Periodic Reporting for period 4 - AMPERE (Accounting for Metallicity, Polarization of the Electrolyte, and Redox reactions in computational Electrochemistry)
Période du rapport: 2022-10-01 au 2023-09-30
Applied electrochemistry plays a key role in many technologies, such as batteries, fuel cells, supercapacitors or solar cells. It is therefore at the core of many research programs all over the world. Yet, fundamental electrochemical investigations remain scarce. In particular, electrochemistry is among the fields for which the gap between theory and experiment is the largest. From the computational point of view, there was no molecular dynamics (MD) software devoted to the simulation of electrochemical systems while other fields had dedicated tools.
This was due to the difficulty of accounting for complex effects arising from (i) the degree of metallicity of the electrode (i.e. from semimetals to perfect conductors), (ii) the mutual polarization occurring at the electrode/electrolyte interface and (iii) the redox reactivity through explicit electron transfers. Our objective was to fill this gap by introducing a whole set of new methods for simulating electrochemical systems. Our molecular dynamics code, MetalWalls, includes all the developments made within the AMPERE project in that direction. It is available under an Open Source license (https://gitlab.com/ampere2/metalwalls). From the software engineering point of view, important efforts were made to adapt MetalWalls to supercomputers, which was acknowledged by the award of the 2021 special prize “Joseph Fourier” by ATOS.
First applications of the new features of MetalWalls aimed at the discovery of new electrolytes for energy storage. Here we have focused on: (i) ‘‘water-in-salts’’ to understand why these revolutionary liquids enable much higher voltage than conventional solutions (ii) redox reactions inside a nanoporous electrode to support the development of future capacitive energy storage devices. Our simulations have allowed to provide to the experimental groups a much better view of the structure of the liquids and of its link with the physical properties, the interfacial reactivity, and the devices performances.
This was due to the difficulty of accounting for complex effects arising from (i) the degree of metallicity of the electrode (i.e. from semimetals to perfect conductors), (ii) the mutual polarization occurring at the electrode/electrolyte interface and (iii) the redox reactivity through explicit electron transfers. Our objective was to fill this gap by introducing a whole set of new methods for simulating electrochemical systems. Our molecular dynamics code, MetalWalls, includes all the developments made within the AMPERE project in that direction. It is available under an Open Source license (https://gitlab.com/ampere2/metalwalls). From the software engineering point of view, important efforts were made to adapt MetalWalls to supercomputers, which was acknowledged by the award of the 2021 special prize “Joseph Fourier” by ATOS.
First applications of the new features of MetalWalls aimed at the discovery of new electrolytes for energy storage. Here we have focused on: (i) ‘‘water-in-salts’’ to understand why these revolutionary liquids enable much higher voltage than conventional solutions (ii) redox reactions inside a nanoporous electrode to support the development of future capacitive energy storage devices. Our simulations have allowed to provide to the experimental groups a much better view of the structure of the liquids and of its link with the physical properties, the interfacial reactivity, and the devices performances.
From the methodological point of view, in a first step we have focused on the development of an approach in which a finite field is applied to an electrochemical cell made of a single electrode in contact with an electrolyte. It has two interesting features: i) it allows to switch from 2D to 3D periodic boundary conditions, which will accelerate the simulations. ii) the method is readily extended to perform finite electric displacement simulations, that is to simulate an electrochemical system under open circuit conditions. Another important methodological advance is the introduction of a mass-zero constrained molecular dynamics algorithm, which has the advantage to be symplectic and time reversible. Finally, we have developed the theoretical framework of a semiclassical Thomas-Fermi model to tune the metallicity of electrodes in molecular simulations. By systematically varyiing the Thomas-Fermi length, we have shown that all the interfacial properties are modified by screening within the metal: the capacitance decreases significantly and both the structure and dynamics of the adsorbed electrolyte are affected. In a last step, we have implemented an algorithm allowing to account for the polarization of the electrolyte together with the polarization of the electrode, as well as the possibility to fix the electrochemical potential of the atoms rather than the electrode potential, allowing to simulate multi-component materials.
In parallel, we have also simulated practical systems for energy storage and conversion applications. A large amount of work focused on the physico-chemical properties of the family of water-in-salts electrolytes. Firstly, we have studied the transport and interfacial properties of the most common water-in-salt, namely the one based on the LiTFSI salt. We have shown that the nanostructuration of the liquid impacts the diffusivities of all the species and the ionic correlations inside the melts. This work was further extended to the study of a large variety of water-in-salt anions. Our simulations have also shown an intriguing feature, namely the polymerization of lithium ions inside the liquid, which can have some important consequencies on the interfacial properties as well. We have also performed many simulations aiming at the interpretation of a series of experimental results: On the one hand, we have shown the occurence of aqueous biphasic systems when mixing water-in-salts with simpler aqueous electrolytes as well as the physical processes at the origin of their formation. On the other hand we have provided further understanding on the formation of a stable solid electrolyte interphase in water-in-salt-based Li-ion batteries by establishing a competitive salt precipitation/dissolution during the reduction of water at the interface. The mechanism of the anion decomposition was studied in detail through the use of ab initio molecular dynamics simulations. As an extension to this work, we have started a study on the effect of confining water in organic electrolytes (instead of ionic ones as in water-in-salts). We have shown that these systems provide very important information on the influence of the chemical speciation of water over its interfacial reactivity, with potential impact in the field of electrocatalysis.
The second series of systems that we targeted, biredox ionic liquids, needed more developments. In particular, there was no force field available to simulate them so we started by performing an (electronic) Density Functional Theory study of their electrochemical properties. The obtained data was then used to develop polarizable force fields in order to perform classical molecular dynamics simulations. We could thus study in detail the structure of the solvation shell of anthraquinone and TEMPO redox-active species in an acetronitrile solvent as well as in a pure ionic liquid. When simulating the pure biredox ionic liquid, we observed some peculiar nanostructural organization, with the formation of nanodomains enriched with the redox-active groups, which can strongly impact the performance of the supercapacitors and explain why higher than expected energy densities are obtained.
In parallel, we have also simulated practical systems for energy storage and conversion applications. A large amount of work focused on the physico-chemical properties of the family of water-in-salts electrolytes. Firstly, we have studied the transport and interfacial properties of the most common water-in-salt, namely the one based on the LiTFSI salt. We have shown that the nanostructuration of the liquid impacts the diffusivities of all the species and the ionic correlations inside the melts. This work was further extended to the study of a large variety of water-in-salt anions. Our simulations have also shown an intriguing feature, namely the polymerization of lithium ions inside the liquid, which can have some important consequencies on the interfacial properties as well. We have also performed many simulations aiming at the interpretation of a series of experimental results: On the one hand, we have shown the occurence of aqueous biphasic systems when mixing water-in-salts with simpler aqueous electrolytes as well as the physical processes at the origin of their formation. On the other hand we have provided further understanding on the formation of a stable solid electrolyte interphase in water-in-salt-based Li-ion batteries by establishing a competitive salt precipitation/dissolution during the reduction of water at the interface. The mechanism of the anion decomposition was studied in detail through the use of ab initio molecular dynamics simulations. As an extension to this work, we have started a study on the effect of confining water in organic electrolytes (instead of ionic ones as in water-in-salts). We have shown that these systems provide very important information on the influence of the chemical speciation of water over its interfacial reactivity, with potential impact in the field of electrocatalysis.
The second series of systems that we targeted, biredox ionic liquids, needed more developments. In particular, there was no force field available to simulate them so we started by performing an (electronic) Density Functional Theory study of their electrochemical properties. The obtained data was then used to develop polarizable force fields in order to perform classical molecular dynamics simulations. We could thus study in detail the structure of the solvation shell of anthraquinone and TEMPO redox-active species in an acetronitrile solvent as well as in a pure ionic liquid. When simulating the pure biredox ionic liquid, we observed some peculiar nanostructural organization, with the formation of nanodomains enriched with the redox-active groups, which can strongly impact the performance of the supercapacitors and explain why higher than expected energy densities are obtained.
Before the ERC AMPERE project started, it was only possible to simulate one type of electrode. Our methodological developments enable to simulate a large variety of systems, ranging from various metals to 2D materials. From our point of view, this is significant step towards the realistic modeling of complex energy storage devices. Among the various methods, although most of them (Thomas-Fermi model, polarizable force field for ionic liquids) were expected within the project, we also developed finite-field methods that were not initially planned. It turns out that these methods may become the most important result from the project since they can be used within the framework of density functional theory-based simulations, which would be a very significant advance. Recently, we have started to extend it to the case of machine-learning approaches, and the results are very promising.