Critical parameters of the three-dimensional Ising spin glass

Marco Baity‐Jesi, R. A. Baños, A. Cruz, L. A. Fernández, J. M. Gil-Narvion, A. Gordillo-Guerrero, D. Íñiguez, A. Maiorano, Filippo Mantovani, Enzo Marinari, V. Martı́n-Mayor, J. Monforte-Garcia, A. Muñoz Sudupe, D. Navarro, Giorgio Parisi, S. Pérez-Gaviro, M. Pivanti, Federico Ricci‐Tersenghi, J. J. Ruiz-Lorenzo, Sebastiano Fabio Schifano, Beatriz Seoane, A. Tarancón, R. Tripiccione, David YllanesPhysical 2013 · DOI

TL;DR

This paper presents a high-precision study of the critical behavior of the 3D Ising spin glass model, a fundamental system in condensed matter physics. By employing extensive simulations on a dedicated supercomputer and a sophisticated analysis technique, the authors have determined the critical exponents and other universal quantities with unprecedented accuracy, settling long-standing questions about universality in this system.

What Problem Does This Paper Address?

Spin glasses, like magnetic alloys, exhibit complex behavior due to disorder in their interactions. Understanding their phase transitions, particularly the nature of the critical point, has been a significant challenge for decades. While it's known that the 3D Ising Edwards-Anderson model undergoes a phase transition, precisely characterizing its critical exponents and confirming universality (the idea that critical behavior is independent of microscopic details) has been difficult. Previous studies often showed apparent discrepancies, which were later attributed to "corrections to scaling" – subtle deviations from ideal critical behavior that become significant at finite system sizes. This paper aims to provide the most accurate determination yet of these critical parameters, including the leading corrections to scaling, to definitively establish the universality class of the 3D Ising spin glass.

Methods

The core of this work relies on finite-size scaling analysis, a powerful technique for studying phase transitions in systems with limited size. The authors use the "quotients method," which involves comparing observables calculated on pairs of lattices of sizes $L$ and $2L$. This method is particularly adept at revealing and quantifying corrections to scaling, which are crucial for accurate results in disordered systems.

Key aspects of their methodology include:

  • Extensive Simulations: They utilized the Janus dedicated computer, a specialized supercomputer, to perform simulations on lattices up to $L=40$. This allowed for a vast number of samples and simulation steps, crucial for reducing statistical errors.
  • Parallel Tempering: This simulation technique helps to efficiently explore the complex energy landscape of spin glasses by allowing the system to "hop" between different temperature replicas, improving the sampling of low-energy states.
  • Joint Fit: Instead of fitting individual quantities, the authors perform a global fit of multiple observables simultaneously. This approach leverages the correlations between different measurements and allows for a more robust determination of parameters, especially the elusive scaling corrections exponent $\omega$.
  • Leading-Order Corrections: The analysis explicitly accounts for the leading-order corrections to scaling, which are essential for bridging the gap between theoretical idealizations and real-world simulation data.

Datasets

Not discussed in the paper. The paper focuses on simulated data generated by the authors.

Key Insights and Ideas

  • Universality Confirmed: The study provides strong evidence that the 3D Ising spin glass belongs to a single universality class, settling previous debates about apparent universality violations.
  • Corrections to Scaling are Key: The accurate determination of critical exponents hinges on properly accounting for corrections to scaling. The authors' meticulous analysis of these corrections has led to significantly improved precision.
  • High-Precision Exponents: The paper achieves a four-fold increase in accuracy for critical exponents compared to previous state-of-the-art studies.
  • New Universal Quantities Computed: Beyond standard exponents, the authors compute several other universal quantities, providing a more complete characterization of the critical point.
  • Methodological Advancement: The use of a joint fit across multiple observables, explicitly including corrections to scaling, represents a refined approach to analyzing finite-size scaling data in complex systems.

Key Results

The paper reports highly precise values for the critical parameters of the 3D Ising spin glass:

  • Critical Temperature ($T_c$): $1.1019(29)$
  • Thermal Exponent ($\nu$): $2.562(42)$
  • Anomalous Dimension ($\eta$): $-0.3900(36)$
  • Corrections to Scaling Exponent ($\omega$): $1.12(10)$

Using these, they derive other critical exponents:

  • Specific Heat Exponent ($\alpha$): $-5.69(13)$
  • Order Parameter Exponent ($\beta$): $0.782(10)$
  • Susceptibility Exponent ($\gamma$): $6.13(11)$

They also computed several dimensionless universal quantities, such as $U_4$, $R_\xi$, $U_{1111}$, $U_{111}$, $U_{22}$, $B_\xi$, and $R_{12}$, with high precision.

Limitations and Future Work

  • Subleading Corrections: While the paper focuses on leading-order corrections to scaling, the authors acknowledge that subleading corrections can be more difficult to control, particularly for certain quantities like $R_{12}$. They suggest that a more conservative approach was taken for $R_{12}$ due to incompatibilities in extrapolations.
  • Numerical Precision: Despite achieving unprecedented precision, there are inherent limitations to numerical simulations. The authors note that the covariance matrix can become singular if too many quantities are included in the fit, limiting the number of observables that can be analyzed simultaneously.
  • Future Directions: The authors highlight that their accurate determination of critical parameters is crucial for future studies, such as investigating correlation functions below $T_c$ and the behavior of spin glasses in an applied magnetic field. They also mention that alternative methods for computing $\eta$ exist and could be explored further.

Summary updated

Abstract

We report a high-precision finite-size scaling study of the critical behavior of the three-dimensional Ising Edwards-Anderson model (the Ising spin glass). We have thermalized lattices up to $L=40$ using the Janus dedicated computer. Our analysis takes into account leading-order corrections to scaling. We obtain ${T}_{\mathrm{c}}=1.1019(29)$ for the critical temperature, $\ensuremath{\nu}=2.562(42)$ for the thermal exponent, $\ensuremath{\eta}=\ensuremath{-}0.3900(36)$ for the anomalous dimension, and $\ensuremath{\omega}=1.12(10)$ for the exponent of the leading corrections to scaling. Standard (hyper)scaling relations yield $\ensuremath{\alpha}=\ensuremath{-}5.69(13)$, $\ensuremath{\beta}=0.782(10)$, and $\ensuremath{\gamma}=6.13(11)$. We also compute several universal quantities at ${T}_{\mathrm{c}}$.
BibTeX
@inproceedings{baityjesi2013critical,
  author = {Marco Baity‐Jesi and R. A. Baños and A. Cruz and L. A. Fernández and J. M. Gil-Narvion and A. Gordillo-Guerrero and D. Íñiguez and A. Maiorano and Filippo Mantovani and Enzo Marinari and V. Martı́n-Mayor and J. Monforte-Garcia and A. Muñoz Sudupe and D. Navarro and Giorgio Parisi and S. Pérez-Gaviro and M. Pivanti and Federico Ricci‐Tersenghi and J. J. Ruiz-Lorenzo and Sebastiano Fabio Schifano and Beatriz Seoane and A. Tarancón and R. Tripiccione and David Yllanes},
  title = {{Critical parameters of the three-dimensional Ising spin glass}},
  booktitle = {Physical},
  year = {2013},
  doi = {10.1103/physrevb.88.224416},
}
Figure