A Generalized Machine Learning Model for Predicting Ionic Conductivity for Ionic Liquids

Dhakal, Pratik; Shah, Jindal K.

doi:10.26434/chemrxiv-2022-nzrr3

Cited by 2 publications

(3 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the model development, we employed the XGBoost ensemble-based method as we found it to perform better than the random forest (RF) method in our previous study. 11 Another reason for choosing an ensemble-based method over a 'black-box' model such as a neural network is to get further insights on the importance of the feature for HOMO/LUMO energy prediction. The hyper-parameters for the XGBoost model are determined using a randomized grid search method with a 5-fold CV implemented in Scikit-learn.…”

Section: Machine Learningmentioning

confidence: 99%

“…[7][8][9] However, several essential electrolyte properties, such as ionic conductivity and electrochemical stability, must be thoroughly understood before using them for battery applications. In our earlier work we focused on developing machine learning models for ionic conductivity predictions and identifying ionic liquids possessing high conductivity through unique combinations of cation and anions 10,11 and binary ionic liquid formation. 10 In this research article, we focus on the electrochemical stability of ionic liquid cations.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Mapping the Frontier Orbital Energies of Imidazolium-based Cations Using Machine Learning

Dhakal

Gassaway

Shah

2023

Preprint

View full text Add to dashboard Cite

The knowledge of frontier orbital, highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO), energies is vital for studying chemical and electrochemical stability of compounds, their corrosion inhibition potential, reactivity, etc. Density functional theory (DFT) calculations provide a direct route to estimate these energies, either in the gas-phase or condensed phase. However, the application of DFT methods becomes computationally intensive when hundreds of thousands of compounds are to be screened. Such is the case when all the isomers for the 1-alkyl-3-alkyllimidazolium cation [C$_n$C$_m$im]$^+$ (n = 1-10, m=1-10) are considered. Enumerating the isomer space of [C$_n$C$_m$im]$^+$ yields close to 316,000 cation structures. Calculating frontier orbital energies for each would be computationally very expensive and time-consuming using DFT. In this article, we develop a machine learning model based on extreme gradient boosting (XGBoost) method using the a small subset of the isomer space and predict the HOMO and LUMO. Using the model, the HOMO energies are predicted with a mean-absolute error (MAE) of 0.4 eV and the LUMO energies with MAE of 0.2 eV. Inferences are also drawn on type of the descriptors deemed important for the HOMO and LUMO energy estimates. Application of the machine learning model results in a drastic reduction in computational time required for such calculations.

show abstract

Section: Machine Learningmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Mapping the Frontier Orbital Energies of Imidazolium-based Cations Using Machine Learning

Dhakal

Gassaway

Shah

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…For example, atomistic simulations of the bulk solvent could produce close-to-experiment estimates for various bulk properties, e.g., mass density and viscosity. [10][11][12][13][14][15] As the accuracy level of simulation outcomes depends on the quality of the Hamiltonian (ranging from fixed-charge force fields, polarizable models to the GPW treatment) and also the sampling convergence, the predictive power exhibits obvious system dependence. Further, large-scale investigations of a spectrum of ion-pair compositions could be computationally demanding.…”

Section: Introductionmentioning

confidence: 99%

Chemical Accuracy Prediction of Molecular Solvation and Partition in Ionic Liquids with Educated Estimators

Zhang,

Wang,

et al. 2023

Preprint

View full text Add to dashboard Cite

Ionic liquids (ILs) derivatives as novel green solvents are widely employed in laboratory and industrial applications. Many computational tools have been developed to compute various physiochemical properties of ILs species, e.g., mass density and viscosity. However, despite their central role as solvents, predictive tools for the solvation of external agents and the partition between water and ILs phases are rather limited. Common selections are atomistic simulations, especially the alchemical method. However, the relatively high computational costs could be harmful to large-scale applications. In this work, we develop a series of low-cost machine-learning estimators to exploit further the chemical-accuracy frontier. 1764 solvation and 1764 water-ILs transfer free energies involving 120 gaseous or drug-like solutes and many commonly applied ILs families are gathered from experimental references to form the dataset with a diverse coverage of chemical spaces. The best-performing tree-based method could predict simultaneously solvation and water-ILs biphasic partition thermodynamics with a state-of-the-art performance of ~0.13 kcal/mol MAE, ~0.25 kcal/mol RMSE, ~0.98 Pearson r and ~0.94 Kendall τ, beating transferable tools such as alchemical free energy calculations in a series of face-to-face comparisons.

show abstract

A Generalized Machine Learning Model for Predicting Ionic Conductivity for Ionic Liquids

Cited by 2 publications

References 66 publications

Mapping the Frontier Orbital Energies of Imidazolium-based Cations Using Machine Learning

Mapping the Frontier Orbital Energies of Imidazolium-based Cations Using Machine Learning

Chemical Accuracy Prediction of Molecular Solvation and Partition in Ionic Liquids with Educated Estimators

Contact Info

Product

Resources

About