Talk:Maxwell's demon

From Wikipedia, the free encyclopedia

WikiProject Physics This article is within the scope of WikiProject Physics, which collaborates on articles related to physics.
??? This article has not yet received a rating on the assessment scale. [FAQ]
??? This article has not yet received an importance rating within physics.

Help with this template Please rate this article, and then leave comments to explain the ratings and/or to identify its strengths and weaknesses.

Contents

[edit] Don't over intellectualize it

It's simple, and you're missing it...

The Second Law of Thermodynamics states that two bodies of equal temperature, when brought into contact with each other and isolated from the rest of the Universe, will not evolve to a state in which one of the two has a significantly higher temperature than the other. The demon would be part of the universe since he is a separate entity not part of the two bodies brought into contact with each other, and once you bring into contact a third party with the two bodies supposed to be being isolated from the rest of the universe you are by definition no longer attempting to test/violate the SLTD.

Maxwell's demon CANNOT violate the SLTD because it's IMPOSSIBLE for him to enter into the closed system defined by the SLTD.

Maxblockm (talk) 18:30, 4 March 2008 (UTC)

Then make the system as the containers + the molecules + the demon + the door. 68.122.147.216 (talk) 21:16, 9 March 2008 (UTC)

Note: I support that the demon wouldn't work, but I think you have the wrong kind of proof. I'm not attacking your concept. 68.122.147.216 (talk) 21:18, 9 March 2008 (UTC)

[edit] Something to keep in mind

There are two points to remember about the second law of thermodynamics.

The second law, like the first, is an empirical statement. Its plausibility derives from empirical results. Its implications are empirical. Its essential character is empirical. The classic writers on thermodynamics point this out frequently and forcefully. The second law makes a statement about the real world - not about the world inside someone's imagination. It is not a mathematical result. It is not the result of a thought experiment.

The second law can only be disproved by an empirical result that contradicts it. To say that one has a thought experiment which "in principle" contradicts the second law, is itself a contradiction. At best, it can lead to discussions about why such a thought experiment will not work when performed empirically. Sometimes these discussions can be fruitful, as in the case of Maxwell's demon.

But what the second law says about Maxwell's demon is this. It is impossible to build a device that does what Maxwell's demon does. That is, if you actually build something that is supposed to work like Maxwell's demon, you will discover that it does not do what you thought it would do. In light of this, the following actually makes no sense...

 "Is Maxwell correct? Could such a demon, as he describes 
  it, actually violate the second law?"

Yes, of course, such a device, as he describes it, will actually violate the second law. I can think of many processes that will do the same. Magic, for instance.

On a different note. The this statement from the article is misleading:

  "The Second Law of Thermodynamics forbids (due to statistical improbability) 
   two bodies of equal temperature, brought in contact with each other and 
   isolated from the rest of the Universe, from evolving to a state in which 
   one of the two has a significantly higher temperature than the other. "

The second law says that perpetual motion machines of the second kind are impossible, not that they are improbable. The second law does not forbid them "due to statistical improbability". The second law forbids them, period. Kinetic theory forbids them due to improbability. Kinetic theory offers elucidations on the second law, but in the classic texts of thermodynamics, the second law is never stated in terms of probabilities. If we take the second law as a statement about improbability, how can we determine if an improbable process within an isolated system has "violated the second law"?

The rest of the paragraph is ambiguous and unclear, since a chemical reaction in one body can raise the temperature. It is helpful to consider a more precise statement Max Planck wrote on pg 85 of his thermodynamics book...

"The principle of Clausius states that heat cannot of itself pass
from a cold to a hot body. As Clausius repeatedly and expressly
pointed out, this principle does not merely say that heat does not
flow directly from a cold to a hot body - that is self-evident,
and is a condition of the definition of temperature - but it expressly
states that heat can in no way and by no process be transported from
a colder to a warmer body without leaving further changes, i.e. without
compensation. (Planck)"

The last phrase can be used as the basis of a better opening paragraph. Because - and this is the whole point of the topic - the question physicists have been asking about Maxwell's demon is: what are those "further changes"? I.e., what are the "further compensations" that a real-world "Maxwell's demon" must leave behind while it does what it does? Various scientists, like Szilard, presented views on this. Korkscru 13:45, 12 February 2007 (UTC)

Korkscru, all laws are emprirical. That doesn't mean that laws can't be violated by thought-experiments. If one can deductively prove from better-founded emperical knowledge that there must be a process that would violate a more tenuous law, then that law is considered to be not always entirely true. However, in the case of the Second Law of Thermodynamics, there are few better known pieces of knowledge that make sense in its absense, so in this case it's hard to imagine violating the law theoretically. More importantly, Maxwell's Demon does NOT do so, so it doesn't really matter.
Furthermore, the Second Law is certainly statistical. While it may be formulated in an absolute sense in older documents, in modern physics it is recognized to simply be a statistical fact. This does not mean it cannot still be stated as a law. With quantum physics in mind, it makes no sense to talk about thermodynamics in a nonstatistical sense, since there is nothing barring large collections of particles from just happening to decrease in entropy--it's just really unlikely. Note that this attitude is reflected on the page about the Laws of Thermodynamics and specifically on the one of the Second Law of Thermodynamics.Eebster the Great (talk) 00:15, 7 April 2008 (UTC)

[edit] Question

I'm not a physicist, but wouldn't the particles in A affect the particles in B, like a cue ball hitting the balls on a table.. Thus the particle entering B would raise the temperature of B only slightly, but what is the likelihood that the first particle that enters B affects the speed of a particle that exits B into A? I think this concept ignores feedback between the two volumes, the demon would have to be incredibly picky and familiar with all the interactions of particles on either side.. I don't even know what the second law of thermodynamics is, but whatever it is it probably reflects this property. The likelihood of a particle exiting one volume into the other is function of the size of the volumes and the likelihood that particles can collide or be influenced by the mass of the other particles in the volume which can slow particles down (??). I think I know how to prove this, think about the probability that a particle might exit through the hole versus the likelihood a particle might hit other particles not moving, then increase the speed of these particles.. You might test this by using a computer model of particles moving from one to the other by adjusting the volume, and using a programmatic demon in between that has a threshold for what is a fast particle and what is a slow particle.. Also from a stupid person point of view, how would the demon know the difference between what was slow or fast if the particles in one volume adopt the momentum of a fast particle having entered the volume of slow particles? Also what is the likelihood of the fast particles heading for the hole versus the volume with the slow particles? If its greater, wouldn't the slow particles become heated before the fast particles in the fast particle volume become cooled?

If you placed a physicist in a vacuum could you hear him yell timber?--Rofthorax 11:05, 24 August 2005 (UTC)

[edit] Reply

I don't think that any of the issues that you raise is a real problem for the demon scenario. Just assume that the demon keeps the hole closed until there is one particle near the hole that would decrease the entropy of the box if it was allowed to pass to the other side; then the demon opens the door and allows that particle through, and only that particle.

News This page has been cited as a source by a media organization. The citation is in:


see also vortex tube?

[edit] Quantum Tunneling Demon

Consider two chambers separated by a wall and place two particles of different energy in one chamber. In the wall there is a port with a potential barrier that only allows the particle of higher energy to cross over into the separate chamber. We can make observations on the system by enabling and disabling access to the port and observe the contents of the two chambers. If the potential barrier is less than the energy of the faster particle it can freely pass back and forth between the chambers while the slower particle is restrained by the barrier. The slower particle can still tunnel through the barrier but the rate at which it does this depends on the height of the barrier. If the barrier is slightly higher than than the energy of the faster particle, it too will have to tunnel through the barrier. So it would seem that the time that the port is open could be used to separate the higher and lower energy particles. If the particles were initially separated, and it was observed later that they were found to be together in one chamber, would this be a violation of the rule that the entropy of a system tends not to decrease? This thought experiment indicates that the entropy of a system can decrease. An number of observations would show that this is not statistically likely for a larger number of particles.Jbergquist 00:51, 26 December 2005 (UTC)

I thought about this, but realised that the any particle with energy to tunnel from A→B would also have enough energy to tunnel from B←A. Unless you can create a "one way tunnel" I dont think this is valid. 210.9.14.2 03:31, 30 August 2006 (UTC) (Mark Churchill)

[edit] along similar lines

Thermionic emission(Edison), Photoelectric effect(Hertz), Joule-Thompson effect, Hall effect, Mass spectrometry, Dispersion(Newton), Catalysis, Quantum tunneling, Gibbs paradox, aperture

[edit] A Suggestion

I find the idea more immediate if it's phrased like this:

Suppose we have a demon with a little (frictionless) trapdoor between gas-filled bodies A and B. The demon is very smart, and can work out what the location of every particle is in both containers going forward in time. When he knows there is a particle moving from A -> B he opens the door, moving from B -> A he shuts the door. Pressure will increase on the B side. A contracting-minded person hooks up a turbine between container A and B, and hooks this engine up to half a ton of concrete. The concrete will be dragged up a hill. In this way we have converted "pure information" into very "concrete" work. It is as if "knowledge" and "work" are, at a base level, fundamentally the same stuff.
Note this also works for probabilities - if the demon knows there is even an infinitessimally higher probability of a particle going from A->B than B->A he can get into the haulage buisness.

I realise this isn't quite the conventional explanation, but it seems to engage people a lot better. 129.94.6.28 08:34, 18 May 2005 (UTC)

You don't need Maxwell's demon to show a relationship between knowledge and energy. Suppose you are looking for Zabar's Bagels. You can either spend a boatload of gasoline and drive around the North America looking for this place, or you can save all that effort by asking someone who knows where it is. Korkscru 02:20, 13 February 2007 (UTC)

I wonder why the paragraph reading

What is so fascinating about the demon is that if it existed, it would be able to break one of the most important laws of physics. If we could tame (or build mechanical) demons, ships for instance would no longer need fuel as they could simply use the heat of sea water as a source of available energy; indeed, it would solve all our energy problems. But at the same time, it is very hard to say why it could not be built. Maxwell's story seems plausible enough - can we yet save the second law by proving that in the actual world no demon can exist? The quest for this proof constitutes the history of Maxwell's Demon. Surprisingly, it is still hotly contested whether this quest has been completed or not.

was removed? In what sense does it add "nothing substansive"? Please rewrite it if you feel its badly done (I can imagine a word like 'fascinating' being considered non-encyclopedic), but pointing out (a) the relevance of the thought experiment and (b) that it is still a subject of controversy seems to me very relevant indeed. Victor Gijsbers 09:00, 26 Oct 2004 (UTC)

Rewrote the description of why the demon doesn't work. It has little to do with quantum mechanics, but rather has more to do with information theory.

Well - that is highly doubtful. Earman & Norton, two very eminent philosophers of physics, suggest in their 1998 article Exorcist XIV: The Wrath of Maxwell's Demon that information theory does not say anything useful about Maxwell's demon. -- Victor Gijsbers

Interesting. Are there any (eminent or significant) physicists who make the same claim?

BTW the link to the Hilsch apparatus needed to make clear that the clever device does not in any way violate the Second Law. You'll find that out if you read the page, but it's better not to raise false hopes. And the Fu & Fu article as of this date is in a "might have been posted by just anybody" status, more like a random Web page than a preprint from arXiv. Dandrake 04:27, May 18, 2004 (UTC)

I thought that diagram looked familiar. The page that describes the Hilsch tube is mostly lifted from "The Amateur Scientist" in Scientific American, November 1958, p. 145. While we're not legally liable for pointing to a copyvio, we shouldn't do it. So now we don't. Dandrake 06:33, May 26, 2004 (UTC)

I just read this article, and the description about why the demon wouldn't work (which references information theory) doesn't make much sense to me. Is there a way to re-write that explanation so it would make sense to the layman? K_R 20:05, 5 Sep 2004 (UTC)

Yes, I will rewrite this article sometime soon. I recently wrote my Master's thesis on Maxwell's Demon, so I hope I know what I'm talking about. ;) Victor Gijsbers 13:41, 23 Sep 2004 (UTC)

excellent - I'm looking forward to reading it K_R 02:43, 25 Sep 2004 (UTC)


"Likely heterosexual"?

I do not understand this at all- I understand entropy and thermodynamics, but Maxwell's demon is bizarre concrete terms for something abstract. Are the people taking the demon to literally? They disprove the theory by saying that the demon would not have enough memory? Are they not aware that demons do not exist, especially scientific men like themselves, and even if they did, how could it be small enough to open a door for molecules? Please help me! I am very confused!!!

The discussion of exorcism comes out of nowhere. Could someone who understands the topic please explain the meaning of "exorcism" in this context?

Myles325 here. I think I can help with the "demon" and "exorcism" business. The demon is not meant to be taken as a literal demon. It is a sort of professorial type of joke, like Shroedinger's Cat. We can understand by "demon" ANY contrivance at all, mechanical or otherwise, which could be used to assess the energy content of each particle within a group of them, so more energetic than others, and divide the less energetic from the more energetic ones. By doing so, the "demon" whatever it is, creates a "hot source" where none existed before, and this can do work. Of course, the "demon" must do some work in separating the high and low energy particles in the first place. So everything rests on whether the "demon" can do the separation task using less energy than he gains by creating the heat source. If he can, then effectively, he is getting energy out of thin air. The business of "exorcising" the "demon" is just carrying this humorous analogy a bit further.

My puzzlement rests on the thought that the whole of life seems to be a "Maxwell's Demon". After all, think of a Polar Bear. Here is a heat source that exists in the midst of frozen emptiness. Yet, that heat was not introduced externally. The bear acts like a Maxwellian Demon, gathering it from subsidiary energy sources like fish, which do the same on a less complex scale. if the Polar Bear had to do more work hunting than he got from its rewards, the exercise would be futile. Before the advent of biological evolution, entropy was close to zero in an environment like this. Later, after the advent of replicating molecules (themselves a maxwellian demon), and the more efficient cell, and then multi-cellular life forms, we have energy hot spots all over the Arctic and Antarctic. I would be honoured if posterity should decide to term this thought experiment "Myles Polar Bear". ````

Life forms do not violate the second law of thermodynamics, simply put Maxwell's Deomn obtains work without releasing heat, a perfectly efficient engine. We can not perform work without expending heat, by doing work (to catch fish) he expends heat, that is WHY he needs to eat, because witout it he will be unable to replace the energy lost by heat while he is hunting.

[edit] Why did I cut everything?

Before my edit, this article was a work in progress -- there were numerous headlines without content, and various notes for future revisions, with promises to flesh out more soon. I checked the history, and to the best I can tell, this semi-finished version has been up for months. It seems to me that an actual article, whatever it's limitations, is better than a worksheet, not least because the latter may intimidate people who have something to add, but feel that they should wait for the contributions of this other user, who has promised to return at some unspecified time.

If you are that other user, just revert my changes; your worksheet is still there in memory, and you can post your completed version when it's completed. Hope you get to it soon; I'd like to read it!

NoahB 18:41, 26 Apr 2005 (UTC)

[edit] Removed references

Nice introduction to thermodynamics, but Feynman uses the ratchet and pawl model instead. Maxwell's demon is mentioned very briefly (lecture 46, middle of page 5), so the explanation is not very clear or tight. Hardly a first choice for a reference.

  • Grayce, Christopher J., "Maxwell's Demon". Educational Applets. University of California, Irvine, Irvine, CA.

404 Not Found

How is this helpful?!

This paper didn't pass peer review, and for a good reason — it demonstrates lack of understanding of the 2nd law and the well-known solutions to the Maxwell's demon paradox. There are ample correct references in the book by Leff and Rex.

This is the first chapter from the book by Leff and Rex, and now linked as such.

[edit] some problems with article

I'm not a mathematician or a physicist, and am certainly not an expert on Maxwell's Demon. However, in browsing around the internet and looking at some other sources, I think there are some serious problems with this article. First of all, the description of the Szilard experiment is vague and doesn't seem to jibe with other accounts I've looked at. I've tried my best to tweak it, but it's hard to know if I've succeeded since, as I said, I'm not an expert. Second, I separated the Bennett experiment from the Szilard experiment, but it too needs to be expanded on and better explained. Finally, the article suggests that Maxwell's Demon has been refuted, and that the Bennett experiment is final. From poking around the web a little, I don't think that's the case...though, again, I don't really have the background to know for sure. Anyone with more knowledge than me able to respond to these points? NoahB 13:53, 3 Jun 2005 (UTC)

[edit] How about Maxwell's demon disproving the SLTD?

Currently, the first line runs:

Maxwell's demon is a character in an 1867 thought experiment by the Scottish physicist James Clerk Maxwell, meant to illustrate the second law of thermodynamics.

But shouldn't it say "...meant to disprove..." or, at least "...meant to complicate the second law ofr thermodynamics"? That's what Maxwell himself claimed it did. I'm ignorant, though, so maybe someone out there who knows more can make the change.

[edit] More References To Maxwell's Daemon

Look in Ken Kesey's Daemon Box - there's a notable regard to Maxwell's Daemon and Psycology

[edit] Oversight?

I am probably missing something, but isnt' Bennett's idea a little short-sighted? Who is to say that information storage space would eventually run out? Surely there must exist or will soon exist an information storage device large enough to catalogue the status of every molecule in a sufficiently small collection of molecules as either 'A' or 'B'.

In that case the 'entropy' of the memory will increase. Initially the memory was empty, but over time it accumulates more and more random information. On the long run this is unsustainable.Count Iblis 15:31, 10 August 2005 (UTC)
If you have a source where that objection is made, cite it and put it in the article. Otherwise it's original research, which Wikipedia usually tries to avoid. Also, you can sign your post by typing four tildes (~). NoahB 18:08, 16 Jun 2005 (UTC)

[edit] my mistake

I see what you're saying. I apologize; it's my nature to argue ;) Taurrandir 16:45, 18 Jun 2005 (UTC)

No apology necessary. My goodness, go look at Wikipedians bashing each other on Talk:Mother Theresa or Talk:Abortion or any number of other places. You raised a legitimate question and presented it politely -- can't ask for more than that. NoahB 13:05, 20 Jun 2005 (UTC)

[edit] experiment deleted

I deleted the following section from the article

Experiments

"In 2005, Xin Yong Fu and Zi Tao Fu of Shanghai Jiao Tong University posted an article that claimed to have performed a comparable experiment using a vacuum tube in a magnetic field. Briefly, two Ag-O-Cs electrodes were attached to a mica substrate in a tube with more than 100 megaohms of resistance. The tube geometry was such that when a magnetic field is present, the trajectory of thermal electrons should only be able to pass in one direction between the electrodes. The experiment was performed at room temperature within a Faraday shield. They reported currents as high as 8x10^-13A. Preparation of the tube seems to be difficult: "The leakage resistance depends mainly on the final exhausting process of the extra cesium." The effect was observed in only two tubes of an apparent series of six. The experimenters credited Yu Ping, an engineer in the Yi Zheng electronic tube factory, for assistance.

The reader will note that the citation given above contains no indication of the journal, if any, in which this study has been published, much less whether it is in a peer-refereed publication. With no supporting data, the report may be given the same credence as any other report that might be found on the Internet."

The relationship of this experiment to Maxwell's Demon is very unclear to me (though I'm no expert) -- if it is put in the article, that relationship needs to be made explicit. Secondly, the caveats expressed in the second paragraph need to be addressed; it seems to me that if they are, in fact, accurate, the article is not noteworthy and does not belong in the article. NoahB 9 July 2005 16:52 (UTC)


[edit] Entropy, Maxwell's Demon and Natural Selection

I have removed the reference to Natural Selection. It is erroneous and stems from an incorrect (although common, unfortunately) understanding of the second law and entropy that has been propagated by some critics of the theory of evolution. Maxwell's Demon was a thought experiment exploring whether it was possible to violate the second law. Natural selection and evolution do not violate the second law, first and foremost because the second law requires a closed system to apply. Secondly, in the popular press, "entropy" is equated with "disorder". This is incorrect. If someone feels that the comparison between natural selection should stay in the article, I can add an explanation of why it is incorrect although it is my first instinct to delete incorrect information instead of explaining why it is incorrect.

-The_Blue_Moose

"first and foremost because the second law requires a closed system to apply."

The universe is usually considered a closed system. This probably has nothing to do with the debate mentioned, but is important re: your point. Why? Because if the universe is not a closed system, then the second rule would not be applicable. Circular logic, but, then again, the universe by definition contains everything. I'd hate to have the baby tossed with the bath, so find a way of disproving the argument without an accidental invalidation of the law. 68.215.226.236

Both your statement "The universe is usually considered a closed system... if... not a closed system, then the second rule would not be applicable" and The_Blue_Moose's statement "the second law requires a closed system to apply" are completely erroneous. Korkscru 05:30, 13 February 2007 (UTC)

[edit] article/site with some of the same content

This page shares some identical, but helpful, text with the Maxwell demon article. It is unclear what is copied from what, as neither page credits the other. The page linked below does have a text date (albeit easily faked) that says "Updated 97-10-17". http://www.maxwellian.demon.co.uk/name.html Contact info for that site is: feedback@maxwellian.demon.co.uk

[edit] Interaction not required for Maxwell's "being"

"...a real-life Maxwell's Demon would need to have some means of measuring molecular speed, and that the act of acquiring information would require an expenditure of energy."

No means of measuring would be required if the being knew the initial state of all the subatomic components of the system and could deduce all subsequent states at any given point in time. Presuming this being understands physics a lot better than we do. :)

And no impartation of energy would be required if the being could open and close the valve without altering any of the other components of the system. Perhaps we should surmise a very sparse set of components for this, to make it seem more... realistic.

A few points -
  • Knowing the initial state does not allow anyone to deduce all subsequent states except in the classical limit. But lets assume classical physics, not quantum.
  • Its true, the energy needed can be made arbitrarily small. What set of components would you suggest?
  • Talking about a classical demon that knows the initial state is getting into undocumented territory (I think.) Do you have any references for such a demon? I'm saying this because I'm interested. It seems to me you could set up a situation where there's a hole in the wall separating two similar gases at equal pressure. Then have a big wheel with its axis perpendicular to the wall. The wheel would be mostly solid, blocking the hole, except at certain places it too would have a hole, which would effectively open a path between the two sides. Then you set the wheel in motion at a particular point in time. It would have the holes placed at just the points where the demon, knowing the initial state, has calculated a particle from one side would hit, letting that one particle through, but preventing particles from the other side from going across. Beginning at the time the wheel is set in motion, the entropy, by the usual definition, will drop, with no compensating increase anywhere else. I have no references for such an idea, so it cannot go into the article, but if you know some, I would be very interested. PAR 00:14, 16 December 2005 (UTC)


[edit] Barriers as gates

Classically, a potential barrier can extract energy from a particle and later return what it took. It can stop and reverse the direction of particles of lower energy. This makes it an interesting choice as a component of the demon. The act of passage is a measurement of the particle's energy. The need for prior knowledge about a particle is avoided. How can one prevent a particle which has cross a barrier from returning? The rate at which a particle of a particular energy interacts with a barrier is a function of concentration of the particles on each side. So this might be a means of controlling passage across a barrier. Another means might be a gate which favors motion in one direction over that in the other. Over time there would be a net movement from one side of the barrier to the other. In this case it would take two barriers to do the work of Maxwell's Demon. A charged particle's interaction with a magnetic field depends on its velocity and the force is transverse to the direction of motion so the energy of the particles would not be affected and particles moving in different directions might be separable by this means. Was Maxwell aware of the Lorentz force? Jbergquist 18:26, 7 January 2006 (UTC)

[edit] What I don't get is this...

Why is everyone ignoring the work that the demon must do in order to open and close the door? Surely it would take more work to open and close the door for each molecule, than could usefully be extracted from the redulting heat difference. The door must be bigger than a single molecule, right? So it's going to have a significant mass, so opening it will take work. What am I missing here? 143.252.80.110 15:07, 24 January 2006 (UTC)

This is EXACTLY what I was thinking. It really puzzles me as well. Can someone explain please ? Gabi - 28 February 2006

Yeah!! Wouldn't energy be used to open and close the "trapdoor?" - Markmichaelh 28 February 2006

Man! Can someone PLEASE answer ? Gabi - 18 March 2006

Well, when you open and close the gate back, the process is reversible and whatever energy Maxwell's demon needs to open the gate (ie. the work done by the demon on the gate) is regained when the door is closed. Erkcan 12:06, 7 May 2006 (UTC)
Think of two boxes, connected via a gate as depicted in the diagrams above. Say the gate was a one way gate (connected to the wall via a spring) that opened into one box only (i.e. a particle from the right box could hit the gate to open it and go into the left compartment, but not the other way round). Now before any particles have hit the gate, it is closed, the spring having just tension to eep it shut. When a particle from the right hits it, the particle slows down using some of its kinetic energy to open the gate, the particle has moved into the left hand side, the second law is broken.... No, the gate now has some internal energy (supplied by the particle that hit it), so therefore is unable to be fully closed again as the spring does not have enough tension to close it fully, making the gate bi-directional, and equilibrium is restored. Second law violation has not occured.

[edit] Care to fix up portion on Bennet

The sentence on Bennet is, um, rather weak. Also, it doesn't follow chronologically with Landauer's portion. Care to fix it anyone? Also, Leff and Rex's work should be the first and emphasized reference. It's a tour de force of the subject.


[edit] Alternate and Improved demons

Is this section necessary? Is it based on anyone's research?

IANAQP, but I've taken through intro quantum and this sounds like bull to me. I tagged it as needing citations, if it remains such for long I think it should be deleted. 24.7.106.155
I agree with your second {{citationneeded}} tag, but it seems to me that the first tag's statement is uncontroversial. So I'm going to remove that one for now, but I'm open to any reasons why it might be in need of citation. — Laura Scudder ☎ 19:40, 16 May 2006 (UTC)
This is so blatantly stupid (a classical system to control quantum mechanics?!!!...please anyone check the forth grader who contributed his ideas of an extended deamon of his, and give him some wikipedia information. And revert whatever he changed, but importantly make sure he doesn`t tinker with other articles in this manner as wellSlicky 08:34, 26 July 2006 (UTC)
This section does not have the rigor of the rest of the article. It seems to be based on an individual's musings. Unless adequate references can be supplied I vote that this section be removed. Complexica 18:46, 6 September 2006 (UTC)
This section is still around? Because of the recent article in CNN, I think I'm going to delete this section. it's unsourced and it makes no sense at all. -sthomson06 (Talk) 23:18, 1 February 2007 (UTC)
Deleted, text dumped here in case anyone freaks out about the deletion. -sthomson06 (Talk) 23:22, 1 February 2007 (UTC)

Maxwell's demon could work this way: imagine a dividing wall in which each element would function as a valve set to allow only those particles of higher velocity/energy/enthalpy through into the other chamber. This would result in accumulation against the entropic gradient, apparently contrary to the second law. Not only would the "work" involved in separating the molecules take a small amount of energy to begin with, after analyzing the location of the molecule, the theoretical "demon" would have to forget the location of the molecule, which would expend more energy than would be created by the energy-generating action of the generator. Simply put, to forget is work by definition, and would prevent the engine from producing any amount of energy.

A slower process which would work just as well as Maxwell's would be a wall in which there was a single valve: by the law of random motion every particle would at some time or another impact this valve and be "assessed" by the valve mechanism and thus either pass through or not into the second chamber. Conceptually this could be done as simply as by having a spring-loaded door: particles with greater momentum would open the door/gate/valve and others would not. In practice, at normal temperatures the dissipation of energy caused by transfer of energy from the bouncing particles to the side walls, to each other, and of course to the valve in their passage through the wall, would soon cause the whole system to lose energy and run down. Recent research suggests that this might not be the case at extremely low temperatures.[1]

[edit] Microscopic vs Macroscopic.

The whole gist of the article is that Maxwell's demon is not possible, because it is against second law of thermodynamics. But second law is derived under assumption of lots of particles, i.e. it is a Macroscopic law. I would like to have some discussion about possibility of Maxwell's demons on microscopic level where second law is not applied

The whole point of the thought experiment is to make the suggestion that the Second Law is only a macroscopic law, i.e., that it can be violated by tinkering with individual molecules. The article as currently written does not take a position on whether or not the Second Law applies at the microscopic level. I don't think it does apply at that level, by the way. Objectors say that the demon needs to expend energy, but we can always imagine that the molecules are much bigger than the demon so that the effects of the demon's work are much greater than the effects of his information-gathering. -- Jdthood 15:04, 27 July 2006 (UTC)
Exactly - opening the door only to specific particles doesn't bring the two quantities of gas into contact as assumed by the SLTd. Contact there means any particles of A can collide or mix with any particles of B. It's a macroscopic law and this refutation is basically a Straw Man.

[edit] Cultural references

I think most of the section "Maxwell's demon in popular culture" is pretty useless. It might be worthwhile to elaborate a bit on how the vivid idea of a "demon" has often inspired references outside of the scientific context, and how Maxwell's conception of a rather general "being" with unusual but ultimately worldly facilities evolved into the cute image of a tiny demon, and at what point in history that happened - I understand it was Kelvin who introduced the term "Maxwell's demon", and so on. I know, if I'd like to read that, I should just write it ;)

But as it stands, we have a long list of items along the lines of "M.d. is mentioned in this book", and rarely do we even get to know, really, of what kind the appearance is that the demon makes. For example, one item reads: Maxwell's demon is mentioned in the Novel Homo Faber by Swiss author Max Frisch, as well as in one of the short stories of The Cyberiad by Stanisław Lem: "The Sixth Sally, or How Trurl and Klaupacius Created a Demon of the Second Kind to Defeat the Pirate Pugg". - now, for one thing, these works are completely unrelated, so why are they mentioned in the same sentence? And second, I happen to know "Homo faber" quite well, and as far as I can see, the only reference to Maxwell's demon in the book is this (transl. by me):

... she thought I was lonely and wanted to be nice, didn't give up until she got me to chat with her - about navigation, radar, earth curvature, electricity, entropy, of which she had never heard. She was not stupid at all. Of those to whom I explained the so-called Maxwell's daemon, not many understood it as quickly as this young girl, whom I called Sabeth, because Elisabeth, in my opinion, is an impossible name. I liked her, but I didn't flirt with her in any way. ..." etc. etc.

How in the world does that amount to a significant occurrence of M.d. in "popular culture"? (Frisch's novels aren't everybody's idea of "pop culture" either, but that's another question.)

Other items range from the completely useless:

  • Maxwell's demon appears in $AUTHOR's $TITLE.

... to the downright silly:

  • In the manga Gundam Wing: Episode Zero, one of the Gundam engineers associates Duo Maxwell's last name with Maxwell's demon.

Should these not simply be deleted and only the helpful and significant entries, such as the band named after the demon, be kept? (Granted, there won't be much left.) --SKopp 12:52, 13 September 2006 (UTC)

It doesn't say "Maxwell's Demon in Helpful and Significant Popular Culture". Manga and the like are popular culture; something doesn't have to be of any major use to be popular. Also, there wouldn't be, for less-known things, a section that isn't popular culture. Having "Maxwell's Demon in Unpopular Culture" would be kind of dumb.
ChristopherEdwards 17:34, 4 October 2007 (UTC)

[edit] About Leo and Leon's explanation...

The demon doesn't really need to be able to measure the molecules, doeth he? From a statistical point of view, if you repeated this experiment enough times, there would eventually be an instance where a faster-than-average (or slower) molecule goes through. You can't just say, "thermodynamics works 90% of the time."

Thermodynamics was understood from experiment. In no experiment has the Second Law ever been violated measurably. However, there was never any a priori reason why the Laws should be true. Enter Statistical Mechanics. Define entropy to be -ln(w), where w is the number of states a given system can be in and still have the same temperature, pressure, and volume. Now, any transition from one system state to another with a different T, P, and V, is (this can be verified analytically, rather than experimentally) enormously more likely to increase the entropy of the system than to decrease it. Or, if you have two containers of different kinds of gas, and you connect them so that the gases can mix freely, then once they are mixed, the probability that they will ever spontaneously separate again, while non-zero, is incredibly small. The expectation value for the time before separation is much much longer than the expected lifetime of the universe. Jon Wilson 24.162.120.52 19:58, 14 September 2006 (UTC)

[edit] super vs. extremely

I changed "super-low temperatures" to "extremely low temperatures", because, especially in science, super means "more than" or "above" in addition to "extremely". The first two meanings could be used by accident, making "super-low temps" mean "a little bit warmer than low temps", which was not the intention of the article. Extremely does not carry this extra baggage, and conveys the intended meaning less ambiguously.

[edit] Uncertainty

When reading about the thought experiment, I was asking myself this question... the experiment talks about a demon (or valve, or trapdoor, or whatever you want) that will allow a molecule to pass from one side to the other when it has above-average energy. To know its energy, one would have to know its momentum. But when one knows the momentum, according to the uncertainty principle, one cannot be sure where the thing is. It may have passed through the door in the time it was open, then again, it may not. (In fact it may have even tunneled back right after the door was closed, but that's irrelevant). Moreover, in the time the hole was open, other particles may have passed through (but if you are sure they did, they may have had either higher or lower velocities than average - there's no way of telling). Anyway, finding out when to open the hole constitutes a measurement and the uncertainty principle will kick in... won't it? Bit confused here, am I missing a point here and does this have nothing to do with the problem, or is this actually a reason of the experiment being physically impossible? --CompuChip 19:54, 31 January 2007 (UTC)

[edit] Current Events

Shouldn't there be some reference to the work described in this CNN article? --TedPavlic 17:31, 4 February 2007 (UTC)

Nevermind. I missed it. I see it now. --TedPavlic 17:32, 4 February 2007 (UTC)


[edit] Leigh's nanomachine

Experimental Work Based on Maxwell's Demon

I briefly peeked on that David Leigh's nanomachine

http://www.s119716185.websitehome.co.uk/home/mdhowitworks.pdf

and it seems that description here

Leigh made a minor change to the axle so that if a light is shone on the device, the center of the axle will thicken, thus restricting the motion of the ring. It only keeps the ring from moving, however, if it is at site A. Over time, therefore, the rings will be bumped from site B to site A and get stuck there, creating an imbalance in the system. In his experiments, Leigh was able to take a pot of "billions of these devices" from 50:50 equilibrium to a 70:30 imbalance within a few minutes.

is a bit oversimplified and as such bit misleading too. What raises my hackles here, is especially this phrase

It only keeps the ring from moving, however, if it is at site A.

Such one way molecular ratchets or mechanical "diodes" which would work without expenditure of energy as explanation here implies, are not possible and does not exist.

Actual working of this nanomachine is more subtle/complicated and can't be easily explained in a couple of sentences.

Particularly, what lacks here, is that act of dethickening of the centre of axle too takes light energy.

[edit] Bands

Do we really need to mention every single band in the world that is or have been named "Maxwell's demon"?

That's a good point. Is there really a purpose to the popular culture section at all? It is just a list, with little to no explanation, that is quickly going to become longer than the rest of the article. z4ns4tsu\talk 21:33, 17 May 2007 (UTC)

[edit] Someone please explain this; why has an important part of SLTD been ignored?

If the second law of thermodynamics requires the two bodies to be “isolated from the rest of the Universe;” wouldn’t Maxwell’s demon, or any other machine/creature/system that tries to interfere or interact with the molecules of these bodies, be ignoring this requirement?

For the two bodies to be isolated (not truly isolated,) they at least must not be intentionally brought in contact with anything other than each other. Bringing in a third factor that would either physically or magically influence the two bodies, clearly violates the most basic concept of isolation. Yes, the demon has a very clever and sneaky way of making a shift in the balance, not actually touching or moving molecules, just allowing them to be distributed in a specific way, but never the less, the demon is there, isn’t it? It’s leaving a mark somehow; its presence makes a difference. Its very significant presence prevents isolation.

Is there a need to scientifically prove that Maxwell’s demon concept, or any of its applications, “requires the expenditure of energy” or “increases the entropy of a system” in order to establish its illegitimacy?

It is a thought experiment, and as a thought, it is simply illogical. Not because of the questionability of the demon; such a being may very well exist (as device or as an actual demon,) but because the idea itself revolves around active, pointed meddling, when the law it tries to associate with is concerned with how X & Y influence each other independently.

Easy enough, isn’t it? Or am I missing something?

[edit] Concerning the energy of knowledge

There is an explanation that the demon does work and consumes energy finding the velocities of the particles. Very well; but once he has acquired this information, there is nothing that prevents him from using solely this information to modify the parameters of the system in whatever way he wants.

For example: He has gathered all the information and knows that is he drops the barrier now, there it will lead to a certain state of the system; if he drops it 5 seconds later, it will lead to a completely different state of the system. What he has done is generate two different energetical states, just by a decision, doing the same amount of work in both cases.

I believe a paradox is still standing here...

Garthenius 22:10, 17 June 2007 (UTC)

[edit] Indie rock band

In the trivia section is says (last on the list) that there is NOT an indie rock band... Is this supposed to be IS or is this a joke of some sort? (i.e. the name of the band is "Not Maxwell's Demon") Micah J. Manary 07:03, 16 September 2007 (UTC)

[edit] About Earman & Norton paragraph

The final paragraph citing Earman and Norton seem to me giving too much importance to their philosophical speculation, allowing a doubt that the physical insight described earlier is somewhat invalidated. To the contrary, understanding -where- the 2nd law is not violated is the whole purpose of the exercise for a physicist, not -whether- it is not violated. I suggest to remove this paragraph or to rephrase it by making this point clear. —Preceding unsigned comment added by 85.168.209.250 (talk) 18:05, 1 January 2008 (UTC)


[edit] An oversight on Maxwell's part?

Newton's third law: "For every action there is an equal and opposite reaction"

From this law we know that when a particle hits another particle, an equal and opposite force is produced, correct? Let's assume the ENTIRE CONTAINER is a single particle, and the demon itself a part of the container. Wouldn't the reactions from the particles on the fast side and the reactions from the particles on the slow side on the container bounce the container around and cause energy to be transferred from one side to another when the particles and the container contact?

The only way the container would not bounce around is if its mass was infinity, and that number is practically unobtainable. Even the speed of light has its limits. Habvc (talk) 01:05, 9 March 2008 (UTC)

[edit] Demon?

Possibly this might be a small point, but doesn't the use of a "demon" defeat the entire purpose of the experiment? As a supernatural entity, a demon would, by definition, exist outside of nature, and therefore wouldn't be affected by natural laws, including the laws of physics. For all intents and purposes, couldn't the same argument be made for using pixie dust or unicorn's breath? The ideas of "work" and "entropy" wouldn't necessarily apply to things that aren't bound by thermodynamics in the first place. Obviously you can break the SLTD if you use magic :) JCub (talk) 17:59, 18 March 2008 (UTC)

See my note in 5. Suggestion above on the convention of using demon as a non-supernatural entity in thought experiments. We’ve gone full circle now. Myles325a (talk) 02:37, 15 April 2008 (UTC)


Could the energy used to operate the trapdoor be extracted from the energy of the colder molecules on side A therefore increasing order and also increasing the temperature difference? (by Nathan Sanders) —Preceding unsigned comment added by 81.138.15.64 (talk) 11:07, 16 April 2008 (UTC)