Centre d’Élaboration de Matériaux et d’Etudes Structurales

Accueil > Recherche > SINanO : Surfaces, Interfaces et Nano-Objets > Thématiques de Recherche

Nano-objects : growth and structure

Selection of our most recent publications on this topic (from 2018) - The names of the team members are in blue


Silica-induced electron loss of silver nanoparticles


M. Benoit, J. Puibasset, C. Bonafos and N. Tarrat

Nanoscale, 2022


Despite the frequent use of silver nanoparticles (Ag NP) embedded in materials for medical or optical applications, the effect of the matrix on the nanoparticles properties remains largely unknown. This study aims to shed light on the effect of an amorphous silica matrix on the structure and charge distribution of 55- and 147-atoms silver nanoparticles by means of dispersion-corrected DFT calculations. Particular attention is paid on nanoparticle size and concentration effects and on the impact of the presence of native defects in the matrix. Covalent bonding between the silver nanoparticles and the matrix are found to occur at the interface. Such interface reconstruction involves the breaking of Si-O bonds, which systematically leads to the formation of Ag-Si bonds, and in some cases, to the formation of Ag-O ones. Interestingly, these interface reconstructions are accompanied by an electron depletion of the nanoparticle, a substantial number of electrons being transferred from the two outer shells of the Ag NP towards the surrounding silica medium. The electrons lost by the nanoparticle are captured by the Si atoms involved in the interface bonds, but also, unexpectedly, by the undercoordinated silica defects that act as electron pumps and by the atoms of the silica network inside a spherical shell of a few angströms around the silver nanoparticle. The number of interface bonds and of electrons transferred to the surrounding silica shell appears to be proportional to the surface area of the Ag NP. The electronic extension within the silica goes beyond that attributable to the Ag NP spill-out. The presence of additional electrons in the matrix, especially on defects, is consistent with the experimental literature.


Periodic DFTB for supported clusters : Implementation and application to benzene dimers deposited on graphene


Mathias Rapacioli and Nathalie Tarrat

Computation 2022, 10, 39


The interest for properties of clusters deposited on surfaces has grown in recent years. In this framework, the Density Functional based Tight Binding (DFTB) method appears as a promising tool due to its ability to treat extended systems at the quantum level with a low computational cost. We report the implementation of periodic boundary conditions for DFTB within the deMonNano code with k-points formalism and corrections for intermolecular interactions. The quality of DFTB results is evaluated by comparison with dispersion-corrected DFT calculations. Optimized lattice properties for a graphene sheet and graphite bulk are in agreement with reference data. The deposition of both benzene monomer and dimers on graphene are investigated and the observed trends are similar at the DFT and DFTB levels. Moreover, interaction energies are of similar orders of magnitude for these two levels of calculation. This study has evidenced the high stability of a structure made of two benzene molecules deposited close to each other on the graphene sheet. This work demonstrates the ability of the new implementation to investigate surface-deposited molecular clusters properties.


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


M. Benoit, J. Amodeo, S. Combettes, A. Roux, I. Khaled, J. Lam

Mach. Learn. : Sci. Technol. 2 025003 (2021)


(Haut) Image des structures testées en dehors de la base de données (Bas) Erreur mesurée sur les forces en fonction de la contrainte de la régression linéaire. Plus les valeurs de alpha sont petites, plus le potentiel est complexe.


De par leur capacité à faire le pont entre la précision des modèles basés sur la structure électronique et la rapidité des potentiels d’interactions classiques, les méthodes machine-learning ont été largement plébiscitées pour le développement de champs de forces entre les atomes. Néanmoins, un inconvénient de ces nouvelles méthodes demeure leur possible incapacité à extrapoler au-delà de la base de données utilisée pendant la phase d’apprentissage.

Dans cet article, nous avons commencé par présenter une nouvelle méthode machine-learning basée sur l’utilisation d’un algorithme de régression linéaire sous contrainte et nous avons montré qu’il était alors possible de construire des champs de forces pour des nanoparticules Fe/Au tout en contrôlant à la fois la complexité et la précision du potentiel. Ensuite, en testant les potentiels obtenus sur des structures non explorées dans la base de données d’apprentissage, nous avons observé un surprenant caractère non-monotone de la précision vis à vis de la complexité.

Ces résultats montrent de manière quantitative une limitation des méthodes machine-learning et tend ainsi vers une meilleure compréhension de leur utilisation.


Plasmon damping and charge transfer pathways in Au@MoSe2 Nanostructures


I. Abid, P. Benzo, B. Pecassou, S. Jia, J. Zhang, J. Yuan, J.B. Dory, O. Gauthier Lafaye, R. Pechou, A. Mlayah, J. Lou

Materials Today Nano 15 (2021) 100131


(A) Cross-section drawing of the sample, (B) optical microscopy image, (C) SEM, and (D) AFM topography images of an MoSe2 flake covered by Au NPs deposited with 4.8-nm Au equivalent thickness. (E) SEM image of an MoSe2 flake covered by Au NPs deposited with 0.8-nm Au equivalent thickness.
(A) Cross-section drawing of the sample, (B) optical microscopy image, (C) SEM, and (D) AFM topography images of an MoSe2 flake covered by Au NPs deposited with 4.8-nm Au equivalent thickness. (E) SEM image of an MoSe2 flake covered by Au NPs deposited with 0.8-nm Au equivalent thickness.

Hybridization of plasmonic and excitonic elementary excitations provides an efficient mean of enhancing the optical absorption and emission properties of metal/semiconductor nanostructures and is a key concept for the design of novel efficient optoelectronic devices. Here we investigate the optical properties of two-dimensional MoSe2 quantum well flakes covered with Au nanoparticles supporting plasmonic resonances. Using spatially resolved confocal spectroscopy, we report the observation of a quenching phenomenon of the Raman scattering and photoluminescence emission of both the MoSe2 layer and the Au nanoparticles. We found that the quenching of the photoluminescence emission from the Au nanoparticles is partial and measurable unlike the one observed for the Au-covered MoSe2 layers, which is total. Its dependence on the thickness of the MoSe2 layer is determined experimentally. Based on electrodynamics calculations and on the electronic band alignment at the Au/MoSe2 interface, the results are interpreted in terms of (1) damping of the plasmonic resonance of the Au nanoparticles due to the optical absorption by the MoSe2 layer and (2) a two-pathways charge transfer scheme where the photoexcited electrons leak from the MoSe2 layer to the Au NPs, whereas the photoexcited holes flow in the opposite direction, that is, from the Au NPs to the MoSe2 layer. The two combined mechanisms account well for the experimental observations and complements the interpretations proposed in the literature for similar metal nanoparticles/transition metal dichalcogenide systems.


Exploring energy landscapes at the DFTB quantum level using the threshold algorithm : the case of the anionic metal cluster Au20-


Mathias Rapacioli, Johannes C. Schoen and Nathalie Tarrat

Theor. Chem. Acc. 2021, 140, 85



We report the combination of the threshold algorithm with the Density Functional-based Tight Binding method allowing for the exploration of complex potential energy surfaces and the evaluation of probability flows between their regions, at the quantum level. This original scheme is used to explore the energy landscape of an anionic 20-atom gold cluster, Au20. On the basis of the relevant structures, 19 structural groups are highlighted, all of them being variations about the pyramidal shape : (1) distorted pyramids, (2) pyramids in which the atom of one of the facets has been removed, leaving a hole, and placed at different positions on the cluster and (3) pyramids on which an atom located at a vertex has been removed and placed on an edge or on a facet. Upper limits of the energies required to connect the basins of the 19 groups on the potential energy surface are evaluated. Moreover, the attractive basins are identified on the basis of the analysis of the probability flows on the landscape. The comparison of the disconnectivity tree with the results of the flux analysis provides a consistent representation of the Au basins’ proximity. Finally, we show how the new scheme allowed for the identification of counter-intuitive transition pathways.


 Importance of Defective and Nonsymmetric Structures in Silver Nanoparticles


David Loffreda, Dawn M. Foster, Richard E. Palmer, and Nathalie Tarrat

J. Phys. Chem. Lett. 2021, 12, 3705−3711


Scanning transmission electron microscopy experiments indicate that face-centered cubic (FCC) is the predominant ordered structure for Ag309 ± 7 nanoclusters, synthesized in vacuum. Historically, experiments do not present a consensus on the morphology at these sizes, whereas theoretical studies find the icosahedral symmetry for Ag309 and the decahedral shape for nearby sizes. We employ density functional theory calculations to rationalize these observations, considering both regular and defective Ag nanoparticles (281−321 atoms). The change of stability induced by the presence of defects, symmetry loss, and change of number of atoms is evaluated by the nanoparticle surface energy, which was measured previously. FCC and decahedral symmetries are found to be more favorable than icosahedral, consistent with our measurements of clusters protected from extended atmospheric exposure. In addition, an energy-free descriptor, surface atomic density, is proposed and qualitatively reproduces the surface energy data. Nonsymmetric and defective structures may be preferred over perfectly regular ones within a given size range.


How interface properties control the equilibrium shape of core–shell Fe–Au and Fe–Ag nanoparticles



Ségolène Combettes, Julien Lam, Patrizio Benzo, Anne Ponchet, Marie-José Casanove, Florent Calvo and Magali Benoit

Nanoscale, 2020, 12, 18079-18090


Growth simulations of Fe-Au and Fe-Ag nanoparticles
Growth simulations of Fe-Au and Fe-Ag nanoparticles

While combining two metals in the same nanoparticle can lead to remarkable novel applications, the resulting structure in terms of crystallinity and shape remains difficult to predict. It is thus essential to provide a detailed atomistic picture of the underlying growth processes. In the present work we address the case of core–shell Fe–Au and Fe–Ag nanoparticles. Interface properties between Fe and the noble metals Au and Ag, computed using DFT, were used to parameterize Fe–Au and Fe–Ag pairwise interactions in combination with available many-body potentials for the pure elements. The growth of Au or Ag shells on nanometric Fe cores with prescribed shapes was then modelled by means of Monte Carlo simulations. The shape of the obtained Fe–Au nanoparticles is found to strongly evolve with the amount of metal deposited on the Fe core, a transition from the polyhedral Wulff shape of bare iron to a cubic shape taking place as the amount of deposited gold exceeds two monolayers. In striking contrast, the growth of silver proceeds in a much more anisotropic, Janus-like way and with a lesser dependence on the iron core shape. In both cases, the predicted morphologies are found to be in good agreement with experimental observations in which the nanoparticles are grown by physical deposition methods. Understanding the origin of these differences, which can be traced back to subtle variations in the electronic structure of the Au/Fe and Ag/Fe interfaces, should further contribute to the better design of core–shell bimetallic nanoparticles.


 Equilibrium shape of core(Fe)–shell(Au) nanoparticles as a function of the metals volume ratio


A. Ponchet, S. Combettes, P. Benzo, N. Tarrat, M. J. Casanove, and M. Benoit

Journal of Applied Physics 2020, 128, 055307


The equilibrium shape of nanoparticles is investigated to elucidate the various core–shell morphologies observed in a bimetallic system associating two immiscible metals, iron and gold, that crystallize in the bcc and fcc lattices, respectively. Fe–Au core–shell nanoparticles present a crystalline Fe core embedded in a polycrystalline Au shell, with core and shell morphologies both depending on the Au/Fe volume ratio. A model is proposed to calculate the energy of these nanoparticles as a function of the Fe volume, Au/Fe volume ratio, and the core and shell shape, using the density functional theory-computed energy densities of the metal surfaces and of the two possible Au/Fe interfaces. Three driving forces leading to equilibrium shapes were identified : the strong adhesion of Au on Fe, the minimization of the Au/Fe interface energy that promotes one of the two possible interface types, and the Au surface energy minimization that promotes a 2D–3D Stranski–Krastanov-like transition of the shell. For a low Au/Fe volume ratio, the wetting is the dominant driving force and leads to the same polyhedral shape for the core and the shell, with an octagonal section. For a large Au/Fe ratio, the surface and interface energy minimizations can act independently to form an almost cube-shaped Fe core surrounded by six Au pyramids. The experimental nanoparticle shapes are well reproduced by the model, for both low and large Au/Fe volume ratios. J. Appl. Phys. 128, 055307 (2020)


Epitaxial growth of a gold shell on intermetallic FeRh nanocrystals


P. Benzo, S. Combettes, C. Garcia, T. Hungria, B. Pecassou, and M. J. Casanove

Cryst. Growth Des. 2020, 20, 6, 4144–4149


In contrast with bimetallics, multimetallic nanoparticles (NPs) can combine different chemical orders in a same NP, which favors an enhanced tuning of their properties. In this work, trimetallic (Fe,Rh,Au) nanocrystals with controlled composition and chemical distribution were grown through a physical vapor deposition route using a two-step process. First FeRh nanocrystals, 8.5 nm of mean diameter, were formed according to a Volmer-Weber growth mode. The growth conditions were tuned so as to achieve the atomic scale chemical order displayed by the intermetallic B2-FeRh phase. Then the gold layer was deposited at a lower temperature. Evidence is given for the complete coverage of the gold shell, which grows epitaxially over the B2-FeRh core exposed facets. These different features are particularly promising for further applications.


Role of the shell thickness in the core transformation of magnetic core(Fe)-shell(Au) nanoparticles


P. Benzo, S. Combettes, B. Pecassou, N. Combe, M. Benoit , M. Respaud, M.J. Casanove

Phys. Rev. Materials 2019, 3, 096001


Fe-Au core-shell nanoparticles embedded in an amorphous alumina matrix are synthesized at high temperature using magnetron sputtering. The nanoparticles display a single-crystal Fe core covered by a crystalline Au shell epitaxially grown on the different core facets. The morphology of the grown nanoparticles is analyzed by transmission electron microscopy, while their structural details are resolved at the atomic scale using probecorrected high-angle annular dark-field scanning transmission microscopy. Two different epitaxial relationships at the Au/Fe interface are observed at low Au coverage, whereas only one type of interface orientation remains when the shell thickness exceeds 3 monolayers (MLs). This leads to a drastic Fe core transformation from a Wulff shaped crystal towards a nanocube. This core transformation drives a surface reconstruction resulting in a combination of open and close-packed strain free facets, offering different opportunities for molecule binding. In addition, our experiments show that the magnetic properties of the Fe core are preserved by the Au shell and that the 10 nm size of the grown nanoparticles favors a superparamagnetic behavior, suitable for biomedical applications.


 Au147 nanoparticles : Ordered or amorphous ?


Nathalie Tarrat, Mathias Rapacioli, Fernand Spiegelman

J. Chem. Phys. 2018, 148, 204308 


Structural aspects of the Au147 nanoparticles have been investigated through a density functional based tight binding global optimization involving a parallel tempering molecular dynamics scheme with quenching followed by geometries relaxation at the Density Functional Theory (DFT) level. The focus is put on the competition between relaxed ordered regular geometries and disordered (or amorphous) structures. The present work shows that Au147 amorphous geometries are relevant low energy candidates and are likely to contribute in finite temperature dynamics and thermodynamics. The structure of the amorphous-like isomers is discussed from the anisotropy parameters, the atomic coordinations, the radial and pair distribution functions, the IR spectra, and the vibrational DOS. With respect to the regular structures, the amorphous geometries are shown to be characterized by a larger number of surface atoms, a less dense volume with reduced coordination number per atom, a propensity to increase the dimension of flat facets at the surface, and a stronger anisotropy. Moreover, all amorphous clusters have similar IR spectra, almost continuous with active frequencies over the whole spectral range, while symmetric clusters are characterized by a few lines with large intensities.