I was checking about the stark effect in the hdyrogen in this site for $n=2$. Now, there are four states to deal with, $\psi_{200}, \psi_{21-1}, \psi_{210}, \psi_{211}$. The perturbation hamiltonian $H_1 = -eEz$. To calculate the degeneracy splitting, we need $\langle2lm|z|2l'm'\rangle$.

This site goes as following: $$ 0 = \langle2lm|[L_z, z]|2l'm'\rangle = (m-m')\langle2lm|z|2l'm'\rangle $$

Therefore, all terms where $m\neq m'$ are zero. And so, assuming the matrix lines are in that order: $|200\rangle$, $|211\rangle$, $|210\rangle$, $|21-1\rangle$ he claims the matrix is of the form: $$ H = \begin{bmatrix} 0 & 0 & \gamma & 0 \\ 0 & 0 & 0 & 0 \\ \gamma & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 \\ \end{bmatrix}, \quad\quad \gamma = \langle200|z|210\rangle $$

However, all the terms in the main diagonal are $\langle2lm|z|2lm\rangle$, and therefore, we have $m=m'$, and thus one shouldn't claim such terms are zero. But, I can't see why should the main diagonal be zero, other than by direct calculation. However, if the main diagonal is not zero, the splitting won't be equidistant. And I guess this is wrong.

Speaking of direct calculation, have a look for instance, in this one: $$ \psi_{210} = \frac{1}{\sqrt{24}}a^{-3/2}\frac{r}{a}\exp\left(-\frac{r}{2a}\right)\left(\frac{3}{4\pi}\right)^{1/2}\cos\theta $$

And, using $z=r\cos\theta$, we have for this coefficient in the main diagonal: $$ \langle210|z|210\rangle = \int_0^\infty\frac{1}{24}a^{-3}\frac{r^2}{a^2}\exp\left(-\frac{r}{a}\right)\frac{3}{4\pi}r^3dr \int_0^\pi\cos^2\theta\sin\theta d\theta \int_0^{2\pi}d\phi \\ \langle210|z|210\rangle = \frac{1}{24}\frac{1}{a^5}\frac{3}{4\pi}\frac{2}{3}2\pi \int_0^\infty r^5\exp\left(-\frac{r}{a}\right)dr \quad\mbox{ since }\quad \int_0^\pi\cos^2\theta\sin\theta d\theta = \frac{2}{3} $$

However, after some manipulations from the definition of gamma function, we can find: $$ \int_0^\infty r^5\exp\left(-\frac{r}{a}\right)dr = 5!a^6 $$

Therefore: $$ \langle210|z|210\rangle = \frac{5!a}{4\cdot 6} = 5a\neq 0 $$

Clearly not zero.


There must a mistake in your calculations since the integral of <210|z|210> must be zero and that' s because the multiplication of two spherical harmonics gives an even function and z is an odd function, so the integral is zero.

Now, because the parity of the spherical harmonics is given by $P=(-1)^l $ and because the above integral must be zero, we conclude that there is a transition rule: $\Delta l = 1 \neq 0 .$

Hope this helps.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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