Thermodynamic glass transition in a spin glass without time-reversal symmetry

Raquel A. Baños, A. Cruz, L. A. Fernández, J. M. Gil-Narvion, A. Gordillo-Guerrero, M. Guidetti, D. Íñiguez, A. Maiorano, Enzo Marinari, V. Martı́n-Mayor, Jorge Monforte-Garcia, A. Muñoz Sudupe, D. Navarro, Giorgio Parisi, S. Pérez-Gaviro, J. J. Ruiz-Lorenzo, Sebastiano Fabio Schifano, Beatriz Seoane, A. Tarancón, Pedro Téllez, R. Tripiccione, David YllanesProceedings of the National Academy of Sciences 2012 · DOI

TL;DR

This paper provides strong evidence for a thermodynamic glass transition in a spin glass system even when time-reversal symmetry is broken by an external magnetic field. Using massive simulations on a specialized computer and a novel finite-size scaling technique, the researchers identified a critical point in four dimensions, settling a long-standing debate and offering insights into the universality class of structural glasses.

What Problem Does This Paper Address?

The study of glass transitions is a fundamental challenge in condensed matter physics. While structural glasses (like supercooled liquids or polymers) exhibit dramatically slowed dynamics without apparent structural changes, spin glasses offer a simpler, theoretically tractable model system. A key difference, however, is that spin glasses possess time-reversal symmetry in the absence of an external magnetic field. This symmetry is known to be broken at the spin glass phase transition. The crucial question this paper addresses is whether a similar thermodynamic phase transition occurs in a spin glass when time-reversal symmetry is explicitly broken by an external magnetic field. Previous theoretical work (like the de Almeida-Thouless line) suggested a transition, but experimental and numerical studies in dimensions relevant to experiments (below the upper critical dimension of 6) have yielded conflicting results, with some suggesting no transition exists in a field. This ambiguity leaves a gap in our understanding of glassy dynamics and its potential connection to structural glasses.

Methods

The researchers employed a combination of advanced computational techniques and a novel analytical approach.

  • Massive Numerical Simulations: The core of the study involved extensive simulations of the four-dimensional Edwards-Anderson model. These simulations were performed on the Janus special-purpose computer, a dedicated machine designed for such complex calculations, allowing for unprecedented system sizes and simulation lengths.
  • Novel Finite-Size Scaling (FSS) Method: Standard FSS techniques, particularly those relying on the second-moment correlation length ($\xi_2$), proved problematic due to an anomalous behavior of the correlation function at zero momentum. This anomaly obscured clear scaling signals. To overcome this, the authors introduced a new FSS method that uses ratios of higher-momentum correlation functions ($R_{12}$). This approach is less sensitive to the problematic zero-momentum mode and provides a more robust signal for scale invariance.
  • Analysis of Scaling Corrections: The authors acknowledge that even with the new $R_{12}$ method, corrections to scaling are significant. They developed a sophisticated analysis to parameterize and account for these corrections, allowing them to extract reliable critical exponents and transition temperatures from their data. This involved studying the evolution of $\xi_2/L$ as a function of $R_{12}$ and fitting the data with explicit terms for scaling corrections.

Datasets

Not discussed in the paper. The paper focuses on numerical simulations of a model system, not on experimental datasets.

Key Insights and Ideas

  • Existence of a Field-Induced Spin Glass Transition: The most significant finding is the conclusive evidence for a thermodynamic phase transition in a spin glass in the presence of an external magnetic field in $D=4$. This settles a long-standing debate and confirms that spin glasses can exhibit glassy dynamics even without time-reversal symmetry.
  • Universality Class Connection: The paper suggests that the universality class of this field-induced spin glass transition might be relevant for understanding the thermodynamic glass transition in structural glasses, a connection previously hypothesized but lacking strong evidence.
  • Power of Novel FSS Techniques: The development and successful application of the $R_{12}$ ratio method for finite-size scaling, which circumvents issues with the zero-momentum correlation function, is a crucial methodological advancement. This demonstrates how overcoming specific analytical challenges can unlock new insights.
  • Strong Corrections to Scaling: The research highlights the significant role of corrections to scaling in spin glass systems, even in dimensions well below the upper critical dimension. Understanding and controlling these corrections is essential for accurate critical exponent determination.

Key Results

  • Conclusive evidence for a phase transition in the four-dimensional Edwards-Anderson model with an external magnetic field.
  • Identification of a de Almeida-Thouless (dAT) line in the temperature-magnetic field phase plane.
  • Determination of critical exponents that differ significantly from the zero-field case. For example, the critical exponent $\nu$ was found to be approximately $1.46(7)$ for $h=0.3$ and $h=0.15$, distinct from the zero-field value of $\nu \approx 0.96(11)$.
  • Estimation of critical temperatures $T_c(h)$ for various field strengths, e.g., $T_c(0.3) \approx 0.906(40)$ and $T_c(0.15) \approx 1.229(30)$.
  • Estimation of the anomalous dimension $\eta$ of the propagator, found to be $\eta \approx -0.30(4)$, consistent with the zero-field value.

Limitations and Future Work

  • Strong Corrections to Scaling: While the authors developed methods to handle them, the presence of strong corrections to scaling means that extrapolations to infinite volume are still subject to some uncertainty, particularly for smaller lattice sizes. The paper notes that for the determination of $\eta$, quadratic corrections to scaling were necessary.
  • Preasymptotic Regime: The authors acknowledge that for some analyses, particularly when comparing with the "no-transition hypothesis," their simulated system sizes might still be in a preasymptotic regime, limiting definitive conclusions in those specific comparisons.
  • Future Work: The paper suggests that their findings settle a long-standing controversy and provide a basis for further theoretical and experimental investigations into the universality class of thermodynamic glass transitions. The successful application of the new FSS method opens avenues for its use in other complex systems. Further studies could explore higher dimensions or different spin glass models to confirm the universality of these findings.

Summary updated

Abstract

Spin glasses are a longstanding model for the sluggish dynamics that appear at the glass transition. However, spin glasses differ from structural glasses in a crucial feature: they enjoy a time reversal symmetry. This symmetry can be broken by applying an external magnetic field, but embarrassingly little is known about the critical behavior of a spin glass in a field. In this context, the space dimension is crucial. Simulations are easier to interpret in a large number of dimensions, but one must work below the upper critical dimension (i.e., in d < 6) in order for results to have relevance for experiments. Here we show conclusive evidence for the presence of a phase transition in a four-dimensional spin glass in a field. Two ingredients were crucial for this achievement: massive numerical simulations were carried out on the Janus special-purpose computer, and a new and powerful finite-size scaling method.
BibTeX
@inproceedings{banos2012thermodynamic,
  author = {Raquel A. Baños and A. Cruz and L. A. Fernández and J. M. Gil-Narvion and A. Gordillo-Guerrero and M. Guidetti and D. Íñiguez and A. Maiorano and Enzo Marinari and V. Martı́n-Mayor and Jorge Monforte-Garcia and A. Muñoz Sudupe and D. Navarro and Giorgio Parisi and S. Pérez-Gaviro and J. J. Ruiz-Lorenzo and Sebastiano Fabio Schifano and Beatriz Seoane and A. Tarancón and Pedro Téllez and R. Tripiccione and David Yllanes},
  title = {{Thermodynamic glass transition in a spin glass without time-reversal symmetry}},
  booktitle = {Proceedings of the National Academy of Sciences},
  year = {2012},
  doi = {10.1073/pnas.1203295109},
}
Figure