In quantum mechanics, we know $\dot{\psi}=-\frac{i}{\hbar}H\psi$,

but why is $U\dot{\psi}=-\frac{i}{\hbar} \left(UHU^\dagger \right) U\psi$?

Does it mean $UHU^\dagger = H$ ? I think $UU^\dagger H = H$, but why can we change the order of the matrices here?


3 Answers 3


You're overthinking this, assuming $U$ is unitary:

$$ U\dot\psi= -\frac{i}{\hbar} UH\psi=-\frac{i}{\hbar} UH\mathbb 1\psi= -\frac{i}{\hbar} UHU^\dagger U\psi.$$

$U$ need not be the time evolution operator and it need not commute with $H$ for this to work, it can be any unitary. This is just saying that if you write $\psi$ in another basis then it evolves with the Hamiltonian written in the new basis. (Or equivalently that a rotated vector evolves with the rotated Hamiltonian).

  1. If the Hamiltonian $\hat{H}$ does not depend on time, and $U$ is supposed to be the time-evolution operator, then $$\hat{U}~=~\exp\left(-\frac{i}{\hbar}\hat{H}\Delta t\right),\tag{A}$$ which commutes$^1$ with $\hat{H}$, so that $$UHU^{\dagger} ~=~ H,\tag{B}$$ cf. OP's question.

  2. If the Hamiltonian $\hat{H}$ do depend on time, then eqs. (A) & (B) need to be modified, cf. e.g. this Phys.SE post.


$^1$ A function $f(\hat{H})$ of $\hat{H}$ commutes with $\hat{H}$, cf. e.g. this & this Phys.SE posts.

  • $\begingroup$ But OP's calculation would be correct even if the Hamiltonian depended on time, right? $\endgroup$ Nov 30, 2020 at 14:31
  • $\begingroup$ Moreover OP never mentioned $U$ is time evolution $\endgroup$ Nov 30, 2020 at 14:32
  • $\begingroup$ Thanks! Could you please explain a little about why H and U commute if U is the time-evolution operator and H does not depend on time? $\endgroup$
    – peachnuts
    Nov 30, 2020 at 17:18
  • 1
    $\begingroup$ I updated the answer. $\endgroup$
    – Qmechanic
    Nov 30, 2020 at 17:35

user2723984 is correct. However, the 2nd part of your question is unresolved: if the Hamiltonian commutes with itself at different times, then the only operator in $U$ is $H$ and, as $H$ commutes with itself, the order of the operators may then be changed.

  • $\begingroup$ OP never mentioned $U$ is time evolution I think $\endgroup$ Nov 30, 2020 at 14:32
  • $\begingroup$ Yes, I assumed this since it is standard notation. $\endgroup$
    – PrawwarP
    Nov 30, 2020 at 14:34

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.