Investigação de Eletrólitos para Armazenamento de Energia: Uma Abordagem Multiescala Utilizando a Teoria do Funcional da Densidade, Dinâmica Molecular e o Método dos Elementos Finitos

Abstract

Computational simulations play an important role in advancing new technologies for energy storage systems (ESSs), such as batteries and supercapacitors, as they allow us, for example, to predict material properties, facilitating experimental work, or even to understand characteristics that are not accessible through experiments, such as the detailed molecular structure of a system. Different methods can be applied depending on the type of study being pursued. In this work, three techniques are used to study problems related to ESSs, to explore different aspects of electrochemical systems, and to understand how each type of simulation can be used in the study of these systems. In the first work, Density Functional Theory (DFT) was used to investigate the structural characteristics of ionic liquids that can maximize the electrochemical stability window of electrolytes. We demonstrated that systems with high electrochemical windows can be obtained with the combination of aliphatic cations attached to electron-donating functional groups and anions attached to electron-withdrawing functional groups. In the second work, Molecular Dynamics (MD) was used to evaluate the influence of parameters (i) simulation box size, (ii) number of independent simulations, and (iii) charge scaling factor on simulation results. We observed that the box size and the number of independent simulations do not significantly affect structural properties; however, transport properties exhibit a strong dependence on these factors. Additionally, it was found that the scaling factor has a significant influence on all parameters. Finally, in the third work, the Finite Element Method (FEM) was used to explore the validity of the Poisson–Nernst–Planck equation for describing the electric double layer formed at the electrode/electrolyte interface. We demonstrated that the linear approximation of the Poisson-Nernst-Planck (PNP) equation is adequate for low applied potentials (≈ 25 mV); however, for higher potentials, the nonlinear solution of the PNP equation is necessary. With this work, it can be seen that for each type of problem to be studied, a specific computational method may be most suitable. Moreover, methods can be combined for a deeper understanding of the system; this combination of different methods is called multiscale simulation.