Perturbative QCD and Partons
Perturbative QCD turns a short-distance question into a controlled chain: define the QCD parameters, identify the measured observable, separate short- and long-distance physics, evolve between scales, perform the convolution, and attach an uncertainty statement with the same scheme and kinematic assumptions. Enter through the observable and hierarchy you actually have—not through the name of a formalism.
Check readiness, then choose a route
Section titled “Check readiness, then choose a route”The chapter has hard dependencies, but it does not have a single mandatory reading order. A required-background link on a leaf is a hard dependency for that result. The order below is a suggested complete pass; readers who can already formulate an infrared-safe observable or a factorization theorem may enter later.
Use this unscored diagnostic before choosing a route. Each symptom is visible in a calculation, and each repair points to the capability that removes it.
| If the calculation currently… | Missing capability | Exact repair |
|---|---|---|
| treats a small coupling as sufficient even when | Power counting with large logarithms | Controlled effective-field-theory expansions |
| changes the number of active quarks without a matching relation | Decoupling at a heavy threshold | Decoupling theorems and threshold corrections |
| cancels an infrared divergence only after assigning a parton-by-parton observable | Inclusive real–virtual cancellation | Bloch–Nordsieck and KLN cancellation |
| calls a jet observable safe without testing a soft emission or collinear split | Infrared and collinear safety | Infrared and collinear safety |
| multiplies a hard cross section by “the PDF” without a scheme, scale, or convolution | Collinear factorization | Collinear factorization and splitting amplitudes |
| uses one renormalization scale for an observable with a hard scale and a much smaller transverse scale | Rapidity as well as virtuality evolution | Rapidity renormalization |
| combines scale, PDF, and fit errors by automatic quadrature | Correlation-aware validation | Validation and theory uncertainties |
The routes appear once, in the order useful for a first full pass.
| Route | Use it when the central question is | Main output |
|---|---|---|
| QCD fields, scales, and the perturbative domain | Which degrees of freedom and scale hierarchies make fixed-order QCD credible? | A convention-complete QCD model and a perturbative-domain test |
| Short-distance parameters, mass schemes, and thresholds | How should and quark masses be run, matched, and converted? | A scheme- and flavor-number-complete parameter record |
| Inclusive annihilation and the emergence of jets | How do color, quark charges, real–virtual cancellation, and jet structure first appear? | The inclusive ratio with its infrared qualifications |
| Deep-inelastic scattering and the parton model | How do structure functions expose momentum fractions and scaling? | A kinematic and tensor decomposition tied to the parton model |
| Collinear factorization and operator-defined PDFs | What exactly is separated into hard coefficients and hadron structure? | An operator-, scheme-, and power-correction-complete factorization statement |
| DGLAP evolution and scaling violation | How do PDFs change between factorization scales? | Spacelike evolution with moment and sum-rule checks |
| Fragmentation functions and timelike evolution | How is an identified final-state hadron described? | A timelike factorization and evolution statement with momentum conservation |
| High-energy QCD and small-x evolution | When do powers of need their own organization? | A BFKL-domain test and its collinear and unitarity boundaries |
| TMD factorization, rapidity evolution, and Glauber limits | How is a measured small transverse momentum retained? | A soft-subtracted, two-scale TMD statement and a process-validity check |
| Hadron-collider factorization and parton luminosities | How do incoming hadrons become a fiducial collider rate? | A cut-aware convolution with channel and PDF correlations |
| QCD radiation, jets, and event shapes | How do partonic singularities become measurable jet observables? | An IRC-safety test and a hierarchy-aware event-shape description |
| QCD prediction and uncertainty record | How should all approximations and correlations be reported? | A reproducible, covariance-aware prediction and uncertainty record |
The prediction chain
Section titled “The prediction chain”For an observable characterized by a hard scale , the reusable structure is
with the obvious deletion or replacement of incoming distributions for annihilation, DIS, fragmentation, or TMD observables. The formula is not yet a prediction until six questions have answers:
This factorized organization and the conditions under which soft attachments cancel are developed in Collins, Soper, and Sterman 1989, §§2–6, pp. 1–67; representative annihilation and DIS derivations appear in Schwartz 2014, chs. 26 and 32.
- What is measured? State the fiducial cuts, jet or event-shape definition, binning, and inclusiveness.
- Which hierarchy is assumed? Record , masses, transverse or endpoint scales, and every logarithm that is counted as large.
- What is factorized? Name the hard coefficient, PDF, fragmentation function, TMD, jet or soft object, and the theorem or controlled approximation connecting them.
- How are scales connected? Give the perturbative order, renormalization and factorization schemes, active flavors, threshold matches, and evolution accuracy.
- What cancels? Verify scale dependence, flavor and momentum sum rules, real–virtual cancellation, and regulator independence to the claimed order.
- What remains uncertain? Preserve correlations among perturbative, PDF, parameter, nonperturbative, numerical, and experimental components instead of counting a shared source twice.
The hierarchy selects the description, not just the numerical size of :
| Observable hierarchy | Natural description | Leading warning sign |
|---|---|---|
| one inclusive hard scale, | fixed order, possibly with collinear factorization | thresholds, endpoints, or vetoes introduce another scale |
| with measured recoil | TMD factorization and rapidity evolution | a process permits uncancelled Glauber exchange |
| for an event shape | hard–jet–soft factorization and resummation | the measurement is not IRC safe or is non-global |
| with | high-energy resummation matched to collinear evolution | linear evolution approaches an unitarity or high-density boundary |
| any nominally hard process with a scale near | nonperturbative input or a different expansion | fixed-order scale variation stops being a control argument |
Informal synthesis check
Section titled “Informal synthesis check”This is an informal, unscored work-product check. Consider production of a color-singlet state of invariant mass in hadron collisions, differential in a recoil variable and restricted by fiducial cuts. A satisfactory analysis should:
- define the measurement and identify separately the inclusive, small-recoil, endpoint, and small- limits;
- write the collinear convolution, including the active-flavor scheme, , , and the order of every coefficient and evolution kernel;
- explain when the small-recoil bin needs TMD or equivalent resummation, and state the process-specific factorization condition;
- verify momentum conservation, fixed-order expansion of any matched result, and cancellation of unphysical scales to the claimed accuracy; and
- provide a correlation-aware uncertainty record that keeps theoretical and experimental covariance distinct until comparison.
Use these acceptance criteria and repairs:
| Criterion | A minimally complete answer contains | Repair if absent |
|---|---|---|
| Observable | a measurement function or an equivalent exact definition of cuts and bins | Phase-space integration and measurement functions |
| Factorization | every convolution variable, scheme, scale, and leading power remainder | Factorization operator structure |
| Evolution | boundary data, kernels, logarithmic accuracy, thresholds, and a sum-rule check | Coefficient evolution |
| Regime control | an explicit test for recoil, endpoint, small-, and Glauber limits | Rapidity divergences and Glauber limits |
| Uncertainty | named variations or nuisance parameters, correlations, validation tests, and provenance | Validation and theory uncertainties |
Purpose-keyed exits
Section titled “Purpose-keyed exits”| When the next task is… | Continue to… |
|---|---|
| prove the hard–collinear–soft separation for a scattering observable | Infrared Structure and Factorization |
| compute matching coefficients and resum a new hierarchy | Multiscale Effective Theories |
| connect collinear distributions to spin, spatial structure, and wider hadron tomography | Partonic structure, spin, and hadron tomography |
| connect QCD scales to hot matter and equilibration | Scale hierarchies in hot gauge theories |
| study energy-flow operators and collider event shapes in a conformal theory | Event shapes and energy correlators |
| assess release-specific precision results or open factorization questions | Amplitudes and precision scattering |
References
Section titled “References”- Collins, John C., Davison E. Soper, and George Sterman. “Factorization of Hard Processes in QCD.” In Perturbative Quantum Chromodynamics, edited by A. H. Mueller, 1–91. World Scientific, 1989. DOI. Open PDF.
- Schwartz, Matthew D. Quantum Field Theory and the Standard Model. Cambridge University Press, 2014, chs. 26 and 32. DOI.