Machine learning predictions of diffusion in bulk and confined ionic liquids using simple descriptors

Bobbitt, N. Scott; Allers, Joshua P.; Harvey, Jacob A.; Poe, Derrick; Wemhoner, Jordyn D.; Keth, Jane; Greathouse, Jeffery A.

doi:10.1039/d3me00033h

Cited by 5 publications

(2 citation statements)

References 98 publications

(163 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The polarizable MD simulations have demonstrated the capacity to calculate SDCs with higher accuracy compared to nonpolarizable MD simulations, whereas polarizable MD simulations require more computational cost than nonpolarizable MD simulations. To significantly reduce the computational resources required for generating diffusion data through MD simulations, Bobbitt et al 26 employed ML methods to accurately predict diffusion properties of 29 ILs at temperatures from 350 to 500 K based on the MD simulation data. In addition to MD simulations, several semi‐empirical methods were developed to describe the SDCs of ILs.…”

Section: Introductionmentioning

confidence: 99%

Modeling self‐diffusion coefficients of ionic liquids using ePC‐SAFT and FVT combined with Einstein relation

Zuo,

Lu,

2024

AIChE Journal

View full text Add to dashboard Cite

The electrolyte perturbed‐chain statistical associating fluids theory (ePC‐SAFT) coupled with free volume theory (FVT) was combined with Einstein relation, that is, ePC‐SAFT‐FVT‐E, to describe self‐diffusion coefficients (SDCs) of ionic liquids (ILs). In modeling, ePC‐SAFT was used to calculate density, while FVT parameters, determined from viscosity data, were utilized to calculate the summation of ionic SDCs through the Einstein relation with one parameter. Two strategies were employed to determine this parameter: fitting experimental data for each IL or estimating a universal parameter from van der Waals volume. Comparative analysis reveals good agreement with experimental data, with average absolute relative deviations (ARDs) of 8.14% (strategy 1) and 10.29% (strategy 2). Subsequently, cationic and anionic SDCs were reliably determined from the summation of ionic SDCs, with average ARDs of 10.80% and 10.21%, respectively. This study indicates the ePC‐SAFT‐FVT‐E model, employing viscosity‐derived parameters and three universal parameters, reliably predicts SDCs of ILs in wide temperature and pressure ranges.

show abstract

Section: Introductionmentioning

confidence: 99%

Modeling self‐diffusion coefficients of ionic liquids using ePC‐SAFT and FVT combined with Einstein relation

Zuo,

Lu,

2024

AIChE Journal

View full text Add to dashboard Cite

show abstract

“…In-depth research in this area has been conducted for years. Molecular dynamics were combined with artificial neural network (ANN) using simple or complex combinations of descriptors to predict the diffusion behavior of a single stream or two-phase fluid accurately and quickly, and the advantages and disadvantages of ANN for diffusion prediction were analyzed in detail. − The main focus of these works was on the selection of feature descriptors (derived from the LJ potential or radial distribution function), and the effects of different features on fluid diffusion were analyzed in detail. It is demonstrated that accurate models can be obtained by using simple physical descriptors.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning-Assisted Exploration of a Two-Dimensional Nanoslit for Blast Furnace Gas Separation

Huan,

Qiu,

Sun

et al. 2023

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Diffusion-induced gas separation is crucial for industrial applications, while the determination of specific conditions is still challenging. Here, molecular dynamics simulation data were used to train machine learning models to identify the effective separation conditions for blast furnace gas confined in nanosilts with different absorption strengths (graphene and graphene oxide). The diffusion coefficients and exponents of the blast furnace gas were obtained as a database by molecular dynamics (MD) simulations. Several environmental and structural controlling factors (such as temperature, layer distance, atomic number, ionization potential, etc.) were extracted through importance analysis. And the relationships between these factors and diffusion properties were further established by the Kernel Ridge Regression algorithm. Based on the differences in diffusion coefficients, specific binary gas mixtures in the blast furnace gas with competitively high separation potential have been screened out by a trained machine learning model and verified by MD simulations. The simulation strategy can provide theoretical guidance for the structural design of membranes for diffusion-controlled gas separation.

show abstract

Screening HFC/HFO and ionic liquid for absorption refrigeration at the atomic scale by the prediction model of machine learning

Chu,

He,

Kontogeorgis

et al. 2024

Green Chemical Engineering

View full text Add to dashboard Cite

Machine learning predictions of diffusion in bulk and confined ionic liquids using simple descriptors

Abstract: Ionic liquids have many intriguing properties and widespread applications such as separations and energy storage.

Cited by 5 publications

References 98 publications

Modeling self‐diffusion coefficients of ionic liquids using ePC‐SAFT and FVT combined with Einstein relation

Modeling self‐diffusion coefficients of ionic liquids using ePC‐SAFT and FVT combined with Einstein relation

Machine Learning-Assisted Exploration of a Two-Dimensional Nanoslit for Blast Furnace Gas Separation

Screening HFC/HFO and ionic liquid for absorption refrigeration at the atomic scale by the prediction model of machine learning

Contact Info

Product

Resources

About