Tuesday, October 26, 2004 ... Deutsch/Español/Related posts from blogosphere

Marvin Cohen, nanotubes, and space elevator

Marvin L. Cohen gave the first Loeb lecture at Harvard, and I liked it. The Loeb lecturers are always very special people - very respected physicists and physicists with exceptional skills to present their subjects. Brian Greene was the Loeb lecturer in the spring.

Cohen is a condensed matter physicist, and he decided to present

  • The Standard Model of Solids
As the president-elect of the American Physical Society (who will probably replace Helen Quinn, our fellow particle physicist, if I understand it well), he had to start with two commercials.
  • One of them was the World Year of Physics 2005. The next year will be celebrated as the international year of physics because it will be 100 years from Einstein's miraculous April 1905 in which he discovered special relativity, the theory of photoelectric effect, and the theory of the Brownian motion, among other things. Cohen's main goal associated with WYP 2005 is to attract many 10-year old girls to physics.
  • The second commercial was about his plans to reorganize APS. If I remember well, Cohen divided APS into four sections: old-fashioned physics (atomic, molecular, optics); astro-particle (which probably include things like string theory); condensed matter physics; plasma and non-linear physics
I did not make any notes, but let me reproduce some of his points about APS. The "classical" section of APS (atomic, molecular, optics) is doing very well. A significant fraction of astro-particle is about the search for a TOE, but also about the Big Bang. Plasma physics continues to develop thermonuclear fusion reactors. And condensed matter physics, which is the largest portion of APS, is disorganized because it develops into too many directions.

Many people who know a lot about science as well its sociological structure and visions start to agree: the interdisciplinary topics in condensed matter physics start to dominate. Biophysics and nanotechnology start to be the most important applications of condensed matter physics - which means, in a sense, that the physicists are moving towards biology, and we must live with that. The task is to preserve the exceptional role of physics among sciences - the role of the most rigorous natural science.

Most of his talk was about physics, of course. He quoted Dirac in 1929 who declared something along these lines:
  • A great portion of physics and all of chemistry can be now reduced to known fundamental physical laws, and it is just a matter of technical difficulty to obtain any results we want.
Cohen showed himself as a reductionist who is simultaneously an emergent person. ;-) Obviously, unlike many of his condensed matter colleagues, he has no problems with the idea of reductionism, and he showed us many examples of the things that have been understood and "reduced", especially if you use the right approximations (pseudopotentials, and so forth).

Take superconductors. It is still believed that all of them are sort of related to BCS superconductors - they are governed by pairs of electrons. If one electron loses momentum (by a collission with an impurity), his partner is still forced to react in the opposite way, and therefore no energy is lost. OK, there are speculations that there can exist other types of superconductors, but nothing conclusive.

It seems that the experimentalists are much ahead of theorists in condensed matter physics. All major new phenomena - such as superconductivity - were first discovered experimentally.

I can imagine that this will change sometime in the future - particle physics is currently just in the opposite situation, because the experimentalists just cannot find anything that the theorists can't predict. Cohen showed how all possible crystallic and superconductive phases of sillicon can be understood, more or less from the basic theoretical principles. Sillicon as a superconductor had a simple, roughly spherical Fermi surface, and therefore it was the first one that was understood. (The first observed superconductor was Mercury, however.) Nevertheless he showed another superconductor, something like MgB_2. Its Fermi surface is composed of four sheets. Nevertheless, today it is probably understood even better than Si, and the pictures looked convincing!

Most of the talk was about nanotubes and their applications, and they were very interesting. Because I must also do other things, let me be sketchy and choose just a couple of examples:
  • Nanotubes are very thin tubes whose thickness is as small as tens or hundreds of nanometers
  • Nanotubes have been seriously proposed as a material to build a "space elevator" - a kind of rope that will be hanging from the skies. You will be able to climb to the outer space, if you wish ;-)
  • Four wine bottles filled with nanotubes, if properly organized, will have greater memory than all human brains in this world altogether - an interesting research direction is to construct computers and memory chips based on these objects
  • Nanotubes may be inserted into one another, and you can build a nanomotor - a motor that rotates whose size is smaller than the radius of a typical virus
  • Nanotubes, if inserted into each other, exhibit friction - the question what does the energy dissipates into is sort of entertaining and possibly deep because there are not too many available degrees of freedom
Marvin Cohen will continue on Wednesday and Thursday, but I am afraid that there are too many other things I must do and attend.

Add to del.icio.us Digg this Add to reddit

snail feedback (8) :

reader Anonymous said...

Thanks for sharing this, it is a very good opportunity for those of us that can't attend such seminars. I'm sure many people would appreciate if you keep on writing things like this.

reader Anonymous said...

I agree.

You become our eyes and ears, where life would have never permitted such possibilties. Keep it up and thanks

reader Anonymous said...

Robert Lauglin has a different perspective I think from a condensed matter Perspective that he might have rejected, any ideas on string theory as emerging as a discriptor of reality.

Likewise, if the very fabric of the Universe is in a quantum-critical state, then the stuff that underlies reality is totally irrelevant-it could be anything, says Laughlin. Even if the string theorists show that strings can give rise to the matter and natural laws we know, they wont have proved that strings are the answer-merely one of the infinite number of possible answers. It could as well be pool balls or Lego bricks or drunk sergeant majors.http://www.fortunecity.com/emachines/e11/86/beneath.html

Would it be wrong to state that string theory is the quantum mechanical discription of the spacetime fabric in a most generalized way?

reader Luboš Motl said...

Yes, Loughlin is exactly the person whom I had in mind when I wrote "unlike others, Cohen has no problems with..." ;-)

Loughlin's antireductionists exhibitions in well-known newspapers were very provoking. ;-) Loughlin, after he achieved "everything" in condensed matter physics, decided to become a leader of quantum gravity, too - and therefore he likes to produces various things that are not accepted too seriously by the mainstream quantum gravitists. But Loughlin probably enjoys it, which is good for him. ;-)

Yes, the descriptions of spacetime and its geometry that appear in string theory are the most "general ones", at least in the sense that they allow to incorporate the principles of quantum mechanics, and deduce which other phenomena can co-exist with gravity in the most general quantum theory.

reader Joel Gilmore said...

I definitely agree with the others - I think it's a useful and interesting resource to have these talk summaries online. I was interested in the splitting up of the APS - do you think that's a "good" division? Would it mean that quantum computing and the like would be in with "old-fashioned" physics? Or would it go in with condensed matter, perhaps?


reader Luboš Motl said...

Thanks! These are all very good questions. Actually, it is likely that quantum computing must belong to one of these four groups, but I am not sure which one it is.

It's probably possible to work with many different divisions, if it's done properly.

Concerning our fields, I think that the interplay between string theory and other particle physics is very important, and they should never be divided and/or lose contact with each other.

There are many relativists who are spiritually outside the "astro-particle" group, because of some sort of religion. I find it hard to accept. There should be one group only. What I want to say is that people just should NOT become professionals in particle physics and/or general relativity without knowing the other subject pretty well. Of course, moreover I think that theoretical physics professionals should know at least basics of string theory. They can think whatever they want, but it is just wrong to be allowed to be a complete ignorant about major related fields.

The organization of APS within these groups won't affect too much, I think. The people will most likely continue to think the same things about one another, and interact with the same frequency.

I have the same feeling that the organization around condensed matter physics - and its corridors to biology - may be more important. It should be open to debate whether the actual science done as "biophysics" is still closer to physics than to biology. If this "biophysics" branch of science grows, it may be a good idea to establish new departments dedicated to it entirely. I am sure that they need people with the usual "physics" viewpoint; they also need the real "biologists".

It's a matter of convention whether this new science "biophysics" remains a part of APS or not. Of course, the division is a social construction to some extent - all natural sciences are parts of PHYSICS if you think in general enough terms. ;-) It's just that the "core" of physics should satisfy some rules of rigor, accuracy, and extensiveness of use of mathematics.

In this sense, I believe - and it may sound as chauvinism to you - that the ultimate reductionists; high-energy theorists; and likes of string theorists should keep on deciding what "is" physics and what is "not" physics. Some fields related to biology can just diverge too far. There's nothing wrong about it and it can be extremely useful for the society. But if it diverges from physics too much, it may be better to split it.

It makes sense for a scientist to be organized as a physicist if he's still able to see the links of her field with all other subfields - well, the connections eventually go to the Standard Model (or string theory) which is the root of the physics tree. If someone thinks that he works for industry and fundamental physics becomes irrelevant for him, it becomes less natural to organize him in APS.

reader Anonymous said...

The link I gave on Lauglin's self organization undertanding. Doesn't this ask the same question of what string theory can do as a emergent property of the third superstring revolution?

Lauglin's hate for the term First Principles, yet what is the position here, when we are trying geometrically to define quantum gravity, and find it is a property of the spacetime fabric?

Thomas Larsson spoke briefly on the symmetry of the circle, and if we saw this emerge from some supersymmetrical reality, how would strings do this?

As to Computers, Gerard't Hooft explains this for us?

Would not Smolins determinations of,

calorimeter design for GLAST produces flashes of light that are used to determine how much energy is in each gamma-ray. A calorimeter ("calorie-meter") is a device that measures the energy (heat: calor) of a particle when it is totally absorbed"Glast's calorimeter(a computerized version of information to simulate the environment, speak to this?

reader Joel Gilmore said...

Interesting thoughts - thanks! I'm actually doing a PhD project in "biophysics", so have had a fair bit of contact with a range of people who call themselves "biophysicists". Some really are doing really interesting physics, looking at delocalisation, coherence, etc. But others are really just biologists who use laser tweezers or atomic force microscopy to do their experiments - and while I'm not in any way slighting the importance of their research, it's hard to justify calling them biophysicists just because they use lasers. Though I'm sure others would hotly disagree!

I guess part of the problem is that defining biophysics is so hard - do you have to be interested in the physics of your biological system? (This perhaps relates also to your comments about deciding what "is" physics and what's not.) Is it enough to be applying techniques from stat mech or field theory to analysing gene sequences?

In my case, I'm looking at the role of quantum mechanics in biology, and how we might model optically active molecules (i.e., find an applicable but useful Hamiltonian). But our interest is two-fold - on the one hand, we want to better understand the biology. But we're interested in using biological systems as a test of physics models - biomolecules are highly tuneable (through genetic engineering, mutation, etc) and can be used to explore a wide range of parameter space. I come from a maths & physics undergrad, with zero biology, so my interests lie very much with understanding the fundamental physics.

Interesting questions indeed.

Illuminating Science

(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) })(window,document,'script','//www.google-analytics.com/analytics.js','ga'); ga('create', 'UA-1828728-1', 'auto'); ga('send', 'pageview');