Whether quantum corrections cause the instanton gas to disappear in four-dimensional QCD — the hypothesis by which Witten (1979) proposed to resolve the discrepancy between instantons, the quark model, and the 1/N expansion — remains open. The tension is sharp: at fixed 't Hooft coupling a single instanton weighs e−cNe^{-cN}, while 1/N counting requires an O(1)O(1) topological susceptibility and mη′2∼1/Nm_{\eta'}^2 \sim 1/N for the flavor-singlet pseudoscalar. Witten demonstrated the proposed disappearance explicitly in the two-dimensional CPN−1\mathbb{C}P^{N-1} sigma model, whose large-N solution has a θ\theta-dependence that no dilute instanton gas can reproduce. Later work resolved the η′\eta'-mass strand through the Witten–Veneziano mechanism (verified on the lattice), established that the dilute gas fails in the confined phase (reorganized phenomenologically as an interacting instanton liquid; the dilute gas is valid only above the deconfinement temperature), and confirmed on the lattice that χt=O(N0)\chi_t = O(N^0) — deepening rather than dissolving the tension with the exponentially suppressed instanton weight. No controlled nonperturbative account of what carries the topological charge of the four-dimensional vacuum exists, and a 2025 refined-instanton analysis even challenges the two-dimensional disappearance argument itself. The surviving problem is a controlled determination of the microscopic topological sector of 4D Yang–Mills/QCD and of the fate — disappearance, refinement, or replacement — of the semiclassical instanton description.

Background

In four-dimensional SU(N)SU(N) Yang–Mills theory and in QCD, Euclidean gauge fields of finite action are classified by an integer winding number (topological charge) Q=(g2/32π2)∫d4x tr GμνG~μνQ=(g^2/32\pi^2)\int d^4x\,\mathrm{tr}\,G_{\mu\nu}\tilde G^{\mu\nu}, where GμνG_{\mu\nu} is the gauge-field strength and G~μν\tilde G^{\mu\nu} its dual. Instantons are the self-dual classical solutions carrying Q=±1Q=\pm 1 that saturate the action bound S≥8π2∣Q∣/g2S\ge 8\pi^2|Q|/g^2; they describe quantum tunneling between topologically distinct classical vacua, and the physical vacuum becomes a superposition (a θ\theta-vacuum) weighted by eiθQe^{i\theta Q}, where θ\theta is the coefficient of the CP-violating topological term in the action. 't Hooft computed the one-instanton tunneling amplitude, including the effective vertex that violates the axial U(1)AU(1)_A symmetry of light quarks (the chiral phase symmetry of the quark fields, whose anomalous quantum breaking is described below), and found an instanton size distribution behaving as ρ−5\rho^{-5} times a running-coupling factor — divergent at large size ρ\rho in a classically scale-invariant theory, which already signals that a naive instanton gas is not a controlled expansion in the infrared. Summing tunneling events as a dilute gas (the dilute instanton gas approximation, DIGA) gives a vacuum energy E(θ)−E(0)E(\theta)-E(0) proportional to the instanton density times (cos⁡θ−1)(\cos\theta-1).

Two countings then collide. In the 't Hooft large-NN limit (N→∞N\to\infty with λ=g2N\lambda=g^2N fixed), the instanton action is 8π2/g2=8π2N/λ8\pi^2/g^2 = 8\pi^2 N/\lambda: a single instanton weighs e−cNe^{-cN} with c=8π2/λ>0c=8\pi^2/\lambda>0, so instanton effects appear exponentially suppressed at large NN. But the 1/N1/N expansion of QCD — and the quark model with its annihilation diagrams (quark-line-disconnected processes, suppressed by 1/N1/N) — require the opposite. The U(1)AU(1)_A problem is the observation that the anomalous axial U(1)U(1) symmetry of the light-quark Lagrangian is not matched by a ninth light pseudoscalar boson; at large NN the resolution is that the flavor-singlet pseudoscalar η′\eta' acquires mη′2∼1/Nm_{\eta'}^2\sim 1/N from annihilation. Through the Witten–Veneziano relation mη′2+mη2−2mK2=2Nf χt/fπ2m_{\eta'}^2 + m_\eta^2 - 2m_K^2 = 2N_f\,\chi_t/f_\pi^2 (with fπf_\pi the pion decay constant in the 93 MeV93\,\mathrm{MeV} normalization and NfN_f the number of light quark flavors), this demands a topological susceptibility χt=∫d4x ⟨q(x)q(0)⟩\chi_t=\int d^4x\,\langle q(x)q(0)\rangle — the fluctuation of the topological charge density qq — of order N0N^0 in pure Yang–Mills, and a vacuum energy that is a nontrivial multibranched function of θ\theta with scaling variable θ/N\theta/N and a cusp at θ=π\theta=\pi.

Witten's 1979 paper 'Instantons, the Quark Model, and the 1/n Expansion' identified this three-way discrepancy and argued that its most attractive resolution is that quantum corrections cause the instanton gas to disappear in QCD: in the true confining vacuum the boundary conditions underlying the semiclassical instanton calculus — the gauge field approaching a pure gauge, Aμ→g−1∂μgA_\mu \to g^{-1}\partial_\mu g, at large distances — are spoiled by quantum fluctuations, and the dilute instanton gas with it. As a controlled demonstration he used the two-dimensional CPN−1\mathbb{C}P^{N-1} sigma model (the complex projective model: a two-dimensional, asymptotically free theory of an NN-component complex field with a local U(1)U(1) gauge invariance, integer-charge instantons, and a controlled 1/N1/N expansion; studied independently by D'Adda, Lüscher, and Di Vecchia). Its large-NN solution indeed has a multibranched θ\theta-dependence with χt=O(1/N)\chi_t=O(1/N) that no e−cNe^{-cN} dilute instanton gas can reproduce — the gas disappears in two dimensions.

What remains unresolved is the four-dimensional question this demonstration was meant to illuminate, in a theory where no controlled large-NN solution is available: does the same quantum disappearance of the instanton gas occur in QCD, and if not, what carries the O(1)O(1) topological susceptibility that the η′\eta' mass, the θ\theta-dependence, and the lattice value χt≈(180 MeV)4\chi_t \approx (180\,\mathrm{MeV})^4 all require?

Problem Statement

Determine, by controlled nonperturbative means (analytic or first-principles numerical), the fate of the semiclassical instanton description of the four-dimensional SU(N)SU(N) Yang–Mills/QCD vacuum in the confined phase — the question Witten (1979) left open when he proposed that quantum corrections cause the instanton gas to disappear in QCD, by analogy with the two-dimensional CPN−1\mathbb{C}P^{N-1} model. The objective comprises the following subgoals, which jointly serve it: (1) settle whether the dilute instanton gas — or any instanton-based semiclassical expansion — is a valid description of the QCD vacuum, or whether quantum fluctuations and confinement invalidate the boundary conditions (Aμ→g−1∂μgA_\mu \to g^{-1}\partial_\mu g at large distances) on which the instanton calculus rests, as happens in the two-dimensional model; (2) identify the field configurations that carry the topological charge of the four-dimensional vacuum and reproduce, consistently with the 1/N1/N expansion, the established anchors — χt=O(N0)\chi_t = O(N^0) with the lattice-compatible magnitude, the multibranched θ\theta-dependence with scaling variable θ/N\theta/N and cusp at θ=π\theta=\pi, and the η′\eta' mass with the 1/N1/N scaling expected from quark-model annihilation (mη′2∼1/Nm_{\eta'}^2 \sim 1/N) — thereby resolving the tension with the naive single-instanton weight e−cNe^{-cN} at fixed λ=g2N\lambda = g^2 N; (3) resolve the infrared pathologies of the instanton expansion — the divergence of the instanton size integral at large ρ\rho and the associated failures of volume extensivity in higher-order instanton computations — either by a mechanism, derived rather than postulated, that suppresses large instantons, or by establishing that instanton methods are not the correct semiclassical expansion of QCD; and (4) make precise, in the light of (1)-(3), under what conditions the two-dimensional mechanism generalizes to four dimensions (dimensionality, gauge-group structure, infrared behavior), including whether the two-dimensional disappearance conclusion itself survives the recently proposed refined-instanton treatments (Nguyen and Ünsal, JHEP 03 (2025) 162), which must be independently assessed rather than taken as settling this part. This is the controlled, microscopic version of the question the source paper posed heuristically; it is not narrowed to any fixed NN, coupling, geometry, or method.

References

  1. G. 't Hooft, 'Computation of the Quantum Effects Due to a Four-Dimensional Pseudoparticle', Phys. Rev. D 14 (1976) 3432. DOI: 10.1103/PhysRevD.14.3432, https://doi.org/10.1103/PhysRevD.14.3432
  2. A. D'Adda, M. Lüscher, P. Di Vecchia, 'A 1/n Expandable Series of Nonlinear Sigma Models with Instantons', Nucl. Phys. B 146 (1978) 63-76. DOI: 10.1016/0550-3213(78)90432-7, https://doi.org/10.1016/0550-3213(78)90432-7
  3. E. Witten, 'Instantons, the Quark Model, and the 1/n Expansion', Nucl. Phys. B 149 (1979) 285-320. DOI: 10.1016/0550-3213(79)90243-8, https://doi.org/10.1016/0550-3213(79)90243-8
  4. E. Witten, 'Current Algebra Theorems for the U(1) Goldstone Boson', Nucl. Phys. B 156 (1979) 269-283. DOI: 10.1016/0550-3213(79)90031-2, https://doi.org/10.1016/0550-3213(79)90031-2
  5. G. Veneziano, 'U(1) Without Instantons', Nucl. Phys. B 159 (1979) 213-224. DOI: 10.1016/0550-3213(79)90332-8, https://doi.org/10.1016/0550-3213(79)90332-8
  6. T. Schäfer, E. V. Shuryak, 'Instantons in QCD', Rev. Mod. Phys. 70 (1998) 323-426. DOI: 10.1103/RevModPhys.70.323, arXiv:hep-ph/9610451, https://arxiv.org/abs/hep-ph/9610451
  7. E. Witten, 'Theta Dependence in the Large N Limit of Four-Dimensional Gauge Theories', Phys. Rev. Lett. 81 (1998) 2862-2865. DOI: 10.1103/PhysRevLett.81.2862, arXiv:hep-th/9807109, https://arxiv.org/abs/hep-th/9807109
  8. R. G. Edwards, U. M. Heller, 'Are topological charge fluctuations in QCD instanton dominated?', Phys. Rev. D 65 (2002) 014505. DOI: 10.1103/PhysRevD.65.014505, arXiv:hep-lat/0105004, https://arxiv.org/abs/hep-lat/0105004
  9. T. Schäfer, 'Instantons in QCD with many colors', Phys. Rev. D 66 (2002) 076009. DOI: 10.1103/PhysRevD.66.076009, arXiv:hep-ph/0206062, https://arxiv.org/abs/hep-ph/0206062
  10. B. Lucini, M. Teper, U. Wenger, 'Topology of SU(N) gauge theories at T ≃ 0 and T ≃ Tc', Nucl. Phys. B 715 (2005) 461-482. DOI: 10.1016/j.nuclphysb.2005.02.037, arXiv:hep-lat/0401028, https://arxiv.org/abs/hep-lat/0401028
  11. E. Vicari, H. Panagopoulos, 'Theta dependence of SU(N) gauge theories in the presence of a topological term', Phys. Rep. 470 (2009) 93-150. DOI: 10.1016/j.physrep.2008.10.001, arXiv:0803.1593, https://arxiv.org/abs/0803.1593
  12. C. Bonati, M. D'Elia, H. Panagopoulos, E. Vicari, 'Change of theta Dependence in 4D SU(N) Gauge Theories Across the Deconfinement Transition', Phys. Rev. Lett. 110 (2013) 252003. DOI: 10.1103/PhysRevLett.110.252003, arXiv:1301.7640, https://arxiv.org/abs/1301.7640
  13. R. D. Pisarski, F. Rennecke, 'Multi-instanton contributions to anomalous quark interactions', Phys. Rev. D 101 (2020) 114019. DOI: 10.1103/PhysRevD.101.114019, arXiv:1910.14052, https://arxiv.org/abs/1910.14052
  14. R. D. Pisarski, F. Rennecke, 'Conjectures about the Chiral Phase Transition in QCD from Anomalous Multi-Instanton Interactions', Phys. Rev. Lett. 132 (2024) 251903. DOI: 10.1103/PhysRevLett.132.251903, arXiv:2401.06130, https://arxiv.org/abs/2401.06130
  15. M. Nguyen, M. Ünsal, 'Refined instanton analysis of the 2D CP^{N-1} model: mass gap, theta dependence, and mirror symmetry', JHEP 03 (2025) 162. DOI: 10.1007/JHEP03(2025)162, arXiv:2309.12178, https://arxiv.org/abs/2309.12178