Exact four-spin correlation functions, cluster-connectivity conformal blocks, and the spin-operator algebra of two-dimensional critical percolation

In the percolation limit ($q\to 1$, central charge $c=0$) of the two-dimensional critical $q$-state Potts model, Dotsenko computed the four-spin correlation function from the Coulomb-gas screening-integral representation - available at the percolation point only because the screening numbers happen to be integers there - and found that the fusion of two spin fields (conformal weight $5/96$) produces, besides the identity, a single new channel due to a field of weight $5/8$. He left open the analytic continuation of four-point Coulomb-gas integrals away from such special points (so the four-spin function along the Potts line $1\le q\le 4$ was not established), the complete mapping between lattice four-spin functions and cluster connectivities and the conformal blocks of the percolation conformal field theory (including his conjecture that the $5/8$ channel appears as a subleading channel in the symmetric connectivities), the exactness of the regularisation used, and the full operator algebra generated by the spin operators. Later work advanced every front: Dotsenko's own general-$q$ construction under an unproven decomposition assumption; numerical and interchiral bootstrap determinations of four-point connectivities; logarithmic conformal field theory at generic central charge; conjectured exact structure constants for critical loop models; and rigorous lattice analyses that confirmed the $5/8$ energy channel, accompanied by a logarithmic partner. What remains open is the exact, assumption-free determination of the four-spin function for general $q$ and its $q\to 1$ limit, the complete conformal-block identification of the lattice four-point functions and connectivities with exact coefficients, and the associated operator algebra of the percolation conformal field theory.

Background

The $q$-state Potts model assigns a spin $\sigma_i\in\{1,\dots,q\}$ to each site of a two-dimensional lattice; in the Fortuin-Kasteleyn (FK) representation its partition function and correlations are rewritten in terms of random clusters of bonded sites, which for $q=1$ become the clusters of ordinary bond percolation. For $0<q\le 4$ the model is critical, and its large-scale behaviour is described by a two-dimensional conformal field theory (CFT), that is, a continuum theory invariant under local angle-preserving transformations; the symmetry algebra is the Virasoro algebra, whose central charge $c$ runs from $1$ at $q=4$ down to $c=0$ at $q\to 1$. At $q=1$ the spins themselves are meaningless (there is nothing to permute) and the fundamental observables are instead cluster connectivities - probabilities that specified points belong to the same FK cluster. The spin (magnetization) operator nevertheless survives as an analytic continuation: at percolation it coincides with the cluster-density field $\psi$, a primary field of conformal weights $h=\bar h=5/96$ (total scaling dimension $5/48$) whose two-point function is the renormalized two-point connectivity.

In a CFT, correlation functions are built from primary fields, each carrying holomorphic and antiholomorphic conformal weights $h,\bar h$; four-point functions decompose into conformal blocks, one per allowed intermediate primary in the operator product of two external fields, and the set of allowed intermediates defines the fusion rules. The rational minimal models are the CFTs with a finite spectrum: for integer $p\ge 2$ they have $c=1-6/(p(p+1))$ and primaries $\Phi_{r,s}$ on a grid (the Kac table). A primary is called degenerate when a linear combination of its descendant fields vanishes (a null vector); the null vector then implies a differential equation that fixes all correlators of that field. The Potts spin operator is $\Phi_{(p+1)/2,(p-1)/2}$, which is degenerate only at special values ($p=3$ Ising, $p=5$ three-state Potts, and higher minimal models). At percolation $p=2$ and the spin indices are half-integers, $\Phi_{3/2,1/2}$ or $\Phi_{3/2,3/2}$: the field is non-degenerate and no null-vector differential equation determines its correlators.

The Coulomb-gas method represents such correlators as expectation values of free-boson vertex operators $V_\alpha$ (with conformal weight fixed by the charge $\alpha$) supplemented by screening charges whose positions are integrated over; the neutrality condition - total charge balance - fixes the numbers $l,k$ of the two screening types. This integral representation can be evaluated in closed form only when $l,k$ are non-negative integers, which for the four-spin function happens only at special points of the Potts line. At the percolation point the two admissible choices of the spin charge give $l=2,k=0$ (two conformal blocks) and $l=2,k=2$ (five blocks); this coincidence is what made Dotsenko's computation possible, and it fails away from percolation. Three-point functions had been settled earlier by field-theoretic and Coulomb-gas methods, including the analytic continuation of three-point screening integrals to non-integer screening numbers; the analogous continuation for four-point integrals is not known.

Four-point functions matter because they encode the actual fusion rules. Dotsenko's result is that the percolation four-spin block decomposes into the identity channel plus a single further channel due to a field of conformal weight $h=\bar h=5/8$ (total scaling dimension $5/4$ - the energy-like field of percolation), with a spurious $h=-1/24$ channel decoupling; the five blocks of the second charge choice must compensate each other after an $\epsilon$-regularisation in which the screening charges are shifted, $\alpha_+\mapsto\alpha_++\epsilon$, and external charges are split. A second, independent subtlety is the lattice-to-continuum dictionary: the lattice spin operator is in general a linear combination of leading and subleading conformal primaries, so lattice four-point functions and the four cluster connectivities defined by Delfino and Viti are linear combinations of several conformal blocks, and identifying which blocks appear with which coefficients is a separate problem that symmetry alone does not settle.

Finally, $c=0$ theories are necessarily non-unitary and, when nontrivial, logarithmic: their correlation functions can contain logarithmic singularities and their symmetry algebra acts indecomposably, so that certain fields appear in Jordan-block pairs with the same conformal weight, one member having a vanishing two-point function. Vasseur, Jacobsen and Saleur showed, from the $Q\to 1$ limit of the Potts model, that the percolation energy operator indeed mixes with such a logarithmic partner; this structure turns out to be essential to the four-spin problem. The current landscape combines a numerical interchiral-conformal-bootstrap program for the four-point connectivity functions of the $Q$-state Potts model, a logarithmic-CFT reconstruction at generic central charge, conjectured exact structure constants for critical loop models, and rigorous probabilistic results on the conformal covariance of percolation connection probabilities; the exact four-spin sector and its lattice matching, however, are still missing.

Problem Statement

The problem is the exact conformal-field-theory determination of four-point functions in the percolation limit of the two-dimensional critical $q$-state Potts model, together with their matching to lattice observables. In the notation of the background, and retaining the generality of the source:

1. (Analytic continuation / general-$q$ four-spin function) Define and compute the four-spin correlation function $\langle\sigma(\infty)\sigma(1)\sigma(z,\bar z)\sigma(0)\rangle$ of the critical Potts model for general $q$ in $1\le q\le 4$ - that is, provide the analytic continuation of the Coulomb-gas four-point integral representation away from the special points where the screening numbers are integers, or an equivalent first-principles construction. The continuation must be model-preserving: it has to fix the finite coefficients that the $\epsilon$-regularisation leaves undetermined, and any assumption replacing the missing continuation (such as the uniqueness of the decomposition of $\sigma\times\sigma$ into degenerate fields $\Phi_{n',n}$) must be justified, not postulated.

2. (Lattice matching) Determine the complete conformal-block decomposition of the lattice four-spin function and of the four-point cluster-connectivity functions of percolation in the sense of Delfino and Viti: which channels appear - in particular the status of the $h=5/8$ channel that Dotsenko conjectured to appear as a subleading channel in the symmetric connectivities, now known to be accompanied by a logarithmic partner - and with which exact coefficients. The identification must account for the mixing of lattice operators into leading, subleading and logarithmic-partner conformal fields.

3. (Operator algebra) Construct the operator algebra of the percolation conformal field theory generated by the spin (density) operators: the full spectrum, fusion rules and structure constants of the leading and subleading spin-like fields and their logarithmic partners, to the level needed to derive multipoint spin and connectivity functions beyond four points.

These three components support one overarching objective - an exact, lattice-matched conformal description of four-point functions in critical percolation. Component 1 supplies the exact continuum functions, component 2 fixes their relation to what the lattice actually measures, and component 3 is the algebraic structure that this matching requires; an answer that omits any one of them leaves the objective incomplete. Full resolution of the problem as stated requires all three.

References

1. V. S. Dotsenko, Correlation function of four spins in the percolation model, Nuclear Physics B 911 (2016) 712-743. DOI: 10.1016/j.nuclphysb.2016.08.032 (arXiv:1606.09162). https://arxiv.org/abs/1606.09162
2. V. S. Dotsenko, Four spins correlation function of the q states Potts model, for general values of q. Its percolation model limit q -> 1, Nuclear Physics B 953 (2020) 114973. DOI: 10.1016/j.nuclphysb.2020.114973 (arXiv:1911.06682). https://arxiv.org/abs/1911.06682
3. G. Delfino and J. Viti, Potts q-color field theory and scaling random cluster model, Nuclear Physics B 852 (2011) 149-173. DOI: 10.1016/j.nuclphysb.2011.06.012. https://doi.org/10.1016/j.nuclphysb.2011.06.012
4. R. Vasseur, J. L. Jacobsen and H. Saleur, Logarithmic observables in critical percolation, Journal of Statistical Mechanics (2012) L07001. DOI: 10.1088/1742-5468/2012/07/L07001 (arXiv:1206.2312). https://arxiv.org/abs/1206.2312
5. M. Picco, S. Ribault and R. Santachiara, A conformal bootstrap approach to critical percolation in two dimensions, SciPost Physics 1 (2016) 009. DOI: 10.21468/SciPostPhys.1.1.009 (arXiv:1607.07224). https://arxiv.org/abs/1607.07224
6. 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 (arXiv:1906.02566). https://arxiv.org/abs/1906.02566
7. J. L. Jacobsen and H. Saleur, Bootstrap approach to geometrical four-point functions in the two-dimensional critical Q-state Potts model: a study of the s-channel spectra, Journal of High Energy Physics 2019 (2019) 084. DOI: 10.1007/JHEP01(2019)084 (arXiv:1809.02191). https://arxiv.org/abs/1809.02191
8. Y. He, L. Grans-Samuelsson, J. L. Jacobsen and H. Saleur, Geometrical four-point functions in the two-dimensional critical Q-state Potts model: connections with the RSOS models, Journal of High Energy Physics 2020 (2020) 156. DOI: 10.1007/JHEP05(2020)156 (arXiv:2002.09071). https://arxiv.org/abs/2002.09071
9. 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, Journal of High Energy Physics 2020 (2020) 019. DOI: 10.1007/JHEP12(2020)019 (arXiv:2005.07258). https://arxiv.org/abs/2005.07258
10. 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 (arXiv:2007.04190). https://arxiv.org/abs/2007.04190
11. R. Nivesvivat, S. Ribault and J. L. Jacobsen, Critical loop models are exactly solvable, SciPost Physics 17 (2024) 029. DOI: 10.21468/SciPostPhys.17.2.029 (arXiv:2311.17558). https://arxiv.org/abs/2311.17558
12. F. Camia, Conformal covariance of connection probabilities and fields in 2D critical percolation, Communications on Pure and Applied Mathematics 77 (2024) 2138-2176. DOI: 10.1002/cpa.22171 (arXiv:2203.08167). https://arxiv.org/abs/2203.08167
13. F. Camia and Y. Feng, Logarithmic correlation functions in 2D critical percolation, Journal of High Energy Physics 2024 (2024) 103. DOI: 10.1007/JHEP08(2024)103 (arXiv:2403.18576). https://arxiv.org/abs/2403.18576