6

https://www.quantamagazine.org/20140416-times-arrow-traced-to-quantum-source

This article says that entanglement drives the arrow of time of increasing entropy towards thermodynamic equilibrium:

Cups of coffee cool, buildings crumble and stars fizzle out, physicists say, because of a strange quantum effect called “entanglement.”

It is hard to believe that entropy is rooted in a quantum phenomenon. Is there any other possible explanation?

AccidentalFourierTransform
  • 53,248
  • 20
  • 131
  • 253
ScienceJournalist01
  • 205
  • 1
    The second law of thermodynamics can be explained classically but only if you assume that time moves in one particular direction. The problem is that an arrow of time can't (superficially) be justified based on the fundamental laws. This quantum entanglement explanation (which I don't understand) supposedly gives time a direction, although it doesn't yet appear to have been accepted into the mainstream. – lemon Apr 09 '16 at 17:34
  • 3
    Entropy is rooted in statistics, which is independent of the QM/CM boundary. The actual question related to entanglement is whether it is responsible for the "distribution" of time, i.e. the fact that your clock and my clock, despite being physically separated, can stay in near perfect synchrony for a long period of time. To be frank, without an actual quantum theory of spacetime such claims are, at best, handwaving with giant foam hands. – CuriousOne Apr 09 '16 at 19:34
  • Entanglement makes a subsystem of the entangled whole be in a mixed state, they say that what we call entanglement entropy is the source of all entropy. https://en.wikipedia.org/wiki/Entropy_of_entanglement – ScienceJournalist01 Apr 10 '16 at 03:54
  • related: this excellent answer by Arnold Neumaier http://physics.stackexchange.com/questions/22745/does-measurement-quantum-in-particular-always-increase-the-total-entropy?rq=1 – Wolpertinger Apr 10 '16 at 17:33

1 Answers1

3

Asking a question about the foundations of statistical mechanics is a good way to start a fight, so don't expect a clear consensus on this.

Stripping away the hype, what these papers try to do is establish that an isolated quantum system will, under certain (but general) conditions, evolve to a state that looks like thermodynamic equilibrium locally, even though the system remains in a pure state and therefore, if taken as a whole, always has zero entropy. If you accept that the universe, or at least some subsystem, is fundamentally quantum mechanical and can be described as a closed system, then this is your ultimate explanation for how equilibration works. However, there are some clear logical leaps there that have not necessarily been worked out, such as whether this picture of spreading entanglement entropy is still a useful picture in a macroscopic, fully decohered system.

I would second CuriousOne in saying that the laws of statistical physics are ultimately rooted in probability rather than quantum theory, and in that sense are probably more fundamental than any other theory we have. It is no coincidence that our most secure beliefs about what quantum gravity says come from making it compatible with thermodynamics. Here is a very nice article from a little while back that gives a similar sentiment (paywalled, sorry). So I would agree that it is, at least, premature to claim that thermodynamics laws are ultimately a byproduct of quantum physics. However, the thermodynamics of isolated quantum systems is clearly a very important particular case.

Rococo
  • 7,671
  • 1
  • 23
  • 54
  • A wave which is in a superposition of states becomes a particle with a probabilistic distribution of states when measured. This measurement is what converts superposition to probability distribution. Double slit experiments have shown this. – ScienceJournalist01 Apr 10 '16 at 17:21
  • Hi @QuantumJournalist- Yes, I am familiar with decoherence (I've even done experiments of this kind myself! With entangled photons it's pretty easy nowadays), but that is similar but not quite the same as the issue here of entanglement-mediated equilibration. Unless I am missing your point, in which case I encourage you to elaborate. – Rococo Apr 12 '16 at 03:15