arXiv subjects

Kevin Rossi

Publications and source records attributed to Kevin Rossi.

18 recordsLinked to original sources

arXiv2026

'pySNOW': a Python Suite for the NanO-World

In computational materials science, numerical simulations are indispensable tools for revealing atomic-scale processes. For building blocks of the nanoworld, such as nanoparticles and nanoalloys, atomistic simulations provide detailed insight into their behaviour under diverse conditions. These simulations enable the study of formation and growth mechanisms, transport phenomena, chemophysical stability, and chemical reactions, including catalytic processes. To unravel the complex structure--property relationships that characterise nanoobjects, it is crucial to develop robust and insightful representations of their atomistic structure at global and local scales. Such descriptions enhance our understanding of nanoparticle behaviour and play a key role in guiding their rational design in silico for targeted applications.

cond-mat.mtrl-sci↗

arXiv2026

Tailoring Germanium Heterostructures for Quantum Devices with Machine Learning

Germanium (Ge) quantum wells are emerging as versatile platforms for quantum devices, supporting high-quality spin qubits and integration with superconducting leads. These applications benefit from strong intrinsic spin-orbit interaction (SOI), enabling efficient electrical control and engineering of spin degrees of freedom. The most advanced Ge/SiGe heterostructures to date, based on compressively strained Ge channels within strain-relaxed silicon-germanium (SiGe) barriers, exhibit weak SOI due to the heavy-hole character of the wave function, posing challenges for spin-based quantum devices and requiring complex device designs for fast qubit manipulation. In this work, we demonstrate that concrete heterostructure modifications can overcome these limitations, enhancing SOI by up to three orders of magnitude. Specifically, we propose to enrich unstrained Ge channels by localized, strained silicon spikes. Leveraging a multi-objective Bayesian optimization, we optimize the spike profile to maximize SOI, while ensuring compatibility with current epitaxial growth processes and robustness against realistic variations of growth parameters. Our heterostructure substantially enhances device performance, yielding up to two orders of magnitude higher quantum-dot spin qubit quality factors than state-of-the-art materials. We also predict GHz-scale spin splittings for hybrid superconducting Andreev spin qubits. These novel Ge heterostructures with engineered Si concentration profiles can open pathways to scalable quantum and spintronic applications.

cond-mat.mes-hall↗

arXiv2026

Journal Research Data Policies in Materials Science

Open and reproducible research in materials science relies on the availability of data, code, and common metadata standards. Journal research data policies (RDPs) remain a primary mechanism by which publication norms are defined and enforced. We survey RDPs for 171 materials science journals spanning 17 publishers, using an expanded coding framework that captures both data-and-code sharing behavior as well as refereeing standards. We find clear signs of progress in comparison to earlier research on RDPs: nearly all journals provide an RDP, and most mention data availability statements. However, enforceable requirements remain uncommon, public deposition of underlying data is rarely mandatory, and FAIR publication is typically encouraged rather than required. Expectations for research software are substantially less developed than those for data, with limited attention to versioning and persistent identifiers, dependency disclosure, reproducible execution environments, or software quality practices. Aggregating the findings on policy features into an open research data score reveals pronounced heterogeneity across journals. Neither impact factor nor access model reliably predicts policy strength. Double-coding further shows that more complex policies and stricter policies can be more challenging to interpret consistently, and we highlight challenges in consistent RDP encoding across studies. Lastly, we conclude with recommended best practice directions for the future.

cs.DL↗

arXiv2026

Long-range electrostatics in atomistic machine learning: a physical perspective

The inclusion of long-range electrostatics in atomistic machine learning (ML) is receiving increasing attention for achieving quantum-mechanical accuracy in predicting a wide range of molecular and material properties. However, there is still no general prescription for how long-range physical effects should be incorporated into a model while preserving the well-established locality principles underlying most transferable ML representations. Here, we provide a physical perspective on this problem by discussing how distinct contributions to a system's electrostatics can be captured through different learning paradigms. Specifically, we distinguish between local charge models, which rely either on explicit charge-density decompositions or implicit auxiliary variables, and models in which a notion of nonlocality is deliberately introduced, either via self-consistent procedures or by using nonlocal descriptors and learning architectures. We further address the related aspect of incorporating finite-field effects through coupling to the system's polarization, relevant for applications involving an external electric bias. We conclude by discussing implications for simulations of electrochemical interfaces, where long-range electrostatics are essential to capture the interplay between charge redistribution, interfacial dynamics, and ionic screening, as well as for ionic transport phenomena, which, although less explored, appear far less sensitive to their inclusion.

cond-mat.mtrl-sci↗

arXiv2025

Bridging Structure and Activity in Nanocatalysts via Machine Learning and Global Structure Representations

Establishing a mapping between nanocatalysts structure and their catalytic properties is essential for efficient design. To this end, we demonstrate the accuracy of a general machine learning framework on a representative and challenging application: predicting the mass activity of Pt nanoparticles for the electrochemical oxygen reduction reaction, estimated via a microkinetic model. Accurate models are obtained when leveraging either a nanocatalyst's structure representation accessible at the computational level, namely the surface site generalized coordination number distributions, or one accessible experimentally, namely the nanoparticle's pair distance distribution function. Building on this result, we demonstrate that our machine learning model, in tandem with Bayesian optimization, efficiently identifies the Top-10 and Top-100 most active structures out of a large pool of candidates comprising more than 50000 different structures, after probing the activity only of a few thousand structures. These findings provide a robust blueprint for accelerated theoretical and experimental identification of active nanocatalysts.

cond-mat.mes-hall↗

arXiv2025

Uncertainty in the era of machine learning for atomistic modeling

The widespread adoption of machine learning surrogate models has significantly improved the scale and complexity of systems and processes that can be explored accurately and efficiently using atomistic modeling. However, the inherently data-driven nature of machine learning models introduces uncertainties that must be quantified, understood, and effectively managed to ensure reliable predictions and conclusions. Building upon these premises, in this Perspective, we first overview state-of-the-art uncertainty estimation methods, from Bayesian frameworks to ensembling techniques, and discuss their application in atomistic modeling. We then examine the interplay between model accuracy, uncertainty, training dataset composition, data acquisition strategies, model transferability, and robustness. In doing so, we synthesize insights from the existing literature and highlight areas of ongoing debate.

physics.chem-ph↗

arXiv2025

Learning the Electrostatic Response of the Electron Density through a Symmetry-Adapted Vector Field Model

A current challenge in atomistic machine learning is that of efficiently predicting the response of the electron density under electric fields. We address this challenge with symmetry-adapted kernel functions that are specifically derived to account for the rotational symmetry of a three-dimensional vector field. We demonstrate the equivariance of the method on a set of rotated water molecules and show its high efficiency with respect to number of training configurations and features for liquid water and naphthalene crystals. We conclude showcasing applications for relaxed configurations of gold nanoparticles, reproducing the scaling law of the electronic polarizability with size, up to systems with more than 2000 atoms. By deriving a natural extension to equivariant learning models of the electron density, our method provides an accurate and inexpensive strategy to predict the electrostatic response of molecules and materials.

cond-mat.mtrl-sci↗

arXiv2021

Accelerating the theoretical study of Li-polysulphide adsorption on single-atom catalysts via machine learning approaches

Unlocking the design of Li-S batteries where no shuttle effects appears, and thus their energy storage capacity does not diminish over time, would enable the manufacturing of energy storage devices more performant than the current Li-ion commercial ones. Computational screening of Li-polysulphide (LiPS) adsorption on single-atom catalyst (SAC) substrates is of great aid to the design of Li-S batteries which are robust against the LiPS shuttling from the cathode to the anode and the electrolyte. To aid this process, we develop a machine learning protocol to accelerate the systematic mapping of dominant local minima found with DFT calculations, and, in turn, fast screen LiPS adsorption properties on SACs. We first validate the approach by probing the potential energy surface for Li-polysulphides adsorbed on graphene decorated with a Fe-N$_4$-C SAC bound to four nitrogen atoms. We identify minima whose binding energy is better or on par with the one previously reported in the literature. We then move to analyse the adsorption trends on Zn-N$_4$-C SAC and observe similar adsorption strength and behaviour with the Fe-N$_4$-C SAC, highlighting the good predictive power of our protocol.

cond-mat.mtrl-sci↗

arXiv2021

Exploring the robust extrapolation of high-dimensional machine learning potentials

We show that, contrary to popular assumptions, predictions from machine learning potentials built upon high-dimensional atom-density representations almost exclusively occur in regions of the representation space which lie outside the convex hull defined by the training set points. We then propose a perspective to rationalize the domain of robust extrapolation and accurate prediction of atomistic machine learning potentials in terms of the probability density induced by training points in the representation space

physics.comp-ph↗

arXiv2021

Data-driven simulation and characterisation of gold nanoparticle melting

The simulation and analysis of the thermal stability of nanoparticles, a stepping stone towards their application in technological devices, require fast and accurate force fields, in conjunction with effective characterisation methods. In this work, we develop efficient, transferable, and interpretable machine learning force fields for gold nanoparticles based on data gathered from Density Functional Theory calculations. We use them to investigate the thermodynamic stability of gold nanoparticles of different sizes (1 to 6 nm), containing up to 6266 atoms, concerning a solid-liquid phase change through molecular dynamics simulations. We predict nanoparticle melting temperatures in good agreement with available experimental data. Furthermore, we characterize the solid-liquid phase change mechanism employing an unsupervised learning scheme to categorize local atomic environments. We thus provide a data-driven definition of liquid atomic arrangements in the inner and surface regions of a nanoparticle and employ it to show that melting initiates at the outer layers.

physics.comp-ph↗

arXiv2021

Compact atomic descriptors enable accurate predictions via linear models

We probe the accuracy of linear ridge regression employing a three-body local density representation derived from the atomic cluster expansion. We benchmark the accuracy of this framework in the prediction of formation energies and atomic forces in molecules and solids. We find that such a simple regression framework performs on par with state-of-the-art machine learning methods which are, in most cases, more complex and more computationally demanding. Subsequently, we look for ways to sparsify the descriptor and further improve the computational efficiency of the method. To this aim, we use both principal component analysis and least absolute shrinkage operator regression for energy fitting on six single-element datasets. Both methods highlight the possibility of constructing a descriptor that is four times smaller than the original with a similar or even improved accuracy. Furthermore, we find that the reduced descriptors share a sizable fraction of their features across the six independent datasets, hinting at the possibility of designing material-agnostic, optimally compressed, and accurate descriptors.

physics.comp-ph↗

arXiv2021

Uncertainty estimation for molecular dynamics and sampling

Machine learning models have emerged as a very effective strategy to sidestep time-consuming electronic-structure calculations, enabling accurate simulations of greater size, time scale and complexity. Given the interpolative nature of these models, the reliability of predictions depends on the position in phase space, and it is crucial to obtain an estimate of the error that derives from the finite number of reference structures included during the training of the model. When using a machine-learning potential to sample a finite-temperature ensemble, the uncertainty on individual configurations translates into an error on thermodynamic averages, and provides an indication for the loss of accuracy when the simulation enters a previously unexplored region. Here we discuss how uncertainty quantification can be used, together with a baseline energy model, or a more robust although less accurate interatomic potential, to obtain more resilient simulations and to support active-learning strategies. Furthermore, we introduce an on-the-fly reweighing scheme that makes it possible to estimate the uncertainty in the thermodynamic averages extracted from long trajectories. We present examples covering different types of structural and thermodynamic properties, and systems as diverse as water and liquid gallium.

physics.chem-ph↗

arXiv2020

Simulating solvation and acidity in complex mixtures with first-principles accuracy: the case of CH$_3$SO$_3$H and H$_2$O$_2$ in phenol

We present a generally-applicable computational framework for the efficient and accurate characterization of molecular structural patterns and acid properties in explicit solvent using H$_2$O$_2$ and CH$_3$SO$_3$H in phenol as an example. In order to address the challenges posed by the complexity of the problem, we resort to a set of data-driven methods and enhanced sampling algorithms. The synergistic application of these techniques makes the first-principle estimation of the chemical properties feasible without renouncing to the use of explicit solvation, involving extensive statistical sampling. Ensembles of neural network potentials are trained on a set of configurations carefully selected out of preliminary simulations performed at a low-cost density-functional tight-binding (DFTB) level. The energy and forces of these configurations are then recomputed at the hybrid density functional theory (DFT) level and used to train the neural networks. The stability of the NN model is enhanced by using DFTB energetics as a baseline, but the efficiency of the direct NN (i.e., baseline-free) is exploited via a multiple-time step integrator. The neural network potentials are combined with enhanced sampling techniques, such as replica exchange and metadynamics, and used to characterize the relevant protonated species and dominant non-covalent interactions in the mixture, also considering nuclear quantum effects.

physics.chem-ph↗

arXiv2019

On Machine Learning Force Fields for Metallic Nanoparticles

Machine learning algorithms have recently emerged as a tool to generate force fields which display accuracies approaching the ones of the ab-initio calculations they are trained on, but are much faster to compute. The enhanced computational speed of machine learning force fields results key for modelling metallic nanoparticles, as their fluxionality and multi-funneled energy landscape needs to be sampled over long time scales. In this review, we first formally introduce the most commonly used machine learning algorithms for force field generation, briefly outlining their structure and properties. We then address the core issue of training database selection, reporting methodologies both already used and yet unused in literature. We finally report and discuss the recent literature regarding machine learning force fields to sample the energy landscape and study the catalytic activity of metallic nanoparticles.

physics.comp-ph↗

arXiv2018

Building machine learning force fields for nanoclusters

We assess Gaussian process (GP) regression as a technique to model interatomic forces in metal nanoclusters by analysing the performance of 2-body, 3-body and many-body kernel functions on a set of 19-atom Ni cluster structures. We find that 2-body GP kernels fail to provide faithful force estimates, despite succeeding in bulk Ni systems. However, both 3- and many-body kernels predict forces within a $\sim$0.1 eV/$\textÅ$ average error even for small training datasets, and achieve high accuracy even on out-of-sample, high temperature, structures. While training and testing on the same structure always provides satisfactory accuracy, cross-testing on dissimilar structures leads to higher prediction errors, posing an extrapolation problem. This can be cured using heterogeneous training on databases that contain more than one structure, which results in a good trade-off between versatility and overall accuracy. Starting from a 3-body kernel trained this way, we build an efficient non-parametric 3-body force field that allows accurate prediction of structural properties at finite temperatures, following a newly developed scheme [Glielmo et al. PRB 97, 184307 (2018)]. We use this to assess the thermal stability of Ni$_{19}$ nanoclusters at a fractional cost of full ab initio calculations.

physics.comp-ph↗

arXiv2018

A genomic characterization of metallic nanoparticles

With a focus on platinum nanoparticles of different sizes (diameter of 1-9 nm) and shapes, we sequence their geometrical genome by recording the relative occurrence of all the non equivalent active site, classified according to the number of neighbours in their first and second coordination shell. The occurrence of sites is morphology and size dependent, with significant changes in the relative occurrence up to 9 nm. Our geometrical genome sequencing approach is immediately transferable to address the effects of the morphological polydispersivity in size-selected samples and the influence of temperature, including ionic vibrations and thermal activated processes. The proposed geometrical genome forecasts an enhancement of the catalytic reduction of molecular oxygen on stellated and anisotropic platinum twinned nanoparticles, with their shortest axes of ~2 nm, and an irreversible disruption of the Pt nanocatalyst's structure above 1000 K.

cond-mat.mes-hall↗

arXiv2017

Size dependent rearrangements in monometallic clusters

Morphology and its stability are essential features to address physicochemical properties of metallic nanoparticles. By means of Molecular Dynamics based simulations we show a complex dependence on the size and material of common structural mechanisms taking place in mono-metallic nanoparticles at icosahedral magic sizes. We show that the well known Lipscomb s Diamond Square Diamond mechanisms, single step screw dislocation motions of the whole cluster, take place only below a given size which is material dependent. Above that size, layer by layer dislocations and/or surface peeling are likely to happen, leading to low symmetry defected motifs. The material dependence of this critical size is similar to the crossover sizes among structural motifs, based on the ration between the bulk modulus and atomic cohesive energy.

cond-mat.mes-hall↗

arXiv2015

Metallic nanoparticles meet Metadynamics

We show how standard Metadynamics coupled with classical Molecular Dynamics can be successfully ap- plied to sample the configurational and free energy space of metallic and bimetallic nanopclusters via the implementation of collective variables related to the pair distance distribution function of the nanoparticle itself. As paradigmatic examples we show an application of our methodology to Ag147, Pt147 and their alloy AgshellPtcore at 1:1 and 2:1 chemical compositions. The proposed scheme is not only able to reproduce known structural transformation pathways, as the five and the six square-diamond mechanisms both in pure and core-shell nanoparticles but also to predict a new route connecting icosahedron to anti-cuboctahedron.

physics.comp-ph↗