I'm not clear why, in case $\hat A$ and $\hat B$ are self-adjoint, then $(\hat A \hat B + \hat B \hat A )$ is also self-adjoint. In order to show that $(\hat A \hat B + \hat B \hat A )=(\hat A \hat B + \hat B \hat A )^\dagger$, I have seen that the property $(\hat a \hat b)^\dagger=\hat b^\dagger \hat a^\dagger$ is applied, so $(\hat A \hat B + \hat B \hat A )^\dagger=\hat B^\dagger \hat A^\dagger + \hat A^\dagger \hat B^\dagger$. However, I don't get with this last is equal to $\hat A \hat B + \hat B \hat A $. Could someone give me a hint?
Asked
Active
Viewed 85 times
1
-
1Firstly, this belongs to maths SE, not physics. Secondly, the property $(AB)^+=B^+A^+$ applies to products, not to sums. – FGSUZ Aug 31 '20 at 09:37
-
@FGSUZ it can be applied to products if you use the propertie $(\hat A + \hat B)^\dagger=\hat A ^\dagger + \hat B ^\dagger$ – Invenietis Aug 31 '20 at 09:43
-
That is not a product... – FGSUZ Aug 31 '20 at 10:02
1 Answers
7
$$\begin{align} \left( \hat A\hat B+\hat B\hat A\right)^\dagger &=\left(\hat A\hat B\right)^\dagger+\left( \hat B\hat A\right)^\dagger \tag{1 }\\ \tag{2} &= \hat B^\dagger\hat A^\dagger+\hat A^\dagger\hat B^\dagger \\ \tag{3} &= \hat A^\dagger\hat B^\dagger+ \hat B^\dagger\hat A^\dagger \\ \tag{4 }&= \hat A\hat B + \hat B\hat A.\end{align}$$
Distribute the Hermitian conjugate between terms.
$(\hat X\hat Y)^\dagger = \hat Y^\dagger \hat X^\dagger$ as you have correctly said.
Matrix addition is commutative.
$\hat A^\dagger=\hat A$ and $\hat B^\dagger =\hat B$ as they are both self-adjoint.
Charlie
- 685
-
I was stuck with something so silly (point 4). Thank you very much! – Invenietis Aug 31 '20 at 09:42
-
-
1Just to add a word of caution: If $\hat A,\hat B$ are general unbounded (self-adjoint) operators one has to ensure that the expression $\hat A\hat B+\hat B\hat A$ is defined at least on a dense subspace of the underlying Hilbert space. For example it could happen that $\operatorname{im}(\hat A)\cap D(\hat B)={0}$ in which case $\hat B\hat A$ does not have a meaning (cf. Chapter VIII.5 in Reed & Simon, "Functional Analysis"). – Frederik vom Ende Sep 13 '20 at 15:25
-
1@FrederikvomEnde A good point to add, but probably quite a lot more advanced than is needed for the target audience of OP's question. – Charlie Sep 13 '20 at 15:46