Talk:Thermodynamics/Archive 3

Archive 1

Archive 2

Archive 3

Wikipedian user boxes for those interested in thermodynamics

This user has a keen interest in Thermodynamics.

— Preceding unsigned comment added by Sadi Carnot (talk • contribs) 02:11, 28 November 2006‎ (UTC)

Thermodynamics is a Class of Theories

One source of confusion in the writing of this article was to treat thermodynamics as one theory, or -- seemingly truth. It is not true: it is science. I've learned at least a dozen different theories that can be expressed by varying the axioms or definitions. Prigogine's non-equilibrium theories number at least four.

When geologists solve geological problems, they will often divide the problem into many smaller ones, then apply one physical theory to each of these. This doesn't mix ideas. Mixing them confuses what you have when you finish.

I suggest you choose which theories to present, or sketch each and link the more important ones to separate articles. In particular, it is bad to mix classical and statistical statements in the same paragraph.

(The good news is, I've been sent a monograph by Jouguet, which I'm translating into English. Many will agree I've contributed enough here. So, it's time to break the champagne! Bye; and best of luck with this most important of articles!)

Bruce Bathurst, PhD Geologist (talk) 23:12, 29 September 2011 (UTC)

Fewer References?

It's occurred to me that the 'theories' of thermodynamics, which number as many as the varied references, may be keeping this article from a 'good' status. Every book is internally consistent with one classical thermodynamics (definitions & axioms), but they mix like apples and oranges: some define equilibrium using a characteristic function, some use homogeneity (and some wrt time, Fermi wrt the environment). In short, each reference gives a different set of opinions about what is important and what isn't.

The article might use one approach (one classical theory), which would dictate the important topics to cover, which might permit a cleaner writing style.

As an example, theories for chemists, which start with Clausius's two laws, use energy and entropy and need a quick existence theorem for temperature: the zeroth law. If one starts with fundamentals, which use thermometers, this law can be dropped.

Physicists are interested in Nernst's Theorem, and it's important in cryogenics; but no natural science but astrophysics can use it. It could be minimized. (Also it requires quantum statistical thermodynamics rather than a steam engine.)

Ironically, reducing the number of authorities might help the article. That said, I can recommend Fermi, E. 1937. Thermodynamics. NJ: Prentice-Hall; and possibly Pippard, A.B. 1957. The Elements of Classical Thermodynamics. Cambridge: University Press. Geologist (talk) 06:27, 31 October 2011 (UTC)

Structure as a branch of physics

The statement:Thermodynamics is one of the best structured branches of physics is obvious and does not need any source.--79.119.210.1 (talk) 10:14, 17 November 2011 (UTC)

It is vague and thus needs clarification and reference. What is structure (of a branch of physics)? How is it evaluated? By whom? Materialscientist (talk) 10:35, 17 November 2011 (UTC)

Without reference, it is complete and internally consistent, within the limits of its assumptions. With reference, "It is the only physical theory of universal content concerning which I am convinced that, within the framework of the applicability of its basic concepts, it will never be overthrown." A. Einstein, Autobiographical Notes in Albert Einstein: Philosopher-Scientist P. A. Schilpp (ed.), Library of Living Philosophers, vol VII, p.33, Cambridge University Press, London, 1970.

This refers to classical thermodynamics, not the explanation and extensions of it provided by statistical mechanics. PAR (talk) 16:37, 17 November 2011 (UTC)

The above statement is not obvious, and some reliable sources would say that it is not true. That Einstein thinks that classical thermodynamics will never be overthrown is not evidence about its structure. And obviousness is not a reason for a statement to go into a Wikipedia article.Chjoaygame (talk) 20:30, 17 November 2011 (UTC)

If there ever existed an equilibrium system, then one can't disprove Einstein's 'doubt': it becomes obvious. Remember, 1st law, 2d law, equilibrium are all axioms. Einstein scientifically states he doubts the theory will be overthrown (not quite the same as fail). A mathematician (and thermodynamics is a field of applied mathematics) could substitute 'A, B, C are true' for the axioms and prove the theory complete and consistent (proof by interpretation, offered by the first sentence of this paragraph). Several axiomatizations have been published. Still, I too haven't a clue what 'structure' means (in this context). Energetics? Geologist (talk) 09:10, 7 January 2012 (UTC)

Searched the article for 'structure' and found a reference by Serrin. He's an applied mathematician, so 'structure' as used by Mr or Ms ip may mean the interpretation of theory 'A, B, C' mentioned above as classical equilibrium thermodynamics. Both Bridgman and Einstein believed thermodynamics differed inexplicably from other theories in physics. Geologist (talk) 09:54, 7 January 2012 (UTC)

Structure in this case could refer to the logical structure (self-consistency etc).--79.119.218.100 (talk) 11:15, 9 January 2012 (UTC)

A question which arises next is how is the logical structure of thermodynamics compared to other branches of theoretical physics like classical mechamics and electromagnetism?--79.119.218.100 (talk) 11:34, 9 January 2012 (UTC)

I'm not a physicist. (I only quote them. :-) Neither theory takes heat into account. There may be something very appealing about heat being work degraded to a microscopic scale. One would have to ask. Geologist (talk) 00:53, 13 January 2012 (UTC)

Perhaps an article called Axiomatization of thermodynamics would be useful in this context. — Preceding unsigned comment added by 84.232.141.36 (talk) 15:44, 4 February 2012 (UTC)

Is entropy an extensive variable?

The article says “entropy in general does not fit the above definition of an extensive variable” (Section “Thermodynamic state variables”). But in the articles Entropy and Intensive and extensive properties entropy is declared as an extensive property. What is true? Greetings 2.38.190.84 (talk) 15:50, 15 December 2011 (UTC)

This is a very reasonable question.

The term of art 'entropy' was invented by Rudolf Clausius in the early days of thermodynamics. Its mathematical equivalent had been used previously by William Rankine and by Clausius. It was a variable of state of a closed thermodynamic system. Its definition has been changed over the years, because it is a difficult concept. Nowadays it is admitted by many writers (example Kondepudi 2008) that it may apply also to an open system. It was by one major classical thermodynamicist, Max Planck, early regarded as a quantity in its own right, subject to its own laws, but later Planck admitted that it had a statistical mechanical basis and could not be understood otherwise. Planck's change of mind was due to his discovery of his law of thermal radiation and to work by Ludwig Boltzmann and by Albert Einstein.

Thermodynamics is severely crippled when it cannot access the concept of entropy. Access to the concept of entropy is very safe for the study of systems, open or closed, in strict thermodynamic equilibrium. For systems not in thermodynamic equilibrium, the concept of entropy is fairly safe in the special case of local thermodynamic equilibrium. Away from local thermodynamic equilibrium the concept is unsafe. For example, a top expert on the question, Walter T. Grandy, Jr, is of the view that we do not know even how to calculate the entropy of a worm (Entropy and the Time Evolution of Macroscopic Systems (2008), Oxford University Press, Oxford UK, ISBN 978-0-19-954617-6). Classical thermodynamics has little or nothing to contribute to the definition of entropy for systems far from thermodynamic equilibrium. A probabilistic or statistical definition seems unavoidable. Grandy and Edwin Thompson Jaynes are perhaps the safest guides to trying to understand the probabilistic approach.

Some scientists do not like to feel crippled by ignorance, and have difficulty saying "I don't know", and this prejudices them, for the present problem, to postulate instead of providing reliable reason. Many writers on non-equilibrium thermodynamics thus simply postulate that an entropy function exists when they need it, and they give the impression that this is sound science. But it's often just wishful thinking, without a proper scientific basis. As part of this wishful thinking, they postulate that entropy for their non-equilibrium studies is an extensive variable.

Entropy is assuredly and safely an extensive variable for suitably defined classical thermodynamic systems in thermodynamic equilibrium.

But that does not make it extensive for far-from-equilibrium systems as wished by such wishful thinkers.

The definition of thermodynamic concepts for non-equilibrium systems must take into account that such systems can undergo major structural fluctuations, and such fluctuations will destroy the postulated basis for simple postulates of entropy for such systems. So far as I know, no one has provided a general way of calculating the entropy of practically encountered far-from-equilibrium systems, precisely for this reason. This means that in general, so far as I can find in the literature, entropy is in general not an extensive variable for far-from-equilibrium systems, although many writers wish that it were so. For a careful discussion of this question I can do no better than recommend Grandy's above-mentioned book, especially Chapter 5, headed 'The presumed extensivity of entropy'. After you have read that, perhaps further discussion might be useful.Chjoaygame (talk) 19:24, 15 December 2011 (UTC)

Many thanks for your detailed and long answer. It helped me a lot. Greetings 2.38.190.84 (talk) 19:17, 17 December 2011 (UTC)

I have been out of contact for a few days. I think that one reason why one can be confident that entropy is not extensive for systems far from thermodynamic equilibrium is that, far from thermodynamic equilibrium, the spatial arrangement of the volume elements is very important in the determination of the evolution of the process, but is entirely ignored by a simple addition of elements of entropy.Chjoaygame (talk) 22:35, 23 December 2011 (UTC)

Truly, I don't mean to be a spoiler, but isn't the above similar to arguing that skunks aren't black & white because furry mammals come in many colors?

The nice thing about positivism is that, because equilibrium fails gradually, or gracefully, we can use instruments of varying accuracy and precision to argue its existence: one need only predict certain thermodynamics quantities (within the abilities of the instruments:-) from those being measured.

The question, however, implies entropy exists; either in exact principle (realist) or fuzzy practice (positivist). So, yes, entropy is extensive if its quantity doubles when a replica of the system is added. Enjoyed Chjoayagame's 'realist' analysis, though. Geologist (talk) 08:36, 7 January 2012 (UTC)

Bias against the Existance of Equilibrium in Nature

One of the reasons, IMO, this article is flawed is its insistence that thermodynamics concludes equilibrium is only approached. (This throws unwarranted importance on theories of non-equilibrium thermodynamics.) No theory of thermodynamics I have ever seen makes any statement whether or not equilibrium is attained. (Text writers, of course, may express opinions; the article appears to misrepresent these.)

Classical thermodynamics assumed equilibrium when deriving practical relations, but the history section has 'equilibrium' Bowdlerized, making it highly flawed. Callen's statement arguing for equilibrium in nature has been used as a reference for the opposite statement. Finally, to demonstrate that natural processes are all non-equilibrium, one obscure process (hysteresis in solids--the model of non-equilibrium cycles for decades) was pulled out, seemingly from a hat. This looks like bias.

Closed systems are represented by a system and its surrounding environment, as stated in the article. Thermodynamics is used, as stated in the article, to predict the response to a perturbation.

Geologists have shown that the response keeps pace with the perturbation, such as extremely slow burial. Such a response is usually accurately predictable from the perturbation. According to Prigogine & Defay's definition of equilibrium, this is a path of equilibrium states.

Those definitions that involve time simply mean the system does not drop its Gibbs energy below a limit, an equilibrium value: a high explosive may reach this value in less than a second. In summary, equilibrium is a limiting state, but thermodynamics does not predict it is approached asymptotically. (The remark about detailed balancing and equilibrium requiring microscopic reversibility is wrong: it is the macroscopic rates of these processes that must be equal.)

Equilibrium is simply the correlation of perturbation and response, which allows prediction. Even geologists don't throw up their hands when asked to predict the perturbation from the measured response because the extents of nuclear reactions were unknown when the chemical reactions stopped. (The widely ranging compositions of univariant oxides in viscous rhyolites that equilibrated when temporarily housed in shallow magma chambers fall flawlessly along a straight line predicted by equiibrium. This serves as a test for equilibrium in nature.)

This article won't be considered a good one (well, by me at least) until the above bias is proved either non-existant or corrected. (Good luck, Mr Phelps. :-)

Geologist (talk) 20:00, 2 February 2012 (UTC)

Life is a manifestation of non-equilibrium processes. The main supply of energy that flows through living things on earth is from the massive emission of heat by the sun, a big-time non-equilibrium process.Chjoaygame (talk) 16:42, 26 February 2012 (UTC)

"No theory of thermodynamics I have ever seen makes any statement whether or not equilibrium is attained" - one of the laws of thermodynamics is the second, which is precisely concerned with the approach to equilibrium. The fact is that nothing in our world is in equilibrium, but many systems are close enough that equilibrium thermodynamics is a good approximation. Waleswatcher (talk) 19:37, 26 February 2012 (UTC)

I read once that, during the first atomic bomb test, Enrico Fermi tossed small pieces of paper straight up into the air. During the explosion, the paper moved away from the explosion. From this distance (between equilibrium states), he estimated the energy of the explosion using equilibrium thermodynamics. In any case, 'natural science' includes more than biology. Geologist (talk) 04:28, 9 March 2012 (UTC)

The Second Law states that a characteristic function will fall to a limit: for a common system, the difference in Gibbs Energy between final and initial equilibrium states is less than or equal to zero. Where here does it say equilibrium is not attained (a limit is not reached)? (If it did, would axiomatic theories have a need for a so-called 'zeroth law', which claims equilibrium is attained?) Geologist (talk) 04:28, 9 March 2012 (UTC)

Chemists us isotopes to measure the rate at which equilibrium is attained. A paper by T.S. Lee, referenced below, has a nice summary. One can even do this with a two mixtures initially at equilibrium: the crystals in one beaker have one stable isotope, those in the other have another. Mix the beakers. Sample the crystals at various times. When no isotopic difference can be measured between any two, equilibrium has been reached. (This is off the top of my head, chemists please correct me.)

This example proposes an entire crystal to be a single phase; geologists consider the rim at which growth stops to be a crystalline phase and would claim the time for equilibrium to be reached is zero, using this definition. Both definitions produce the equilibrium composition predicted by classical thermodynamics. Both claim equilibrium is attained, just at different times. Geologist (talk) 04:28, 9 March 2012 (UTC)

The 'zeroth law' is positivist: it claims equilibrium exists when no difference in predicted equilibrium qualities can be observed or measured. The definition used in this article appears to claim 'equilibrium' is a theoretical object, incorrectly predicted to not be attained by either a classical or a statistical theory; or perhaps such a statement gives 'non-equilibrium' theories an undeserved importance. Geologist (talk) 04:41, 9 March 2012 (UTC)

Metamorphic petrologists do not assume that all previous thermodynamic states, now preserved, were at equilibrium: they have their tests. T.S. Lee has these appropriate quotes: 'Needless to say, many reactions are so slow that they never come to equilibrium. An activation energy of 45 kcal. is large enough to prevent any appreciable reaction in a million years at 25 deg. C.' If one want to give it a few more million, 'As has been pointed out by many authors, there is no geological evidence that nitric acid has bee produced at moderate temperatures by the reaction of air with ocean water... .' T.S. Lee, 1959, 'Chemical Equilibrium and the Thermodynamics of Reactions', in Treatise on Analytical Chemistry, I.M. Kolthoff & P.J. Elving, eds., pt.I, v.1, p.185-317. Geologist (talk) 21:59, 12 March 2012 (UTC)

I honestly cannot figure out what you're trying to say with these comments. Do you claim that true equilibrium exists in nature? If so, that's trivial to falsify (the universe is evolving, that suffices). Are you claiming that many systems are close enough to equilibrium that they can be approximated as such? If so, we all agree with you (or at least I do). Is there some specific language in the article you find objectionable? If so, what is it? Waleswatcher (talk) 13:30, 13 March 2012 (UTC)

I claim true equilibrium exists in nature if observations or measuremets do not measurably deviate from those predicted by equilibrium thermodynamics. I claim the surface of a pond remains in equilibrium with the air just above it during a 24-hour period.

I claim thermodynamic theory makes no statement that natural states or even natural processes are equilibrium or non-equilibrium: I believe an experiment is necessary.

A careful reading of the article shows strange & awkward wording throughout to avoid claiming that equilibrium systems exist. Gibbs wrote on whether 'a process would occur spontaneously' (not about equilibrium among phases); neither Duhem, Lewis, nor Guggenheim derived 'equilibrium' relations. I claim this is not good, and ask myself why the article was so written.

I claim 'classical thermodynamics' is synonymous with 'equilibrium thermodynamics'. I claim some systems are in equilibrium, some nearly so, some far from so, from very slow to frozen (no swimming in nitric acid). I claim (as Gibbs did) that stable (water), neutral (water & ice), & unstable (water vapor nuclei) exist. I claim (with Einstein) that classical thermodynamics is the most logically deductive & accurate (predicting) thermodynamic theory. (No quantas necessary.)

I claim more systems exist in nature than the universe. I claim the article should be fixed. Geologist (talk) 00:01, 16 March 2012 (UTC)

chatty edit removed

I reverted two edits (478965001, 478965255; dated 26 Feb 2012) that were chatty editorializing, not about thermodynamics directly, but about people's responses to it and its possible significance. If that editor wants to write about such things in the Wikipedia, I suggest he work out a way of doing it other than by putting chatty comments, not supported by the cited "sources", in the present article which is about physics, not sociology.Chjoaygame (talk) 16:37, 26 February 2012 (UTC)

Waleswatcher's re-wording

Waleswatcher has apparently not read the references that supported the wording change that he re-worded and that he leaves in support of his re-wording.Chjoaygame (talk) 01:11, 14 March 2012 (UTC)

Eh, Wot? One excellent paragraph of Waleswatcher contained 'the present article takes a gradual approach to the subject, starting with a focus on cyclic processes and thermodynamic equilibrium, and then gradually beginning to further consider non-equilibrium systems'. I wish this were true. A rejection of natural, equilibrium states apparently resulted in the less happy 'chemical thermodynamics studies the role of entropy in chemical reactions'. Geologist (talk) 00:27, 16 March 2012 (UTC)

demotion of importance of concept of thermodynamic equilibrium

There has been some complaint recently that the article was underemphasizing the importance of thermodynamic equilibrium. The sentence "Thermodynamic equilibrium is one of the most important concepts for thermodynamics" was not enough to deflect this complaint. Now the new edit has deliberately demoted the importance of the concept of thermodynamic equilibrium by replacing that sentence with "Thermodynamic equilibrium is an important concept in thermodynamics."

Callen 1985 on page 26 writes: "The single all-encompassing problem of thermodynamics is the determination of the equilibrium state that eventually results after the removal of internal constraints in a closed composite system." Callen is one of the most widely cited theoreticians of thermodynamics. I think that statement of his goes too far, and is the work of a theoretician, and I am not particularly enamoured of Callen's presentation, though I recognize that it has merits. (When Callen writes "closed system", we understand that he means "isolated system".)

Most thermodynamics, theoretical as well as practical, works with at least the local thermodynamic equilibrium concept as one of its most important elements, and so I would say that the demotion was not a good move. I would like to restore the sentence "Thermodynamic equilibrium is one of the most important concepts for thermodynamics."Chjoaygame (talk) 14:58, 30 March 2012 (UTC)

The new edit downplays the fact that there are many processes of nature that progress so rapidly or irregularly that they cannot feasibly be followed by currently known thermodynamics, because they are too far from local thermodynamic equilibrium. For example, the new edit might seem to suggest that nature tends to limit her activities to those that are within the scope of a version of non-equilibrium thermodynamics based on the assumption of local thermodynamic equilibrium, which makes their study only "a little more involved but of much more practical importance".

The new edit in the introduction refers to inhomogeneity of systems, studied in engineering, before the concept of homogeneity of a system has been defined in the article. It does not make it clear at this point that there is a close relation between the concept of local thermodynamic equilibrium and the proviso required by the edit, that "the thermodynamic parameters are well defined."

The lead has already told the reader that "Thermodynamics can be applied to a wide variety of topics in science and engineering", but, in the introductory section of the article, the new edit, apparently not concerned with processes in nature, wants to further emphasize the practical importance of the engineering use of thermodynamics.Chjoaygame (talk) 00:24, 31 March 2012 (UTC)

I agree - to downplay the role of equilibrium in thermodynamics is like downplaying the role of space and time in relativity. Local thermodynamic equilibrium is simply equilibrium thermodynamics applied to a continuum of systems. It does not downplay the role of equilibrium, it emphasizes it. Other methods of dealing with non-equilibrium thermodynamics involve perturbations of the equilibrium state, again, emphasizing the importance of equilibrium. The four laws are stated for systems in equilibrium.

Yes, please restore the statement. PAR (talk) 06:42, 31 March 2012 (UTC)

Done.Chjoaygame (talk) 07:12, 31 March 2012 (UTC)