Classical spin liquids from frustrated Ising models in hyperbolic space

Fabian K\"ohler; Johanna Erdmenger; Matthias Vojta; Roderich Moessner

REVIEW 3 major objections 5 minor 59 references

Classical spin liquids from frustrated Ising models in hyperbolic space

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The antiferromagnetic Ising model on hyperbolic {3,q} tilings with type-A boundaries realizes a classical spin liquid, with residual entropy density 0.102(2) per spin for {3,7}; a type-B boundary instead produces a unique ordered…

desk verdict Solid, genuinely new result: exact FKT ground-state counting on hyperbolic {3,q} tilings plus a boundary-shape switch between spin liquid and ferrimagnet; the one soft spot is the extrapolation behind sres = 0.102(2). read the letter →

arxiv 2506.02113 v2 pith:IWAFBP27 submitted 2025-06-02 cond-mat.str-el cond-mat.mes-hallhep-th

Fabian Köhler , Johanna Erdmenger , Roderich Moessner , Matthias Vojta This is my paper

classification cond-mat.str-elcond-mat.mes-hallhep-th

keywords hyperbolicclassicalspinliquidgeometricfrustrationantiferromagneticIsingmodel{3q}tilingsresidualentropyFisher-Kasteleyn-Temperleyalgorithmboundarycontrolferrimagneticorder

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Antiferromagnetic nearest-neighbor Ising spins on hyperbolic triangular tessellations realize a classical spin liquid: exponentially many ground states, no thermal phase transition, and no global magnetic order at zero temperature. The paper proves this for $\{3,q\}$ tilings with smooth open (type-A) boundaries by counting all ground states exactly through a dimer-covering mapping, obtaining a residual entropy density $s_{\mathrm{res}}=0.102(2)$ for $\{3,7\}$ and finite values for all $q\geq 7$. A second result is boundary control: adding one triangle to every boundary bond (type-B boundary) removes the degeneracy and produces a unique ferrimagnetic ground state. This matters because hyperbolic space makes the boundary an extensive part of the system, so the boundary shape is not a perturbation but a switch between liquid and ordered low-energy behavior.

What carries the argument

The load-bearing object is the duality between Ising ground states and dimer coverings on the dual lattice, evaluated with the Fisher-Kasteleyn-Temperley (FKT) algorithm. Because every plaquette has odd length, each plaquette needs at least one frustrated bond; a ground state chooses one frustrated bond per plaquette, which pairs plaquettes through shared edges and is equivalent to a perfect matching of the plaquette graph. FKT computes the number of such matchings as a Pfaffian, hence $S_{\mathrm{res}}=\ln 2+\frac{1}{2}\ln\det A'$. The lattices themselves are generated by a recursive inflation rule in which boundary vertices of exposure 3 and 4 spawn sequences $(3)\to(34)$, $(4)\to(344)$, giving closed-form expressions for the numbers of vertices, edges, plaquettes, and the boundary fraction $N_{\mathrm{bnd}}/N\to(\sqrt{5}-1)/2\approx0.618$ for $\{3,7\}$. These closed forms are what turn the dimer count into a residual entropy per spin and what justify discarding odd-plaquette boundary corrections of order $O(\ln N/N)$.

What would settle it

Compute exact FKT ground-state counts for a different natural infinite sequence of type-A $\{3,7\}$ lattices, for instance a shifted or differently truncated layer construction, and check whether $S_{\mathrm{res}}/N$ extrapolates to $0.102(2)$; a different limit, or a limit of zero, would show the spin liquid is a boundary artifact rather than a bulk phase.

Watch

Extended reading notes

Core claim

The paper's central claim is that geometric frustration acts in negatively curved space to produce a classical spin liquid. For the Hamiltonian $H=J\sum_{\langle ij\rangle}\sigma_i\sigma_j$ on $\{3,q\}$ tessellations with type-A boundaries, every ground state corresponds to a perfect dimer covering of the dual graph, and the Fisher-Kasteleyn-Temperley algorithm counts these coverings exactly: $S_{\mathrm{res}}=\ln 2+\frac{1}{2}\ln\det A'$. Monte Carlo simulations show a broad, size-independent specific-heat maximum, no diverging susceptibility, no long-range order at $T=0$, and a low-temperature Curie tail whose prefactor shrinks with system size. In the $N\to\infty$ limit the entropy per spin for $\{3,7\}$ converges to $0.102(2)$, and the ground-state energy per spin approaches $-J$. If the boundary is changed to type-B, the unique dimer covering yields a ferrimagnet with ferromagnetically ordered layers antiferromagnetically stacked, and the residual entropy vanishes; the spin-liquid configurations remain as low-lying excited states.

Load-bearing premise

The load-bearing premise is that the thermodynamic limit of the residual entropy per spin for type-A lattices is unique, so that different layer truncations and odd-plaquette systems all extrapolate to the same value once corrections of order $O(\ln N/N)$ are discarded.

Editorial extensions

If this is right

The finite residual entropy density is not special to $\{3,7\}$: FKT data for $\{3,q\}$ with $q=8,10,12,14,16$ all extrapolate to nonzero values that decrease with $q$.
For type-A systems the ground-state manifold contains magnetized states, so the low-temperature uniform susceptibility obeys a Curie law $\chi\sim1/T$ whose prefactor decreases with system size.
Changing the boundary to type-B selects a unique ground state, so the boundary shape controls whether the low-temperature phase is a spin liquid or an ordered ferrimagnet.
The ferromagnetic Ising model on the same type-A lattices orders at $T_c/J\approx3.3$, confirming that the spin-liquid signatures in the antiferromagnet are not an artifact of the hyperbolic geometry or of Monte Carlo freezing at low temperatures.
The ground-state energy per spin approaches $-J$ in the thermodynamic limit, consistent with a maximally satisfied set of antiferromagnetic bonds coexisting with an exponentially large ground-state degeneracy.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A direct test that would separate bulk from boundary entropy: compute $s_{\mathrm{res}}$ for type-A boundaries with a different boundary fraction $N_{\mathrm{bnd}}/N$ and compare against the $q$-dependence reported here; the paper's data suggest the trend but do not isolate the boundary contribution.
The same dimer-counting logic should extend to other odd-sided hyperbolic tilings such as $\{5,q\}$; if the residual entropy remains finite, geometric-frustration spin liquidity is generic to odd-loop hyperbolic tessellations rather than specific to triangles.
If the holographic motivation is taken seriously, the boundary-controlled switch between a degenerate liquid and a unique ordered state implies that the cutoff surface in a tensor-network realization is itself a physical tuning parameter for the low-energy sector; this consequence is not developed in the paper.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Bottom line: this is a genuinely new and mostly solid paper. The antiferromagnetic nearest-neighbor Ising model on {3,q} hyperbolic tilings has not been studied before, and the authors use the right tool for the central quantity: exact FKT dimer counting for the ground-state degeneracy, corroborated by Monte Carlo. The headline result is the boundary-shape switch — type-A boundaries give a classical spin liquid with finite residual entropy density sres ≈ 0.10, while type-B boundaries pin a unique ferrimagnetic ground state, provable by a clean inductive argument on the dimer covering. Because the boundary is an extensive fraction of the system (about 62% for {3,7} type-A), this is not a surface correction; it is a genuine control parameter. That is the real contribution.

What is done well: the FKT counting is exact for even-plaquette systems and the MC agrees; the ferromagnetic benchmark in the SM shows the same simulations resolve a clear transition with a Binder crossing, which makes the negative evidence (no transition, no order) in the AFM case credible; the ground-state energy formula checks against MC; and the Curie-tail analysis, traced to the magnetization distribution within the ground-state manifold, is good physics. The citation pattern is fair, including the prior hyperbolic Ising and FKT literature.

Soft spots, in proportion. The headline value sres = 0.102(2) rests on the Fig. 5 extrapolation, which mixes even-plaquette exact FKT points with odd-plaquette systems (N=232, 617) where FKT does not directly apply. The paper announces an O(ln N/N) boundary-entropy correction but does not show it being subtracted; at N=617 that term is about 0.01, roughly five times the quoted error. So the intercept could drift by a few thousandths. That does not threaten the qualitative conclusion — the data for {3,8} through {3,16} all converge to something near 0.10, and the spin-liquid label only needs finiteness — but the third significant figure is not secured, and the claim of a 'unique' limit would be cleaner with a second type-A construction tested, given the paper's own caveat that hyperbolic OBC thermodynamic limits are boundary-dependent. Minor: sres appears as 0.102(2) in one place and 0.102(1) in another, and there is no code or data deposit. The AdS/CFT paragraph in the outlook is speculative but labeled as such.

Who it is for: people working on hyperbolic lattice systems, frustrated magnetism in curved space, and discrete holography. It deserves a serious referee. Accept-if-sound, with a request to clarify the odd-plaquette points and the extrapolation.

Referee Report

3 major / 5 minor

Summary. The paper studies the nearest-neighbor antiferromagnetic Ising model on hyperbolic {3,q} tessellations with open boundaries. For type-A boundaries (central-vertex growth), the authors argue that the system realizes a classical spin liquid: an exponentially large ground-state degeneracy computed exactly via the Fisher-Kasteleyn-Temperley (FKT) algorithm, a finite residual entropy density (s_res = 0.102(2) for {3,7}), no thermal phase transition, and no global long-range order at T=0. For type-B boundaries, obtained by adding a triangle to every boundary edge, the ground state is unique up to global spin flip and is a ferrimagnet. The claims are supported by exact finite-size ground-state counts, Monte Carlo thermodynamics, and correlation-function data, with the FKT and Monte Carlo results agreeing where they overlap.

Significance. If established, this is a genuinely novel contribution: it extends the classical spin liquid paradigm from flat-space frustrated lattices to hyperbolic space, and it demonstrates that the boundary of a hyperbolic open system can select between disordered spin-liquid and ordered ferrimagnetic ground states. The main technical strength is the exact FKT counting of ground states for finite even-plaquette type-A systems, which is a parameter-free determination of the ground-state degeneracy, together with the explicit Monte Carlo cross-checks. The exact inductive proof of the unique type-B ground state is also a strong result. The central quantitative claim, however, rests on a finite-size extrapolation that mixes data sets of different parity and boundary structure, and the uniqueness of the thermodynamic limit within the type-A family is not demonstrated.

major comments (3)

[Ground-state manifold; Eq. (4) and Fig. 5] The quoted value s_res = 0.102(2) for {3,7} is obtained from an extrapolation that mixes even-plaquette FKT data with odd-plaquette systems N=232 and N=617 for which the FKT algorithm is not directly applicable. The paper states that these odd-plaquette systems carry an additional boundary entropy of order O(ln(N_bnd)), but this correction is never subtracted or propagated into the fit. For N=617, ln(617)/617 is approximately 0.010, which is comparable to the vertical scatter in Fig. 5 and about five times the quoted error 0.002. A fit that includes these points without a quantitative correction therefore does not determine the intercept to the claimed precision. Please repeat the extrapolation using only even-plaquette FKT systems, or implement the O(ln N/N) correction explicitly and show that the intercept and error bar are stable.
[SM S1.C and S4.A; Summary and outlook] The claim that the type-A family has a well-defined, unique large-N limit is not established. Only a single growth sequence (central-vertex inflation with full layers plus a few partially completed layers) is used, while SM S4.A explicitly notes that the thermodynamic limit in hyperbolic open-boundary systems cannot be discussed independently of boundary conditions. Because the boundary fraction is finite (about 0.62 for {3,7}), a boundary-dependent finite-size term could shift the extrapolated intercept. Please provide either a second type-A-like construction (for example, growth from a central plaquette or a different layer truncation) showing the same intercept, or a quantitative argument that all type-A constructions share the same limiting residual entropy density.
[Classical spin liquid; Fig. 2] The absence of a thermal phase transition for type-A systems is inferred mainly from a broad specific-heat maximum whose height and shape are essentially size independent. This diagnostic is not conclusive for hyperbolic open systems: the ferromagnetic Ising model on the same lattices shows a genuine transition (SM S4.A, Figs. S5-S7) even though its specific heat also remains broad and nonsingular. Please provide an order-parameter susceptibility or Binder cumulant for the antiferromagnetic type-A systems showing no crossing or divergence, or soften the claim to state that no transition is observed in the accessible size range.

minor comments (5)

[Title and abstract] The title contains a typo ('s pace' instead of 'space'), and 'tesselations' should be 'tessellations' throughout the manuscript.
[Fig. 5 caption] The caption should unambiguously identify which symbols are FKT data, which are Monte Carlo data, and which data points (notably N=232 and N=617) are not direct FKT counts because of odd plaquette parity.
[Eq. (2)] The fitting parameter gamma in Eq. (2) is described as an unknown ratio of ground and excited states; it would help to state explicitly that this is a fitted quantity and that it does not enter the exact FKT residual-entropy determination.
[SM S2.B] The sentence 'This opens up the use of numerical approximations for the logdet to increase calculation speed' is vague; please specify the approximation used for the reported system sizes.
[SM S3.B] The statement about the variance of the Bernoulli estimator is correct, but the notation D is introduced without defining the number of independent samples; please define it explicitly.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular reduction: FKT ground-state counting and MC thermodynamics are independent; the only self-citation is for lattice construction and is not load-bearing.

full rationale

The central quantitative claims are derived in-paper, not imported from a fit or from a self-citation. The ground-state degeneracy is computed exactly via the FKT Pfaffian (Eq. 4 and SM Eq. S10) on the dual graph of each finite lattice, and the residual entropy sres = 0.102(2) is an extrapolation of these exact counts shown in Fig. 5. The Monte-Carlo entropy integration (Eq. 3) is an independent check: the only fitted parameter is the ratio gamma in the activated form Eq. (2), with the gap Delta = 2J fixed by the single-spin-flip energy cost, and the resulting MC sres is consistent with the FKT values rather than being used to define them. The type-B unique ground state is proven by an inductive dimer-covering argument in SM S2.C, not assumed. The self-citation to Ref. 12 supplies the iterative lattice-inflation algorithm, but the same rules are restated and analyzed in SM S1.C, and the spin-liquid conclusion does not rest on any uniqueness theorem imported from that reference. The paper's own caveats — that OBC thermodynamic limits in hyperbolic space are boundary-dependent (SM S4.A) and that the Fig. 5 extrapolation includes odd-plaquette systems with only an O(ln N/N) boundary correction — are finite-size and boundary-uniqueness concerns, not circular reductions. Against external benchmarks (triangular-lattice sres = 0.32306 from Wannier; the ferromagnetic transition consistent with Refs. 6, 18–21) the paper is self-contained. No specific circular step can be exhibited; the score of 2 merely acknowledges the minor, non-load-bearing self-citation for the lattice construction.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The FKT-derived residual entropy is the main contribution and is not grounded in any fitted parameter. The assumptions are the standard planar-graph Pfaffian method, the dimer mapping for frustrated odd-loop Ising models, and the assertion that a unique thermodynamic limit exists for hyperbolic open-boundary lattices. No new entities are introduced.

free parameters (1)

gamma (fitted degeneracy ratio in low-T activation fit) = not quoted (fit parameter)
Appears in Eq. (2) for the low-temperature energy/entropy of type-A MC data; it is the unknown ratio of ground to excited state counts. It affects the MC entropy estimate but not the FKT-based residual entropy.

assumptions (4)

standard math FKT algorithm yields the number of perfect matchings of a planar graph via the Pfaffian of a skew-symmetric signed adjacency matrix.
Invoked in 'Ground-state manifold' and SM S2.A to count dimer coverings on the dual {7,3} graph.
domain assumption Ground states of the AF Ising model on a planar graph with odd plaquette lengths correspond to dimer coverings of the dual graph, with one frustrated bond per plaquette.
Used to map the spin problem to perfect matchings; requires even number of plaquettes or an outside plaquette, as discussed in SM S2.
domain assumption The nearest-neighbor exchange coupling is independent of curvature and takes the flat-space Ising form H = J sum sigma_i sigma_j.
Stated in the introduction: 'assuming an interaction of the form (1) irrespective of curvature'.
domain assumption For type-A hyperbolic tilings, the thermodynamic limit N to infinity with open boundaries yields a unique residual entropy per spin, with odd-plaquette boundary corrections O(ln N/N) vanishing in the limit.
Used to extract sres from finite-size FKT data in Fig. 5 and to justify the large-system spin-liquid statement; argued but not proven.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Classical spin liquids from frustrated Ising models in hyperbolic space." pith.science (2026). https://pith.science/paper/IWAFBP27

@misc{pith2026250602113,
  author       = {Pith},
  title        = {Pith review of: Classical spin liquids from frustrated Ising models in hyperbolic space},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IWAFBP27}},
  note         = {Machine review of arXiv:2506.02113}
}

read the original abstract

Antiferromagnetic Ising models on frustrated lattices can realize classical spin liquids, with highly degenerate ground states and, possibly, fractionalized excitations and emergent gauge fields. Motivated by the recent interest in many-body system in negatively curved space, we study hyperbolic frustrated Ising models. Specifically, we consider nearest-neighbor Ising models on tesselations with odd-length loops in two-dimensional hyperbolic space. For finite systems with open boundaries we determine the ground-state degeneracy exactly, and we perform extensive finite-temperature Monte-Carlo simulations to obtain thermodynamic data as well as correlation functions. We show that the shape of the boundary, constituting an extensive part of the system, can be used to control low-energy states: Depending on the boundary, we find ordered or disordered ground states. Our results demonstrate how geometric frustration acts in curved space to produce classical spin liquids.

Figures

Figures reproduced from arXiv: 2506.02113 by the authors.

**Figure 3.** We observe a very broad maximum in χ(T)/N around T /J = 1.2. Below T /J ∼ 0.3 data for different system sizes strongly differ, each obeying a lowtemperature Curie law, χ ∼ 1/T, with the prefactor decreasing with system size. As detailed below and in SM [36], the reason is that the ground-state manifold contains states with non-zero magnetization, but the magnetization distribution in this manifold has a non-tri… view at source ↗

**Figure 5.** FIG. 5. Residual entropy per spin of [PITH_FULL_IMAGE:figures/full_fig_p004_5.png] view at source ↗

Figures from the paper (1 more)

**Figure 7.** Figure 7: FIG. 7. NN bond correlations [PITH_FULL_IMAGE:figures/full_fig_p005_7.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

59 extracted references · 50 canonical work pages

[1]

Maciejko and S

J. Maciejko and S. Rayan, Sci. Adv. 7, eabe9170 (2021)

work page 2021
[2]

Cheng, F

N. Cheng, F. Seraﬁn, J. McInerney1, Z. Rocklin, K. Sun, and X. Mao, Phys. Rev. Lett. 129, 088002 (2022)

work page 2022
[3]

Boettcher, A

I. Boettcher, A. V. Gorshkov, A. J. Kollar, J. Maciejko, S. Rayan, and R. Thomale, Phys. Rev. B 105, 125118 (2022)

work page 2022
[4]

C. G. Callan and F. Wilczek, Nucl. Phys. B 340, 366 (1990)

work page 1990
[7]

J. M. Maldacena, Int. J. Theor. Phys. 38, 1113 (1999)

1999
[8]

Witten, Adv

E. Witten, Adv. Theor. Math. Phys. 2, 253 (1998)

1998
[9]

S. S. Gubser, I. R. Klebanov, and A. M. Polyakov, Phys. Lett. B 428, 105 (1998)

work page 1998
[10]

Axenides, E

M. Axenides, E. G. Floratos and S. Nicolis, J. High En- erg. Phys. 02, 109 (2014)

work page 2014

Show all 59 references

[11]

R. C. Brower, C. V. Cogburn, A. L. Fitzpatrick, D. Howarth, and C. Tan, Phys. Rev. D 103, 094507 (2021)

2021
[13]

S. Dey, A. Chen, P. Basteiro, A. Fritzsche, M. Greiter, M. Kaminski, P. M. Lenggenhager, R. Meyer, R. Sorbello, A. Stegmaier, R. Thomale, J. Erdmenger, and I. Boettcher, Phys. Rev. Lett. 133, 061603 (2024)

2024
[14]

Erdmenger, J

J. Erdmenger, J. Karl, Y. Thurn, M. Vojta, and Z.-Y. Xian, J. High Energ. Phys. 05, 165 (2025)

2025
[15]

N. P. Breuckmann, C. Vuillot, E. Campbell, A. Krishna, and B. M. Terhal, Quantum Sci. Technol. 2, 035007 (2017)

2017
[16]

Mosseri and J

R. Mosseri and J. Sadoc, J. Phys. Lett. 43, 249 (1982)

1982
[17]

S. K. Baek, H. Shima, and B. J. Kim, Phys. Rev. E 79, 060106(R) (2009)

2009
[18]

Rietman, B

R. Rietman, B. Nienhuis, and J. Oitmaa, J. Phys. A: Math. Gen. 25, 6577 (1992)

1992
[19]

Krcmar, A

R. Krcmar, A. Gendiar, K. Ueda, and T. Nishino, J. Phys. A: Math. Gen. 41, 125001 (2008)

2008
[20]

Gendiar, R

A. Gendiar, R. Krcmar, S. Andergassen, M. Daniska, and T. Nishino, Phys. Rev. E 86, 021105 (2012)

2012
[21]

Asaduzzaman, S

M. Asaduzzaman, S. Catterall, J. Hubisz, R. Nelson, and J. Unmuth-Yockey, Phys. Rev. D 106, 054506 (2022)

2022
[22]

Krcmar, T

R. Krcmar, T. Iharagi, A. Gendiar, and T. Nishino, Phys. Rev. E 78, 061119 (2008)

2008
[23]

Placke and N

B. Placke and N. P. Breuckmann, Phys. Rev. E 107, 024125 (2023)

2023
[24]

D. R. Nelson, Phys. Rev. Lett. 50, 982 (1983)

1983
[25]

Balents, Nature 464, 199 (2010)

L. Balents, Nature 464, 199 (2010)

2010
[26]

Savary and L

L. Savary and L. Balents, Rep. Prog. Phys. 80, 016502 (2017)

2017
[27]

Knolle and R

J. Knolle and R. Moessner, Annu. Rev. Condens. Matter Phys. 10, 451 (2019)

2019
[28]

P. M. Lenggenhager, S. Dey, T. Bzduˆ sek, and J. Ma- ciejko, Phys. Rev. Lett. 135, 076604 (2025)

2025
[29]

Dusel, T

F. Dusel, T. Hofmann, A. Maity, Y. Iqbal, M. Greiter, and R. Thomale, Phys. Rev. Lett. 134, 256604 (2025)

2025
[30]

Elser and C

V. Elser and C. Zeng, Phys. Rev. B 48, 13647 (1993). 6

1993
[31]

G. H. Wannier, Phys. Rev. 79, 357 (1950)

1950
[32]

S. T. Bramwell and M. J. P. Gingras, Science 294, 1495 (2001)

2001
[33]

Castelnovo, R

C. Castelnovo, R. Moessner, and S. L. Sondhi, Nature 451, 42 (2008)

2008
[34]

Pauling, J

L. Pauling, J. Am. Chem. Soc. 57, 2680 (1935)

1935
[35]

Boyle, M

L. Boyle, M. Dickens, and F. Flicker, Phys. Rev. X 10, 011009 (2020)

2020
[36]

See supplement for a detailed discussion of the hyperboli c tilings, additional numerical results for the Ising spin- liquid, as well as results for type-B antiferromagnets and ferromagnets
[37]

Standard cluster update schemes do not work well in MC simulations on hyperbolic lattices because of the large boundary fraction of sites
[38]

Alternatively, a high-temperature expansion can be used to obtain the exact behavior of CV (T ) in the high- T limit
[41]

Schrauth, Y

M. Schrauth, Y. Thurn, F. Goth, J. S. E. Portela, D. Herdt, and F. Dusel, arXiv:2309.10844

arXiv
[42]

J. Chen, F. Chen, L. Yang, Y. Yang, Z. Chen, Y. Wu, Y. Meng, B. Yan, X. Xi, Z. Zhu, M. Cheng, G.-G. Liu, P. P. Shum, H. Chen, R.-G. Cai, R. Yang, Y. Yang, and Z. Gao, arXiv:2305.04862

arXiv
[43]

Sachdev and J

S. Sachdev and J. Ye, Phys. Rev. Lett. 70, 3339 (1992)

1992
[44]

Since the Sachdev-Ye model is a spin model, our results raise the possibility that a similar du- ality may be realized between spin chains and antiferro- magnetic Ising models

– a result which triggered intense research estab- lishing holographic dualities for the Sachdev-Ye-Kitaev (SYK) model [45]. Since the Sachdev-Ye model is a spin model, our results raise the possibility that a similar du- ality may be realized between spin chains and antiferro...
[45]

Sachdev, Phys

S. Sachdev, Phys. Rev. X 5, 041025 (2015)

2015
[46]

Chowdhury, A

D. Chowdhury, A. Georges, O. Parcollet, and S. Sachdev, Rev. Mod. Phys. 94, 035004 (2022)

2022
[47]

Basteiro, R

P. Basteiro, R. N. Das, G. Di Giulio, and J. Erdmenger, SciPost Phys. 15, 218 (2023). Supplemental material for: Classical spin liquids from frustrated Ising models in hyperbolic s pace Fabian K¨ ohler,1 Johanna Erdmenger, 2 Roderich Moessner, 3 and Matthias Vojta 1 1Institut ...

2023
[48]

set a vertex of a {p, q} tesselation as the origin of the ﬂake (layer L0)
[49]

identify all vertices at the current boundary, i.e., with coordination less than q
[50]

then insert the missing neighbors of that vertex and close the newly started plaquettes with loop length p (layer +1)
[51]

This construction yields a “smooth” boundary, compat- ible with the FKT algorithm described below

repeat steps 2 and 3 until layer Ln is reached. This construction yields a “smooth” boundary, compat- ible with the FKT algorithm described below. The con- crete inﬂation rules for step 3 will be described now. S2 3 4 3 4 4 3 4 4 3 3 4 4 Li Li+1 FIG. S1. Inﬂation rules for one...
[52]

due to the exponential dependence between the to- tal and the linear system size (i.e. radius). Contrary to ﬂat space, the number of boundary sites of OBC systems approaches a ﬁxed ratio of the total number of spins, see Sec. S1 C. As a result, the system is eﬀectively in- hom...
[53]

distances

4 0 1 2 3 4 S ( T )/ N 0 0 .25 0 .50 l n( 2 ) N = 15 N = 50 N = 141 N = 379 FIG. S8. Speciﬁc heat (main) and entropy (inset) per site of the Ising antiferromagnet on type-B {3, 7} lattices of diﬀerent size. The residual entropy vanishes within error bars. The data spread in CV...
[54]

Boettcher, A

I. Boettcher, A. V. Gorshkov, A. J. Kollar, J. Maciejko, S. Rayan and R. Thomale, Phy. Rev. B 105, 125118 (2022)

2022
[55]

N. P. Breuckmann, B. Placke, and A. Roy, Phys. Rev. E 101, 022124 (2020)

2020
[56]

Basteiro, G

P. Basteiro, G. D. Giulio, J. Erdmenger, J. Karl, R. Meyer, and Z.-Y. Xian, SciPost Phys. 13, 103 (2022)

2022
[57]

Euler, Commentarii Academiae Scientiarum Imperi- alis Petropolitanae, 8, 128 (1736)

L. Euler, Commentarii Academiae Scientiarum Imperi- alis Petropolitanae, 8, 128 (1736)
[58]

H. N. V. Temperley and M. Fisher, Phil. Mag. 6, 1061 (1961)

1961
[59]

P. W. Kasteleyn, Physica 27, 1209 (1961)

1961
[60]

Mnasri, B

K. Mnasri, B. Jeevanesan, and J. Schmalian, Phys. Rev. B 92, 134423 (2015)

2015
[61]

Thomopoulos, Essentials of Monte Carlo Simulations , (Springer New York, NY, 2013)

N. Thomopoulos, Essentials of Monte Carlo Simulations , (Springer New York, NY, 2013)

2013
[62]

K. Ueda, R. Krcmar, A. Gendiar, and T. Nishino, J. Phys. Soc. Jpn. 76, 084004 (2007)

2007
[63]

F. R. Lux and E. Prodan, Phys. Rev. Lett. 131, 176603 (2023)

2023
[64]

G¨ otz, G

A. G¨ otz, G. Rein, J. Carvalho Inacio, and F. F. Assaad, Phys. Rev. B 110, 235105 (2024)

2024

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Maciejko and S

J. Maciejko and S. Rayan, Sci. Adv. 7, eabe9170 (2021)

work page 2021

[2] [2]

Cheng, F

N. Cheng, F. Seraﬁn, J. McInerney1, Z. Rocklin, K. Sun, and X. Mao, Phys. Rev. Lett. 129, 088002 (2022)

work page 2022

[3] [3]

Boettcher, A

I. Boettcher, A. V. Gorshkov, A. J. Kollar, J. Maciejko, S. Rayan, and R. Thomale, Phys. Rev. B 105, 125118 (2022)

work page 2022

[4] [4]

C. G. Callan and F. Wilczek, Nucl. Phys. B 340, 366 (1990)

work page 1990

[5] [7]

J. M. Maldacena, Int. J. Theor. Phys. 38, 1113 (1999)

1999

[6] [8]

Witten, Adv

E. Witten, Adv. Theor. Math. Phys. 2, 253 (1998)

1998

[7] [9]

S. S. Gubser, I. R. Klebanov, and A. M. Polyakov, Phys. Lett. B 428, 105 (1998)

work page 1998

[8] [10]

Axenides, E

M. Axenides, E. G. Floratos and S. Nicolis, J. High En- erg. Phys. 02, 109 (2014)

work page 2014