research32 min studyLesson 708 in Physics

Background-field gauge and renormalization

Preserve background covariance while integrating quantum fluctuations and interpret the domain of one-loop running.

Before you begin

On this page

The question for today

What you’ll understand

Distinguish background transformations from quantum gauge transformations.
Explain why background covariance constrains renormalization.
Calculate one-loop coupling evolution with explicit matter and scale assumptions.

The background-field method reorganizes a gauge theory around a chosen connection while integrating its fluctuations. It preserves a useful covariance with respect to the background even though the quantum redundancy is gauge fixed. This makes some effective-action constraints resemble familiar gauge-invariant classical expressions. The convenience must be interpreted carefully: an off-shell background effective action can still depend on the quantum gauge parameter, and a perturbative beta function does not determine the strongly coupled spectrum.

Split a connection without changing the theory

Write the full connection as background B plus fluctuation a. Under a background transformation, B changes as a connection and a changes homogeneously in the adjoint. Their sum therefore has exactly the connection transformation fixed earlier. A different operation, called a quantum gauge transformation, keeps B fixed and assigns the required full-connection change to a. The gauge condition must specify which of these operations it fixes.

Choose the background-covariant divergence of a as the gauge-fixing function. It transforms in the adjoint under simultaneous background transformations, making its squared norm invariant. The ghost operator then involves a background derivative followed by the derivative built from the full connection. It is generally not the ordinary Lorenz-gauge operator with a decorative extra field label. Vertices and propagators reorganize accordingly, while a consistently calculated physical result must agree with other valid formulations.

A_\mu=B_\mu+a_\mu,\quad B_\mu'=UB_\mu U^{-1}+\frac{i}{g}U\partial_\mu U^{-1},\quad a_\mu'=Ua_\mu U^{-1},\qquad G=\overline D_\mu a_\mu

The background derivative is bar-D X=partial-X-minus-i-g[B,X]. This is background covariance in the module convention, not a statement that the fluctuation alone is a connection under that transformation.↔ Scroll wide equations horizontally.

Use covariance to organize counterterms

After integrating fluctuations, the effective action of the background respects background gauge invariance when the regulator and counterterms are compatible. Its two-point divergence is then completed into a background curvature invariant, tying higher background vertices to the same structure. In a conventional background renormalization scheme, the coupling and background wave-function factors satisfy a reciprocal square-root relation. This permits a coupling counterterm to be inferred efficiently from the background two-point calculation.

The background field is still an argument of an effective action, not an automatically measurable classical field. Dependence on the quantum gauge parameter can survive in off-shell terms even while background covariance is exact. Truncated derivative expansions and omitted diagrams can violate the intended identities. One must state the subtraction scheme, regulator, and approximation order before using background quantities as quantitative predictions or comparing their coefficients with a different gauge organization.

Z_gZ_B^{1/2}=1,\qquad \mu\frac{dg}{d\mu}=-\frac{b_0g^3}{16\pi^2}+O(g^5),\qquad b_0=\frac{11N}{3}-\frac{2N_f}{3}

The beta coefficient is for four-dimensional SU(N) with N-f active Dirac fermions in the fundamental representation and no scalars. The displayed Z relation uses a compatible background-field normalization and renormalization prescription.↔ Scroll wide equations horizontally.

Integrate running only in its controlled regime

At one loop the derivative of inverse g squared is constant with respect to logarithmic scale. Integrating between two scales gives an additive logarithm in inverse coupling squared. Positive b-zero makes the coupling decrease toward higher scales, which is asymptotic freedom. The formula applies when the coupling stays perturbatively small and the active degrees of freedom do not change across a mass threshold. Threshold matching and higher orders modify a longer-range extrapolation.

The opposite running direction can drive the perturbative expression toward strong coupling, but continuing it until its denominator vanishes is not a controlled prediction of a physical pole. Confinement, possible infrared fixed points, and mass generation depend on the theory and nonperturbative dynamics. Matter content also changes the leading coefficient, so the pure Yang-Mills conclusion cannot be transferred to every gauge theory. A useful calculation separates the sign diagnosis at weak coupling from claims about an uncomputed infrared phase.

\frac1{g^2(\mu)}=\frac1{g^2(\mu_0)}+\frac{b_0}{8\pi^2}\ln\frac{\mu}{\mu_0}\quad\text{at one loop}

Both scales have the same energy units and their logarithmic ratio is dimensionless. The relation neglects higher-loop terms and assumes the stated fixed active matter content over the scale interval.↔ Scroll wide equations horizontally.

Work it through

A weak-coupling scale comparison

For pure SU(3), take $g(\mu_0)=0.5$ and $\mu=e\mu_0$ . Use the one-loop formula to estimate the new coupling, taking $\pi^2\approx9.87$ . State why this does not predict an infrared mass spectrum.

Pure SU(3) has $N_f=0$ and $b_0=11$ . The initial inverse coupling squared is $1/(0.5)^2=4$ , and the logarithmic scale ratio is one.
The added inverse-coupling term is $11/(8\times9.87)\approx0.1393$ . Therefore $1/g^2(\mu)\approx4.1393$ , giving $g(\mu)\approx0.4915$ , slightly smaller at the higher scale.
The small change is a weak-coupling one-loop result for a fixed theory. It neither computes a gauge-invariant bound-state mass nor controls evolution into a strongly coupled infrared region, where this approximation can cease to apply.

The result

The one-loop estimate gives coupling approximately 0.4915 at the higher scale, illustrating asymptotic freedom without determining the infrared spectrum.

The living notebook

Explore the model

Interactive lab

Interactive lab · scattering

A schematic sensitivity to coupling

Reset and set coupling $g=0.5$ and angle ninety degrees; predict illustrative weight $g^4(1+\cos^2\theta)=0.0625$ . Compare a fresh Reset run with $g=1$ , angle ninety degrees; predict one. This is a schematic dimensionless weight, not a Yang-Mills cross section, beta function, background-field calculation, or measured scale evolution.

Incoming termsOutgoing termsAngle θ

Coupling factor g⁴: 0.24
Angular factor 1 + cos²θ: 1.329
Illustrative weight: 0.319

Coupling g0.7Scattering angle θ55°

t = 0 s

A schematic term diagram with dimensionless weight g⁴(1 + cos²θ). Lines denote terms, not particle paths; this is not a process cross-section or loop calculation. Parameter changes pause playback and recompute the model at the current elapsed time. Reset restores both parameters and returns time to zero. Playback starts paused. Adjust either parameter, then choose Play or Step.

Look forA coupling change can strongly affect the size of a term, but the lab does not determine how that coupling runs with scale.

Pause & reason

Check your understanding

Interactive checks need JavaScript. You can still read the explanation.

Read the checkpoint explanation

Background covariance constrains functional form and counterterms. It is distinct from independence of the quantum gauge-fixing parameter for arbitrary off-shell terms or from a solution of strong-coupling dynamics.

Your turn

Try it yourself

For SU(3) with no scalars, compute b-zero for three and eighteen active fundamental Dirac fermions. At weak $g=0.4$ , determine the one-loop beta-function sign in each case and estimate its magnitude using $\pi^2=9.87$ .

Take a moment to make a prediction or work it out on paper. Open a hint when you need one.

Give me a hint

Insert the active flavor count before multiplying by g cubed; a negative b-zero reverses the leading running direction.

Compare with the solution

For three flavors $b_0=11-2=9$ , giving beta $-9(0.4)^3/(16\times9.87)\approx-0.00365$ . For eighteen flavors $b_0=11-12=-1$ , giving beta approximately $+0.000405$ . The first theory is asymptotically free at this order, while the second has increasing coupling toward higher scales in the weak-coupling calculation. These signs alone do not establish either theory's infrared phase or global ultraviolet completion.

Keep the idea

Take it with you

Background and quantum gauge transformations act differently on the split fields.
Background covariance constrains renormalization without making every off-shell term physical.
One-loop running is conditional on weak coupling and fixed active matter content.

Read further

Original sources and deeper explanations to follow at your own pace.

Optional progress, saved in this browser when available. Revisit any lesson at any time.

Lesson not marked complete.