Talk:Second law of thermodynamics

From Wikipedia, the free encyclopedia
Jump to: navigation, search

RfC on the lead[edit]

NAC:There are good argument both #1 and #2. There is rough consensus for #1. The closer's assessment is that the arguments in favor of #1 are strong because they include simplicity, because the lede doesn't have to summarize everything, and #1 does summarize the key points. Robert McClenon (talk) 00:47, 12 June 2016 (UTC)
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Which lead is better? Indicate option #1 or #2. 17:38, 30 April 2016 (UTC)

  • Option #1
Text of option 1
The following discussion has been closed. Please do not modify it.

The second law of thermodynamics states that in every real process the sum of the entropies of all participating bodies is increased. In the idealized limiting case of a reversible process, this sum remains unchanged. The increase in entropy accounts for the irreversibility of natural processes, and the asymmetry between future and past.

While often applied to more general processes, the law technically pertains to an event in which bodies initially in thermodynamic equilibrium are put into contact and allowed to come to a new equilibrium. This equilibration process involves the spread, dispersal, or dissipation[1] of matter or energy and results in an increase of entropy.

The second law is an empirical finding that has been accepted as an axiom of thermodynamic theory. Statistical thermodynamics, classical or quantum, explains the microscopic origin of the law.

The second law has been expressed in many ways. Its first formulation is credited to the French scientist Sadi Carnot in 1824 (see Timeline of thermodynamics).

  • Option #2
Text of option 2
The following discussion has been closed. Please do not modify it.

The second law of thermodynamics states that for a thermodynamically defined process to actually occur, the sum of the entropies of the participating bodies must increase. In an idealized limiting case, that of a reversible process, this sum remains unchanged. A simplified version of the law states that the flow of heat is from a hotter to a colder body.

A thermodynamically defined process consists of transfers of matter and energy between bodies of matter and radiation, each participating body being initially in its own state of internal thermodynamic equilibrium. The bodies are initially separated from one another by walls that obstruct the passage of matter and energy between them. The transfers are initiated by a thermodynamic operation: some external agency intervenes[2] to make one or more of the walls less obstructive.[3] This establishes new equilibrium states in the bodies. If, instead of making the walls less obstructive, the thermodynamic operation makes them more obstructive, no transfers are occasioned, and there is no effect on an established thermodynamic equilibrium.

The law expresses the irreversibility of the process. The transfers invariably bring about spread,[4][5][6] dispersal, or dissipation[1] of matter or energy, or both, amongst the bodies. They occur because more kinds of transfer through the walls have become possible.[7] Irreversibility in thermodynamic processes is a consequence of the asymmetric character of thermodynamic operations, and not of any internally irreversible microscopic properties of the bodies.

The second law is an empirical finding that has been accepted as an axiom of thermodynamic theory. When its presuppositions may be only approximately fulfilled, often enough, the law can give a very useful approximation to the observed facts. Statistical thermodynamics, classical or quantum, explains the microscopic origin of the law. The second law has been expressed in many ways. Its first formulation is credited to the French scientist Sadi Carnot in 1824 (see Timeline of thermodynamics). Carnot showed that there is an upper limit to the efficiency of conversion of heat to work in a cyclic heat engine operating between two given temperatures.

References
The following discussion has been closed. Please do not modify it.

References

  1. ^ a b W. Thomson (1852).
  2. ^ Guggenheim, E.A. (1949), p.454: "It is usually when a system is tampered with that changes take place."
  3. ^ Pippard, A.B. (1957/1966), p. 96: "In the first two examples the changes treated involved the transition from one equilibrium state to another, and were effected by altering the constraints imposed upon the systems, in the first by removal of an adiabatic wall and in the second by altering the volume to which the gas was confined."
  4. ^ Guggenheim, E.A. (1949).
  5. ^ Denbigh, K. (1954/1981), p. 75.
  6. ^ Atkins, P.W., de Paula, J. (2006), p. 78: "The opposite change, the spreading of the object’s energy into the surroundings as thermal motion, is natural. It may seem very puzzling that the spreading out of energy and matter, the collapse into disorder, can lead to the formation of such ordered structures as crystals or proteins. Nevertheless, in due course, we shall see that dispersal of energy and matter accounts for change in all its forms."
  7. ^ Pippard, A.B. (1957/1966), p. 97: "it is the act of removing the wall and not the subsequent flow of heat which increases the entropy."

Survey[edit]

  • #2: It is a bit more precise and gives a bit more detail. See my remarks in the threads linked under discussion. Kingsindian   17:38, 30 April 2016 (UTC)
  • #1. There's no need for "walls" and "external agents" to talk about the second law, and no need to limit its applicability to systems initially and eventually in equilibrium. It is misleading. Walls, separate systems initially and eventually in equilibrium and an experimentalist removing the wall would be a good first example though. YohanN7 (talk) 10:14, 2 May 2016 (UTC)
  • #2 is better because it includes two important points which are omitted in #1. (See discussion below). Dirac66 (talk) 11:28, 2 May 2016 (UTC)
  • #1 as per YohanN7. However as discussed below this is a false dichotomy and I think trying to decide between two clearly imperfect versions is a waste of time. It's better to edit as usual for wikipedia and try to arrive at a consensus version. In that spirit I will make an edit that attempts to take on board as many of the comments below as possible. If anyone strongly objects they can always revert. Waleswatcher (talk) 16:47, 3 May 2016 (UTC)
  • #2. My concern is to have a an accurate statement of the law, as opposed to a loose one. Reliable high quality texts give prominence to walls and external agents and equilibrium.Chjoaygame (talk) 16:56, 3 May 2016 (UTC)
  • #1. It is comprehensive, clear, and well-written. Indeed, it is so well-written that I would like to congratulate those responsible, I wish more Wikipedia articles could have such good leads. I could not understand #2, though I have a science degree: "transfer through the walls"? Wtf? Maproom (talk) 06:57, 5 May 2016 (UTC)
  • #1. As someone with a vaguely science-y background, #1 was far more readable. If there are inaccuracies then those should be corrected but I think what #2 has in detail it loses in clarity; too many concepts and technical terms for a lead. Those details can be included in the article proper. Nat2 (talk) 02:09, 7 May 2016 (UTC)
  • #1 is much clearer, which is important for the lead. A lot of non-technical readers will come here and have no interest beyond getting a basic understanding from the lead. It seems to hit the main points well. #2 tries to be more technical/rigorous, but the message gets muddled. Save those concerns for the body of the article. The 'walls' and 'external agent' also seem to be an unhelpful distraction. It was particularly odd to discuss the case of making walls more obstructive. Alsee (talk) 05:45, 7 May 2016 (UTC)
  • #3 (see below) as proposer. If really it comes to a two-options choice, #1, but that is a false dichotomy. TigraanClick here to contact me 08:47, 11 May 2016 (UTC)
  • #Neither Back to the drawing board. And on the way rethink the function of the lede. That function is to inform the possibly uncomprehending visitor what the subject is about, and in particular, rather than to explain it, to indicate whether he need bother to read the main body of the article. (The topic might be irrelevant to his needs, or incomprehensible to anyone without the necessary background.) Accordingly the lede must be clear and precise, but need not be substantially explanatory. Both #1 & #2 need rewording for precision and clarity, and in the light of the proper functions of the lede, probably need rewriting as well. User:Tigraan's proposal might be a lot closer, but given all the points at issue, it might be best to take precautions to prevent (or at least mitigate) a wall-of-text campaign, so I am inclined to propose a table of three columns in this discussion; they should respectively nominate items proposed or repudiated as being suitable for the lede, or agreed to be relegated to the text body. (eg "(thermodynamic) walls" might best be in column 3) It then should be simple to assess the relative merits of proposals, and to cooperate in constructing a generally satisfactory lede in short order instead of engaging in protracted and repetitive debate. I would be willing to start such a table if there is sufficient interest, but, given that I am neither a physicist nor engineer, It might be better if someone else had a go.JonRichfield (talk) 06:21, 17 May 2016 (UTC)
  • #1 much clearer. The lead should only summarize the result and does not need to go into the details. Precision can go later.--Salix alba (talk): 09:06, 19 May 2016 (UTC)

Discussion[edit]

See discussion threads here, here, here, here, among others. Kingsindian   17:38, 30 April 2016 (UTC)

  • #2 is better because it includes two important points which are omitted in #1.
a) the mention of walls between participating bodies (or subsystems) which become less obstructive for the process to occur, an important concept.
b) the other is the Carnot limit on engine efficiency, which is important to include in the intro because it constitutes the initial experimental basis of the law, plus its initial practical application. Non-expert readers may want to know at the outset: why do we think this law is true, and what it is good for?
This said, I would accept two of the changes suggested in option #1.
a)I would describe the processes not only as thermodynamically defined but also as real (or natural or spontaneous), since I believe (perhaps I am wrong) that the reverse of a real process would also be thermodynamically defined if it occurred, which of course it does not.
b)I would delete the sentence about making the walls more obstructive, which seems excessively detailed for the intro. It seems quite obvious that if the process is not occurring and the walls become more obstructive, nothing will happen. Dirac66 (talk) 20:38, 30 April 2016 (UTC)
  • First, let me say that there is no reason to choose between 1 and 2. Clearly they can both be improved, so this RfC is ill-conceived and potentially counterproductive. We should be iteratively improving the lede through the usual process of editing, not force some kind of false dichotomy. That complaint aside, if I must choose one of the two, I pick #1 (hence my initial edit). 2 is incorrect or potentially seriously misleading in several ways. Its most serious deficiency is that it makes it sound as though the second law applies only in some idealized thermodynamic sense. In reality, the second law is routinely applied in situations where the system(s) are not in strict thermodynamic equilibrium. In fact, it is always applied in such situations, because no real system is ever strictly in equilibrium. Restricting to "thermodynamically defined" processes makes it sound as though the second law only applies in artificial situations that never actually occur. That's just about the worst possible misconception the article could give to the reader.
  • A less important objection is that there is no necessity for "walls". One can consider many situations where the 2nd law applies even in the strictest sense that do not literally involve walls or their removal. For instance, two confined (by a force, say, not walls) systems that are initially very far apart and are then brought together. Or two populations of particles that are intermingled but exactly non-interacting until some change is made to the system by an external agency. Another questionable statement is that there must be spread, dispersal, or dissipation of matter or energy. Suppose there are particles in state A in one half of a container that is separated by a wall from the other half, in which there are particles in state B. States A and B have the same energy, and both are stable (if you want a specific example, these could be non-interacting photons with either L or R circular polarization). Now remove the wall, and the two populations mingle. Where is the spread of matter or energy? Waleswatcher (talk) 23:28, 30 April 2016 (UTC)
@Waleswatcher: Of course both 1 and 2 can be improved (nobody suggested otherwise), but they are fundamentally different, and I have been seeing disagreement between you and Chjoaygame spanning over many months, the edit you made recently being only the latest in this story. For instance, you have been making this point of the 2nd law being applicable in systems not in equilibrium for many months now. On this point, it is unlikely that you or Chjoaygame will agree; there comes a time when one has to cut the Gordian knot and choose. In my experience, long-term disagreements over irreconcilable points can lead to acrimony and frustration. As to your point that the lead should be iteratively improved, I have two rejoinders. Firstly, that is not the spirit in which you made the wholesale edit. Secondly, there is nothing stopping the lead from being iteratively improved after the RfC. The RfC is about which vision of the lead is appropriate. Kingsindian   00:22, 1 May 2016 (UTC)
In that case, this RfC isn't actually about which version is preferred, it's about which "vision" is preferred. You should make clear (a) what you see as the "fundamental" difference between the two (because I don't see one), and (b) what you see as the "vision" represented by each. Note that the version I restored contained the sentence "While often applied to more general processes, the law technically pertains to an event in which bodies initially in thermodynamic equilibrium are put into contact and allowed to come to a new equilibrium." (It also contained similar language about spread of energy, which I intended to edit before your revert.) As for Chjoaygame and acrimony, s/he has already been indefinitely topic-banned by the community from editing quantum physics articles. Arguably the second law is one such article, although I won't try to make that case at least for the moment. But I bring this up to make the point that Chjoaygame's editing certainly does create acrimony. Waleswatcher (talk) 00:39, 1 May 2016 (UTC)
@Waleswatcher: I apologize for being unclear. My use of "vision" was simply a shorthand. The main differences which I see are that the second version is a bit more detailed and tries to give a more formal description of the 2nd law, while the first version tips the balance more towards a general outline and leaves more details for the body of the article. This is just my impression of the two leads; as I said elsewhere, I am not an expert on the topic. If you think a more iterative approach will work, I am happy to withdraw this RfC. Kingsindian   01:06, 1 May 2016 (UTC)
Editor Waleswatcher mentions three matters of physics. Walls, approximation to equilibrium, and spread. As for walls, they are emphasized in important texts such as Callen. It is possible, as does Editor Waleswatcher, to separate systems by means that are not literally walls. But there are virtual walls. As for approximation to equilibrium, there are no systems that are observed for ever and so verified to be in thermodynamic equilibrium. The law is idealized. For systems not in ideal thermodynamic equilibrium, the law is often approximately valid. But an approximation is an approximation. I think it clumsy and partly misleading or evasive to talk about the law as "technically pertain[ing]". The law is exact for ideal cases, and approximate for approximate cases. As for spread, the A particles spread into the former B compartment, and vice versa.Chjoaygame (talk) 01:20, 1 May 2016 (UTC)
"For systems not in ideal thermodynamic equilibrium, the law is often approximately valid." Please give an example of a system where the second law of thermodynamics is not valid, and please explain what "approximately valid" means. Note that "not applicable" is not the same as "not valid". As for walls, I'm glad you acknowledge the language is imprecise. As for spread, yes, A particles spread, but if A and B are photons with different polarizations, that is neither a spread of matter nor of energy (photons of equal energy are moving equally in all directions). My point? It's very hard to be really precise, and the lede is not the appropriate place to try. The lede should be correct, but it can also be somewhat vague or general. Most importantly, it shouldn't mislead the reader (like into thinking that walls have anything essential to do with the second law). Waleswatcher (talk) 01:28, 1 May 2016 (UTC)
Re the Gordian knot invoked above by Kingsindian: everyone please remember that unlike Alexander the Great, Wikipedians are supposed to resolve controversies by reference to reliable sources. Dirac66 (talk) 01:36, 1 May 2016 (UTC)
Thank you for this response. I am happy to replace my above talk-page word 'valid' with your word 'applicable'. I think walls (actual or virtual) do have something important to do with the law. They are the targets of thermodynamic operations, which are conceptually essential. The internal dynamics of the materials are reversible. As for the spread of matter or energy when talking about polarized photons, the important idea that people need is the one emphasized by Guggenheim and others: spread. For the case you raise, of polarization, it is polarization that spreads. Perhaps that could be added if it is felt to be important. But spread is unquestionably important.Chjoaygame (talk) 01:44, 1 May 2016 (UTC)
Good - you've now agreed with me that "For systems not in ideal thermodynamic equilibrium, the law is often approximately valid/applicable" and that in the real world there "are no systems...in thermodynamic equilibrium". Those are two of my central points - the lede needs to clearly explain that. As for your other comments above, polarization doesn't spread in my example. The system goes from being highly polarized to being unpolarized, so it's quite the opposite. And it's not difficult to think of situations where the prevailing sense of motion is not "spread", but contraction. For instance, water vapor is brought into contact with ice. The ice warms slightly, barely changing shape or volume, but the vapor condenses to a liquid. There's no "spread" of energy really, it just flows from vapor to ice. Again, the point is that these attempts to make the lede specific often end up making it actually wrong or at least misleading. There simply isn't room in the lede to give the appropriate background for the reader to understand the limitations and nuances of what is being said. Instead, the reader is (rightly) expecting to see only the most central and essential concepts briefly explained. Waleswatcher (talk) 11:47, 1 May 2016 (UTC)
There is spread at the microscopic level, which perhaps should be specified. The photons with L polarization spread from one half of the container to the other, and those with R polarization spread from the second half to the first. This happens to have the macroscopic effect of decreasing the total net polarization to zero. And in your second example, the energy in the vapor molecules spreads to the ice molecules, which happens to have the macroscopic effect of decreasing the system volume. Dirac66 (talk) 12:26, 1 May 2016 (UTC)
It's true that the L (and R) photons spread, but the polarized region contracts. For the water vapor there is a transfer of energy (it's a bit of a stretch to call that a "spread", but OK), but the system's volume contracts. So there is a major problem with asserting that "spread" is important - in both these examples you could consider the time reverse (a process that would violate the second law) and just as naturally (more naturally, really) say that something spreads: polarization in the first case, and volume in the second. Hence there is nothing about "spreading" (versus its opposite, contracting) that is actually intrinsic to the character of the second law, else it wouldn't apply to the time reverse. Again, my point is not that we shouldn't discuss spreading in the article, it's that we need more context and careful detail than can possibly go in the lede. Waleswatcher (talk) 14:24, 1 May 2016 (UTC)
The talk of spreading went into the lead because users were asking for intuitive accounts of entropy. As I see it, we could choose simply to say that an intuitive account of entropy has no place in the lead. That is a reasonable option. If the requests for an intuitive account are acceded to by putting something in the lead, then the question arises of what one to use. There are two obvious candidates, disorder and spreading. Disorder is traditional but objectionable. Spreading, so far as I know, subject to better information, came on the scene with Guggenheim in 1949. Editor PAR has expressed concern about it, and now Editor Waleswatcher is concerned about it. I am not wedded to putting an intuitive account of entropy in the lead. I did it only to respond to requests. I am happy to fit in with other editors on this point.Chjoaygame (talk) 16:57, 1 May 2016 (UTC)
I think Editor Dirac66 is right to point to the need to read the spreading story as often referring to spread amongst microscopic degrees of freedom as distinct from less refined ideas of spread. That is clear in Guggenheim. It is an argument that the idea of spread is not strictly macroscopic, a significant point, I think. On the other hand, the disorder tradition also appeals to microscopic degrees of freedom. Guggenheim couples his talk of spread with talk of increase of accessibility. That is in other words hardly different from decrease of obstruction by walls. This describes the thermodynamic operation more than the effect on the bodies. The section of the article, headed Intuitive meaning of the law, says this as follows: "This reasoning is of intuitive interest, but is essentially about microstates, and therefore does not belong to macroscopic equilibrium thermodynamics, which studiously ignores consideration of microstates, and non-equilibrium considerations of this kind."Chjoaygame (talk) 01:49, 2 May 2016 (UTC)
As for approximation. I think the law should be stated primarily in its ideal form, and accurately. I am quite happy to see a statement to the effect that ideal scenarios are ideal, and that approximations can also be made. But I would be unhappy to see the law stated in some loose form as a way of combining the ideal accurate form with the statement that approximations are possible. For example, I am happy with the present statement in the lead: "When its presuppositions may be only approximately fulfilled, often enough, the law can give a very useful approximation to the observed facts."Chjoaygame (talk) 17:08, 1 May 2016 (UTC)

I object to two things: First, the statement of the second law should not be only in terms of entropy, but also irreversibility. As it stands, it begs the question of entropy. The second law says "Define entropy as such and such. Its a state function. In the thermodynamic limit (infinite quantity of material), it never decreases". That's the quantitative statement, but gives no intuitive understanding. I favor the inclusion of an intuitive understanding along with the quantitative statement, and that would involve irreversibility. Second, as mentioned above, "spreading" not always true when taken at face value, therefore not fundamental. If we back off and say its some sort of spreading in phase space, we've not only strayed from thermodynamics into statistical mechanics, but we also have to realize that it is not energy or matter that is spreading, but the probability distribution of microstates which is spreading, (therefore the probability distribution of energy and matter). This spread is according to a very particular measure of "spread" which corresponds to the information entropy of that probability distribution. The spreading metaphor gives a warm, fuzzy, but ultimately false sense of understanding. PAR (talk) 12:45, 2 May 2016 (UTC)

Agreed on all points. An intuitive understanding in the lede is absolutely essential, in my view, as is a clear statement that this is a law of nature that applies far beyond the narrow, idealized world of classical equilibrium thermodynamics. Many people come to this article to try to gain some understanding of what this is all about, but aren't interested in/can't understand technical details. Those people should come away having learned something. The sentence "When its presuppositions may be only approximately fulfilled, often enough, the law can give a very useful approximation to the observed facts" is so convoluted as to be essentially incomprehensible. Again, the law is constantly and correctly applied in all sorts of situations that are not even close to the conditions stated in the current lede. When people ask - as they very often do - why evolution doesn't violate the 2nd law, the best answer is that the earth is not a closed system and the entropy of the earth/sun system is increasing. The best answer is not "the earth and the sun were not initially in internal thermodynamic equilibrium, so the law is inapplicable" or "there was no wall between the earth and the sun that was removed by an external agency, so the law is inapplicable", etc. Waleswatcher (talk) 14:09, 2 May 2016 (UTC)
  • Responding to Editor PAR.
I object to two things: First, the statement of the second law should not be only in terms of entropy, but also irreversibility. As it stands, it begs the question of entropy. The second law says "Define entropy as such and such. Its a state function. In the thermodynamic limit (infinite quantity of material), it never decreases". That's the quantitative statement, but gives no intuitive understanding. I favor the inclusion of an intuitive understanding along with the quantitative statement, and that would involve irreversibility.
Editor PAR raises the question of how the law should be stated in the lead. As I read him, he favors a statement in terms of entropy provided there is further expression of intuitive understanding, with particular reference to irreversibility. This raises the question, of how to refer to entropy in the lead, and how to refer to irreversibility. He is right that both versions under discussion rely on entropy without defining it. It may be a problem to define entropy in the lead. Carathéodory, and recently Lieb & Yngvason, have derived the concept of entropy from a presupposition of knowledge of irreversibility of processes in a sort of empirical data-base. Both versions under discussion talk of irreversibility. The nature of irreversibility is not quite simple. The microscopic dynamics are in general reversible, while the macroscopic processes of thermodynamic are in general irreversible; yet that is no real contradiction. On another tack, it is possible to state the second law in terms, for example of cyclic processes, that do not rely on entropy; at present I do not see this as proposed here.Chjoaygame (talk) 17:53, 2 May 2016 (UTC)
Second, as mentioned above, "spreading" not always true when taken at face value, therefore not fundamental. If we back off and say its some sort of spreading in phase space, we've not only strayed from thermodynamics into statistical mechanics, but we also have to realize that it is not energy or matter that is spreading, but the probability distribution of microstates which is spreading, (therefore the probability distribution of energy and matter). This spread is according to a very particular measure of "spread" which corresponds to the information entropy of that probability distribution. The spreading metaphor gives a warm, fuzzy, but ultimately false sense of understanding.
Editor PAR has objections to Guggenheim's idea of spread. That idea is intended to give an intuitive understanding, and is open to criticism. There is another, more traditional idea, that might be regarded as competing with spread, namely 'disorder'. Such ideas are debatable and perhaps do not fit into the lead.Chjoaygame (talk) 17:53, 2 May 2016 (UTC)
  • Responding to Editor Waleswatcher.
I think the notion that somehow evolution might challenge the second law is caused by loose statements of the law. I see the remedy as based on tackling that cause: the law is about passages from equilibrium to equilibrium, and evolution does not conflict with that. I do not see the remedy as yet another loose statement. There is as yet no established full theory of non-equilibrium thermodynamics. It would be misleading to suggest that there is, by making loose statements in the lead.Chjoaygame (talk) 18:11, 2 May 2016 (UTC)
I think the answer to "why doesn't evolution violate the 2nd law" is to say that this is too vague to answer; one can't apply the second law in such a loose fashion. Kingsindian   04:21, 3 May 2016 (UTC)
Then it sounds as though both of you disagree with the current lede. Reference 5 is quoted there as follows: "It may seem very puzzling that the spreading out of energy and matter, the collapse into disorder, can lead to the formation of such ordered structures as crystals or proteins. Nevertheless, in due course, we shall see that dispersal of energy and matter accounts for change in all its forms." Waleswatcher (talk) 16:45, 3 May 2016 (UTC)
Since I posted in the article the quote from a respected text, about spreading, copied just above, I am puzzled to read that I seem to disagree with it.Chjoaygame (talk) 17:01, 3 May 2016 (UTC)
Just above, you said "I think the notion that somehow evolution might challenge the second law is caused by loose statements of the law. I see the remedy as based on tackling that cause: the law is about passages from equilibrium to equilibrium, and evolution does not conflict with that." In other words, the second law is inapplicable to evolution because evolution doesn't fit some narrowly defined requirements. That is consistent with your view that we should only discuss idealized thermo and ignore the multitude of other applications of the second law. I'm simply pointing out that according to your own source, it's quite natural to be puzzled by "the formation of such ordered structures as...proteins" (just as I said above, many people find that puzzling and will come to this article to learn more), and (again, according to your source) there's no problem asking about such things and answering them in the context of the second law. Waleswatcher (talk) 17:09, 3 May 2016 (UTC)
Editor Waleswatcher, writing this, "your view that we should only discuss idealized thermo and ignore the multitude of other applications of the second law", is grievously misrepresenting my view. My view is that loose statement of something that someone might think was the law is not good for Wikipedia. I think Editor Waleswatcher is pushing such a loose statement.Chjoaygame (talk) 17:34, 3 May 2016 (UTC)
As for "the collapse into disorder, can lead to the formation of such ordered structures as crystals or proteins": Indeed the 'disorder' notion of entropy can lead to puzzlement. 'Spread' is offered as perhaps a way past it.Chjoaygame (talk) 20:28, 5 May 2016 (UTC)

For those who are unhappy about using the word walls, we could say instead that in the initial state the subsystems are separated in space, and that during the process some matter (distinct molecules, or energy, or photons with different polarization states, etc.) passes from one subsystem to another. We don't have to specify what exactly separates the subsystems, or we could use some synonym for (real or virtual) walls. Dirac66 (talk) 17:38, 5 May 2016 (UTC)

Good point. The word 'walls' is deeply engraved in reliable sources, but seems like a red rag to some readers. I have thought about such clumsy phrases as 'more or less semi-permeable walls'. Not an easy one. Needs thought. Perhaps 'selectively permeable walls' would do?Chjoaygame (talk) 20:20, 5 May 2016 (UTC)Chjoaygame (talk) 17:10, 8 May 2016 (UTC)

I think Editor Waleswatcher is right, that this RfC is ill-conceived and potentially counter-productive. I suggest Editor Kingsindian may agree and act on his above proposal "If you think a more iterative approach will work, I am happy to withdraw this RfC."Chjoaygame (talk) 17:20, 8 May 2016 (UTC)

I will be happy to withdraw the RfC if Waleswatcher agrees, as I said above. They only have to let me know. Kingsindian   17:46, 8 May 2016 (UTC)
  • Summoned by bot; did not read previous discussion threads (do say if I repeat an argument that was rejected before). Basically I agree with everything that User:Waleswatcher said before; notably, that the RfC looks like a false dichotomy, and it is far more important to be clear and concise than to be precise in the lead.
I would mention first (what I think to be) an important point. In its conventional modern form, the second law is about a single (closed) system (though historically the exchange aspect was crucial, see Clausius theorem). The current proposals make it look like there has to be at least two systems in play. Of course, that makes no operational difference because as soon as something happens in system A you can artificially cut it into system A1 and A2 with exchanges between A1 and A2, but the thing is, one can readily see what A1 and A2 are in the case of thermal machines and other stuff with heat-work conversions, but it is far less easy in other cases (e.g. salt dissolving in water). So while I have no quarrel with the word "walls" in particular (I would prefer "interface" but really that is splitting hairs) I would think they should not even be mentioned in the lead.
Physics background required to understand this paragraph. On the various cases where the law does not apply (needs to be at equilibrium, etc.): that is almost irrelevant. Yes, I know my physics and technically the second law only applies to an infinitely large system where the transitions between microstates are infinitely faster than any macroscopic process (i.e. the system is in a steady state, macroscopically). But the very first operational step is to say that it is good enough that the macroscopic system moves "slowly enough" compared to the microscopic transitions ("quasi-equilibrium"), and that the Avogadro number is large enough that we ignore statistical deviations from the central limit theorem.
Putting the "dirty" but useful version in the lead is much better than putting a ton of mathematics mumbo-jumbo right off the bat. The lead of Addition is a visual depiction of apples grouped together, not the first of 400 pages of set theory, even if those are the current rigorous definition of addition. (And the interested reader has a taste of it in Addition#Natural_numbers)
With that in mind, here is the lead I would write:
Radical proposal (option 3)
The following discussion has been closed. Please do not modify it.

The second law of thermodynamics states that the total entropy of an isolated system always increases over time, or remains constant in the ideal cases where the system is in a steady state or undergoing a reversible process. The increase in entropy accounts for the irreversibility of natural processes, and the asymmetry between future and past.

Historically, the second law is an empirical finding that has been accepted as an axiom of thermodynamic theory. Statistical thermodynamics, classical or quantum, explains the microscopic origin of the law.

The second law has been expressed in many ways. Its first formulation is credited to the French scientist Sadi Carnot in 1824 (see Timeline of thermodynamics).

Inspired from the previous proposals of course - I removed the reference to the "energy spread" interpretation that does not belong in the lead (if it even belongs to WP), and I am not quite sure that Carnot ought to be mentioned in the lead. TigraanClick here to contact me 08:46, 11 May 2016 (UTC)
Much though one may respect the authority of Editor Tigraan, Wikipedia editors also respect reliable sources, which conflict with Editor Tigraan's authority. Reliable sources explicitly presuppose the existence of states of thermodynamic equilibrium, and say that the second law is about transfers between systems. Slick abbreviated statements leave the reader to work that out for himself, and dismiss the more systematic and logical development of thermodynamics that is found in reliable sources. If the reader can work that out for himself, he is hardly helped by the slick abbreviated statement.Chjoaygame (talk) 13:05, 11 May 2016 (UTC)
Is
  • Mandl, F. (1989). Statistical physics (third ed.). 
reliable? It agrees pretty much (I need to check details, don't have it in my lap) with the first sentence in the "radical proposal". The first example does involve a wall though. YohanN7 (talk) 13:55, 11 May 2016 (UTC)
Mandl's book is about statistical physics. This article is about macroscopic physics.Chjoaygame (talk) 14:29, 11 May 2016 (UTC)
It blends thermodynamics (even a la pre 1850) with "statistical physics". And he definitely calls the law the second law of thermodynamics. Is it reliable? YohanN7 (talk) 14:44, 11 May 2016 (UTC)
As far as I can see, there is no third edition. The latest I traced was the second edition, 1988. No, it's not reliable on this question, taken along with other reliable sources that are primarily on this topic.Chjoaygame (talk) 14:51, 11 May 2016 (UTC)
If your argument is that a consensus of other sources disagree, I see the logic but I disagree with the premise. If it is that statistical physics is too far from thermodynamics, and hence a textbook on SP is not a RS for thermodynamics, I strongly object. TigraanClick here to contact me 16:02, 11 May 2016 (UTC)
As I said, I didn't have the book in my lap. Right, the 1989 reprint of the second edition of 1988. Statistical physics is macroscopic physics. It is the message of the first sentence in the first chapter of the book. Why don't you admit that you have not read the book, and accordingly do not have a clue? It would be more honest by you, instead of leaving summaries suggesting pipe dreams about different editions on my part. YohanN7 (talk) 08:46, 12 May 2016 (UTC)
My "authority" is irrelevant. I would very much like to see your sources that "say the second law is about transfers between systems". Not some mention of transfer between systems, because that will be mentioned, of course, but something like "the second law is as follows: *(blah blah transfers)*". Off the top of my head, the Landau and Lifshitz Statistical Physics does not mention that (need to check though).
In any case, that is fairly irrelevant, too. What we need RS for is for facts, not for the presentation of facts. An outline that dishonestly presents a string of facts to mislead the reader into an incorrect conclusion would constitute WP:SYNTH, but we do not have to follow the closely the outline of an RS (arguably, doing so too closely would be copyvio). Which is a good thing, because most RS on the subjects are intended for physics graduates, while WP is intended for a (relatively smart) person with no former knowledge of the subject.
I would bet you can find a reliable source à la Bourbaki to express the second law as "every isolated system has a nonconstant state function that is nondecreasing with time", but it does not mean this is a helpful formulation in the lead. TigraanClick here to contact me 15:58, 11 May 2016 (UTC)
Adkins 3rd edition (1983), p. 53: The Clausius statement: "No process is possible whose sole result is the transfer of heat from a colder to a hotter body."
Denbigh 4th edition (1981), p. 26: "STATEMENT A. It is impossible to make a transfer of heat from a heat bath, at a uniform temperature, and obtain an equivalent amount of work, without causing a change in the thermodynamic state of some other body."
As you say, our job here is to render the facts to be found in reliable sources. Readers ask for layman-accessible accounts. They need to know that thermodynamics is based on the equilibrium properties of materials, as revealed by transfers when walls are removed or made more permeable. A statement that the entropy of an isolated system never decreases is nice for experts such as the assembled company, but near meaningless to non-experts. It is slick, but not accurate. In general, the entropy of a system is defined only when it is in its own state of internal thermodynamic equilibrium. It is a pipe-dream to have a proper definition of a single entropy for a non-equilibrium system. A pipe-dream widely shared, but not regarded as established by reliable thermodynamics sources. Unless one makes a bundle of assumptions that experts might make, about sub-systems, initially present walls or the like, removal of the walls, an isolated system is not something that makes sense to a layman for the notion of increase of entropy. Entropy for non-equilibrium systems is something for the future of physics. I think it will be more or less as Phil Attard proposes, a matter of a hierarchy of several-time entropies, but such is way beyond the scope of the present article, and hardly within the scope of Wikipedia at all. To make grandiosely general statements that more or less hint or imply that one has a concept of non-equilibrium entropy is to go far beyond what reliable sources can support, no matter how gratifying one might find it for oneself. The spreading story as an intuitive view of entropy is new, so far as I know only dating from 1949 (Guggenheim). It seems to find favor in few eyes here. Without it, the layman needs more concrete ideas, such as walls and external tamperers, not a retreat into the abstraction of entropy. If you doubt the importance of equilibrium, Callen (1985) writes in italics on p. 6: "By definition, suggested by the nature of macroscopic observations, thermodynamics describes only static states of macroscopic systems."Chjoaygame (talk) 18:16, 11 May 2016 (UTC)
So, your suggestion is that we use the Clausius theorem (which has a standalone article, BTW) as the formulation of the second law? If so, we should remove any reference to entropy. That does not seem very reasonable to me, and I really, really hope that whoever edits the second law of thermodynamics does not intend to remove entropy from its lead (would you suggest to remove energy from the first law?). The Clausius theorem is basically the historical way the second law was discovered, which is important, but not the focus of the lede.
Whether there is no ironclad definition of entropy in out-of-equilibrium systems is, again, irrelevant, in the same way that it is irrelevant that the lead of addition refers to "numbers" as if that was a natural thing easily defined in maths. Using the equilibrium definition is good enough for quasi-steady processes, that is what everyone does, and the fine details can be left to the article. TigraanClick here to contact me 08:53, 12 May 2016 (UTC)
The second law as stated in terms of entropy of isolated systems in an arbitrary state is of postulatory nature. The postulates needed aren't proved from more basic principles, but they can be made plausible, and are further motivated by their universal success so far. This is what I meant in a section below about the second law having quite some status in present day physics that is hidden by Chjoaygame's definition. For instance black holes are believed to possess entropy, even calculable entropy. For those believers, the second law overrides general relativity that (when considered alone) suggests the entropy is zero (if any) for the simplest version of black hole due to the singularity. YohanN7 (talk) 10:26, 12 May 2016 (UTC)

":Mandl's book is about statistical physics. This article is about macroscopic physics." Mandl is of course a reliable source for the second law of thermodynamics, and to suggest otherwise (because the title contains the word "statistical"?) is, frankly, absurd. As for the proposed wording, I think it's pretty good and gets the tone and level right on (I'd make some small changes, like ideal --> idealized). Waleswatcher (talk) 16:36, 11 May 2016 (UTC)

preempting the course of the discussion[edit]

Editor Waleswatcher has now posted a vote with an announcement that he intends to preempt the course of this discussion. Others have refrained out of respect for the discussion. I disapprove of a preemptive edit.Chjoaygame (talk) 17:07, 3 May 2016 (UTC)

Actually, the discussion began with a revert of my original edit. Anyway, it's pretty clear that there is no consensus on which of those two versions is better, and part of the reason is that that is not the right question. Neither version is perfect, and there is no need to choose between them. Rather than waste time arguing over a false dichotomy, let's try to find a consensus version. This was my attempt. Waleswatcher (talk) 17:13, 3 May 2016 (UTC)
Looking at the edit here, covered by the edit summary "edited the lede, attempting to find a middle ground that takes on board various talk page comments". I disapprove. The edit summary is misleading. The edit is close to a total imposition of version #1 that Editor Waleswatcher made and led to this RfC.
Though objecting strongly, I am not accepting Editor Waleswatcher's invitation to "revert" his new edit because to do so would be to participate in his preemption of the course of this discussion.Chjoaygame (talk) 17:20, 3 May 2016 (UTC)

What this RFC is about[edit]

The RFC is quite misleadingly suggesting that the issue is about Waleswatcher overthrowing a consensus "reached on this talk page". Not so. The real issue is Chjoaygame's January 9 edit. It is there that the second law is stripped of every sort of general applicability.

I dare say that the second law – in a sense – has the same or even higher "truth value" than even special relativity or quantum mechanics in the physics society. The latter theories are thought of as pretty solid. They aren't completely set in stone though. But no one, afaik, expects the second law to ever fail. Therefore, I believe it is important for this article (in the lead) to suggest establish that it has a pretty wide applicability. YohanN7 (talk) 11:57, 6 May 2016 (UTC)

That "no one ... expects the law to ever fail" is no argument that it should be stated loosely or so as to have "pretty wide applicability". It is an argument that the law should be stated accurately, so as to precisely define the scope of its applicability. The accuracy should be enough to make the reversible process a well defined limiting case. If the law is stated in terms entropy, the definability of entropy needs to be established.Chjoaygame (talk) 03:01, 7 May 2016 (UTC)

too many concepts and technical terms for a lead[edit]

Editor Nat2 is concerned that #2 has "too many concepts and technical terms for a lead." Perhaps so. If a remedy is needed, it is not to state the law loosely in the lead, because many readers will take the loose statement at face value and run into error with it. If a remedy is needed, it is to state the law in the lead in such a way that readers will not be encouraged to run with a loose statement. It is not the task of Wikipedia to state laws in seductively over-simplified ways, for the benefit of those not interested in accuracy.Chjoaygame (talk) 03:14, 7 May 2016 (UTC)

Summary so far[edit]

This RfC has been useful in at least one respect - a clear majority has now emerged that agrees that the old lede (#2) was indeed too technical, too detailed, and overly specific and restrictive. The comment on this talk page that inspired my original edit was "The introduction section of this page needs to be translated back into English. The kind of English that most people will understand." Hopefully that is addressed whether whether we end up with #1, #3, or some similar version. Waleswatcher (talk) 16:58, 11 May 2016 (UTC)

Evidently #2 is massively outvoted. Only three supporters. To save further time spent on this point, I suggest something like #1 or #3.Chjoaygame (talk) 19:54, 11 May 2016 (UTC)
While I do not like #2, it is WP:NOTAVOTE, and 3 !votes out of 9 is still a significant fraction of the people taking part. But anyways, the polling format is disturbing for a lead when many more options could be proposed. TigraanClick here to contact me 08:39, 12 May 2016 (UTC)
RfCs, in my experience, work best when there is a clear, short and neutral question posed. I have opened many RfCs on the same model as the one here. Regarding the "false dichotomy" point, the answer is that there isn't any. Of course, things can always be improved, but as the RfC makes clear, the starting point should be #1 or #3, rather than #2. I stated this point explicitly from the very beginning. I think the situation right now is an advance from the starting point of a rather sterile argument between two people who weren't going to agree with each other. YMMV. Kingsindian   05:11, 19 May 2016 (UTC)

Consistency of article[edit]

According to the summary just above, the lead seems to be rescued (for now). But then, what about the section intuitive meaning of the law immediately following the lead? This near infinite ramble, including many mile long literal quotes, is much in line with Chjoaygame's January 9 attack on the lead (and quite possibly part of it). No layman reader, and few other, will endure that section. Not only is it strongly POV-pushing. It is horribly written and far too long. YohanN7 (talk) 12:19, 12 May 2016 (UTC)

Fully agreed. That subsection should either be deleted or drastically trimmed and re-written. If it is to explain something "intuitively" it should be clear, comprehensible, and simple. It is quite the opposite. Waleswatcher (talk) 12:22, 12 May 2016 (UTC)
I am late to the party, but for feedback, that section is unreadable. How could anyone be bothered to read it? A list of everyday examples where entropy increases would be far better e.g. when an ice cube melts into water, the motion of water molecules in the liquid are more disordered than in the solid cube. MŜc2ħεИτlk 16:50, 12 May 2016 (UTC)
You would heat the ice cube to melt it, hence bringing energy into the system, which makes the example awkward. But your basic suggestion is good, e.g. with salt dissolving in water. TigraanClick here to contact me 16:59, 12 May 2016 (UTC)
The list of everyday examples would have to indicate when they are valid. For example ice cubes do not melt into water when they are outdoors in Northern Canada in January. Under these conditions water will freeze. To explain why there is a difference we need a quantitative criterion such as entropy or a related property. Dirac66 (talk) 17:46, 12 May 2016 (UTC)
Bring ice cubes in environment hotter than 0°C and they will melt. Bring water in environment colder than 0°C and it will freeze. But that is an extremely poor example for the second law, because energy is transfered; in the second case, the entropy of the water molecules actually decreases (though the total entropy increases when take into account the environment's entropy).
The relevance of the second law in that example is that ice cubes in winter Siberia do not spontaneously melt and make the environment even colder. But I do not think it is really easy to understand. A similar classic example is the "icemaking boat" (see e.g. [1], "one might propose a scheme..."). TigraanClick here to contact me 11:17, 13 May 2016 (UTC)
Yes, the ice cube was just a very quick and incomplete example and I didn't care to be detailed in the last post. Ideally we should choose examples which people can actually connect to, requiring only brief descriptions of conditions. The point was simply that examples are superior to a rambling wall of text. MŜc2ħεИτlk 11:28, 13 May 2016 (UTC)
Agreed as well. "Some energy, inside or outside the system, is degraded in its ability to do work." - is that supposed to mean anything for a layman? TigraanClick here to contact me 16:59, 12 May 2016 (UTC)
It is meaningless because energy does not "do work". Work is a form of energy. MŜc2ħεИτlk 11:33, 13 May 2016 (UTC) Amended MŜc2ħεИτlk 11:52, 13 May 2016 (UTC)

A section "Intuitive meaning of the law" should quite leisurely relate to everyday experience. A good such section should imo be possible to summarize with "there is nothing such as a free lunch". Conclusions drawn directly from the classical formulations lend themselves to quite vivid and even entertaining examples. The modern notion is a bit more boring, and is perhaps best described in terms of "disorder". Everything will, given time, end up in a total mess. Don't take this post too seriously. But I do want to find something quite the opposite of the present version. These "mostly-for-fun" examples I have in mind can be reliably sourced. YohanN7 (talk) 11:45, 13 May 2016 (UTC)

The ice cube discussed above is a simple example which relates to everyday experience, but it is a valid example only if we point out that melting is spontaneous under some conditions (T > 0oC) and freezing under other conditions. And to explain why the direction of the process depends on the surroundings, we need to mention entropy of both system and surroundings, and energy transfer between the two. If we only discuss melting as an example of system entropy increase, then a naïve reader may conclude that system entropy always increases, which would be incorrect. Dirac66 (talk) 16:12, 13 May 2016 (UTC)
Fair enough, but this section doesn't need to be logically impeccable. One example of what I had in mind is this. It refers to Kelvin's formulation: Let the system be a submarine in a heat bat (the ocean). If we had a mechanism turning heat into work (in a fashion violating Kelvin's statement), then we could transfer that work to the propeller axis of the submarine, temporarily cooling the sea. At the end of the journey, all heat have been returned to the sea due to frictional forces heating the submarine and mechanical forces stirring up the water, and equilibrium is retained in the end. In particular, the first law is not violated, not even temporarily. (I can reference this example to an undergraduate level book (in Swedish, don't think it is translated).) (This would be a perpetuum mobile of the second kind according to my source.)
Referring to the above (or something similar), one might argue that (assuming a calm sea) it is utterly improbable (if not impossible for other reasons) that a collection of unusually fast (with respect to water temperature) water molecules would conspire as to set any macroscopic mechanical device into motion when the water is treated from the point of view of statistical mechanics. Generalize this to a statement that disordered motion doesn't not by itself become ordered motion. Then introduce the entropy S as a qualitative (statistical, not precise) measure of the disorder. The intuitively supported statement is now dS ≥ 0. (I cannot reference this particular example, but similar examples exist in the literature: "It would be utterly surprising if...")
The level of rigor in an "intuitive meaning" section need not be higher. Rigor is like clothing. It should suit the occasion. YohanN7 (talk) 09:26, 16 May 2016 (UTC)

Big enough topic for two articles?[edit]

Second law of thermodynamics and Second law of thermodynamics (classical thermodynamics)? Just an idea...

Chj is a documented POV pusher. This does not mean that his/hers POV is always invalid (downright false). With two separate articles, the Chj version of affairs can be kept in the new article, except for the cranky tone entering it. This tone probably enters because he/she feels the need to ARGUE inside the articles. Example:"It will be evident to the reader that the account given in the present article does not follow such a mainstream view... " The present version of intuitive meaning of the law contains much of the very same sort of ARGUING. At any rate, it can't be allowed in the present (or any) article. The talk pages are more, much more, than enough for that purpose. YohanN7 (talk) 12:38, 13 May 2016 (UTC)

I do not approve of a content fork which is purposed to make a designated NPOV battle arena and a serious article, and I doubt it would be efficient. When WP:CIR is the problem, WP:ROPE is the solution. TigraanClick here to contact me 13:34, 13 May 2016 (UTC)
I agree with Tigraan that two articles isn't a great idea. There is only one second law of thermodynamics, so there should only be one article on it. It originated in classical thermodynamics, but the modern understanding of it stems from quantum statistical mechanics. With competent editing it should not be difficult to make that clear. If Chjoaygame won't allow such editing to happen, he should be topic-banned from all physics articles, not just quantum physics (as is the case for now). Waleswatcher (talk) 14:49, 13 May 2016 (UTC)
I agree with both of you. I simply raised the question. We do have two articles on entropy. This article refers first thing in the lead to entropy, then later to entropy (classical thermodynamics) (only). A fork would have resolved such issues, and the definition dispute (not only Chjoaygame favors a classical definition), but good editing will too. YohanN7 (talk) 08:44, 16 May 2016 (UTC)

Enough discussion?[edit]

It seems that a clear consensus has now emerged: #2 is too detailed and too technical, and that either #1, #3, or something similar is preferable (this consensus includes even Chjoaygame as per his/her comment above). Therefore I'm now going to edit the lede, using #1 and #3 as models. Please feel free to improve it. Waleswatcher (talk) 16:53, 20 May 2016 (UTC)


The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

The Second Law is about entropy maximization, not heat transfer[edit]

Entropy is a measure of progress towards the state of thermodynamic equilibrium. It is NOT a measure of disorder, whatever that means anyway. That state of maximum entropy occurs when all unbalanced energy potentials have dissipated. Those energy potentials need not only involve molecular kinetic energy, that is, temperature. They include gravitational potential energy of molecules as well. The Clausius statement is now understood by us physicists to be merely a corollary of the Second Law which only applies when only kinetic energy affects entropy - hence only in a horizontal plane in a gravitational field. When the world starts to understand entropy correctly, it will be realized that there can be an increase in entropy associated with DOWNWARD heat transfer in a planet's troposphere, and this also happens in other force fields, notably in a vortex tube along any radius as there is heat from the cooling central regions to the outer warming regions. There is no other valid explanation for planetary core and surface temperatures. It happens when a prior state of thermodynamic equilibrium (which DOES have a non-zero temperature gradient vertically in a planet's troposphere) is disturbed with the absorption of new thermal energy at the higher altitudes - as happens mostly in the morning. That's how the surface warms each morning even under thick cloud cover. The colder clouds do NOT raise the surface temperature with their radiation. The surface temperature cannot be explained with solar radiation, or any other radiation. — Preceding unsigned comment added by 202.172.115.20 (talk) 01:19, 18 July 2016 (UTC)

Fine, the second law of Thermodynamics still deals with trends (maximization of entropy over time), right? --154.69.29.251 (talk) 14:51, 30 October 2016 (UTC)
The Second Law of Thermodynamics only holds of an isolated system. The system that people like 202.172.115.20 and Doug Cotton have described in terms like the above is clearly far from isolated.

Why there MUST be a non-zero temperature gradient in a force field like gravity[edit]

The state of maximum entropy (which the Second Law says will be approached) is called "thermodynamic equilibrium" by physicists. In that state entropy is homogeneous and there are thus no unbalanced energy potentials. In other words, nothing can happen spontaneously as no net energy tends to move across any internal plane. But molecular gravitational potential energy is just as relevant as molecular kinetic energy, the mean of which is related to temperature. If you consider a horizontal plane in the troposphere, at thermodynamic equilibrium the pressure from above the plane equals that from below the plane. But, by the Ideal Gas Law, pressure is proportional to the product of temperature and density. At thermodynamic equilibrium there is also no net transfer of mass across any internal plane. It follows that there must be equal numbers of molecules passing upwards across that horizontal plane as there are crossing downwards. Hence, since the pressure is equal, the mean kinetic energy of the molecules passing upwards equals the mean kinetic energy of those passing downwards. Now, gravity increases the percentage chance that a molecule will move downwards after a collision, that probability thus becoming a little above 0.5. Hence there must have been a lower density of molecules above the plane than below it, and so we have a stable density gradient which results from the Second Law process of entropy maximization. Likewise, because gravity accelerates the motion of molecules between collisions, there must have been a lower mean molecular kinetic energy above the plane and a higher one below the plane. Hence the Second Law process of entropy maximization is what causes the stable temperature gradient in any planet's troposphere. Inter-molecular radiation between radiating molecules works against the gravitationally induced temperature gradient, reducing it by up to about a third of its theoretical magnitude, as we know happens in more moist regions. But the fundamental IPCC assumption that the surface temperature would have been the same as that found at the radiating altitude in the absence of radiating molecules is wrong. — Preceding unsigned comment added by 202.172.115.20 (talk) 01:44, 18 July 2016 (UTC)

The Second Law of Thermodynamics only holds of an isolated system. The system that people like 202.172.115.20 and Doug Cotton have described in terms like the above is clearly far from isolated. Vaughan Pratt (talk) 05:52, 12 July 2017 (UTC)