The ability to predict transport properties, such as viscosity, through molecular simulation is a valuable advantage. Obtaining reliable estimates for these properties can be more challenging compared to equilibrium properties. Our manuscript focuses on an in-depth discussion of two prominent methods for predicting transport properties: equilibrium molecular dynamics (EMD) using the Green–Kubo approach and nonequilibrium molecular dynamics (NEMD) with the periodic perturbation method. These methods were utilized to calculate the shear viscosity of highly viscous room-temperature ionic liquids (RTILs), specifically 1-ethyl-3-methylimidazolium tetrafluoroborate [EMIM][BF4] and 1-butyl-3-methylimidazolium tetrafluoroborate [BMIM][BF4]. Additionally, we assessed both all-atom and coarse-grained models for describing the RTILs, as well as examined the impact of various parameters such as barostat, ensemble, acceleration step, length, and number of trajectories on viscosity calculation. Our analysis revealed that the NEMD method provided the most accurate predictions of shear viscosity for short-chain imidazolium ionic liquids. The influence of barostat and ensemble choice was found to be minimal, and reducing the acceleration step in the NEMD method can help decrease simulation time while maintaining accuracy.
