Search

International audience; The morphology, epitaxy, band-alignment and optical response of silver nanoparticles on ZnO(1010) has been analysed by microscopy, photoemission, electron diraction and dier-ential reectivity. At 700 K, clusters tend to nu-cleate and grow exclusively along [010]-oriented polar steps and corners, and not on at terraces or non-polar [001] steps. This preference leads to a specic epitaxy Ag(111)[110] ZnO[010](1010) with a large lattice mismatch along the dense row of the metal. The experiments have been rationalized through ab ini-tio simulations. A much lower adsorption energy is obtained for the O-terminated steps with a charge transfer from silver to oxygen, that corroborates the variation of ionization energy observed in photoemission and the presence of cationic silver. This anisotropy of growth reverberates in the plasmonic response of the metallic particles. The growth and epitaxy of metals on the polar (0001) and non-polar (1010) ori-entations of ZnO is discussed in terms of sur-face/step polarity compensation in the light of the present ndings.

The morphology, epitaxy, band alignment, and optical response of silver nanoparticles on ZnO(10 (1) over bar0) have been analyzed by microscopy, electron diffraction, photoemission and differential reflectivity. At 700 K, clusters tend to nucleate and grow exclusively along [010]-oriented polar steps and corners and not on flat terraces or nonpolar [001] steps. This preference leads to a specific epitaxy Ag[1 (1) over bar0](111) II ZnO[010] (10 (1) over bar0) with a large lattice mismatch along the dense row of the metal. The experiments have been rationalized through ab initio simulations. A much lower adsorption energy is obtained for the O-terminated steps with a charge transfer from silver to oxygen that corroborates the variation of ionization energy observed in photoemission and the presence of cationic silver. This anisotropy of growth reverberates in the plasmonic response of the metallic particles. The growth and epitaxy of metals on polar (0001) and nonpolar (10 (1) over bar0) orientations of ZnO are discussed in terms of surface/step polarity compensation in the light of the present findings.

We conducted ab initio molecular dynamics simulations of the delta phase of the hydrous mineral AlOOH at ambient temperature and high pressure. Nuclear quantum effects were included through a Langevin dynamics in a bath of quantum harmonic oscillators. We confirm that under increasing pressure delta-AlOOH undergoes a phase transition from a $P2_1nm$ structure with asymmetric and disordered O-H bonds to a stiffer $Pnnm$ phase with symmetric hydrogen bonds, which should be stable within the pressure and temperature ranges typical for the Earth's mantle. The transition is initially triggered by proton tunneling, which makes the mean proton position to coincide with the midpoint of the O-O distance, at pressures as low as 10 GPa. However, only at much larger pressures, around 30 GPa as previously found by other calculations, the Pnnm phase with symmetric hydrogen bonds is stable from the classical point of view. The transition is also characterized through the analysis of the H-O stretching modes, which soften considerably and fade out around 10 GPa in the P2_1nm structure, when thermal and nuclear quantum effects are taken into account in the simulations. At variance, the harmonic picture is not adequate to describe the highly anharmonic effective potential that is seen by the protons at the transition. Finally, we propose that the picture of a dynamical transition to the high-symmetry and proton-centered Pnnm phase, which is brought about by the onset of proton tunneling, could be confirmed by quasi-elastic neutron scattering and vibrational spectroscopy under pressure.

International audience; The quantum thermal bath (QTB) has been presented as analternative to path-integral-based methods to introduce nuclear quantumeffects in molecular dynamics simulations. The method has proved to beefficient, yielding accurate results for various systems. However, the QTBmethod is prone to zero-point energy leakage (ZPEL) in highly anharmonicsystems. This is a well-known problem in methods based on classicaltrajectories where part of the energy of the high-frequency modes istransferred to the low-frequency modes leading to a wrong energydistribution. In some cases, the ZPEL can have dramatic consequences onthe properties of the system. Thus, we investigate the ZPEL by testing theQTB method on selected systems with increasing complexity in order to studythe conditions and the parameters that influence the leakage. We also analyze the consequences of the ZPEL on the structuraland vibrational properties of the system. We find that the leakage is particularly dependent on the damping coefficient and thatincreasing its value can reduce and, in some cases, completely remove the ZPEL. When using sufficiently high values for thedamping coefficient, the expected energy distribution among the vibrational modes is ensured. In this case, the QTB methodgives very encouraging results. In particular, the structural properties are well-reproduced. The dynamical properties should beregarded with caution although valuable information can still be extracted from the vibrational spectrum, even for large values ofthe damping term.

Increasing interest has risen for nuclear quantum effects (NQE) in the recent past. Indeed, NQE such as proton tunneling and zero point energy often play a crucial role in the properties of hydrogen-containing materials. The standard methods to simulate NQE are based on path integrals. An alternative to these methods is the Quantum Thermal Bath (QTB): it is based on a Langevin equation where the classical degrees of freedom are coupled to an ensemble of quantum harmonic oscillators. In the classical Langevin equation, the random force is a white noise and fulfills the classical fluctuation-dissipation theorem, while within the QTB formalism, it fulfills the quantum fluctuation-dissipation theorem. We investigate through simple models the reliability and the limits of the QTB and show that the QTB enables realistic simulations including NQE of condensed-phase systems, generating static and dynamic information such as pair correlation functions and vibrational spectra which can be confronted with experimental results. We show that the QTB is particularly successful in the study of the symmetrization of hydrogen bonds in several systems. Indeed, the difficulty lies in the identification of a precise transition pressure since this phase transition is often blurred by quantum or thermal fluctuations. In high-pressure ice, it depends on the oxygen-oxygen distance but it can be affected by ionic impurities and by the asymmetric environment of hydrogen bonds as in the delta phase of AlOOH. Moreover, by comparing results from QTB and standard ab initio simulations, we are able to disentangle the respective roles of NQE and thermal fluctuations in these phase transitions.; L’étude des effets quantiques nucléaires (NQE) suscite de plus en plus d’intérêt. En effet, les effets quantiques comme l’effet tunnel ou l’énergie de point zéro, peuvent profondément modifier les propriétés de matériaux constitués d'atomes légers comme l'hydrogène. Les méthodes standards de simulation des NQE sont basées sur les intégrales de chemin. Le bain thermique quantique (QTB) constitue une alternative à ces méthodes: le principe est que les degrés de liberté classiques du système obéissent à une équation de Langevin et sont couplés à des oscillateurs harmoniques quantiques. Dans l’équation de Langevin classique, la force aléatoire est un bruit blanc et le théorème de fluctuation-dissipation classique est vérifié; avec le QTB, le théorème de fluctuation-dissipation quantique est vérifié. Nous étudierons à travers des modèles simples la validité et les limites du QTB et montrerons qu'il permet de simuler des systèmes de la matière condensée en incluant les NQE en générant leurs propriétés structurales et dynamiques. Nous montrerons que le QTB est particulièrement adapté à l’étude de la symétrisation de liaisons hydrogènes et permet d'identifier précisément une pression de transition. Celle-ci dépend de la distance entre deux oxygènes voisins comme dans la glace sous haute pression, mais est modifiée par la présence d'impuretés ioniques ou par l'environnement atomique des liaisons hydrogènes comme dans la phase delta de AlOOH. De plus, en comparant des simulations classiques à des simulations QTB, nous pouvons identifier les rôles respectifs des effets quantiques et thermiques dans ces transitions de phase.