The road to nowhere

October 16, 2013

An experts panel of 21 top scientists in the field of fusion research has recently met to assign new objectives to ITER and define a new roadmap: please have a look at the article on Nature here.

They are top-level researchers, engineers, directors of labs with a load of experience which completely crushes my modest know-how. And yet, I don’t understand what they try to do. Or I understand too much: they try to do politics.

To quickly summarize the new roadmap, they want to focus everything on achieving break-even with a gain factor of 10, i.e., you produce ten times more power than you feed in your tokamak. And to do that, they skip every physics experiments which were supposed to pave the way to this goal (well, they postpone these experiments until 2030). The idea is simple: we know how a tokamak work, we have an interpolation figure for the confinement time in function of the machine’s parameters; we just have to build it: this is an engineering problem. The physics will help in tuning the system.

This is a very risky bet. We are far from understanding all aspects of a fusion plasma, even on present machines. A simple example: the H-mode, which is an improved mode of confinement of the plasma. There is no accepted theory of it. The associated instabilities, the ELMs (Edge Localized Modes), are not completely understood. We throw all this problems in the freezer and say: no worry, we will take huge margins and we will manage to reach Q=10.


This is a short-sighted approach, driven by the problems of funding: ITER managers need a quick visible result; something that will create the buzz, that will compete with the announcements of the NIF that they have already reached break-even. When I say quickly, the next opportunity is in 2027: this machine requires time to be built. But with this goal, you lose the big picture: ITER is supposed to provide the experimental foundations for a future power plant. And you get them by doing basic physics experiments. Even if they manage to achieve this primary goal, what will they do next? The physicists who were supposed to work as soon as 2020 will have to wait one more decade: they will not wait, they will go to other fields. In the end, you will have a machine for engineers with great chances that they do not understand, without the help of physicists, why it does not work.

I love plasma physics and fusion research: this is a beautiful area of physics; perhaps not so elegant as fundamental physics, but not so muddy as fluid mechanics: the plasma has a coherence that classical fluids do not have: you can elaborate nice theories of turbulence, gyrokinetics, hamiltonian and even better, you can see them in action in the tokamak. Huge progresses have been made since the concepts was proposed in the 50s. There still are some major hurdles and ITER, even if it was not the best choice, would have been a great tool to improve our knowledge. I fear that we are about to miss the moon.

Short illustrated plasma course – part III

February 22, 2012

Last part about particle trajectories in a plasma with magnetic field for today. With that we will be able to understand the configuration of a tokamak and how particles behave in its particular magnetic field.

Short illustrated plasma course – part II

February 22, 2012

Next step in understanding plasma confinement for fusion energy: the trajectory of plasma particles inside a magnetic field.

Short illustrated plasma course – part 1

February 22, 2012

To understand fusion energy with magnetic confinement, the first thing to know is the nature of a plasma.

ITER facts

February 22, 2012

ITER will be the biggest tokamak to be built to test nuclear fusion with magnetic confinement.
It is also one of the biggest experiments to be built.
M. Merola, the head of ITER Divertor group have found good comparisons to make feel the order of magnitude of this machine’s size.
You can find details of his presentation here:


US to enter the Wendelstein 7X project

July 8, 2011

This is now official: several american labs will cooperate with the Max Planck Institute to the development of the stellarator Wendelstein 7X (W7X) located in Greifswald.


Assembly of W7X - Credits Max Planck Institute for Plasma physics

You will find a good technical overview of the project in this presentation by T. Klinger.

One one hand, the physics of a stellarator is a bit simpler than the physics of tokamaks. On the other hand, the engineering can be seen as a nightmare because the facility is not axi-symmetric: just have a look at the shape of each coil: they all are unique.

Configuration of magnetic coils on W7X - Credit MPI-Plasma Physics

W7X will be the biggest stellarator in operation and has a budget of about 420M€; it should start in 2015.

The project was, at its beginning, controversial, because most of the funding was supported by Germany (with 30% help of the European Union) and it was seen as a competitor of ITER: some people did not see the point of spending some much on several big fusion facilities.  In addition, given the complexity, the project faced schedule and cost challenges. But the management was improved in the german way and now the construction is really on a good track. I find the smoothness very impressing, at least in comparison with ITER. But there is, I think, a good reason for this difference: there is only one main contributor, Germany. Just one head to drive and the management of interfaces becomes easier. Of course, there are other countries involved, especially Poland, but the decision process is far clearer than on ITER.

The participation of the US is a very good sign that the project is in good shape. Indeed, the budget allocated to magnetic confinement fusion is quite low (lower than on Inertial Fusion, which is also supported by military budgets) there and they already have several machines (DIII-D, Alcator C-Mod,… ) to run. I have, for instance, the feeling that ITER is not one of their priorities. That they are ready to collaborate in a foreign project, is really a very good news.

For those who are interested to work there (and they often have open positions), these are some pictures of the city of Greifswald (which has a big university):

But with hard winters!

EPS Conference: part II

July 4, 2011

I took an extended week-end to recover from the particularly busy month of June, a week-end far away from plasma physics and fusion research. But I was not finished with the summary from last week’s EPS conference in Strasbourg. This is the story of the second part of the conference.

The thursday’s session  had a taste of AIAA conference with a talk on reentry plasmas (more generally on kinetics in molecular gases) by Capitelli and one on plasma space propulsion. I must admit that I did not completely follow the  logic behind the presentation, it was a bit too fast for me; if someone has some complements, or an easy way to develop the presentation, he is welcome . I saw that there was a big consortium called Phys4entry organised for the 7th framework program of the european union. I should write a post on the issue of this FP: look at the list of participants and tell me how many they are: a lot, indeed! I would like that somebody from the EU explain me how to manage with limited funding so many interfaces between big companies, labs and startups. But this is another story.

The second talk on plasma thrusters, by E. Ahedo, had my full attention: the presentation of the requirements for space engines was good, with highlights on the two main drivers: specific impulse and thrust and the trade-off to do to achieve mission objectives. I was a bit disappointed by the presentation of the different types of thrusters. It was more an enumeration of the technical specs of each engine than an explanation of their principle. I think he should have narrowed the scope of its talk and focused on a small number of systems, for instance the Hall thruster and the Vasimr engine, which really have a lot of common parts with auxiliary systems for tokamaks and stellerators  (helicon source like in NBI, ICRF heating…). But well, he had a stringent time limit. Anyway, there also were some posters on the topic of helicon sources which have applications for these space thrusters.

We had again a full series of talks on transport (ok it is normal, it is issue #1), with ion heat transport on JET, and a special emphasis on the effect of plasma rotation. A particular noteworthy presentation concerned the I mode on Alcator C-Mod and ASDEX Upgrade by Hubbard. The I mode is an intermediate mode between the Low confinement mode and the High confinement mode, the famous H-mode. In the I mode, we have an independent control of energy and particle confinement, with a pedestal in Te and Ti but without density barrier: the energy is confinement but the particles not and fusion ashes can be swept out without extinguishing the reaction.

On the last day of the conference, we had a contact between two worlds: the world of plasma physics and the world of medicine,, with a presentation on synergistic effects in plasma-surface interactions by D. Graves and on application of non-equilibrium plasma by G. Kroesen. I won’t go into the details but the main idea is that a plasma is created in the air at the interface with the skin; the plasma enhances some  specific chemical reactions which can help, for instance, to treat wounds (with some nice pictures of them at 8.30, the day after the conference dinner!). I have a lot admiration for these people who create bridges between two continents of science: this requires to be open-minded, self-confident and of course  a bit crazy.

I followed, to finish, the parallel sessions dedicated to magnetic confinement fusion, to have a look at the numerical tools used to design and develop the plasma scenarii for ITER (by G. Giruzzi), with codes like CRONOS, METIS, HELIOS. The purpose is to develop scenarios with a strongly enhanced level of self-organization, with self-heating, Alfven eigenmodes and high bootstrap current. Well each time, I hear the phrase “self-organization”, I do not make a connection with tokamaks but with spheromaks.

Some other talks on current ramp on MAST and on current overshoot on ASDEX Upgrade to reach the improved H-mode were among the last noticeable talk I would like to mention.

All in all, an interesting conference which makes it possible thanks to a great effort of communication  to discover new areas of research. The big topic is transport driven by turbulence in tokamaks, with a subsection related to rotation. ASDEX Upgrade was very present with the use of RMF coils for the first time and it was the only big tokamak in Europe in operation last year.  People from CCFE had also very good talks, well prepared, cleared, proving once again the seriousness of this reference in scientific research.

I hope that I gave to people who were not present a feeling of the content of this conference and if you have some questions or comments, please do not hesitate.

EPS Conference: first comments

June 29, 2011

The 38th EPS Conference on Plasma Physics kicked off on Monday in Strasbourg with an incredibly hot and sunny weather. We have some free time this afternoon and will use it to give some of my feelings on this first part of the conference. If you want to add some other points, please comment.

This conference was, as usual, rich and dense in information, covering a wide range of topics from space plasmas to fusion plasmas,  and it would be difficult to summarize these last three days in few lines.  However,  I’d like to highlight some points of the program.

Plasma turbulence is certainly one of the most active area of research in Magnetic Confinement Fusion (and astrophysics) and three major contributors, A. Hasegawa, K. Mima and P. Diamond were rewarded this year with the Alfven Prize. In their prize lecture, they gave a (mostly) theoretical overview of electrostatic turbulence (and the formulation of the Hasegawa-Mima equation) and on zonal flows (you can find a review by Diamond, Itoh and Hahm here) and drift wave turbulence. Although the presentations were for several items a bit fast, they made it possible to feel the physics at stake. And there are good reasons that plasma turbulence is now so popular: they managed to build a solid system of theories and developed clever tools.

The experimental overview was given on Wednesday by A. Fujisawa but with emphasis on the drift wave  turbulence and not on interchange instabilities; he presented more particularly a nice diagram showing the change of paradigm: in the 60s, turbulence was seen as the effect of a system of drift waves governed by the Hasegawa-Mima equation. Thanks to Diamond, the situation evolved with the discovery of mesoscale structures called zonal flows which acted as an energy sink for the energy carried by the drift waves (inverted cascade). Turbulence is now seen as the result of non-linear interactions between drift waves, meso and micro scale structures. This interaction can be modeled through powerful predator-prey models and experimentally analyzed with tools like bicoherence.

On the subject of turbulence bifurcation, a very good talk by E.G. Highcock, a PhD student. He gave two references, here and here, for details on his work.

But why all this interest for turbulence? Because this is one of the major phenomena which governs the transverse transport, thus the confinement, in magnetic confinement devices.

In this field, a huge effort is invested in the numerical tools to simulate and understand turbulent effects on transport. F. Jenko and A. Bottino gave an overview of the numerical methods (global gyrokinetic PIC code  with adjustable control variates method to address the cancellation problem) used at the IPP-Garching; one idea to keep in mind: the analysis of turbulence is a multiscale challenge which cannot be solved on tokamaks in a brute-force style (i.e. with gyrokinetic Vlasov codes); a better understanding of the physics is necessary to develop simpler models. The two main points to take into account: the non-local effects (the meso-scale structures evoked previously) but also transport barriers and the electromagnetic effects (turbulent reconnection of magnetic lines).

I would have liked to attend more sessions on space plasmas to have a look at their methods but they were almost always in parallel with the sessions on Magnetic Confinement Fusion. I saw at least the plenary talk by about sun-earth connections by T. Pulkkinen. I keep in mind this spectacular video from the STEREO Satellite, which shows the comet Encke’s tail torn off by a solar flare.

In the reviews, we also had a talk by Piero Martin from RFX Consortium on MHD stability and active control of plasmas. There are two possibilities to avoid MHD instabilities: we choose the right parameters (β) to have stability (but we limit ourselves) or we accept to be instable but we try to control the instabilities and their effects. Among such instabilities, the kink modes which can be stabilized with a grid of saddle coils around the vacuum vessel, a solution already applied on RFX. Another type of control is the injection of ECRH to pace the sawteeth.

Another interesting review was the talk by Butanov on Extreme Field Science, interesting but very difficult to grasp: the slides were overloaded with equations. It is a pity because it addressed the physics of facilities like NIF, ELI or Hiper, based on the Laser Wakefield acceleration.

That’s it for the reviews, there were also announcements of interesting results.

We were lucky  to have the first presentations on ASDEX Upgrade (AUG) operations with RMP coils, for ELM mitigation; W. Suttrop showed that the coils indeed suppressed type-I ELMs. K. Lang showed that central pellet injection) could even be used during this phase to increase the central density without triggering ELMs (contrary to DIII-D or JET), making possible operations above the Greenwald limit (proving once again that this limit is an pedestal effect).

A talk by I. Chapman on sawtooth control in tokamaks with the original idea of using off-axis ICRH to improve sawtooth mitigation with ECRH.

To finish the second day of the conference, we have a status review of ITER by its Director-General O. Motojima. I would not like to have its job at this moment: in addition to the drastic cost reduction from last year, he has now to deal with the consequences of the earthquake in Japan and the associated delays on the program. In spite of the simplification he brought to the management, the internal structure of ITER remains heavy and probably delicate to handle. It is perhaps only a feeling but Motojima looked and sounded like a bit tired, what I can perfectly understand, given the circumstances.

While hearing the long (more than 1 hour) talk, I felt a persistent discomfort on the project itself due to one fact: I got bored. Well, this Fusion Science! We work on state-of-the-art technologies and pioneer the borders of physics! ITER should be motivating, INSPIRING; instead, I was bored and I am pretty sure that I was not the only one (who can confirm?). It is a point that should really be improved: of course, ITER is a serious project with huge amounts of money at stake. Yet, it should also be the accomplishment of a dream, it should make people dream; I would like to have the same pleasure  on ITER as on ASDEX Upgrade.

All in all, this conference makes it possible to discover other topics: most (but not all) presentations about physics are very well prepared to introduce the subject and present the basic questions and methods at stake. There are still a bit too long and some slides heavily loaded with equations that you do not have time to read in 30s. Some speakers even start to adopt the “TED” method by trying to narrate a story and give enlightening examples instead of the traditional exhaustive academic approach which is badly suited to the allocated time slot.

For those interested, the four pages papers are already online (I am not sure if there is a public access).

Let’s have a game

June 23, 2011

Next week, I will attend the EPS Conference on Plasma Physics in Strasbourg. A lot of things to see and here. I will present a poster there; for those who also will attend the conference, I propose you to find me and, as only hint, I give you a plot from the poster:

Good luck!

Tour of tokamaks: Alcator C-Mod

June 15, 2011

I had the opportunity two weeks ago to visit the Plasma Science & Fusion Center (PSFC) at MIT (that was my first visit and I was quite impressed to be in this temple of science and technology, but this is another story)) and to have a look at their tokamak: Alcator C-Mod.

The Alcator name is derived from the italian words Alto Campo Torus (the program was initiated by Bruno Coppi, born in Italy, and current leader of the very hyped Ignitor project.), which mean High Field Torus. The machine is the third of its class (yes, the predecessors were called A (operated in 1975, B designed but never built,  C operated in 1982 but which is a completely different machine) and started its operations in 1993.

At first sight, I was surprised to see how compact the machine is: a major radius of 0.67m but the torus itself and the coils are hidden behind thick concrete walls. However it has a very high magnetic field (5T on average, up to 8T, world-record) and can produce plasma at high pressure (for a tokamak plasma). AUxiliary heating includes ICRH (6MW) and Lower Hybrid (1.6MW).

Among the latest results from the machine, the possibility to decrease the local deposition of exhaust power by injection of impurities (see details). Indeed, a hot and dense plasma makes it possible to produce high amounts of power (which is what we want in the end), but too much exhaust power can lead to local overheating of the walls: a balance has to be found between these two requirements. To decrease the local flux of power, the idea is to seed some impurities (like Neon or Nitrogen) at the plasma edge: these impurities will radiate at other locations than the core plasma leading to a better distribution of the exhaust power.

Other results presented in non-technical languages are presented on the website of PSFC.

%d bloggers like this: