Part VI, Chapter 2 | Page 2 of 3

The Stark Effect in Hydrogen

Linear energy splitting of the n=2 level in an external electric field

The Stark effect in the $n=2$ level of hydrogen is the quintessential application of degenerate perturbation theory. It reveals a linear splitting of energy levels in the electric field -- qualitatively different from the quadratic Stark effect seen in non-degenerate levels.

The n=2 Hydrogen States

At the $n=2$ level, hydrogen has a 4-fold degeneracy. The four states are:

$$|2,0,0\rangle \quad (\text{2s state}), \quad |2,1,0\rangle, \quad |2,1,1\rangle, \quad |2,1,-1\rangle \quad (\text{2p states})$$

All four states have the same unperturbed energy:

$$E_2^{(0)} = -\frac{13.6\text{ eV}}{4} = -3.4\text{ eV}$$

This degeneracy is accidental (specific to the Coulomb potential) and arises from the hidden SO(4) symmetry of hydrogen. The $2s$ state has $l=0$ (even parity), while the $2p$ states have $l=1$ (odd parity).

The Perturbation

An external uniform electric field $\vec{\mathcal{E}} = \mathcal{E}_0\hat{z}$ produces the perturbation:

$$\hat{H}' = e\mathcal{E}_0 z = e\mathcal{E}_0 r\cos\theta$$

where we use $z = r\cos\theta$ in spherical coordinates. This perturbation is odd under parity ($z \to -z$) and preserves the $z$-component of angular momentum ($[\hat{H}', \hat{L}_z] = 0$).

Selection Rules

Before computing the $4 \times 4$ $W$-matrix, symmetry eliminates most matrix elements:

Rule 1: $\Delta m = 0$

Since $[\hat{H}', \hat{L}_z] = 0$, the perturbation cannot change $m$. Therefore $\langle n,l',m'|r\cos\theta|n,l,m\rangle = 0$ unless $m' = m$.

Rule 2: $\Delta l = \pm 1$

Since $\cos\theta \propto Y_1^0$, the angular integral vanishes unless $l' = l \pm 1$ (from the properties of spherical harmonics and the Wigner-Eckart theorem).

Consequence:

The states $|2,1,1\rangle$ and $|2,1,-1\rangle$ have $m = \pm 1$ and cannot couple to any other $n=2$ state (since $|2,0,0\rangle$ has $m=0$ and $|2,1,0\rangle$ has $m=0$). Only $|2,0,0\rangle$ and $|2,1,0\rangle$ can mix.

The W-Matrix

In the basis $\{|2,0,0\rangle, |2,1,0\rangle, |2,1,1\rangle, |2,1,-1\rangle\}$, the $W$-matrix is:

$$W = \begin{pmatrix} 0 & W_{12} & 0 & 0 \\ W_{12}^* & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 \end{pmatrix}$$

The diagonal elements vanish because $\hat{H}'$ is odd under parity and each basis state has definite parity. The only non-zero element is:

$$W_{12} = \langle 2,0,0|e\mathcal{E}_0 r\cos\theta|2,1,0\rangle$$

Computing the Matrix Element

Using the hydrogen wave functions:

$$\psi_{200} = \frac{1}{4\sqrt{2\pi}}\frac{1}{a_0^{3/2}}\left(2 - \frac{r}{a_0}\right)e^{-r/2a_0}, \quad \psi_{210} = \frac{1}{4\sqrt{2\pi}}\frac{1}{a_0^{3/2}}\frac{r}{a_0}e^{-r/2a_0}\cos\theta$$

The matrix element separates into radial and angular parts:

$$W_{12} = e\mathcal{E}_0 \underbrace{\int_0^\infty R_{20}(r)\, r\, R_{21}(r)\, r^2\, dr}_{\text{radial}} \times \underbrace{\int Y_0^0 \cos\theta\, Y_1^0\, d\Omega}_{\text{angular}}$$

The angular integral gives $1/\sqrt{3}$ (using standard spherical harmonic identities). The radial integral evaluates to:

$$\int_0^\infty R_{20}(r)\, r\, R_{21}(r)\, r^2\, dr = -3\sqrt{6}\, a_0$$

Combining everything:

$$\boxed{W_{12} = -3e\mathcal{E}_0 a_0}$$

This is real, so $W_{12} = W_{21}$.

Solving the Secular Equation

The $4 \times 4$ matrix block-diagonalizes. The $|2,1,\pm 1\rangle$ states are unaffected ($E^{(1)} = 0$). The $2 \times 2$ block for $|2,0,0\rangle$ and $|2,1,0\rangle$ is:

$$\det\begin{pmatrix} -E^{(1)} & -3e\mathcal{E}_0 a_0 \\ -3e\mathcal{E}_0 a_0 & -E^{(1)} \end{pmatrix} = 0$$
$$(E^{(1)})^2 - (3e\mathcal{E}_0 a_0)^2 = 0$$

The eigenvalues are:

$$\boxed{E^{(1)} = \pm 3e\mathcal{E}_0 a_0}$$

The Energy Splitting

The four originally degenerate $n=2$ states split into three levels:

$$E_2 = -3.4\text{ eV} + \begin{cases} +3e\mathcal{E}_0 a_0 & \text{(one state: } |+\rangle = \frac{1}{\sqrt{2}}(|200\rangle - |210\rangle)\text{)} \\ 0 & \text{(two states: } |211\rangle, |21{-}1\rangle\text{)} \\ -3e\mathcal{E}_0 a_0 & \text{(one state: } |-\rangle = \frac{1}{\sqrt{2}}(|200\rangle + |210\rangle)\text{)} \end{cases}$$

The total splitting is:

$$\Delta E = 6e\mathcal{E}_0 a_0$$

This is the linear Stark effect: the energy splitting is proportional to the first power of the electric field. This is in sharp contrast to the quadratic Stark effect for non-degenerate levels.

Linear vs Quadratic Stark Effect

The distinction between linear and quadratic Stark effects is fundamental:

$$\text{Linear: } \Delta E \propto \mathcal{E}_0 \quad (\text{degenerate levels, first-order PT})$$
$$\text{Quadratic: } \Delta E \propto \mathcal{E}_0^2 \quad (\text{non-degenerate levels, second-order PT})$$

The linear effect is much larger for typical field strengths. Physically, the "good" states $|\pm\rangle$ are superpositions of $s$ and $p$ states, which have permanent electric dipole moments:

$$\langle \pm|z|\pm\rangle = \mp 3a_0 \neq 0$$

These mixed-parity states possess permanent electric dipole moments that interact linearly with the applied field.

Physical Interpretation

The "good" states $|\pm\rangle = \frac{1}{\sqrt{2}}(|200\rangle \mp |210\rangle)$ are no longer eigenstates of parity. By mixing opposite-parity states, the electron cloud becomes displaced along the $z$-axis, creating a permanent electric dipole moment. The state $|+\rangle$ has the electron displaced against the field (higher energy), while $|-\rangle$ has it displaced with the field (lower energy).