Would like to ask a question, but first i would like to say Hello Everybody in a way that plays the system, since some geniouses decided that one should not be able to say hello in a question.

The uncertainty principle in quantum mechanics is well known and considered one of most basic properties of natural reality. The 2nd Law of thermodynamics is also well known and also considered one of the most basic processes of natural reality.

The uncertainty principle uses and is related to Planck's constant. Planck's constant has the dimensions of action and in a statistical mechanics approach, also relates nicely with the partitioning of the phase-space providing the basic measure for the entropy functional (this answer provides a nice outline of this).

Apart from that, there are relatively recent papers which relate the Heisenberg Uncertainty Principle in quantum mechanics directly and intuitively to the 2nd Law of Thermodynamics.

Is this relation correct? And if so can we derive one from the other?

Thank you

PS. One can also check this question, which although not the same, is related in an interesting way.


anna's answer is accepted since by mentioning the derivation of (part of) the 2nd law from unitary dynamics, answers the question at least in one way. Please consider this as still open so you can add another answer. There are more alternatives (and one of which is my stance, ie thermodynamics -> uncertainty)

  • $\begingroup$ Nikos, when you look at the PhysicsSE home page it shows just the first paragraph of your question i.e. just the complaint about not being able to say hello. It would increase the chance of your question being answered if you move the first paragraph to the end of your question. $\endgroup$ May 30, 2014 at 7:43
  • 1
    $\begingroup$ @John, thank you very much for your comment and indeed it has a rationale (not seen so far in .se), however, isnt the title of the question supposed to be the shortest summary of the question itself? And additionally, lets not be "robots" in our communication, humour, phrasing, expression et al. in general can be half the fun :) $\endgroup$
    – Nikos M.
    May 30, 2014 at 7:48
  • $\begingroup$ @John, and of course one always takes the responsibility for how to phrase the question and the associated results of not being answered fast or whatever, i would not like the editor to take away my hello if i want to say so. i dont talk to the editor i address people $\endgroup$
    – Nikos M.
    May 30, 2014 at 7:51
  • $\begingroup$ @guntbert, i do not agree with the edit to remove the chatty, unnecessary beginning (aka a hello message with an edge indeed), and it does not seem to give me a way to undo... $\endgroup$
    – Nikos M.
    May 30, 2014 at 14:41

2 Answers 2


That is a very nice question and the article (which I did not know) is interesting as it states that thermodynamics can't be true and the microscopic uncertainty relations false at the same time. Regardless of whether it's true or not, it is an interesting claim as it departs from the traditional wisdom of the full bottom-up approach.

I haven't checked whether the argument they put forward is sound or not but the way they do it is definitely reasonable for different reasons:

  • If uncertainty relations have to do with thermodynamics, the link is likely to be at the information level. That is because, on the one hand, uncertainty relations imply that there exist such things as incompatible events which in turn says that the maximum amount of information we can have on a quantum system is a set of eigenvalues of a complete set of commuting observables. This information is seemingly always less than or equal to the classical information we could have on such a system. On the other hand, Landauer has proposed that the minimum amount of energy one has to give to change one bit of information is $k_BT \ln 2$ in order not to violate the second principle of thermodynamics. This limit is very important has it was a first reasonable solution to the Maxwell demon problem.

  • At first glance, the authors of the paper seem to propose a thought experiment akin to the Maxwell's demon one but accounting for the quantum properties of the particles. From their reasoning they conclude that one can in principle obtain an infinite energy source (i.e. we would be back to the original problem of the Maxwell's demon) unless the uncertainty relations are true.

I still need to study more thoroughly the paper, but my take on it is that they have showed that one can seemingly violate the Landauer bound (and hence the second principle of thermodynamics) over one cycle if uncertainty relations are not enforced.

It has been shown recently with a classical system that indeed one could violate the original Landauer bound in an erasure procedure of information. However, this does not violate a more elaborate version of the Landauer bound that takes into account the rate of success of the erasure protocol and I wonder if the result from Hanggi et al. could be interpreted in this broader context.

Unfortunately, I have no definite opinion neither about the work you point out not about the actual claim that thermodynamics implies uncertainty relations. But I do think that if it were to be true, then assessing the cost of information erasure would be the right direction to look into and also, in my view, how fluctuation relations behave in the quantum world needs to be better understood to provide more definite arguments on this issue.

  • $\begingroup$ thanks for the article on fluctuation relations, will have to take a look at it $\endgroup$
    – Nikos M.
    May 30, 2014 at 14:37
  • $\begingroup$ On the other hand an interpretation of uncertainty relations is exactly that the phase-space cannot be compressed arbitrarily (reality is indeed real), moreover information can also be derived itself. Regarding thermodynamics of computation (Landauer, Bennett et al), i (at least at present) do not consider whether it is based on erasure or on acquisition of information a problem, i think that both can be used equally well $\endgroup$
    – Nikos M.
    May 30, 2014 at 14:59
  • $\begingroup$ Maxwell's daemon and variations (Szillard Engine, Szillard-Popper engine, etc..) are frequently used metaphors and frameworks in thermodynamic reasoning related to information. There are however objections (e.g Norton's no-go reasoning) $\endgroup$
    – Nikos M.
    May 30, 2014 at 15:08
  • $\begingroup$ There was a recent paper on arxiv (circa 2012), that actually derived gravity from entropy and information reasoning (will have to find it, i have it somewhere) $\endgroup$
    – Nikos M.
    May 30, 2014 at 15:11
  • $\begingroup$ The paper on fluctuations is quite good, has the math from the Clausius inequality,to Fokker-Planck and Path Integrals.However it is phrased and based on older/gloomy notions of entropy (heat death,disorder etc..). i suggest to look at Prigogine's work which is one of those to breath new life into the genre, and dispelling the myths of gloomy notions of entropy. Finally a more elementary analysis on entropy,fluctuations, etc.. on another post of mine $\endgroup$
    – Nikos M.
    May 30, 2014 at 15:34

You say your self :

The uncertainty principle in quantum mechanics is well known and considered one of most basic properties of natural reality.

In fact quantum mechanics and its postulates and laws are the underlying framework on which any classical theory is built.

The "laws" of classical theories emerge from the underlying quantum mechanical framework. In the paper you quote they claim that :

More precisely, we show that violating the uncertainty relations in quantum mechanics leads to a thermodynamic cycle with positive net work gain, which is very unlikely to exist in nature.

As an experimentalist I am in no position to check whether their conclusion is correct, this is the work of peer review in journals, and it has been accepted in Nature and , I hope, peer reviewed. Well done if it is correct, because it is one more validation of the underlying quantum mechanical framework.

I do not know whether it is related to the statement in the wiki article :

In statistical thermodynamics, the second law is a consequence of unitarity in quantum mechanics

It seems from the references to be connected to the many worlds interpretation , so this new derivation might be a more mainstream connection of the quantum mechanical framework to the second law.

  • $\begingroup$ nice answer, thanks, although i do not favor the mainstream interpretation, specifically the Copenhagen interpretation and the ("artificial") fragmentation into "quantum" and "classical" worlds. So i try to find connections (and interpretations) in these directions, i think this is relevant. Nevertheless, i like the answer very much, will wait for any other answer, also $\endgroup$
    – Nikos M.
    May 30, 2014 at 8:18
  • $\begingroup$ more specifically i tend towards quantum mechanics as a result of statistical mechanics (or sth like this), of course if they are equivalent (in some way), this makes things easier. On the other hand if this is the way it is , this is the way it is. No point involved in physics otherwise $\endgroup$
    – Nikos M.
    May 30, 2014 at 8:21
  • $\begingroup$ Yes there is a result which "derives" the 2nd law from unitarity of the quantum formalism, however, can we go the other way? And moreover how about the measurement collapse and 2nd law? i think these are interesting by themselves but also, in what might be plausible quantum phenomena at large $\endgroup$
    – Nikos M.
    May 30, 2014 at 8:31
  • $\begingroup$ There is also the time-asymetry of the 2nd law and the time-symmetric formalism of current quantum mechanics, There is a paper on an alternative formulation of quantum formalism (encompassing entropy), i can give the link if you like $\endgroup$
    – Nikos M.
    May 30, 2014 at 8:34
  • 1
    $\begingroup$ Look Nikos, at my age (74) I am to be allowed to have decided on my basic framework of physics, having worked professionally for 35 years and seen the standard model build up gradually. In my view it is QM that it is at the bottom the mathematics of nature and everything else emerges from it. Quantum measurement paradoxes are navel gazing by philosophically inclined theorists, again in my opinion. $\endgroup$
    – anna v
    May 31, 2014 at 4:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.