Modern Physics · Upper Division

Scattering Theory

Scattering is how we probe the structure of matter — from Rutherford discovering the nucleus to the LHC discovering the Higgs. The formal theory of quantum scattering connects asymptotic states to the S-matrix, and produces Feynman diagrams in the relativistic limit.

PrerequisitesQuantum mechanics (Ch. 22) · Green's functions (Ch. GF) · Angular momentum (Ch. SP)
Learning Goals
  • Define the scattering amplitude f(θ)f(\theta) and relate it to the differential cross section dσ/dΩ=∣f∣2d\sigma/d\Omega = |f|^{2}.
  • Apply the Born approximation to compute scattering amplitudes as Fourier transforms of the potential.
  • Expand the scattering amplitude in partial waves, extract phase shifts δl\delta_{l}, and apply the optical theorem.
  • Derive the Breit-Wigner resonance formula and explain why nuclear resonances can have σ≫πR2\sigma ≫ \pi R^{2}.
  • Write down the Lippmann-Schwinger equation for the T-matrix and relate its Born expansion to Feynman diagrams.

SC.1 Scattering Formalism

A particle of momentum ℏk is scattered by a potential V(r), localized near the origin. Far from the target, the wavefunction is:

ψ(r)→eik⋅r+f(k,k′) eikrr(asymptotic scattered wave)\psi(\mathbf{r}) \to e^{i\mathbf{k}\cdot\mathbf{r}} + f(\mathbf{k}, \mathbf{k}')\, \frac{e^{ikr}}{r} \qquad \text{(asymptotic scattered wave)}(SC.1)

where f(k, k') is the scattering amplitude(k' = k r̂). The differential cross section is:

dσdΩ=∣f(θ,ϕ)∣2(differential cross section)\frac{d\sigma}{d\Omega} = |f(\theta, \phi)|^2 \qquad \text{(differential cross section)}(SC.2)

Total cross section: σtotal=∫∣f∣2 dΩ\sigma_{\text{total}} = \int |f|^2\, d\Omega. The optical theorem relates σtotal\sigma_{\text{total}} to the forward scattering amplitude:

σtotal=4πk Im⁡f(θ=0)(optical theorem)\sigma_{\text{total}} = \frac{4\pi}{k}\, \operatorname{Im} f(\theta = 0) \qquad \text{(optical theorem)}(SC.3)

This is exact — it follows from unitarity of the S-matrix: probability is conserved. Physically: the forward scattered wave must interfere destructively with the incident wave to remove probability from the forward beam (shadow scattering).

SC.2 Born Approximation

For a weak potential V ≪ typical kinetic energies, the Born approximation gives:

f(q)=−m2πℏ2∫e−iq⋅r V(r) d3r=−m2πℏ2 V~(q)f(\mathbf{q}) = -\frac{m}{2\pi\hbar^2} \int e^{-i\mathbf{q}\cdot\mathbf{r}}\, V(\mathbf{r})\, d^3r = -\frac{m}{2\pi\hbar^2}\, \tilde{V}(\mathbf{q})(SC.4)

where q = k − k' is the momentum transfer (|q| = 2k sin(θ/2)). The scattering amplitude is the Fourier transform of V at the momentum transfer.

For Coulomb potential V=Ze2/(4πε0r)V = Ze^2/(4\pi\varepsilon_0 r): V~(q)=Ze2/(ε0q2)\tilde{V}(q) = Ze^2/(\varepsilon_0 q^2). Born cross section: dσdΩ=(2mZe24πε0ℏ2)2(2ksin⁡(θ/2))4=(Ze216πε0E)21sin⁡4(θ/2)\dfrac{d\sigma}{d\Omega} = \dfrac{\left(\frac{2mZe^2}{4\pi\varepsilon_0\hbar^2}\right)^2}{(2k\sin(\theta/2))^4} = \left(\dfrac{Ze^2}{16\pi\varepsilon_0 E}\right)^2 \dfrac{1}{\sin^4(\theta/2)}. This is the Rutherford formula — and it is exact for Coulomb scattering (higher-order Born terms are zero due to the long range of Coulomb).

SC.3 Partial Waves

For a spherically symmetric potential V(r), expand in angular momentum eigenstates. The scattering amplitude:

f(θ)=1k∑l(2l+1) eiδlsin⁡δl Pl(cos⁡θ)(partial wave expansion)f(\theta) = \frac{1}{k} \sum_l (2l+1)\, e^{i\delta_l} \sin\delta_l\, P_l(\cos\theta) \qquad \text{(partial wave expansion)}(SC.5)

where δₗ is the phase shift of the l-th partial wave — the phase acquired relative to free propagation. The partial wave cross section:

σl=4πk2(2l+1)sin⁡2δl≤4πk2(2l+1)(unitarity bound)\sigma_l = \frac{4\pi}{k^2}(2l+1) \sin^2\delta_l \le \frac{4\pi}{k^2}(2l+1) \qquad \text{(unitarity bound)}(SC.6)

Resonances: when δₗ passes through π/2, the partial wave cross section reaches its maximum (unitarity limit). A Breit-Wigner resonance at energy ErE_r with width Γ:

σl(E)≈4πk2(2l+1) Γ2/4(E−Er)2+Γ2/4(Breit-Wigner resonance)\sigma_l(E) \approx \frac{4\pi}{k^2}(2l+1)\, \frac{\Gamma^2/4}{(E - E_r)^2 + \Gamma^2/4} \qquad \text{(Breit-Wigner resonance)}(SC.7)
Example SC.1 — S-wave Scattering from a Hard Sphere

Find the s-wave phase shift and cross section for a hard sphere of radius a(V=∞a (V = \infty for r < a, 0 for r > a) at low energy ka ≪ 1.

Radial equation:Forl=0For l=0, the radial equation gives u(r)=rψ0(r)u(r) = r \psi_{0}(r). Outside: u=Asin⁡(kr+δ0)u = A \sin(kr + \delta_{0}). Atr=a:u(a)=0→ka+δ0=nπ→δ0=−kaAt r=a: u(a) = 0 \to ka + \delta_{0} = n\pi \to \delta_{0} = -ka (for small ka).
Scattering length:Define the scattering length as=−lim⁡k→0δ0/k=aa_s = -\lim_{k\to 0} \delta_0/k = a (positive for hard sphere). Low-energy limit: f≈−as=−a.σtotal=4πa2f \approx -a_{s} = -a. \sigma_{\text{total}} = 4\pi a^{2} (at low energy, four times the geometric cross section!).
Physical:The factor of 4 arises because quantum diffraction occurs in all directions, not just the shadow. At ka ≫ 1 (high energy): σ→2πa2\sigma \to 2\pi a^{2} (geometric πa2\pi a^{2} from shadow + πa2\pi a^{2} from diffraction ring).
Universality:All short-range potentials with the same scattering length asa_{s} have the same low-energy cross section σ→4πas2\sigma \to 4\pi a_{s}^{2}. The scattering length fully characterizes low-energy scattering — this is why ultracold atoms can be described by just asa_{s} regardless of the microscopic potential details.

SC.4 The S-Matrix

The S-matrix (scattering matrix) maps incoming asymptotic states to outgoing asymptotic states: |out⟩ = S|in⟩. For elastic scattering:Sl=e2iδlS_l = e^{2i\delta_l} — a pure phase (unitarity: S†S=1→∣Sl∣=1S^\dagger S = 1 \to |S_l| = 1).

The T-matrix: S = 1 + 2iT. The cross section: dσ/dΩ=∣⟨k′∣T∣k⟩∣2d\sigma/d\Omega = |\langle k'|T|k\rangle|^2. Lippmann-Schwinger equation:

T=V+V G0(E+iε) T(Lippmann-Schwinger for the T-matrix)T = V + V\, G_0(E + i\varepsilon)\, T \qquad \text{(Lippmann-Schwinger for the } T\text{-matrix)}(SC.8)

where G₀ = 1/(E − H₀ + iε) is the free-particle Green's function. Iterating: T = V + VG₀V + VG₀VG₀V + ... — the Born series. Each term is a Feynman diagram in position-space (non-relativistic).

SC.5 Inelastic Scattering and Form Factors

For scattering from an extended object (nucleus, atom), the cross section involves the form factor F(q) — the Fourier transform of the charge distribution:

dσdΩ=(dσdΩ)point∣F(q)∣2(form factor)\frac{d\sigma}{d\Omega} = \left(\frac{d\sigma}{d\Omega}\right)_{\text{point}} |F(q)|^2 \qquad \text{(form factor)}(SC.9)

At small q: F(q) ≈ 1 − q²⟨r²⟩/6 + ... where ⟨r²⟩ is the mean-square radius. The proton charge radius was measured by electron-proton scattering (Hofstadter, 1961):rp≈0.84r_p \approx 0.84 fm. This became the proton radius puzzle when muonic hydrogen measurements gave rp=0.84087r_p = 0.84087 fm vs ordinary hydrogen spectroscopy giving 0.8775 fm — a 7σ discrepancy (since resolved to ~0.84 fm by 2019 CODATA).

Definition SC.1 — Common Traps
  • Cross sections are probabilities per flux: they are not geometric areas except in special limits.
  • Phase shifts contain the interaction: different partial waves contribute differently by energy.
  • Resonances have widths: lifetime and energy uncertainty are linked.
  • Born approximation is weak-scattering: strong potentials require nonperturbative treatment.
Exercises — SC.1–SC.5 Scattering Theory
1.
Verify the optical theorem σ=(4π/k)Imf(0)\sigma = (4\pi/k) Im f(0) for s-wave scattering from a hard sphere of radius a at low energy.
× πa²
Straightforward
2.
Find the differential and total cross section for Born scattering from a Yukawa potential V=V0e−μr/rV = V_{0} e^{-\mu r}/r. How does σ behave at low and high energies?
fm
Intermediate
3.Describe s-wave resonance scattering from an attractive square well. Derive the Breit-Wigner form for the cross section and explain why nuclear resonances can have σ≫πR2\sigma ≫ \pi R^{2}.
Intermediate
4.Derive the eikonal approximation for scattering at high energy (ka ≫ 1). Show how it reduces to the Born approximation for weak potentials, and to Rutherford for Coulomb.
Challenging
Key Takeaways
  • Scattering amplitude f(θ):dσ/dΩ=∣f∣2f(\theta): d\sigma/d\Omega = |f|^{2}. Optical theorem: σ=(4π/k)Imf(0)—\sigma = (4\pi/k) Im f(0) — from unitarity.
  • Born approximation: f=−m/(2πℏ2)V~(q)f = -m/(2\pi\hbar^{2}) \tilde{V}(q). Cross section = |FT of potential|². Rutherford = exact Born.
  • Partial waves: f=(1/k)∑(2l+1)eiδlsin⁡δlPl(cos⁡θ)f = (1/k)\sum(2l+1) e^{i\delta_{l}} \sin\delta_{l} P_{l}(\cos\theta). Phase shifts encode all scattering information.
  • S-wave at low energy: σ→4πas2(as\sigma \to 4\pi a_{s}^{2} (a_{s} = scattering length). Hard sphere: σ=4πa2(4\sigma = 4\pi a^{2} (4× geometric).
  • Resonances: δl→π/2,σl\delta_{l} \to \pi/2, \sigma_{l} → unitarity limit 4π(2l+1)/k24\pi(2l+1)/k^{2}. Breit-Wigner: Lorentzian in energy.
  • S-matrix: S=e2iδS = e^{2i\delta}. T=V+VG0TT = V + VG_{0}T (Lippmann-Schwinger). Born series = Feynman diagrams.