Study of Binary ionic liquids by dielectric spectroscopy and Data Mining

Abstract

This work focuses on the physicochemical study and dielectric characterization of imidazolium-based ionic liquids, combining experimental analysis and data-driven modeling better to understand their structural, thermal, and electrical behaviors. The study integrates data mining techniques, including Principal Component Analysis (PCA) and Partial Least Squares Regression (PLS), with Broadband Dielectric Spectroscopy (BDS) and thermal analysis methods such as Differential Scanning Calorimetry (DSC) and Thermogravimetric Analysis (TGA). Using the Vogel–Fulcher–Tammann (VFT) dielectric equation, the PLS approach was applied to predict the dielectric properties of binary ionic liquid mixtures based on the known parameters of their individual components. The PCA method enabled the classification and visualization of property relationships, facilitating the identification of correlations and trends within complex datasets. These techniques provided valuable insight into the structure–property relationships governing ionic liquid behavior. Experimentally, BDS measurements were performed on 1-butyl-3-methylimidazolium tetrafluoroborate ([BMIM][BF₄]) and other imidazolium-based systems to investigate frequency- and temperature-dependent dielectric properties. The results revealed electrode polarization effects, relaxation dynamics, and scaling laws characteristic of ion exchange processes between interfacial layers and the bulk liquid. The relaxation peaks followed a Vogel–Fulcher–Tammann temperature dependence, confirming the kinetic nature of charge transport mechanisms. Furthermore, the thermal and dielectric behaviors of two newly synthesized ionic liquids, [EtOHVIM][HSO₄] (monomeric) and [P-EtOHVIM][H₂PO₄] (polymeric), were investigated using NMR, FTIR, DSC, and TGA. Both materials exhibited good thermal stability, low glass transition temperatures, and strong hydrogen bonding networks that influence their hydrophilicity and conductivity. Dielectric analysis showed high ionic conductivities with temperature-dependent transitions from VFT to Arrhenius-type behavior. Overall, this study demonstrates that combining data mining methods with dielectric and thermal analyses offers a comprehensive and predictive framework for understanding and designing ionic liquids. The results highlight the potential of [EtOHVIM][HSO₄] and [P-EtOHVIM][H₂PO₄] as promising candidates for solid-acid electrolytes and proton exchange membranes, contributing to the development of next-generation electrochemical and energy storage materials.