Perturbation Theory and the Variational Method
CSIR NET quantum mechanics: time-independent and degenerate perturbation theory, Stark effect, variational method, Fermi's golden rule, selection rules.
- 20 min
On this page · 15 sections
- Prerequisites
- The Hook: Helium Has No Exact Solution
- Non-Degenerate Perturbation Theory
- Degenerate Perturbation Theory
- The Variational Method
- Time-Dependent Perturbation Theory
- Electric Dipole Transitions and Selection Rules
- Sudden and Adiabatic Limits
- Memory Tricks
- Don’t Confuse
- Common Mistakes
- Practice
- Quick Revision Table
- Key Takeaways
- What’s Next
Prerequisites
- One-Dimensional Problems: Box, Oscillator and Tunnelling: the unperturbed systems, and ladder-operator matrix elements.
- Angular Momentum, Spin and the Hydrogen Atom: hydrogen states and spherical harmonics, needed for the Stark effect and selection rules.
- Operators, Commutators, Dirac Notation and Uncertainty: matrix elements, completeness and parity.
The Hook: Helium Has No Exact Solution
Non-Degenerate Perturbation Theory
Write \( \hat H = \hat H_0 + \hat H' \), where \( \hat H_0\lvert n^{(0)}\rangle = E_n^{(0)}\lvert n^{(0)}\rangle \) is solved and \( \hat H' \) is small. Expanding in powers of \( \hat H' \):
$$\boxed{E_n^{(1)} = \langle n^{(0)}\rvert\hat H'\lvert n^{(0)}\rangle}$$$$\boxed{E_n^{(2)} = \sum_{m\neq n}\frac{\lvert\langle m^{(0)}\rvert\hat H'\lvert n^{(0)}\rangle\rvert^2}{E_n^{(0)} - E_m^{(0)}}}$$$$\lvert n^{(1)}\rangle = \sum_{m\neq n}\frac{\langle m^{(0)}\rvert\hat H'\lvert n^{(0)}\rangle}{E_n^{(0)} - E_m^{(0)}}\lvert m^{(0)}\rangle$$The first-order energy is the perturbation averaged over the unperturbed state. The expansion is valid when \( \lvert H'_{mn}\rvert \ll \lvert E_n^{(0)} - E_m^{(0)}\rvert \) for every pair that couples.
Three structural facts that settle many questions without calculation:
- For the ground state, every denominator in \( E^{(2)} \) is negative, so \( E_0^{(2)} \leq 0 \). Second order always pushes the ground state down.
- Levels repel: the contribution of a coupled level above pushes \( n \) down, and one below pushes it up.
- Parity: if \( \hat H_0 \) is symmetric and \( \hat H' \) is odd (like \( x \) or \( z \)), \( E_n^{(1)} = 0 \) for every non-degenerate level, and the first non-zero shift is second order.
Degenerate Perturbation Theory
If \( E_n^{(0)} \) is degenerate, the denominators \( E_n^{(0)} - E_m^{(0)} \) vanish for the partner states and the formulas above blow up. The fix:
- Build the matrix \( W_{ij} = \langle i\rvert\hat H'\lvert j\rangle \) within the degenerate subspace only.
- Its eigenvalues are the first-order shifts; its eigenvectors are the “good” zeroth-order states.
For a two-fold degeneracy,
$$E^{(1)}_\pm = \frac{W_{aa} + W_{bb}}{2} \pm \sqrt{\left(\frac{W_{aa} - W_{bb}}{2}\right)^2 + \lvert W_{ab}\rvert^2}$$If you can find a Hermitian operator \( \hat A \) that commutes with both \( \hat H_0 \) and \( \hat H' \) and has distinct eigenvalues on the degenerate states, then its eigenstates are already the good states and \( W \) is diagonal. That is why we use the \( \lvert j, m_j\rangle \) basis for spin–orbit coupling and the \( \lvert n, l, m\rangle \) basis with \( m \) conserved for a field along \( z \).
The linear Stark effect in hydrogen
A field \( \mathcal E\hat z \) gives \( \hat H' = e\mathcal Ez \). The ground state is non-degenerate and even, so it has no first-order shift; its shift is quadratic in \( \mathcal E \).
The \( n = 2 \) level has four states: \( 200 \), \( 210 \), \( 211 \), \( 21{-1} \). Since \( z \) conserves \( m \) and is odd, the only non-zero element of \( W \) connects \( 200 \) and \( 210 \): \( \langle200\rvert z\lvert210\rangle = -3a_0 \). Diagonalising:
$$\boxed{E^{(1)} = \pm3e\mathcal Ea_0 \ \text{for}\ \tfrac{1}{\sqrt{2}}(\psi_{200} \pm \psi_{210}), \qquad 0 \ \text{for}\ \psi_{211},\ \psi_{21-1}}$$So \( n = 2 \) splits into three levels, with the middle one doubly degenerate. The shift is linear in the field because the \( l \)-degeneracy of hydrogen lets the mixed states carry a permanent dipole moment. Atoms without this \( l \)-degeneracy show only a quadratic Stark shift in fields that are weak compared with their level spacings.
The Variational Method
For any normalised trial state \( \lvert\psi\rangle \),
$$\boxed{E_{\text{gs}} \leq \langle\psi\rvert\hat H\lvert\psi\rangle}$$Proof. Expand \( \lvert\psi\rangle = \sum c_n\lvert n\rangle \) in exact eigenstates: \( \langle H\rangle = \sum\lvert c_n\rvert^2E_n \geq E_{\text{gs}}\sum\lvert c_n\rvert^2 = E_{\text{gs}} \).
Recipe: choose a trial function with one or more parameters, compute \( \langle H\rangle \), minimise. The result is an upper bound on the ground-state energy. A trial function orthogonal to the true ground state (for a symmetric potential, any odd function) gives an upper bound on the first excited state instead.
The energy estimate is accurate to second order in the error of the wavefunction, so a mediocre trial function still gives a good energy.
Other standard results:
| System | Trial function | Variational estimate | Exact |
|---|---|---|---|
| Harmonic oscillator | Gaussian \( e^{-bx^2} \) | \( \frac{1}{2}\hbar\omega \) | \( \frac{1}{2}\hbar\omega \) (trial family contains the answer) |
| Delta well | Gaussian | \( -\frac{m\alpha^2}{\pi\hbar^2} \) | \( -\frac{m\alpha^2}{2\hbar^2} \) |
| Hydrogen | Gaussian \( e^{-br^2} \) | \( -\frac{8}{3\pi}\times13.6 \approx -11.5 \) eV | \( -13.6 \) eV |
| Helium | product of 1s orbitals with \( Z_{\text{eff}} \) | \( -77.5 \) eV at \( Z_{\text{eff}} = \frac{27}{16} \) | \( \approx -79.0 \) eV (experiment) |
Helium in four numbers
In atomic units (1 hartree \( \approx 27.2 \) eV), with both electrons in hydrogen-like 1s orbitals of charge \( Z \):
- No repulsion: \( E = 2\times(-4\times13.6) = -108.8 \) eV.
- First-order perturbation in the repulsion, \( \langle e^2/4\pi\epsilon_0r_{12}\rangle = \frac{5}{8}Z \) hartree \( = \frac{5}{4}Z\times13.6\ \text{eV} = 34 \) eV for \( Z = 2 \): \( E \approx -74.8 \) eV.
- Variational with \( Z \) as the parameter: \( \langle H\rangle = \left(Z^2 - \frac{27}{8}Z\right) \) hartree, minimised at \( Z_{\text{eff}} = \frac{27}{16} \approx 1.69 \), giving \( -\left(\frac{27}{16}\right)^2 \) hartree \( \approx -77.5 \) eV.
- Experiment: about \( -79.0 \) eV.
Each electron screens the nucleus from the other, so the effective charge drops below 2.
Time-Dependent Perturbation Theory
Now let \( \hat H = \hat H_0 + \hat H'(t) \) and expand \( \lvert\psi(t)\rangle = \sum c_n(t)e^{-iE_nt/\hbar}\lvert n\rangle \). Starting in \( \lvert i\rangle \) at \( t = 0 \), to first order,
$$\boxed{c_f^{(1)}(t) = -\frac{i}{\hbar}\int_0^t\langle f\rvert\hat H'(t')\lvert i\rangle\,e^{i\omega_{fi}t'}\,dt', \qquad \omega_{fi} = \frac{E_f - E_i}{\hbar}}$$and \( P_{i\to f} = \lvert c_f^{(1)}\rvert^2 \) for \( f \neq i \). The transition amplitude is the Fourier component of the perturbation at the Bohr frequency. A perturbation that switches on and off slowly compared with \( 1/\omega_{fi} \) has almost no such component and causes almost no transitions.
Constant perturbation switched on at t = 0
$$P_{i\to f}(t) = \frac{4\lvert V_{fi}\rvert^2}{\hbar^2}\,\frac{\sin^2(\omega_{fi}t/2)}{\omega_{fi}^2}$$As a function of \( \omega_{fi} \), this has a central peak of height \( \propto t^2 \) and width \( \propto 1/t \), so its area grows linearly in \( t \). For long times it becomes \( \frac{2\pi}{\hbar}\lvert V_{fi}\rvert^2\,t\,\delta(E_f - E_i) \): energy is conserved, and the probability grows at a constant rate.
Fermi’s golden rule
When the final states form a continuum with density \( \rho(E_f) \) (states per unit energy), integrating over them gives a transition rate:
$$\boxed{\Gamma_{i\to f} = \frac{2\pi}{\hbar}\lvert V_{fi}\rvert^2\rho(E_f)}$$For a harmonic perturbation \( \hat H' = \hat Ve^{-i\omega t} + \hat V^\dagger e^{i\omega t} \), the same rule holds with \( E_f = E_i + \hbar\omega \) (absorption, from the \( \hat Ve^{-i\omega t} \) term) or \( E_f = E_i - \hbar\omega \) (stimulated emission, from \( \hat V^\dagger e^{i\omega t} \)). Absorption and stimulated emission between the same two levels have equal rates, which is the quantum basis of Einstein’s \( B_{12} = B_{21} \) (for equal degeneracies).
The golden rule needs \( t \) long enough for the energy to be sharply conserved, yet short enough that the initial state is not depleted (\( \Gamma t \ll 1 \)). The resulting exponential decay has lifetime \( \tau = 1/\Gamma \) and an energy width \( \hbar\Gamma \), consistent with the energy–time uncertainty relation.
Electric Dipole Transitions and Selection Rules
For light of wavelength much larger than the atom (\( \lambda \sim 100 \) nm versus \( a_0 \sim 0.05 \) nm), \( e^{i\mathbf k\cdot\mathbf r} \approx 1 \) and the coupling is \( \hat H' = -\hat{\mathbf d}\cdot\mathbf E(t) \) with \( \hat{\mathbf d} = -e\hat{\mathbf r} \). Rates are proportional to \( \lvert\langle f\rvert\hat{\mathbf r}\lvert i\rangle\rvert^2 \), so a transition is allowed only if this matrix element is non-zero.
Since \( z \propto Y_1^0 \) and \( x \pm iy \propto Y_1^{\pm1} \), the angular integrals give, for a one-electron atom:
$$\boxed{\Delta l = \pm1, \qquad \Delta m = 0, \pm1, \qquad \Delta s = 0, \qquad \text{parity changes}, \qquad \Delta n \text{ unrestricted}}$$\( \Delta m = 0 \) goes with light polarised along \( z \); \( \Delta m = \pm1 \) with circular polarisation in the \( xy \) plane. Including spin–orbit coupling, \( \Delta j = 0, \pm1 \) but not \( j = 0 \to 0 \).
| System | Dipole selection rule | Reason |
|---|---|---|
| Hydrogen-like atom | \( \Delta l = \pm1 \), \( \Delta m = 0, \pm1 \) | \( \mathbf r \) is a rank-1 (vector) operator |
| Harmonic oscillator | \( \Delta n = \pm1 \) | \( x \propto \hat a + \hat a^\dagger \) |
| Rigid rotor (polar molecule) | \( \Delta J = \pm1 \) | same as \( \Delta l = \pm1 \) |
| Infinite well \( (0, L) \) | \( n \leftrightarrow n' \) with \( n + n' \) odd | parity about the centre |
So \( 2p \to 1s \) is allowed but \( 2s \to 1s \) is not: the \( 2s \) state is metastable and decays mainly by two-photon emission. “Forbidden” means forbidden at dipole order; magnetic-dipole and electric-quadrupole transitions are much slower but not zero.
The spontaneous emission rate is
$$A = \frac{\omega^3\lvert\mathbf d_{fi}\rvert^2}{3\pi\epsilon_0\hbar c^3}$$The \( \omega^3 \) factor is why spontaneous emission dominates at optical and higher frequencies and is negligible for radio transitions. For hydrogen \( 2p \to 1s \) it gives a lifetime of about 1.6 ns.
Sudden and Adiabatic Limits
- Sudden approximation: if \( \hat H \) changes in a time much shorter than \( \hbar/\Delta E \), the state is unchanged at that instant. Expand it in the new eigenstates. Example: in the beta decay of tritium (\( Z = 1 \to 2 \)), the probability that the 1s electron ends in the 1s state of He\( ^+ \) is \( \left(\frac{2\sqrt{Z_1Z_2}}{Z_1 + Z_2}\right)^6 = \left(\frac{2\sqrt{2}}{3}\right)^6 = \frac{512}{729} \approx 0.70 \).
- Adiabatic theorem: if \( \hat H \) changes slowly compared with \( \hbar/\Delta E \) (and the level stays non-degenerate), a system starting in the \( n \)th eigenstate ends in the \( n \)th eigenstate of the final Hamiltonian, acquiring only a phase (dynamic plus geometric, or Berry, phase).
Memory Tricks
Don’t Confuse
| This | vs | That |
|---|---|---|
| Non-degenerate: \( E^{(1)} = H'_{nn} \) | Degenerate: eigenvalues of \( W \) in the subspace | |
| Ground state \( E^{(2)} \leq 0 \) always | Excited state \( E^{(2)} \) can have either sign | |
| Variational energy: upper bound | Perturbative energy: in general not a bound (but \( E_0^{(0)} + E_0^{(1)} \) for the ground state is \( \langle H\rangle \) in the unperturbed state, hence an upper bound) | |
| Linear Stark: hydrogen \( n = 2 \) (degenerate) | Quadratic Stark: hydrogen ground state, all non-degenerate levels | |
| Transition probability \( \propto t^2 \) at short times (single final state) | Rate \( \Gamma \) constant (continuum, golden rule) | |
| Sudden: state frozen, re-expanded | Adiabatic: state follows the instantaneous eigenstate |
Common Mistakes
Practice
All questions below are original practice questions written for this page. They are not from any CSIR NET paper.
Level 1: Foundation
- A constant perturbation \( \hat H' = V_0 \) acts everywhere. What is the first-order shift of every level?
- What is the sign of the second-order correction to a non-degenerate ground state?
- Is \( 2s \to 1s \) in hydrogen an electric-dipole-allowed transition?
Answers
- \( V_0 \). And all higher orders vanish, since \( \langle m\rvert V_0\lvert n\rangle = 0 \) for \( m \neq n \).
- Negative (or zero). Every denominator \( E_0 - E_m \) is negative.
- No. \( \Delta l = 0 \) and no parity change.
Level 2: Part B Style
Practice Q1. A particle in an infinite well \( 0 < x < L \) is perturbed by \( \hat H' = V_0 \) for \( 0 < x < L/2 \) and zero elsewhere. The first-order energy correction to level \( n \) is:
(a) \( V_0 \) (b) \( V_0/2 \) (c) \( V_0/2 \) for odd \( n \), 0 for even \( n \) (d) \( V_0/n^2 \)
Answer: (b) V₀/2
\( E_n^{(1)} = V_0\int_0^{L/2}\frac{2}{L}\sin^2\frac{n\pi x}{L}dx \). Every \( \lvert\psi_n\rvert^2 \) is symmetric about \( L/2 \), so exactly half the probability lies in the left half: \( E_n^{(1)} = V_0/2 \) for all \( n \).
Practice Q2. A harmonic oscillator is perturbed by \( \hat H' = \lambda\hat x^4 \). The first-order correction to the ground-state energy is:
(a) \( \frac{\lambda\hbar^2}{4m^2\omega^2} \) (b) \( \frac{3\lambda\hbar^2}{4m^2\omega^2} \) (c) \( \frac{3\lambda\hbar^2}{2m^2\omega^2} \) (d) 0
Answer: (b)
\( \langle0\rvert\hat x^4\lvert0\rangle = 3\left(\frac{\hbar}{2m\omega}\right)^2 = \frac{3\hbar^2}{4m^2\omega^2} \). Equivalently, for a Gaussian \( \langle x^4\rangle = 3\langle x^2\rangle^2 \) with \( \langle x^2\rangle = \hbar/2m\omega \). Option (a) uses \( \langle x^2\rangle^2 \) without the factor 3; (d) misapplies the parity argument (\( x^4 \) is even).
Practice Q3. A harmonic oscillator of mass \( m \) and frequency \( \omega \) is placed in a uniform force field, \( \hat H' = b\hat x \). The second-order shift of the \( n \)th level is:
(a) \( -\frac{b^2}{2m\omega^2} \) for all \( n \) (b) \( -\frac{(2n+1)b^2}{2m\omega^2} \) (c) \( +\frac{b^2}{2m\omega^2} \) (d) 0
Answer: (a)
\( E_n^{(2)} = \frac{\lvert\langle n+1\rvert b\hat x\lvert n\rangle\rvert^2}{-\hbar\omega} + \frac{\lvert\langle n-1\rvert b\hat x\lvert n\rangle\rvert^2}{+\hbar\omega} = \frac{b^2\hbar}{2m\omega}\cdot\frac{-(n+1) + n}{\hbar\omega} = -\frac{b^2}{2m\omega^2} \).
Independent of \( n \), as the exact solution (completing the square) requires. Option (b) adds the two contributions instead of letting them partially cancel.
Practice Q4. A normalised Gaussian trial function is used for the attractive delta potential \( -\alpha\delta(x) \). The ratio of the best variational energy to the exact ground-state energy is:
(a) \( 1 \) (b) \( \pi/2 \) (c) \( 2/\pi \) (d) \( 1/2 \)
Answer: (c) 2/π
\( E_{\text{var}} = -\frac{m\alpha^2}{\pi\hbar^2} \), \( E_{\text{exact}} = -\frac{m\alpha^2}{2\hbar^2} \). Ratio \( = \frac{2}{\pi} \approx 0.64 \). It is less than 1 because the variational energy is less negative (higher). Option (b) would put the variational energy below the exact one, which is impossible.
Level 3: Part C Style
Practice Q5. For the helium ground state, a trial function of two hydrogen-like 1s orbitals with effective charge \( Z \) gives \( \langle H\rangle = \left(Z^2 - \frac{27}{8}Z\right) \) hartree (1 hartree \( = 27.2 \) eV). Consider:
- The optimum is \( Z = 27/16 \).
- The variational energy is about \( -77.5 \) eV.
- The first-order perturbation result (repulsion treated as a perturbation on \( Z = 2 \) orbitals) is about \( -74.8 \) eV.
- The true ground-state energy lies above \( -77.5 \) eV.
Which are correct?
(a) 1 and 2 only (b) 1, 2 and 3 only (c) 2, 3 and 4 only (d) 1, 2, 3 and 4
Answer: (b) 1, 2 and 3 only
- \( \frac{d}{dZ}\left(Z^2 - \frac{27}{8}Z\right) = 0 \Rightarrow Z = \frac{27}{16} \approx 1.69 \). True.
- \( \langle H\rangle_{\min} = -\left(\frac{27}{16}\right)^2 = -2.848 \) hartree \( \approx -77.5 \) eV. True.
- At \( Z = 2 \) the same expression gives \( 4 - \frac{27}{4} = -2.75 \) hartree \( \approx -74.8 \) eV, which is exactly the first-order perturbation result (the trial function at \( Z = 2 \) is the unperturbed state). True.
- The variational estimate is an upper bound, so the true energy lies below \( -77.5 \) eV (experiment: about \( -79.0 \) eV). False.
Practice Q6. A hydrogen atom in its \( n = 2 \) level is placed in a uniform electric field \( \mathcal E = 1.0\times10^7 \) V/m along \( z \). Ignoring spin and fine structure, and using \( a_0 = 0.0529 \) nm, consider:
- The level splits into three distinct energies.
- The largest shift is about \( 1.6\times10^{-3} \) eV.
- The states \( \psi_{211} \) and \( \psi_{21-1} \) are unshifted at first order.
- The perturbed states with non-zero shift are \( \psi_{200} \) and \( \psi_{210} \) separately.
Which are correct?
(a) 1, 2 and 3 only (b) 1 and 4 only (c) 2 and 3 only (d) 1, 2, 3 and 4
Answer: (a) 1, 2 and 3 only
The first-order shifts are \( \pm3e\mathcal Ea_0 \) and 0 (twice): three distinct energies. Statement 1 true.
\( 3e\mathcal Ea_0 = 3\times(1.0\times10^7\ \text{V/m})\times(5.29\times10^{-11}\ \text{m})\times e = 1.59\times10^{-3} \) eV. Statement 2 true.
\( z \) conserves \( m \), and the only \( m = \pm1 \) states at \( n = 2 \) are \( \psi_{21\pm1} \), each alone in its \( m \) sector with \( \langle z\rangle = 0 \). Statement 3 true.
The good states are \( \frac{1}{\sqrt{2}}(\psi_{200} \pm \psi_{210}) \), not the separate states. Statement 4 false.
Practice Q7. A charged oscillator (charge \( q \), mass \( m \), frequency \( \omega \)), initially in \( \lvert0\rangle \), is exposed to a pulse \( \hat H' = -q\mathcal E_0e^{-t^2/\tau^2}\hat x \) for \( -\infty < t < \infty \). Consider, to first order:
- \( P_{0\to1} = \frac{\pi q^2\mathcal E_0^2\tau^2}{2m\hbar\omega}e^{-\omega^2\tau^2/2} \).
- \( P_{0\to2} \neq 0 \).
- For fixed \( \mathcal E_0 \), \( P_{0\to1} \) is largest at \( \tau = \sqrt{2}/\omega \).
- \( P_{0\to1} \to 0 \) as \( \tau \to \infty \).
Which are correct?
(a) 1, 3 and 4 only (b) 1 and 2 only (c) 2, 3 and 4 only (d) 1, 2, 3 and 4
Answer: (a) 1, 3 and 4 only
\( c_1 = -\frac{i}{\hbar}(-q\mathcal E_0)\langle1\rvert\hat x\lvert0\rangle\int_{-\infty}^{\infty}e^{-t^2/\tau^2}e^{i\omega t}dt = \frac{iq\mathcal E_0}{\hbar}\sqrt{\frac{\hbar}{2m\omega}}\,\sqrt{\pi}\,\tau\,e^{-\omega^2\tau^2/4} \).
Squaring: \( P_{0\to1} = \frac{q^2\mathcal E_0^2}{\hbar^2}\cdot\frac{\hbar}{2m\omega}\cdot\pi\tau^2e^{-\omega^2\tau^2/2} \). Statement 1 true.
\( \langle2\rvert\hat x\lvert0\rangle = 0 \), so there is no first-order \( 0 \to 2 \) amplitude. Statement 2 false.
\( \frac{d}{d\tau}\left(\tau^2e^{-\omega^2\tau^2/2}\right) = 0 \Rightarrow \tau^2 = 2/\omega^2 \). Statement 3 true.
A slow pulse (\( \omega\tau \gg 1 \)) has no Fourier component at \( \omega \): the adiabatic limit. Statement 4 true. (For fixed \( \mathcal E_0 \), a very short pulse also gives \( P \to 0 \), simply because it delivers almost no impulse.)
Quick Revision Table
| Idea | Formula or fact |
|---|---|
| First order | \( E_n^{(1)} = \langle n\rvert H'\lvert n\rangle \) |
| Second order | \( E_n^{(2)} = \sum_{m\neq n}\frac{\lvert H'_{mn}\rvert^2}{E_n - E_m} \); ground state \( \leq 0 \) |
| Degenerate | diagonalise \( W \) in the subspace |
| Oscillator \( + bx \) | \( -\frac{b^2}{2m\omega^2} \), exact |
| Oscillator \( + \lambda x^4 \) | \( E_0^{(1)} = \frac{3\lambda\hbar^2}{4m^2\omega^2} \) |
| Linear Stark (\( n = 2 \)) | \( \pm3e\mathcal Ea_0 \), 0, 0 |
| Variational principle | \( E_{\text{gs}} \leq \langle\psi\rvert H\lvert\psi\rangle \) |
| Helium | \( Z_{\text{eff}} = 27/16 \), \( -77.5 \) eV |
| First-order amplitude | \( c_f = -\frac{i}{\hbar}\int H'_{fi}e^{i\omega_{fi}t}dt \) |
| Golden rule | \( \Gamma = \frac{2\pi}{\hbar}\lvert V_{fi}\rvert^2\rho(E_f) \) |
| Dipole rules | \( \Delta l = \pm1 \), \( \Delta m = 0, \pm1 \), parity changes |
| Spontaneous emission | \( A \propto \omega^3\lvert d_{fi}\rvert^2 \); H \( 2p \): \( \tau \approx 1.6 \) ns |
| Sudden, tritium to He\( ^+ \) | \( P(1s \to 1s) = 512/729 \approx 0.70 \) |
Key Takeaways
What’s Next
- Identical Particles, Pauli Principle and Spin-Statistics: helium returns, now with exchange symmetry.
- Quantum Mechanics: Quick Revision: every result above on one page.
- The Physical Sciences overview lists the atomic and molecular physics unit, where Zeeman, Stark and spectroscopic selection rules are developed further.