You are here : Home > L_Sim team > Multiscale modeling of lithium transport in solid and hybrid Li-ion electrolytes and their interfaces

Joël Martin Dalmas

Multiscale modeling of lithium transport in solid and hybrid Li-ion electrolytes and their interfaces

Published on 18 December 2023
Thesis presented December 18, 2023

Abstract:
Hybrid Solid Electrolytes (HSEs) offer a promising alternative to conventional liquid electrolytes in the field of Li-ion batteries. These HSEs incorporate ceramic fillers, typically in nanoparticle form, into polymeric electrolytes. This integration aims to address the primary challenge encountered by Solid Polymeric Electrolytes (SPEs): their lower conductivity when compared to alternatives such as liquid or ceramic electrolytes. However, it remains uncertain whether the addition of ceramic fillers to pure SPEs yields a positive impact. The literature presents two distinct sets of findings. The first, stemming from early experimental research conducted two decades ago, advocates a significant improvement in SPE conductivity through the incorporation of passive ceramic fillers such as silica or alumina across various concentrations and temperatures. Conversely, an opposing perspective has emerged, highlighting outcomes that demonstrate an adverse effect of ceramics on the ionic mobility within SPEs, particularly when the polymer is in its amorphous phase.
The ongoing debate in this field calls for a needed clarification. In this thesis, we seek to provide answers to a critical question: Does the inclusion of ceramic nanoparticles in Solid Polymeric Electrolytes enhance or impede ion mobility? To address this inquiry, we employ molecular dynamics simulation techniques to analyze two hybrid systems comprised of Polyethylene Oxide (PEO) as the polymer, LiTFSI as the lithium salt, and either silica or alumina as the ceramic components. Our approach involves classical molecular dynamics simulations using the OPLS-AA force field, enabling us to explore the dynamic behaviors and interactions of these materials over extended time scales, typically spanning tenths of nanoseconds. The force field parameters are examinated from various literature sources, each having undergone individual validation through comparisons with experimental data.
We carried out an analysis of their structural properties, closely examining their correlation with the dynamic behavior of ions. This analysis provides a detailed account of the shifts in the system's dynamics. Our results demonstrate a high precision in replicating the temperature-dependent behavior observed in experimental studies of pure SPEs. Moreover, our simulations reproduce the solvation mechanisms of the salt on PEO, serving as a robust validation of our findings for pure SPEs.
Our findings concerning the use of silica nanoparticles reveal a substantial reduction in conductivity upon their addition, regardless of the ionic concentration. Most of this reduction can be accounted for by the diffusion equation, resulting from the fact that the space occupied by the nanoparticles is made inactive and unable to sustain ionic diffusion, interupting the movement of the ions. We identify two distinct concentration regimes: one above and one below a threshold concentration of 2 mol/L, which coincides with the point of maximum conductivity. These regimes exhibit contrasting ionic distributions and coordination properties among species. In the low-concentration regime, lithium ions are predominantly coupled to oxygen atoms within the PEO, leading to its saturation at 2 mol/L. In the second regime, the surplus of lithium ions interacts with TFSI anions, influencing interactions among other ions in the system.
The absence of conductivity enhancement observed in our simulations aligns with recent experimental measurements, contrary to earlier reports on hybrid ceramic/polyethylene-oxide electrolytes. Similar outcomes are evident in our results for alumina nanoparticles. In the specific case of alumina nanoparticles, we explored the utilization of a new set of force field parameters, resulting in significant alterations in the internal organization of the electrolyte. Despite these variations, our simulations consistently indicate a reduction in conductivity upon the addition of alumina nanoparticles.

Keywords:
Electrolytes, Interfaces, Lithium, Hybrid, Molecular Dynamics, Nanoparticles