arXiv subjects

Ludovic Berthier

Publications and source records attributed to Ludovic Berthier.

At least 19 recordsLinked to original sources

arXiv2026

Identifying the relevant parameters in design strategies for stable glasses

A glass is conventionally obtained by cooling a bulk supercooled liquid through its glass transition temperature. The discovery of ultrastable glasses prepared using physical vapor deposition, together with the recent multiplication of numerical algorithms created to increase the stability of glasses, demonstrates the existence of a variety of strategies for designing glasses with different physical properties. This raises a broader question: which parameters most strongly govern the enhancement of glass stability? Existing computational strategies often produce highly stable glasses by optimizing certain physical properties through dynamical changes in particle diameters. We challenge the idea that these physical quantities are causally responsible for glass stability and suggest instead that diameter dynamics is the principal source of enhanced stability. To support our view, we introduce computational methods to optimize physical quantities without changing the particle diameters. Using the examples of enhanced hyperuniformity at large scale and local ordering at small scale, we design glass configurations with highly optimized values compared to bulk equilibrium states. However, these glasses do not show enhanced stability. The proposed physical quantities are correlated with glass stability, but are not causally responsible for ultrastability. These findings indicate that design rules for stable glasses should be reinterpreted in terms of the dynamical processes that generate stability, rather than the optimized physical quantities they target.

cond-mat.stat-mech↗

arXiv2026

How to quantify long-time rotational motion in molecular systems

We show that all existing methods quantifying rotational motion in molecular fluids eventually have severe limitations in systems undergoing complex rotational motion characterized by slow, heterogeneous, or intermittent dynamics. This impacts in particular the study of rotational dynamics in molecular supercooled liquids near their glass transition, as well as discussions of the decoupling between rotational and translational motion and violations of the Debye-Stokes-Einstein relation. We present a brief overview of existing methods and explain why none of them can accurately capture the evolution of rotational dynamics from a diffusive fluid to an arrested solid, thus resolving inconsistent literature results. We then introduce an empirical method that efficiently solves all issues. We benchmark our method devising a family of continuous time random walk models for rotational dynamics. Our method correctly quantifies the statistics of free and caged rotational motion, as well as non-Gaussian and non-Fickian rotational dynamics, and should allow a better characterization of dynamic heterogeneity in the rotational motion of supercooled molecular fluids.

cond-mat.stat-mech↗

arXiv2026

Glass and jamming transitions in a random organization model

We study a two-dimensional, off-lattice particle model introduced to describe absorbing phase transitions in driven non-Brownian suspensions. We numerically explore the $(\phi,\epsilon)$ phase diagram, where $\phi$ is the packing fraction and $\epsilon$ controls the amplitude of particle jumps. We use a binary mixture to suppress crystallization, which allows us to disentangle the model's distinct phase transitions between amorphous states. At large $\phi$, we find that the approach to the absorbing transition is preceded by a non-equilibrium glass transition to a non-diffusive amorphous state. This dynamic arrest makes the location of the critical absorbing transitions protocol-dependent. The $\epsilon \to 0$ end-point of the transition line defines a jamming transition whose location is shown to vary continuously with the preparation protocol, and cannot serve as a unique definition of random close packing. Near jamming, we observe a complex landscape and marginal stability, reminiscent of Gardner phases found in thermal glasses. The critical exponents characterizing packings at the jamming transition numerically agree with alternative approaches based on energy minimization, and with analytic predictions from mean-field replica theory. We analyze hyperuniformity in fluid and glass phases, where it emerges with qualitatively distinct signatures, and show that random organization dynamics does not determine the hyperuniformity observed in jammed packings, which is found to be non-universal. Our results show that random organization models share deep physical similarities with thermal soft-particle systems undergoing glass and jamming transitions, with little impact of the non-equilibrium nature of the microscopic dynamics on emerging physical properties.

cond-mat.stat-mech↗

arXiv2026

Direct Boltzmann inversion method from particle configurations at arbitrary state points

We introduce a direct Boltzmann inversion method to infer the interaction potential in particle systems using as input particle configurations generated at an arbitrary state point of the system. Unlike iterative Boltzmann inversion, the proposed method does not require performing a new Monte Carlo simulation at each step of the iteration process. It relies instead on enforcing consistency between two independent estimates of the pair correlation function, respectively obtained from interparticle distances and from pairwise forces. As a result, the approach is computationally inexpensive and straightforward to implement. Because it relies on the sole expression of interparticle forces, our method naturally applies to any state point, including when the density is large and alternative methods may fail. Here we present the basic principles of the method and benchmark its performance on a diverse set of test potentials studied using computer simulations. Practical aspects and detailed implementation of the method are also discussed. Owing to its simplicity and generality, the method should be broadly applicable, from the construction of coarse-grained interaction potentials to the inference of effective interactions in non-equilibrium systems.

cond-mat.stat-mech↗

arXiv2025

Boltzmann generators for amorphous particle systems

Sampling configurations in thermodynamic equilibrium is a long-standing challenge in statistical physics. Boltzmann generators address this problem by employing generative models to propose independent configurations, which are then reweighted via importance sampling using exact likelihood evaluations. Recent Boltzmann Generators based on continuous normalizing flows and flow matching have achieved significant success for particle systems and biomolecules. However, these approaches have not been extended to amorphous materials (glasses), for which equilibrium sampling is notoriously slow. Because of their disordered structure, the invariances and geometrical constraints of amorphous materials differ from those of crystals and biomolecules, preventing the direct use of existing generative models. Here, we develop Boltzmann Generators tailored to amorphous materials by building the required equivariances directly into Riemannian stochastic interpolants. Our framework incorporates periodic boundary conditions and particle symmetries using equivariant graph neural networks. Numerical experiments demonstrate that enforcing physical symmetries significantly improves the accuracy of Boltzmann Generators, but also reveal an intrinsic limitation of the continuous-flow formulation: accumulated numerical errors during likelihood integration break time-reversibility, compromising exact thermodynamic reweighting. These results reveal a fundamental challenge for continuous-flow generative models in statistical mechanics and call for alternative approaches that preserve exact thermodynamic consistency.

stat.ML↗

arXiv2025

Directed percolation transition to active turbulence driven by non-reciprocal forces

We numerically study the collective dynamics of dense particle assemblies driven by non-reciprocal pairwise forces of amplitude $\kappa$. At a critical value $\kappa_{\rm c}$, the system undergoes a dynamical phase transition from an absorbing state ($\kappa < \kappa_{\rm c}$) to a chaotic steady state ($\kappa > \kappa_{\rm c}$). The chaotic phase is marked by nontrivial spatiotemporal velocity correlations and mixing, reminiscent of active turbulence in self-propelled systems. The sharp onset of chaos shows critical scaling consistent with the universality class of directed percolation. We argue that this transition is generic to a broad class of locally-driven, dense disordered materials.

cond-mat.stat-mech↗

arXiv2025

Hyperuniform interfaces in non-equilibrium phase coexistence

We show that long-wavelength interfacial fluctuations are strongly suppressed in non-equilibrium phase coexistence between bulk hyperuniform systems. Using simulations of three distinct microscopic models, we demonstrate that hyperuniform interfaces are much smoother than equilibrium ones, with a universal reduction of height fluctuations at large scale. We derive a non-equilibrium interface equation from the field theory of the bulk order parameter, and predict a reduction in height fluctuations, $S_h(\boldsymbol k)\equiv \langle |h(\boldsymbol k)|^2\rangle\sim |\boldsymbol k|^{-1}$, in stark contrast to equilibrium capillary wave theory where $S_h(\boldsymbol k)\sim |\boldsymbol k|^{-2}$. Our results establish a new universality class for non-equilibrium interfaces, highlighting the fundamental role of suppressed bulk fluctuations in shaping interfacial dynamics far from equilibrium.

cond-mat.stat-mech↗

arXiv2025

Numerical investigation of the equilibrium Kauzmann transition in a two-dimensional atomistic glass

Dense liquids gradually transform into non-equilibrium amorphous solids as they pass through the experimental glass transition. Experimentally, ergodicity is lost because measurements are conducted within a finite time window. More than seventy years ago, Kauzmann posed a fundamental question: If experiments could run indefinitely, would there exist a critical temperature at which an ergodicity-breaking phase transition occurs? Random first-order transitions represent the modern theoretical framework for this idea, rigorously established in the mean-field limit of high-dimensional atomistic systems and several idealized physical models. However, achieving theoretical understanding in finite dimensions is challenging, while experimental and numerical limitations on accessible timescales hinder direct observation of the putative Kauzmann transition. Here, we overcome this longstanding barrier by developing a computational strategy that combines three advanced Monte Carlo methods to access the equilibrium thermodynamic properties of a two-dimensional atomistic glass-former down to zero temperature across a range of system sizes. This enables us to directly measure thermodynamic and structural observables that provide unambiguous evidence that the system undergoes a Kauzmann transition at a temperature that vanishes in the thermodynamic limit. This transition is towards an ideal glass state characterized by a complex energy landscape with a hierarchical organization of low-lying states. Our results are the first demonstration that computer simulations can fully probe the statistical mechanics of the bulk transition to a non-ergodic glass state. We anticipate that our study will serve as a foundation for future simulation work on larger systems, three-dimensional materials, and more complex glass-forming models to fully elucidate the nature of the glass state of matter.

cond-mat.soft↗

arXiv2025

Molecular motion at the experimental glass transition

We propose a novel computational strategy to study the glass transition of molecular fluids. Our approach combines the construction of simple yet realistic models with the development of Monte Carlo algorithms to accelerate equilibration and sampling. Inspired by the well-studied ortho-terphenyl glass-former, we construct a molecular model with an analogous triangular geometry and construct a `flip' Monte Carlo algorithm. We demonstrate that the flip Monte Carlo algorithm achieves a sampling speedup of about $10^9$ at the experimental glass transition temperature $T_g$. This allows us to systematically analyze the equilibrium structure and molecular dynamics of the model over a temperature regime previously inaccessible. We carefully compare the observed physical behavior to earlier studies that used atomistic models. In particular, we find that the glass fragility and the departure from the Stokes-Einstein relation are much closer to experimental observations. We characterize the development and temperature evolution of spatial correlations in the relaxation dynamics, using both orientational and translational degrees of freedom. Excess wings emerge at intermediate frequencies in dynamic rotational spectra, and we directly visualize the corresponding molecular motion near $T_g$. Our approach can be generalized to a \rev{broad range of molecular geometries and paves the way to a deeper} understanding of how molecular details may affect more universal physical aspects characterizing molecular liquids approaching their glass transition.

cond-mat.stat-mech↗

arXiv2025

Scaling the glassy dynamics of active particles: Tunable fragility and reentrance

Understanding the influence of activity on dense amorphous assemblies is crucial for biological processes such as wound healing, embryogenesis, or cancer progression. Here, we study the effect of self-propulsion forces of amplitude $f_0$ and persistence time $\tau_p$ in dense assemblies of soft repulsive particles by simulating a model particle system that interpolates between particulate active matter and biological tissues. We identify the fluid and glass phases of the three-dimensional phase diagram obtained by varying $f_0$, $\tau_p$, and the packing fraction $\phi$. The morphology of the phase diagram directly accounts for a non-monotonic evolution of the relaxation time with $\tau_p$, which is a direct consequence of the crossover in the dominant relaxation mechanism, from glassy to jamming. A second major consequence is the evolution of the glassy dynamics from sub-Arrhenius to super-Arrhenius. We show that this tunable glass fragility extends to active systems analogous observations reported for passive particles. This analogy allows us to apply a dynamic scaling analysis proposed for the passive case, in order to account for our results for active systems. Finally, we discuss similarities and differences between our results and recent findings in the context of computational models of biological tissues.

cond-mat.soft↗

arXiv2025

Bacterial Glass Transition

Bacterial assemblies exhibit rich collective behaviors that control their biological functions, making them a relevant object of study from an active matter physics perspective. Dense bacterial suspensions self-organize into distinct physical phases with intriguing dynamical properties. Here, we study dense two-dimensional films of swimming bacteria using advanced imaging techniques and machine learning. By varying density, we uncover a bacterial glass transition, a direct active matter analogue of equilibrium glass transitions in colloidal and molecular fluids. The transition is marked by a dramatic slowdown of dynamics with minimal structural change. Strong dynamic heterogeneity emerges in space and time, leading to an anomalous violation of the Stokes-Einstein relation and a growing dynamic correlation length, universally observed across five bacterial strains. Our results establish that bacterial colonies exhibit glassy dynamics, but their living, active nature gives them unique properties, paving the way for new research regarding how non-equilibrium physics impacts biology.

cond-mat.soft↗

arXiv2025

Athermal creep deformation of ultrastable amorphous solids

We numerically investigate the athermal creep deformation of amorphous materials having a wide range of stability. The imposed shear stress serves as the control parameter, allowing us to examine the time-dependent transient response through both the macroscopic strain and microscopic observables. Least stable samples exhibit monotonicity in the transient strain rate versus time, while more stable samples display a pronounced non-monotonic S-shaped curve, corresponding to failure by sharp shear band formation. We identify a diverging timescale associated with the fluidization process and extract the corresponding critical exponents. Our results are compared with predictions from existing scaling theories relevant to soft matter systems. The numerical findings for stable, brittle-like materials represent a challenge for theoretical descriptions. We monitor the microscopic initiation of shear bands during creep responses. Our study encompasses creep deformation across a variety of materials ranging from ductile soft matter to brittle metallic and oxide glasses, all within the same numerical framework.

cond-mat.soft↗

arXiv2025

Irreversible swap algorithms for soft sphere glasses

We extend to soft repulsive interaction potentials a recently proposed irreversible swap algorithm originally designed for polydisperse hard spheres. The original algorithm performs rejection-free, irreversible, collective swap moves. We show that event-driven cluster updates of particle diameters can also be performed in continuous potentials by introducing a factorised Metropolis probability. However, the Metropolis factorisation needed to deal with continuous potentials decreases the efficiency of the algorithm and mitigates the benefits of breaking detailed balance. This leads us to propose another irreversible swap algorithm using the standard Metropolis probability that accelerates the relaxation of soft sphere glasses at low temperatures, compared to the original swap algorithm. We apply these efficient swap algorithms to produce very stable inherent structures with vibrational density of states lacking the quasi-localised excitations observed in conventional glasses.

cond-mat.soft↗

arXiv2024

Initial Stages of Rejuvenation of Vapor-Deposited Glasses during Isothermal Annealing: Contrast Between Experiment and Simulation

Physical vapor deposition can prepare organic glasses with high kinetic stability. When heated, these glassy solids slowly transform into the supercooled liquid in a process known as rejuvenation. In this study, we anneal vapor-deposited glasses of methyl-m-toluate (MMT) for six hours at 0.98Tg to observe rejuvenation using dielectric spectroscopy. Glasses of moderate stability exhibited partial or full rejuvenation in six hours. For highly stable glasses, prepared at substrate temperatures of 0.85Tg and 0.80Tg, the six-hour annealing time is ~2% of the estimated transformation time, and no change in the onset temperature for the {\alpha} relaxation process was observed, as expected. Surprisingly, for these highly stable glasses, annealing resulted in significant increases in the storage component of the dielectric susceptibility, without corresponding increases in the loss component. These changes are interpreted to indicate that short-term annealing rejuvenates a high frequency relaxation (e.g., the boson peak) within the stable glass. We compare these results to computer simulations of the rejuvenation of highly stable glasses generated by the swap Monte Carlo algorithm. The in silico glasses, in contrast to the experiment, show no evidence of rejuvenation within the stable glass at times shorter than the alpha relaxation process.

cond-mat.soft↗

arXiv2024

Clever algorithms for glasses work by time reparametrization

The ultraslow dynamics of glass-formers has been explained by two views considered as mutually exclusive: one invokes locally hindered mobility, the other rests on the complexity of the configuration space. Here we demonstrate that the evolution responds strongly to the details of the dynamics by changing the speed of time-flow: it has time-reparametrization softness. This finding reconciles both views: while local constraints reparametrize the flow of time, the global landscape determines relationships between different correlations at the same times. We show that modern algorithms developed to accelerate the relaxation to equilibrium act by changing the time reparametrization. Their success thus relies on their ability to exploit reparametrization softness. We conjecture that these results extend beyond the realm of glasses to the optimization of more general constraint satisfaction problems and to broader classes of algorithms.

cond-mat.dis-nn↗

arXiv2024

Characterising the slow dynamics of the swap Monte Carlo algorithm

The swap Monte Carlo algorithm introduces non-physical dynamic rules to accelerate the exploration of the configuration space of supercooled liquids. Its success raises deep questions regarding the nature and physical origin of the slow dynamics of dense liquids, and how it is affected by swap moves. We provide a detailed analysis of the slow dynamics generated by the swap Monte Carlo algorithm at very low temperatures in two glass-forming models. We find that the slowing down of the swap dynamics is qualitatively distinct from its local Monte Carlo counterpart, with considerably suppressed dynamic heterogeneity both at single-particle and collective levels. Our results suggest that local kinetic constraints are drastically reduced by swap moves, leading to nearly Gaussian and diffusive dynamics and weakly growing dynamic correlation lengthscales. The comparison between static and dynamic fluctuations shows that swap Monte Carlo is a nearly optimal local equilibrium algorithm, suggesting that further progress should necessarily involve collective or driven algorithms.

cond-mat.soft↗

arXiv2024

Monte Carlo simulations of glass-forming liquids beyond Metropolis

Monte Carlo simulations are widely employed to measure the physical properties of glass-forming liquids in thermal equilibrium. Combined with local Monte Carlo moves, the Metropolis algorithm can also be used to simulate the relaxation dynamics, thus offering an efficient alternative to molecular dynamics. Monte Carlo simulations are however more versatile, because carefully designed Monte Carlo algorithms can more efficiently sample the rugged free energy landscape characteristic of glassy systems. After a brief overview of Monte Carlo studies of glass-formers, we define and implement a series of Monte Carlo algorithms in a three-dimensional model of polydisperse hard spheres. We show that the standard local Metropolis algorithm is the slowest, and that implementing collective moves or breaking detailed balance enhances the efficiency of the Monte Carlo sampling. We use time correlation functions to provide a microscopic interpretation of these observations. Seventy years after its invention, the Monte Carlo method remains the most efficient and versatile tool to compute low-temperatures properties in supercooled liquids.

cond-mat.stat-mech↗

arXiv2024

Normalizing flows as an enhanced sampling method for atomistic supercooled liquids

Normalizing flows can transform a simple prior probability distribution into a more complex target distribution. Here, we evaluate the ability and efficiency of generative machine learning methods to sample the Boltzmann distribution of an atomistic model for glass-forming liquids. This is a notoriously difficult task, as it amounts to ergodically exploring the complex free energy landscape of a disordered and frustrated many-body system. We optimize a normalizing flow model to successfully transform high-temperature configurations of a dense liquid into low-temperature ones, near the glass transition. We perform a detailed comparative analysis with established enhanced sampling techniques developed in the physics literature to assess and rank the performance of normalizing flows against state-of-the-art algorithms. We demonstrate that machine learning methods are very promising, showing a large speedup over conventional molecular dynamics. Normalizing flows show performances comparable to parallel tempering and population annealing, while still falling far behind the swap Monte Carlo algorithm. Our study highlights the potential of generative machine learning models in scientific computing for complex systems, but also points to some of its current limitations and the need for further improvement.

cond-mat.soft↗