Understanding Interfaces of Molybdenum Disulfide and Ionic Liquids Using Electronic Structure Methods

Neuromorphic devices and circuits mimic the structure and function of the nervous system and are desirable due to their ability to learn and make decisions in an energy efficient manner. Advancements in neuromorphic computing could lead to accelerated progress and expanded possibilities of what is feasible with the current computational methods available. Recent work in this field focuses on integrating two-dimensional layered semiconductors and organic ionic liquids for applications in neuromorphic devices. One such system of interest incorporates thin film Molybdenum Disulfide (MoS₂) and an organic ionic liquid (IL). However, current experimental methods do not provide a complete understanding of the system, its properties, and its characteristics. This necessitates the application of computational methods to fill current gaps in knowledge about MoS₂-IL composites.
In this work, Density Functional Theory (DFT) calculations were employed to analyze the nanoscale structure present within the composite. These composites contain a thickness distribution of MoS₂ flakes including monolayer, bilayer, and beyond, all of which interact differently with the ionic liquid cation-anion pairs within the material. DFT relaxation calculations, with continuum solvation effects and van der Waals approximations, are used to parse equilibrium configurations of both adsorbed and intercalated ion pairs within these systems to supplement operando X-ray data and understand the material’s structure on the atomic scale.
Frequency and spatial dependent optical properties are extremely valuable for understanding how the composite properties differ away from the MoS₂ surface under both static and non-static conditions. Accurately capturing the dielectric constant as a function of frequency, however, is not feasible using standard DFT. In order to properly take quasiparticle interactions and partial coulombic screening into account, large-scale GW calculations were carried out using the Without Empty STates (WEST) code developed at the Midwest Center for Computational Materials (MiCCoM) . These calculations were performed on the DFT-relaxed structures in order to obtain the dielectric function as both a function of frequency and space, linking different structures with desirable properties for neuromorphic computing. While molecular dynamics is needed to capture the ionic dielectric response, many body perturbation theories such as GW capture the electronic screening which is critical for a semiconductor-organic interface.
Using these calculations, finite element models were developed. Experimental collaborators validated computed values by collecting spectral ellipsometry data and data from MoS₂-IL composites to validate the modeled system. This study provides fundamental insights that can help guide the development of future neuromorphic computing technologies.