Scattering Theory

Scattering theory asks what a collision reveals about an interaction. Prepare an incoming particle or wave packet, let it encounter a target, and measure the outgoing directions, energies, and internal states. Quantum mechanics predicts complex amplitudes; their interference and probability currents determine the measured cross sections.

This volume begins with nonrelativistic potential scattering, chiefly in three dimensions. It then develops angular-momentum methods, low-energy behavior, resonances, and coupled channels. One-dimensional scattering, long-range Coulomb forces, identical particles, and connections to field theory receive separate treatments where their assumptions differ.

Required background. Free particles explain incident waves; probability current explains flux; continuum normalization explains how stationary scattering states relate to normalized wave packets.

From a wave to a cross section

For a short-range potential with one elastic channel, take an incident wave of unit amplitude and wave vector $\mathbf k$. With time dependence $e^{-iEt/\hbar}$, the large-distance wave has the form

$$\psi^{(+)}(\mathbf r)\sim e^{i\mathbf k\cdot\mathbf r} +f(\hat{\mathbf r})\frac{e^{ikr}}r.$$

Here $k$ is a wave number and the momentum magnitude is $\hbar k$. The first term describes the incident beam; the second describes the outgoing spherical wave. The superscript $(+)$ specifies the outgoing boundary condition on that scattered wave. It is not a sign of the energy.

The coefficient $f$ has length units. Comparing the outgoing rate in a solid angle $d\Omega$ with the incident flux gives

$$\frac{d\sigma}{d\Omega}=|f(\hat{\mathbf r})|^2.$$

Thus a cross section is an area characterizing the scattering response, rather than a dimensionless probability for a single state. The phase of $f$ still matters: amplitudes interfere before their modulus is squared. States and observables develops this connection from preparation to detection.

Choose a route through the volume

Your question	Chapter
How are measured counts related to a scattering wave?	Scattering states and observables
How can the exact scattering problem be organized and approximated?	Operators and unitarity
How do rotational symmetry and low energy simplify a collision?	Partial waves and thresholds
What do a resonance or a new outgoing channel reveal?	Poles, resonances, and channels
How do the methods work on solvable potentials?	Worked problems
How are observables extracted from a numerical radial solution?	Scattering computations
How can formulas from different sources be compared?	Conventions and connections

For a first reading, follow the first three chapters in order, using the worked problems along the way. Start with the scattering experiment if cross sections are new, or with one-dimensional scattering if reflection and transmission from barriers are more familiar.

Three ways to simplify a collision

Weak interaction. The first Born approximation replaces the distorted wave inside the interaction region by the incident wave. The result is related to the potential’s Fourier transform. A simple transform does not establish the approximation’s accuracy; Born validity checks repeated scattering and threshold effects.

Rotational symmetry. For a central potential, partial waves separate angular-momentum channels. This separation is exact; approximations enter when phase shifts are estimated or the channel sum is truncated. It remains useful for strong potentials where first Born theory fails.

Low energy. For suitably short-range interactions, threshold scattering can often be described by a scattering length and effective range. This is a low-momentum expansion, not a weak-potential expansion. Large scattering lengths and near-threshold poles require particular care.

These approaches can be combined. For example, Born theory can estimate partial-wave phases, and an accurate radial calculation can test those estimates.

Know when the setting changes

An unscreened Coulomb potential has a long-range phase, so its asymptotic states require Coulomb scattering. For identical particles, indistinguishable alternatives must be combined as amplitudes and final states counted consistently. Inelastic channels can have different outgoing speeds, introducing flux factors absent from the simplest elastic formula.

Resonances connect energy-dependent scattering with long-lived states and complex poles. A peak can be shifted or hidden by interference and thresholds, so a cross-section maximum is not the definition of a resonance. The field-theory bridge retains amplitudes, unitarity, and poles while adding relativistic kinematics and processes with changing particle number.

Exercises

Check the dimensions. Why must $f$ have length units in the asymptotic wave above?

Solution

The incident term is dimensionless in the chosen unit-amplitude convention. The scattered term $f e^{ikr}/r$ must have the same units, so $f$ has length units. Accordingly $|f|^2$ has area units, as required for a differential cross section per dimensionless solid angle.

Choose a starting method. A hard sphere is a strong interaction, but it is spherically symmetric. Should you begin with first Born theory or angular-momentum channels? What changes for a weak, smooth Gaussian potential?

Solution

Use partial waves and exact boundary matching for the hard sphere. A weak Gaussian suggests first Born theory, provided its coupling, range, and collision energy satisfy the validity conditions. Rotational symmetry can still be used for the Gaussian; the choices describe different simplifications and are compatible.

References

• Newton, R. G. Scattering Theory of Waves and Particles. 2nd ed., Dover (2002; reprint of the 1982 edition). Partial waves, analytic structure, and inverse problems.
• Sakurai, J. J., and J. Napolitano. Modern Quantum Mechanics. 3rd ed., Cambridge University Press (2020). Introductory potential scattering and approximation methods.
• Taylor, J. R. Scattering Theory: The Quantum Theory of Nonrelativistic Collisions. Dover (2006; reprint of the 1972 edition). States, amplitudes, operators, and collision observables.