Orateur : Dr. Thijs Stuyver (Chimie ParisTech, PSL)

Résumé :

Machine learning (ML) has had a significant impact on various subfields of science in recent years. Part of the reason for the success of ML is that it enables the generation of predictive models with only limited domain knowledge: usually, only minor modifications to a generic ML algorithm needs to be made to generate effective models for a specific application. That is of course under the condition that sufficient data is available, and then we usually mean hundreds of thousands or even millions of datapoints. In chemistry, there are several predictive tasks for which we have this abundancy of data, but for most specialized applications – particularly those related to chemical reactivity – we do not have this luxury.

              In principle, computational chemistry offers a way out when limited experimental data is available, since it enables data generation in a cheap and easy-to-automate manner. In the first part of my talk, I will explore this approach in a bit more detail and focus specifically on an ML accelerated computational workflow to screen for promising bioorthogonal click reactions that I recently developed.

              While quantum chemical simulations of reactivity tend to be relatively cheap compared to experimental characterizations, the cost of generating sufficient training data for a machine learning model still becomes prohibitive, fast. As such, in the second part of my talk I will discuss strategies to improve the data efficiency of ML-based computational workflows for reactivity prediction. Specifically, I will focus on models based on intermediate valence bond inspired representations, and demonstrate that these outperform conventional machine learning models by a wide margin for hydrogen atom transfer reactions in the low data regime.

References

• Casetti, N.; Alfonso-Ramos, J. E.; Coley, C. W.*; Stuyver, T.*, Combining Molecular Modeling and Machine learning for accelerated reaction screening and discovery, Chem. Eur. J. 2023, e202301957.
• Stuyver, T.*; Jorner, K.; Coley, C. W.*, Reaction profiles for quantum chemistry-computed [3+2] cycloaddition reactions, Sci. Data 2023, 10, 66.
• Stuyver, T.*; Coley, C. W.*, Machine learning‐guided computational screening of new candidate reactions with high bioorthogonal click potential, Chem. Eur. J. 2023, e202300387.
• Alfonso-Ramos, J.; Neeser, R.; Stuyver, T. Repurposing Quantum Chemical Descriptor Datasets for on-the-Fly Generation of Informative Reaction Representations: Application to Hydrogen Atom Transfer Reactions, ChemRxiv 2023, 10.26434/chemrxiv-2023-2n281.

Orateur : Vikram Deshpande, University of Utah, USA

Résumé : Topological materials have burgeoned of late due to their implications for the fields of electronics, spintronics and quantum computing, among others. While their electronic properties are important in their own right, they can also couple in fascinating ways to the lattice. We have developed techniques to deform materials controllably and study their resulting electronic properties in-situ, while complementarily sensing the electronic ground state through the mechanical degree of freedom. In this talk, by way of introduction, I will first present purely electrical measurements on the prototypical topological material, the three-dimensional (3D) topological insulator (TI), wherein we hybridize Dirac cones of 3D TI surfaces controllably to realize the quantum spin Hall effect in the ultrathin limit. Then I will present our recent results applying the above-mentioned mechanical techniques to two different topological materials, namely twisted bilayer graphene (TBG) and the intrinsic magnetic topological insulator (MTI) MnBi2Te4, respectively. We are able to tune the Hofstadter’s spectrum of non-magic angle TBG and induce magnetism in non-magnetic correlated insulating states of magic-angle TBG using isotropic strain, for example, and detect various magnetic states and measure magnetoelastic couplings in the case of MTIs. Our advances present unique routes to tuning and sensing the parameter space of these exciting materials.

Contact : Stéphane BERCIAUD (berciaud@unistra.fr)

Orateur : Benjamin Besga, ILM Lyon

Résumé : The aim of stochastic thermodynamics is to study small non-equilibrium systems subject to thermal fluctuations. Some results from this field will be illustrated using experiments carried out on opto-mechanical systems, mainly colloidal particles in an optical trap. Using non-equilibrium statistical physics will see how we can accelerate the natural dynamics of a system, shorten the mean first passage time on a target, or measure the forces acting on a non-equilibrium probe. Finally, we’ll ask how we can interrogate the quantum limit of these results by looking at the opto-mechanical coupling of a self assembled supercrystal of quantum dots in an optical trap.

Dr. Veiko Karu (responsable de projet principal pour les projets de recherche et développement de l’EIT à l’Université de Technologie de Tallinn (TalTech)

Orateur : Dr. Murielle Chavarot-Kerlidou (Univ. Grenoble Alpes, CNRS, CEA, IRIG, Laboratoire de Chimie et Biologie des Métaux (UMR 5249), F-38000, Grenoble, France.)

Résumé :

Making fuels and chemicals from sunlight and abundant raw materials such as water and CO₂ represents a major challenge to meet for a clean energy future. To achieve sunlight-driven water splitting, the integration of molecular H₂-evolving catalysts into photoelectrochemical cells is a promising, yet challenging strategy. Our contribution to the field will be highlighted through different examples covering all aspects from catalyst design to the construction of functional water-splitting devices. Photoelectrochemical activity of dye-sensitized NiO photocathodes based on covalent dye – cobalt catalyst assemblies will be discussed in detail, together with some key insights into the factors currently affecting their performances, obtained from advanced spectroscopy and post-operando characterizations

Orateur : Franklin Luis dos Santos Rodrigues, University of Stuttgart

Résumé :

Quantum thermodynamics allows for the interconversion of quantum coherence and mechanical work. Quantum coherence is thus a potential physical resource for quantum machines. However, formulating a general nonequilibrium thermodynamics of quantum coherence has turned out to be challenging. In particular, precise conditions under which coherence is beneficial to or, on the contrary, detrimental for work extraction from a system have remained elusive. We here develop an approach that allows us to study the far-from-equilibrium thermodynamics of coherence. We concretely derive generalized fluctuation relations and a maximum-work theorem that fully account for quantum coherence at all times, for both closed and open dynamics. We obtain criteria for successful coherence-to-work conversion, and identify a nonequilibrium regime where maximum work extraction is increased by quantum coherence for fast processes beyond linear response

Contact :
Contact : Martin Bowen (martin.bowen@ipcms.unistra.fr)

Fernanda de Avila Abreu (Microbiology Institute, Federal University of Rio de Janeiro (UFRJ)

Résumé :

Magnetotactic microorganisms are capable of synthesizing membrane-enclosed intracytoplasmic magnetic nanocrystals. The ability to produce these structures is shared among living beings that use the geomagnetic field as a form of guidance in migratory processes. Due to their structural simplicity concerning macroorganisms, magnetotactic bacteria have been widely studied regarding the synthesis of magnetic nanocrystals and magnetic orientation. The nanocrystals produced by magnetotactic bacteria have a narrow size range, consistent shape, chemical purity, crystallographic perfection, and stable properties. Furthermore, the synthesis process is considered entirely sustainable. Therefore, scaling up the production of these magnetic nanoparticles of biological origin, as well as innovative biotechnological approaches, are being studied towards a sustainable future.

Contact : Ovidiu ERSEN,  ovidiu.ersen@ipcms.unistra.fr