Talk:Maxwell's demon/Archive 1

Archive 1

Basic Flaw

I'm no rocket scientist, (I'm a brain surgeon!) but isn't Maxwell basically saying "If magic existed, it could circumvent science"? And if so, why is anyone bothering to discuss the entire "Maxwell's Demon" concept? Again, I don't claim to be a genius, but I'm pretty sure that if I said "Given a Flying Carpet, I could defy the laws of gravity", it'd be about as relevant to scientific discussion as Maxwell's Demon. I doubt anyone would spend a lot of time arguing about whether my hypothetical flying carpet could really defy gravity. 75.169.105.175 (talk) 05:19, 9 May 2009 (UTC)

I think the term "demon" is confusing. I believe the quote by Maxwell in the article makes clear that by "demon", Maxwell never intended to imply a supernatural or 'magic' entity, just an intelligent creature or machine, obeying all the laws of physics, only small enough to manipulate individual particles. --Chetvorno^TALK 14:37, 9 May 2009 (UTC)

Actually Maxwell did not come up with term "Demon," It was Lord Kelvin who did. Maxwell therefore could not have been referring to magic. — Preceding unsigned comment added by 98.174.211.60 (talk) 18:23, 13 September 2011 (UTC)

Oops, you're right. Thank you> --Chetvorno^TALK 06:16, 14 September 2011 (UTC)

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)

The flaw in Maxwells theory (and I cannot believe that all the intellectuals have missed it) is that he has two closed systems which are not closed systems as they have a door between them which gets opened. These two objects exist in a closed system containing a demon this third outer system is attempting to achieve equilibrium by transferring high energy particles from a to b and low enegry particles from b to a at some point equilibrium will be achieved when halfway through the swapping there will be an equal number of low and high energy particles in both a and b. The system will then continue off into an unbalanced state in the opposite direction to the starting point. This oscillating of energy is actually quite natural. The theory of entropy is also about a closed that left alone will attempt to achieve equilibrium if you have a demon interfering you have not disproved the theory in fact I would argue that you have proved it as it required the effort of a demon to thwart the natural tendency for the two bodies to attempt to achieve equilibrium (DarrylHas). —Preceding unsigned comment added by 69.7.138.1 (talk) 11:01, 14 August 2009 (UTC)

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)

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)

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.

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)

Tunneling both ways is not a problem. If low-energy particles stay on one side, and high-energy particles diffuse both ways, entropy is still reduced, which violates the SLTD.--128.187.80.2 (talk) 18:39, 17 October 2008 (UTC)

along similar lines

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

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.

Where did you find the requirement that Maxwell's Demon be "perfectly efficient"? I claim that no such requirement exists. A demon that expends some small amount of energy to open an shut the door could still perform all functions necessary for the experiment. Such a demon could, for example, be powered by a Heat_engine connected to the very temperature differential that it creates. As long as it consumes less energy than it produces, the system will still converge to lower entropy, which would violate SLTD. —Preceding unsigned comment added by 128.187.80.2 (talk) 18:49, 6 March 2009 (UTC)

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)

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)

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)

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)

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)

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)

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)

There is no need to "prevent a particle which has cross a barrier from returning". If high-energy particles can diffuse across the barrier (in both directions) and low-energy particles cannot, then entropy is reduced when only part of the high-energy particles move to the other side, and the second law of thermodynamics has already been violated. If you really want total separation of the particles, you can just periodically enclose and remove particles that have crossed the barrier. --76.27.96.159 (talk) 04:25, 11 September 2008 (UTC)

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.

It doesn't matter into which box the door swings. It could just as well slide open and shut parallel to the wall between the two compartments. Interactions with particles in partially-open states can also be ignored by assuming the demon is just really fast. Issues with the spring or locking doors are not important either. It doesn't even matter if the demon does use a little bit of energy to open and shut the door, as long as it uses less energy than a heat-motor could generate by undoing the process. All details such as the mass of the door or the interactions with a locking mechanism can be made negligible. The demon is the only necessary assumption in the whole thought experiment. If there is a very low-energy way to make the decision about when to open and shut a *hypothetical* door of unspecified mass and unspecified mechanism, then the thought experiment is successful. The answer is not found in some detail that Maxwell overlooked.--76.27.96.159 (talk) 01:31, 3 March 2009 (UTC)

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]

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.

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)

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)

Your quantum mechanical arguments might be an additional reason why Maxwell's Demon fails, in quantum systems, in addition to the 2nd Law. But I believe the 2nd law doesn't depend on quantum mechanics; it also works with classical particles, or particles that are big enough that the Heisenberg uncertainty in their position or momentum is negligible. In other words, any particular Demon's particle sorting process may fail through quantum uncertainty or other reasons. But even if the sorting process is successful, it fails to create a temperature difference because the Demon and his process are part of the system, and consume energy and create entropy at a greater rate than the entropy reduction due to the sorting. --Chetvorno^TALK 04:30, 18 October 2008 (UTC)

Why do you suggest that the sorting process necessarily consumes energy faster than it can produce it? (You cannot say "because SLTD says so", since it is the thing on trial here.) Making a decision consumes a constant amount of energy, while the energy gained by sorting particles depends on the energy in those particles. Suppose the particles are super-massive, or are moving really fast. These values can be increased arbitrarily, without affecting the cost of making a decision, so there must be some point at which more is gained by sorting than is lost by making the decision. I don't think anyone yet knows why Maxwell's demon fails, and all the people in this discussion who claim they know why it fails have yet to put forth a complete explanation.--128.187.80.2 (talk) 20:19, 5 May 2010 (UTC)

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)

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)

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?

I believe the phrase isolated from the rest of the universe applies to the entire system, bodies plus Demon. The Demon and his trapdoor are part of the system, and can be considered part of either body. --Chetvorno^TALK 04:39, 18 October 2008 (UTC)

With regard to your second question; yes, the key to the experiment is that the Demon and his activities require energy and create entropy at a greater rate than the entropy reduction caused by the particle sorting. --Chetvorno^TALK 04:44, 18 October 2008 (UTC)

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)

The two different decisions you describe would result in different amounts of energy difference between the two containers, thus different final temperatures. So yes, the demon, by how he manipulates the trap door, can create anywhere between no energy difference (if he doesn't open the door or opens it randomly, without regard to the particle's energy), and some maximum energy difference (if he admits to container 2 all particles moving faster than the average velocity, and admits to container 1 all particles moving slower than the average velocity) --Chetvorno^TALK 03:49, 18 October 2008 (UTC)

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)

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)

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)

Nope. It doesn't matter if a small amount of energy leaks to the other side, as long as the effect of the demon is faster than the leak for some significant period of time. The experiment doesn't have to last for eternity, just long enough to separate a significant amount of the high-energy particles from the low-energy particles. After that happens, you declare success and end the experiment. The leak could be made to be negligible with containers of much less than infinite mass. (Isn't the speed of light supposedly constant? Limits only make sense for variable things. If the speed of light is not constant, then on what do we base its limits?) --128.187.80.2 (talk) 18:29, 17 October 2008 (UTC)

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)

I believe that by Demon, Maxwell never intended to imply a supernatural or 'magic' entity, just an intelligent creature or machine, obeying all the laws of physics, only small enough to manipulate individual particles. --Chetvorno^TALK 04:56, 18 October 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)

No, that violates the 2nd Law. Energy can't be extracted from molecules at equilibrium (constant temperature) unless another reservoir of molecules at a lower temperature is available --Chetvorno^TALK 05:04, 18 October 2008 (UTC)

I am sorry, but that cannot be a valid explanation. This article is challenging the Law, so so you cannot quote the Law as the justification. This is not proper science. Luciano --124.168.201.115 (talk) 03:54, 5 February 2009 (UTC)

How would the energy be 'extracted' from the reservoir? Particle collisions exchange energy reciprocally. To have a net flow of energy you have to have a collection of particles somewhere that has less kinetic energy than the ones in the reservoir. Give a specific example. --Chetvorno^TALK 12:35, 5 February 2009 (UTC)

I am not arguing that the original statement is right or wrong. I am simply saying that the explanation that it is wrong is invalid. Luciano--124.168.194.49 (talk) 23:52, 5 February 2009 (UTC)

I think the confusing issue here is the apparent chicken-and-egg problem. If some of the temperature-differential is consumed in order to generate the energy that operates the door that creates the temperature-differential, then how does the system get started? Note, however, that there is no requirement that the system begin with both sides in equilibrium. If they start unbalanced, and the demon causes that unbalance to become amplified, then the experiment is successful, so I argue that it is perfectly valid to extract necessary energy from the temperature-differential.--128.187.80.2 (talk) 17:39, 9 April 2009 (UTC)

Those are both perfectly good, valid points; the door can be operated by the temperature differential. And Luciano's point that my 2nd Law argument was circular was correct too. Actually, Nathan's whole question about the energy needed to operate the "door" is beside the point. The reason Maxwell's demon fails is not because of the energy needed to open the "door" (which can presumably be made small), but because of the energy needed to make a decision of which particles to let through. This requires interaction with the particles that always results in no net energy going into the higher temp reservoir. This is the connection between thermodynamic entropy and information entropy. See my discussion of the spring loaded trapdoor in the section [[#Just a simple question, I hope|]] below, which is shamelessly cribbed from Feynmann's Lectures on Physics, Vol.1, Ch.46. --Chetvorno^TALK 18:54, 9 April 2009 (UTC)

Questionable source

I've removed the following source from the "Further reading" section:

Splasho.com page on "Maxwell: Thermodynamics meets the demon"

I don't believe that splasho.com meets the criteria for being a reliable source; it's essentially a personal website. I do appreciate the amount of work that the author has put into what is on that page - and certainly think that Wikipedia editors should look at the sources listed at the bottom of the page, for valid sources (to the extent they aren't already in the Wikipedia article, and that more sources are really needed) - but the page is purely original research. And yes, it has advertising. -- John Broughton (♫♫) 19:10, 27 June 2008 (UTC)

It is a technical article directly related to the topic. Please do not remove external links to technical text until you can provide either mathematical or experimental proof that the article is in error, and please be specific. Thanks. --PaulLowrance (talk) 20:27, 27 June 2008 (UTC)

Did you just make up a policy? There is no Wikipedia policy or guideline that I'm aware of that says that "external links to technical text" should not be removed unless an editor can "provide either mathematical or experimental proof that the article is in error". Where did that come from? -- John Broughton (♫♫) 19:54, 28 June 2008 (UTC)

The technical article is related to the topic. I see no valid reason why links to such technical articles should be removed regardless if such articles support or challenge present conventional physics. As far as I aware, science is about sharing of information while being open to new mathematical derivations, experiments, theories, and interpretations. Thanks --PaulLowrance (talk) 04:34, 29 June 2008 (UTC)

Yes, the article can be linked from the "see-also" section in the way it is done now. We make clear that it is a non-peer reviewed article. It cannot be used as an authoritative source. I.e. we cannot make some statement in the wiki article and then give a ref to this non-peer reviewed article. If we wanted to do that we would have to say that the author of the article has made a some claim in a non-peer reviewed article and then give the ref. But then, since the opinions of that author are not very notable, that would not be appropriate.

Now, I've looked at the article, and it does look ok. to me. Also, the article itself cites references, so if that article were itself a wiki article there wouldn't be much wrong with it. That's why I don't object to include it in the see-also section in the way it is done now. Count Iblis (talk) 13:25, 29 June 2008 (UTC)

Peer review is a necessary part of science, but lets not forget the issues and limitations with such peer reviews. Please see Criticisms of peer review and Peer review failures --PaulLowrance (talk) 15:37, 29 June 2008 (UTC)

Ok, but then the fact that innocent people get jailed from time to time due to prosecutorial abuse, juries who vote "guilty" even if there is reasonable doubt or for other reasons, is not really a strong argument for having trials without a prosecutor, jury and judge. Count Iblis (talk) 15:49, 29 June 2008 (UTC)

Did you read the WP references? Peer review abuse is one of many complaints in the science community. One reason to consider all work regardless if it's yet to be peer reviewed is the complaint that peer review can often take considerable time for various reasons. Big claims can take decades to receive full recognition. One reason is due to firm beliefs in present conventional science where a reviewer may falsely believe conventional science is flawless. Simply because a law *appears* to have stood the test of time is no indication such a law is flawless. For example I have observed that a high percentage of physicists falsely believe 2LoT functions flawlessly at a microscopic scale when in fact it is a theory of macroscopic systems at equilibrium, not 100% accurate a microscopic scale. See Microscopic 2LoT. --PaulLowrance (talk) 16:46, 29 June 2008 (UTC)

(undent) As far as I aware, science is about sharing of information while being open to new mathematical derivations, experiments, theories, and interpretations. This is Wikipedia. There are specific rules here about what can and cannot be included in articles, and how to include that - policy like WP:V and guidelines like WP:RS, WP:EL, and WP:GTL. If you disagree with those rules, you're welcome to try to change them. And until they change, you need to abide by them; if you don't want to, you're welcome to post elsewhere on the Web. -- John Broughton (♫♫) 00:41, 30 June 2008 (UTC)

You are pointing the finger at me when you are the one who removed valid links. I merely undid your link removals. There's no need to preach at me. Just please be more considerate to others who abide by WP rules. Thanks. --PaulLowrance (talk) 01:06, 30 June 2008 (UTC)

I suggest you recheck the article history. I removed exactly one link that is now back in the article - and that is a questionable/disputed link - to splasho.com, currently at the top of the "Further reading section. Please either retract your statement about my removing multiple "valid" links, or point out specifically what link removals I made that you "undid", other than the splasho.com link.

I am leaving the splasho.com link in place, for the moment, to see what other editors think. I continue to believe that it is wrong, per the policies and guidelines cited, but I'll defer to other editors on the matter. -- John Broughton (♫♫) 23:18, 30 June 2008 (UTC)

Please check the history so you can see that you removed two links. The second link was a broken link, which I fixed with ease by replacing it with an archive.org link. As far as the other link that you removed, please read "Count Iblis" comment where he believes the article seems fine after reading through it and noted the references contained in the article. There's no question the article is valid. Please be specific as to why you believe this article is invalid. —Preceding unsigned comment added by PaulLowrance (talk • contribs) 23:33, 30 June 2008 (UTC)

My two cents: Wikipedia does not allow original research means research made by you, the user, not research made by others who are not you. Peer review is not a criteria for inclusion, although it is usually preferred. That being said, Wikipedia is not the grounds for pushing an idea that goes against currently accepted science. This is an encyclopedia after all, not a scientific journal. The Splasho website seems neutral in its treatments, nor do it seems to be pushing for ideas that are not commonly held in science (glanced over it for a minute). I don't see any reason to discard it. Headbomb {ταλκ – WP Physics: PotW} 07:16, 1 July 2008 (UTC)

The problem is that the policy WP:V says Articles should rely on reliable, third-party published sources with a reputation for fact-checking and accuracy. What we have here is a source that doesn't meet this criteria - the site splasho.com has no such "reputation". It's not our role, as individual editors, to evaluate neutrality and accuracy of a webpage that is self-published, by an unidentified author, and decide, based on our evaluations, whether it seems okay. 17:04, 1 July 2008 (UTC)

If it's not your role to evaluate the neutrality of a source, then you can't argue that the source shouldn't be used because it's "not neutral". Splasho's website doesn't contradict anything out there, gives a good overview of Maxwell's demon, and backs up its claims by using references. I don't care that Splasho's website is not "recognized" as being the "authority on thermodynamics", the page is accurate, of high-quality, in line with mainstream science, gives a different take on Maxwell's demons, properly referenced, and covers things appropriately. To remove it from this page would deprive readers of an easily accessible (a lot more accessible than going to the library and finding a dusty old book) and very useful resource. Headbomb {ταλκ – WP Physics: PotW} 17:45, 1 July 2008 (UTC)

It's my understanding that "reputation" is a point of view based on a community, or at least that is my interpretation of the unabridged dictionary. The article includes fact-checking, references, and so far appears accurate. So far three people agree the link is valid. IMO removal of the link falls under the category of suppressing valid information and is taking WP rules beyond reasonable interpretation. --PaulLowrance (talk) 17:53, 1 July 2008 (UTC)

(undent) There are a number of ways (particularly RfC and 3rd opinion) for getting more of the Wikipedia community involved in a question like this one, as opposed to simply the editors who happen to stop by, or have this article on their watchlist. But again, because I don't think the matter is that important, I'm going to leave the link in and see if others object to it. -- John Broughton (♫♫) 00:50, 2 July 2008 (UTC)

Link submission requests

Would it be appropriate to add this section for questionable links? Lets please consider the nature of this topic, Maxwell's demon, which is still under debate in the science community. I believe links to articles containing acceptable mathematics, experiments, or references to such data is acceptable. --PaulLowrance (talk) 16:12, 1 July 2008 (UTC)

I believe this link is valid for this WP article. Coupled Brownian Motors, by Prof. Ryoichi Kawai in collaboration with Peter Reimann, Christian Van den Broeck, and Peter Hänggi:

http://www.kawai.phy.uab.edu/research/motor/ —Preceding unsigned comment added by PaulLowrance (talk • contribs) 16:49, 1 July 2008 (UTC)

By the way, the source doesn't say that Brownian motors can extract energy from a system at equilibrium, which would be a Maxwell's Demon in violation of SLoT, only that they can extract energy from nonequilibrium systems, as macroscopic engines do. --Chetvorno^TALK 05:16, 18 October 2008 (UTC)

Hi. I've spent years researching the concept of equilibrium. The concept is fictitious. Conventional physics accepts that LoT is a macroscopic system dealing with averages, not microscopic-- ref. Second_law_of_thermodynamics#Microscopic_systems. At a microscopic scale it is a violent world with constant temperature gradients. There is no scale at which such natural temperature gradients suddenly vanish. Take a common magnifying glass, say 10X, sprinkle some pollen particles on the surface of the water and you will see Brownian motion, a macro scale event. Actually it's possible to see Brownian motion with the naked eye with some patience. Equilibrium requires infinite insulation to say the least. Even with infinite insulation there's no equilibrium due to flicker noise, which is now believed to be a Universal effect. Presently I am researching ultra high frequency ZBD's (zero bias diodes; e.g., SMS7630) that are producing a consistent DC voltage within extensive shielding. --PaulLowrance (talk) 13:26, 18 October 2008 (UTC)

A simple question, I hope

If the demon were replaced with some odd ball semi-permeable membrane which allowed high temperature molecules to go one way, and lower ones the other violate the law, since a membrane just sits there, using no energy? I realize this membrane does not currently exist but humor me. (user Tackleberry61) —Preceding unsigned comment added by Tackleberry61 (talk • contribs) 20:44, 16 November 2008 (UTC)

In that case, every pore of that membrane would act as Maxwell's demon.Avocadohead (talk) 11:53, 26 November 2008 (UTC)

The membrane couldn't act like ordinary membranes, which are symmetrical and have the same properties for particles striking them from both directions; it would have to be asymmetrical, and act like it had tiny trapdoors in it that only opened one way. The idea of a passive trapdoor as a Maxwell's demon has already been analyzed; it doesn't work. Consider a one-way trapdoor between the two chambers, held closed by a spring. When a molecule from the low temp side hits the trapdoor with enough energy, it would push the door open against the spring and pass through. The problem with this is that the molecule does work against the spring in opening the door, leaving it with less energy once it gets through. The spring has to be stiff enough to keep molecules with less that the average high temp energy out, so that is the minimum amount of energy that is transferred to the trapdoor. If the energy transferred to the trapdoor could be kept in the high temp chamber, it would work, but it can't be, because the door is exposed to both chambers. If the door is perfectly elastic, it will keep the energy it receives from the particles, banging open and closed. It will spend most of it's time open, so low energy particles can slip through to the high energy chamber. In addition since it is receiving collisions from particles on both sides while it is open, it acts to transfer momentum from the particles on one side to the other. Since the particles on the high temp side have more energy and momentum, the net energy transfer will be from the high to the low temp chamber. The trapdoor becomes a heat leak.

The microscopic potential barriers of any passive filter would act analogously to the spring in the trapdoor, transferring energy both ways. --Chetvorno^TALK 13:33, 26 November 2008 (UTC)

A simple proposal (I think)

Firstly it seems that Maxwell did not pay any consideration to the business about whether the demon is a real physical object and whether any energy is required to observe and allow certain particles through. Instead it seems that Maxwell assumed the demon had no physical properties, but had the "magic" ability to alter the behaviour of a physical world. It seems a contradiction, much like whether ghosts can move objects. So, consequently it appears the whole experiment is flawed, and no conclusion can be derived.

However, let me propose another model that is possibly much easier to analyse. I don't know what the conclusion will be, but I would love to see concurrence of mathematicians and physicists where it clearly can be seen that the explanation is rigorous and does not have unexplained assumptions, ambiguities, circularities, etc.

Take this example, say you have the box, separated into two equal partitions, but with one hole in it, such that the shape of the hole is like a funnel. Let the narrow part of the funnel be just a little larger than the size of the particle. Let the other side of the funnel be much larger. Now let's put exactly one particle in each partition with exactly the same energy each. What is the probability of each particle traversing to the other side? When two particles happen to be on the same side (I hope no one is going to say it cannot happen), what is the probability of a particle going back to the other side? Once we know this for each side, can someone calculate whether a particular side is likely to have 2 particles for a longer period than the other side?

If the result is the same for both sides then we haven't disproved the law. If one side has a higher probability, we have, because now someone can design a machine that will do work as the pressure on one side on average is higher than that on the other.

Perhaps an adventurous person can do the calculation for a large volume of particles.--124.168.216.16 (talk) 11:39, 4 February 2009 (UTC) Luciano

I believe, from the quote in the article, by "demon" Maxwell intended to mean not a supernatural being but simply an intelligent creature small enough to manipulate individual molecules, but I certainly agree that he didn't consider the energy the demon would expend in observing and filtering the molecules.

About your funnel example: in a container of gas of any shape, if you partition it into two volumes with a dividing surface, the probability of a gas molecule crossing the surface in one direction is equal to the probability of a molecule crossing it in the other direction. So your container will have equal gas pressures, densities, and velocity distributions (temperatures) on both sides, even if there are only two particles, as in your example. The probability of finding a particle in each volume is simply proportional to the ratio of the volumes. For a Maxwell's Demon, there must be some sort of energy-selective mechanism. Cheers. --Chetvorno^TALK 13:47, 4 February 2009 (UTC)

Thank you. I would agree wholeheartedly if the size of the hole is considerably larger than the size of the molecule. However if it is only slightly larger would not one expect other effects? For example if all the possible trajectories of a molecule are equally likely, then only a very small proportion would lead to the molecule entering the hole from the small end of the funnel. However, on the other side I would expect a much larger proportion, as the funnel would "guide" the molecule into the hole. So I am expecting the probability of a molecule to traverse to the other side to be much smaller at the narrow end of the funnel.

In a sense one could do a comparison with liquids in capillaries. Liquids there behave very differently from liquids in larger tubes. So, is it not possible that a gas in a similar circumstance would behave differently from gases in larger containes? Cheers Luciano--124.168.201.115 (talk) 03:45, 5 February 2009 (UTC)

Greenhouse effect

Isn't the greenhouse effect a particular kind of Maxwell's demon? It could be a perfect illustrating example. --pgimeno (talk) 09:08, 4 June 2009 (UTC)

It's an interesting idea, but the Earth's atmosphere isn't a Maxwell's Demon because it is not in equilibrium. If the Sun didn't exist, and the Earth's atmosphere spontaneously got warmer, it would be a Maxwell's Demon in violation of SLoT. But the Earth constantly receives a flux of energy from the Sun, so it is not in thermodynamic equilibrium. Global warming is just a redistribution in the heat transfer within the atmosphere. --Chetvorno^TALK 19:04, 4 June 2009 (UTC)

Information Theory and Deletion

One important thing about this thought experiment was realized in an information-theoretic context. If this temperature difference was ever to be "used," i.e, if the trapdoor was opened, then the state of the vessel where hot and cold particles were separated would be erased. Erasure of the state constitutes an increase in entropy. This is where the "missing" entropy goes. Slachterman (talk) 22:37, 4 July 2009 (UTC)

What Maxwell intended for his thought experiment

I have reverted the introduction, regarding Maxwell's reason for inventing the concept, because it contradicts what Maxwell wrote.

Maxwell stated bluntly, in an undated letter to Peter Guthrie Tait, "Concerning Demons: ... 3. What was their chief end? To show that the 2nd Law of Thermodynamics has only a statistical certainty."

Maxwell's letters and his book, in which he introduced the concept, are consistent with this claim.

For example, in his 1867 to Tait, Maxwell wrote:

If I can help you in any way with your book I shall be glad, as any contributions I could make to that study are in the way of altering the point of view here and there for clearness or variety and picking holes here and there to ensure strength and stability.

As for instance I think you might make something of the theory of absolute scale of temperature by reasoning pretty loud about it and paying it due honour, at its entrance. To pick a hole -- say in the 2nd law of thermodynamics that if two things are in contact the hotter cannot take heat from the colder without external agency.

Now let A & B be two vessels divided by a diaphragm....

[The action of Maxwell's demon is described.]

...the hot system has hot hotter and the cold system colder and yet no work has been done, only the intelligence of a very observant and neat-fingered being has been employed.

Or in short if the heat is the motion of finite portions of matter and if we can apply tools to such portions of matter as to deal with them separately, then we can take advantage of the different motion of different proportions to restore a uniform hot system to unequal temperatures or to motions of large masses.

Only we can't, not being clever enough.

Maxwell is suggesting that Tait may improve his presentation of the 2nd law by including a thought experiment, imagining what it would take to violate it -- thereby imparting a deeper appreciation of the (statistical) nature of the law.

In his 1870 letter to Lord Rayleigh, Maxwell again described the thought experiment, and concludes:

"Moral The 2nd law of thermodynamics has the same degree of truth as the statement that if you throw a tumberful of water into the sea you cannot get the same tumberful of water out again."

And in his book, Theory of Heat, immediately following the quote that's presently in the article, Maxwell continues:

This is only one of the instances in which conclusions which we have drawn from our experience of bodies consisting of an immense number of molecules may be found not to be applicable to the more delicate observations and experiments which we may suppose made by one who can perceive and handle the individual molecules which we deal with only in large masses, In dealing with masses of matter, while we do not perceive the individual molecules, we are compelled to adopt what I have described as the statistical method of calculations, and to abandon the strict dynamical method, in which we follow every motion by the calculus.

Nowhere does Maxwell say that he intended to encourage questions about the possibility of violating the 2nd law.

J-Wiki (talk) 15:57, 29 August 2009 (UTC)

Maxwell was trying to show that there exist devices that violate the second law. You must agree with that. If I say "well, energy usually seems to be conserved, but if I build this particular arrangement of magnets then I can get a consistent energy output without depleting any fuel", I'm saying that the law of conservation of energy doesn't hold for this device. I'm challenging the validity of that law of physics. I'm saying that it's inconsistent with the laws I invoked in support of the statement that the device will work. I could say I had shown that the law has "only statistical validity", but it would be a strange thing to say because the only statistical part of it is the uncertainty of human action. Anyone who wants to violate energy conservation can build one of my devices and do it.

Whether or not Maxwell used the phrase "only statistical certainty", it's clear that he was challenging the validity of the second law. The problem is that there's another, entirely different, sense in which the second law is only statistically true, and if we use the phrase "only statistical certainty" it sounds like we are talking about that, which we aren't. There is no reason to use Maxwell's exact words here. We aren't quoting him, we are saying what he did, and what he did is challenge the second law with a thought experiment. -- BenRG (talk) 16:57, 29 August 2009 (UTC)

I agree that it would be better not to use of Maxwell's exact words, mostly for reasons of style. However, what is stated about his intentions in introducing the concept must properly represent what he wrote on the subject. Until it is agreed exactly what his words meant, and how to represent them accurately, I suggest that the direct quote remain.

Upon re-reading what Maxwell wrote in his letters, and in his book, it seems he was saying that the 2nd law is applicable only when masses of molecules must be dealt with, and that it doesn't apply if molecules can be dealt with individually. He was attempting to help clarify its definition by showing its (supposed) limitations -- "picking a hole" in it, a place where it doesn't apply, not to violate it per se. To say that he intended to encourage questions about the possibility of violating the 2nd law is to project back to his time our modern understanding. J-Wiki (talk) 21:37, 30 August 2009 (UTC)

My Suggestion for Solution

The article on entropy says it is the the same as losing information. In the thought experiment here, the Demon has an unbounded amount of information (the location and direction of all particles, for all time, moving forward). The extra information, or order, is where the unusual result comes from, not from any violation of the 2nd law. —Preceding unsigned comment added by JoshNarins (talk • contribs) 21:27, 27 March 2010 (UTC)

Thanks, but our job is not to find solutions. If you can find sources making this claim, by all means add it. Regrettably, we're not at liberty to regard you as a reliable source. ;) Paradoctor (talk) 22:14, 27 March 2010 (UTC)

To address JoshNarins point: the extra information IS the violation of the Second Law. The Second Law says the entropy of an isolated system cannot decrease, which means that the net "information" about a system cannot increase due to any activities within the system. The Demon doesn't "have" the information on the positions and velocities of the particles; he must get it. The activities of the demon in measuring the velocities of the particles to choose which go through the trapdoor consume energy. This produces enough entropy (disorder, loss of information), at some other point in the system, to compensate for the increased order due to the demon's sorting operations. --Chetvorno^TALK 22:59, 27 March 2010 (UTC)

WP:NOTFORUM, you can discuss this at one of your talkpages. BTW, your explanation is wrong. Paradoctor (talk) 23:37, 27 March 2010 (UTC)

Violation of 2nd law Measured

There is a paper called "Experimental Demonstration of Violations of the Second Law of Thermodynamics for Small Systems and Short Time Scales" by G. M. Wang, E. M. Sevick, Emil Mittag, Debra J. Searles and Denis J. Evans1 DOI: 10.1103/PhysRevLett.89.050601 They observed Entropy losses in mikroscopoc systems.

I dont think the measurment argument repairing the violation in valid as you dont need to measure the energies. If you just close the gate randomly it will be very unlkikely that you split energies equaly. (213.47.167.58 (talk) 18:38, 18 April 2010 (UTC))

Misleading introduction?

Is anyone bothered by the introduction? It seems to imply that the 2nd Law can be violated: "The thought experiment demonstrates Maxwell's point by describing how to violate the 2nd Law." I believe the consensus in physics is that a Maxwell's demon is impossible. Shouldn't this be made clear in the intro? --Chetvorno^TALK 14:06, 27 June 2010 (UTC)

I agree, but then this is what you get when you give a historical introducton to this topic. We could rewrite it, giving the modern correct perspective first. Count Iblis (talk) 14:17, 27 June 2010 (UTC)

The point of the demon is to give an argument against the second law. If you wish to state that there is a broad consensus that a demon is impossible, please source the statement to a reliable source. Otherwise I have no problem with the proposed change. Paradoctor (talk) 17:33, 27 June 2010 (UTC)

Maxwell was right!

The introduction to this article says Maxwell's objective in this thought experiment was to "show that the 2nd Law of Thermodynamics has only a statistical certainty." If that is the case, he was absolutely correct. Consider the objections to the thought experiment. They claim that the demon would increase the entropy of the universe through his efforts in collecting information about the particle speeds. Now any set of information can be stored as a set of zeroes and ones. So instead of actually going out and measuring particle speeds, our demon could just flip a coin N times, and then assume that this binary sequence gives the correct information about particle speeds. Of course, by far the most likely outcome of this will be that the binary string will give him incorrect information, and thus he won't achieve his goal of achieving a temperature difference between the two boxes. However once in every 2^N times, he will get lucky and get a binary sequence which will allow him to achieve a temperature difference, thus decreasing the total entropy of the universe. Of course, the probability that entropy will increase in any given trial is far greater than the probability that it will decrease, but the point is, as Maxwell said, "that the 2nd Law of Thermodynamics has only a statistical certainty." For further evidence that Maxwell's insight was correct, see this article: http://en.wikipedia.org/wiki/Fluctuation_theorem —Preceding unsigned comment added by 69.248.123.28 (talk) 13:38, 9 August 2010 (UTC)

This doesn't have anything to do with the Demon and is simply a property of statistical systems of particles. Even without any Demon or trapdoor, if you mentally partition a container of gas into two parts, there will be some probability that at any given instant the particles in the left part will all be moving significantly faster than the particles in the right part, and so have a higher "temperature". The point of the fluctuation theorem is that as the number of particles in the container is increased, the time of such a momentary departure from thermodynamic equilibrium decreases exponentially. An ordinary-sized container has of the order of 10²⁶ particles, and the probability of finding a macroscopic temperature difference is an exponential function of that. So your number N above is of the order of 10^10²⁶. You could run your experiment for a million times the age of the universe before the Demon got lucky.

Notwithstanding Maxwell's quote, the point of Maxwell's demon is not that there are extremely low-probability events in any system that violate the Second Law - that is generally known and accepted. The idea of Maxwell's demon it that (if it could exist) it would violate the Second Law continuously, all the time, and produce macroscopic energy without a temperature difference. --Chetvorno^TALK 22:46, 9 August 2010 (UTC)

New development:

http://www.foxnews.com/scitech/2010/11/15/scientists-convert-information-demonic-energy/ —Preceding unsigned comment added by 128.244.9.9 (talk) 18:06, 15 November 2010 (UTC)

You, don't need a demon, it is fairly easy to envision a device that turns heat directly into electricity. Maxwell was right. You can read a description of such a device at Thirm.com.

Warren Harding Thirm.com —Preceding unsigned comment added by 199.126.136.167 (talk) 01:59, 28 December 2010 (UTC)

Acknowledgement of Image

The second image looks like it may have been create for the purpose of this article, but the first one looks like it was scan from a book. Shouldn't its source be acknowledged?

(I mean the image that carries the following as a legend: "Lord Kelvin, the first to use the word "demon" for Maxwell's concept, implied that he intended the mediating, rather than malevolent, meaning of the word.") 212.219.21.109 (talk) 12:28, 7 January 2011 (UTC)

From looking at the file page, it looks like the image is in public domain because it's from a very old book. The creator of the image is unknown and can't be credited. At any rate, image captions are not used to credit the image creator unless it's relevant to the article; the image page serves as the credit. --Danger (talk) 14:24, 7 January 2011 (UTC)

True. However, it is also true that the fact that an image "looks like [it] is in the public domain" due to the fact that it might be "from a very old book" proves nothing, and it is merely speculation. Likewise, we don't know for sure that "the creator of the image is unknown and can't be credited"; we simply know that his/her identity is unknown to us at this very point in time and, thus, we have failed to credit him/her. I do not mean to be a stick-in-the-mud, but they are not the same thing. Surely the person who uploaded the image in the first place is aware of its provenance. And acknowledging a source always trumps not acknowledging it.81.99.193.239 (talk) 12:45, 15 January 2011 (UTC)

Well the uploader lists a source, but it is in Cyrillic text that I can't read. They also list the creator as "unknown", so that is how we know that the creator is unknown. At any rate, the credit will not appear in the article text because the file page serves as that attribution unless the creator of the image is directly relevant to the text. --Danger (talk) 13:47, 15 January 2011 (UTC)

According to Google Translate, the source listed in Cyrillic by the uploader reads "Scan from the book." Not the most accurate of acknowledgements. Sloppy at best. And let me reiterate that its author is not unknown. The author of Sir Gawain and the Green Knight is unknown. The author of this image is merely unlisted, as he/she has gone unacknowledged. That's all. For all we know the book the image comes from is not in fact that old, and its use could represent copyright infringement. Even if the acknowledgement does not belong in the article, it is absolutely necessary in the credits of the image page. 149.5.64.141 (talk) 14:22, 24 January 2011 (UTC)

see also vortex tube?

Removed references

• Feynman, Richard P., Feynman Lectures on Computation (Perseus: 1996). ISBN 0201489910.

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

• First International Conference on Quantum Limits to the Second Law of Thermodynamics, San Diego, CA, July 2002

How is this helpful?!

• Xin Yong Fu and Zi Tao Fu, "Realization of Maxwell's Hypothesis : An Experiment Against the Second Law of Thermodynamics". Shanghai Jiao Tong University, Shanghai, P.R. China (arXiv.org is a site for pre-prints of scientific papers; but this paper is not endorsed or sponsored under arXiv's procedures by any known scientist including the authors.)

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.

• Vlatko Vedral's Lengthy discussion of Maxwell's Demon (PDF)

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

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.

More References To Maxwell's Daemon

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

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)

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

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.

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.

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.

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)

I don't know why that section failed to mention the Anime "El Cazador de la Bruja". —Preceding unsigned comment added by 69.131.23.208 (talk) 19:48, 22 May 2009 (UTC)

This section bothers me the most. It's a mess. My first urge wasn't to delete (I'm an inclusionist- storage is cheap) it was to sort it into Books, Music, Movies, TV. Then it will look like a list and someone will say 'WP is not for lists!' Shall I do it? Connectionfailure (talk) 01:05, 4 June 2011 (UTC)

Did Maxwell wonder about the evolution of "informed" systems?

Maxwell's scientific intent in posing this puzzle was mainly about statistical dynamics, but did he also wonder about agents of change? At the time, he may have wondered about evolved biologic systems such as photosynthesis, respiration, and digestion which function so reliably that they seem "informed". (Contemporary biology holds that living systems do not violate the second law, but still puzzles over the nature of information in DNA molecules and in systems.) An "informed" demon running cellular respiration is not so far fetched when one considers the organization of respiratory enzymes and atp synthase machines. We now know that these systems were naturally selected because they worked so well that reproduction of their host organisms was made more successful. But was their work due to the information in the dna that produced these enzymes?

Darwin's ideas had just been published (and have recently been verified with scientific certainty through subsequent fossil evidence and genomics), and Maxwell would have known about natural selection as an historical agency of evolution. Did he also wonder about the natural selection process as a change agent that leads to the selection of information systems? Can science respond to the puzzles of selection and information as apparent agencies, or must this be relegated to either rational or mystical philosophy?

I would rather see science or the philosophy of science give this a try, so here goes. Electron transport enzymes and atp synthase are formations of matter/energy deemed by our brains as valuable because we find meaning in the workings. It then becomes tempting to call these "informed" systems. When the thermodynamics is investigated, the second law is found to be intact, and therefore these valued selections or "informed" system end up being, not agencies, but merely intriguing after-thoughts of our brains.

Maxwell's demon lives on as to tempt us to deal with it, and I bet Maxwell would be delighted. Khobler (talk) 04:06, 25 March 2011 (UTC)

I do not think so, I believe that when people stretch it out to information entropy as the main argument that is playing the loophole. I have an article which I can email you if you want the solution to the dilema. — Preceding unsigned comment added by AlanS333 (talk • contribs) 21:35, 11 November 2011 (UTC)

New content

Hello everyone, I have done some work recently on Maxwell's demon and the solution. I have published a paper on it and would like to add a short excerpt on the findings to the criticism and development section. Is everyone okay with it? — Preceding unsigned comment added by AlanS333 (talk • contribs)

Not as long as it sourced to Facebook. Wknight94 ^talk 19:27, 11 November 2011 (UTC)

Thanks for bringing this to the talk page, Alan. I'm afraid I agree with Wknight; in order to be included, material needs to be sourced to reliable sources as defined by Wikipedia. In this case, it would need to be from a peer reviewed journal or equivalent. I don't believe that's the case, right? --Floquenbeam (talk) 19:32, 11 November 2011 (UTC)

No it is not in a magazine or in any other corporate owned publishing because I am mainly against corporations, they put profit first above anything at the expense of values. It is however a scientific paper which has been published. It was sourced to the face book notes so that people could see the actual paper if they need it to, but I guess i could put that in further reading or elsewhere or not at all if it causes such pandemonium. However on my last edit i took the facebook part out and just left it as the paper itself. — Preceding unsigned comment added by AlanS333 (talk • contribs) 20:44, 11 November 2011 (UTC)

Where was it published. If it was published in something that is peer reviewed, even if it is an amateur publication, it can sometimes be valid (I'm thinking for example the Society of Ancients publication, Slingshot. Yes these are all amateur submissions, but my G-d, if you mix your Parthians up with your Sassenids, heaven help you, because these guys won't. Elen of the Roads (talk) 21:19, 11 November 2011 (UTC)

Independent Publishing. — Preceding unsigned comment added by AlanS333 (talk • contribs) 23:28, 11 November 2011 (UTC)

That sounds like it was self-published. So, it is not okay for Wikipedia. (Self-published sources can be used on Wikipedia in only a few special circumstances). Cardamon (talk) 00:39, 12 November 2011 (UTC)

Self publishing is as verifiable as other types of publishings. Just because you aren't paying a company does not mean is less verifiable. what are the special circumstances? — Preceding unsigned comment added by 68.226.103.44 (talk) 04:28, 12 November 2011 (UTC)

Have you looked at the link to reliable sources that Floquenbeam provided you? Also, maybe you could read more about Wikipedia means by verifiabilty. It doesn’t mean that we can verify that it got posted it on Facebook yesterday. As an aside, some journals are published by non-profits and some journals do not have page charges. The circumstances when a self-published source can be used are mentioned in the link in my previous reply. Roughly, one of them is when a recognized expert in a field is taking about his or her own field, and saying something not too surprising, and we can’t find a better source for what they are saying. The others are as a source for information about the author(s) that there is no real reason to doubt (one example might be a notable person saying in their blog what their birthday is), or as a source for the actual words of the author(s). Yet another requirement is notability. Generally, something posted recently on Facebook would not qualify. Cardamon (talk) 07:01, 12 November 2011 (UTC)

Bennett's reasoning

The article implies that the demon works if he has an unlimited memory. I read Bennett's papers and it does seem to be what he says, explaining himself by saying an unlimited memory means it's not cyclical and therefore does not violate Carnot cycle rules. In order to store the memory bit, Landauer's minimal energy limit for a bit, E=kT*ln(2), it has to be extracted from the molecule's energy E=3/2*kT to change the state of the memory, but Bennett does not mention this.

If Bennett is right, then why not use a reversible AND gate to detect the presence of 2 molecules coming towards the gate and then open it? Even Bennett says a reversible gate like this would cost only 1 bit. It's waiting on 2 "bits" of energy to come along before spending 1 bit to open the door. My best guess is that the memory bit necessary is the same as the bit energy needed in Feynman's ratchet and pawl or any other minimal "door", which is itself a memory bit. I mean, the minimal bit energy has a 50% chance of being in the wrong place, and if the door has to wait a lot longer for 2 molecules to be coming, then the door is going to be accidentally open a lot longer, exactly offsetting since both the gas molecule energy and the bit energy are based on thermal fluctuations. Ywaz (talk) 18:25, 22 December 2011 (UTC)

Author of ref 13 is Christensen, not Penrose and Hameroff

Sir Roger Penrose & Dr. Stuart Hameroff (2011). "Consciousness and the Universe: Quantum Physics, Evolution, Brain and Mind," Chapter XIV Does the Universe have Cosmological Memory? If so does this imply Cosmic Consciousness?. Oxford University Press.

Maybe I don't understand reference rules, but usually, isn't the author noted and not the editors? From the article, you can see that the author is Walter J. Christensen Jr.; from Volume XIV, you can see that the editors are Penrose and Hameroff. The referencing rules seem to be fairly clear?

On second look, that journal seems to have nothing to do with Oxford University Press.

--Labreuer (talk) 19:47, 5 October 2012 (UTC)

Christopher Stasheff's "A Wizard in Rhyme" series has the demon as a character

http://en.wikipedia.org/wiki/A_Wizard_in_Rhyme

Maxwell's Demon is introduced as a character in this fantasy series by Christopher Stasheff, the demon is quick to distance himself from the Hell crowd, and aids the eponymous wizard (Matthew Mantrell) by increasing entropy in various ways to combat an opposing army. One such way is by decreasing the amount of time metal takes to rust, so that the army's armor and weapons fall apart overnight. 68.97.202.187 (talk) 15:12, 7 September 2013 (UTC)

Ratchet

A molecular ratchet that converts shaftwork into usable chemical energy might be designed. Or at least a brownian-motion-sized nano-ratchet that stores random collisions as nano-spring energy. Self winding watches convert random motions into usable macroscopic spring energy - which could be released by driving a generator to electrolyze water, and the spring become available for further recharge. None of these schemes require information storage. Sillybilly (talk) 14:16, 22 February 2014 (UTC)

Interestingly, genetic life on Earth has not come up with shaft and continuously rotating structures. An owl's head can turn 360° but not continuously in the same direction. A dog's tail can rotate, and I can do stirring motions with my hands, but at the joints there is no shaft-like continuously rotating slip. So life on Earth may have never come up with a ratcheting "enzyme" that would violate the 2nd law, something that nevertheless may be possible to create. Sillybilly (talk) 14:16, 22 February 2014 (UTC)

A Maxwell's Demon converts thermal energy, the random motion of molecules at thermodynamic equilibrium, to work, or to a temperature difference which could be used to produce work. This is what is prohibited by the Second Law of Thermodynamics. The self-winding watch example is not a Maxwell's Demon because it merely converts one form of nonequilibrium energy, the mechanical motion of the arm, to another. The motion of the arm is not due to random thermal motion of molecules, but to an ordinary nonequilibrium energy source, the conversion of chemical energy in food. --Chetvorno^TALK 16:09, 22 February 2014 (UTC)

There are 3 different scales: molecules, brownian motion, and watch. While a 2nd-law breaking device could in theory work at the brownian motion nano-scale, it would be much less efficient than a molecular device. The principle would not work at all on the macroscopic scale - a watch lever harnessing the impact of air molecules: a macroscopic ratchet has too much inertia and lacks the finesse in the ratcheting spring(not to be confused with the energy storing spring) to activate and click from a single molecular impact - but a similar mechanism, similar device on the molecular or brownian scale might work. The brownian scale may be easier to construct with present technology, than a molecular scale device, to illustrate a principle. We have a hard enough time with taxol total synthesis of the western yew, let alone synthesizing a complex 10000 atom enzyme molecule with paddle-levers, ratchets, springs, and release of spring strain energy as a higher energy chemical compound at some absorption site created from a lower energy one. But maybe such a molecule could be constructed on the scale of 200 atoms. I'm not smart enough to tell.Sillybilly (talk) 07:51, 23 February 2014 (UTC)

That's what I was referring to. Such a microscopic device has already been considered; see Brownian ratchet. And no, it doesn't work on a small scale, either. --Chetvorno^TALK 08:25, 23 February 2014 (UTC)

Where are the reliable sources claiming that such a device is possible? Please note that arguing without sources violates WP:NOTFORUM. Paradoctor (talk) 09:21, 23 February 2014 (UTC)

Have you read the articles? They are NOT possible. Neither functioning Brownian ratchets nor Maxwell's demons. You can't violate the 2nd Law of Thermodynamics. That is what I am trying to get across to Sillybilly. Both articles are adequately sourced. --Chetvorno^TALK 14:34, 23 February 2014 (UTC)

There seems to be a misunderstanding. 1) My reply was to Sillybilly, not you. 2) I wasn't challenging the articles you linked to, I asked for sources supporting Sillybilly's claim that "a 2nd-law breaking device could in theory work", which is the proper response to unsupported claims. Discussing the subject is not proper talk page usage, and Wikipedia articles are not reliable sources. Teaching Sillybilly about the 2nd law is not our job, he can always use our reference desk. Paradoctor (talk) 15:03, 23 February 2014 (UTC)

Sorry, I misunderstood who you were addressing. On the issue of NOTFORUM, editors periodically add examples of Brownian motors and other nanotech to these articles in the mistaken impression that they can produce energy from systems at equilibrium. I just wanted to forestall such an addition. --Chetvorno^TALK 19:01, 23 February 2014 (UTC)

No problem. I understand the impulse to explain what appears obvious, but generally, best practice is to avoid subject discussions and explanations. Being partial to paradoxes, I have a pretty good idea what that leads to. Paradoctor (talk) 20:10, 23 February 2014 (UTC)

Right you are. --Chetvorno^TALK 21:40, 23 February 2014 (UTC)

I added the Brownian ratchet link to the see also list of the article. Sillybilly (talk) 14:11, 24 February 2014 (UTC)

That's a good idea, surprised it wasn't there already. BTW, I didn't mean to be abrupt in my answer above. The issue of rotational motion on a nano scale is interesting, it's just that it is not relevant to the article. --Chetvorno^TALK 17:54, 24 February 2014 (UTC)

Raizon's cooling experiments

The paragraphs in the "Experimental work" section discussing Mark Raizon's experiments in single-atom cooling describe it as "...the first realization of Maxwell's demon...". I don't have access to the Sci. Am. article, and don't know how it works. Maybe someone who does could add to the section how it fails to violate the 2nd Law (I assume it doesn't violate it) and cannot produce a temperature difference without an equal or greater expenditure of energy, and thus cannot function like Maxwell's classical Demon and produce useful energy from a system at equilibrium. --Chetvorno^TALK 04:19, 21 March 2014 (UTC)

Container has only 2 sides, so wtf?

Quote:

...a container of gas molecules at equilibrium is divided into two parts by an insulated wall, with a door that can be opened and closed by what came to be called "Maxwell's demon". The demon opens the door to allow only the faster than average molecules to flow through to a favored side of the chamber, and only the slower than average molecules to the other side, causing the favored side to gradually heat up while the other side cools down, thus decreasing entropy

This does not make sense. I corrected it to be consistent with the prefatory note that there are only TWO parts (compartments) to the container. If gas is not being introduced from a third compartment, then all the gas is initially in one part of that container. Thus, if the Demon allows the faster gas molecules to move to the other side, then the slower ones stay on the side that they were in originally. I don't know how Maxwell explained this in his text, but if he said anything like what is said here, then he simply made a semantic error. As it is, the description will simply confuse people who read it. I will allow some time for comments and, unless someone can come up with a rational defence, I will change it back. Myles325a (talk) 03:54, 9 April 2014 (UTC)

If the demon only allows molecules from one side through to the other, then the number of molecules and thus the pressure will gradually rise on the destination side, an irrelevant complication he apparently wished to avoid. By allowing the same number of slower-moving molecules through from side B to side A as the number of faster-moving molecules allowed from side A to side B, the density on the two sides can be kept equal while the temperature diverges. --Chetvorno^TALK 04:24, 9 April 2014 (UTC)

Ah, ok, now I understand. I had thought that all the gas was on just one side, but it is on both sides, and there is two-way traffic with the energetic molecules moving one way, and the slow ones the other way. If the text included the Side A and Side B terms that you invoke here, then it would be clearer. But my work is finished here. I'll let someone else to it. Myles325a (talk) 05:38, 9 April 2014 (UTC)

I'm afraid you're both mistaken.

Myles325a: In "a container of gas molecules at equilibrium", molecules are evenly distributed everywhere. If this container is then "divided into two parts", we will have molecules on both sides of the "insulated wall", don't we? I don't really see how you get the idea that the gas would be all on one side of the partition, initially.

Chetvorno: Even if there is no net particle transfer, the "pressure will gradually rise" on the hot side. At constant volume and density, pressure is proportional to temperature, as in $pV=NkT$, right? Second-guessing people like Maxwell rarely works, at least for me. Paradoctor (talk) 07:53, 9 April 2014 (UTC)

I said the density would be equal, not the pressure. For that matter, the demon could violate the 2nd law without even determining the speed of the molecules by letting all the molecules through in only one direction; the pressure and temperature would rise on the destination side. Maxwell just wanted to present a symmetrical, easy to understand thought experiment, without complications. --Chetvorno^TALK 08:21, 9 April 2014 (UTC)

You said "pressure will gradually rise" … "a" … "complication he" … "apparently wished to avoid". There is nothing that makes this claimed intention "apparent". Maxwell simply did not discuss pressure. If I was to guess, I'd say he wanted to keep the numbers constant to make the resulting temperature difference more immediately apparent from the stipulation about the energy of the molecules. But I won't claim this because I know of no reliable source for it. But you do make a claim, namely that Maxwell "wished to avoid" a pressure difference, so please WP:PROVEIT.

"allowing the same number of" … "molecules" actually necessitates a pressure difference. As a follow-up to the previous sentence of yours (which talks about pressure), it just doesn't work. The only way to get a state change at constant density and constant volume is to have a change in pressure and temperature simultaneously. Paradoctor (talk) 20:36, 9 April 2014 (UTC)

Just to be clear, isn't this concept more simply demonstrated if the demon only permits traffic to one side of the chamber, regardless of the temperature of the molecule? The demon could be a simple mechanical protocol (1 - Open door when approached from the East. 2 - Close door when approached from the West. [We don't even need the account for the door being approached simultaneously from both sides because the net movement will still be from East to West]). The pressure would increase on one side, reducing entropy. The demon could even slide a turbine in place of the door and harvest free energy until the pressure equilibrates. --Shabd_sound 15:55, 28 Dec 2014 (CT) — Preceding unsigned comment added by Shabd sound (talk • contribs)

I would think so. I would guess Maxwell didn't use that scenario because it didn't have as obvious an application to the 2nd Law. I don't think Maxwell thought about the details much - he just wanted a simple example of how a temperature difference could be obtained in violation of the 2nd law. I'm sure he would be very surprised at all the analysis that has been devoted to it since his time. --Chetvorno^TALK 01:51, 29 December 2014 (UTC)

The "Comparison to Feynman's Ratchet" section

This section is pretty confusing, and is nowhere near adequately sourced. It may or may not be a good analysis of Feynman's ratchet as a Maxwell's demon, but it looks like either WP:ORIGINAL RESEARCH or WP:SYNTHESIS. Everything on Wikipedia, each fact, has to be WP:VERIFIABLE from a reliable source. Nor are Wikipedia articles allowed to "synthesize" conclusions from different sources. Whole paragraphs of this section are unsourced. Another thing is that the section discusses the ratchet in terms of chemical bonds, and I don't see that in the sources. Also, the Jarzynski paper, besides having no evidence it was published anywhere but on Arxiv, analyzes a discrete "model" of the Feynman ratchet consisting of a particle on a lattice, not the ratchet itself. This essay might be more appropriate on Wikibooks, where original research is allowed. --Chetvorno^TALK 20:28, 19 November 2014 (UTC)

Deleted it. --Chetvorno^TALK 02:02, 29 December 2014 (UTC)

I think the opening sentence is misleading

The opening sentence says that Maxwell's Demon "shows that the Second Law of Thermodynamics has only a statistical certainty." I really dislike this phrasing and I think it will cause readers to get the wrong impression. The fact is that Maxwell's Demon, as invented, will always violate the 2nd Law. There is no statistics about it. It's not like the Demon sometimes violates the 2nd Law, but on average obeys it. I'd like to remove this quotation from the opening sentence, but I thought I'd open it up for discussion first because it would be a major change. Let me know what you think. Tedsanders (talk) 18:16, 8 June 2015 (UTC)

I agree, that "statistical" phrase has always bothered me, too. --Chetvorno^TALK 20:08, 8 June 2015 (UTC)

However, I think the current lead sentence

"...Maxwell's demon is a thought experiment created by the physicist James Clerk Maxwell to show that the Second Law of Thermodynamics can be violated."

is too strong; it contradicts the body of the article and needs some modification. It is not clear to me whether Maxwell actually believed the 2nd Law could be violated (this would be a revolutionary stand in the 1800s) or whether he invented his demon to clarify its workings. Even if he did, it is important to make clear to general readers that the 2nd law has survived. Although research into Maxwell's Demon has elucidated the working of the law, it has not shown that the 2nd Law can be violated in the sense that energy or temperature differences can be produced from a system at thermodynamic equilibrium. I think most modern physicists and sources would disagree with the lead sentence as it stands. What about something like: "...Maxwell's demon is a thought experiment created by the physicist James Clerk Maxwell which raises the question of whether the Second Law of Thermodynamics can be violated." --Chetvorno^TALK 05:00, 9 June 2015 (UTC)

I reverted the change, pending consensus on an appropriate revision. I also think that Maxwell's summary is too abstruse for the purpose of an encyclopedia article, and should be improved upon. My proposal:

"...Maxwell's demon is a thought experiment created by the physicist James Clerk Maxwell in which he suggested how the Second Law of Thermodynamics could hypothetically be violated."J-Wiki (talk) 18:18, 9 June 2015 (UTC)

That version would be fine with me. --Chetvorno^TALK 19:01, 9 June 2015 (UTC)

Great rewording! Thanks. (Also, sorry if my edit earlier was too unilateral. I wanted to discuss it first, but it looked like this talk page was pretty dormant, so I just decided to go ahead with it.) Tedsanders (talk) 19:23, 9 June 2015 (UTC)

No, you did the right thing; BRD. Glad someone had the guts to get rid of that "statistical" phrase; I didn't. --Chetvorno^TALK 20:09, 9 June 2015 (UTC)

"Feynman's thought experiment"

The first sentence of the 'Experimental work' section refers to "Feynman's thought experiment". Should 'Feynman' read 'Maxwell', here, or is this a reference to some other thought experiment by Feynman? If the latter, it ought to be described. --MichaelMaggs (talk) 13:31, 17 September 2015 (UTC)

It's a reference to Feynman's Brownian ratchet. Yeah, that term should be in there. Put it in. --Chetvorno^TALK 17:12, 17 September 2015 (UTC)

That makes more sense. Thanks. --MichaelMaggs (talk) 18:08, 17 September 2015 (UTC)

Biological systems

Happens in biology all the time, eg, gate proteins. And of course, the 2nd Law is never violated. 213.1.8.114 (talk) 02:03, 28 May 2016 (UTC)

Computer Daemons

The unsourced claim that computing daemons are named after Maxwell's demon is simply wrong. I am removing it. If anyone feels strongly enough to try to restore it, please find a citation. 69.255.61.212 (talk) 22:43, 1 May 2017 (UTC)

I've copied over a source from Daemon (computing) which supports the assertion. clpo13_(talk) 22:55, 1 May 2017 (UTC)