OpenWorldLab

Loading interactive view…

Past the textbook

The Zeeman effect

A magnetic field splits each of hydrogen’s levels into states with different $m_j$. Lyman-α breaks into a pattern of components, and above a tesla or so the pattern changes as spin lets go of orbit.

What you are looking at

Hydrogen in 2p, m = +1, with a 1 T field along z. The panel plots n = 2’s eight states against field: the 2s pair in amber, the six 2p states in grey. Below, Lyman-α’s components: π light in amber, σ in blue.

The physics

In a weak field each fine-structure level ∣nljmj⟩|n l j m_j\rangle shifts by

ΔE=gJ mj μBB,gJ=1+j(j+1)+s(s+1)−l(l+1)2j(j+1),\Delta E = g_J\, m_j\, \mu_B B, \qquad g_J = 1 + \frac{j(j+1) + s(s+1) - l(l+1)}{2j(j+1)},

with gs≈2.0023g_s \approx 2.0023 in place of 2. 2p³⁄₂ has gJ=4/3g_J = 4/3, 2p½ has 2/32/3, 2s½ has 2.

When μBB\mu_B B outgrows the 10.97 GHz fine structure (near 0.8 T), spin and orbit decouple. Each state then shifts by (ml+gsms)μBB(m_l + g_s m_s)\mu_B B: the Paschen–Back effect. The lab diagonalises the fine structure and the field together in each n l, so it follows the whole change; its slopes match both limits.

What to measure

Read the splitting of 2p³⁄₂ at 0.1 T and find its g factor.

  1. At 0.1 T, read the spacing of 2p³⁄₂’s four states and divide by μ_B B = 1.40 GHz: the g factor.
  2. Slide to 3 T and read the 2p slopes: close to m_l + 2m_s, that is −2, −1, 0, 0, 1 and 2 times μ_B.

A common misconception

“Lines split into three in a magnetic field.” That is the “normal” Zeeman effect, of a spinless atom. Real hydrogen has spin, so its lines split into more components with uneven spacings; the three-line pattern returns only in the Paschen–Back limit, roughly.

What the model leaves out

  • Hyperfine structure from the proton’s magnetic moment, about 1.4 GHz in 1s and smaller above.
  • The shift of the cloud itself: the cloud shows the state you choose, and the field acts on the energies drawn in the panel.
  • The diamagnetic term, B²⟨x² + y²⟩/8: a few MHz for n ≤ 4 at the panel’s 3 T, but it grows as n⁴B², is hundreds of MHz for n = 4 at 10 T, and dominates for Rydberg states.

Things to try

  • Switch to Hα: many more components, from three l pairs.
  • Watch for avoided crossings: states with the same mjm_j never cross.

References

  • C. J. Foot, Atomic Physics (Oxford, 2005), chapters 1, 2 and 5.
  • H. A. Bethe and E. E. Salpeter, Quantum Mechanics of One- and Two-Electron Atoms (Springer, 1957), parts I and III.

More experiments, and how the lab works, are in the overview of atoms and orbitals.