Tuesday, November 09, 2004

Fusion reactor (ITER)

The European guys seem pretty convinced that the research of thermonuclear fusion can lead somewhere, although not much progress was seen in the last 50 years.

The International Thermonuclear Experimental Reactor (ITER) may be built in Cadarache, France, or in Japan, and the different countries are competing to "win" the project.

Brussels has "warned" that it may go ahead and build the first reactor with anyone who will be friendly enough.

http://www.cnn.com/ ...

I am not sure what are the estimated chances that such a reactor will work and it seems as a more important question than the location. ;-) It's pretty clear that if the thermonuclear fusion reactor worked, many other developed countries would try to build their own - because it would be a big deal.

Such a reactor would be much more environmentally friendly and more efficient than the fission reactors. Moreover, the "fuel" could essentially be the water in the world's oceans.


  1. If you check the ITER site http://www.iter.org you get the impression that they have solved the problem with the high temperature and the materials to be used for the containment of the plasma. How many hours can they run the machine until they have to replace the blankets in the vacuum vessel? http://www.iter.org/ITERPublic/ITER/b5.html But, this is perhaps not the biggest problem, maybe that is the stability of the plasma? Have they solved that, they do not mention it here http://www.iter.org/ITERPublic/ITER/rd_text.html

    Mike Ros

  2. Dear Mike Ros! Your links are very interesting! Thanks.

  3. Someone mentioned the ``first wall problem'' already...

    For plasma performance, the issue is transport. Specifically,
    turbulence and transport. Current thinking is that the core
    temperature profile is not very far from marginal stability
    with respect to several types of ``drift modes'' whose
    frequencies are on the long side of omega = c_s/L_T (sound
    speed divided by temperature scale length). In a hot plasma
    running quasi-steady, the ion/electron temperatures should be
    close. You want to have the core T above 20 keV. Marginal
    stability is given in terms of R/L_T, with R the major radius.
    It is not a large step to realise that this temperature will
    be given by the edge temperature times some factor. So it
    all boils down to the edge... my subject :-)

    Modern tokamaks are run in what is called an ``H-mode'' in
    which the edge turbulence is semi=suppressed and hence there
    is a steep gradient region over the last 2 cm or so of the
    plasma. This is called the ``temperature pedestal'' since
    the T(r) profile in the core sees this edge as a boundary
    condition, and T(a) is some finite value significantly above
    zero (r is the local minor radius and a is the boundary minor
    radius... all this gets generalised to shaped magnetic

    The whole ball game for ITER plasma performance is basically
    the value of T at the top of this pedestal. Since the gradient is limited by a stability boundary (thought to be
    due to MHD... magnetic fluid dynamics), what we need to know
    is the _width_ of this pedestal. So far, no theory can give
    us this, for the reason that the strong parameter inhomogeneity has to be faced, and we're still getting there.
    (Note here this MHD boundary is different from the threshold
    of the core instabilities... the parameters are different,
    mainly a parallel/perpendicular scale ratio issue, and the
    ``crossover point'' may well determine the pedestal width,
    but we don't know).

    ITER will be a hallmark experiment, because we will find out
    experimentally how this pedestal scales with major radius...
    does it go like R, or like the ion gyroradius (or a few tens
    thereof), or something else? We have to build ITER to find out, and know the answer to one of the two questions whose
    answers will tell us whether and how design of a tokamak reactor is possible (the other is the first wall problem,
    i.e., the materials interface one). Current tokamaks are
    too close in size... we need to know if it is possible to
    have a pedestal whose width can be 100s (or 1000s) of
    gyroradii but still be much smaller than a.


    ps the other performance question... global MHD instabilities, has seen enough progress that it isn't the
    main showstopper anymore

  4. As far as I can determine, there is no new information on the theory of how fusion works. Using a high temperature inside a "magnetic bottle" will not work. There is a company in Utah called ceramic works or something, that discovered heating water to high temperature separates the hydrogen-oxygen molecule bond and is trying to isloate the hydrogen gas from this high temperature steam. With a correct design, this could lead to fusion of the hydrogen atoms. Four atoms of hydrogen would have to fuse at once to make one helium atom. Current theory does not explain this type of fusion. Iter will not work. For more information, there is my research on this matter - search barry nix fusion

  5. Some time ago, I contacted the Department of Energy about some information concerning Fusion, the power of the Sun. Basically, It was a preliminary report on some interesting facts about heat energy that is not recognized by today's physicist. It seems these interesting facts are still being ignored. Without considering these facts about heat energy, the fusion programs, such as ITER (International Thermonuclear Experimental Reactor) and FIRE (Fusion Ignition Research Experiment) will not work.

    Here is a fact as an example; there is a candle experiment that demonstrates the property of heat. Heat, or the force that causes Heat, is a two-dimensional force. To illustrate this two-dimensional property of Heat, there is a simple experiment one can do that demonstrates this quality. Two candles are placed on the ends of a yard stick, enclosed in glass to keep the flames from going out, then the yard stick is set in motion on its own axis. As the yardstick revolves, the flame of the candles will bend at an angle of 90 degrees towards the middle of the revolution, indicating that Heat has the property of traveling towards the center of a revolving body.

    The way fusion works is this; the right angle, the bend of the flame, forces a particle, say a hydrogen atom, towards the center. In fusion, four atoms fuses all at once to make a helium atom, the resulting atomic weight of helium is slightly less than the combined atomic weight total of the four hydrogen atoms with the difference going into energy. What rubs the atomic physicist the wrong way is the fact that fusing four atoms at once is not their theory. This then is the problem in a nutshell.

  6. I work at the Culham Science Center, England. We have two machines, one called "MAST" and one called "JET". Fusion works, that has been demonstrated, particularly on JET, where a power of 17MW of fusion has been demonstrated. The problem is, that to run JET along with all of its associated components takes nearly 1GW of electrical power to run it up. In fact so much that the energy has to be stored on site by way of two enourmous 400 ton spinning machines, to even the load on the power grid.

    ITER has a couple of main advantages over JET - it is bigger. Arguably the main barrier to fusion is turblence, this transports the heat and energy out of the plasma quickly, but if you make the plasma very big, it takes longer and therefore you get better results. The problem is a bigger machine is much more expensive, but fusion power certainly seems to suit economies of scale quite well, unlike what "spiderman2" would have you believe. Secondly ITER will use superconducting field coils. The intense magnetic field required to contain the plasma has to be generated by driving several millions of amps of current around coils. A high school physicist will tell you the power lost goes like current squared, so superconding coils will greatly cut down the power requirements of ITER, and move us a bit closer to the goal of producing a machine that can make more power from fusion than the power it takes to run it!

    The question is not whether ITER will "work", the question is how well will it work. It is hoped it will provide 500MW of power (not to be turned into electricity) and have a plasma for several minutes at a time. Many theoretical work has been done, this machine costs 5 billion dollars, no one wants it to be a white elephant. The only thing to do now is build it and see what happens, if the predictions are right, and begin on some of the other technological and scientific challenges in fusion, such as materials.

    I think people under-estimate the importance of the fusion project. This really is the only geniune alternative to fossil fuel production, and unlike solar or wind power, can easily provide the "base load" and doesn't take up any more space than a conventional power plant.

  7. This is the kind of answer that I really appreciate, olikea, thanks, good luck, and a happy new year!

  8. I would agree that ITER might work, but according to the timeline on ITER's website, we probably won't get a commercial fusion reactor based on ITER until around 2050. ITER, as olikea said, is simply a larger tokamak reactor, large enough to overcome all the losses that smaller tokamak reactors experience through bohm diffusion.

    There are, however, alternatives to tokamak reactors like ITER that have the potential to be developed into commercial reactors long before conventional tokamaks.

    One is the Field-Reversed Configuration (FRC). The FRC can achieve higher power densities, meaning that it does not have to be as large as conventional tokamak reactors to achieve breakeven. Also, in the case of the Colliding Beam Fusion reactor here at UCI where I'm currently studying, it can achieve diffusion rates closer to classical diffusion, which is much lower than that of bohm diffusion.

    Another alternative is Fast Ignition, a form of inertial confinement that, unlike large laser fusion experiments such as NIF and LMJ, uses two different sets of laser pulses to compress and heat a plasma. Fast ignition reactors can be built with less precision than traditional laser fusion reactors because they do not have the same symmetry problems that NIF has with heating and compressing the plasma.

    To me, the issue that I have with ITER is not whether it will work, but whether it is the best way to build a fusion reactor. There are many alternative confinement concepts that I believe we should spend more time and effort pursuing instead of simply building bigger tokamak reactors. The alternatives I mentioned above both have the potential to be developed at a much lower cost and a shorter timeframe than fusion reactors based on ITER.