next up previous
Next: The Preperation of Rydberg Up: The Aufbau Principal, Kramers Previous: Kramer's Relation and the

Selection Rules and Available Transitions

A change in energy levels for a Rydberg atom in a circular state must obey selection rules so that $ \Delta m = \pm 1$ or 0, and $ \Delta l = \pm 1$ . Thus an atom in a Rydberg state under guarded environmental condtions can only transition as

$\displaystyle n \ \rightarrow \ n-1, \ l \ \rightarrow \ l-1, \vert m\vert \ \rightarrow \ \vert m\vert -1$

Thus a Rydberg atom approximates a two-level system.

We demonstrate the theory behind elementary selection rules with a simple example. A two-state system in the presence of puturbing Hamiltonian can be described as,

$\displaystyle \Psi(t)=c_a(t)\psi_ae^{-iE_at/\hbar} + c_b(t)\psi_b e^{-iE_bt/\hbar}$ (3.1)

Its evolution is described by

$\displaystyle (H+H'(t))\Psi = i\hbar \frac{\partial \Psi}{\partial t}$

from which we have,

$\displaystyle c_a(t)\langle\psi_a\vert H' \vert\psi_a\rangle e^{-iE_at/\hbar} +...
...\vert H' \vert\psi_b\rangle e^{-iE_bt/\hbar}=i\hbar \dot{c_a(t)}e^{-iE_at/hbar}$ (3.2)

Rejoining convention and writing $ H'_{ij}=\langle \psi_i\vert H'\vert\psi_j\rangle$ , $ \omega_0 = (E_b - E_a)/\hbar$ , and assuming (as is warented in the experiments we discuss) that the diagonal of the perturbing part of the Hamiltonian is zero, we can obtain a set of equations for the prefactors,
$\displaystyle \dot{c_a}$ $\displaystyle =$ $\displaystyle -\frac{i}{\hbar}H'_{ab}e^{-i\omega_0t}c_b$ (3.3)
$\displaystyle \dot{c_b}$ $\displaystyle =$ $\displaystyle -\frac{i}{\hbar}H'_{ba}e^{i\omega_0t}c_a$ (3.4)

Following [4] we consider a basic hydrogen atom in the $ n=2$ state in an electric field so that $ H'=-eEr\cos\theta$ . As we showed in a previous compendium, the wave functions for hydrogen, $ n=1, n=2$ , are


$\displaystyle \psi_{100}$ $\displaystyle =$ $\displaystyle \frac{1}{\sqrt{\pi a^3}}e^{-r/a}$  
$\displaystyle \psi_{200}$ $\displaystyle =$ $\displaystyle \frac{1}{\sqrt{8\pi a^3}}\left(1-\frac{r}{2a}\right)e^{-r/2a}$  
$\displaystyle \psi_{210}$ $\displaystyle =$ $\displaystyle \frac{1}{\sqrt{32 \pi a^3}}\frac{r}{a}e^{r/a}\cos \theta$  
$\displaystyle \psi_{211}$ $\displaystyle =$ $\displaystyle -\frac{1}{\sqrt{64 \pi a^3}}\frac{r}{a}e^{-r/2a}\sin \theta e^{i\phi}$  
$\displaystyle \psi_{21-1}$ $\displaystyle =$ $\displaystyle \frac{1}{\sqrt{64 \pi a^3}}\frac{r}{a}e^{-r/2a}\sin \theta e^{-i\phi}$  

The perturbation matrix is simple in that all but one of these will be even in $ r\cos \theta$ , so the perturbation matrix ( $ \langle n' l' m'\vert H'\vert nlm\rangle$ ) will have all zero elements except

$\displaystyle \langle 100\vert H'\vert 210 \rangle = - \frac{eE}{4\sqrt{2}\pi a...
...2 \cos^2\theta r^2 \sin\theta dr d\theta \ d\phi = -\frac{2^8}{\sqrt{2} 3^5}eEa$ (3.5)



  $ \psi_{100}$ $ \psi_{200}$ $ \psi_{210}$ $ \psi_{211}$ $ \psi_{21-1}$
$ \psi_{100}$ 0 0 $ -\frac{2^8}{\sqrt{2} 3^5}eEa$ 0 0
$ \psi_{200}$ 0 0 0 0 0
$ \psi_{210}$ $ -\frac{2^8}{\sqrt{2} 3^5}eEa$ 0 0 0 0
$ \psi_{211}$ 0 0 0 0 0
$ \psi_{21-1}$ 0 0 0 0 0


Thus we see that under this perturbation, the $ 100 \leftrightarrow 210$ levels are ``selected' and we have an approximate two-state system under the right conditions (ideal).

In general, we can derive selection rules for m and l transitions. In the case of m, we consider that $ [L_z,z]=0$ so that

$\displaystyle 0=\langle n'l'm'\vert[L_z,z]\vert nlm\rangle=(m'-m)\langle n'l'm'\vert z\vert nlm\rangle$ (3.6)

That $ [L_z,x]=i\hbar y$ gives,

$\displaystyle (m'-m)\langle n'l'm'\vert x\vert nlm\rangle=i\langle n'l'm'\vert y\vert nlm\rangle$ (3.7)

Finally, $ [L_z,y]=-i\hbar x$ gives,

$\displaystyle (m'-m)\langle n'l'm'\vert y\vert nlm\rangle=-i\langle n'l'm'\vert x\vert nlm\rangle$ (3.8)

Thus $ (m'-m)^2 =1$ or $ \langle\vert x,y\vert\rangle=0$ . For the $ l,l'$ case, it can be shown that

$\displaystyle L^2,[L^2,r]]=2\hbar^2(rL^2 + L^2r)$

From this we can show that

$\displaystyle 2\hbar^4(l(l+1)+l'(l'+1))\langle n' l' m'\vert r\vert nlm\rangle = \hbar^4(l'(l'_1)-l(l+1))^2)\langle n' l' m'\vert r\vert nlm\rangle$ (3.9)

Rewriting

$\displaystyle l'(l'+1)-l(l+1)=(l'+l+1)(l'-l), \ $   and$\displaystyle \ 2(l(l+1)+l'(l'+1))=(l'+l+1)^2+(l'-l)^2-1 $

We thus conclude that $ l'=l\pm1$ and $ m'=m\pm1 \ $   or$ \ m$
next up previous
Next: The Preperation of Rydberg Up: The Aufbau Principal, Kramers Previous: Kramer's Relation and the
tim jones 2007-04-09