Gori and Viti (JHEP 12 (2018) 131) conjectured an exact conformal-field-theory expression for the universal ratio R(η) of four-point boundary connectivities of Fortuin-Kasteleyn (FK) clusters in the critical two-dimensional Q-state Potts model, for integer Q = 1, 2, 3, 4, by solving a third-order null-vector differential equation for the boundary field φ_{1,3}; at Q = 1 (critical percolation, central charge c = 0) the answer develops a logarithmic singularity and is expressed through generalized hypergeometric functions 3F2. Their computation leaves open a coherent set of problems: (i) a rigorous derivation of the null-vector decoupling and of the identification of the selected ODE solution with the lattice connectivity ratio in the non-unitary, logarithmic c = 0 regime, beyond the self-consistency checks; (ii) a detailed study of the finite-size corrections to R(η) at Q = 4, where power-law and logarithmic corrections coexist and are difficult to disentangle; and (iii) a systematic theory of the connection formulas and analytic continuation of the 3F2 representations appearing at Q = 1. Later literature (2019-2024) has determined the bulk four-point connectivities at generic Q numerically and semi-analytically, has confirmed the logarithmic operator-product-expansion structure of the boundary φ_{1,3} four-point function by probabilistic methods, and has determined the c = 0 identity-module structure; however, the rigorous identification, the Q = 4 finite-size theory, and the systematic 3F2 connection theory all remain open.

Background

The Q-state Potts model can be represented as a random-cluster model of Fortuin-Kasteleyn (FK) clusters: bonds of a lattice are occupied with probability p, and each connected cluster of occupied bonds carries a common factor Q in the Boltzmann weight, so that the model is defined for any real Q ≥ 0 (Q = 1 is bond percolation, Q = 2 the Ising model). A central class of observables are cluster connectivities, i.e. the probabilities that a set of marked points is partitioned among FK clusters in a specified way. For n points on the boundary of a simply connected planar domain, only the non-crossing partitions give independent connectivities; for n = 4 there are exactly three, denoted P(12)(34), P(14)(23) and P(1234), while P(13)(24) = 0 by planarity. Gori and Viti studied the dimensionless, conformally invariant ratio

R(η)=P(14)(23)P(14)(23)+P(12)(34)+P(1234),R(\eta)=\frac{P_{(14)(23)}}{P_{(14)(23)}+P_{(12)(34)}+P_{(1234)}},

where η is the anharmonic (cross) ratio of the four boundary points, which at criticality is expected to be a universal function of η alone.

At criticality the scaling limit is described by a two-dimensional conformal field theory (CFT) with central charge c = 1 − 6/(p(p+1)) and Q = 4cos²(π/(p+1)) (p = 2 percolation, p = 3 Ising, p = 5 three-state Potts, p → ∞ four-state Potts). In boundary CFT, changing the boundary condition at a boundary point is implemented by a boundary-condition-changing operator, which Gori and Viti identify with the Kac-table field φ_{1,3} (the Kac table is the grid of primary fields of a minimal model, labelled by two integer indices r, s) of chiral, i.e. holomorphic, scaling dimension h = (p−1)/(p+1) (h = 1/3 at Q = 1; h = 1/2 at Q = 2, where it coincides with the boundary scaling dimension of the Ising order parameter). A null vector is a null (vanishing) descendant in a Virasoro highest-weight representation; its decoupling implies that correlation functions of the corresponding primary satisfy a finite-order differential equation. For φ_{1,3} the null vector at level three yields a third-order Fuchsian ordinary differential equation (an equation whose singular points are all regular) for the four-point function, with regular singular points at η = 0, 1, ∞ and indicial roots (the local power-law exponents of the solutions at each singular point) equal to the Kac dimensions h_{1,1} = 0, h_{1,3} = h and h_{1,5} = 3h+1. The solutions of this equation are the chiral conformal blocks, and the ratio R is a linear combination of them whose coefficients Gori and Viti fixed through lattice duality arguments and operator-product-expansion selection rules, obtaining closed conjectures for R(η) at integer Q, verified by high-precision Monte Carlo simulations.

At Q = 1 the central charge vanishes and the theory is non-unitary and logarithmic: the dilatation operator L0 is non-diagonalizable, fields appear in Jordan cells, and correlation functions can contain logarithmic singularities. Such logarithmic CFTs are characterized by an indecomposability parameter b (a universal number entering the regularized chiral operator product expansion, conjectured by Gurarie and Ludwig to be b = −5/8 for boundary percolation and re-derived algebraically by Dubail, Jacobsen and Saleur). Gori and Viti's Q = 1 result provides a fully explicit logarithmic observable: the identity conformal block is singular at c = 0 and must be regularized, and the resulting symmetric solution is a logarithmic combination of functions expressible through generalized hypergeometric functions 3F2 of the special argument 27(1−η)²η²/(4(1−η+η²)³), the several pieces being stitched together by hand through continuity conditions at η = 1/2. For the classical 2F1 equation a complete theory of Kummer solutions (the 24 canonical local solutions related by the Kummer transformations) and connection formulas exists; for 3F2 the corresponding theory is much less developed (partial results go back to Thomae, Bühring and, more recently, Mimachi's twisted-homology determination of connection matrices).

At Q = 4 the Potts critical point is marginal: the continuum theory contains marginally irrelevant operators (perturbations whose coupling vanishes only logarithmically along the renormalization-group flow) that generate logarithmic corrections to scaling, known classically from the renormalization-group analysis of Cardy, Nauenberg and Scalapino and from finite-size-scaling studies of Hamer, Batchelor and Barber. In Gori and Viti's Monte Carlo data the distance between the measured ratio and the conformal-field-theory prediction at Q = 4 can be fitted almost equally well by a pure power law and by a logarithm, and the boundary operator φ_{1,3} itself becomes marginal exactly at Q = 4, so that both correction types are expected to coexist.

For orientation on rigor: the two-point version of boundary connectivity, Cardy's crossing formula, has been rigorously proved for site percolation on the triangular lattice by Smirnov, and the form-factor approach of Delfino and Viti (exact matrix elements of fields between asymptotic particle states) determines bulk two- and three-point connectivities; the four-point boundary problem studied here is the first to require the full null-vector machinery inside a non-unitary logarithmic CFT, which is why its foundations and its Q = 4 endpoint remain open.

Problem Statement

The common objective is to place the conformal-field-theory determination of the four-point boundary FK-connectivity ratio

R(η)=P(14)(23)P(14)(23)+P(12)(34)+P(1234)R(\eta)=\frac{P_{(14)(23)}}{P_{(14)(23)}+P_{(12)(34)}+P_{(1234)}}

of the critical two-dimensional Q-state Potts model on a complete and rigorous footing, completing the program initiated by Gori and Viti (JHEP 12 (2018) 131). Within this single objective, the interrelated open problems left by that computation are:

(i) Rigorous foundations of the differential-equation method in the logarithmic regime. Prove, or rigorously derive from the lattice random-cluster model, that null-vector decoupling holds for the boundary field φ_{1,3} of dimension h = (p−1)/(p+1) in the non-unitary, logarithmic CFTs at central charge c = 1 − 6/(p(p+1)) - in particular at c = 0 (Q = 1, where the identity conformal block is singular and must be regularized) - so that the boundary four-point function of φ_{1,3} satisfies the third-order Fuchsian null-vector equation with indicial roots (the local power-law exponents at the singular points η = 0, 1, ∞) 0, h and 3h+1; and prove that the specific linear combination of its solutions selected in Gori and Viti's construction indeed equals the lattice connectivity ratio R(η) in the scaling limit. This is the analogue, for the four-point FK-connectivity problem, of what Smirnov's theorem proves for Cardy's formula, and it goes beyond the self-consistency checks, Monte Carlo agreement at Q = 1, 2, 3, and algebraic analyses on which the conjecture currently rests.

(ii) Finite-size corrections at Q = 4. Develop a detailed study of the finite-size corrections to R(η) at the marginal endpoint Q = 4 of the critical regime, where the data of Gori and Viti are compatible both with a pure power law and with a logarithmic scaling and the boundary operator φ_{1,3} is marginal; the study should parametrize and disentangle the simultaneous power-law and logarithmic terms (with their exponents and η-dependence) so as to settle the asymptotic convergence to the conformal-field-theory prediction.

(iii) Hypergeometric connection theory at Q = 1. Develop a systematic theory of the connection formulas and analytic-continuation properties of the 3F2 representations in which the Q = 1 conformal blocks are expressed - the analogue for 3F2 of the classical theory of Kummer solutions and connection coefficients for 2F1 - sufficient to relate the various representations of the blocks (the Frobenius power-series solutions at the singular points, the closed 3F2 forms of argument 27(1−η)²η²/(4(1−η+η²)³), and the logarithmic solutions) without ad hoc stitching at η = 1/2, to fix branch-cut choices, and to handle the degenerate rational-parameter cases that arise; this machinery underlies the closed-form manipulation of the percolation results in (i).

The three parts (i)-(iii) are jointly required for completion of this single objective: resolving it requires the rigorous identification of (i), the Q = 4 finite-size theory of (ii), and the systematic 3F2 connection theory of (iii) together. Each part is an independently attackable open problem, and a result that settles a proper subset of them is a partial advance on the objective that leaves the remaining parts open.

References

  1. G. Gori and J. Viti, Four-point boundary connectivities in critical two-dimensional percolation from conformal invariance, JHEP 12 (2018) 131, DOI 10.1007/JHEP12(2018)131, https://doi.org/10.1007/JHEP12(2018)131 (arXiv:1806.02330).
  2. G. Gori and J. Viti, Exact logarithmic four-point functions in the critical two-dimensional Ising model, Physical Review Letters 119 (2017) 191601, DOI 10.1103/PhysRevLett.119.191601, https://doi.org/10.1103/PhysRevLett.119.191601 (arXiv:1704.02893).
  3. J. L. Cardy, Critical percolation in finite geometries, Journal of Physics A: Mathematical and General 25 (1992) L201, DOI 10.1088/0305-4470/25/4/009, https://doi.org/10.1088/0305-4470/25/4/009.
  4. S. Smirnov, Critical percolation in the plane: conformal invariance, Cardy's formula, scaling limits, Comptes Rendus de l'Academie des Sciences - Series I - Mathematics 333 (2001) 239, DOI 10.1016/S0764-4442(01)01991-7, https://doi.org/10.1016/S0764-4442(01)01991-7 (arXiv:0909.4499).
  5. V. Gurarie and A. W. W. Ludwig, Conformal algebras of two-dimensional disordered systems, Journal of Physics A: Mathematical and General 35 (2002) L377, DOI 10.1088/0305-4470/35/27/101, https://doi.org/10.1088/0305-4470/35/27/101 (arXiv:cond-mat/9911392).
  6. J. Dubail, J. L. Jacobsen and H. Saleur, Conformal field theory at central charge c = 0: A measure of the indecomposability (b) parameters, Nuclear Physics B 834 (2010) 399, DOI 10.1016/j.nuclphysb.2010.02.016, https://doi.org/10.1016/j.nuclphysb.2010.02.016 (arXiv:1001.1151).
  7. G. Delfino and J. Viti, Potts q-color field theory and scaling random cluster model, Nuclear Physics B 852 (2011) 149, DOI 10.1016/j.nuclphysb.2011.06.012, https://doi.org/10.1016/j.nuclphysb.2011.06.012 (arXiv:1104.4323).
  8. J. L. Cardy, M. Nauenberg and D. J. Scalapino, Scaling theory of the Potts-model multicritical point, Physical Review B 22 (1980) 2560, DOI 10.1103/PhysRevB.22.2560, https://doi.org/10.1103/PhysRevB.22.2560.
  9. C. J. Hamer, M. T. Batchelor and M. N. Barber, Logarithmic corrections to finite-size scaling in the four-state Potts model, Journal of Statistical Physics 52 (1988) 679, DOI 10.1007/BF01019724, https://doi.org/10.1007/BF01019724.
  10. M. Picco, S. Ribault and R. Santachiara, On four-point connectivities in the critical 2d Potts model, SciPost Physics 7 (2019) 044, DOI 10.21468/SciPostPhys.7.4.044, https://doi.org/10.21468/SciPostPhys.7.4.044 (arXiv:1906.02566).
  11. Y. He, J. L. Jacobsen and H. Saleur, Geometrical four-point functions in the two-dimensional critical Q-state Potts model: The interchiral conformal bootstrap, JHEP 12 (2020) 019, DOI 10.1007/JHEP12(2020)019, https://doi.org/10.1007/JHEP12(2020)019 (arXiv:2005.07258).
  12. L. Grans-Samuelsson, L. Liu, Y. He, J. L. Jacobsen and H. Saleur, The action of the Virasoro algebra in the two-dimensional Potts and loop models at generic Q, JHEP 10 (2020) 109, DOI 10.1007/JHEP10(2020)109, https://doi.org/10.1007/JHEP10(2020)109 (arXiv:2007.11539).
  13. R. Nivesvivat and S. Ribault, Logarithmic CFT at generic central charge: from Liouville theory to the Q-state Potts model, SciPost Physics 10 (2021) 021, DOI 10.21468/SciPostPhys.10.1.021, https://doi.org/10.21468/SciPostPhys.10.1.021 (arXiv:2007.04190).
  14. Y. He and H. Saleur, A note on the identity module in c = 0 CFTs, SciPost Physics 12 (2022) 100, DOI 10.21468/SciPostPhys.12.3.100, https://doi.org/10.21468/SciPostPhys.12.3.100 (arXiv:2109.05050).
  15. F. Camia and Y. Feng, Logarithmic correlation functions in 2D critical percolation, JHEP 08 (2024) 103, DOI 10.1007/JHEP08(2024)103, https://doi.org/10.1007/JHEP08(2024)103 (arXiv:2403.18576).
  16. F. Camia and Y. Feng, Conformally covariant probabilities, operator product expansions, and logarithmic correlations in two-dimensional critical percolation, arXiv:2407.04246, https://arxiv.org/abs/2407.04246.
  17. G. Delfino, Particles, conformal invariance and criticality in pure and disordered systems, European Physical Journal B 94 (2021) 65, DOI 10.1140/epjb/s10051-021-00076-0, https://doi.org/10.1140/epjb/s10051-021-00076-0 (arXiv:2010.12275).
  18. K. Fukushima and K. Sakai, Numerical study on a crossing probability for the four-state Potts model: Logarithmic correction to the finite-size scaling, Progress of Theoretical and Experimental Physics 2019 (2019) 091A01, DOI 10.1093/ptep/ptz101, https://doi.org/10.1093/ptep/ptz101.
  19. N. Javerzat, M. Picco and R. Santachiara, Two-point connectivity of two-dimensional critical Q-Potts random clusters on the torus, Journal of Statistical Mechanics: Theory and Experiment 2020 (2020) 033101, DOI 10.1088/1742-5468/ab6331, https://doi.org/10.1088/1742-5468/ab6331 (arXiv:1907.11041).
  20. W. Bühring, An analytic continuation formula for the generalized hypergeometric function, SIAM Journal on Mathematical Analysis 19 (1988) 1249, DOI 10.1137/0519088, https://doi.org/10.1137/0519088.
  21. K. Mimachi, Connection matrices associated with the generalized hypergeometric function 3F2, Funkcialaj Ekvacioj 51 (2008) 107, DOI 10.1619/fesi.51.107, https://doi.org/10.1619/fesi.51.107.