4.8 Article

Origin of Ultrastability in Vapor-Deposited Glasses

期刊

PHYSICAL REVIEW LETTERS
卷 119, 期 18, 页码 -

出版社

AMER PHYSICAL SOC
DOI: 10.1103/PhysRevLett.119.188002

关键词

-

资金

  1. NSF [DMR-1608086]
  2. Simons Foundation [454933, 454937, 454955]

向作者/读者索取更多资源

Glass films created by vapor-depositing molecules onto a substrate can exhibit properties similar to those of ordinary glasses aged for thousands of years. It is believed that enhanced surface mobility is the mechanism that allows vapor deposition to create such exceptional glasses, but it is unclear how this effect is related to the final state of the film. Here we use molecular dynamics simulations to model vapor deposition and an efficient Monte Carlo algorithm to determine the deposition rate needed to create ultrastable glassy films. We obtain a scaling relation that quantitatively captures the efficiency gain of vapor deposition over bulk annealing, and demonstrates that surface relaxation plays the same role in the formation of vapor-deposited glasses as bulk relaxation does in ordinary glass formation.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.8
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

Article Chemistry, Physical

Supervised perceptron learning vs unsupervised Hebbian unlearning: Approaching optimal memory retrieval in Hopfield-like networks

Marco Benedetti, Enrico Ventura, Enzo Marinari, Giancarlo Ruocco, Francesco Zamponi

Summary: This study compares the Hebbian unlearning algorithm with a symmetric perceptron algorithm and finds that they have similar properties in terms of basin size and convergence region. The study also proposes a geometric interpretation of Hebbian unlearning and explores its potential applications in memory storage in materials.

JOURNAL OF CHEMICAL PHYSICS (2022)

Article Biochemistry & Molecular Biology

Modeling Sequence-Space Exploration and Emergence of Epistatic Signals in Protein Evolution

Matteo Bisardi, Juan Rodriguez-Rivas, Francesco Zamponi, Martin Weigt

Summary: The study proposes stochastic models of experimental protein evolution based on reconstructed data-driven fitness landscapes, predicting important features of experimentally evolved sequence libraries. It showcases the potential of the approach in determining protein structure from epistasis signals in experimental sequence libraries, offering a quantitative explanation for outcomes of recent experiments and enabling forecasts for future evolution experiments.

MOLECULAR BIOLOGY AND EVOLUTION (2022)

Article Chemistry, Physical

The dimensional evolution of structure and dynamics in hard sphere liquids

Patrick Charbonneau, Yi Hu, Joyjit Kundu, Peter K. Morse

Summary: The formulation of mean-field infinite-dimensional solution of hard sphere glasses is a significant advancement in theoretical physics. However, its relevance in understanding low-dimensional glass-forming liquids is still unclear. By revisiting simulation techniques and leveraging standard liquid-state theory, the study of hard sphere liquids up to d = 13 has become possible, helping to bridge the gap between different approaches.

JOURNAL OF CHEMICAL PHYSICS (2022)

Correction Chemistry, Physical

The dimensional evolution of structure and dynamics in hard sphere liquids (vol 156, 134502, 2022)

Patrick Charbonneau, Yi Hu, Joyjit Kundu, Peter K. Morse

JOURNAL OF CHEMICAL PHYSICS (2022)

Article Physics, Multidisciplinary

Local Dynamical Heterogeneity in Simple Glass Formers

Giulio Biroli, Patrick Charbonneau, Giampaolo Folena, Yi Hu, Francesco Zamponi

Summary: This study reveals that the local dynamical fluctuations of particles in glass-forming models display divergent behavior around the glass transition due to nontrivial correlations between displacements along different directions. This phenomenon leads to the emergence of a divergent non-Gaussian parameter and provides a first-principle explanation for its growth around the glass transition without relying on multiparticle correlations.

PHYSICAL REVIEW LETTERS (2022)

Correction Multidisciplinary Sciences

Efficient generative modeling of protein sequences using simple autoregressive models (vol 12, pg 5800, 2021)

Jeanne Trinquier, Guido Uguzzoni, Andrea Pagnani, Francesco Zamponi, Martin Weigt

NATURE COMMUNICATIONS (2022)

Article Physics, Multidisciplinary

Gradient descent dynamics and the jamming transition in infinite dimensions

Alessandro Manacorda, Francesco Zamponi

Summary: This article analyzes the gradient descent dynamics in complex energy landscapes, focusing on the case of soft repulsive particles in infinite spatial dimension. The study shows that the dynamics undergoes a jamming transition, characterized by the elimination of particle overlaps at low density and the persistence of overlaps at high density. Self-consistent dynamical equations are derived via mean field theory, and some progress towards their solutions is presented. The article also examines the jamming transition in the random Lorentz gas and discusses its analogy with the capacity transition in supervised learning.

JOURNAL OF PHYSICS A-MATHEMATICAL AND THEORETICAL (2022)

Article Chemistry, Physical

Microscopic observation of two-level systems in a metallic glass model

Felix C. C. Mocanu, Ludovic Berthier, Simone Ciarella, Dmytro Khomenko, David R. R. Reichman, Camille Scalliet, Francesco Zamponi

Summary: The low-temperature behavior of amorphous solids is influenced by localized tunneling defects in the potential energy landscape. Computational models are used to study these defects and recent simulation work has generated stable glassy configurations for metallic glass models using the swap Monte Carlo algorithm. In this study, the glassy metabasins of a widely used model of metallic glasses are explored, and the depletion of tunneling defects with increased glass stability is observed.

JOURNAL OF CHEMICAL PHYSICS (2023)

Article Physics, Multidisciplinary

Finite-Disorder Critical Point in the Yielding Transition of Elastoplastic Models

Saverio Rossi, Giulio Biroli, Misaki Ozawa, Gilles Tarjus, Francesco Zamponi

Summary: This study provides numerical evidence for the existence of a finite-disorder critical point that separates brittle and ductile yielding. The critical exponents in 2D and 3D are estimated.

PHYSICAL REVIEW LETTERS (2022)

Article Multidisciplinary Sciences

Creating bulk ultrastable glasses by random particle bonding

Misaki Ozawa, Yasutaka Iwashita, Walter Kob, Francesco Zamponi

Summary: A recent breakthrough in glass science involves the synthesis of ultrastable glasses using physical vapor deposition techniques. These glasses exhibit increased stability in terms of thermodynamics, kinetics, and mechanics, with significant implications for both fundamental research and practical applications. However, the current deposition technique is limited to specific types of glass-formers and can only produce thin film samples. In this study, a novel approach using random particle bonding is proposed to generate ultrastable glassy configurations in bulk materials. Computer simulations show the effectiveness of this method, which can be applied to various molecular and soft matter systems, offering great potential for the design of a wide range of ultrastable glasses.

NATURE COMMUNICATIONS (2023)

Article Computer Science, Artificial Intelligence

Machine-learning-assisted Monte Carlo fails at sampling computationally hard problems

Simone Ciarella, Jeanne Trinquier, Martin Weigt, Francesco Zamponi

Summary: Several strategies using machine learning tools have been proposed to improve Monte Carlo sampling efficiency, but they are challenged by a class of problems that are exponentially hard to sample at low temperatures using conventional local Monte Carlo. We specifically studied the antiferromagnetic Potts model on a random graph, which reduces to the coloring of random graphs at zero temperature. We tested several machine-learning-assisted Monte Carlo approaches and found that they all failed. Our work provides important benchmarks for future proposals of smart sampling algorithms.

MACHINE LEARNING-SCIENCE AND TECHNOLOGY (2023)

Article Physics, Fluids & Plasmas

Equilibrium fluctuations in mean-field disordered models

Giampaolo Folena, Giulio Biroli, Patrick Charbonneau, Yi Hu, Francesco Zamponi

Summary: This study investigates mean-field models of glasses using the replica method, revealing the fluctuation characteristics under thermodynamic limits. It analyzes the heterogeneous fluctuations between different states and achieves good quantitative agreement through numerical simulations.

PHYSICAL REVIEW E (2022)

Article Chemistry, Physical

Supervised perceptron learning vs unsupervised Hebbian unlearning: Approaching optimal memory retrieval in Hopfield-like networks

Marco Benedetti, Enrico Ventura, Enzo Marinari, Giancarlo Ruocco, Francesco Zamponi

Summary: This study compared the Hebbian unlearning algorithm with a supervised algorithm in neural networks, finding similarities in stability of stored memories and learning paths. The proposed geometric interpretation explained the optimal performance of the Hebbian unlearning algorithm.

JOURNAL OF CHEMICAL PHYSICS (2022)

暂无数据