Nature of the low-temperature phase and the lower critical dimension of the three-dimensional Edwards-Anderson Ising spin glass

A 1997 comparison study of the three-dimensional (3D) Edwards-Anderson Ising spin glass and the two-dimensional ferromagnetic XY model (Iñiguez, Marinari, Parisi and Ruiz-Lorenzo) left two questions open — the shape of the overlap distribution $P(q)$ at lattice sizes well beyond $L=16$, and whether the Edwards-Anderson order parameter $q_{\mathrm{EA}}$ vanishes in the thermodynamic limit, as a competing power-law fit to the peak position $q_M$ allowed — and placed them in a coupled context: it contrasted a replica-symmetry-breaking (RSB) description of the low-temperature phase with the droplet picture through the expected behavior of $P(0)$, and it quoted the lower critical dimension $d_l$ only as bracketed, $2 \le d_l < 3$, with an estimate $d_l \approx 2.5$, noting that the system lies very close to $d_l$ while stressing that the influence of that proximity is highly non-trivial. Later work has substantially settled both questions at the level of numerical-physics evidence: equilibrium simulations now reach $L = 32$–$40$, and a continuous finite-temperature transition with nonzero $q_{\mathrm{EA}}$ is established, excluding the Kosterlitz–Thouless-like trivial alternative. The surviving open core, actively contested as of 2026, is the identity of the theory of the 3D spin-glass phase — RSB versus droplet scaling, which subsumes the asymptotic structure of $P(q)$ such as the limiting value of $P(0)$ — and the exact determination of $d_l$, for which current estimates cluster at 2.43–2.50 with a candidate exact value of $5/2$.

Background

A spin glass is a magnet whose interactions are simultaneously random and frustrated, so that not all spin pairs can be satisfied at once; the canonical lattice model is the Edwards-Anderson (EA) Ising spin glass, with Ising spins on a cubic lattice of linear size $L$ and quenched (frozen during the thermal equilibration of the spins) nearest-neighbor couplings drawn independently at random. The central observable is the overlap $q$ between two independent equilibrium copies (replicas) of a system sharing the same couplings, and its probability distribution $P(q)$ plays the role of an order-parameter distribution. The Edwards-Anderson order parameter $q_{\mathrm{EA}}$ is the typical self-overlap of an equilibrium state, i.e. the edge of the support of the limiting $P(q)$. A standard transition diagnostic is the Binder cumulant, a dimensionless ratio of moments of $P(q)$ whose curves for successive lattice sizes merge below a Kosterlitz–Thouless (KT)-type transition but cross at a genuine one; the KT transition of the two-dimensional (2D) XY ferromagnet is an infinite-order transition below which the magnetization nevertheless vanishes in the thermodynamic limit.

The source of the present problem is a 1997 study by Iñiguez, Marinari, Parisi and Ruiz-Lorenzo that compared the finite-size order-parameter distributions of the 3D EA spin glass with the magnetization distributions of the 2D ferromagnetic XY model. The comparison was a warning: at accessible sizes the XY magnetization distribution looks essentially identical to the spin-glass $P(q)$ — double-peaked at finite $L$ with an apparently nonzero weight at the origin — even though its thermodynamic limit is trivial. At the time, the largest equilibrated 3D spin-glass lattices had $L = 16$; the peak position $q_M$ of $P(q)$ at $T = 0.7$ extrapolated to $q_{\mathrm{EA}} \approx 0.70$, consistent with an independent off-equilibrium determination by Marinari, Parisi, Ruiz-Lorenzo and Ritort, yet a pure power-law decay $q_M \propto L^{-0.12}$, implying $q_{\mathrm{EA}} = 0$ and a KT-like trivial limit, fit the same data equally well. Evidence for a transition rested on Binder-cumulant crossings at a critical temperature $T_c \approx 1.1$ in units of the coupling (Kawashima and Young 1996), and the lower critical dimension $d_l$ — the dimension below which no finite-temperature spin-glass phase exists — was bracketed only as $2 \le d_l < 3$, with early estimates $d_l \approx 2.5$ from the anomalous dimension $\eta$ (the critical exponent governing the power-law decay of the two-point correlation function at criticality) and from mean-field interface arguments (Franz, Parisi and Virasoro 1994).

Two competing theories describe what the ordered phase should be. In the replica symmetry breaking (RSB) scenario descending from Parisi's mean-field solution, the phase contains many pure thermodynamic states organized ultrametrically (their pairwise distances obey the ultrametric inequality, i.e. the states form a nested hierarchy), and the thermodynamic $P(q)$ is nontrivial, with nonzero weight at $q = 0$. In the droplet picture of Fisher and Huse there is only a single pair of pure states related by the global spin flip, and $P(q)$ collapses onto the two points $\pm q_{\mathrm{EA}}$; large-scale excitations are droplets whose free-energy cost grows as $L^{\theta}$ with the stiffness exponent $\theta$, and $d_l$ is where $\theta$ changes sign. The asymptotic overlap distribution, the stiffness and interface scalings, and $d_l$ are therefore coupled facets of one question: which theory describes the finite-dimensional spin-glass phase.

Problem Statement

Determine the thermodynamic-limit structure of the low-temperature phase of the three-dimensional Edwards-Anderson Ising spin glass in zero magnetic field — for either of the two standard nearest-neighbor coupling distributions, bimodal $\pm J$ or Gaussian, which are generally expected to yield the same phase structure in three dimensions — as posed by the 1997 source study of Iñiguez, Marinari, Parisi and Ruiz-Lorenzo and refined by the subsequent evidence, together with the associated critical-dimension question:

1. Asymptotic overlap distribution. Establish, with controlled finite-size extrapolation from equilibrium data below $T_c$, the $L \to \infty$ form of $P(q)$: does it remain nontrivial, with a nonzero limiting central weight $P(0)$ and broad internal structure, or does it collapse to the trivial pair of delta functions at $\pm q_{\mathrm{EA}}$? (The 1997 data questions — measuring $P(q)$ at sizes well beyond $L = 16$, and excluding a vanishing $q_{\mathrm{EA}}$ — have been met by equilibrium simulations up to $L \approx 40$ and by the established continuous transition at $T_c > 0$; what remains open is the asymptotic structure itself.)

2. Theoretical scenario. Decide whether the low-temperature phase is described by the replica-symmetry-breaking scenario (many pure thermodynamic states, ultrametric organization, marginal stability) or by the droplet/scaling scenario of Fisher and Huse (a single pair of pure states, stiffness-controlled excitations) — or establish a third possibility with comparable rigor.

3. Lower critical dimension. Determine $d_l$ for the Ising spin glass — exactly, if it admits a closed value such as the candidate $5/2$, or otherwise to a precision that discriminates the candidate values (current estimates span roughly 2.43 to 2.50).

These three targets are coupled facets of one scientific objective, the identity of the theory of the finite-dimensional spin-glass phase: the asymptotic $P(q)$ and the stiffness/interface scalings are the observables through which the scenario is decided, and $d_l$ is fixed by the same stiffness exponent through $\theta(d_l) = 0$.

References

1. D. Iñiguez, E. Marinari, G. Parisi, J. J. Ruiz-Lorenzo, '3D spin glass and 2D ferromagnetic XY model: a comparison', J. Phys. A: Math. Gen. 30, 7337-7347 (1997). DOI: 10.1088/0305-4470/30/21/010, https://doi.org/10.1088/0305-4470/30/21/010 (preprint: arXiv:cond-mat/9707050, https://arxiv.org/abs/cond-mat/9707050)
2. N. Kawashima, A. P. Young, 'Phase transition in the three-dimensional ±J Ising spin glass', Phys. Rev. B 53, R484-R487 (1996). DOI: 10.1103/PhysRevB.53.R484, https://doi.org/10.1103/PhysRevB.53.R484
3. E. Marinari, G. Parisi, J. J. Ruiz-Lorenzo, F. Ritort, 'Numerical evidence for spontaneously broken replica symmetry in 3D spin glasses', Phys. Rev. Lett. 76, 843-846 (1996). DOI: 10.1103/PhysRevLett.76.843, https://doi.org/10.1103/PhysRevLett.76.843
4. M. Palassini, S. Caracciolo, 'Universal finite-size scaling functions in the 3D Ising spin glass', Phys. Rev. Lett. 82, 5128-5131 (1999). DOI: 10.1103/PhysRevLett.82.5128, https://doi.org/10.1103/PhysRevLett.82.5128 (preprint: arXiv:cond-mat/9904246, https://arxiv.org/abs/cond-mat/9904246)
5. The Janus Collaboration: R. Alvarez Baños, A. Cruz, L. A. Fernández, J. M. Gil-Narvión, A. Gordillo-Guerrero, M. Guidetti, D. Iñiguez, A. Maiorano, F. Mantovani, E. Marinari, V. Martín-Mayor, J. Monforte-García, A. Muñoz Sudupe, D. Navarro, G. Parisi, S. Perez-Gaviro, F. Ricci-Tersenghi, J. J. Ruiz-Lorenzo, S. F. Schifano, B. Seoane, A. Tarancón, R. Tripiccione, D. Yllanes, 'Sample-to-sample fluctuations of the overlap distributions in the three-dimensional Edwards-Anderson spin glass', Phys. Rev. B 84, 174209 (2011). DOI: 10.1103/PhysRevB.84.174209, https://doi.org/10.1103/PhysRevB.84.174209 (preprint: arXiv:1107.5772, https://arxiv.org/abs/1107.5772)
6. The Janus Collaboration: M. Baity-Jesi, E. Calore, A. Cruz, L. A. Fernandez, J. M. Gil-Narvión, A. Gordillo-Guerrero, D. Iñiguez, A. Maiorano, E. Marinari, V. Martin-Mayor, J. Monforte-García, A. Muñoz Sudupe, D. Navarro, G. Parisi, S. Perez-Gaviro, F. Ricci-Tersenghi, J. J. Ruiz-Lorenzo, S. F. Schifano, B. Seoane, A. Tarancón, R. Tripiccione, D. Yllanes, 'A statics-dynamics equivalence through the fluctuation-dissipation ratio provides a window into the spin-glass phase from nonequilibrium measurements', Proc. Natl. Acad. Sci. USA 114, 1838-1843 (2017). DOI: 10.1073/pnas.1621242114, https://doi.org/10.1073/pnas.1621242114 (preprint: arXiv:1610.01418, https://arxiv.org/abs/1610.01418)
7. P. Contucci, C. Giardinà, C. Giberti, G. Parisi, C. Vernia, 'Ultrametricity in the Edwards-Anderson model', Phys. Rev. Lett. 99, 057206 (2007). DOI: 10.1103/PhysRevLett.99.057206, https://doi.org/10.1103/PhysRevLett.99.057206 (preprint: arXiv:cond-mat/0607376, https://arxiv.org/abs/cond-mat/0607376)
8. T. Jorg, F. Krzakala, 'Comment on Ultrametricity in the Edwards-Anderson Model', Phys. Rev. Lett. 100, 159701 (2008). DOI: 10.1103/PhysRevLett.100.159701, https://doi.org/10.1103/PhysRevLett.100.159701 (preprint: arXiv:0709.0894, https://arxiv.org/abs/0709.0894)
9. W. Wang, J. Machta, H. G. Katzgraber, 'Evidence against a mean-field description of short-range spin glasses revealed through thermal boundary conditions', Phys. Rev. B 90, 184412 (2014). DOI: 10.1103/PhysRevB.90.184412, https://doi.org/10.1103/PhysRevB.90.184412
10. W. Wang, J. Machta, H. Munoz-Bauza, H. G. Katzgraber, 'Number of thermodynamic states in the three-dimensional Edwards-Anderson spin glass', Phys. Rev. B 96, 184417 (2017). DOI: 10.1103/PhysRevB.96.184417, https://doi.org/10.1103/PhysRevB.96.184417 (preprint: arXiv:1710.08004, https://arxiv.org/abs/1710.08004)
11. B. Yucesoy, H. G. Katzgraber, J. Machta, 'Evidence of non-mean-field-like low-temperature behavior in the Edwards-Anderson spin-glass model', Phys. Rev. Lett. 109, 177204 (2012). DOI: 10.1103/PhysRevLett.109.177204, https://doi.org/10.1103/PhysRevLett.109.177204 (preprint: arXiv:1206.0783, https://arxiv.org/abs/1206.0783)
12. A. Billoire, L. A. Fernandez, A. Maiorano, E. Marinari, V. Martin-Mayor, G. Parisi, F. Ricci-Tersenghi, J. J. Ruiz-Lorenzo, D. Yllanes, 'Comment on Evidence of Non-Mean-Field-Like Low-Temperature Behavior in the Edwards-Anderson Spin-Glass Model', Phys. Rev. Lett. 110, 219701 (2013). DOI: 10.1103/PhysRevLett.110.219701, https://doi.org/10.1103/PhysRevLett.110.219701 (preprint: arXiv:1211.0843, https://arxiv.org/abs/1211.0843)
13. W. Wang, M. Wallin, J. Lidmar, 'Evidence of many thermodynamic states of the three-dimensional Ising spin glass', Phys. Rev. Research 2, 043241 (2020). DOI: 10.1103/PhysRevResearch.2.043241, https://doi.org/10.1103/PhysRevResearch.2.043241 (preprint: arXiv:2005.05805, https://arxiv.org/abs/2005.05805)
14. M. A. Moore, 'The droplet-scaling versus replica symmetry breaking debate in spin glasses revisited', Phys. Rev. E 103, 062111 (2021). DOI: 10.1103/PhysRevE.103.062111, https://doi.org/10.1103/PhysRevE.103.062111 (preprint: arXiv:2103.02973, https://arxiv.org/abs/2103.02973)
15. S. Franz, G. Parisi, M. A. Virasoro, 'Interfaces and lower critical dimension in a spin glass model', J. Phys. I France 4, 1657-1667 (1994). DOI: 10.1051/jp1:1994213, https://doi.org/10.1051/jp1:1994213
16. C. Amoruso, E. Marinari, O. C. Martin, A. Pagnani, 'Scalings of domain wall energies in two dimensional Ising spin glasses', Phys. Rev. Lett. 91, 087201 (2003). DOI: 10.1103/PhysRevLett.91.087201, https://doi.org/10.1103/PhysRevLett.91.087201 (preprint: arXiv:cond-mat/0305042, https://arxiv.org/abs/cond-mat/0305042)
17. S. Boettcher, 'Stiffness of the Edwards-Anderson model in all dimensions', Phys. Rev. Lett. 95, 197205 (2005). DOI: 10.1103/PhysRevLett.95.197205, https://doi.org/10.1103/PhysRevLett.95.197205
18. A. Maiorano, G. Parisi, 'Support for the value 5/2 for the spin glass lower critical dimension at zero magnetic field', Proc. Natl. Acad. Sci. USA 115, 5129-5134 (2018). DOI: 10.1073/pnas.1720832115, https://doi.org/10.1073/pnas.1720832115 (preprint: arXiv:1711.05590, https://arxiv.org/abs/1711.05590)
19. M. Aguilar-Janita, V. Martin-Mayor, J. Moreno-Gordo, J. J. Ruiz-Lorenzo, 'Evidence of de Almeida-Thouless line below six dimensions', J. Stat. Mech. 2025, 113301 (2025). DOI: 10.1088/1742-5468/ae1a4a, https://doi.org/10.1088/1742-5468/ae1a4a (preprint: arXiv:2510.14446, https://arxiv.org/abs/2510.14446)
20. M. Bernaschi, L. A. Fernández, I. González-Adalid Pemartín, V. Martín-Mayor, G. Parisi, F. Ricci-Tersenghi, 'Low energy excitations in a long prism geometry: testing the lower critical dimension of the Ising spin glass', arXiv:2601.07926 (2026), https://arxiv.org/abs/2601.07926
21. G. Delfino, 'On the nature of the spin glass transition', J. Stat. Mech. 2026, 043201 (2026). DOI: 10.1088/1742-5468/ae5a20, https://doi.org/10.1088/1742-5468/ae5a20 (preprint: arXiv:2511.20163, https://arxiv.org/abs/2511.20163)
22. C. Chilin, E. Marinari, V. Martín-Mayor, G. Parisi, J. J. Ruiz-Lorenzo, D. Yllanes, 'On the true low-energy excitations of the three-dimensional spin glass', arXiv:2606.07197 (2026), https://arxiv.org/abs/2606.07197
23. D. S. Fisher, D. A. Huse, 'Ordered phase of short-range Ising spin-glasses', Phys. Rev. Lett. 56, 1601-1604 (1986). DOI: 10.1103/PhysRevLett.56.1601, https://doi.org/10.1103/PhysRevLett.56.1601
24. G. Parisi, 'A sequence of approximated solutions to the S-K model for spin glasses', J. Phys. A: Math. Gen. 13, L115-L121 (1980). DOI: 10.1088/0305-4470/13/4/009, https://doi.org/10.1088/0305-4470/13/4/009