For the uniform self-avoiding walk of length n on the integer lattice , the closing probability that the walk's endpoint neighbours its starting point is predicted to decay as with in , via the scaling relation between the polygon exponent , the walk exponent , and the closing exponent. Hammond's snake-method paper (Trans. Amer. Math. Soc. 372, 2019, reference 1) proved the general-dimension upper bound and recorded that the existence and values of these exponents remain open in low dimensions. The problem of rigorously establishing the existence and value of the closing exponent in and is open.
Background
A self-avoiding walk (SAW) of length on the integer lattice is a path taking nearest-neighbour steps that never revisits a vertex; it is the standard lattice model of a linear polymer. Writing for the number of such walks starting at the origin, the limit , called the connective constant, exists by a submultiplicativity argument of Hammersley, and Kesten (1963) established the corresponding ratio-limit behaviour. A self-avoiding polygon (SAP) of length is a self-avoiding walk of length whose endpoint neighbours the start, closed by one further step; counts such polygons up to translation. If is chosen uniformly among length- walks, the closing probability — the probability that the endpoint is a nearest neighbour of the origin — satisfies the exact identity
so closing-probability decay and polygon/walk enumeration are two views of one question.
Physics predicts power laws: for even and , whence with the closing exponent . In , Nienhuis's Coulomb-gas prediction (1982) — obtained by mapping the planar model onto an exactly solvable two-dimensional Coulomb gas — gives and , so is predicted; the same values follow from conformal-field-theory heuristics. Rigorous knowledge is sharply dimension-dependent: for , above the upper critical dimension (the dimension above which self-avoidance becomes a weak perturbation and Gaussian mean-field behaviour holds), the lace expansion (a perturbative expansion exploiting the weak effect of self-avoidance) yields Gaussian scaling and the exponents (Hara–Slade, 1992), and on the hexagonal lattice the connective constant is known exactly (Duminil-Copin–Smirnov, 2012); but in no critical exponent for the self-avoiding walk has ever been proved to exist.
The best rigorous closing-probability bounds in low dimension come from the snake method, a probabilistic technique introduced by Duminil-Copin, Glazman, Hammond and Manolescu (2016). It decomposes a closing walk into two parts at the walk's lexicographically maximal vertex and exploits Kesten's pattern theorem — which guarantees that any fixed finite walk pattern appears linearly often in all but exponentially few long self-avoiding walks — together with the Gaussian fluctuation of pattern counts on the scale . That work gave in general dimension and proved that the walk is sub-ballistic, meaning its typical displacement grows slower than linearly in . Hammond's paper (Trans. Amer. Math. Soc. 372, 2019, reference 1) reworked the method to reach , for any , for every and all large odd ; its companion papers strengthened this to along a positive-density set of odd in (via polygon joining) and to along even of full density in . That paper notes that the method could not exceed the barrier, that the stronger decay rates are confined to subsequences or to the plane, and that the predicted value lies far beyond all current bounds.
Problem Statement
Determine, rigorously, the power-law decay of the closing probability of the uniform self-avoiding walk on the nearest-neighbour lattice in the low dimensions and .
- (Existence) Prove, or disprove, that there exist exponents and such that (for even ) and , and hence — via the exact identity — that the closing probability satisfies with the closing exponent .
- (Value) If the exponents exist, determine their values; in particular, prove or refute the prediction (equivalently and ).
A rigorous proof that the exponent fails to exist in or equally resolves the question by refuting the predicted form. Any resolution must hold along the full sequence of lengths (for , all odd — lattice parity forces for even on ), not merely along a subsequence.
References
- Alan Hammond, On self-avoiding polygons and walks: The snake method via pattern fluctuation, Transactions of the American Mathematical Society 372 (2019) 2335–2356, DOI 10.1090/tran/7494 (arXiv:1808.09597), https://doi.org/10.1090/tran/7494
- Alan Hammond, On self-avoiding polygons and walks: the snake method via polygon joining, Electronic Journal of Probability 24 (2019), Paper No. 49, 43 pp., DOI 10.1214/18-EJP249 (arXiv:1808.10500), https://doi.org/10.1214/18-EJP249
- Alan Hammond, An upper bound on the number of self-avoiding polygons via joining, The Annals of Probability 46 (2018) 175–206, DOI 10.1214/17-AOP1182 (arXiv:1808.09032), https://doi.org/10.1214/17-AOP1182
- Hugo Duminil-Copin, Alexander Glazman, Alan Hammond, Ioan Manolescu, On the probability that self-avoiding walk ends at a given point, The Annals of Probability 44 (2016) 955–983, DOI 10.1214/14-AOP993 (arXiv:1305.1257), https://doi.org/10.1214/14-AOP993
- Hugo Duminil-Copin, Shirshendu Ganguly, Alan Hammond, Ioan Manolescu, Bounding the number of self-avoiding walks: Hammersley–Welsh with polygon insertion, The Annals of Probability 48 (2020) 1644–1692, DOI 10.1214/19-AOP1400 (arXiv:1809.00760), https://doi.org/10.1214/19-AOP1400
- Hugo Duminil-Copin, Alan Hammond, Self-avoiding walk is sub-ballistic, Communications in Mathematical Physics 324 (2013) 401–423, DOI 10.1007/s00220-013-1811-1, https://doi.org/10.1007/s00220-013-1811-1
- Anthony J. Guttmann, Iwan Jensen, On the existence of critical exponents for self-avoiding walks, Journal of Physics A: Mathematical and Theoretical 55 (2022) 415206, DOI 10.1088/1751-8121/ac943a (arXiv:2207.08433), https://doi.org/10.1088/1751-8121/ac943a
- Dmitrii Krachun, Christoforos Panagiotis, Quantitative sub-ballisticity of self-avoiding walk on the hexagonal lattice, The Annals of Probability 54 (2026) 1109–1125, DOI 10.1214/24-AOP1730 (arXiv:2310.17299), https://doi.org/10.1214/24-AOP1730
- Alan Hammond, Tyler Helmuth, Self-attracting self-avoiding walk, Probability Theory and Related Fields 175 (2019) 677–719, DOI 10.1007/s00440-018-00898-7, https://doi.org/10.1007/s00440-018-00898-7
- Takashi Hara, Gordon Slade, Self-avoiding walk in five or more dimensions I. The critical behaviour, Communications in Mathematical Physics 147 (1992) 101–136, DOI 10.1007/bf02099530, https://doi.org/10.1007/bf02099530
- Bernard Nienhuis, Exact critical point and critical exponents of O(n) models in two dimensions, Physical Review Letters 49 (1982) 1062–1065, DOI 10.1103/PhysRevLett.49.1062, https://doi.org/10.1103/PhysRevLett.49.1062
- Harry Kesten, On the number of self-avoiding walks, Journal of Mathematical Physics 4 (1963) 960–969, DOI 10.1063/1.1704022, https://doi.org/10.1063/1.1704022
- Hugo Duminil-Copin, Stanislav Smirnov, The connective constant of the honeycomb lattice equals sqrt(2+sqrt 2), Annals of Mathematics 175 (2012) 1653–1665, DOI 10.4007/annals.2012.175.3.14, https://doi.org/10.4007/annals.2012.175.3.14
- Florian Lehner, Christian Lindorfer, Christoforos Panagiotis, Self-avoiding walk is ballistic on graphs with more than one end, Forum of Mathematics, Sigma 13 (2025) e179, DOI 10.1017/fms.2025.10131 (arXiv:2403.13121), https://doi.org/10.1017/fms.2025.10131