Ghost Vertices in Gauge Theory & Pure-Spinor Formalism

• Ghost vertices are interaction points involving ghost degrees of freedom introduced via gauge fixing, crucial for maintaining BRST symmetry in gauge theories.
• Studies in Landau gauge show that the ghost–gluon vertex remains near its tree-level value, indicating controlled nonperturbative effects and finite infrared behavior.
• In pure-spinor and supersymmetric frameworks, ghost vertices are essential for cohomological consistency and ultraviolet finiteness, enabling precise vertex constructions.

Ghost vertices are interaction vertices involving ghost degrees of freedom introduced by gauge fixing or, in pure-spinor formalisms, vertex operators classified by ghost number. In non-Abelian gauge theory the central example is the Faddeev–Popov ghost–gluon three-point function, which enters lattice calculations, Schwinger–Dyson equations, functional renormalization-group truncations, MOM renormalization schemes, and Gribov-horizon analyses; in the 11D pure-spinor worldline formalism, by contrast, “ghost vertices” denotes operators such as $U^{(3)}$, $U^{(1)}$, and $U^{(0)}$ carrying distinct pure-spinor ghost numbers (Brito et al., 29 May 2025, Guillen et al., 27 Aug 2025).

1. Definition and kinematic structure

In covariantly gauge-fixed Yang–Mills theory, gauge fixing introduces a Faddeev–Popov determinant represented by Grassmann-valued ghost fields $c^a$ and $\bar c^a$. In Landau gauge,

$\partial_\mu A_\mu^a(x)=0,$

the ghost sector contains both the ghost propagator and the ghost–gluon interaction. In the supplied literature, the Landau-gauge ghost–gluon vertex is written in two equivalent tensor decompositions,

$$\Gamma_\mu^{abc}(k,k',q)= i g f^{abc}\big(k'_\mu H_1 + q_\mu H_2\big),$$

and

$$\Gamma_\mu^{abc}(k,p,q)= i g f^{abc}\big[q_\mu\,\alpha(k,p,q)+k_\mu\,\beta(k,p,q)\big],$$

with conventions differing only by momentum routing and basis choice (Brito et al., 29 May 2025, Wilson et al., 2012).

Landau-gauge transversality is the decisive simplification. Since the gluon propagator is transverse, components proportional to the gluon momentum are projected out. In the notation of $H_1,H_2$, this means that $H_2$ is not accessible in the lattice analysis of the Landau-gauge vertex; in the notation of $U^{(1)}$0 or $U^{(1)}$1, only the form factor multiplying the ghost or anti-ghost momentum contributes to the ghost Schwinger–Dyson equation (Brito et al., 29 May 2025, Aguilar, 2014). This is why much of the nonperturbative literature concentrates on a single scalar dressing function even though the full 1PI tensor decomposition is larger.

A recurrent structural constraint is Taylor’s theorem. In Landau gauge, the renormalization constant of the ghost–gluon vertex in the vanishing incoming ghost momentum configuration is unity, and the specific Taylor combination of form factors is fixed to its tree-level value. The supplied studies are explicit that this does not imply that the full vertex is bare for all momenta or in all kinematics; it constrains a particular kinematic limit and a particular linear combination of tensor structures (Aguilar, 2014).

2. Landau-gauge lattice determinations

A high-statistics lattice determination of the Landau-gauge ghost–gluon vertex in pure SU(3) Yang–Mills theory was carried out on $U^{(1)}$2 and $U^{(1)}$3 lattices at $U^{(1)}$4, corresponding to $U^{(1)}$5. For the ghost–gluon analysis, $U^{(1)}$6 configurations on $U^{(1)}$7 and $U^{(1)}$8 on $U^{(1)}$9 were used, and the calculation focused on the soft-gluon limit $U^{(0)}$0, where the vertex depends effectively on a single momentum scale (Brito et al., 29 May 2025).

The extraction of the scalar form factor was based on amputation and projection,

$U^{(0)}$1

with $U^{(0)}$2 taken either as the lattice tree-level vertex

$U^{(0)}$3

or as the continuum-like projector

$U^{(0)}$4

Using both projectors provided a direct test of tree-level lattice artifacts (Brito et al., 29 May 2025).

The principal numerical result is that the soft-gluon Landau-gauge ghost–gluon vertex remains very close to tree level over a wide momentum interval. In the infrared, $U^{(0)}$5 is close to its tree-level value and shows no strong enhancement or suppression. Up to about $U^{(0)}$6, the two lattice volumes and the two projector choices agree well, which was taken to indicate that finite-volume and discretization effects are under control in that range. Above about $U^{(0)}$7, lattice-spacing effects become more important and were not fully resolved in the single-$U^{(0)}$8 setup (Brito et al., 29 May 2025).

This lattice result reinforces an older pattern already present in previous SU(3) lattice studies: in Landau gauge, the ghost–gluon vertex deviates only mildly from tree level, especially in the infrared and intermediate-momentum regimes. A plausible implication is that nonperturbative Landau-gauge dynamics is not driven by a strongly singular basic ghost–gluon interaction, but by the coupled behavior of propagators and higher gluonic vertices.

3. Functional equations, RGZ, and infrared interpretation

Continuum Schwinger–Dyson studies place the ghost vertex at the center of the coupled ghost–gluon system. In Euclidean Landau gauge, the ghost propagator is written as

$U^{(0)}$9

and the ghost SDE depends explicitly on the ghost–gluon vertex. A central result of the coupled SDE analysis is that when the renormalization condition is imposed at a perturbative scale, $c^a$0, the ghost dressing is finite in the infrared; but obtaining a self-consistent coupled gluon–ghost solution requires nontrivial vertex input, particularly in the small-momentum regime of the gluon propagator. In these studies, a bare ghost–gluon vertex is inadequate once realistic symmetric three-gluon vertices are used, whereas Taylor-consistent nontrivial ghost–gluon ansätze supply the positive infrared ghost-loop contribution needed to support a masslike gluon propagator (Wilson et al., 2012, Pennington et al., 2011).

A complementary analysis derived approximate dynamical equations for the ghost–gluon form factor $c^a$1 in two special limits: the soft-gluon limit and the soft-ghost limit. In the one-loop dressed approximation, using fully dressed propagators and tree-level vertices in the relevant diagrams, the resulting $c^a$2 shows a substantial departure from tree level: in the soft-gluon configuration it develops a peak around $c^a$3, and in the soft-ghost configuration it peaks around $c^a$4. Solving the coupled system formed by the ghost dressing equation and the vertex equation in the soft-ghost limit yields excellent agreement with lattice ghost data without artificially enhancing the coupling constant (Aguilar, 2014).

Within the Refined Gribov–Zwanziger framework, the ghost–gluon vertex was first computed at one loop in the soft-gluon limit and then extended to general kinematics. The soft-gluon analysis found a finite infrared form factor that rises from its perturbative value, reaches a maximum around $c^a$5, and then returns toward perturbative behavior; the Taylor identities remain valid even in the presence of the Gribov horizon and condensates (Mintz et al., 2018). The later general-kinematics calculation in $c^a$6 for SU(2) and SU(3) introduced a toy model for the running coupling and found that RGZ results match fairly closely those from lattice simulations, Schwinger–Dyson equations, and the Curci–Ferrari model for symmetric, orthogonal, and soft-gluon configurations (Barrios et al., 2024).

Taken together, these results support a consistent infrared picture: decoupling or massive-type propagator behavior is compatible with a ghost–gluon vertex that is nontrivially dressed at intermediate scales but infrared regular. This suggests that the dominant nonperturbative role of ghosts in Landau gauge is mediated by a tightly constrained vertex rather than by a dramatic ghost–gluon singularity.

4. Temperature dependence and alternative gauges

At finite temperature, the soft chromomagnetic ghost–gluon vertex in SU(2) minimal Landau gauge was found to be remarkably inert. In the soft Matsubara sector, with purely spatial gluon legs, the dressing function $c^a$7 remains essentially tree level both below and above the deconfinement temperature $c^a$8. Within statistical accuracy, almost no temperature dependence was observed; moreover, data at $c^a$9, $\bar c^a$0, and the dimensionally reduced $\bar c^a$1 limit are essentially the same within relatively large statistical errors (Fister et al., 2014). In the supplied literature, this behavior is contrasted sharply with the three-gluon vertex, whose tree-level tensor dressing shows pronounced temperature dependence around the phase transition.

Coulomb gauge offers a structurally different but conceptually related setting. In the Hamiltonian approach, the ghost–gluon vertex is defined through the inverse Faddeev–Popov operator $\bar c^a$2 and, due to transversality of the spatial gluon, retains the same color and tensor structure as the bare vertex,

$\bar c^a$3

A one-loop solution based on nonperturbative input propagators found the vertex to be infrared finite but infrared enhanced compared to the bare one by $\bar c^a$4 to $\bar c^a$5, depending on the kinematical momentum regime (Campagnari et al., 2011). A later self-consistent canonical recursive DSE analysis, including back-coupling effects and additional one-loop diagrams, again found an infrared finite ghost–gluon vertex and an infrared diverging three-gluon vertex (Huber et al., 2014).

The contrast between the two vertices is significant. In both finite-temperature Landau gauge and Coulomb gauge Hamiltonian studies, the ghost–gluon vertex remains comparatively stable, while stronger infrared or thermal restructuring appears in the purely gluonic sector. This suggests a recurring division of labor in nonperturbative Yang–Mills theory: ghost vertices are essential, but often regular; the most dramatic infrared structures are displaced toward propagators and multi-gluon vertices.

5. Higher ghost-related vertices, perturbative schemes, and supersymmetric finiteness

Beyond the primitively divergent ghost–gluon vertex, two non-primitively divergent ghost-related four-point functions have been computed nonperturbatively in Landau gauge: the two-ghost–two-gluon vertex $\bar c^a$6 and the four-ghost vertex $\bar c^a$7. Their full transverse tensor structures were resolved, and a clear hierarchy was found with regard to the color structure, which reduces the number of relevant dressing functions. The impact of the two-ghost–two-gluon vertex on the three-gluon vertex is negligible, and only in the ghost–gluon vertex is a small net effect below $\bar c^a$8 seen. The four-ghost vertex is extremely small in general (Huber, 2017). This is an important correction to any expectation that higher ghost vertices might strongly back-react on lower Green functions.

At the perturbative level, the ghost–gluon vertex is also a renormalization-scheme anchor. A two-loop computation of the triple gluon, quark–gluon, and ghost–gluon vertices at the symmetric subtraction point in the $\bar c^a$9 scheme provided the basis for defining three MOM schemes, including MOMh from the ghost–gluon vertex. The conversion functions of all the wave functions, coupling constant, and gauge-parameter renormalization constants relative to $\partial_\mu A_\mu^a(x)=0,$0 were determined analytically, and these were then used to derive the three-loop anomalous dimensions of the Faddeev–Popov ghost and the MOM-scheme beta-functions in an arbitrary linear covariant gauge (Gracey, 2011).

In $\partial_\mu A_\mu^a(x)=0,$1 supersymmetric gauge theories regularized by higher covariant derivatives, the relevant ghost vertices are the triple gauge–ghost vertices with one leg of the quantum gauge superfield and two legs corresponding to the Faddeev–Popov ghost and antighost. The matter contribution to these vertices was shown to be UV finite at two loops in a general $\partial_\mu A_\mu^a(x)=0,$2-gauge, confirming a theorem that the triple gauge–ghost vertices are UV finite in all orders (Kuzmichev et al., 2021). A subsequent explicit two-loop verification established the full finiteness of the triple gauge–ghost vertices in a general renormalizable $\partial_\mu A_\mu^a(x)=0,$3 supersymmetric gauge theory, again for a general $\partial_\mu A_\mu^a(x)=0,$4-gauge; this was identified as an important step for the all-loop derivation of the exact NSVZ $\partial_\mu A_\mu^a(x)=0,$5-function (Kuzmichev et al., 2021).

6. Ghost-numbered vertices in the pure-spinor formalism

In the 11D pure-spinor worldline formalism, “ghost vertices” has a distinct technical meaning: vertex operators are classified by pure-spinor ghost number rather than by Faddeev–Popov field content. The minimal formalism contains a BRST charge $\partial_\mu A_\mu^a(x)=0,$6, a ghost number three vertex operator $\partial_\mu A_\mu^a(x)=0,$7 encoding linearized 11D supergravity states, and a ghost number one vertex operator $\partial_\mu A_\mu^a(x)=0,$8 arising from the linearized deformation of the BRST charge. A no-go theorem showed that a ghost number zero vertex operator compatible with 11D supergravity cannot be constructed in the minimal formalism (Guillen et al., 27 Aug 2025).

The obstruction is removed in the non-minimal formalism, which introduces $\partial_\mu A_\mu^a(x)=0,$9, $$\Gamma_\mu^{abc}(k,k',q)= i g f^{abc}\big(k'_\mu H_1 + q_\mu H_2\big),$$0, $$\Gamma_\mu^{abc}(k,k',q)= i g f^{abc}\big(k'_\mu H_1 + q_\mu H_2\big),$$1, and $$\Gamma_\mu^{abc}(k,k',q)= i g f^{abc}\big(k'_\mu H_1 + q_\mu H_2\big),$$2, and enlarges the BRST operator to

$$\Gamma_\mu^{abc}(k,k',q)= i g f^{abc}\big(k'_\mu H_1 + q_\mu H_2\big),$$3

In this setting, a ghost number zero vertex operator $$\Gamma_\mu^{abc}(k,k',q)= i g f^{abc}\big(k'_\mu H_1 + q_\mu H_2\big),$$4 was constructed for the first time. Its structure becomes compact when written in terms of “physical operators” built from the non-minimal variables, and it satisfies the expected descent relation with the ghost number one vertex operator. In addition, its commutator with the ghost number three single-particle vertex reproduces the two-particle superfield introduced previously in the literature (Guillen et al., 27 Aug 2025).

This usage of “ghost vertices” is formally different from the Yang–Mills ghost–gluon literature, but it preserves the same organizing principle: ghost number controls BRST cohomology, and vertex operators are the basic objects through which physical states and amplitudes are assembled.

Across these frameworks, ghost vertices are highly constrained objects. In gauge-fixed Yang–Mills theory, the basic ghost–gluon vertex is usually infrared finite, close to tree level in large kinematic regions, and strongly shaped by transversality and Taylor identities; higher ghost vertices are generally small, and their back-reaction on lower Green functions is limited. In supersymmetric and pure-spinor settings, the emphasis shifts from infrared dressing to cohomological consistency and ultraviolet finiteness, but the central role of ghost vertices as BRST-governed building blocks remains unchanged.