A5015199441- Machine Learning in Materials Science
- Metal-Organic Frameworks: Synthesis and Applications
- Computational Drug Discovery Methods
- X-ray Diffraction in Crystallography
- Enhanced Oil Recovery Techniques
- Advanced Condensed Matter Physics
- Advanced Materials Characterization Techniques
- Protein Structure and Dynamics
- High Entropy Alloys Studies
- Theoretical and Computational Physics
- Clay minerals and soil interactions
- Covalent Organic Framework Applications
- Advanced Chemical Physics Studies
- Soil and Unsaturated Flow
- Perovskite Materials and Applications
- Chemical Synthesis and Reactions
- Bioinformatics and Genomic Networks
- Nonlinear Optical Materials Research
- Quantum, superfluid, helium dynamics
- Lanthanide and Transition Metal Complexes
- Hydrocarbon exploration and reservoir analysis
- Lipid Membrane Structure and Behavior
- Innovative Microfluidic and Catalytic Techniques Innovation
- Scientific Computing and Data Management
- Magnetism in coordination complexes
Queen's University Belfast
2025
Université Paris Cité
2017-2025
Institut Pasteur
2025
Centre National de la Recherche Scientifique
2014-2025
École Polytechnique Fédérale de Lausanne
2020-2024
Advanced Neural Dynamics (United States)
2021
Chimie ParisTech
2016-2019
Université Paris Sciences et Lettres
2016-2019
Institut de Recherche de Chimie Paris
2016-2019
Délégation Paris 6
2017
Alloys composed of several elements in roughly equimolar composition, often referred to as high-entropy alloys, have long been interest for their thermodynamics and peculiar mechanical properties, more recently potential application catalysis. They are a considerable challenge traditional atomistic modeling, also data-driven potentials that the most part memory footprint, computational effort, data requirements which scale poorly with number included. We apply proposed scheme compress...
Atomic-scale simulations have progressed tremendously over the past decade, largely thanks to availability of machine-learning interatomic potentials. These potentials combine accuracy electronic structure calculations with ability reach extensive length and time scales. The i-PI package facilitates integrating latest developments in this field advanced modeling techniques a modular software architecture based on inter-process communication through socket interface. choice Python for...
In computational physics, chemistry, and biology, the implementation of new techniques in shared open-source software lowers barriers to entry promotes rapid scientific progress. However, effectively training users presents several challenges. Common methods like direct knowledge transfer in-person workshops are limited reach comprehensiveness. Furthermore, while COVID-19 pandemic highlighted benefits online training, traditional tutorials can quickly become outdated may not cover all...
We present here the computational chemistry methods our group uses to investigate physical and chemical properties of nanoporous materials adsorbed fluids. highlight multiple time length scales at which these can be examined discuss tools relevant each scale. Furthermore, we include key points consider—upsides, downsides, possible pitfalls—for methods.
In this paper, we parametrized in a consistent way new force field for range of different zeolitic imidazolate framework systems (ZIF-8, ZIF-8(H), ZIF-8(Br), and ZIF-8(Cl)), extending the MOF-FF parametrization methodology two aspects. First, implemented possibility to use periodic reference data order prevent difficulty generating representative finite clusters. Second, optimizer based on covariance matrix adaptation evolutionary strategy (CMA-ES) was employed during process. We confirmed...
Physically motivated and mathematically robust atom-centered representations of molecular structures are key to the success modern atomistic machine learning. They lie at foundation a wide range methods predict properties both materials molecules explore visualize their chemical compositions. Recently, it has become clear that many most effective share fundamental formal connection. can all be expressed as discretization n-body correlation functions local atom density, suggesting opportunity...
Abstract Efficient, physically-inspired descriptors of the structure and composition molecules materials play a key role in application machine-learning techniques to atomistic simulations. The proliferation approaches, as well fact that each choice features can lead very different behavior depending on how they are used, e.g. by introducing non-linear kernels non-Euclidean metrics manipulate them, makes it difficult objectively compare methods, address fundamental questions one feature...
Data-driven schemes that associate molecular and crystal structures with their microscopic properties share the need for a concise, effective description of arrangement atomic constituents. Many types models rely on descriptions atom-centered environments, which are associated an property or contribution to extensive macroscopic quantity. Frameworks in this class can be understood terms density correlations (ACDC), used as basis body-ordered, symmetry-adapted expansion targets. Several other...
Abstract High-entropy alloys (HEAs), containing several metallic elements in near-equimolar proportions, have long been of interest for their unique mechanical properties. More recently, they emerged as a promising platform the development novel heterogeneous catalysts, because large design space, and synergistic effects between components. In this work we use machine-learning potential that can model simultaneously up to 25 transition metals study tendency different segregate at surface...
Zeolitic Imidazolate Frameworks (ZIFs) represent a thriving subclass of metal–organic frameworks (MOFs) owing to the large variety their topologies, which some them are common with zeolites, and ability modulate chemistry as well hydrophobicity/hydrophilicity balance, making perfect examples isoreticular concept. One peculiar structural feature ZIFs is potential for transitions by rotation (or swing) linkers under external stimuli (guest adsorption, mechanical constraints, etc.). This...
Abstract Data analyses based on linear methods constitute the simplest, most robust, and transparent approaches to automatic processing of large amounts data for building supervised or unsupervised machine learning models. Principal covariates regression (PCovR) is an underappreciated method that interpolates between principal component analysis can be used conveniently reveal structure-property relations in terms simple-to-interpret, low-dimensional maps. Here we provide a pedagogic...
The number of materials or molecules that can be created by combining different chemical elements in various proportions and spatial arrangements is enormous.Computational chemistry used to generate databases containing billions potential structures (Ruddigkeit, Deursen, Blum, & Reymond, 2012), predict some the associated properties (Montavon et al., 2013;Ramakrishnan, Dral, Rupp, Lilienfeld, 2014).Unfortunately, very large makes exploring such database -to understand structureproperty...
Here we highlight recent progress in the field of computational chemistry nanoporous materials, focusing on methods and studies that address extraordinary dynamic nature these systems: high flexibility their frameworks, large-scale structural changes upon external physical or chemical stimulation, presence defects disorder. The wide variety behavior demonstrated soft porous crystals, including topical class metal-organic opens new challenges for at all scales.
We report here the properties of LiCl aqueous solutions at various concentrations confined inside pores ZIF-8 metal–organic framework, on basis classical molecular dynamics simulations. This system has been proposed for applications in storage or dissipation mechanical energy, using liquid-phase intrusion concentrated electrolytes hydrophobic framework. describe structure liquids and influence confinement, their dynamics, ZIF-8, impact liquid them. show that presence electrolyte a moderate...
We have studied the properties of water adsorbed inside nanotubes hydrophilic imogolite, an aluminum silicate clay mineral, by means molecular simulations. used a classical force field to describe and flexible imogolite nanotube validated it against data obtained from first-principles dynamics. With it, we observe strong structuration confined in nanotube, with specific adsorption sites distribution hydrogen bond patterns. The combination number sites, their geometry, preferential...
<ns7:p>Easy-to-use libraries such as scikit-learn have accelerated the adoption and application of machine learning (ML) workflows data-driven methods. While many algorithms implemented in these originated specific scientific fields, they gained popularity part because their generalisability across multiple domains. Over past two decades, researchers chemical materials science community put forward general-purpose The deployment methods into other domains, however, is often burdensome due to...
<ns3:p>Easy-to-use libraries such as scikit-learn have accelerated the adoption and application of machine learning (ML) workflows data-driven methods. While many algorithms implemented in these originated specific scientific fields, they gained popularity part because their generalisability across multiple domains. Over past two decades, researchers chemical materials science community put forward general-purpose The deployment methods into other domains, however, is often burdensome due to...
Vibrational spectroscopy is a fundamental tool to investigate local atomic arrangements and the effect of environment, provided that spectral features can be correctly assigned. This challenging in experiments simulations when double peaks are present because they have different origins. Fermi dyads common class such doublets, stemming from resonance excitation mode with overtone another. We new, efficient approach unambiguously characterize resonances density functional theory (DFT) based...
The last decade has seen an explosion of the family framework materials and their study, from both experimental computational points view. We propose here a short highlight current state methodologies for modelling at multiple scales, putting together brief review new methods recent endeavours in this area, as well outlining some open challenges field. will detail advances atomistic simulation methods, development material databases growing use machine learning prediction properties. This...
Spherical harmonics provide a smooth, orthogonal, and symmetry-adapted basis to expand functions on sphere, they are used routinely in physical theoretical chemistry as well different fields of science technology, from geology atmospheric sciences signal processing computer graphics. More recently, have become key component rotationally equivariant models geometric machine learning, including applications atomic-scale modeling molecules materials. We present an elegant efficient algorithm...
We have developed a simple fabrication process for fluorescent silver-coated liquid particles. The synthesis involves the encapsulation of quasi-monodisperse micronic soybean oil-in-water emulsion droplets in polydopamine shell followed by an electroless silver plating surface. Due to presence thin layer, exhibit broad range fluorescence that shows no photobleaching upon illumination. method multimodal colloidal microparticles can be produced via high-yield from sustainable, off-the-shelf...
The last decade has seen an explosion of the family framework materials and their study, both from experimental computational point view. We propose here a short highlight current state methodologies for modelling at multiple scales, putting together brief review new methods recent endeavours in this area, as well outlining some open challenges field. will detail advances atomistic simulation methods, development databases, growing use machine learning properties prediction.