I am trying formally derive the projection of the energy eiegenstates of the 1D quantum harmonic oscillator into the $x$ basis $$ \phi_n(x) = \langle x | n \rangle = \langle x | \frac{{a^{\dagger}}^{n}}{\sqrt{n!}} | 0 \rangle = \frac{{a^{\dagger}}^{n}}{\sqrt{n!}} \, \phi_0(x) $$

I'm not sure how to make the last step

$$ \langle x |\frac{{a^{\dagger}}^{n}}{\sqrt{n!}} | 0 \rangle = \frac{{a^{\dagger}}^{n}(x)}{\sqrt{n!}} \langle x | 0 \rangle $$

where I use ${a^{\dagger}}^{n}(x)$ to clarify that the operator is in the $x$ basis.

I feel I need to use the resolution of identity

$$ \int \langle x | \frac{{a^{\dagger}}^{n}}{\sqrt{n!}}| x' \rangle \langle x'| 0 \rangle \: \mathrm{d} x'$$

I can do this integral for a single creation operator

$$ \int \langle x | a^{\dagger}| x' \rangle \langle x'| 0 \rangle \: \mathrm{d} x' = \int \langle x | \gamma X - i \epsilon P| x' \rangle \langle x'| 0 \rangle \: \mathrm{d} x' $$

where $ \gamma = \sqrt{\frac{m \omega}{2 \hbar}}$ and $ \epsilon = \sqrt{\frac{1}{2m \omega \hbar}}$. Using the matrix representations of the $X$ and $P$ operators in the $x$ basis

$$ \langle x |\gamma X - i \epsilon P| x' \rangle = x' \gamma \delta(x - x') + \hbar \epsilon\delta^{(1)}(x - x')$$

where $\delta^{(1)}(x)$ is the first derivative of the Dirac delta function. So the integral is

$$ \phi_1(x) = \int \left[ x' \gamma \delta(x - x') + \hbar \epsilon\delta^{(1)}(x - x') \right] \phi_0 (x') \: \mathrm{d} x' = \left[ \gamma x - \hbar \epsilon\frac{\mathrm{d}}{\mathrm{d} x} \right]\phi_0 (x)$$

So how do I continue with this derivation? Do I make some sort of inductive step? Is this even the right path to go down, is there a much simpler method?

  • 1
    $\begingroup$ You've already done it! What do you even want help with? $\endgroup$
    – Prahar
    Apr 20, 2017 at 13:15
  • $\begingroup$ To check the recursive generation of Hermite polynomials, consult WP. $\endgroup$ Apr 20, 2017 at 14:35
  • $\begingroup$ That is, your already know that $\phi_{n+1}(x)\propto \left ( x- {d\over dx}\right ) \phi_n(x)$. $\endgroup$ Aug 1, 2020 at 15:29

1 Answer 1


I have not checked the details and this is too long for a comment but it looks right and yes it is induction. Instead of starting with $\phi_n(x)=\langle x\vert \frac{(\hat a^\dagger)^n}{\sqrt{n!}}\vert 0\rangle$, start with \begin{align} \phi_{n+1}(x)&=\frac{1}{\sqrt{n+1}}\langle x\vert \hat a^\dagger\vert n\rangle\, ,\\ &= \frac{1}{\sqrt{n+1}}\left[\gamma x-\hbar \epsilon \frac{d}{dx}\right] \phi_n(x) \tag{1} \end{align} where you obtain (1) by repeating your steps starting with $\vert n\rangle$ rather than $\vert 0\rangle$. You can then start the induction.


Your Answer

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

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