arXiv subjects

Julien Lam

Publications and source records attributed to Julien Lam.

At least 19 recordsLinked to original sources

arXiv2026

Deciphering borophene growth pathways with data-driven simulations

Deterministic synthesis of borophene remains challenging because many polymorphs compete during nucleation and growth. Here we combine a reactive machine-learned interatomic potential with grand-canonical Monte Carlo simulations and data-driven structural classification to track borophene formation from early nuclei to extended layers on Ag(111) and Ag(100). We build temperature-pressure substrate growth maps and resolve how vacancy motifs, phase intermixing and seed structure govern polymorph selection. The simulations reproduce key experimental trends, including the prevalence of $\beta_{12}$/$\chi_3$ phases and their temperature-dependent competition, while revealing kinetic pathways that connect metastable nuclei to long-range order. We identify conditions that suppress competing motifs and promote targeted phases, providing actionable synthesis windows. These results establish a predictive framework for directing borophene growth and, more broadly, for controlling polymorphism in low-dimensional materials by coupling atomistic simulation with machine-learning-enabled phase recognition.

cond-mat.mtrl-sci↗

arXiv2025

Growth driven phase transitions in Zinc Oxide nanoparticles through machine-learning assisted simulations

This study investigates the formation of zinc oxide (ZnO) nanoparticles, a material of significant technological interest with complex structural properties, through atom-by-atom deposition modeling a process common in bottom-up synthesis. Our findings demonstrate that, although the body-centered tetragonal (BCT) structure is thermodynamically stable at equilibrium for small particle sizes, the deposition process induces a crystal-to-crystal phase transition into the more stable wurtzite (WRZ) phase. This transformation is facilitated by a specific redistribution of the nanoparticle ions, which effectively compensates the emerging polar facets at the moment of transition. These insights offer a deeper understanding of oxide nanoparticle formation, which should ultimately help the design of materials with targeted structural features.

cond-mat.mtrl-sci↗

arXiv2025

Nonclassical Nucleation Pathways in Liquid Condensation Revealed by Simulation and Theory

Using state-of-the-art rare-event sampling simulations, we precisely characterize the nucleation of liquid droplets from a supersaturated Lennard-Jones gas and uncover a key physical feature: critical clusters nucleate with a density that differs substantially from that of the macroscopic equilibrium liquid. Our atomistic simulations also reveal a nonclassical nucleation pathway showing simultaneous growth and densification in liquid condensation. We then exploit these insights to develop a two-variable nucleation theory, in which the cluster density is allowed to vary. Our accessible model based on the capillary approximation is able to quantitatively retrieve the numerical results in nucleation rate and critical cluster properties over a large range of supersaturation. Remarkably, the two-variable model successfully captures the observed nucleation pathway. The effectiveness of this integrated numerical and theoretical framework demonstrates that the cluster density is a decisive variable in nucleation, highlighting the limitations of the single-variable description while offering a robust foundation for its refinement.

physics.chem-ph↗

arXiv2024

Competing nucleation pathways in nanocrystal formation

Despite numerous efforts from numerical approaches to complement experimental measurements, several fundamental challenges have still hindered one's ability to truly provide an atomistic picture of the nucleation process in nanocrystals. Among them, our study resolves three obstacles: (1) Machine-learning force fields including long-range interactions able to capture the finesse of the underlying atomic interactions, (2) Data-driven characterization of the local ordering in a complex structural landscape associated with several crystal polymorphs and (3) Comparing results from a large range of temperatures using both brute-force and rare-event sampling. Altogether, our simulation strategy has allowed us to study zinc oxide crystallization from nano-droplet melt. Remarkably, our results show that different nucleation pathways compete depending on the investigated degree of supercooling.

cond-mat.stat-mech↗

arXiv2024

Exploring the formation of gold/silver nanoalloys with gas-phase synthesis and machine-learning assisted simulations

While nanoalloys are of paramount scientific and practical interests, the main processes leading to their formation are still poorly understood. Key structural features in the alloy systems, including crystal phase, chemical ordering, and morphology, are challenging to control at the nanoscale, making it difficult to transfer their usage to industrial applications. In this contribution, we focus on the gold/silver system that has two of the most prevalent noble metals, and combine experiments with simulations to uncover the formation mechanisms at the atomic-level. Nanoparticles are produced using state-of-the-art inert-gas aggregation source and analyzed using transmission electron microscopy and energy-dispersive x-ray spectroscopy. Machine-learning-assisted molecular dynamics simulations are employed to model the crystallization process from liquid droplets to nanocrystals. Our study finds a preponderance of nanoparticles with five-fold symmetric morphology, including icosahedron and decahedron which is consistent with previous results on mono-metallic nanoparticles. However, we observe that gold atoms, rather than silver atoms, segregate at the surface of the obtained nanoparticles for all the considered alloy compositions. These segregation tendencies are in contrast to previous studies and have consequences on the crystallization dynamics and the subsequent crystal ordering. We finally show that the underpinnings of this surprising segregation dynamics is due to charge transfer and electrostatic interactions rather than surface energy considerations.

cond-mat.mtrl-sci↗

arXiv2024

Harvesting nucleating structures in nanoparticle crystallization: The example of gold, silver and iron

The thermodynamics and kinetics of crystallization of nanoparticles, as opposed to bulk phases, may be influenced by surface and size effects. We investigate the importance of such factors in the crystallization process of gold, silver, and iron nanodroplets using numerical simulations in the form of molecular dynamics combined with path sampling. This modeling strategy is targeted at obtaining representative ensembles of structures located at the transition state of the crystallization process. A structural analysis of the transition state ensembles reveals that both the average size and location of the critical nucleation cluster are influenced by surface and nanoscale size effets. Furthermore, we also show that transition state structures in smaller nanodroplets exhibit a more ordered liquid phase, and differentiating between a well-ordered critical cluster and its surrounding disordered liquid phase becomes less evident. All in all, these findings demonstrate that crystallization mechanisms in nanoparticles go beyond the assumptions of classical nucleation theory.

cond-mat.stat-mech↗

arXiv2023

A critical comparison of general-purpose collective variables for crystal nucleation

The nucleation of crystals is a prominent phenomenon in science and technology that still lacks a full atomic-scale understanding. Much work has been devoted to identifying order parameters able to track the process, from the inception of early nuclei to their maturing to critical size until growth of an extended crystal. We critically assess and compare two powerful distance-based collective variables, an effective entropy derived from liquid state theory and the path variable based on permutation invariant vectors using the Kob-Andersen binary mixture and a combination of enhanced-sampling techniques. Our findings reveal a comparable ability to drive nucleation when a bias potential is applied, and comparable free-energy barriers and structural features. Yet, we also found an imperfect correlation with the committor probability on the barrier top which was bypassed by changing the order parameter definition.

physics.comp-ph↗

arXiv2021

Non-classical nucleation of zinc oxide from a physically-motivated machine-learning approach

Observing non-classical nucleation pathways remains challenging in simulations of complex materials with technological interests. This is because it requires very accurate force fields that can capture the whole complexity of their underlying interatomic interactions and an advanced structural analysis. Here, we first report the construction of a machine-learning force field for zinc oxide interactions using the Physical LassoLars Interaction Potentials approach which allows us to be predictive even for untrained structures. Then, we carried out freezing simulations from a liquid and observed the crystal formation with atomistic precision. Our results, which are analyzed using a data-driven approach based on bond order parameters, demonstrate the presence of both prenucleation clusters and two-step nucleation scenarios thus retrieving seminal predictions of non-classical nucleation pathways made on much simpler models.

cond-mat.mtrl-sci↗

arXiv2021

Strong and fast rising pressure waves emitted by plasmonic vapor nanobubbles

Plasmonic vapour nanobubbles are currently considered for a wide variety of applications ranging from solar energy harvesting and photoacoustic imaging to nanoparticle-assisted cancer therapy. Yet, due their small size and unstable nature, their generation and consequences remain difficult to characterize. Here, building on a phase-field model, we report on the existence of strong pressure waves that are emitted when vapor nanobubbles first form around a laser-heated nanoparticle immersed in water, and subsequently after bubble rebound. These effects are strongest when the fluid is locally brought high in its supercritical state, which may be realized with a short laser pulse. Because of the highly out-of-equilibrium nature of nanobubble generation, the waves combine a high pressure peak with a fast pressure rising time, and propagate in water over micron distances, opening the way to induce spatially and temporally localized damage. Discussing the consequences on biological cell membranes, we conclude that acoustic-mediated perforation is more efficient than nanobubble expansion to breach membrane. Our findings should serve as guide for optimizing the thermoacoustic conversion efficiency of plasmonic vapor nanobubbles.

physics.optics↗

arXiv2020

Infrared spectra of neutral polycyclic aromatic hydrocarbons by machine learning

The Interest in polycyclic aromatic hydrocarbons (PAHs) spans numerous fields and infrared spectroscopy is usually the method of choice to disentangle their molecular structure. In order to compute vibrational frequencies, numerous theoretical studies employ either quantum calculation methods, or empirical potentials, but it remains difficult to combine the accuracy of the first approach with the computational cost of the second. In this work, we employed Machine Learning techniques to develop a potential energy surface and a dipole mapping based on an artificial neural network (ANN) architecture. Altogether, while trained on only 11 small PAH molecules, the obtained ANNs are able to retrieve the infrared spectra of those small molecules, but more importantly of 8 large PAHs different from the training set, thus demonstrating the transferability of our approach.

physics.chem-ph↗

arXiv2020

Out-of-equilibrium Polymorph Selection in Nanoparticle Freezing

The ability to design synthesis processes that are out of equilibrium has opened the possibility of creating nanomaterials with remarkable physico-chemical properties, choosing from a much richer palette of possible atomic architectures compared to equilibrium processes in extended systems. In this work, we employ atomistic simulations to demonstrate how to control polymorph selection via cooling rate during nanoparticle freezing in the case of Ni$_3$Al, a material with a rich structural landscape. State-of-the-art free-energy calculations allow to rationalize the complex nucleation process, discovering a switch between two kinetic pathways, yielding the equilibrium structure at room temperature and an alternative metastable one at higher temperature. Our findings address the key challenge in the synthesis of nano-alloys for technological applications, i.e., rationally exploiting the competition between kinetics and thermodynamics by designing a treatment history that forces the system into desirable metastable states.

cond-mat.mtrl-sci↗

arXiv2020

Long-wavelength density fluctuations as nucleation precursors

Recent theories of nucleation that go beyond Classical Nucleation Theory predict that diffusion-limited nucleation of both liquid droplets and of crystals from a low-density vapor (or weak solution) begins with long-wavelength density fluctuations. This means that in the early stages of nucleation, 'clusters' can have low density but large spatial extent, which is at odds with the classical picture of arbitrarily small clusters of the condensed phase. We present the results of kinetic Monte Carlo simulations using Forward Flux Sampling to show that these predictions are confirmed: namely that on average, nucleation begins in the presence of low-amplitude, but spatially extended density fluctuations thus confirming a significant prediction of the non-classical theory.

cond-mat.stat-mech↗

arXiv2020

Comparison of aluminum oxide empirical potentials from cluster to nanoparticle

Aluminum oxide nanoparticles are increasingly sought in numerous technological applications. However, as the nanoparticles grow during the synthesis, two phase transitions occur. At the nanoscale, numerical simulation of the stability of the alumina phases requires the use of empirical potentials that are reliable over a large range of system sizes going from a few atoms to several hundred thousand atoms. In this work, we confronted four different empirical potentials that are currently employed for bulk alumina. We found that only two of them are correct at the molecular level when compared to DFT calculations. Furthermore, the two potentials remain the best at the nanoscale as they reproduce one or two phase transitions that were observed experimentally: from amorphous solid to cubic crystal (γ) and from cubic to hexagonal (α, i.e. corundum) crystal.

cond-mat.mes-hall↗

arXiv2019

Measuring transferability issues in machine-learning force fields: The example of Gold-Iron interactions with linearized potentials

Machine-learning force fields have been increasingly employed in order to extend the possibility of current first-principles calculations. However, the transferability of the obtained potential can not always be guaranteed in situations that are outside the original database. To study such limitation, we examined the very difficult case of the interactions in gold-iron nanoparticles. For the machine-learning potential, we employed a linearized formulation that is parameterized using a penalizing regression scheme which allows us to control the complexity of the obtained potential. We showed that while having a more complex potential allows for a better agreement with the training database, it can also lead to overfitting issues and a lower accuracy in untrained systems.

physics.chem-ph↗

arXiv2019

Entropy-driven impurity-induced nematic-isotropic transition of liquid crystals

Phase behavior of liquid crystals is of long-standing interest due to numerous applications, with one of the key issues being how the presence of impurities affects the liquid crystalline order. Here we study the orientational order of 4-cyano-4$^{'}$-pentylbiphenyl (5CB) and 4-cyano-4$^{'}$-hexylbiphenyl (6CB) nematic liquid crystals in the presence of varying concentrations of water and n-hexane molecules serving as impurities, by carrying out both fully atomistic simulations and experiments. Our results reveal that mixing of the impurities (in case of hexane) with the host liquid crystals causes a nematic-to-isotropic phase transition with hexane concentration as the control parameter while demixing (in case of water) results in only weak impurity-induced perturbations to the nematic liquid-crystalline order. We develop a coarse-grained model illustrating the general nature and entropic origin of the mixing-induced phase transition.

cond-mat.soft↗

arXiv2019

Combining quantum mechanics and machine-learning calculations for anharmonic corrections to vibrational frequencies

Several methods are available to compute the anharmonicity in semi-rigid molecules. However, such methods are not routinely employed yet because of their large computational cost, especially for large molecules. The potential energy surface is required and generally approximated by a quartic force field potential based on ab initio calculation, thus limiting this approach to medium-sized molecules. We developed a new, fast and accurate hybrid Quantum Mechanic/Machine learning (QM//ML) approach to reduce the computational time for large systems. With this novel approach, we evaluated anharmonic frequencies of 37 molecules thus covering a broad range of vibrational modes and chemical environments. The obtained fundamental frequencies reproduce results obtained using B2PLYP/def2tzvpp with a root-mean-square deviation (RMSD) of 21 cm-1 and experimental results with a RMSD of 23 cm-1. Along with this very good accuracy, the computational time with our hybrid QM//ML approach scales linearly with N while the traditional full ab initio method scales as N2, where N is the number of atoms.

physics.chem-ph↗

arXiv2019

Alloy, Janus and core-shell nanoparticles: Numerical modeling of their nucleation and growth in physical synthesis

While alloy, core-shell and Janus binary nanoclusters are found in more and more technological applications, their formation mechanisms are still poorly understood, especially during synthesis methods involving physical approaches. In this work, we employ a very simple model of such complex systems using Lennard-Jones interactions and inert gas quenching. After demonstrating the ability of the model to well reproduce the formation of alloy, core-shell or Janus nanoparticles, we studied their temporal evolution from the gas via droplets to nanocrystalline particles. In particular, we showed that the growth mechanisms exhibit qualitative differences between these three chemical orderings. Then, we determined how the quenching rate can be used to finely tune structural characteristics of the final nanoparticles, including size, shape and crystallinity.

cond-mat.mtrl-sci↗

arXiv2018

Solvent-mediated interactions between nanostructures: from water to Lennard-Jones liquid

Solvent-mediated interactions emerge from complex mechanisms that depend on the solute structure, its wetting properties and the nature of the liquid. While numerous studies have focused on the two first influences, here, we compare results from water and Lennard-Jones liquid in order to reveal to what extent solvent-mediated interactions are universal with respect to the nature of the liquid. Besides the influence of the liquid, results were obtained with classical density functional theory and brute-force molecular dynamics simulations which allows us to contrast these two numerical techniques.

cond-mat.soft↗