2
$\begingroup$

I have a question about the meaning of complex conjugation in time reversal symmetry in quantum mechanics.

$T$-symmetry in classical mechanics is defined simply by the substitution $t \to -t$. If I have an external magnetic field it is not enough and I have to substitute $ \textbf{B} \to - \textbf{B} $. This makes sense because reversing time makes the "external current" generating the magnetic field run backwards, therefore a time symmetry that acts on the system as a whole is indeed supposed to reverse the signs of the magnetic fields as well.

In quantum mechanics T symmetry is given by an operator that acts on a generic wave function as $ T \psi\left(\textbf{x}, t \right) = \psi^{*} \left(\textbf{x}, -t \right) $. The meaning of $t \to -t$ is clear but what about complex conjugation? I know it makes Schroedinger equation invariant if the Hamilton is invariant but what does it have to do with time reversal? Is there a way to justify it like we justify the correspondence $ \textbf{B} \to - \textbf{B} $ in electromagnetism? Because if not it appears to me that $T$ kind of has to do with reversing time but it isn't really a time reversal of the system.

$\endgroup$
3
  • $\begingroup$ The Schroedinger equation is not going to be a priori time-reversal symmetric. It depends on whether the Hamiltonian is time-reversal symmetric. $\endgroup$ Commented Nov 2, 2020 at 21:38
  • $\begingroup$ Okay, assuming we have a T symmetry in the Hamiltonian then. Let me correct the question $\endgroup$
    – Masterme
    Commented Nov 2, 2020 at 21:43
  • $\begingroup$ You can define time-reversal to act however you want. Your choice may or may not be a symmetry of your system though, so some choices are more useful than other. You may want to have a look at this PSE post for more on this. $\endgroup$ Commented Nov 2, 2020 at 22:36

1 Answer 1

2
$\begingroup$

Time reversal operator $T$, when acting on $i$, must result in $TiT^{-1} = -i$. This comes from the fact that $TxT^{-1} = x, TpT^{-1} = -p$, and commutation relation $[x, p] = i\hbar$.

Combine this fact with linearity of time reversal operator, we conclude that $T$ is antiunitary operator, and can be decomposed in the form $KU$, where $U$ is unitary and $K$ is complex conjugation operator. (For reference, it is proved by Wigner that all symmetries of quantum mechanics must be unitary or anti-unitary)

$U$ can of course vary depending on the system you are working on.

$\endgroup$
4
  • 1
    $\begingroup$ This is the same as requiring that T leaves the commutation relations unchanged. Why do we want that? $\endgroup$
    – Masterme
    Commented Nov 2, 2020 at 22:26
  • 1
    $\begingroup$ It is not 'requiring' commutation relation unchanged, but requiring appropriate transformation rules for x and p. $\endgroup$
    – hwang
    Commented Nov 2, 2020 at 22:30
  • $\begingroup$ commutation relation [x, p] = i is, in some sense, axiom of quantum mechanics. Also, by the definition of time reversal, x should go to x and p should go to -p. $\endgroup$
    – hwang
    Commented Nov 2, 2020 at 22:32
  • $\begingroup$ So if you agree with that commutation relation, and agree with the fact that under your transformation x->x and p->-p, your transformation ''should include" complex conjugate. Now, if you assume such operation is a 'symmetry' of your system, then we can further conclude your operator is anti-unitary. $\endgroup$
    – hwang
    Commented Nov 2, 2020 at 22:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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