1
$\begingroup$

I have been asked to decompose the state $|\Psi \rangle$, with wavefunction in spherical coordinates $\langle r \, \theta \, \phi|\Psi \rangle = \sqrt{\frac{3}{\pi}}e^{-r}\sin{\theta}\cos{\phi}$ into the $|L = 1 \, m_z\rangle$ basis.

I know $\langle \theta \, \phi|\ell \, m\rangle = Y^m_\ell(\theta, \phi)$, and $\Psi(r,\theta,\phi) = \sqrt{2}e^{-r}(Y_1^{-1} - Y_1^1)$, but I'm unsure whether this is already the proper decomposition into that basis. It honestly sounds like a bogus question to me.

My point is, I can't express $|\Psi\rangle = \frac{1}{\sqrt{2}}(|L = 1 \, m_z=-1\rangle - |L = 1 \, m_z=1\rangle)$ without leaving out the radial part of the wavefunction. Is there a way to do this, or is there no proper answer to this question apart from the decomposition of the wavefunction itself into spherical harmonics?

$\endgroup$

1 Answer 1

2
$\begingroup$

Your decomposition is fine. Clearly you cannot write down $|\Psi\rangle$ solely by angular momentum eigenstates, or spherical harmonics for that matter, since they give only the angular part of the wavefunction. So either you write your state as a product of radial wavefunction and spherical harmonics, as you did $$\Psi(r,\theta,\phi) = \sqrt{2}e^{-r}\left(Y^{-1}_1(\theta,\phi)-Y^{1}_1(\theta,\phi)\right)$$ or you can use the braket notation $$|\Psi\rangle = N|n\rangle(|1,-1\rangle-|1,1\rangle)$$ where $N$ is some normalization constant.

In any case you'll need to specify three quantum numbers $n,l,m$ which are encoded either into the radial part of the wavefunction or the anguar part. There's no way to get rid of the radial part since by doing so you'll be giving away a quantum number.

$\endgroup$
9
  • 1
    $\begingroup$ I've been knocking myself in the head for far too long, this makes sense. Just one question, I assume I'd be defining $|n\rangle = e^{-r}$ myself, right? $\endgroup$ Commented Jun 9, 2020 at 21:23
  • $\begingroup$ Well that's not good enough. You should check the radial wavefunction solutions to the hydrogen atom $R_{nl}(r)$ which will depend on the $n$ quantum number. When you find the right $R_{nl}$ then you'll have the $n$ to write down the associated ket $|n\rangle$ $\endgroup$ Commented Jun 9, 2020 at 21:28
  • $\begingroup$ If I'm not mistaken there could be an error in the definition of the $\Psi$ wavefunction since, on the top of my head, the only radial solution proportional to $e^{-r}$ should be $R_{10}$ which is in contrast with the $l=1$ of the angular part. But I could be wrong here $\endgroup$ Commented Jun 9, 2020 at 21:31
  • $\begingroup$ It's not defined to be an electron in a hydrogen-like atom, just the wavefunction of a spinless particle. $\endgroup$ Commented Jun 9, 2020 at 21:34
  • $\begingroup$ Ohw ok, then it's perfectly fine. But still you cannot identify a ket $|n\rangle$ with a wavefunction. Since you're only interested in the radial part, and so on the angular momenta, you can just call it $|n\rangle$ and leave it at that since you don't have the specifics of the radial part. It'll just count as a placeholder for it. $\endgroup$ Commented Jun 9, 2020 at 21:37

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.