Researcher at Max Planck Institute for Plasma Physics, Garching
Following the dream of fusion power, which would provide humanity with the same abundant energy as produced by the sun, scientific research is focusing on creating and controlling high- temperature, burning plasma, as it exists in stars, using deuterium, a type of hydrogen abundant in the oceans. Significant progress toward this goal has been made in the area of tokamak research that aims to confine the plasma through magnetic fields and provide the basis for ITER, the next-step fusion device already presented at the 2009 Falling Walls. New technologies have been developed to protect the walls of fusion reactors against temperatures of 200 million degrees. Fusion scientist Rachael McDermott received her PhD in Plasma Physics and Fusion Technology from the Department of Nuclear Science and Engineering at the Massachusetts Institute of Technology. She is now a member of the ASDEX Upgrade team, a tokamak project of the Max-Planck-Institut für Plasmaphysik, and of the Helmholtz Association Nachwuchsgruppe, and will present, in 15 minutes, the most recent advances on the journey to the ultimate clean energy source.
Breaking the Wall to Control Fusion Power. How Tokamak Research is Paving the Way for Successful Fusion Energy Reactors
Good afternoon everyone. It is my pleasure to be here today to speak to all of you about my field, the field of fusion energy research and the progress that we have made towards making fusion energy a reality. Lets go to my title slide: in the last few years, the fusion community has developed some very innovative solutions to some of the most challenging problems science and engineering has ever had to face. It is these solutions, or at least some of these solutions, that I will be presenting to you today. But first, I would like to start by answering two questions that are key to appreciating this progress. First, what is fusion energy? Second, how do we actually propose to create it?
Fusion energy is the energy of the sun and the stars. It also has the potential to be a clean, competitive, and nearly unlimited source of energy on earth as well. The sun is composed mainly of hydrogen atoms, which due to the extreme temperatures, exists in a plasma state. This means that unlike in a gas, the electrons, the little blue guys, are stripped off of the atomic nuclei, and all the particles are free to move about entirely independently of one another. Now, in fusion plasma, the particles have a lot of energy. Sometimes, when two nuclei collide, they can actually fuse together and form a new, slightly heavier, element. This process, the combination of two light elements to form a slightly heavier element, is fusion. In addition to just forming a new element, it also releases a significant amount of energy per reaction.
This comes about, because in fusion processes, the mass of the final product is actually always lighter than the mass of all the things that came together to make it up. So, helium atoms are actually lighter than the neutrons and the protons that come together. The difference in those masses is released as a net energy gain, according to Einstein’s famous formula: E=mc2. Now, the most promising reaction for fusion on earth is the D+T reaction shown here, where the “D” stands for deuterium; this is heavy hydrogen. It is found in the earth’s oceans. It is comprised of one proton and one neutron, opposed to hydrogen, which just has the one proton. The “T” is for tritium; it has got two neutrons and one proton. When these two come together, they form a helium atom, a neutron and 17.6 MeV per reaction.
Now I know a lot of people don’t go around thinking about energy in terms of mega- electronvolts—that is just something that we do—but that is actually a lot of energy. If you compare the amount of energy you can get from two and a half grams of D-T; that is about one spoonful if you were to condense it into a liquid form. It is the equivalent of 28 tons of coal: factor of 11 million times more energy. Alternatively, you could compare it to oil, which is slightly more efficient than coal, but not really. You get the same amount of energy in one spoonful of D-T than 20 tons or 143 barrels of oil—the factor of 11 million times more efficient.
So, how do we actually propose to create fusion energy on earth? The sun is a giant hot burning ball of plasma. How do we create something that is millions of degrees on the earth? Well, first off, it is millions of degrees; so, you cannot let it touch anything. So, you need to find a way to confine plasma without letting it come into contact with anything. The way you do that is with a magnetic field. So, in a tokamak, you have a magnetic field, and since the particles and the plasma are charged, they are bound to those magnetic field lines and forced to process around them. But in a tokamak, the magnetic field lines aren’t straight like they are here; they are actually bent into a circle. This means that the particles are bound to these magnetic field lines and go around and around and around and can be trapped into a finite donut- shaped volume, and in this way prevents it from coming into contact with anything and melting whatever it touches.
This is actually something that we do everyday; we have tomamaks, we can find plasmas. Once we have them, we heat them up to fusion relevant temperatures to give the particles enough energy to fuse and to start creating basically heat and energy of their own. But the problem is: if you think of another energy source, if you think of a light bulb. You go to the hardware store; you buy a light bulb. You plug it in at home; you turn it on, and it does two things. One, it produces light. Two, it gets hot. If you touch the glass, it will be hot. Now, instead of a 50-watt light bulb, think of a 500-megawatt light bulb—so 10 million of these light bulbs. It is a lot of light. It is also a lot of heat, and that is our problem. We have a lot of heat.
A tokamak is a little bit like a light bulb, except that a light bulb has the advantage that the light goes in all directions, and it uniformly distributes that heat. But a tokamak has directionality to it; all of that heat gets thrown down here into this area at the bottom. So, whatever material we put down there has to be able to withstand a very high heat load—absolutely a requirement: you cannot let your tokamak melt. So, the most obvious thing to do, the first step, is to create your tokamak—the walls of your tokamak out of something that has a very high melting point.
This has been done, for example, at the ASDEX Upgrade tokamak that is here in Germany; this is a picture of ASDEX here. All of these tiles on the wall are made out of tungsten, which has a melting point of about 3,400 degrees. That is pretty good, and for most everyday plasmas that we created at ASDEX is enough, because the plasma is confined with the magnetic field; it doesn’t actually touch the walls. The tungsten by itself can pretty much handle the heat.
For example, this is a discharge we ran just a few weeks ago. We had about 10 megawatts of power in this. I slowed it down a little bit so that you can get a better feel for what these things look like. In this discharge everything went perfectly; nothing melted, nothing sparked; everything was fine, which is why I am showing it first, because this is not always the case. Occasionally things get a little too hot. For example, in this discharge we increased the amount of power we put into it. We went to 13 megawatts opposed to the ten that we did before. Here, when I play this, you will actually see that the tiles down here start to actually glow. This is tungsten that is getting so hot that it starts to glow. Glowing is one step before melting, and that is generally what we want to avoid.
Alternatively, in this discharge, when I play it, there was one tile along that wall that was not perfectly aligned with the other one. It stuck out just a little bit. You want your surfaces to be smooth. When something sticks out, it catches more heat than the ones around it. And, in this discharge, that tile—it is going to be somewhere right here—it actually got so much heat that the tungsten became molten and went flying everywhere, and that is the end of your plasma discharge. This is not what you want if you are trying to make a steady-state reactor.
This is the problem: both of these plasmas were suffering from the same problem, and that is that all the heat that we put in, and all the heat the plasma produces itself, also has to flow out of that reactor at some point in time. You cannot let all of that heat get deposited locally anywhere. You need to distribute it evenly along all of the surface, along all of the wall. You need to control how much comes out, how quickly, and where it goes.
One of the ways that we have developed in the last few years of doing this—and this is a brand new research that has just really been proven to work, and we are all feeling much better about the chances of fusion now—is through radiative cooling. Now when I say “radiative”, I am not talking about radiation; I am not talking about Madam Curie, I am talking more like the radiators we have in our house that keep us warm in the winter. Actually, it is a lot like a light bulb that distributes the light in all directions—going back to that analogy.
So, electrons are bound to atomic nuclei, and these electrons can bounce up and down in their atomic shells. When they do that, they emit light in the form of photons. That is an even distribution in all directions, which is exactly what we want—which point in time, you might now be saying, “Wait a minute Rachel, you told us earlier that in fusion plasma, the electrons are stripped off of the nuclei: how can they bounce up and down?” Well, that is true for hydrogen, which only has the one electron. But, if you go to higher-z elements, like nitrogen with 7 electrons or argon with 18, then they can retain their electrons even in a hot fusion plasma and have enough electrons to jump and down and to distribute the light.
So, this is what we do. We inject argon and nitrogen into fusion plasmas: argon in the centre and nitrogen down here at the bottom in order to control how much we radiate light and how much energy is distributed. We actually monitor the temperature of the walls, and as the walls get hotter we inject more gas. That cools the walls, and we can keep the temperature of the walls where we want them. So, this, for example, is done in this discharge. I can show that it actually works. This has got 14 megawatts of cooling. I am going to show identical discharges. The only difference is that this one has radiative cooling; we are puffing gas, and this one we are not. You can see for yourself what happens. This is a thermal image. So hot things show up as red, cold things as blue. Without the nitrogen, this diverter very clearly starts to glow. So, this is a proof that what we are doing actually works.
Actually, in this last campaign, just a few months ago, we put the maximum amount of power available on ASDEX Upgrade; that was 23 megawatts, into the machine, all at the same time, which we had never before, because it was a guaranteed way of melting the diverter. This time we did it with argon, and argon and nitrogen gas: here is the traces of these two elements right here. We monitored the wall temperature. We had a set temperature: the black line. This is where we want it to be. Then we measured the wall temperature, we adjust the gas that we puff into it to control the radiation, which I am showing up here. Everything worked perfectly. Nothing started to glow. Everything was exactly where we wanted it to be, which is just a really, really nice thing to have in your hand when going to a next generation sort of device, the ability to control the steady state fluxes.
Now, unfortunately there is another problem. This is the steady-state problem, and we have that solved. But, there is a transient problem in a tokamak as well. This comes about, because if you look at the temperature or the pressure profile across the middle of the tokamak, it would look something like this: where you have a very steep gradient here at the edge. That steep gradient is necessary, because it allows you to get to very high temperatures in the core where the fusion is actually happening, but it is also a problem. Because, if the pedestal gets too steep, it will actually collapse. When it collapses, it ejects plasma out of the confined region where we want it to be and into the walls where it can damage the walls; it can melt the walls, which is not what you want to have happen.
This is a movie that I will show in just a second of actual measured data. This is the measurement of the edge pedestal and how it behaves. This instability, when this happens, when it reaches a critical gradient and ejects plasma, it is called an ELM or edge localised mode. When it does that, this is the power that will hit the wall. You will see the gradient build, build, build; it will get too steep. It will eject plasma, which hits the wall, and it will happen—we have it twice here in this film. So, this is the temperature; it building, building, building...it is going to hit the critical gradient here in just a second and... flattens. All the plasma that was in that area in between flies out. So, we can also see this on fast cameras. We have lots of images of this sort of thing; so you can see it with your eyes. You will see the countdown coming down to zero, and that is when you will see the plasma get expelled. There it is; that is the ELM. You can see the plasma coming out.
Now in present-day devices, this is not that much of a problem, because our plasmas are relatively small. So, the amount of material that gets ejected is also relatively small. But in future devices, which will be much bigger, the amount of plasma that gets ejected will be much more. But the melting point of tungsten is not going to increase; it is going to stay exactly the same, which is why for future devices it is absolutely imperative that we get rid of this instability.
So, how do we do that? Well, this has been a topic of on-going research for a very long time. We have just started having success at this. What you do, if you remember I said the plasma consists of charged particles. We can find those particles with magnetic fields. So, if we destroy the magnetic field, at the edge, just a little bit, not a lot—just a little bit—then you can effect the confinement of the plasma just enough to stop that instability from happening. You can keep that pedestal from reaching the point where it suddenly has to collapse.
This is an example of how we do it. We have coils that we have installed outside the vacuum vessel of ASDEX Upgrade, and we run current through those coils and create a magnetic field. That magnetic field interacts with the tokamak magnetic field and can actually adjust it enough that the instability no longer occurs.
So, here is an example of this actually working in ASDEX Upgrade. Here is the coil current at the bottom. This is the magnetic field that we are creating with these external coils that interacts with the tokamak field. Here, this is the power going to the tiles. Each one of these spikes is an ELM. You can see when the coils are on, there are no more ELMS. So, it works.
With that I am going to go to my conclusion slide. So, in the last few years the fusion energy community has been working very hard to come up with solutions to the practical challenges of making fusion energy work. I hope, after this quick presentation, you will agree that we are having success at coming up with these solutions. We are actually having a lot of success. ITER will be the next step fusion reactor being right now in France. It will be the first device to show definitively that fusion energy can work. It will demonstrate the feasibility of this by producing 500 megawatts of fusion power. Thank you very much.