Skip to content

Partial-Wave Expansion

Partial-wave expansion solves central-potential scattering by decomposing the wavefunction into angular-momentum channels. For a rotationally invariant potential V(r)V(r), different ℓ\ell values do not mix, and scattering is encoded in one phase shift δℓ\delta_\ell per partial wave.

The central result is the elastic scattering amplitude

f(θ)=1k∑ℓ=0∞(2ℓ+1)eiδℓsin⁡δℓ Pℓ(cos⁡θ),f(\theta) = \frac{1}{k} \sum_{\ell=0}^\infty (2\ell+1) e^{i\delta_\ell} \sin\delta_\ell\, P_\ell(\cos\theta),

or equivalently

f(θ)=12ik∑ℓ=0∞(2ℓ+1)(e2iδℓ−1)Pℓ(cos⁡θ).f(\theta) = \frac{1}{2ik} \sum_{\ell=0}^\infty (2\ell+1) \left( e^{2i\delta_\ell}-1 \right) P_\ell(\cos\theta).

This page is the canonical home for the expansion. The interpretation and extraction of δℓ\delta_\ell are treated in Phase Shifts.

For an incident wave along the zz-axis,

eikz=∑ℓ=0∞(2ℓ+1)iℓjℓ(kr)Pℓ(cos⁡θ),e^{ikz} = \sum_{\ell=0}^\infty (2\ell+1)i^\ell j_\ell(kr) P_\ell(\cos\theta),

where jℓj_\ell are spherical Bessel functions and PℓP_\ell are Legendre polynomials.

For the ordinary, spherical, and Riccati-Bessel conventions used in radial matching, see Bessel Functions.

For a central potential, the full stationary wavefunction can be written as

ψ(r)=∑ℓ=0∞Rℓ(r)Pℓ(cos⁡θ)\psi(\mathbf r) = \sum_{\ell=0}^\infty R_\ell(r) P_\ell(\cos\theta)

for an azimuthally symmetric incident beam. More generally, spherical harmonics YℓmY_\ell^m diagonalize angular momentum.

Writing

Rℓ(r)=uℓ(r)r,R_\ell(r)=\frac{u_\ell(r)}{r},

the radial Schrödinger equation is

−ℏ22md2uℓdr2+[V(r)+ℏ2ℓ(ℓ+1)2mr2]uℓ=Euℓ.- \frac{\hbar^2}{2m} \frac{d^2u_\ell}{dr^2} + \left[ V(r) + \frac{\hbar^2\ell(\ell+1)}{2mr^2} \right] u_\ell = Eu_\ell.

The term

ℏ2ℓ(ℓ+1)2mr2\frac{\hbar^2\ell(\ell+1)}{2mr^2}

is the centrifugal barrier. At low energy it suppresses high-ℓ\ell penetration into the interaction region.

For a short-range potential, outside the interaction region each radial channel is a free spherical wave with a phase shift:

uℓ(r)∼Aℓsin⁡(kr−ℓπ2+δℓ).u_\ell(r) \sim A_\ell \sin\left( kr-\frac{\ell\pi}{2}+\delta_\ell \right).

The free solution would have δℓ=0\delta_\ell=0. The potential changes the phase of the outgoing wave relative to the free radial wave.

For elastic scattering in one channel, define

Sℓ=e2iδℓ.S_\ell=e^{2i\delta_\ell}.

Unitarity requires

∣Sℓ∣=1|S_\ell|=1

for each partial wave when no absorption or inelastic channel is open. The factor of 22 appears because SℓS_\ell compares outgoing and incoming radial waves, while δℓ\delta_\ell is the standing-wave phase shift.

Unitarity owns the Argand-disk geometry, the ηℓ<1\eta_\ell\lt1 inelastic case, and elastic and reaction bounds.

The scattered amplitude is

f(θ)=12ik∑ℓ=0∞(2ℓ+1)(Sℓ−1)Pℓ(cos⁡θ).f(\theta) = \frac{1}{2ik} \sum_{\ell=0}^\infty (2\ell+1) (S_\ell-1) P_\ell(\cos\theta).

For elastic scattering with Sℓ=e2iδℓS_\ell=e^{2i\delta_\ell}, this becomes

f(θ)=1k∑ℓ=0∞(2ℓ+1)eiδℓsin⁡δℓPℓ(cos⁡θ).f(\theta) = \frac{1}{k} \sum_{\ell=0}^\infty (2\ell+1) e^{i\delta_\ell} \sin\delta_\ell P_\ell(\cos\theta).

This expansion converts a three-dimensional scattering problem into a set of one-dimensional radial problems.

Partial-Wave Cross Sections carries this decomposition through the angular integration and develops elastic, reaction, and total channel sums, threshold scaling, angular-cut interference, and practical truncation tests.

For a finite-range potential of range RR, the largest important angular momenta are roughly

ℓmax⁡∼kR.\ell_{\max}\sim kR.

Classically, ℓ/k\ell/k corresponds to an impact parameter. Partial waves with impact parameter much larger than the range of the potential barely interact.

At low energy, the ss-wave ℓ=0\ell=0 often dominates. At higher energy, many partial waves contribute and semiclassical impact-parameter intuition becomes useful. Stationary Phase in Quantum Mechanics shows how saddles of the large-ℓ\ell sum recover the deflection function and why coalescing saddles require a rainbow uniform approximation.

Photoelectron Spectroscopy applies partial-wave amplitudes and scattering phases to photoionization thresholds, Cooper minima, and electron angular distributions.

  • Using partial waves for a noncentral potential without accounting for angular-momentum mixing.
  • Forgetting the centrifugal barrier in the radial equation.
  • Confusing δℓ\delta_\ell with SℓS_\ell; they are related by Sℓ=e2iδℓS_\ell=e^{2i\delta_\ell} for elastic scattering.
  • Truncating the partial-wave sum without checking convergence.
  • Applying short-range formulas directly to unscreened Coulomb scattering.
  1. Show that if all phase shifts vanish, the partial-wave scattering amplitude is zero.
Solution

Using

f(θ)=1k∑ℓ=0∞(2ℓ+1)eiδℓsin⁡δℓPℓ(cos⁡θ),f(\theta) = \frac{1}{k} \sum_{\ell=0}^\infty (2\ell+1) e^{i\delta_\ell} \sin\delta_\ell P_\ell(\cos\theta),

if δℓ=0\delta_\ell=0 for every ℓ\ell, then sin⁡δℓ=0\sin\delta_\ell=0 for every term. Hence f(θ)=0f(\theta)=0, as expected for no scattering.

  1. Explain why low-energy scattering is often dominated by ℓ=0\ell=0.
Solution

The effective radial potential includes the centrifugal barrier

ℏ2ℓ(ℓ+1)2mr2.\frac{\hbar^2\ell(\ell+1)}{2mr^2}.

For ℓ>0\ell>0, this barrier suppresses penetration into the short-range interaction region at low energy. The ℓ=0\ell=0 channel has no centrifugal barrier, so it usually dominates.

  1. If a potential has range RR, give a semiclassical estimate for the number of important partial waves at wave number kk.
Solution

The angular momentum is roughly ℓ∼kb\ell\sim kb, where bb is the impact parameter. Scattering is important mainly for b≲Rb\lesssim R, so

ℓmax⁡∼kR.\ell_{\max}\sim kR.

Thus the number of contributing partial waves grows with kRkR.

  • J. R. Taylor, Scattering Theory: The Quantum Theory of Nonrelativistic Collisions, Dover, 2006.
  • R. Shankar, Principles of Quantum Mechanics, 2nd ed., Springer, 1994.
  • A. Messiah, Quantum Mechanics, Dover, 1999.
  • R. G. Newton, Scattering Theory of Waves and Particles, 2nd ed., Springer, 1982.