The Robertson uncertainty relation is

$\sigma^2_A \sigma^2_B \geq \big|\dfrac{1}{2} \langle\{A,B\}\rangle - \langle A \rangle \langle B \rangle \big|^2 + \big| \dfrac{1}{2i} \langle [A,B] \rangle \big|^2.$

Where $\sigma^2_X$ is the variance of the operator $X$ and $\{A,B\}$, $[A,B]$ are the anti-commutator and the commutator of the Hermitian operators $A$ and $B$, respectively.

The uncertainty relation is more commom presented in the form

$\sigma^2_A \sigma^2_B \geq \big| \dfrac{1}{2i} \langle [A,B] \rangle \big|^2.$

Where there are commom physical examples which have that satisfied, e.g. $[x,p] \geq \dfrac{\hbar}{2}$, but these examples have $\big|\dfrac{1}{2} \langle\{A,B\}\rangle - \langle A \rangle \langle B \rangle \big|^2=0$.

I am trying to find a quantum system where the term $\big|\dfrac{1}{2} \langle\{A,B\}\rangle - \langle A \rangle \langle B \rangle \big|^2 \neq 0$, so that the lowest limit of the product of the variances of $A$ and $B$ have a dependence on the latter. So, to answer my question, it is necessary to give a possible physical system where $\big|\dfrac{1}{2} \langle\{A,B\}\rangle - \langle A \rangle \langle B \rangle \big|^2 \neq 0$ for $A$ and $B$ Hermitian.

Any help or ideas are welcome.


1 Answer 1


You might profit from calculating a few expectation values for, e.g. the oscillator, for which $a^\dagger |n\rangle= \sqrt{n+1} | n+1\rangle$. Take $A=a$ and $B=a^\dagger$, so that $$ [a,a^\dagger ]=1, \qquad \{ a,a^\dagger \} =a a^\dagger + a^\dagger a=1+2N. $$

Look at the first excited state, $|1\rangle$, so $a|1\rangle=|0\rangle$, so your expectation values are $$ \langle 1| a|1\rangle=\langle 1|a^\dagger| 1\rangle=0,\\ \langle 1| \{ a,a^\dagger \} |1\rangle= 3,\\ \langle 1| [a,a^\dagger ]|1\rangle= 1, $$ so that, for this state $$ \sigma_a^2 \sigma^2_{a^\dagger}\geq 9/4 +1/4= 10/4. $$ The anticommutator, of course, is not a constant, unlike the commutator, nor should you expect it to be.

  • $\begingroup$ Thanks for the attention and how would I interpret $a$ and $a^{\dagger}$ as physical quantities in this example? $\endgroup$
    – Slayer147
    Commented Apr 11, 2018 at 22:58
  • 1
    $\begingroup$ They are the usual linear combinations of x and p. You could rerun all this with plain x and p on the 1st excited site of the oscillator, the first termite polynomial times the usual Gaussian.... $\endgroup$ Commented Apr 11, 2018 at 23:49

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.