KUTE: Green-Kubo Uncertainty-Based Transport Coefficient Estimator.

Area of Science:

• Computational chemistry
• Materials science
• Chemical engineering

Background:

• Calculating transport properties from molecular dynamics (MD) simulations is crucial for understanding material behavior.
• Existing methods often rely on arbitrary cutoffs or external parameters, introducing uncertainty.
• The Green-Kubo (G-K) formalism and Einstein relations are common theoretical frameworks.

Purpose of the Study:

• To introduce and evaluate a novel algorithm, kute, for calculating transport properties from MD simulations.
• To assess the performance of kute against established methods.
• To address the limitations of arbitrary parameters in transport property calculations.

Main Methods:

• Developed the kute algorithm, which estimates integrals from the Green-Kubo theorem.
• Incorporated uncertainty quantification of correlation functions to avoid arbitrary cutoffs.
• Tested kute's performance using MD simulations of a protic ionic liquid for various transport properties.

Main Results:

• kute demonstrated comparable accuracy to the Einstein relations for the studied transport properties.
• kute outperformed other Green-Kubo-based methods in accuracy.
• The algorithm effectively handles uncertainties in correlation functions.

Conclusions:

• The kute algorithm provides a robust and accurate method for calculating transport properties from MD simulations.
• kute offers an improvement over existing Green-Kubo implementations by mitigating the impact of arbitrary parameters.
• This method enhances the reliability of transport property predictions in materials simulations.