Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Physical review. B, Solid state

Evidence for Fisher's Droplet Model in Simulated Two-Dimensional Cluster Distributions

E. Stoll, Kurt Binder, T. Schneider

Published Oct 1, 1972• 73 citations
Publisher

Abstract

The distribution function ${n}_{l}$ of clusters with $l$ reversed spins has been calculated for two square $N\ifmmode\times\else\texttimes\fi{}N$ Ising lattices ($N=55 \mathrm{and} 110$) at various temperatures $T<{T}_{c}$ using periodic boundary conditions. These calculations strongly support the cluster model proposed by Fisher and yield the exponent $\ensuremath{\tau}=2.1\ifmmode\pm\else\textpm\fi{}0.1$. However, for small $l$, $T$ not close to ${T}_{c}$, and $H\ensuremath{\ne}0$, we find considerable deviations.

DOI
10.1103/physrevb.6.2777

Related papers

Droplet singularity

Douglas B. Abraham, Paul J. Upton

By an exact calculation on the planar Ising model we show that for all subcritical temperatures, the bubble model predicts the correct average distribution of matter contributing to the pair-spin correlation function in the two-particle sector. This considerably substantiates the bubble model which is generalized to higher dimensions with an external field h and then used to show that the free energy has an essential singularity in h of Andreev-Fisher form at all points along the phase boundary. We argue that such a singularity would also occur in the Baxter model.

Physical Review Letters•Mar 15, 1993•15 citations

Evidence for the Droplet Picture of Spin Glasses

M A Moore, Hemant Bokil, Barbara Drossel

We have studied the Parisi overlap distribution for the three-dimensional Ising spin glass in the Migdal-Kadanoff approximation. For temperatures $T\ensuremath{\simeq}{0.7T}_{c}$ and system sizes up to $L\phantom{\rule{0ex}{0ex}}=\phantom{\rule{0ex}{0ex}}32$, we found a $P(q)$ as expected for full Parisi replica-symmetry breaking, just as was also observed in recent Monte Carlo simulations on a cubic lattice. However, for lower temperatures our data agree with predictions from the droplet or scaling picture. The failure to see droplet model behavior in Monte Carlo simulations is due to the fact that all existing simulations are done at temperatures too close to the transition temperature so that system sizes larger than the correlation length have not been achieved.

arXiv:cond-mat/9808140•Physical Review Letters•Nov 9, 1998•113 citations

Cluster-size heterogeneity in the two-dimensional Ising model

Woo Seong Jo, Sudo Yi, Seung Ki Baek +1 more

We numerically investigate the heterogeneity in cluster sizes in the two-dimensional Ising model and verify its scaling form recently proposed in the context of percolation problems [Phys. Rev. E 84, 010101(R) (2011)]. The scaling exponents obtained via the finite-size scaling analysis are shown to be consistent with theoretical values of the fractal dimension d(f) and the Fisher exponent τ for the cluster distribution. We also point out that strong finite-size effects exist due to the geometric nature of the cluster-size heterogeneity.

arXiv:1209.2568•Physical Review E•Sep 10, 2012•11 citations

Magnetic exponents of two-dimensional Ising spin glasses

Frauke Liers, Olivier Martin

The magnetic critical properties of two-dimensional Ising spin glasses are controversial. Using an exact ground-state determination, we extract the properties of clusters flipped when a uniform field is increased continuously. We show that these clusters have many holes but otherwise have statistical properties similar to those of zero-field droplets. A detailed analysis gives for the magnetization exponent $\ensuremath{\delta}\ensuremath{\approx}1.30\ifmmode\pm\else\textpm\fi{}0.02$ using lattice sizes up to $80\ifmmode\times\else\texttimes\fi{}80$; this is compatible with the droplet model prediction, $\ensuremath{\delta}=1.282$. The reason for previous disagreements stems from the need to analyze both singular and analytic contributions in the low-field regime.

arXiv:0704.1626•Physical Review B•Aug 15, 2007•6 citations

Geometrical clusters in two-dimensional random-field Ising models

László Környei, Ferenc Iglói

We consider geometrical or Ising clusters (i.e., domains of parallel spins) in the square lattice random-field Ising model by varying the strength of the Gaussian random field $\ensuremath{\Delta}$. In agreement with the conclusion of a previous investigation [Phys. Rev. E 63, 066109 (2001)], the geometrical correlation length, i.e., the average size of the clusters $\ensuremath{\xi}$ is finite for $\ensuremath{\Delta}>{\ensuremath{\Delta}}_{c}\ensuremath{\approx}1.65$ and divergent for $\ensuremath{\Delta}\ensuremath{\leqslant}{\ensuremath{\Delta}}_{c}$. The scaling function of the distribution of the mass of the clusters as well as the geometrical correlation function are found to involve the scaling exponents of critical percolation. On the other hand, the divergence of the correlation length, $\ensuremath{\xi}(\ensuremath{\Delta})\ensuremath{\sim}{(\ensuremath{\Delta}\ensuremath{-}{\ensuremath{\Delta}}_{c})}^{\ensuremath{-}\ensuremath{\nu}}$, with $\ensuremath{\nu}\ensuremath{\approx}2$, is related to that of tricritical percolation. It is verified numerically that critical geometrical correlations transform conformally.

arXiv:cond-mat/0610110•Physical Review E•Jan 29, 2007•15 citations

Study of droplets for correlated site-bond percolation in two dimensions

Naeem Jan, Antonio Coniglio A, Dietrich Stauffer

The authors study the droplet size distribution of the correlated site-bond percolation model introduced by Coniglio and Klein (1980), and also the usual clusters of two-dimensional Ising models near the critical point. Equilibrium configurations of the Ising model with nearest-neighbour interaction and also one with nearest- and next-nearest-neighbour interactions are generated through a Monte Carlo simulation, and then a cluster analysis is performed. The exponents beta and gamma for the Coniglio and Klein droplet distribution are found to agree, for both the nearest-neighbour and the next-nearest-neighbour model, with the corresponding exponents of the Ising model. The usual Ising clusters diverge only at T c in the Ising model with nearest-neighbour interaction but not for the model with next-nearest-neighbour interaction. The Potts model formulation is used to predict the behaviour of the droplet for general further-neighbour interactions.

Journal of Physics A Mathematical and General•Dec 1, 1982•16 citations
The distribution function nl{n}_{l}nl​ of clusters with lll reversed spins has been calculated for two square N\ifmmode×\else\texttimes\fiNN\ifmmode\times\else\texttimes\fi{}NN\ifmmode×\else\texttimes\fiN Ising lattices (N=55and110N=55 \mathrm{and} 110N=55and110) at various temperatures T<TcT<{T}_{c}T<Tc​ using periodic boundary conditions. These calculations strongly support the cluster model proposed by Fisher and yield the exponent \ensuremathτ=2.1\ifmmode±\else\textpm\fi0.1\ensuremath{\tau}=2.1\ifmmode\pm\else\textpm\fi{}0.1\ensuremathτ=2.1\ifmmode±\else\textpm\fi0.1. However, for small lll, TTT not close to Tc{T}_{c}Tc​, and H\ensuremath≠0H\ensuremath{\ne}0H\ensuremath=0, we find considerable deviations.
By an exact calculation on the planar Ising model we show that for all subcritical temperatures, the bubble model predicts the correct average distribution of matter contributing to the pair-spin correlation function in the two-particle sector. This considerably substantiates the bubble model which is generalized to higher dimensions with an external field h and then used to show that the free energy has an essential singularity in h of Andreev-Fisher form at all points along the phase boundary. We argue that such a singularity would also occur in the Baxter model.
We have studied the Parisi overlap distribution for the three-dimensional Ising spin glass in the Migdal-Kadanoff approximation. For temperatures $T\ensuremath{\simeq}{0.7T}_{c}$ and system sizes up to L=32L\phantom{\rule{0ex}{0ex}}=\phantom{\rule{0ex}{0ex}}32L=32, we found a P(q)P(q)P(q) as expected for full Parisi replica-symmetry breaking, just as was also observed in recent Monte Carlo simulations on a cubic lattice. However, for lower temperatures our data agree with predictions from the droplet or scaling picture. The failure to see droplet model behavior in Monte Carlo simulations is due to the fact that all existing simulations are done at temperatures too close to the transition temperature so that system sizes larger than the correlation length have not been achieved.
We numerically investigate the heterogeneity in cluster sizes in the two-dimensional Ising model and verify its scaling form recently proposed in the context of percolation problems [Phys. Rev. E 84, 010101(R) (2011)]. The scaling exponents obtained via the finite-size scaling analysis are shown to be consistent with theoretical values of the fractal dimension d(f) and the Fisher exponent τ for the cluster distribution. We also point out that strong finite-size effects exist due to the geometric nature of the cluster-size heterogeneity.
The magnetic critical properties of two-dimensional Ising spin glasses are controversial. Using an exact ground-state determination, we extract the properties of clusters flipped when a uniform field is increased continuously. We show that these clusters have many holes but otherwise have statistical properties similar to those of zero-field droplets. A detailed analysis gives for the magnetization exponent $\ensuremath{\delta}\ensuremath{\approx}1.30\ifmmode\pm\else\textpm\fi{}0.02$ using lattice sizes up to $80\ifmmode\times\else\texttimes\fi{}80$; this is compatible with the droplet model prediction, $\ensuremath{\delta}=1.282$. The reason for previous disagreements stems from the need to analyze both singular and analytic contributions in the low-field regime.
We consider geometrical or Ising clusters (i.e., domains of parallel spins) in the square lattice random-field Ising model by varying the strength of the Gaussian random field $\ensuremath{\Delta}$. In agreement with the conclusion of a previous investigation [Phys. Rev. E 63, 066109 (2001)], the geometrical correlation length, i.e., the average size of the clusters $\ensuremath{\xi}$ is finite for $\ensuremath{\Delta}>{\ensuremath{\Delta}}_{c}\ensuremath{\approx}1.65$ and divergent for $\ensuremath{\Delta}\ensuremath{\leqslant}{\ensuremath{\Delta}}_{c}$. The scaling function of the distribution of the mass of the clusters as well as the geometrical correlation function are found to involve the scaling exponents of critical percolation. On the other hand, the divergence of the correlation length, $\ensuremath{\xi}(\ensuremath{\Delta})\ensuremath{\sim}{(\ensuremath{\Delta}\ensuremath{-}{\ensuremath{\Delta}}_{c})}^{\ensuremath{-}\ensuremath{\nu}}$, with $\ensuremath{\nu}\ensuremath{\approx}2$, is related to that of tricritical percolation. It is verified numerically that critical geometrical correlations transform conformally.
The authors study the droplet size distribution of the correlated site-bond percolation model introduced by Coniglio and Klein (1980), and also the usual clusters of two-dimensional Ising models near the critical point. Equilibrium configurations of the Ising model with nearest-neighbour interaction and also one with nearest- and next-nearest-neighbour interactions are generated through a Monte Carlo simulation, and then a cluster analysis is performed. The exponents beta and gamma for the Coniglio and Klein droplet distribution are found to agree, for both the nearest-neighbour and the next-nearest-neighbour model, with the corresponding exponents of the Ising model. The usual Ising clusters diverge only at T c in the Ising model with nearest-neighbour interaction but not for the model with next-nearest-neighbour interaction. The Potts model formulation is used to predict the behaviour of the droplet for general further-neighbour interactions.