Abstract
Room-temperature ionic liquids (RTILs) are novel materials that can be used in separation processes in chemical industries such as liquid-liquid extraction (LLE) and CO2/CH4 separation. Due to the high cost of RTILs and limitations of experimental methods, in-silico methods are becoming more widespread in studying RTILs. In this research work, molecular dynamics (MD) simulations and density functional theory (DFT) calculations have been used to investigate the molecular basis of RTIL-based separation processes, particularly LLE of aromatics from aliphatics, and CO2/CH4 separation with RTIL-based composite membrane. In RTIL-based LLE process a method based on MD simulations was proposed, which calculates the number of molecules dissolved in each layer from the molecular structure generated by MD simulations. The proposed method successfully predicted the extractive properties such as selectivity and distribution ratio in LLE, where its reliability has been validated against experimental data. Afterwards, the extraction behaviour of seven different RTILs were tested using the proposed method where [C2mim][Tf2N] gives the best LLE performance. The deviations between the LLE performances of different RTIL extractants have been explored by the binding energies produced from DFT calculations. In CO2/CH4 separation, the mechanisms behind the selectivity enhancement by adding RTIL into ZIF-8 membrane (RTIL@ZIF-8) has been explored from the aspect of aperture configurations. The energy barrier values of CO2/CH4 penetrating the ZIF-8 aperture of different configurations have been obtained from the potential energy curves generated by DFT calculations. The MD-simulated results indicate that the presence of RTILs have stabilised the aperture configuration, which gives better selectivity compared to varied aperture configurations as evidenced from DFT calculations. Subsequently, the RTIL@ZIF-8 technique was further improved from the gas diffusion routes and feed CO2 concentration by MD simulation of CO2/CH4 diffusing through the ZIF-8 supercell. Results from MD simulations suggest that the gas molecules favour the route of minimum length, which may serve as objective of future membrane synthesis techniques. The selectivity value rises with increasing proportion of CO2 in feed mixture. When CO2 concentration achieves beyond 50%, the selectivity value still increases but does not ensure a better separation performance.