« avril 2024 »
L M M J V S D
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 1 2 3 4 5
 
Tous les évènements de Physique à venir

Tous les évènements de Biologie / Chimie à venir

Tous les évènements à venir

Les évènements relevant de la Physique et de la Biologie / Chimie sont représentés en turquoise

Neural Network Approach for a Rapid Prediction of Metal-Supported Borophene Properties

Jeudi 7 mars 11:00 - Duree : 1 heure
Lieu : Science Building Seminar Room, SB-036, ILL, EPN Campus

Orateur : C. BOUSIGE (Université Claude Bernard Lyon1, Villeurbanne, France)

Résumé :

Single layer materials have drawn a lot of attention due to their peculiar physical properties (opto-electronic properties, high conductivity, flexibility…). In particular, it has been predicted that boron could exist as a single atomic layer in distinctive crystallographic configurations (allotropes), called borophene – in reference to the carbon equivalent, graphene. Borophene is one of the only 2D material with metallic behaviour, among other interesting properties [1]. Recent studies have focused on the synthesis of such material under various allotropic forms, the obtained allotrope depending on the substrate used and experimental parameters such as synthesis temperature [2–5].

To identify and assess borophene allotropes synthesised on metallic substrates, we propose to produce an extended database of simulated structures and their corresponding STM images, allowing a facilitated allotrope identification from experimental STM imaging using image classification. For that purpose, one needs a large database of accurate extended models comprising several unit cells of substrate and borophene – such large models are needed to allow describing Moiré patterns and/or borophene corrugation. In the first stage of this work, we have developed a new atomic potential using a machine learning approach [6–8], which allows us to explore multiple structural arrangements of borophene allotropes on metal substrates. The developed potential [9] presents the advantage of performing fast simulations with a level of accuracy comparable to ab initio calculations [10] – moreover, no classical potential pre-existed for this type of system.

In this talk, I will present the methodology to develop this machine learning potential as well as the various borophene allotropes that have been simulated on Ag surfaces. I will also show how structural properties of given allotropes can be retrieved from their STM images, and discuss the prospects of this work [9] for the prediction of structures on various metals, phase transitions, as well as allotrope identification through experimental vs simulated STM image comparison.

[1] A. Mannix et al., Nat. Nanotech 13 (2018), 444–450

[2] A. Mannix et al., Science 350 (2015), 1513–1516

[3] B. Kiraly et al., ACS Nano 13 (2019), 3816–3822

[4] B. Feng et al., Nat. Chem 8 (2016), 563–568

[5] M. Cuxart et al., Sci. Adv. 7 (2021), eabk1490

[6] J. Behler & M. Parrinello, Phys. Rev. Lett. 98 (2007), 146401

[7] J. Behler, Int. J. Quantum Chem. 115 (2015), 1032-1050

[8] A. Singraber et al., J. Chem. Theory Comput. 15 (2019), 3075–3092

[9] P. Mignon et al., J. Am. Chem. Soc. (2023)

[10] A. Singraber et al., J. Chem. Theory Comput. 15 (2019), 1827–1840

Contact : tellier@ill.fr



Prévenir un ami par email

Télécharger dans mon agenda

Cafés sciences de Grenoble | UdPPC de Grenoble | Sauvons Le Climat | Cafe des sciences de Vizille
Accueil du site | Secretariat | Espace privé | Suivre la vie du site RSS 2.0 : Tous les evenements Suivre la vie du site RSS 2.0 : Evenements de Physique Suivre la vie du site RSS 2.0 : Evenements de Biologie & Chimie