# Would $[\hat{Q},\hat{H}]$ correspond to an observable? [closed]

Would $[\hat{Q},\hat{H}]$ correspond to an observable? Where $\hat{Q}$ is an observable and $\hat{H}$ is the Hamiltonian.

Surely that would just mean that $[\hat{Q},\hat{H}]$ would commute i.e. = 0?:

$[\hat{Q},\hat{H}]\phi_{n} = \hat{Q}\hat{H}\phi_{n} - \hat{H}\hat{Q}\phi_{n} = \hat{Q}E_{n}\phi_{n} - \hat{H}q_{n}\phi_{n} = q_{n}E_{n}\phi_{n} - E_{n}q_{n}\phi_{n} = 0$ ? Hence the commutation does NOT correspond to an observable?

## closed as unclear what you're asking by ACuriousMind♦, innisfree, Danu, John Rennie, Kyle KanosJun 5 '15 at 13:22

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

• An observable is just a self-adjoint operator. Why do you suppose it commutes with $H$, or rather, why do you assume $\hat{Q}\phi_n = q_n \phi_n$? What's the actual question here? – ACuriousMind Jun 4 '15 at 20:18
• Not all observables are conserved/commute with the Hamiltonian – innisfree Jun 4 '15 at 20:18
• And furthermore, why should 0 not be an observable? (Although a boring one) – Sebastian Riese Jun 4 '15 at 20:19
• You've a hidden assumption that they can be diagonalized simultaneously, which is equivalent to assuming they commute. – Omry Jun 4 '15 at 20:42

$[A, B]$ for two observables $A$ and $B$ is an observable if, and only if, $A$ and $B$ commute.

Proof: $$[A, B]^\dagger = (AB)^\dagger - (BA)^\dagger = B^\dagger A^\dagger - A^\dagger B^\dagger = BA - AB = -[A, B].$$

Note: An observable is any Hermitian operator. The commutator of two Hermitian operators is anti-Hermitian, as the proof shows. $0$ is an observable, but a "boring" one. (Actually any real number multiplied by the identity is an observable).

• I like this +1, but in what sense is $0$ an observable? – innisfree Jun 4 '15 at 21:34
• It is a corner case, admittedly. But it is a Hermitian operator, in all states we can give the probability ($p = 1$) of measuring the eigenvalue 0, all states are eigenstates with the eigenvalue 0. – Sebastian Riese Jun 4 '15 at 21:55

As noted by Sebastian, $[Q,H]$ will be anti-Hermitian and therefore generally not an observable (except in the trivial case).

However $i[Q,H]$ is an important observable. This corresponds to the classical Poisson bracket which can be see in the following formula,

$$\frac{d \langle Q\rangle}{d t} = \frac{i}{\hbar} \langle [H,Q]\rangle + \langle \frac{\partial Q}{\partial t} \rangle.$$