Role of Structure and Entropy in Determining Differences in Dynamics for Glass Formers with Different Interaction Potentials

Atreyee Banerjee, S. Sengupta, Srikanth Sastry, Sarika Maitra BhattacharyyaPhysical 2014 · DOI · arXiv

Abstract

We present a study of two model liquids with different interaction potentials, exhibiting similar structure but significantly different dynamics at low temperatures. By evaluating the configurational entropy, we show that the differences in the dynamics of these systems can be understood in terms of their thermodynamic differences. Analyzing their structure, we demonstrate that differences in pair correlation functions between the two systems, through their contribution to the entropy, dominate the differences in their dynamics, and indeed overestimate the differences. Including the contribution of higher order structural correlations to the entropy leads to smaller estimates for the relaxation times, as well as smaller differences between the two studied systems.

Introduction

A central question in the physics of glass-forming liquids is the relationship between a liquid's static structure and its dynamic relaxation. Traditionally, the WCA (Weeks-Chandler-Andersen) theory posited that the repulsive part of the interatomic potential primarily determines the structure, with attractive forces acting as a minor perturbation. However, recent studies have revealed a paradox: a Lennard-Jones (LJ) liquid (with attractions) and its corresponding WCA liquid (purely repulsive) exhibit nearly identical pair-wise structures but vastly different dynamics as they cool. Specifically, the LJ liquid's relaxation time diverges much more rapidly than that of the WCA liquid. This study investigates whether thermodynamics, specifically configurational entropy, can resolve this discrepancy.

Methodology

The researchers employed molecular dynamics simulations of the Kob-Andersen binary mixture at a fixed density ($\rho = 1.2$). They compared the standard LJ system with a truncated WCA version. Key metrics included:

  • Relaxation Time ($\tau$): Measured via the self-intermediate scattering function.
  • Configurational Entropy ($S_c$): Calculated as the difference between the total liquid entropy and the vibrational entropy of the corresponding inherent structures.
  • Entropy Decomposition: The excess entropy was decomposed into a two-body term ($S_2$), derived from radial distribution functions, and a residual multiparticle entropy (RMPE, $\Delta S = S_{ex} - S_2$).
  • Adam-Gibbs (AG) Relation: The validity of $\tau(T) = \tau_0 \exp(A / T S_c)$ was tested for both systems.

Results and Discussion

The Success of the Adam-Gibbs Relation

The study confirms that the divergence in relaxation times between the LJ and WCA systems is fundamentally linked to their thermodynamic differences. As shown in Figure 1, while the relaxation times $\tau$ diverge differently with temperature (panel a), they collapse onto a single linear trend when plotted against $1/TS_c$ (panel d), validating the Adam-Gibbs relation for both potentials.

Figure 1: (a) Relaxation time $\tau$ vs $1/T$. (b) Vibrational entropy $S_{vib}$ vs $T$. (c) $TS_c$ vs $T/T_K$. (d) Adam-Gibbs plot of $\log \tau$ vs $1/TS_c$.

The Two-Body Entropy Paradox

A surprising finding is that the two-body structure ($S_2$) actually overestimates the difference in dynamics. If one were to predict relaxation times based solely on pair correlations, the LJ system would appear even more sluggish relative to the WCA system than it is in reality.

As illustrated in Figure 4, the relaxation times predicted by the two-body entropy ($\tau_2^{AG}$) show a much larger gap between LJ and WCA than the actual measured $\tau$. The inclusion of higher-order correlations through the full configurational entropy $S_c$ is necessary to accurately capture the dynamics. The residual multiparticle entropy (RMPE) is positive at low temperatures, effectively "speeding up" the dynamics compared to the two-body prediction.

Figure 4: Relaxation times $\tau$ compared with predictions from the Adam-Gibbs relation using full configurational entropy and two-body entropy. The inset shows the ratios of these times between the LJ and WCA systems.

Conclusion

This research demonstrates that the "mysterious" dynamic differences between structurally similar glass formers are quantitatively explained by their configurational entropy. While pair-wise structures are similar, subtle differences in their thermodynamics—and crucially, the role of many-body correlations—dictate their relaxation behavior. Future work should explore these relationships across varying densities and investigate the specific geometric motifs responsible for the higher-order entropy contributions.

Summary updated

Abstract

We present a study of two model liquids with different interaction potentials, exhibiting similar structure but significantly different dynamics at low temperatures. By evaluating the configurational entropy, we show that the differences in the dynamics of these systems can be understood in terms of their thermodynamic differences. Analyzing their structure, we demonstrate that differences in pair correlation functions between the two systems, through their contribution to the entropy, dominate the differences in their dynamics, and indeed overestimate the differences. Including the contribution of higher order structural correlations to the entropy leads to smaller estimates for the relaxation times, as well as smaller differences between the two studied systems.
BibTeX
@inproceedings{banerjee2014role,
  author = {Atreyee Banerjee and S. Sengupta and Srikanth Sastry and Sarika Maitra Bhattacharyya},
  title = {{Role of Structure and Entropy in Determining Differences in Dynamics for Glass Formers with Different Interaction Potentials}},
  booktitle = {Physical},
  year = {2014},
  doi = {10.1103/physrevlett.113.225701},
}
Figure