Talk:Thermodynamics
|
|
| Contents |
General discussion
Why did user CDC revert my edits to this page (as 129.170.x.x)? I am attempting to transform the page into a more logically coherent sequential exposition of the basics of thermodynamics and its relationship to statistical mechanics, and I can't for the life of me figure out why my edits were reverted. 64.222.115.30
Vhung wrote what looks like a brand new Thermodynamics article. Why? Why not try to adapt the old one, or, if the old one is irremediably bad, why not move it to a /Talk page? Basically, we don't want to have to vote about which version is the best. That's not the wiki way... --LMS 0th-law:
- It is possible to build a thermometer.
- That is: If objects A and B are
in thermal equilibrium with each other, and objects B and C are in thermal equilibrium with each other, then objects A and C are also in thermal equilibrium with each other.
- Two objects are in thermal equilibrium
with each other, if their macroscopic properties, such as electrical resistance or volume, do not change with time when these objects are brought into thermal contact with each other.
Excised old version
1st-law:
- Convervation of Energy
2nd-law:
- Degradation of Energy (irreversibilities)
- "Nothing goes without loss"
- concept of Entropy (s)
- T.ds = du + p.dv
- T.ds = dh - v.dp
In lay terms,
- You can't win.
- You can't even break even.
- And you can't get out of the game.
3rd-law:
- At absolute zero, the Entropy of a Perfect crystal is zero.
Would it be unfair of me to suggest that this page requires some serious refactoring? :-) --LMS
Refactoring thermodynamics is best attempted by madmen or fools, or by foolish madmen such as Clifford Truesdell.
There should be a reference to the relationship between Thermodinamics and Statistical Mechanics
Rephrased second law. I am very wary about making statements that involve the entire universe.
I don't like the version of the third law given here. I just did a search on the Internet, and it looks like it's fairly popular, but I have no idea why. The first thing someone's going to say when they see it is "who cares about perfect crystals". What about
- All processes cease as temperature approaches zero
or
- Absolute zero can only be approached asymptotically
I also don't like the ones that say S(T=0)=0, because in the usual derivation of the third law, S(T=0) is a constant, which is set to zero merely for convenience. It's similar to the way you can set gravitational potential energy at r -> ∞ to zero.
If no-one answers, I'll just change it. -- Tim
Okay, no-one answered. It's changed. -- Tim Starling
"If A and B are at the same temperature, and B and C are at the same temperature, then A and C are also at the same temperature."
Actually, I'm under the strong impression that the zeroth law is: "If A and B are in thermal equilibrium, and B and C are in thermal equilibrium, then A and C are also in thermal equilibrium."
The way it is written down now, the zeroth law becomes a trivial fact. Since '=' is an equivalence relation, 'having the same temperature' is necessarily transitive. If no-one disagrees, I'll change the definition shortly. (More information can be found in Status_of_the_zeroth_law_of_thermodynamics.) --Victor Gijsbers
- Quite so, well spotted. I like your new article, but note that you don't need to put underscores in links, so [[Status of the zeroth law of thermodynamics]] is fine. Also, although titles are case sensitive, the first letter is automatically capitalised, so [[status of the zeroth law of thermodynamics]] goes to the right place too. See Wikipedia:How to edit a page. -- Tim Starling 23:46 7 Jul 2003 (UTC)
- Ok, thanks for the tip. --Victor Gijsbers
Current statement of the third law: "All processes cease as temperature approaches zero."
This is not only wrong, it is positively meaningless in the context of thermodynamics. In this theory, temperature is defined for equilibrium states only (the theory only talks about equilibrium states), so no processes take place whenever a temperature is defined, not just at absolute zero. Something along these lines could perhaps be the 3rd law of statistical mechanics, but it cannot be a law od thermodynamics.
I suggest replacing it by Buchdahl's (1966, The concepts of classical thermodynamics) formulation: The entropy of any given system attains the same finite least value for every state of least energy. --Victor Gijsbers
How about, The entropy of a system reaches some finite minimal value for every state of minimal energy. Pizza Puzzle
You say that the idea of a change or process taking place is absolutely meaningless. This is incorrect, indeed the bulk of any undergraduate level thermodynamics course deals with precisely such processes. This is done formally by defining a process as being a succession of infintesimally separated equilibrium states. This is called a "quasistatic" process. There is no time variable as such, instead relationships between various thermodynamic properties are calculated. I'm taking this from Chapter 1 of Sears & Salinger, by the way.
The statement of the 3rd law I was paraphrasing (again from S&S) is:
- It is impossible to reduce the temperature of a system to absolute zero in any finite number of operations
It's true that I allude to the presence of a time variable, but I do this to maintain comprehensibility. I don't believe it significantly affects the accuracy of the statement. After all, I don't define the word "cease" -- I could easily make up a definition on the spot which fits in with the quasistatic picture.
The advantage of this statement is that a lay-person can understand it. A person not trained in thermodynamics has very little conception of it means for entropy to approach zero. "You can't cool something to absolute zero" is a statement that your Mum would understand, and it is sufficiently rigorous at the level of this article.
An alternative statement, again saying exactly the same thing, is this:
- The derivative of temperature with respect to any macroscopic property, with entropy held constant, is zero when temperature is zero.
But please, save it for an article with a more advanced target audience. Specifically, save it for third law of thermodynamics, where a proper explanation can be given of such complexities.
-- Tim Starling 14:16 8 Jul 2003 (UTC]
Can someone explain to me the reason the edits of 129.170.29.119 are unacceptable? PAR 22:03, 7 Apr 2005 (UTC)
1st Law
"The sum of heat flowing into a system and work done by the system is zero. " - No, non-zero heat may flow into a system (e.g. a metal bar) but the system do no work. It might just get hot.
Perhaps the more standard "The heat flowing into a system is equal to the change in internal energy minus the work done by the system." would be better.
- Oops. Good point, sorry. -- Tim Starling 09:40, 31 Aug 2003 (UTC)
"The work exchanged in an adiabatic process depends only on the initial and the final state and not on the details of the process. " - What is "work exchanged"? You obviously have a context in mind here but it is not spelt out. Anyway, work is "done", not exchanged.
- Quite so. Those alternate statements were added by Reddi, I didn't really pay much attention to their accuracy. He claimed to get them out of physicsworld.wolfram.com, maybe there was a bit of paraphrasing going on there as well. See [1] (http://www.wikipedia.org/w/wiki.phtml?title=Thermodynamics&oldid=1146861) for a pre-Reddi revision. -- Tim Starling 09:40, 31 Aug 2003 (UTC)
- They are from physicsworld.wolfram.com (and can be cited from other sources). JDR
- The way I remember this stuff, just like the zeroth law is a statement about the existence of a transitive thermodynnamic equilibrium and can be used to define temperature, the first law (as it is stated) requires an explanation of what is meant by heat transfer (or absence thereof: adiabatic) and can be used to define work. I need to research the details so I can add it to the article. — Miguel 03:15, 2004 Feb 26 (UTC)
Without context information, it's not clear (to me) that the "entropy" version is really the same as the 1st law. I think that you need to clarify this statement.
2nd Law
I think that the laws as stated on this page need some work.
The 2 versions of the zeroth law are too similar. Please pick one.
- I think I left them because I was trying to avoid pissing Reddi off for the fifth time in one day. Obviously I should have paid a bit more attention. -- Tim Starling 09:40, 31 Aug 2003 (UTC)
- Don't worry about that JDR
Have already posted comments on 1st law
"A system operating in a cycle cannot produce a positive heat flow from a colder body to a hotter body " - so how does a refrigerator work?
- By using external work. The right statement is "A system operating in a cycle without using external work..." Miguel 03:17, 2004 Feb 26 (UTC)
- See Lord Kelvin's work in Thermodynamics ...
"A system operating in contact with a thermal reservoir cannot produce positive work in its surroundings" - Does this mean that a steam engine doesn't work?
- See Rudolf Clausius work in Thermodynamics ...
- Delete them all and join me in the fun pastime of Reddi watching - see User:Tim Starling/Reddi watchlist. He's made a few edits outside Nikola Tesla recently that I didn't bother checking up. Welcome to Wikipedia, Sokane. Please sign entries on talk pages with ~~~~, which is automatically converted to a name and a date. Note that you don't need to use <p> here, just pressing enter twice does the trick. I'll put a few more tips on your user talk page. -- Tim Starling 09:40, 31 Aug 2003 (UTC)
- 1st, don't delete the factual info ... 2nd, Have you updated that list lately? =-D JDR
We have: "The entropy of a closed macroscopic system never decreases". Shouldn't this say "isolated" since we are giving the system theoretic definition: "closed systems: exchanging energy (heat and work) but not matter with their environment". I have changed this to "thermally isolated" as in the Entropy article. Physics texts usually define "closed" to mean "thermally isolated". I came here because of a discussion with someone who was using the statements in this article to argue that the Earth is essentially a closed system so it's entropy must increase.
- Whether the Earth is a closed system depends on the level of approximation. The Earth's atmosphere loses hydrogen and gains small amounts of matter from solar wind, the interstellar medium and meteorites. It also exchanges work on other celestial bodies (notably through lunar and solar tides). Finally, the Earth clearly exchanges heat with its "environment" - it radiates energy away as a black body because it is hotter than the CMB. The energy radiated should (on average) match that absorbed from solar radiation, resulting in net entropy production. The Earth is also not in local equilibrium internally, so entropy gets constantly produced. — Miguel 03:58, 2004 Oct 5 (UTC)
Time
User:Moink said "took out "thermodynamics is not concerned with time"... not true in my experience". It's true in my experience. In all my time studying thermodynamics and statistical physics, I don't think I saw a single t. Time is mentioned obliquely -- for example it's often noted that a Carnot engine would take an infinite amount of time per stroke. But the statement you removed was, I felt, quite accurate. Can you explain? -- Tim Starling 06:03, Jan 13, 2004 (UTC)
- Think about the second law of thermodynamics: Entropy increases with time. If that's not a dynamic law I don't know what is. Almost all the thermodynamics I've ever done (I'll admit it is not my central field, fluid dynamics is) has included time in some way or another. Here's a problem for you: A reservoir is separated into two cells with a thermally insulating membrane. The temperature on side A is 400K, on side B is 100K. The membrane is removed. What is the resulting temperature as a function of space and time? If you agree that this is indeed a thermodynamic problem, then we can say that thermodynamics is more than thermostatics. moink 22:29, 13 Jan 2004 (UTC)
- Thermodynamics uses time qualitatively, but not quantitatively. This is the fundamental point the original author was trying to make. The question of temperature as a function of space and time is not thermodynamics. The question of entropy differences between initial and final states is thermodynamics. The term "thermostatics" reminds me of quasistatic equilibrium, and interesting concept in itself. -- Tim Starling 23:23, Jan 14, 2004 (UTC)
- If that's the point the article is trying to make, it doesn't make it very clearly. And I disagree that the study of flow with temperature changes is not thermodynamics. This article states that "Thermodynamics is the study of energy, its conversions between various forms..." and if a viscous flow with temperature differences is not a study of the conversions of energy between forms, I don't know what is. moink 22:29, 15 Jan 2004 (UTC)
- I don't think specific mechanisms of heat transport such as conduction or convection are considered part of thermodynamics. -- Tim Starling 00:25, Jan 16, 2004 (UTC)
- I've taken a class called "thermodynamics" and one called "fluid dynamics". We never took time into consideration in thermo, but we did do the problem Moink suggested in fluid dynamics. I think thermo tells you what a system tends to do, but not how it does it.
- I think part of what's going on here is that most people have taken only an introductory thermo course. Introductory courses tend to deal only with equilibrium states, mostly because they're easier to solve for beginners, and also because solving for an equilibrium state gives you quite a bit of valuable information. But I do think thermo, in its more advanced incarnations, does deal with how it gets there. moink 22:15, 16 Jan 2004 (UTC)
- I've studied thermodynamics, kinetic theory and statistical physics to third year level as a physics major. All up I've done about about 50 hours of lectures on the subject. -- Tim Starling 10:44, Jan 17, 2004 (UTC)
- Could this be pointed out in the article? JDR
- Maybe not heat convection and conduction, but definitely transferring energy between types, like heat to velocity. moink 22:15, 16 Jan 2004 (UTC)
Velocity is more the domain of kinetic theory not thermodynamics.
From Sears & Salinger chapter 1-1:
- Thermodynamics is an experimental science based on a small number of principles that are generalizations made from experience. It is concerned only with macroscopic or large-scale properties of matter and it makes no hypotheses about the small-scale or microscopic structure of matter. From the principles of thermodynamics one can derive general relations between such quantities as coefficients of expansion, compressibilities, specific heat capacities, heats of transformation, and magnetic and dielectric coefficients, especially as these are affected by temperature. The principles of thermodynamics also tell us which of these relations must be determined experimentally in order to completely specify all the properties of the system...
- Thermodynamics is complementary to kinetic theory and statistical thermodynamics. Thermodynamics provides relationships between physical properties of any system once certain measurements are made. Kinetic theory and statistical thermodynamics enable one to calculate the mangitudes of these properties for those systems whose energy states can be determined.
-- Tim Starling 10:44, Jan 17, 2004 (UTC)
Phenomenological study of time-dependent phenomena such as fluid mechanics or chemical reaction kinetics is separate from kinetic theory or statistical mechanics. Just like in thermodynamics on can introduce a phenomenological "heat capacity" whose value can be calculated from first principles using statistical mechanics, one can introduce a phenomenological "heat conductivity" whose value can only be calculated using kinetic theory. See my stub on non-equilibrium thermodynamics for more on this. Miguel 03:26, 2004 Feb 26 (UTC)
I think the issue is almost a semantical one with the occasional more generalized usage of "thermodynamics" to exclusively mean "equilibrium thermodynamics". But both "equilibrium thermodynamics" and "nonequilibrium thermodynamics" contain the word thermodynamics, so it doesn't seem quite right to claim that time is not involved in thermodynamics. Even introductory thermodynamics courses typically deal with things like heat transfer rates, so I don't think it's an educational level issue so much as a semantical one. Perhaps the divide between the two can be explained more clearly without excluding time from thermodynamics. — Cortonin | Talk 10:53, 24 Mar 2005 (UTC)
The quality of the article
I am actually surprised to see that the article's quality has decreased (in my very humble opinion) since I was last involved with it about a year ago. The presentation has lost all coherence and it reads just like a collection of equations!
I think some discussion needs to take place as to what exactly we want to emphasize about thermodynamics. — Miguel 20:03, 2005 May 19 (UTC)