Rigorous establishment of the gluon-condensate route to confinement in Yang-Mills theory

A 2026 study by Schierholz (Eur. Phys. J. C 86, 646) proposes a renormalization-group approach to color confinement built on the gradient flow: a nonvanishing, finite, renormalized gluon condensate $G$ is identified as a sufficient condition from which the infrared behavior $\alpha_S(\mu)\simeq\Lambda_S^2/\mu^2$, a positive mass gap, and a linearly rising static potential (Wilson confinement) follow. The approach assumes but does not prove $G>0$; its lattice support is not continuum-extrapolated in the deepest infrared; and the microscopic structures generating the condensate (instantons, monopoles, center vortices) are unidentified in a model-independent way. The problem is to complete this program on a first-principles footing: prove (or refute) the existence and positivity of the renormalized gluon condensate in continuum four-dimensional Yang-Mills theory, which by the framework's sufficiency argument places the confinement argument on a rigorous footing. This route has no accepted resolution as of August 2026.

Background

Quantum chromodynamics (QCD), the theory of the strong interaction, is a non-Abelian Yang-Mills gauge theory: its fundamental fields are the gluons, gauge bosons carrying the color charge of the SU(3) gauge group. A hallmark of such theories is asymptotic freedom — the effective coupling $\alpha_S(\mu)$ measured at momentum scale $\mu$ decreases logarithmically as $\mu$ grows, so short-distance physics is accessible to perturbation theory. At large distances the coupling grows and perturbation theory fails. The expected physical consequence is confinement: isolated colored particles (quarks and gluons) are never observed, only color-neutral bound states. A standard diagnostic is Wilson's criterion — the potential $V(r)$ between a static quark-antiquark pair separated by distance $r$ rises linearly, $V(r)\simeq\sigma r$, with string tension $\sigma>0$, so pulling the charges apart costs arbitrarily large energy. A related property is the mass gap: a strictly positive lower bound $m_\sigma$ on the energies of excitations above the vacuum. A mass gap is necessary but not sufficient for confinement (a theory of massive free gluons would have a gap but no confinement).

The gluon condensate, introduced by Shifman, Vainshtein and Zakharov in 1979 in their QCD sum-rule program, is the vacuum expectation value $G=(\alpha_S/\pi)\langle F^a_{\mu\nu}F^a_{\mu\nu}\rangle$ (with $F^a_{\mu\nu}$ the gluon field-strength tensor and the sum over color indices $a$ implied); it measures the fluctuation strength of the gauge-field energy density in the vacuum and anomalously breaks the scale invariance of the classical action. An early physics argument for a nonzero condensate is Savvidy's 1977 observation that the perturbative Yang-Mills vacuum is unstable against the formation of a constant chromomagnetic field, suggesting the true vacuum has nonzero field strength.

The technical tool underlying the 2026 framework is the gradient flow introduced by Lüscher in 2010: a diffusion equation that evolves the gauge field towards smoother configurations, with flow time $t$ playing the role of a resolution scale. Crucially, expectation values of local gauge-invariant expressions in the flowed field at $t>0$ are finite, renormalized quantities; the flow thus realizes, infinitesimally, the coarse-graining step of a momentum-space renormalization group (RG) transformation — a systematic change of the scale $\mu=(8t)^{-1/2}$ at which the theory is viewed. Building on earlier work in which the energy density of the flowed field defines a running coupling $\alpha_{GF}(\mu)$, Schierholz (Eur. Phys. J. C 86 (2026) 646) argues that:

• the renormalized gluon condensate $G$ is invariant under the RG flow (scale invariant), reflecting self-similar behavior across scales;
• the existence of a nonvanishing, finite $G$ is a sufficient condition for confinement;
• from it follows the infrared (long-distance) behavior $\alpha_S(\mu)\simeq\Lambda_S^2/\mu^2$ as $\mu\to 0$ — infrared slavery, the counterpart of asymptotic freedom — driving the flow to the infrared fixed point $1/\alpha_S=0$;
• the static potential then rises linearly with $\sigma=(2/3)\Lambda_V^2$ ($\Lambda_V$ the scale parameter of the potential scheme), satisfying Wilson's criterion; and
• the lowest excitation above the vacuum obeys a mass-gap relation, $m_\sigma^2=(\pi^3/576)\,|\langle 0|(\alpha_{GF}/\pi)G^a_{\mu\nu}G^a_{\mu\nu}|\sigma\rangle|^2/\Lambda_{GF}^6$, so that $m_\sigma>0$ (here $G^a_{\mu\nu}$ is the field strength of the flowed gauge field, whose vacuum expectation defines the condensate in this framework, $\alpha_{GF}$ and $\Lambda_{GF}$ are the coupling and scale parameter of the gradient-flow scheme, and $|\sigma\rangle$ is the lowest-lying excited state).

The analytical claims are supported by lattice simulations in the pure SU(3) gauge theory: at flow times up to $t\simeq 0.2L^2$ (with $L$ the lattice spatial extent) the flowed energy density behaves as $1/t$, the coupling grows linearly with flow time, and restricting the path integral to fixed topological charge $Q$ reproduces the expected instanton content. The relevant microscopic structures known to populate the vacuum at intermediate scales are instantons (localized classical solutions of the Yang-Mills field equations carrying topological charge), monopoles (Abelian magnetic excitations, relevant to the dual-superconductor picture of confinement), and center vortices (magnetic flux tubes linked to the center of the gauge group, another confinement mechanism candidate). The 2026 study estimates that instantons account for roughly 60% of the condensate magnitude at its scales, but by universality — the expectation that many microscopic configurations flow to the same long-distance behavior — it leaves the dominant microscopic origin model-differentiating and unresolved.

What the framework does not deliver, and states as its own limitations, is threefold. First, the existence of $G>0$ is argued from physical indications (scale-anomaly phenomenology, vacuum instability, lattice data) but not proven from first principles in the continuum theory; without such a proof the confinement argument is not rigorous. The mass-gap component of this gap in rigor contains, as a special case, the celebrated Clay Mathematics Institute Millennium Prize problem of Yang-Mills existence and mass gap, whose official statement explicitly notes that no proof of the mass-gap property is known. Second, the numerical evidence rests on simulations at a single lattice spacing ($a=0.082$ fm) in a fixed volume, with identified artifacts (lattice discretization errors of order $a^2$, dislocation-induced topological charge loss), so the deep-infrared results are not yet established as continuum statements; and the extension to dynamical (sea) quarks requires a so-far-undeveloped procedure of retuning the bare quark masses and coupling as the flow time increases so that the physics stays fixed. Third, the microscopic mechanism generating the condensate remains unidentified. A further, phenomenologically adjacent question the source flags as unsolved is how experimental determinations of the effective coupling at small momentum transfer — where perturbative operator-product-expansion techniques fail because leading-twist Wilson coefficients mix with higher-twist matrix elements (twist characterizing the collinear scaling of operators) — could ever confront the predicted $1/\mu^2$ infrared behavior.

Problem Statement

Complete the gluon-condensate renormalization-group approach to confinement on a first-principles footing for continuum four-dimensional SU(3) Yang-Mills theory. The single overarching objective is a rigorous account of confinement through the scale-invariant renormalized gluon condensate:

Prove, from first principles in the continuum theory — not from the lattice evidence — that the renormalized gluon condensate $G$ exists and is nonvanishing and finite, i.e. that $G>0$; or, alternatively, prove that $G$ vanishes or is ill-defined, refuting the approach. In the framework's sufficiency argument these confinement properties follow as corollaries of $G>0$ rather than standing as separate objectives: the infrared behavior $\alpha_S(\mu)\simeq\Lambda_S^2/\mu^2$ with the fixed point $1/\alpha_S=0$, the positivity of the lowest excitation mass $m_\sigma$ through the mass-gap relation, and the linearly rising static potential satisfying Wilson's confinement criterion. This question contains, as a special case along this specific route, the mass-gap question of the Clay Millennium problem; the problem stated here is the derived condensate-route formulation, not the Millennium problem itself, and a Millennium-problem solution by any route would settle only the mass-gap component.

In the source framework, a nonvanishing, finite condensate is by itself the sufficient condition from which these confinement properties follow, and the microscopic properties of the vacuum are largely averaged out by universality; the confinement argument therefore rests on this analytic property alone. The identification of the microscopic vacuum structures generating the condensate, and the continuum-limit verification of the lattice evidence — including the extension to dynamical quarks, whose bare-parameter retuning procedure the source defers to a separate publication — are recorded here as related open questions; they are not folded into the problem statement.

The question is resolved by establishing, from first principles in the continuum theory, the existence and positivity of the renormalized condensate — or its failure; the confinement properties above then follow from the framework's sufficiency argument.

References

1. Gerrit Schierholz, "Renormalization group approach to confinement", Eur. Phys. J. C 86 (2026) 646, DOI: 10.1140/epjc/s10052-026-15857-z, arXiv:2509.10658, https://arxiv.org/abs/2509.10658
2. Clay Mathematics Institute, "Yang-Mills & the Mass Gap" (Millennium Prize Problem, listed unsolved, accessed 2026-08-26), https://www.claymath.org/millennium/yang-mills-the-maths-gap/
3. Martin Lüscher, "Properties and uses of the Wilson flow in lattice QCD", JHEP 08 (2010) 071, DOI: 10.1007/JHEP08(2010)071, arXiv:1006.4518, https://arxiv.org/abs/1006.4518
4. Hiroshi Suzuki, "Energy-momentum tensor from the Yang-Mills gradient flow", Prog. Theor. Exp. Phys. 2013 (2013) 083B03, DOI: 10.1093/ptep/ptt059, arXiv:1304.0533, https://arxiv.org/abs/1304.0533
5. Gerrit Schierholz, "QCD lambda parameter from gradient flow", Eur. Phys. J. C 85 (2025) 1007, DOI: 10.1140/epjc/s10052-025-14743-4, arXiv:2410.17677, https://arxiv.org/abs/2410.17677
6. Yoshifumi Nakamura and Gerrit Schierholz, "The strong CP problem solved by itself due to long-distance vacuum effects", Nucl. Phys. B 986 (2023) 116063, DOI: 10.1016/j.nuclphysb.2022.116063, arXiv:2106.11369, https://arxiv.org/abs/2106.11369
7. G. K. Savvidy, "Infrared instability of the vacuum state of gauge theories and asymptotic freedom", Phys. Lett. B 71 (1977) 133, DOI: 10.1016/0370-2693(77)90759-6, https://doi.org/10.1016/0370-2693(77)90759-6
8. M. A. Shifman, A. I. Vainshtein and V. I. Zakharov, "QCD and resonance physics. Theoretical foundations", Nucl. Phys. B 147 (1979) 385, DOI: 10.1016/0550-3213(79)90022-1, https://doi.org/10.1016/0550-3213(79)90022-1
9. Guy F. de Téramond, Arpon Paul, Stanley J. Brodsky, Alexandre Deur, Hans Günter Dosch, Tianbo Liu, Raza Sabbir Sufian, "QCD Running Coupling in the Nonperturbative and Near-Perturbative Regimes", Phys. Rev. Lett. 133 (2024) 181901, DOI: 10.1103/PhysRevLett.133.181901, arXiv:2403.16126, https://arxiv.org/abs/2403.16126
10. Martin Beneke and Hiromasa Takaura, "Gradient-flowed operator product expansion without IR renormalons", JHEP 03 (2026) 033, DOI: 10.1007/JHEP03(2026)033, arXiv:2510.12193, https://arxiv.org/abs/2510.12193
11. Stephan Dürr and Gianluca Fuwa, "Topological susceptibility and excess kurtosis in SU(3) Yang-Mills theory", Phys. Rev. D 113 (2026) 054508, DOI: 10.1103/dw3m-vcdm, arXiv:2501.08217, https://arxiv.org/abs/2501.08217
12. James M. Zanotti, Kadir Utku Can, Roger Horsley, Yoshifumi Nakamura, Holger Perlt, P. E. L. Rakow, Gerrit Schierholz, Hinnerk Stueben, R. D. Young, "Advances in lattice hadron physics calculations using the gradient flow", PoS LATTICE2021 (2022) 055, DOI: 10.22323/1.396.0055, arXiv:2112.06355, https://arxiv.org/abs/2112.06355
13. Hidenori Sonoda and Hiroshi Suzuki, "A basis of the gradient flow exact renormalization group for gauge theory", Prog. Theor. Exp. Phys. 2025 (2025) 093B05, DOI: 10.1093/ptep/ptaf115, arXiv:2506.23567, https://arxiv.org/abs/2506.23567
14. Hiroshi Shiba and Tsuneo Suzuki, "Monopole action and condensation in SU(2) QCD", Phys. Lett. B 351 (1995) 519, DOI: 10.1016/0370-2693(95)00422-H, arXiv:hep-lat/9408004, https://arxiv.org/abs/hep-lat/9408004
15. M. Engelhardt and H. Reinhardt, "Center vortex model for the infrared sector of Yang-Mills theory: Confinement and deconfinement", Nucl. Phys. B 585 (2000) 591, DOI: 10.1016/S0550-3213(00)00445-4, arXiv:hep-lat/9912003, https://arxiv.org/abs/hep-lat/9912003
16. Sourav Chatterjee, "Yang-Mills for probabilists", arXiv:1803.01950, https://arxiv.org/abs/1803.01950
17. Mir Faizal and Arshid Shabir, "Reflection-Positive Construction of a Four-Dimensional SU(N) Yang-Mills Theory With Mass Gap and Confinement", Fortschr. Phys. 74 (2026) e70097, DOI: 10.1002/prop.70097, arXiv:2606.19362, https://arxiv.org/abs/2606.19362
18. V. P. Nair, "Towards a Proof of Mass Gap in 3d Yang-Mills Theory", arXiv:2608.10133, https://arxiv.org/abs/2608.10133