the ion mixing in the isotope mixture plasma because we need an effective efficient mixing of process to exhaust the helium, which is produced by the fusion reaction. So this is a highlight of today's presentation for the turbulence transport. I will introduce the LHD and the Vendor Science 7X comparison experiment and bottle powder injection and isotope effect. The turbulent spreading, I will show the impact to the adiabata heat load and also the first turbulence are spreading. And third topics is wave particle interaction, especially I would like to emphasize mass dependence of collisionless energy transfer. Then the last topic is the ion mixing. And right graph show the fraction of international collaboration and domestic collaboration. And this is how many proponent for LHD experiment last year. And one third of proponent is international collaborators. And in fact, the proposal from the collaborator exceed 50% of the total. Okay, so a transport characteristic. So the, we study the impact of magnetic field configuration on the turbulence with the collaboration that IPP. And so this is the result of the comparison LHD and the Vendor Science 7X. And compare with the neoclassical transport, Vendor Science is strongly suppressed of neoclassical. So the Vendor Science is a low neoclassical and compared with the Vendor Science, LHD has a higher neoclassical transport. And what is interesting is that we see a difference over the turbulence transport. For example, and this is the calculation, the right figure show the calculation and how much is the heat flux due to the turbulence. And the Vendor Science ITG is higher than the LHD ITG. So in other words, in terms of the ITG that LHD is, ITG is suppressed. But on the other hand, if we compare the ITG, the LHD heat flux contributed by ITG is, is larger than the Vendor Science 7X. So that even the steriliter configuration, depending on the detail of the magnetic configuration, the characteristic of your turbulence is quite different. And this has been published by Walmart, Plex, Walmart, and PRL last year. And we also have studied the mass dependence of electron ITB. And the right figure shows the electron temperature profile with ECH for deuterium is a top and for hydrogen is the bottom. And we found the difference of the threshold power, threshold power for ITB. In other words, in the deuterium plasma, threshold power is lower. So if we compare the marginal or marginal power, we see a significant difference between the deuterium and the hydrogen. And we also have a threshold power experiment for the mixed plasma. The other interesting result is we have a, we provide the off-axis ECH and produce the hollow temperature profile. And that is a bit unique in steriliter or LHD. And in many cases, Tokamak showed the so-called the heat pinch. And in other words, even the off-axis heating, electron temperature is sometimes heat. But in LHD, we see a hollow profile and also we do the modulation experiment and try to get the temperature gradient and heat flux, so-called flux gradient relationship. And what we found is the clear hysteresis of gradient, flux gradient relation. And also at the zero temperature gradient, we see a finite heat flux. So that is the outworld heat flux is observed clearly in LHD experiment. And for the transport improvement. So this biograph show the volume powder experiment with a collaboration in the PPPL. And we drop the volume powder from the top of the machine. And then we see a significant increase of electron temperature and ion temperature. So this biograph show the comparison with reference. Reference means no volume powder. Both the electron and ion temperature is increased. And if we look the fluctuation, the characteristic more in detail. And by applying the volume powder, the fluctuation, the around 20 or 30 kHz range, just be a low frequency range. That frequency goes down. And so this is also the good collaboration experiment with PPPL. And we also study the so-called the core edge topping. And this is the experiment with diverter pumping. And so one can expect the diverter pumping and reduce the neutral and edge neutral. And we compare the cost of effective thermal diffstivity. And what's happening is we see almost no difference at edge thermal diffstivity or no impact of the diverter pump. But we see a core improvement with diverter pump. So this is the clear evidence of edge core transport coupling. In other words, when we reduce the recycling, then we see improvement in the quality rather than the edge region. Okay, then the next issue is the reduction of the peak heat load. And this is experiment with magnetic island self-regulated oscillation. So the top picture showed the time evolution of magnetic island size. So magnetic islands become big and small and show the oscillation. And also the interaction with bootstrap current, this oscillation is self-regulated. In other words, in terms of the operation that is a steady state, but magnetic island itself has an oscillation. And at the same time, we see a clear reduction of heat load at the diverter plate. So the right figure shows the relation between the diverter peak heat load as a function of island expansion rate. So when the island becomes big, then the diverter peak heat load is reduced. So this is the experiment, the magnetic island contributes to reduction of heat load through the detachment. The other experiment is the impact of a terminal spreading on the heat load. So in this experiment, the top figure shows the turbulence amplitude control. This is time versus space. And 0.48, that is the elastic loss flux surface. So when the magnetic fluctuation appears, then we see turbulence spreading from core region to the outer region in the stochastic layer. And so when the turbulence spreading occurs, then the edge density fluctuation increases. But associated with the increase of edge density fluctuation, we see a reduction of diverter peak heat load like this. And this is also the company with the broadening of the heat load. So the bottom figure shows the full radius of a half maximum of the diverter heat load. So as fluctuation increase, we see a broadening of the heat load at the diverter plate. So this is also another useful tool or mechanism to reduce the peak heat load on the diverter plate. And also the turbulence spreading, what we found is a very first turbulence spreading or radial propagation of the turbulence. And usually, the so-called the avalanche model predicts the simultaneous radial propagation of the heat pulse and turbulence. In other words, the speed of the heat pulse is the same of the propagation speed of the turbulence pulse. But in this experiment, we found the turbulence pulse propagated six times more than the heat pulse. In other words, the heat pulse propagate is more or less what we expect from avalanche model. Better turbulence propagation is much faster. And this is also the quite interesting phenomena. And one of the possibility of a first propagation would be a turbulence, turbulence, IK and low K turbulence coupling. Okay, then I would like to move on to the next one. In the disciplinary research and here I would like to show you the direct observation of rendezvous and transient time dumping and also the collision energy transfer. So the rendezvous dumping is a well-known mechanism. So the particle which has the slightly higher energy of the resonance with the energy and the particle with lower phase velocity has gained the energy. And at the end, the net energy flow is from wave to the particle. That is so called the rendezvous dumping. And right figure show the cartoon of ion velocity distribution and the resonance phase velocity. The resonance phase velocity is brought in green and due to the random dumping and the population of higher velocity increase and the population of lower velocity decrease. So we take the difference between the Maxwellian distribution we see a so called the bipolar signature. And that is expected in the rendezvous dumping. So in fact, we observe this bipolar structure. Direct figure show the time evolution of this discharge. And we see a drop of a neutron associated with the image to the past with this. And what we found is increase of second moment of. I am not. Velocity distribution. The second moment is the course month to the energy. Okay, and if we look the I am velocity distribution more in detail. And that deformation appears only a parallel to magnetic field, not perpendicular to magnetic field. And also the localize at the 0.79. We are the way we is excited. And if we take the difference from before and after the burst, we see a very, very clear the bipolar signature. And then we can evaluate how much is the original energy transfer. And from deduction of neutral neutral, we can calculate how much of energetic particle lose the energy or transfer to the MHD wave. Then they're from the carbon impurity and the bulk impute bulk iron. And since we measure the iron velocity distribution by integrating the population and the along the velocity, we can evaluate how much is the energy. So, the top figure and second figure is how the iron velocity distribution change associated with burst. So, at the, in this case, is a 200 kilo. Meta per second, and we see a deformation of iron velocity distribution at the same as velocity. And we can evaluate how much is a game game is an increase of the energy. So, the carbon and the energy game is somewhat 20%. Well, that is like a 5% energy gain. So, we see a significant difference or mass dependence of energy transfer from the wave to the carbon and bulk iron. That is because the, in the case of the carbon is close to the summer velocity. However, in the bulk iron, some other velocity is much higher than the resonance velocity. So, depending on the slope and efficiency is higher in the case of the carbon impurity. And that this result suggests the, the collision less energy transfer from wave to iron. For, for instance, the alpha channeling should have a somewhere I stop and defend this. In the world, the heavy iron has a higher efficiency for collision less energy transfer. Okay, last topic is that I am mixing. And this is an experiment of I stop by mixing it in LHD. And I showed you a tour profile of a duty and the hydrogen. And most cases, if we compare the electron density profile that they're almost the same almost identical. In that non mixing case, we have a peak heading density and follow a duty and density. And pick the helium density is due to the helium beam feeling and follow a dense duty and is due to recycling from the wall. But when the I stop by mixing occurs, regardless the source and the hydrogen and duty and profile become same. And that transition is related to the. So, top it on state, whether that you TM dominant thing or IDG dominant. So, in the case of it is dominant we can expect the I stop by mixing. And when the TM is dominant, we expect the ISO top are not mixing. And we also see a transition. This is a time evolution of electron density and electron temperature and in this discharge. This is a close to the density limit. And the plasma is a collapse like this, the temperature significantly decreases like this. And in this discharge we see a clear the transition from I stop, no mixing to mixing. This is hydrogen fraction. So you see a hydrogen fraction is higher in the quality and the lower in the near the age, but at the end. The ratio is almost a uniform and right figure show the electron temperature and I on temperature profile and. Um, fraction of a hydrogen. So, in the non mixing state. The hydrogen is a peak. And the duty of his horror. However, when, when the I stop by mixing occurs, the ratio is almost a constant in space. So, uh, this is the evidence of, uh, I stop a mixing occurs. And the. With the density increases or towards the density limit. Okay, so this is the conclusion. So a significant progress in the scientific research has been made in the experiment. On turbulence transport and Ireland. And this is particle. And I really experiment to provide the following important. With a vision. A role of turbulence, a splitting. And core edge diverter coupling. And non diffusive transport. And the interaction between the magnetic island and transport and the land. And trying to time dumping. So. However, our knowledge of these issues is limited. So, um, therefore further study using a sophisticated diagnostic is necessary. For a comprehensive understanding for the observation of. Thank you very much for your attention. Thank you very much. We have a question already on the chat, but we'll take it after the other talks. So this is a list of the recent, uh, please release. Just I leave the advertisement. Very helpful. So now we go. We welcome Dr. Sibuyon director of K-star research center at the Korean Institute of Fusion Energy. But can you see my screen? Yes, we see, we see as we see ourself on the web. I don't know why it works previously, but not now. Do you want to share the PowerPoint? You see the slide. Yes, we can see the slide. Okay, then let me, let me start. Okay, so today I would like to introduce listen visual to K-star. I mean, I see what could you, can you go full screen? Yeah, it is full screen right now, but no. No, not for us. Hmm. How about this? We don't see full screen. We see basically see the PowerPoint, but without going full screen. I don't know if you can, can you, there's a, if you go full screen, what happens? Yeah, I did the full screen, but because I share, I'm sharing the application window. Okay. Okay. I mean, I think it's, it's, but if you, if you click on the bottom, on the bottom right, the full screen icon on the bottom right next to the magnifier, the plus or minus, there's. Yes. Yes. Okay. Nothing happens. Okay. I don't know why, I mean, it just worked the previously and it doesn't right now. Maybe because previously you were sharing the screen and now you're sharing the application. I'm trying to share in the screen, but trying to show the screen. Yes. Try to go to the PowerPoint. Okay. So, is it working right now? Yes. Yes, great. Okay, so then let me start with. Thank you for your presentation for the recent case. Okay, so let me talk shortly about what case is aiming at. So the mission is quite clear that the machine like K-star is to explore the science and technology for the high performance state-state operation, which is most important. Of the future demo. And so we are going to do, do some steady state hybrid operation development and also the instability control and also try to understand some fundamental processes in this hybrid discharges and also touch up touching upon the engineering and technological issues. Yes, K-star is a mid-sized superconducting tokamak. So, and also beyond, we have, I should say we have their unique capabilities for physics and engineering research for K-star. The most of them, most of the one is that we have a better plasma symmetry, which includes the low-lipple and also low airfield, and also better controllability with investor control coils is quite flexible and reliable system. And also we have a fairly good imaging diagnostic to look at the dynamics and the structures in some details. And also we have a fairly good MBI system, which is fairly tangential and it's very efficient for the current drive and also the rotation drive. So based on this capability, K-star has been doing up to now several milestones and 2021 we got the high temperature plasma sustainment up to 30 seconds. So in this talk, so I would like to address several topics starting with the scenario development, which will include the high internal inductance scenario. And also the internal transport barrier with eye mode edge is so-called fire, a fast time regulated mode. And second topic is MHT instability. So, I'm going to explain the control of the edge localized mode with the Legend magnetic perturbation, which is, I mean, very important technique for ether and future reactor and also the disruption mitigation with multiple shattered pellet injection. And suddenly about the fundamental projects. So, I shortly addressed the self organized structure of the electron temperature so called the profile correlation. And also the turbulence spreading near the magnetic island. And also the key mechanism for the density pump up by the Legend magnetic perturbation I'll shortly talk about the virtual platform at the end. So starting with the study scenario development is high AI scenario has a pitch and force. It has a fairly picked current profile and due to the strong magnetic shear that we have a better compartment than the conventional or shear plasma. And we actually used optimized ECCD and NBI mixtures so we can get the fairly high level of it better normalized with allies around one. And then this scenario provide a good candidate for the steady state. Hi beta scenario. And in terms of fusion gain, it's fairly high and Q 95 around four. So it's quite attractive scenario for for. It's a steady state and also the future demo machine. And, and actually the longest capability of this mode has been, I mean, explored and with the beta normalize 2.8 we can be successfully sustained this mode more than 20 seconds. And it is really, really, I mean, long compared with the relaxation time and I sustained steadily above one. So this is a guarantee of this mode can be, I mean, extrapolated to the more steady state and longer person with a high beta at the same time. So when you look at this simulation, I'm a predictive modeling based on the on the radio transport. And we, as you compare to hear that the method profile and the simulated profile looks very similar, except for the some deviation at the center for the for the rotation. But the others are used fairly in good shape. So, I think we understand that the basic transport by this model. And this is, I have to emphasize that this one is based on the strong magnetic shear part of the turbulent suppression. And next scenario is internal transport barrier, which is quite straight station only at high ion temperature. And it actually got the, the ITV mode with the devoted at starting and also it can be also extended to the eye model like pedestal. So, so it also give us additional gain in the performance. And basically the performance itself is quite similar to the depth of the H mode. And without M's, of course, it's a mod edge and the loop voltage is quite low. And it also has a good prospect for the for the future reactor mode. The one of the issue here is we, as you can see it's around the three seconds here, there is a collapse of the ITV due to the MHD, but we can recover it quite with the control of the of the profile with additional hitting. So, this can, this kind of MHD mode, we can, we have a knobs to control. So it can be a translated to the future reactor mode. So even though that right now the density is a little bit low. So the next job we have to do is the increase density still sustaining the internal transport barrier. So the, this is the ITV for the ion temperature only we don't see the clear ITV in electron temperature. And we are we use the upper singular, which means the unfavorable grad B drift direction, which will limit the LH transition with high power operation. So when you look at the put of the ITV, then it's really related with the past time populations. So we think that the, the, the role of the past time in this mode is quite clear. So when you do the data, linear data accumulation to understand the basic mechanism, there is a effect of the past time clearly, and also the electromagnetic stabilizing stabilization factors also present. The effect of the shopping shift and also the delusion of delusion of the summer lion. So all of them actually works together to decrease the linear growth rate. Quite a lot that we use the basic mechanism of the ITV formation we believe. And when you do that see gyro nonlinear simulation also predict the reduction of the other hip flux quite significant based on similar mechanism. So this mode can be could be applied to the future reactor mode. Without any severe mhd event in the edge and can be sustained in long purse with the high iron temperature at the core. So the next topic is the mhd control. So K star is working on very hard on the, the MPM suppression. So we succeeded the low end MPM control. And as you can see here, we can go down to the Q 95 is 3.5 and even up to data similar shape. And also the, we have a fairly high beta level to sustain with RMP. And one of the important characteristic of case to RMP separation is we are. It's the 3D field spectrum. So, so what we call is edge localized RMP is the key to for the M suppression and still we got the improved the compartment. So this kind of a configuration is, is unique for case style because it is possible only this only possible with, with the several local loads of the RMP call otherwise this optimization is very difficult. So using the three rows of the RMP call employed our course, we are optimizing the RMP amplitude in at the edge, the component of the RMP amplitude at the edge, and still reducing the core component significantly, then we are reducing the effect of the of the core degradation from RMP, but still we got successful M suppression at the edge. So, basically, this is also predicted by the linear plasma response models, as you can see here the edge. And when you change that the RMP configuration, we can decrease the component in the core significantly. Then we can, if you use like lead like configuration, still at the edge we have enough RMP component, then this is ideal spectrum, we can have for the M suppression. So this is so called edge localized mode. So it has a lower impact on the pastime transport. And so this has a reduced impact on the core degradation. So based on this technique and also the implying this some, some real time adaptive M control technique. All together, we can go to the higher beta normalized operation with the M suppression and also we can get a long pulse M suppression up to 30 or 40 seconds in the figure. So the control logic developed a case star using the ERMP together with adaptive control provide a good tool for the long pulse M suppression is required for the future reactors. So, one of the important implication of this low end ERMP is it provide a pass to the express a core configuration, which will, which would survive in react environment so people understand that the investor control code is kind of nightmare for the react environment. So that's why I want to move it outside. However, without this kind of technique, the data, there is a strong I mean legend component in the core also so it's not so ideal. But when we apply this kind of ERMP configuration successfully, then there is chance to use this express a core configuration. With the successful suppression of Agile climbers. So these are the older collaboration work for in this regard to searching for the, for the possibility of this express a core configuration. So we do have multiple pellet injectors and we realize that the very synchronized within the pre-summer pinch time, then they are synergistic effect. So, it is very important to shot multiple pellet within a certain time range and can be assimilated very similarly with the high temperature plasma. So then we will have a strong impact on the, on the current, the density lies and also the, the colloidal and colloidal symmetry. So this is a fairly good news for the operation because it is using multiple pellet at the same time so the proper I mean tuning of the timing provide the predicted performance of the adoption mitigation system. And for the fundamental processes. So I'd like to introduce the very old problem so called self organization in the electron temperature in the emerald plasma. So it leads to the major scale structure. With the electron temperature congregation, as you can see here, so we don't, we, there is a prediction of this phenomenon by the gyro kinetic simulation, but finally case test succeeded in, in the measurement of this structure. As you can see here in the to the imaging diagnostic, you can see the major scale structure here, profile congregation. And we found that this step size is following the typical pressure like distribution. So we are quite clear, we are quite sure that what you may measure is the, is the electron temperature collusion which is predicted by the simulation. So, yeah. So this is kind of a good experimental validation of the ongoing assimilation with a good agreement. So for the table is spreading prophecy already mentioned that this physics and we found that the, this table is spreading around the magnetic island, it's actually the rapid heat transport across the magnetic island and promote reconstruct the connection inside and that could help the, the open, opening the magnetic island, and which can go to the disruption. So, when we look at the case test disruption data, there are certain level of the minor disruption so it also always related with the increase of the temperature fluctuation and it actually helps the minor disruption happens. So, we, we think that the disturbance spreading is important mechanism for the, the disruption event. Near the magnetic island. So, finally, this is the density pump out. We all know that the dense pump is quite mysterious. I mean, phenomena, we understand the M suppression but dense pump out is another thing. And when you look at the case data data data multiple steps of the density pump out. Many people think that it is, it is only the legend and component important important for this, this phenomena, but we realize that this is not that simple. And when you include the, and put a non linear image called the Joe rack. So we include this image the legend component. Also the plasma component and both the, and also the no critical. No critical to the viscosity effect, which is, it's not relevant, but it has a strong impact on the rotation and the, and the particle transport. So, if you include these two successfully. Then we can reproduce what we measure from the experiment in good agreement as you can see here. So, finally, we think that it is related with the different dense pump out correspond to a different Q 95 surface and in sequence. So it is. We think that it's, it's a good agreement with what we measure at k-star right now. And finally, I will briefly explained what we are going to do in this virtual platform development. So we actually develop the virtual case. So cold. There is a lot of application for this model. It's coming from the very simple simulation, but also about to the possible to the first principle simulation all together. We have a real time monitoring system for the machine operation included already. And you will use this platform for the fusion flight simulator. And also the inclusion of the synthetic synthetic diagnostics to all of the, the physics and engineering part will be included integrated in this platform to understand better for the what's going on in this complex system. So, this is the summary. So, as I mentioned before, we have you found several can good candidate for the, for the scenario, and can be I mean extrapolated to the reactor relevant condition. And also you found the novel control technique. So cold ER MP and multiple pellet shuttle pellet injector, which is urgent issue for the eater or baseline operation. And we are going, we are providing the key physics. I understand the fundamental, fundamental projects. I mean, the self organization in the electron temperature and the turbulence splitting. And also the density pump out. It is so there is a measurement and the simulation all together to, you know, in a good shape for each topics. I believe. Of course, we need more. I mean, this analysis to understand further, but for the moment, there is a really good agreement right now. And we have a virtual platform development. Okay, so this is the all I have and stay tuned with the case. And we have more heating systems are coming and the new. So tungsten. One of the cassette, which is, is the eater class. So, and also the new current drive systems called Helicon current drive. And then we are going, we are going to move on to the even higher beta with a longer person scenario development with the MHD control together. So thank you for your attention. Thank you for sharing this results with us. So now we welcome our final speaker, Dr. David Gates, Chief Technology Officer at Princeton, the letters incorporated. Okay. Thank you, Mattel. Can you hear me okay. Yes. Good I switched to a different microphone so I just want to make sure it's still working. So I'm going to share my screen. Someone needs to stop sharing first. There you go. Okay, can you see my screen. Yes, yes, not full screen. Thank you. Okay, so today I'm going to talk about using arrays of simple dipoles to simplify still readers. For those of you who know me. I am now on leave it from Princeton plasma physics lab as of this month. So we have a new company called Princeton sellers incorporated, which has been in operation for a few months now. So I'm going to tell you about our, our goals and our plans and sort of how we got to this point. So just to remind people what, what still readers have that make them advantageous. In particular, they don't have current right so we don't have to drive current and that's really important if you're looking to build reactors current drive and reactors is expensive. It requires recirculating power. Any currents from the current drive can induce heating into important components that need to be cooled. So for for efficiency still readers are a very good idea. And if you look at these two pictures, I think, you know, a lot of people have identified the magnets as a big issue for still readers, they're they're not simple, or haven't been to date. And I think that's an issue that we've been trying to address now for some time looking at alternative ways of constructing still readers. Just to remind people how still readers have been designed in recent times. There's been a great deal of progress in optimizing stellar equilibrium to have good neoclassical confinement. That's an idea that goes back to the 80s. And then a method to build the coils that give you those configurations was developed in the late 1980s by Peter Merkel. So it's a straightforward procedure, but it's, it leads to coils that that haven't always had the best control over their shape. So you end up defining a shape that's based on the equilibrium we're trying to create. So in the processes that you define your optimized equilibrium, which is done through some physics optimization code. There are several of them around the world that do this. You know, you can put in various physics criteria for what you're trying to optimize, particularly can usually people optimize for neoclassical confinement, which has been the historical Achilles heel of accelerators, and that's been shown to work. I mean that process does work. W7 X has shown improved neoclassical confinement for for ions and unambiguously. That was a very major achievement in recent times. It can also be optimized for MHD stability and various other fast particle confinement various other parameters that might be chosen to be optimized it can also design diverters. So to make the equilibrium that you've now defined through physics parameters. The process is that you create usually what's called a winding surface there are various methods but this, this is the first method and it's the basic method that's still used for most elevator designs. So you create what's called the winding service where the coils are going to sit. And on that wanting surface you use the virtual casing theorem to create fields that set the normal component of the field to be zero on the last closed flux service. This is a requirement of course you know normal fields don't exist on flux surfaces if they're actually flux surfaces. And there's usually at least one other constraint requirement is that you match the turtle flux or or a similar constraint that gives you the right value of the so you don't get the trivial solution of be equals zero that's not a very nice solution. And then, you know, using the virtual casing theorem you come up to the current potential on that surface that you've found. And this is actually the figures from Merkel's paper. The top figure there is the potential current potential. That's for one period. And then those current potential lines are the starting point for a more complex optimization that involves, you know, coils with finite thickness and extent and find current density. And but when you end up as what looks at what's on the bottom there which is this is a early W seven X like configuration from the 1980s. There are further optimizationizations of the coils here that that take into account the realistic requirements for building things that can't the coils can't go through each other. They have to be. There's a radius of curvature constraint. You can't bend conductors at arbitrary angles. But you end up with coils that look that looked like the things on the bottom right. Recently, a pair Helander visited PPPL and was talking to Mike Sarnsdorf about other ways to build stellar errors. And this actually is a good story. So I'm going to tell it Mike was building a high school science project with his, his son, and he wanted to son wanted to build a real gun. And they decided for the high school science project, it wasn't really plausible to use to use, you know, electromagnets because it was going to be too expensive, but permanent magnets are quite inexpensive. And so, and they actually make amazingly high fields in modern permanent magnets. So he had the idea of making shaping fields with permanent magnets. So, and pair and Steve Cowley and Michael Drevlock at IPP got very excited about this idea and wrote a PRL showing that this is actually plausible. And in the meantime, so this is this is a figure from that. That paper, the method is actually essentially identical. The difference is that permanent magnets can't make to roto fields. You can show that it's a one line proof that, you know, amperes law says it. If you want a field that goes around the long way around the tourists, there has to be a current up through the middle. But you subtract off a total component of the field. And you end up with contours that instead of looking like these closed wiggly contours are actually closed contours in the plane. They don't encircle the the tourists any longer in the political direction, and you get these helical stripes of of required additional field. And the numbers here are plausible. You can actually fairly easily show that that you can generate the fields you need using just definitions of magnetism. I won't go through this in any detail. I point you to Paris paper. It's a very, it's a very nice proof. But you can show that surface currents and magnetic dipoles can give you these solutions. Just an example of what that looks like this is the first calculation you do you look at you define the winding surface where the dipoles will lie. It's not really a winding surface anymore with permanent magnets. It's more of a mounting surface. And you can work out the current potentially you need with this is actually we did for a hypothesized experiment at PPL where we used the toroidal field coils that were originally built for the NCS X project. And mounted permanent imagine mounting permanent magnets on the vacuum vessel surface. This is the zero thickness version of that. The code was developed. I should acknowledge. Who did most of the work. I'm going to show you in the next couple slides. And, you know, the color here is the direction of the magnetic dipole is read in and blew out or get the other way around. I'm not. Totally sure but And you can see that that, you know, they make nice helical stripes of. Magnetic moment or in this case current potential, which are directly relatable to each other. Really quickly, you know, I mean, there's been a series of papers by Chao Zhang and Ken Hammond. This is a figure from one of Chao Zhang's papers looking at using normal directed. Magnetic moments. This is still not quite realistic. It's these are perfectly stacked. There's no support structure. There's, but there's also a series of papers that include, you know, leaving gaps for for steel to hold things and glue. But you can actually find many, many solutions to this problem. In fact, I think you can prove that it is an infinite number of solutions to this problem. But it seemed like we should probably find a realistic solution. So we made a proposal. The goal was demonstrated a permanent magnet design that was constructable with sufficient field accuracy to make the target equilibrium. There's a series of additional papers that will be coming out on this looking at errors and material. Real material properties and and it turns out when you buy a magnet, the magnetic moments, not perfectly aligned. It's off by some manufacturing error and we included all those variations with stochastic projections. But we named the project PM for style. And we had a goal of building a half period magnet block. We have completed the final design. PPL also was actually interested in looking at what it would take to build such a device, which was run as a separate project. We're using some components of the vacuum vessel on the trail field coils. And so it was based around the vacuum vessel for NCSX was completed. So we base the design around that. And the TF coil set had a field on access of a half a Tesla. So this was a design with realistic parameters. We had real steel support structures. We had an assembly method that that project was obviated by the realization that we can extend to this idea to superconducting dipole arrays. So rather than using just permanent magnets, which have limited field generation capability. We can build, you know, dipole coils that are mounted on the same vessel surface, or maybe further back, but with effectively no limit on the field that can be generated relative to the Toronto field because you're going to use the same. Same material to build the dipole coils as you use to build a toroidal coils. And of course, there's much smaller requirement on these on the dipole calls and then the total field is, you know, dominant Lee, the main field. And so the idea here is that small tiles mounted on panels that are demountable. This actually creates both a control scheme and a maintenance maintenance scheme for still readers which has always been a bit of an issue. So with complex modular coils, it's hard to imagine large sector maintenance like in a tokamak, but by mounting these dipoles on panels, we should be able to get a large sector access. And so this patented concept led to a spin out company which I'm now have joined several other things have made this idea really attractive. There's been a lot of work on optimizing equilibria. In particular, work done by Matt Lenderman, Elizabeth Paul, Elizabeth is a hope that she's quite gotten there yet but she was a postdoc at Princeton and is now going to be a professor at Columbia, that's of course at University of Maryland. But there's been a lot of work and this is funded by the Simons Foundation, in large part through a project called the hidden symmetries activity. And the idea is to make better stellar equilibria. And this, this equilibria is from a paper by Matt and Elizabeth. And they achieve quasi symmetry under the order of 10 of minus six. So, you know, the theoretical existence of these equilibria is now established. The particular focus of this article is to demonstrate excellent fast particle confinement. So fast particles are controlled and confined sufficiently really for a reactor in a theoretical sense anyways. And, you know, I think, you know, this is the simpler and more accurate equilibria allow for imagining a stellar reactor that actually substantially easier to operate and sustain. And this, this should lead to lower cost for construction as well. So this idea, the thing on the right is a cartoon when I'm comparing it to things on the left, which are not cartoons, but I think you can get the idea here. You know, the coils on the left, those are of W7X, HSX is in the University of Wisconsin, NCSX, the coils are completed but never assembled. I don't really think you need to look very hard to see the complexity in the coil set. But also there's a real implications of that complexity because it's not just complexity. It's highly precise complexity, which means that those coils are very complex shapes, but they have to be accurate to a very high precision. This has been done. It's not impossible, but it's sometimes just because you can do something maybe it's not the best idea to do it that way. And I think the idea on the right, we've taken the complexity out of the shape of the coil and put it now into a 3D current distribution on a surface. So each of those individuals small dipoles has a different current through it. So I've taken the complexity out of the coils and put it into basically a control system. You know, in fairness, I think, you know, when these objects were designed, the idea of a control system that independently control the 500 dipole coils would have been pretty hard to imagine. But in a modern sense, controlling large numbers of identical objects is pretty straightforward. You know, a lot of this in the design and in the control scheme, really it comes down to improved computer power so that we can imagine doing such things. Also, you know, the existence of high current densities, superconductors will make this much easier to achieve as well. And of course, high temperature will need to lower power consumption. So our company's goal here is to build a first generation fusion system. It'll be completely prototypical of a power plant will do DD fusion only will be aiming at being target and the idea is to generate a neutron source and demonstrate the real advantages of accelerators this way. We can show that we have the fast particle confinement that's predicted theoretically now. We can use modern technology, negative ion neutral beams, which are have been in use for some time now but nobody's ever really looked at optimizing them as neutron generators so we're going to run up at the near the peak of the deuterium deuterium fusion cross section and make a DD neutron source. The idea here this device will have a blanket. It'll be a tritium generation blanket. So it will be also prototypical in that sense, we'll be able to generate, we want to use this and generate some electricity won't be a net generation of electricity but a complete prototype of a complete fusion power plant and take you know and really demonstrate that we can run for days on end with with each generation we will have a source for revenue we can sell the rest of the fusion program tritium course we'd like to sell it to ourselves for our own electrical power plant. But also this device could be used for making medical isotopes. So this is our near term commercialization, we're going to run it fields that have been used in the past, high higher fields and accelerators have ever achieved. We're going to aim it, you know, six Tesla. This will allow us to take advantage of commercial development of either like gyro trons. We cannot order those I mean you can just purchase them for there's another company that's been developed that will sell you a 170 gigahertz gyro tron and run those also in steady state. So the idea here is to have a near term generation of revenue. You know, on a time scale that's interesting to to venture capitalists that really do care about you having a product. I think a quick shout out to my co founders. We're running this as a business. I am not the CEO. I have no training as a CEO I'm going to be in charge of the technology development. I have actual real business people as my partners. I really want to met Miller who was a co founder of the stellar energy foundation helped us get going. He stepped into the president and chairman of the board role. And I have a very capable and energetic partner in Brian Bersin as the CEO. And I finished a little bit early, but I think that's probably okay. People have been listening to an hour and a half of talks already. So thanks for your attention. And thank you, Dave. Also for saving and saving some time for questions. We have some questions coming. I'm going to stop sharing real quick. Okay, so thank you all again. So I invite the attendees to put the questions on the chat as they are already and I'm going to go through them. Mostly as they as they get typed. I will go back to either son. There was a question for LHD. They're asking if radio frequency eating has ever been used to control instabilities. He doesn't. Yes, sorry. What is a Christian deal. You think that I cannot hear. Can you hear my voice? Yes. Did you see the question on the RF eating? The question is I see. What. Not specified. So how can I see. I think by chat or. I. Is a way to just click the chat, then you can see the, see the questions. Yes, I did. I click the chart, but. Oh, okay. So while you go through the chat, you will find there's the question about it's, it's basically the first question you'll see at 326. The first question is. I have a theory that a certain type of electromagnetic signal. Can be used in. RF eating to control the turbulence in the plasma has radio frequency heating ever been used to control instabilities. In LHD. Okay, okay. As RF eating has been used to control the instability. Which instability talking about it's not quite clear. We can move ahead for now. Dave, I'll come to you. Do you they were asking if you know exactly what medical isotope. You guys are going to focus on for production. So, just to be clear, we're not. We're going to provide a neutron source. We've had some discussions with companies that have expertise in generating malibdom 99 is in fact the most. It's the biggest market. But the process is actually fairly complex to generate malibdom 99 and then purified and it has to be done very quickly because it decays very fast. So, our, our, our model would be that we design and build the neutron source and, you know, offer space in on the machine to do this work. We would go to a company that has that expertise. It's, it's, I've, I know enough about it to know that I think I don't know how to do it. There's a, but there are companies out there that already do malibdom 99 generation, but we can produce neutrons. I think much more effectively and more, more cheaply than other methods. To be precise, I mean, and it's really all this new technology that makes that possible. We can get high electron temperature and a relatively small volume. We can inject fairly high energy do rounds and get a, we estimate a 10 to the 18th per second neutron generation, which is on the order of, you know, vision reactors. In terms of optimizing the device to generate DD neutrons really isn't something people focused on. I typically wanted to have high iron thermal iron temperatures, which isn't really a requirement for being target, which you really want is high electron temperature for being target. So, they were also asking if there's any, I mean, this is going to be a pure fusion systems. Is there any plan for. I don't, I mean, it's a possibility. It's always been kind of my point of view that that fusion is really the, it's my goal. I want to make fusion energy, not vision energy. You know, I wouldn't rule that out as something this machine can do if somebody else wants to do it, but it's not something I, first off, I don't have any expertise in, in generating. What did they want to make. Basically, they were asking if the you would. To make it out of a vision hybrid system. Yeah, I mean, I think, you know, one of the attractions of a pure fusion system that it doesn't have any, it has much reduced. Perliferation risk. And I mean, obviously this would undo that I think. So, yeah, I think that would. It's not my target. But in the, you didn't specify any timeline. I don't know if you can, can you tell us a little bit more. The target of building this neutron source, which is our prototype reactor in 5 years. There's nothing since we're not challenging any engineering boundaries. We're aiming at 6 desolate it's been done we're aiming at using commercially available gyatrons we want to. We actually want to downsize negative on neutral beams right now we want low power. A couple of megawatts of NNBI will go a long way. We're not, we think that that's all doable in a very short term. We have some technology demonstrations to do in the near term using. Presumably, HTS superconducting magnets that are coupled in strong ways. And then we can go ahead and just build this. Now there, there, the big issue here will be supply chain, I think. I mean, we'll have to, you know, but the, the HTS industry actually has very aggressive plans for increasing the availability of HTS tapes. And we've looked into this, we think it's plausible that timeline. Okay, thank you. There was a question about how would the permanent magnets behave under extreme conditions. So permanent magnets. And much like superconductors the permanent magnets would have to be behind shields and blankets so that they don't get neutron flux. So, because neither superconductors nor permanent magnets want to really see neutron flux. The neodymium permanent magnets are actually extremely robust permanent magnets are almost ideal. And it's, I was really impressed actually I did quite I learned quite a bit doing this project I had never really worked with permanent magnets before. The key thing is to, there's two issues and one is the total field you can make which is called the residual field. And then there's the what's called the coercivity which is what it takes to demagnetize magnet. And it turns out if you cool neodymium magnets to even liquid nitrogen temperatures, they have extremely high coercivity, you can use them in fields up to eight or nine Tesla. So, now that said, you know it's not easy to work with things that have that much inherent field on them. It's really forces are quite substantial. And assembly is not a small task. A lot of the appeal of moving to superconducting dipoles as opposed to permanent magnet dipoles is that, you know, really the limit on the field I can generate from a superconducting dipoles the same as the field I can generate using the turtle field coils. So that there's no, you know, mismatch in the total field you can generate. So there's our question. There's a quite long question you'll see in the chat if you if you had the time to read it I'll come back to you. And he does and now I think the person who had asked you the question brought more on the chat so if you could probably if you could possibly also read that and then I'll ask you again. I asked a question to see what because I was I was curious to know. Do you think how would you see a possible collaboration with the international partner partners on the simulation platform on the virtual simulator on the flat simulator on this project for integrating physics and technology how would you how do you think how would you picture a possible collaboration in this area with the international I mean, yeah, it's just but so every every country, I think it has its own plan to develop this virtual system. And I believe that the most important part is that the basic engine it's it's it's from the game industry I mean the graphical user interface and all the, all the things so we have many things in common in in game I mean business and also the because this platform actually integrated many things starting from the synthetic diagnostic so we have a certain level of development in USA for for this technique and also we need the very sophisticated first principle simulation toolkit also so maybe you can collaborate with with other institution for the very fundamental gyrokinetic simulate simulator and transport it to the to the virtual system and integration or the integration so there is plenty of a country collaboration point here so but up to now it's development is is is domestic basically so we are building a basic blocks right now, maybe we can add up the additional part from the the other country so this is, I mean, a very interesting topic to collaborate with with with other countries I think. Okay, thank you. So, it doesn't do you. Yes, I read the additional questions and. You usually the our heating, while the heating is the big energy source in the. You didn't divide it and that creating the so called the temperature gradient and that's temperature gradient drive a lot of the instability and turbulence. So, that is the main effect of a heating or RF wave. And we do not see a much effect of RF stabilization of the turbulence by RF in other words, RF is kind of the source of the turbulence. Rather than the control the instability. Of course, one of the idea is once we apply the RF RF, we created the so called the energetic particle and energetic particle has some contribution to stabilize the turbulence that is also the presented in the case. So, I think there is a somewhat indirect effect of stabilization or effect on the turbulence, but there is no direct turbulence control by RF wave. Thank you. Thank you. So, Dave now asking those questions, I'll read them just so that everyone can can hear the question. So the question was, will the resulting resulting field produced by the dipole calls give non-optimal particle engagement onto the first world diverse and will the dipole field lines intersect the plasma phasing components on the small cleansing angles that are currently used to decrease the heat flux density. So, yeah, it's a very, I mean, it's a good question. It's a very detailed calculation that needs to be done there. We haven't, I mean, we haven't done that calculation yet. We're designing the, you know, sort of the bulk field elements. You know, I think this is a question for essentially any fusion device, right? Can you protect the first wall from fast particles? And I have my ideas for how to deal with that problem. I think, you know, even Eater has difficulty with possible fast particle laws due to alphane waves or the elm control perturbations that, you know, come from the elm control coils. So, you know, I think this is an area where we the whole fusion community needs to really decide how to deal with fast particle laws near the wall. We will look at this problem, but it's not our first, our first design activity. Okay. Thank you very much for answering. I don't see any other question, I think on the chat. So I think we can close it here for today. Thank you all again for your time and for the great presentations and thank you all for participants for staying with us. Thank you for the invitation. Thank you. Thank you very much. Yeah, thank you very much. Bye bye.