« The next wave of the web | Main | Simple sounds make for sound investments »

Mini fridge exploits brownian motion

Nano paddle could, in principle, cool a pool of molecules.

Just the thing to keep your nano beers cold: an idea for the smallest refrigerator in the world.

Read the story here.

Posted by Nicola Jones on May 30, 2006 10:26 AM |

TrackBack

TrackBack URL for this entry:
http://blogs.nature.com/cgi-bin/mt/mt-tb.cgi/655

Listed below are links to weblogs that reference Mini fridge exploits brownian motion:

» Refrigerator Magnets from Refrigerator Magnets
Can your refrigerator magnet do this?Magnets.com is a leader in the custom design, manufacture, marketing ... [Read More]

Tracked on June 17, 2006 10:55 PM

Comments

All surfaces are stepped at the same atomic scale as the working fluid's structure and are thermally jiggling their bond distances. One doubts discrete analysis spawns the Demon in the given example.

Why not corannulene facing one direction atop and the other below like a cup anemometer? It would be much easier to make and just as silly to apply.

Posted by: Uncle Al | May 30, 2006 04:29 PM

This appears to be another variant of Feynman's Brownian Ratchet, and thus represents a perpetual motion machine of the 2nd kind. The thermal agitation of the paddles themselves as well as the friction between the shaft and the barrier would contribute to defeating the net clockwise movement.

I am surprised that a publication with the stature of Nature would even entertain this absurd idea seriously.

Perhaps this is a fine example of how classical thinking can be misleading in the quantum world of small scales.

Something to consider is that, for example, molecules impacting the wedges and the paddles at various angles will impart angular movement at right angles to the shaft, thus causing the shaft to impact with the barriar, thus negating any apparent gain from the wedge shape vs. paddles on the underside. In my thought experiments, there would be no net angular movement.

Posted by: Fred Mitchell | May 30, 2006 08:14 PM

Not sure how this differs from an old idea I developed years ago - which I never thought would work and have dismissed as wrong. This was to create a "Maxwell's Demon" by manufacturing a flexible membrane seal between 2 gas filled insulated chambers. A VERY small rigid rod would be stuck to one side of the membrane, projecting into one chamber and ending in a small metal loop which wold be positioned between the poles of a permanent magnet. Brownian motion by molecules on BOTH sides of the membrane should cause the flexible seal, rod and lop to vibrate, and the resulting induced currents in the loop should result in the motion being damped with consequent heat generation only in the "loop chamber". This would create a small temperature differential that could be exploited with thermocouples - hence, perpetual motion??? I don't think so!

Posted by: Phrustrated Physicist | May 31, 2006 03:13 AM

I agree with Uncle Al; I strongly suspect that it is not possible to construct a 'triangle' on an atomic/molecular scale.
It is easy to create paradoxes/almost paradoxical if one does not play by the rules of reality.

Posted by: Spud Gun | May 31, 2006 09:14 AM

Léon Brillouin, then in the US, showed, shortly after the war, at the time information theory was most popular, that Maxwell's demon could not, even in theory, create a temperature gradient because of the "cost" of information relative to the speed of molecules (the uncertainty principle). How about applying such an analysis to the "mini fridge" ?

Posted by: Jean-François Foncin | May 31, 2006 03:49 PM

i) If the authors are right, then mechanical motion follows from heating the two fluids at different temperatures. For example, eq. (1) of the arxiv preprint provides the resulting velocity for a simpler, translational motion of this kind. It looks rather similar to a perpetuum mobile, or, better, to creation of (angular) momentum ex nihilo.

ii) The authors assume low density of both fluids (page 2). Moreover, they invoke Onsager symmetry (L21 = L12, same page). In turn, the latter involves thermodynamical (hence macroscopic) quantities, which are hardly meaningful in the low density (i.e., large mean free path) limit.

iii) Linear Irreversible Thermodynamics of Ref. [4] requires that the product of a 'thermodynamical force' and a 'thermodynamical flux' has the dimension of an entropy production density, i.e. energy/(temperature*time*volume). I am not sure that this requirement is satisfied here. In particular, if the thermodynamical flux has the dimension of the inverse of a temperature (as suggested in the preprint, page 2) then the corresponding thermodynamical flux should have the dimension of a heat flux, not of a velocity.

iv) The computations involving the master equation (4) provide no independent confirmation of the equation (1) quoted above, as both follow from the same ref. [3] with the same authors.

v) The authors require a thermal conductivity proportional to the inverse square root of the temperature in order to ensure self-consistency of Onsager formalism (page 3). To my knowledge, no known material exhibits such property.

vi) The authors provide no proof of validity of Onsager symmetry to their own system.


Item vi) deserves further discussion. In my opinion, Onsager symmetry should never be given as granted. As pointed out by a classical paper of Casimir in 1943 -and implicitly assumed in Onsager's proof itself- Onsager symmetry requires fluctuations of any quantity at a point x in space and at a time t to be related to fluctuations of the same quantity at the same point x and at a different time t'. Then, Onsager symmetry follows from time-invariance of the microscopic physics underlying the fluctuations. Of course, spatial distance may be unrelevant in crystals where the same microscopic structure is observed everywhere, but is crucial to transport phenomena. This is likely to be the underlying reason of both the splendid success of Onsager symmetry relationships in solid state physics (e.g. the Seebeck effect quoted by the authors) and their simultaneous failure in the physics of energy transport. Not only the thermal conductivity quoted above in unphysical, but even in plasma physics the Onsager-based 'neoclassical' theory of energy transport fails to fit experiments.

Posted by: Andrea Di Vita | August 30, 2006 05:06 PM

Еhanks for the reference, the useful article

Posted by: Marina | November 12, 2006 01:08 PM

I cannot see anything special about that device. Maybe I got it wrong, what it should be able to do. But they clearly state, that the "motor" won't move without a temperature gradient, now would it cool anything without an external driving force.
This is nothing else than my refrigerator, any Peltier element ans any gas turbine is doing.
The interesting point would be: Can it do this with less/no entropy production?

A device that converts Brownian motion in a reservoir of one temperature T1 into mechanical work applicable to the environment, thus cooling the reservoir, is something else.

Posted by: Steffen | August 5, 2008 12:31 PM

Post a comment

Comments will be reviewed by staff before being published. You can be as critical or controversial as you like, but please don't get personal or offensive, and do keep it brief. Excessively long entries may be cropped. Remember this is for feedback and discussion - not for publishing papers or press releases.

We strongly encourage you to use your real, full name. Email addresses are required: this is just in case we need to discuss your comment with you privately. They won’t be published.


Please enter the numbers you see below - this helps us to cut down on spam. Note that attempting to post within 30 seconds of hitting ‘preview’ or ‘post’ can cause the system to think you are spamming the site. If you are having trouble with this system, you can instead e-mail a comment to 'inthefield at nature.com'.

Search