Talk:Brownian ratchet

History Error
This ratchet is also a restatement of Maxwell's Demon more or less. Feynman may have made this idea famous but he did not invent it... It was being discussed by Smoluchowski in 1912. Less full out incorrect statements on Wikipedia please. — Preceding unsigned comment added by 128.32.166.162 (talk) 23:27, 13 November 2012 (UTC)

I am altering the article from "Feynman invented" to "idea made famous by Feynman". I will not add Smoluchowski without a reference. 131.111.8.96 (talk) 22:33, 8 December 2009 (UTC)


 * Glad you caught that. I looked up the references and added Lippmann and Smoluchowski's contributions. Thanks.  -- Chetvorno TALK 22:38, 15 January 2010 (UTC)

Could someone explain simply why a brownian ratchet is self defeating?


 * It's not an incredibly scientific explaination, but I did add sort of a plausibility argument for its failure that I've heard at talks on Brownian motors. Laura Scudder 01:13, 17 Feb 2005 (UTC)

This article makes a claim that ratchet slippage somehow prevents the ratchet device from producing work. The article should reference such a claim as an unproven theory until an appropriate reference is included in the article showing proof that slippage prevents the ratchet device from performing any work. Being one who has written many of simulations, include a trapdoor simulation, I believe such slippage is merely a technological issue, not a limitation. Thanks.--PaulLowrance (talk) 18:41, 28 May 2008 (UTC)


 * Good point. I added the reference.  Feynmann works out the mathematics of the slippage in Lectures on Physics. -- Chetvorno TALK 14:47, 1 July 2008 (UTC)

I added fact noted by Feynman that wheel would rotate in the other (backwards) direction if the temperature difference were reversed, as well as the parallel to why electrical diodes cannot extract useful work by rectifying thermal noise in a circuit at constant temperature.CharlesHBennett (talk) 04:07, 30 March 2010 (UTC)

could someone add a picture to this page? I found a couple (including an animated one here: http://web.mit.edu/8.592/www/lectures/lec20/ (hover over the pic to see the animation)) but I didn't add them to the article because I wasn't sure of the copyright stuff.

The picture would be clearer if the ratchet were drawn with unsymmetrical teeth, with a gentle ramp leading up to a sudden drop, as in Feynman's drawing.CharlesHBennett (talk) 04:07, 30 March 2010 (UTC)

could someone add a picture to this page? I found a couple (including an animated one here: http://web.mit.edu/8.592/www/lectures/lec20/ (hover over the pic to see the animation)) but I didn't add them to the article because I wasn't sure of the copyright stuff.

Thanks. -PB

scale
On a small scale brownian motion could be seen a lots of pools of different temperature. -Jeff

Pawls are not fundamental
It is not absolutely neccesary to use a pawl or even to have the machine rotate in one preferential direction to do work. An simple electric generator will output power no matter if the rotor goes one way or the other, AC currents carry useful energy even if they are null on average. Also at the brownian scale there are instantaneous large temperature fluctuations between the pawl and the ratchet even if they average over time.-Felix


 * The generator example is a good alternate illustration. If, for example, the generator is used to do work by sending current through a resistor at the same temperature as the generator, the random AC current from the generator will be no bigger than the random fluctuations of thermal Johnson noise current in the resistor. The resistor will act like a 'generator' itself some of the time, driving current through the generator.  So on average, an equal amount of noise power will travel both ways, and no net power will flow into the resistor.  On the other hand, if the resistor is at a lower temperature than the generator, it will produce less Johnson noise current, so net power will flow from the generator to the resistor, acting like a normal heat engine. -- Chetvorno TALK 00:00, 1 July 2008 (UTC)


 * I think the microscopic generator idea Felix was describing is an attempt to produce rectification of such thermal energies, and therefore resistance from the resistor would merely help. Such a hypothetical microscopic generator would convert higher frequency thermal energies to lower frequency energies. If a 1MHz signal is modulated by the same 1MHz signal the result is DC plus 2MHz. I believe that is the basis of rectification for diodes, but I'm uncertain as to the details of what Felix is referring to. --PaulLowrance (talk) 01:26, 1 July 2008 (UTC)


 * I'm not totally clear on what you and Felix were referring to, but the diode is an electronic analog of the ratchet, and a circuit with a diode and a resistor is an exact electrical analog of Feynmann's experiment, and fails to generate power at equilibrium for the same reasons. The exponential torque vs. velocity equation Feynmann derives for his brownian ratchet is exactly the same as the voltage/current relation in a diode.  See eq.46-1, p.46-4 Lectures on Physics, compare to Schockley diode equation in Diode.  -- Chetvorno TALK 00:31, 3 July 2008 (UTC)


 * Hi. I am well aware of such data, as I have spent years studying the possibility of diodes rectifying thermal energy. I am studying highly shielded measurements on 50GHz ZBD (Zero Bias Diodes) conducted out in the California desert that is showing DC voltage conversions from natural ambient thermal energy. It is well known that at microscopic levels vast temperature gradients always exist in all matter. In fact, it is impossible to get rid of such thermal fluctuations. Several years ago I spent a great deal of time measuring small *macro* temperature gradients at room temperature. There's no size scale at which natural thermal fluctuations vanish, but the fluctuations are greater at small scales. In terms of theory, the micro to macro scale occurrences in real semiconductor matter are vast and too difficult to compress into a nice little hand written equation. The quantum physics of diodes is based on mathematical approximations. One could write a book on just how controversial the math and science is behind diodes at thermal equilibrium rectifying ambient thermal energy. There's even question how to properly mathematically model thermal and shot noise. For example, there are papers that mathematically prove Gaussian thermal noise violates 2LoT. So from a distance, yes, it may appear conventional physics dictates that a diode cannot rectify thermal energy at thermal equilibrium, but when one becomes evolved in dirty details, present research from various scientists, etc. things become questionable. IMO in ~ a decade technology will reach a level where devices will convert thermal AC noise energy as sustained by the Sun into usable DC energy. Theoretically such devices would work day and night at any hospitable location. --PaulLowrance (talk) 16:29, 3 July 2008 (UTC)


 * I sure hope you're right; we certainly need new sources of energy. -- Chetvorno TALK 03:45, 4 July 2008 (UTC)


 * We figured it out how to extract limitless energy that graphene generates. It’s only a little power, but limitless energy nonetheless. Walid.Miran (talk) 12:37, 24 February 2021 (UTC)
 * See WP:NOTFORUM. --Hob Gadling (talk) 15:53, 24 February 2021 (UTC)

Remove claim that Brownian motors extract work from noise?
I don't know of any support for the claim that Brownian motors extract useful work from thermal noise. Since this seems to violate the 2nd Law, and the 'citation needed' tag has been on a while, I think it should be removed. Comments? -- Chetvorno TALK 09:54, 30 August 2008 (UTC)
 * I don't know who placed the claim you are talking about, but I would like to just pop in (no debating for now please) and say that my research in this field is showing that it is merely an interpretation and a failed one at that to say thermodynamics claims it is impossible to extract usable energy from natural ambient thermal energy. The following link is not an attempt at self-promotion. I do not sell anything and I do not accept $ donations, as it's 100% self funded. It's merely an outline, so the mathematics is excluded from the site for now until time permits --> http://aemcomp.googlepages.com --PaulLowrance (talk) 15:16, 30 August 2008 (UTC)
 * Good luck with your research. Meanwhile, I'm going to remove the unsupported Brownian motor claim. -- Chetvorno TALK 19:49, 11 September 2008 (UTC)

I know I am late to this debate but. Noise would need to be defined before stating whether or not the claim is possible or not. For example, equilibrium thermal noise cannot be used to do work; however, non-equilibrium noise can be used to do work. If the fluctuations are random but higher than the ambient thermal noise then yes they can be used to do work. This is how wrist watches work that neither have batteries nor a wind-up mechanism inside. — Preceding unsigned comment added by 128.32.166.162 (talk) 23:32, 13 November 2012 (UTC)

Granular gas section mis-represents the experiment
As far as I can see, this experiment is mis-quoted or mis-described. The abstract in phys rev lett does not describe any ratchet at all, the symmetry breaking is provided by the vanes having different coating (soft, i.e. energy-absorbing inelastic) on the clockwise side. Furthermore, the phrasing at the beginning ("despite being theoretically proven") is a gross overstatement; the original paper does not claim to provide thermal equilibrium but explicitly refers to the "necessary" out of equilibrium environment given by the shaking of the gas. The experiment is probably worth mentioning but it should be done accurately. 207.172.169.230 (talk) 14:56, 10 September 2010 (UTC)

Massless rod

 * "The ratchet is connected by a massless and frictionless rod to a paddle wheel that is immersed in a fluid of molecules at temperature T1."

This implies that other parts of the apparatus, notably the paddle and cog wheel, need not be massless. If that is so, and since these parts are all connected and moving in unison, why does it matter that the rod is massless? 86.160.83.115 (talk) 01:16, 14 August 2011 (UTC)
 * I agree. I have never, in any description of the thought experiment, come across a requirement that the shaft be massless and frictionless.  I think it should be removed.  Good catch.  -- Chetvorno TALK 07:40, 14 August 2011 (UTC)
 * Removed it. -- Chetvorno TALK 07:24, 18 August 2011 (UTC)

If T2 is a Vacuum or there is a differential hydrostatic pressure
If the pawl is encased in a vacuum, no particles hit the pawl to loosen it. Does this violate the second law of thermodynamics now since the ratchet should function? — Preceding unsigned comment added by LarsOlson (talk • contribs) 00:47, 1 March 2014 (UTC)
 * No, if the pawl itself is at the same temperature as the rest of the device (as it must be if the system is at thermodynamic equilibrium) the atoms it is made of will themselves be undergoing Brownian motion, causing the pawl to spontaneously bounce up and down, even if it is in vacuum. The Equipartition theorem says that every degree of freedom in a dynamic system at temperature T has thermal energy ½kT.  -- Chetvorno TALK 01:34, 1 March 2014 (UTC)

Response to the above: sorry I should have explained further or drawn a diagram of what I meant. If the pawl and spring (and gear teeth) are located in a completely separate container, enclosed, and the shaft of the rod is connected only via a nano magnet so that the paddle side of the rod spins the other rod... this creates a vacuum in one section and room temperature in the other section. This creates an apparent differential in temperature (actually pressure) without having to continually maintain a difference in temperature since one side is a permanent vacuum (imperfect or perfect one).

Also there are some serious flaws in your argument about brownian motion of the pawl in a vacuum. Brownian motion of a solid object is not the same as a gas or liquid. Imagine the brownian motion of aluminum or steel. The atoms move extremely small amounts internally, compared to gas or liquid which moves much more. The whole point of the browian ratchet is to move and vibrate due to other molecules bombarding it, not internal atoms of the actual paddle moving (a solid with low vibrations). If it were true that the brownian ratchet were vibrating without any molecules or atoms bombarding it and it just moved all by itself due to internal vibrations.. there would be no need to draw these diagrams of the brownian ratchet with any particles hitting the paddle. Why not simplify the diagrams and remove all the particles hitting the paddle since you claim the particles aren't even necessary for massive vibrations to occur? Likely what you are doing here is simply trying to prove that the second law is always true (rather than trying to falsify it) and coming up with creative conspiracy theories that you have not actually tested experimentally nor thought through carefully. Are you saying for example that solid steel has just as much brownian motion as air particles smashing into the metal.. Look at some online websites of brownian motion of solid objects vs air and liquid.

Also, creating a hydrostatic pressure differential using two containers of water is another option. One container is taller than the other, and has more hydrostatic pressure at the bottom of the container due to gravity. Pressure is like energy per area... so two different pressures are like two different temperatures, creating a differential. It costs no energy or money to create a difference in hydrostatic pressure - gravity does the work for you. Consider two containers and the rod of the ratchet is connected via a nano magnet. Each compartment is separate, No seals are even needed due to the magnetism allowing two completely separate containers to be used for both halves of the system. The hydrostatic pressure is more in the paddle side of the ratchet, and the hydrostatic pressure is lower in the pawl/spring/gear side of the ratchet. One does not have to continually maintain a cold temperature on one side of the system - hydrostatic pressure is permanent without costing energy or money. Yes this could in fact be falsified and proven to not work: that is how science works and I am simply posting these thought experiments to allow people to further study the subject without such straw man inventions like the simple brownian ratchet which was fairly easily knocked down. — Preceding unsigned comment added by LarsOlson (talk • contribs) 03:52, 18 July 2015 (UTC)


 * Well, in the first place, "I" and the other editors who wrote this article, are not trying to "prove" anything. That is not allowed on Wikipedia (see WP:SYNTHESIS).  The article summarizes scientific research published in WP:RELIABLE SOURCEs.   The sources are given in the 12 references at the bottom of the page.  Rather than just arguing off the top of your head on this page, you could actually look up the sources and see if they answer your objections.     The Feynman ratchet is a very well known and analyzed device and there are a lot of explanations of it on the web.  If you think the article does not explain what is in the sources correctly, or that we are using unreliable or bogus sources, that is something to discuss here.    If you think the sources themselves are wrong, that would be something to take up with the authors.


 * Assuming the explanation in the article is inadequate, I can try to figure out how to explain it better to you, as far as I understand it (I am not an expert in thermodynamics) so we can improve the explanation in the article. I admit the explanation in the article is brief and does not go into all the details.  However we can't just have a general discussion of the ratchet;  this page is for discussion of the article, not the subject (see WP:TALK).  -- Chetvorno TALK 06:16, 18 July 2015 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 1 one external link on Brownian ratchet. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20031206054642/http://monet.physik.unibas.ch:80/~elmer/bm/ to http://monet.physik.unibas.ch/~elmer/bm/#why

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.— InternetArchiveBot  (Report bug) 13:29, 9 November 2016 (UTC)

Not an apparent perpetual motion machine.
Even if the brownian ratchet were to work (i.e. it were to do useful work), it would not satisfy the definition of a perpetual motion machine: "A perpetual motion machine is a hypothetical machine that can do work indefinitely without an energy source." The energy source is the ambient temperature of the air. This is a finite resource, and the ratchet would eventually reduce the air temperature via reducing the velocity of gas particles. Therefore, it is not an "aparrent perpetual motion machine". This phrase should be removed. Cowlinator (talk) 22:03, 20 February 2020 (UTC)
 * No, because all forms of energy, in particular the mechanical energy produced by the ratchet, are eventually degraded to ambient heat (random motion of molecules), which is the power source of the ratchet. So if the ratchet could work it would be a true perpetual motion machine.  A perpetual motion machine that uses ambient heat is called a perpetual motion machine of the second kind. --ChetvornoTALK 01:40, 21 February 2020 (UTC)

Please update with: "Fluctuation-induced current from freestanding graphene"
I think the article should probably contain some information on this study. It may also be relevant to other articles (which?). It's currently featured in 2020 in science in short form like so:

A rippling graphene-based energy-harvesting circuit with the potential to deliver "clean, limitless, low-voltage power for small devices" if adequately incorporated into a chip is demonstrated.

Additional news report here. From this report of Physics World:

In the early 1960s, the Nobel laureate Richard Feynman popularized a thought experiment known as the “Brownian ratchet”, which had been conceived in 1912 by the Polish physicist Marian Smoluchowski. This involves a paddle wheel that is connected by an axle to a ratcheted gear. Both the paddle wheel and the ratchet are immersed in fluids. The system is imagined as being small enough so that the impact of a single molecule is sufficient to turn the paddle. Because of the ratchet, the paddle can only turn in one direction and therefore it appears that the Brownian motion of the paddle can be harnessed to do the work of turning the axle.

However, Feynman showed that if the two fluids were at the same temperature, collisions throughout the system would prevent this from happening. The only way work could be done, argued Feynman, is if the fluids are a different temperature, making the Brownian ratchet a heat engine.

In their new study, University of Arkansas physicist Paul Thibado and colleagues replaced the paddle with a freestanding sheet of graphene – a single layer of carbon atoms.

--Prototyperspective (talk) 10:25, 4 December 2020 (UTC)

Appeal to authority
I keep reading arguments where people are using “Feynman proved it therefore it must be true” to support their claim. I’ve listened to Feynman’s explanation, and no offense to his genius, but it makes no sense to me at all. Frankly, everyone who used the argument “Feynman said so, therefore it must be true” was appealing to authority, because as it turns out, the reason why what he was saying sounded like nonsense is because it was. We can now extract limitless energy produced by graphene, albeit little power, but still limitless energy. Yes, perpetual motion has finally been realized thanks to Brownian motion. Walid.Miran (talk) 12:51, 24 February 2021 (UTC)
 * By that reasoning, no proof is valid. You can always say "it sounds like nonsense to me", turning the proof into just something someone said and thus into an appeal to authority.
 * Your refutation is invalid. --Hob Gadling (talk) 15:53, 24 February 2021 (UTC)
 * Let me try to give an explanation.
 * Firstly, for simplicity I will consider the temperature of the paddle subsystem (T1), and that of the ratchet subsystem (T2), to be equal, otherwise a part of the work can come from the thermal potential, with the paddle and ratchet working as a heat engine.
 * For the ratchet to rotate at all, the pawl must be very weak, as the force from Brownian motion is very small; and because of the high randomness of the Brownian motion, the paddle may sometimes rotate backwards quickly while the pawl is up, and repeating Feynman, the pawl probably needs to be so small that it will also be affected by Brownian motion.
 * Making the paddle or ratchet larger to increase the force, which would let us use a stronger pawl, is practically not possible, as this would also increase the mass of the paddle-ratchet system, which would require more force to move.
 * Finally, the second law of thermodynamics is heuristic, so there will generally be statistical arguments in disproofs of perpetual motion machines.
 * Also, can you provide a source for your claim of a perpetual motion machine? Yodo9000 (talk) 23:15, 18 March 2024 (UTC)
 * Also, can you provide a source for your claim of a perpetual motion machine? Yodo9000 (talk) 23:15, 18 March 2024 (UTC)

Shape of teeth
I've undone the removal the "failed verification" tag on the passage that claimed Magnesco proved there was no net motion regardless of the shape of the teeth of the ratchet. The tag originally pointed to section but that does not seem to have any relevant discussion. I can't find any of the words "teeth", "tooth", or "shape" anywhere in the paper. Now Magnesco talks about periodic "potential functions" which we can interpret as the teeth of the ratchet, but I'm not seeing a claim either that any shape protential function will result in no net motion. Indeed, the whole thrust of the paper is that "net drift speed" is possible in some circumstances. If you think that this paper does support the claim, then please specify exactly where it does so before removing the tag again. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 14:16, 26 September 2022 (UTC)