Skip to content

Scattering formulas

An amplitude becomes a cross section only after specifying incoming flux, final-state counting, and the observable. This guide collects four-dimensional formulas for stable asymptotic particles with relativistic normalization, then checks them on elastic scattering of identical real scalars. It covers vacuum scattering; confinement, unstable external particles, and unresolved infrared singularities need additional treatment.

Helpful background. In and Out States explains the asymptotic-state premise. Cross Sections and Decay Rates derives the normalization and phase-space factors used here.

Use S=1+iTS=1+iT and define the invariant amplitude by

⟨f∣T∣i⟩=(2π)4δ(4)(Pf−Pi)Mfi.\langle f|T|i\rangle=(2\pi)^4\delta^{(4)}(P_f-P_i)\mathcal M_{fi}.

The one-particle normalization is ⟨p′∣p⟩=2Ep(2π)3δ(3)(p′−p)\langle\mathbf p'|\mathbf p\rangle=2E_{\mathbf p}(2\pi)^3\delta^{(3)}(\mathbf p'-\mathbf p), with an additional Kronecker delta for discrete labels. The labeled nn-particle phase-space measure is

dΦn(P)=(2π)4δ(4) ⁣(P−∑a=1npa)∏a=1nd3pa(2π)3 2Ea.\mathrm d\Phi_n(P) =(2\pi)^4\delta^{(4)}\!\left(P-\sum_{a=1}^n p_a\right) \prod_{a=1}^n\frac{\mathrm d^3\mathbf p_a}{(2\pi)^3\,2E_a}.

For two incoming particles with momenta p1,p2p_1,p_2 and masses m1,m2m_1,m_2, define

F=4(p1⋅p2)2−m12m22.\mathcal F=4\sqrt{(p_1\cdot p_2)^2-m_1^2m_2^2}.

Then, for a phase-space integration that counts every labeling,

dσ=∣M∣2‾FdΦn∏sns!.\mathrm d\sigma =\frac{\overline{|\mathcal M|^2}}{\mathcal F} \frac{\mathrm d\Phi_n}{\prod_s n_s!}.

Here nsn_s is the number of identical final particles of species ss. The bar means a sum over unobserved final labels and an average over an explicitly unpolarized incoming ensemble. Do not average a polarization that was prepared or sum over a final label that the measurement resolves. These factors follow from the wave-packet normalization and transition probability in Schwartz 2014, § 5.1, pp. 57–63.

In the center-of-mass frame, let s=(p1+p2)2s=(p_1+p_2)^2, and let pip_i and pfp_f be the magnitudes of the initial and final three-momenta. For distinguishable final particles,

dσdΩ=164π2spfpi ∣M∣2‾.\frac{\mathrm d\sigma}{\mathrm d\Omega} =\frac{1}{64\pi^2s}\frac{p_f}{p_i}\, \overline{|\mathcal M|^2}.

For two identical final particles integrated over the full labeled sphere, multiply by 1/2!1/2!. An equivalent convention integrates only one representative of each final-state pair and omits that factorial. Use one convention consistently.

For equal-mass elastic scattering above threshold, s>4m2s>4m^2, the ratio pf/pip_f/p_i equals one. A canonically normalized real scalar with Lint=−λϕ4/4!\mathcal L_{\mathrm{int}}=-\lambda\phi^4/4! has a tree vertex −iλ-i\lambda, so Mtree=−λ\mathcal M_{\mathrm{tree}}=-\lambda. Therefore,

dσtreedΩ=λ2128π2s,σtree=λ232πs.\frac{\mathrm d\sigma_{\mathrm{tree}}}{\mathrm d\Omega} =\frac{\lambda^2}{128\pi^2s}, \qquad \sigma_{\mathrm{tree}}=\frac{\lambda^2}{32\pi s}.

The second equation follows by integrating the first over 4π4\pi. Its mass dimension is −2-2, as required for a cross section. The 4!4! in the Lagrangian cancels the attachments at a vertex; it does not replace the final-state 2!2!. There is no incoming factorial for two separately specified beams. The momentum-space rules explain the vertex, and the scalar calculation checks the phase-space Jacobian.

For a parent of mass MM at rest, the analogous differential decay width is

dΓ=12M ∣M∣2‾dΦn∏sns!.\mathrm d\Gamma=\frac{1}{2M}\, \overline{|\mathcal M|^2} \frac{\mathrm d\Phi_n}{\prod_s n_s!}.

For a narrow unstable parent this expression is used within a controlled resonance approximation; it does not declare the parent to be an exact stable asymptotic state. For genuinely stable external legs, LSZ reduction extracts pole residues and divides by the interpolating-field overlaps. Amputation and physical state normalization are both needed.

The observable must say which final states and angular region are included. Beyond tree level, verify that the chosen inclusive observable permits cancellation of the relevant infrared singularities; a bare exclusive expression can diverge. The optical theorem supplies a unitarity check on the imaginary part of an amplitude, while Measurement Functions and Inclusive Observables explains how to define the measured rate.

  • Schwartz, Matthew D. Quantum Field Theory and the Standard Model. Cambridge University Press, 2014. DOI.