REVIEW
Nonequilibrium dynamic exponent and spin-glass transitions
Disordered Systems and Neural Networks · Submitted 3 Mar 2006 · 59 references
Pith's verdict: Careful MC study confirming the 1/T dynamic exponent law and reporting a possible chiral/spin separation in 3D Heisenberg spin glass, but the separation rests on a weak chiral signal that needs cross-checking.
T0 · AI review
Why Pith says this
Highlight any text to comment on it.
Referee report
The paper
The paper claims
The 3D Heisenberg spin glass freezes chirality at 0.23 and spins at 0.18, with one shared dynamic exponent z = 5.1.
The reading
What carries the argument
The load-bearing object is the nonequilibrium dynamic exponent $z(T)$, defined by the algebraic growth of the dynamic correlation length $\xi(t)\sim t^{1/z(T)}$ in the relaxation after a quench. Its empirical law $z_s(T)=z_{\mathrm{sg}}T_{\mathrm{sg}}/T$ is inserted into the finite-time-scaling ansatz for the spin-glass susceptibility, $\chi_{\mathrm{sg}}(t)t^{-\gamma/(z_s(T)\nu)}$ equal to a universal function of $t|T-T_{\mathrm{sg}}|^{-z_s(T)\nu}$; this is what turns time into a substitute for system size. The chiral-glass analog uses the same machinery with local chirality variables, and the mechanism that decides the transition is the crossing of the decreasing $z_s(T)$ with the constant $z_{\mathrm{cg}}$: above the crossing temperature chirality grows independently and freezes first, below it chirality follows spin dynamics. The scaling analysis is made robust by requiring the universal function to be independent of the fitting procedure, temperature range, and discarded initial steps.
Extended reading notes
Core claim
On the paper's own terms, the discovery is a temperature-dependent nonequilibrium dynamic exponent that behaves differently for spin and chiral correlations. The spin-glass correlation length grows as $\xi_{\mathrm{sg}}(t)\sim t^{1/z_s(T)}$ with $z_s(T)=z_{\mathrm{sg}}T_{\mathrm{sg}}/T$ throughout the paramagnetic and spin-glass phases; the chiral-glass correlation length in the XY and Heisenberg models instead grows with a constant exponent $z_{\mathrm{cg}}\simeq5.0$–$5.1$ above $T_{\mathrm{sg}}$ and only becomes temperature-dependent below it. Feeding this law into finite-time scaling yields $T_{\mathrm{sg}}=1.18(2)$, $\nu_{\mathrm{sg}}=1.75(10)$, $\gamma_{\mathrm{sg}}=3.4(2)$ for Ising; $T_{\mathrm{sg}}=0.435(15)$, $\nu_{\mathrm{sg}}=1.25(15)$, $\gamma_{\mathrm{sg}}=2.30(25)$ and $T_{\mathrm{cg}}=0.445(15)$, $\nu_{\mathrm{cg}}=1.05(10)$, $\gamma_{\mathrm{cg}}=1.65(20)$ for XY; and $T_{\mathrm{sg}}=0.18(1)$, $\nu_{\mathrm{sg}}=1.55(15)$, $\gamma_{\mathrm{sg}}=2.9(2)$ with $T_{\mathrm{cg}}=0.23(1)$, $\nu_{\mathrm{cg}}=0.70(15)$, $\gamma_{\mathrm{cg}}=0.65(15)$ for Heisenberg. The dynamic critical exponents at all five transitions agree within error bars ($z\simeq5.0$–$5.1$), which the paper reads as dynamic universality of 3D spin-glass transitions.
Load-bearing premise
The load-bearing premise is that the empirical law $z_s(T)=z_{\mathrm{sg}}T_{\mathrm{sg}}/T$, combined with the finite-time-scaling form built from it, describes the spin-glass susceptibility over the whole fitted temperature range with no important correction terms; the paper itself shows that switching to an alternative scaling variable changes $\nu_{\mathrm{sg}}$ substantially (from 1.75 to 2.18 in Ising and from 1.55 to 3.7 in Heisenberg) while $T_{\mathrm{sg}}$ and $\eta_{\mathrm{sg}}$ remain stable.
Editorial extensions
If this is right
- In the 3D Heisenberg spin glass there is a temperature window $T_{\mathrm{sg}}<T<T_{\mathrm{cg}}$ where chirality has frozen but spins have not, so chiral-glass order exists without spin-glass order.
- All the 3D spin-glass and chiral-glass transitions studied share the dynamic critical exponent $z\simeq5.1$, so a single dynamic universality class appears to control all of them.
- Finite-time-scaling analyses that assume a temperature-independent dynamic exponent will bias estimates of $\nu_{\mathrm{sg}}$ and $\gamma_{\mathrm{sg}}$, even when $T_{\mathrm{sg}}$ is stable; the temperature-dependent law must be built into the analysis.
- Because $z_s(T)=z_{\mathrm{sg}}T_{\mathrm{sg}}/T$ holds with no anomaly, high-temperature nonequilibrium relaxation data can be used to estimate the dynamic critical exponent at the transition.
- The observed separation in the Heisenberg model means that measurements or simulations that overlook the temperature dependence of the dynamic exponent will see a smeared, apparently simultaneous transition rather than the two distinct transitions.
Reading between the lines
- One can infer a sharper experimental signature: in the intermediate phase $T_{\mathrm{sg}}<T<T_{\mathrm{cg}}$, chiral correlations should keep growing at the constant rate $z_{\mathrm{cg}}\simeq5.1$ while spin correlations, governed by the temperature-dependent $z_s(T)$, change their growth rate with temperature; observing this differential growth would confirm the separation.
- The method also suggests that past equilibrium simulations reporting a simultaneous transition may have been seeing the short-scale spin-chirality coupling regime; reanalyzing those data with the temperature-dependent exponent may reveal the decoupling without longer runs.
- Because $T_{\mathrm{sg}}$ and $\eta_{\mathrm{sg}}$ are robust to the choice of scaling variable while $\nu_{\mathrm{sg}}$ and $\gamma_{\mathrm{sg}}$ are not, the transition temperature is the most reliable number to compare across models; for the chiral transition, no equivalent robustness check is reported, so a similar stability analysis would test whether $T_{\mathrm{cg}}=0.23$ is as solid.
- A natural extension is to compute $z$ for other strongly frustrated 3D models, such as gauge glasses or random-bond $O(n)$ models, to see whether $z\simeq5.1$ is truly a universal feature of 3D spin-glass dynamics or specific to the $\pm J$ family.
Original abstract
Nonequilibrium dynamics of the $\pm J$ Ising, the {\it XY}, and the Heisenberg spin-glass models are investigated in three dimensions. A nonequilibrium dynamic exponent is calculated from the dynamic correlation length. The spin-glass dynamic exponent continuously depends on the temperature. There is no anomaly at the critical temperature as is recently reported by Katzgraber and Campbell. On the other hand, the chiral-glass dynamic exponent takes a constant value above the spin-glass transition temperature ($T_\mathrm{sg}$), and becomes temperature-dependent below $T_\mathrm{sg}$.The finite-time scaling analyses on the spin- and the chiral-glass susceptibility are performed using the temperature dependence of the dynamic exponent. A difference of the spin- and the chiral-glass transition temperatures is resolved in the Heisenberg model. The dynamic critical exponent takes almost the same value for all transitions. It suggests that the spin-glass and the chiral-glass transitions in three dimensions are dynamically universal.
The compiled record
The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
The abstract, sentence by sentence
Nonequilibrium dynamics of the $\pm J$ Ising, the {\it XY}, and the Heisenberg spin-glass models are investigated in three dimensions. A nonequilibrium dynamic exponent is calculated from the dynamic correlation length. The spin-glass dynamic exponent continuously depends on the temperature. There is no anomaly at the critical temperature as is recently reported by Katzgraber and Campbell. On the other hand, the chiral-glass dynamic exponent takes a constant value above the spin-glass transition temperature ($T_\mathrm{sg}$), and becomes temperature-dependent below $T_\mathrm{sg}$.The finite-time scaling analyses on the spin- and the chiral-glass susceptibility are performed using the temperature dependence of the dynamic exponent. A difference of the spin- and the chiral-glass transition temperatures is resolved in the Heisenberg model. The dynamic critical exponent takes almost the same value for all transitions. It suggests that the spin-glass and the chiral-glass transitions in three dimensions are dynamically universal.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Nonequilibrium dynamics of the $\pm J$ Ising, the {\it XY}, and the Heisenberg spin-glass models are investigated in three dimensions. | contribution 0.93 | 0.89 | 0.03 | asserted 0.96 |
| s1 | A nonequilibrium dynamic exponent is calculated from the dynamic correlation length. | method 0.87 | 0.91 | 0.05 | asserted 1.00 |
| s2 | The spin-glass dynamic exponent continuously depends on the temperature. | result 1.00 | 0.85 | 0.24 | asserted 1.00 |
| s3 | There is no anomaly at the critical temperature as is recently reported by Katzgraber and Campbell. | result 1.00 | 0.63 | 0.13 | asserted 1.00 |
| s4 | On the other hand, the chiral-glass dynamic exponent takes a constant value above the spin-glass transition temperature ($T_\mathrm{sg}$), and becomes temperature-dependent below $T_\mathrm{sg}$.The finite-time scaling analyses on the spin- and the chiral-glass susceptibility are performed using the temperature dependence of the dynamic exponent. | result 0.96 | 0.86 | 0.15 | asserted 1.00 |
| s5 | A difference of the spin- and the chiral-glass transition temperatures is resolved in the Heisenberg model. | result 1.00 | 0.88 | 0.12 | asserted 1.00 |
| s6 | The dynamic critical exponent takes almost the same value for all transitions. | result 0.99 | 0.89 | 0.25 | asserted 0.90 |
| s7 | It suggests that the spin-glass and the chiral-glass transitions in three dimensions are dynamically universal. | interpretation 1.00 | 0.71 | 0.71 | hedged possibility 1.00 |
What was examined, and what was not
- arXiv listing abstract v2
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
Figures
Showing 15 of the paper's 16 figures. See all figures in the PDF ↗ Figures reproduced from arXiv:cond-mat/0603062 by the authors.
References
How the literature uses it
3 arXiv papers cite this in 14 sentences
Papers citing it · 10 sentences without a confident label
Uses its method or tool 1 paper
-
To estimate the critical temperature and some critical exponents of the 3D $±J$ XY model, we used the following dynamical scaling law for SG susceptibility [ref] : [equation] where [equation] [equation] [equation]
Dynamical scaling study for the estimation of dynamical exponent $z$ of three-dimensional XY spin glass model · Models and Observables for Estimating the Critical Temperature · 0.99
Compares with it 2 papers
-
To examine this point more closely, we compared our results with those of Ref. [ref] under conditions that mimic the earlier study: we imposed the same lower temperature cutoff ($T≥0.460$) and truncated our relaxation data at $1×10^5$ MCS, which was the maximum simulation time used in Ref. [ref] $.$
Dynamical scaling study for the estimation of dynamical exponent $z$ of three-dimensional XY spin glass model · Dynamical Scaling Analysis · 0.87
-
The estimated CG critical exponents are compatible with the ones reported in earlier literature on the 3D XY SG [ref] .
Monte Carlo simulations of the three-dimensional XY spin glass focusing on the chiral and the spin order · CRITICAL PROPERTIES · 0.99
Discussion
Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.