Browsing by Author "Wilson, Tayla Lee"

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Open Access
    Development and application of the CL&Pol polarisable force field for ionic liquid-based electrolytes
    (2025) Wilson, Tayla Lee; Venter, Gerhard
    Ionic liquids (ILs) are a fascinating class of molecular systems due to diverse and promising applications, including the potential use as electrolyte systems, replacing traditional volatile organic electrolytes. The appeal of ILs for such applications lies in the favourable properties, such as high diffusivity, thermal and electrochemical stability, and low volatility. Molecular dynamics (MD) simulation is a powerful tool for studying the physical properties of liquids; however, whereas traditional solvents are typically well described using classical, fixed-charge force fields (FFs), explicit inclusion of polarisation is essential for accurate description of IL dynamics. Consequently, the development and application of polarisable FFs for ILs is a necessary focus within IL research. The first aim of this work was therefore to extensively validate the recently developed Drude-based CL&Pol FF for six pure ILs and five alkali-earth containing IL-electrolytes. Reproducibility and precision are not often addressed when MD simulation is used to calculate thermophysical properties, yet without quantification of uncertainty, the value of a validation study is questionable. Hence, statistically meaningful uncertainties were reported for all properties as a 95 % confidence interval of the mean over replicate simulations. The simulation protocol was further validated by using well-known theoretical relationships (e.g., the Stokes Einstein and Nernst-Einstein equations) to confirm the internal consistency of key calculated transport properties. The accuracy of calculated properties of pure ILs varied, with average errors as low as 1 % for density to 35 % for viscosity, and 50 % for conductivity. Most properties could be calculated with uncertainties of ~20 %, while calculated conductivities had uncertainties of ~50 %. The second aim of this work involved the further development of the van der Waals component of the intermolecular interaction. The CL&Pol Lennard-Jones (LJ) parameters are carried over from its fixed-charge predecessor, CL&P. The parameters are then adjusted with prescribed scaling of the LJ well-depth (ε) to remove the induction contribution, making it transferable to the Drude FF featuring explicit polarisation. While scaling of ε produces reasonably accurate ii transport properties, the resulting induction-free LJ potentials for interactions involving the alkali metals do not reproduce the ab initio exchange-dispersion potentials, producing theoretically unsatisfactory van der Waals interactions. Furthermore, the existing OPLS ε parameters for the alkali metals do not correlate with the strength of the ab initio dispersion interactions, warranting reconsideration of these parameters. Thus, this work presents a stable and robust protocol for obtaining van der Waals potential parameters compatible with a Drude FF without the need of scaling, based on first-principles resolution of the dispersion and exchange components of the potential using Symmetry-Adapted Perturbation Theory (SAPT). However, while overall interaction energies can be accurately obtained with small basis sets, this is due to error cancellation in the components. Consequently, the f2+SAPT0 complete basis set (CBS) methodology was developed using the extended AHB21x5 dataset of anion-neutral dimers. This method uses basis set extrapolation and scaling of SAPT0 components to provide dispersion and exchange components with an average deviation of ~10 % from higher order SAPT2+/CBS equivalents. Finally, investigation into various potentials showed that Halgren's buffered LJ potential provides a description of the van der Waals interactions more consistent with the equivalent f2+SAPT0/CBS components, particularly at short range, than is offered by the CL&Pol scaled LJ potential.
  • Thumbnail Image
    Item
    Open Access
    Prediction of isobaric heat capacities of room temperature ionic liquids
    (2020) Wilson, Tayla Lee; Venter, Gerhard A.
    Ionic liquids (ILs) have many potential applications that require knowledge of a variety of physical and thermodynamic properties. While these properties can often be determined experimentally, this is impossible for novel, yet to be synthesised ILs; thus, property prediction from first principles is essential to unlock new developments in the rational design of ILs. The isobaric heat capacity (CP ) is an important thermodynamic property that quantifies the amount of heat needed to increase the temperature of a material and is thus of great importance in engineering applications involving the design of heat-transfer systems. From a theoretical viewpoint, the heat capacity is a fundamental quantity that expresses the temperature dependence of enthalpy and entropy. Several models for the prediction of CP have been developed to date; however, these are often trained on limited data sets and published model performance is largely dependent on the judicious choice of the testing data. Moreover, popular techniques such as group contribution methods (GCMs) cannot always be applied to structurally novel ILs and quantitative structure property relationships (QSPRs) are highly dependent on the diversity of training data. In this work, predictive models for CP have been developed using linear and nonlinear machine-learning methods. A large data set of 2463 temperature-dependent CP values, spanning 208 ILs, was obtained from the ILThermo database. Molecular volumes, features based on the electrostatic potential (ESP) and other molecular descriptors were calculated for each cation and anion in the data set. Following this, several multiple linear regression models were developed, for which Lasso regression was used reduce the number of features, where necessary. The models were developed using a methodology that attempts to reduce the dependency of the results on the identity of the specific species in the training set. The complexity of these models was gradually increased from a simple volume-based model (inspired by the success of the Volume Based Thermodynamics (VBT) approach of Glasser and Jenkins [L. Glasser and H. D. B. Jenkins, Chem. Soc. Rev., 2005, 34, 866], which was applied to ionic liquids and augmented by Krossing and co-workers [W. Beichel et al., J. Mol. Liq., 2014, 192, 3]), to the addition of electrostatic potential surface areas and finally including the General Interaction Properties Functions (GIPFs) of Murray and Politzer [J. S. Murray et al., J. Mol. Struct. (THEOCHEM), 1994, 307, 55], which are statistically well-defined quantities derived from ESP data, and a Feed Forward Neural Network (FFNN) was developed using the most effective of the aforementioned feature sets. In addition to reporting test-set errors, an external data set was carefully compiled, containing ILs with components (either the cation or anion) not present in the training data, and structurally distinct. This was done to assess the general applicability and flexibility of the final models, and to allow for a fair comparison of model performance. Of the linear models developed, that using interacting features consisting of molecular volumes and GIPFs produced the lowest errors; this is likely due to the ability of the interaction features to describe intermolecular interactions between cations and anions. Consequently, molecular volumes and GIPFs features were also used to develop a nonlinear FFNN. Finally, the linear interacting GIPFs model and FFNN also produced the lowest errors of 3.2 1.5% and 3.8 2.4%, respectively, when applied to the external data set.