Skip to content

Fine Structure, Hyperfine Structure and LS/jj Coupling

CSIR NET Physical Sciences Part C: relativistic corrections, hyperfine structure, isotope shift, natural, Doppler and pressure line widths, LS and jj coupling.

  • 22 min read
  • Updated
On this page · 15 sections
  1. Prerequisites
  2. The Hook: One Red Line, Four Stories
  3. The Hierarchy of Atomic Energies
  4. Relativistic Corrections: Fine Structure of Hydrogen
  5. Hyperfine Structure
  6. Isotope Shift
  7. Line Widths
  8. LS and jj Coupling
  9. Memory Tricks
  10. Don’t Confuse
  11. Common Mistakes
  12. Practice
  13. Quick Revision Table
  14. Key Takeaways
  15. What’s Next

Prerequisites

The Hook: One Red Line, Four Stories

The Hierarchy of Atomic Energies

EffectTypical size (light atoms)OriginQuantum numbers it introduces
Gross structureeVCoulomb attraction, central field\( n, l \) (configuration)
Residual electrostatic\( \sim 0.1 \)–1 eVelectron–electron repulsion beyond the central field, exchange\( L, S \) (terms)
Fine structure\( \sim 10^{-5} \)–\( 10^{-2} \) eVspin–orbit coupling and relativity, order \( \alpha^2 \)\( J \) (levels)
Hyperfine structure\( \sim 10^{-6} \)–\( 10^{-5} \) eVnuclear moments; smaller by about \( m_e/m_p \)\( F \)
Lamb shift\( \sim 10^{-6} \) eV in HQED (vacuum fluctuations)separates \( 2s_{1/2} \), \( 2p_{1/2} \)

Relativistic Corrections: Fine Structure of Hydrogen

For a hydrogen-like atom, three corrections of order \( \alpha^2 \) relative to the Bohr energies act together:

  1. Relativistic kinetic energy, \( -p^4/8m^3c^2 \): lowers every level, most for low \( l \).
  2. Spin–orbit coupling, \( \xi(r)\,\mathbf L\cdot\mathbf S \) with the Thomas factor of \( \frac{1}{2} \): zero for \( l = 0 \).
  3. Darwin term, proportional to \( \nabla^2V \): acts only at the nucleus, so only on \( l = 0 \).

Their sum depends only on \( n \) and \( j \):

$$\boxed{E_{nj} = E_n\left[1 + \frac{\alpha^2Z^2}{n^2}\left(\frac{n}{j + \frac{1}{2}} - \frac{3}{4}\right)\right], \qquad E_n = -\frac{13.6\ \text{eV}\,Z^2}{n^2}}$$

with \( \alpha \approx 1/137 \). Consequences:

  • Every level is pushed down (the bracket term is positive and \( E_n < 0 \)).
  • Levels with the same \( n \) and \( j \) but different \( l \) stay degenerate: \( 2s_{1/2} \) and \( 2p_{1/2} \); \( 3p_{3/2} \) and \( 3d_{3/2} \). This is an exact feature of the Dirac equation for a Coulomb potential.
  • The effect scales as \( Z^4/n^3 \): large for heavy ions, small for highly excited states.

The Lamb shift. Measurements in 1947 by Lamb and Retherford showed that \( 2s_{1/2} \) lies above \( 2p_{1/2} \) by about 1.06 GHz (roughly a tenth of the fine-structure splitting). Dirac theory predicts exact degeneracy; quantum electrodynamics explains the shift as the interaction of the electron with vacuum fluctuations of the radiation field. It is the reason “fine structure depends only on \( n, j \)” carries the qualifier “in Dirac theory”.

Fine structure in alkali atoms

For one valence electron outside a closed core, the spin–orbit energy is

$$E_{\text{so}} = \frac{\zeta_{nl}}{2}\left[j(j+1) - l(l+1) - \frac{3}{4}\right], \qquad \Delta E(j = l + \tfrac{1}{2}) - \Delta E(j = l - \tfrac{1}{2}) = \zeta_{nl}\left(l + \tfrac{1}{2}\right)$$

with \( \zeta_{nl} > 0 \), so the higher \( j \) lies above. The sodium 3p splitting of 17.2 cm\( ^{-1} \) is far larger than anything in hydrogen’s \( n = 3 \) shell because the 3p electron penetrates to where it feels a much larger nuclear charge.

Hyperfine Structure

A nucleus with spin \( I \neq 0 \) has a magnetic dipole moment \( \boldsymbol\mu_I = g_I\mu_N\mathbf I/\hbar \), and it couples to the magnetic field the electrons produce at the nucleus. The total angular momentum becomes \( \mathbf F = \mathbf I + \mathbf J \):

$$\boxed{H_{\text{hf}} = A\,\mathbf I\cdot\mathbf J, \qquad E_F = \frac{A}{2}\left[F(F+1) - I(I+1) - J(J+1)\right], \qquad F = \lvert I - J\rvert, \ldots, I + J}$$

(with \( A \) in energy units, and \( \mathbf I \), \( \mathbf J \) measured in units of \( \hbar \)). The number of hyperfine levels is \( 2\min(I, J) + 1 \). Each \( F \) level has \( 2F + 1 \) states.

Interval rule for hyperfine levels:

$$E_F - E_{F-1} = AF$$

the same form as the Landé interval rule below. Nuclei with \( I \geq 1 \) can also have an electric quadrupole moment, which adds a term that breaks the simple interval rule.

Size: \( \mu_N/\mu_B = m_e/m_p \approx 1/1836 \), so hyperfine splittings are roughly a thousand times smaller than fine-structure splittings for comparable electron states.

System\( I \), \( J \)\( F \) valuesSplitting
Hydrogen 1s\( \frac{1}{2} \), \( \frac{1}{2} \)0, 11420.4 MHz (the 21 cm line), \( F = 1 \) above
Caesium-133 ground state\( \frac{7}{2} \), \( \frac{1}{2} \)3, 49 192 631 770 Hz exactly, by the SI definition of the second
Deuterium 1s1, \( \frac{1}{2} \)\( \frac{1}{2} \), \( \frac{3}{2} \)smaller than hydrogen’s (smaller \( g_I \))

The hydrogen \( F = 1 \to F = 0 \) transition is a magnetic dipole transition (no parity change, \( \Delta F = 1 \)). It is extremely slow, but there is so much hydrogen in the galaxy that the 21 cm line maps it.

Isotope Shift

Two isotopes of an element have slightly different line positions. Two separate effects contribute:

1. Mass shift

The nucleus is not infinitely heavy. For a one-electron atom, the Rydberg constant becomes

$$R_M = \frac{R_\infty}{1 + m_e/M}$$

A heavier isotope has a larger \( R_M \) and lines at slightly higher wavenumber (shorter wavelength). For hydrogen and deuterium,

$$\frac{\Delta\lambda}{\lambda} \approx m_e\left(\frac{1}{m_H} - \frac{1}{m_D}\right) \approx 2.72\times10^{-4}$$

so Balmer-\( \alpha \) at 656.28 nm shifts by about 0.18 nm for deuterium. Urey and co-workers found deuterium in 1932 from exactly this faint blue-shifted companion.

In many-electron atoms the mass shift has a normal part (the reduced-mass correction above) and a specific part from correlated motion of the electrons (the \( \mathbf p_i\cdot\mathbf p_j \) terms in the nuclear recoil). The specific mass shift can have either sign and has to be calculated.

2. Field (volume) shift

The nucleus is not a point. Its charge is spread over a radius of a few femtometres, and an s electron, which has density inside the nucleus, feels a weaker attraction there than from a point charge. A larger nucleus (usually the heavier isotope) raises s-level energies slightly. The shift is proportional to \( \lvert\psi(0)\rvert^2\,\delta\langle r^2\rangle_{\text{nuc}} \).

ShiftDominant inScalingDepends on
Mass (normal + specific)light elements (\( Z \lesssim 30 \))\( \propto \Delta M/M^2 \), falls with \( A \)nuclear mass
Field (volume)heavy elementsgrows with \( Z \)nuclear charge radius, s-electron density at nucleus

Line Widths

A spectral line is never infinitely sharp. The shape is described by a profile \( g(\nu) \) normalised to unit area, and the width quoted is the full width at half maximum (FWHM).

Natural (radiative) broadening

An excited state with lifetime \( \tau \) decays exponentially, so its emitted field is a damped oscillation. Its Fourier transform is a Lorentzian:

$$\boxed{g_L(\nu) = \frac{\Delta\nu/2\pi}{(\nu - \nu_0)^2 + (\Delta\nu/2)^2}, \qquad \Delta\nu_{\text{nat}} = \frac{1}{2\pi\tau}}$$

The same result follows from the energy–time uncertainty relation \( \Delta E\,\tau \sim \hbar \). If both the upper and lower levels decay, \( \Delta\nu = \frac{1}{2\pi}\left(\frac{1}{\tau_u} + \frac{1}{\tau_l}\right) \). An allowed optical transition with \( \tau \approx 10 \) ns has \( \Delta\nu_{\text{nat}} \approx 16 \) MHz.

Doppler broadening

An atom moving with velocity component \( v_x \) towards the observer emits at \( \nu_0(1 + v_x/c) \). The Maxwell distribution of \( v_x \) is a Gaussian, so the line is a Gaussian:

$$\boxed{\Delta\nu_D = \frac{2\nu_0}{c}\sqrt{\frac{2k_BT\ln2}{m}} \approx 7.16\times10^{-7}\,\nu_0\sqrt{\frac{T}{M}}}$$

with \( T \) in kelvin and \( M \) the mass in atomic mass units. Doppler width grows as \( \sqrt T \) and as \( 1/\sqrt M \). For optical lines at room temperature it is typically around a gigahertz, much larger than the natural width.

Pressure (collision) broadening

Collisions interrupt the phase of the emitting oscillator. If the mean time between phase-interrupting collisions is \( \tau_c \), the line is again Lorentzian with

$$\Delta\nu_{\text{coll}} = \frac{1}{\pi\tau_c}$$

and since \( 1/\tau_c \propto \) pressure \( \times \) mean speed, the width rises linearly with pressure. Collisions also shift the line centre slightly.

Homogeneous versus inhomogeneous

TypeEvery atom has the same profile?ShapeExamples
HomogeneousyesLorentziannatural, collision, power broadening
Inhomogeneousno: different atoms at different centresGaussian (usually)Doppler, crystal-field variation in solids

When both act, the profile is a convolution called the Voigt profile: Gaussian near the centre and Lorentzian in the far wings, because the Lorentzian falls off as \( 1/\Delta\nu^2 \) and the Gaussian much faster. The distinction matters directly for lasers: a homogeneously broadened gain medium tends to oscillate on one mode, an inhomogeneously broadened one can “burn holes” and run on several (see Lasers).

Ways to beat the Doppler width: cooled atomic beams observed at right angles, saturated absorption spectroscopy (the Lamb dip), and two-photon absorption with counter-propagating beams.

LS and jj Coupling

In a many-electron atom, beyond the central field there are two perturbations:

  • \( H_{\text{es}} \): the non-central part of the electron–electron repulsion, which depends on how the orbital and spin angular momenta are arranged.
  • \( H_{\text{so}} = \sum_i\xi(r_i)\,\mathbf l_i\cdot\mathbf s_i \): spin–orbit coupling of each electron.

Which one is larger decides the coupling scheme.

graph TD
    C[Configuration, e.g. np2] --> Q{Which perturbation dominates?}
    Q -->|electrostatic much larger: light atoms| LS[LS coupling: L = sum of l, S = sum of s]
    Q -->|spin-orbit much larger: heavy atoms, inner shells| JJ[jj coupling: each j = l + s, then J = sum of j]
    LS --> T[Terms 2S+1 L, split by spin-orbit into levels J]
    JJ --> G[Groups j1 j2, split by electrostatics into levels J]
    T --> J[Same set of J values and same number of states]
    G --> J

LS (Russell–Saunders) coupling

\( H_{\text{es}} \gg H_{\text{so}} \). The orbital angular momenta add to \( \mathbf L = \sum\mathbf l_i \), the spins to \( \mathbf S = \sum\mathbf s_i \), and then \( \mathbf J = \mathbf L + \mathbf S \). States are labelled by the term symbol

$$^{2S+1}L_J, \qquad J = \lvert L - S\rvert, \ldots, L + S$$

The electrostatic interaction splits a configuration into terms (\( L, S \)); spin–orbit coupling then splits each term into levels (\( J \)) with energies

$$E_J = \frac{A}{2}\left[J(J+1) - L(L+1) - S(S+1)\right]$$

Landé interval rule:

$$\boxed{E_J - E_{J-1} = AJ}$$

The interval between adjacent levels of a multiplet is proportional to the larger \( J \). For a \( ^3P \) term the intervals \( (J = 0\to1) : (1\to2) \) are \( 1 : 2 \); for \( ^3D \), \( 2 : 3 \); for \( ^3F \), \( 3 : 4 \). The interval rule is a test of LS coupling: helium and heavy atoms violate it.

Hund’s rules for the ground term of an equivalent-electron configuration:

  1. Maximum \( S \) (exchange keeps parallel-spin electrons apart).
  2. For that \( S \), maximum \( L \).
  3. Less than half-filled shell: lowest \( J = \lvert L - S\rvert \) lies lowest (normal multiplet, \( A > 0 \)). More than half-filled: \( J = L + S \) lowest (inverted, \( A < 0 \)). Exactly half-filled: \( L = 0 \), so one level.
ConfigurationTermsStatesGround level
\( p^1 \), \( p^5 \)\( ^2P \)6\( ^2P_{1/2} \) (\( p^1 \)), \( ^2P_{3/2} \) (\( p^5 \))
\( p^2 \), \( p^4 \)\( ^1S \), \( ^1D \), \( ^3P \)15\( ^3P_0 \) (C), \( ^3P_2 \) (O)
\( p^3 \)\( ^4S \), \( ^2D \), \( ^2P \)20\( ^4S_{3/2} \) (N)
\( d^2 \)\( ^1S \), \( ^1D \), \( ^1G \), \( ^3P \), \( ^3F \)45\( ^3F_2 \)
\( np\,n'p \) (non-equivalent)\( ^{1,3}S \), \( ^{1,3}P \), \( ^{1,3}D \)36not fixed by Hund’s rules in general

Equivalent electrons (same \( n, l \)) have fewer terms than non-equivalent ones, because the Pauli principle removes some combinations: \( p^2 \) gives 15 states, \( np\,n'p \) gives \( 6\times6 = 36 \). The full microstate derivation is in Term Symbols and Atomic Spectra.

LS selection rules (electric dipole):

RuleComment
\( \Delta S = 0 \)breaks down as spin–orbit coupling grows
\( \Delta L = 0, \pm1 \), not \( 0 \leftrightarrow 0 \)
\( \Delta J = 0, \pm1 \), not \( 0 \leftrightarrow 0 \)rigorous for dipole radiation
parity changes; one electron jumps with \( \Delta l = \pm1 \)rigorous (parity)

jj coupling

\( H_{\text{so}} \gg H_{\text{es}} \): heavy atoms, and inner-shell or highly excited configurations. Each electron first forms its own \( \mathbf j_i = \mathbf l_i + \mathbf s_i \), with \( j_i = l_i \pm \frac{1}{2} \), and then \( \mathbf J = \sum\mathbf j_i \). States are labelled \( (j_1, j_2)_J \). \( L \) and \( S \) are no longer good quantum numbers, so \( \Delta S = 0 \) is no longer a selection rule; \( \Delta J = 0, \pm1 \) and parity remain.

The np² check. For two equivalent p electrons, each \( j \) is \( \frac{1}{2} \) or \( \frac{3}{2} \):

jj groupAllowed \( J \) (Pauli applied)States
\( (\frac{1}{2}, \frac{1}{2}) \)0 only1
\( (\frac{1}{2}, \frac{3}{2}) \)1, 28
\( (\frac{3}{2}, \frac{3}{2}) \)0, 2 only6

Total 15 states, with \( J = 0 \) twice, \( J = 1 \) once, \( J = 2 \) twice. LS gives \( ^1S_0 \), \( ^1D_2 \), \( ^3P_{0,1,2} \): again \( J = 0 \) twice, \( J = 1 \) once, \( J = 2 \) twice. The number of levels with each \( J \) is the same in both schemes; only their grouping and spacing differ. Going down group 14 (C, Si, Ge, Sn, Pb), the level pattern of the \( np^2 \) ground configuration moves smoothly from the LS pattern to the jj pattern. Real atoms in between show intermediate coupling.

Memory Tricks

Short anchors for the results on this page. Say each one aloud before a mock test; if one does not trigger the formula, reread its section.

Don’t Confuse

ThisvsThat
Fine structure: \( \sim\alpha^2 \times \) Bohr energy, quantum number \( J \)Hyperfine: smaller by \( \sim m_e/m_p \), quantum number \( F \)
Mass isotope shift: light atoms, from reduced massField isotope shift: heavy atoms, from nuclear size
Natural width \( 1/(2\pi\tau) \): homogeneous, LorentzianDoppler width \( \propto\nu_0\sqrt{T/M} \): inhomogeneous, Gaussian
Collision width \( 1/(\pi\tau_c) \), rises with pressureNatural width, independent of pressure
Normal multiplet: lowest \( J \) lowest (less than half filled)Inverted: highest \( J \) lowest (more than half filled)
LS: \( (L, S) \) then \( J \)jj: \( (j_1, j_2) \) then \( J \)

Common Mistakes

Practice

All questions below are original practice questions written for this page. They are not from any CSIR NET paper. Take \( \alpha = 1/137.036 \), \( k_B = 1.381\times10^{-23} \) J K\( ^{-1} \), \( h = 6.626\times10^{-34} \) J s.

Level 1: Foundation

  1. What are the possible \( F \) values for the hydrogen ground state?
  2. List the levels of a \( ^3D \) term.
  3. An excited state has lifetime 10 ns and decays to the ground state. What is the natural line width in MHz?
Answers
  1. \( F = 0, 1 \). \( I = \frac{1}{2} \), \( J = \frac{1}{2} \).
  2. \( ^3D_1 \), \( ^3D_2 \), \( ^3D_3 \). \( J = \lvert2 - 1\rvert \) to \( 2 + 1 \).
  3. About 15.9 MHz. \( 1/(2\pi\times10^{-8}) \); the ground state has infinite lifetime.

Level 2: Part B Style

Practice Q1. The fine-structure splitting between \( 2p_{3/2} \) and \( 2p_{1/2} \) in hydrogen is closest to:

(a) 0.046 cm\( ^{-1} \) (b) 0.365 cm\( ^{-1} \) (c) 0.73 cm\( ^{-1} \) (d) 3.65 cm\( ^{-1} \)

Answer: (b)

\( \Delta E = \lvert E_2\rvert\frac{\alpha^2}{4}\left[\left(2 - \frac{3}{4}\right) - \left(1 - \frac{3}{4}\right)\right] = \frac{3.40\alpha^2}{4} = 4.53\times10^{-5} \) eV \( = 0.365 \) cm\( ^{-1} \). Option (c) is twice the correct value and option (a) one-eighth of it. For scale, the Lamb shift (about 1.06 GHz) is about 0.035 cm\( ^{-1} \).

Practice Q2. A multiplet of an LS-coupled atom shows three levels with intervals 30 cm\( ^{-1} \) and 45 cm\( ^{-1} \), in order of increasing energy. The term is:

(a) \( ^3P \) (b) \( ^3D \) (c) \( ^3F \) (d) \( ^4D \)

Answer: (b) ³D

Interval rule: \( E_J - E_{J-1} = AJ \). The ratio \( 30 : 45 = 2 : 3 \) means the upper \( J \) values of the two intervals are 2 and 3, so \( J = 1, 2, 3 \) with \( A = 15 \) cm\( ^{-1} \). Three levels with \( J = 1, 2, 3 \) means \( ^3D \). \( ^3P \) would be \( 1 : 2 \); \( ^3F \) would be \( 3 : 4 \); \( ^4D \) has four levels.

Practice Q3. The configuration \( 2p\,3p \) (non-equivalent electrons) in LS coupling gives:

(a) 3 terms, 15 states (b) 6 terms, 36 states (c) 6 terms, 15 states (d) 4 terms, 36 states

Answer: (b)

\( L = 0, 1, 2 \) and \( S = 0, 1 \), all allowed because the electrons differ in \( n \): \( ^1S \), \( ^3S \), \( ^1P \), \( ^3P \), \( ^1D \), \( ^3D \). States \( = 6\times6 = 36 \) (check: \( 1 + 3 + 3 + 9 + 5 + 15 = 36 \)). Option (a) is the equivalent \( 2p^2 \) result.

Practice Q4. The Balmer-\( \alpha \) line of hydrogen is at 656.28 nm. The corresponding deuterium line is shifted by about:

(a) 0.18 nm to shorter wavelength (b) 0.18 nm to longer wavelength (c) 0.36 nm to shorter wavelength (d) 0.09 nm to shorter wavelength

Answer: (a)

\( \lambda \propto 1 + m_e/M \). \( \Delta\lambda = 656.28\times m_e\left(\frac{1}{m_p} - \frac{1}{m_d}\right) = 656.28\times2.72\times10^{-4} = 0.179 \) nm, to the blue because deuterium’s larger reduced mass binds the electron more strongly.

Level 3: Part C Style

Practice Q5. A sodium D line (\( \lambda = 589 \) nm) is observed from a vapour at 500 K. Take the upper-state lifetime as 16 ns and the atomic mass as 23 u. Consider:

  1. The Doppler width (FWHM) is about 1.7 GHz.
  2. The natural width is about 10 MHz.
  3. Near the line centre the profile is essentially Gaussian.
  4. Raising the temperature to 1000 K doubles the Doppler width.

Which are correct?

(a) 1 and 2 only (b) 1, 2 and 3 only (c) 2 and 4 only (d) 1, 2, 3 and 4

Answer: (b) 1, 2 and 3 only
  1. \( \nu_0 = c/\lambda = 5.09\times10^{14} \) Hz; \( \Delta\nu_D = 7.16\times10^{-7}\times5.09\times10^{14}\times\sqrt{500/23} = 1.70\times10^9 \) Hz. True.
  2. \( 1/(2\pi\times16\ \text{ns}) = 9.9 \) MHz. True.
  3. The Doppler width is about 170 times the natural width, so the Voigt profile is Gaussian near the centre (the Lorentzian shows only in the far wings). True.
  4. False: \( \Delta\nu_D \propto\sqrt T \), so it rises by \( \sqrt2 \), to about 2.4 GHz.

Practice Q6. For the \( np^2 \) configuration, consider:

  1. LS coupling gives \( ^1S \), \( ^1D \), \( ^3P \), 15 states in all.
  2. For carbon the ground level is \( ^3P_0 \).
  3. In jj coupling the \( (\frac{3}{2}, \frac{3}{2}) \) group gives \( J = 0, 1, 2, 3 \).
  4. Both schemes give two levels with \( J = 0 \), one with \( J = 1 \) and two with \( J = 2 \).

Which are correct?

(a) 1, 2 and 4 only (b) 1 and 3 only (c) 2, 3 and 4 only (d) 1, 2, 3 and 4

Answer: (a) 1, 2 and 4 only
  1. True; \( 1 + 5 + 9 = 15 = \binom{6}{2} \).
  2. True: Hund gives \( ^3P \); \( p^2 \) is less than half filled, so lowest \( J = 0 \).
  3. False: Pauli allows only the antisymmetric combinations of two identical \( j = \frac{3}{2} \) electrons, \( J = 0 \) and 2 (6 states). \( J = 1, 3 \) are excluded.
  4. True. LS: \( ^1S_0 \), \( ^3P_0 \) (two \( J = 0 \)); \( ^3P_1 \); \( ^1D_2 \), \( ^3P_2 \). jj: \( (\frac12,\frac12)_0 \), \( (\frac32,\frac32)_0 \); \( (\frac12,\frac32)_1 \); \( (\frac12,\frac32)_2 \), \( (\frac32,\frac32)_2 \).

Practice Q7. The ground-state hyperfine transition of atomic hydrogen has frequency 1420.4 MHz. Consider:

  1. Its wavelength is about 21.1 cm.
  2. The splitting is about 5.9 μeV, equivalent to a temperature \( \Delta E/k_B \) of about 0.07 K.
  3. In interstellar gas at 100 K, the population ratio \( N(F = 1)/N(F = 0) \) is close to 3.
  4. The transition is an electric dipole transition with \( \Delta F = 1 \).

Which are correct?

(a) 1 and 2 only (b) 1, 2 and 3 only (c) 3 and 4 only (d) 1, 2, 3 and 4

Answer: (b) 1, 2 and 3 only
  1. \( \lambda = 2.998\times10^8/1.4204\times10^9 = 0.211 \) m. True.
  2. \( h\nu = 6.626\times10^{-34}\times1.4204\times10^9 = 9.41\times10^{-25} \) J \( = 5.87 \) μeV; \( /k_B = 0.068 \) K. True.
  3. \( N_1/N_0 = \frac{3}{1}e^{-0.068/100} \approx 3.00 \): the Boltzmann factor is essentially 1, so the ratio is the degeneracy ratio \( (2F + 1) \). True.
  4. False: both states are 1s (even parity). It is a magnetic dipole transition.

Quick Revision Table

IdeaFormula or fact
Hydrogen fine structure\( E_{nj} = E_n\left[1 + \frac{\alpha^2Z^2}{n^2}\left(\frac{n}{j + 1/2} - \frac34\right)\right] \)
H, \( n = 2 \) splitting0.365 cm\( ^{-1} \), 10.9 GHz
Lamb shift\( 2s_{1/2} \) above \( 2p_{1/2} \) by about 1.06 GHz
Hyperfine\( E_F = \frac{A}{2}[F(F+1) - I(I+1) - J(J+1)] \); \( E_F - E_{F-1} = AF \)
21 cm1420.4 MHz, \( F = 1\to0 \), magnetic dipole
Cs clock9 192 631 770 Hz, \( F = 3, 4 \)
Mass isotope shift\( R_M = R_\infty/(1 + m_e/M) \); H–D Balmer-\( \alpha \): 0.18 nm
Field isotope shiftnuclear size, heavy atoms, s levels
Natural width\( 1/(2\pi\tau) \), Lorentzian
Doppler width\( 7.16\times10^{-7}\nu_0\sqrt{T/M} \), Gaussian
Collision width\( 1/(\pi\tau_c) \propto \) pressure, Lorentzian
LS levels\( E_J - E_{J-1} = AJ \)
Hundmax \( S \), max \( L \), \( J \) low if less than half full
jj\( j_i = l_i\pm\frac12 \), then \( J \); same \( J \) count as LS

Key Takeaways

What’s Next