Global warming is now widely recognized as being the most significant global issue facing human beings. Mitigating CO2 emission from fossil-fueled power plants as well as from transports has become an urgent and worldwide research topic, in which CO2 separation is often needed. Technologies have been developed and commercialized, whereas the cost is still high. Developing new technologies for CO2 separation is one focus research area. Ionic liquids (ILs) are promising absorbents for CO2 separation due to their very low vapor pressure, high solubility and selectivity for CO2 as well as low energy usage for solvent regeneration. A drawback of using IL for CO2 separation is the high viscosities, and using supported ILs has been proposed as a promising solution. This can take advantage of the high selectivity of gas in ILs, and also the high surface area of materials can reduce the impact of viscosity, improve the gas transfer, and hence increase the absorption rate. To develop IL-based technologies, thermodynamic properties (density, heat capacity, gas solubility, etc.), viscosity, and surface tension of ILs as well as the thermodynamic properties for confined ILs are the prerequisites. As the number of ILs that can be theoretically synthesized is up to an order of 1018, determining all the properties experimentally is impractical, not to mention the time-consuming with high cost. It is desirable to develop theoretical tools to predict the thermodynamic and transport properties of ILs and IL-containing mixtures in a wide temperature and pressure range. In our previous work, the framework of ion-specific electrolyte perturbed-chain statistical associating fluid theory (ePC-SAFT) has been developed with reliable results. The developed ePC-SAFT model was further combined with Free Volume Theory (ePC-SAFT-FVT) and Density Gradient Theory (ePC-SAFT-DGT or DGT-ePC-SAFT) to represent the viscosity and surface tension of ILs, respectively. However, the work is limited to the imidazolium-based ILs, and the model performance for other commonly used ILs is still unclear. It has been pointed out that the model with the parameters fitted to the experimental data may result in pitfalls, and further validation is needed. Meanwhile, DGT-ePC-SAFT for pure ILs needs to be further developed to be consistent with the ion-specific ePC-SAFT model and extended to IL-mixtures. To describe the properties of confined ILs, in our previous work, ePC-SAFT model was combined with DFT (classical Density Functional Theory) (i.e., ePC-SAFT-DFT) to describe the properties of the IL and CO2/IL confined in nanopores. Still, the algorithm based on equal mesh width leads to intensive computations, making it inefficient. In this thesis, these ion-specific ePC-SAFT-based models were further developed and extended to the systems containing the ILs which are composed of the IL-cations ([Cnmim]+, [Cnpy]+, [Cnmpy]+, [Cnmpyr]+, and [THTDP]+) and the IL-anions ([Tf2N]-,[PF6]-, [BF4]-, [tfo]-, [DCA]-, [SCN]-, [C1SO4]-, [C2SO4]-, [eFAP]-, Cl-, [Ac]-, and Br-). Before modeling the properties, a method and scheme were developed to investigate the pitfall when modeling IL and IL-gas systems with ePC-SAFT. All 96 ILs considered in the thesis and their binary mixtures with CO2, H2S, CO, O2, CH4, N2, and H2 were covered. The results show that ePC-SAFT with the density-fitted parameters is virtually free of the undesired pitfall for almost all ILs with only one exemption ([C8mpy][BF4]) and the parameters for [Cnmpy]+ may need to be modified in future work. Afterward, ePC-SAFT was used to predict gas solubilities and second-order thermodynamic properties, such as heat capacities, isothermal and isentropic compressibilities, speeds of sound, thermal expansion coefficients, and internal pressures. The model predictions were evaluated by comparing with the experimental data, showing reliable results. ePC-SAFT-FVT was used to model the viscosities of ILs and IL-mixtures and compared with the available experimental data. It shows the model can represent the viscosity of pure ILs in a wide temperature and pressure range, and the parameters obtained from pure ILs can be used to predict the viscosity of IL-mixtures reliably. The model performance of Cl-based ILs at low temperatures is poor, and the temperature-dependent FVT parameter may be used to improve the model results. DGT-ePC-SAFT can provide reliable results for pure ILs in a wide temperature range, and it can be further used to describe the density profile in the interface. Furthermore, spot-DGT-ePC-SAFT based on the approximate density profile was proposed to predict the surface tension of IL-IL and IL-CO2 systems, and the reliable predictions imply the promising of spot-DGT-ePC-SAFT. To calculate the properties of confined ILs efficiently, Chebyshev pseudo-spectral collocation method was applied to accelerate the ePC-SAFT-DFT calculation. The feasibility of accelerating the ePC-SAFT-DFT calculation with the Chebyshev pseudo-spectral collocation method was discussed for the confined IL-CO2 systems. It was found that the Chebyshev pseudo-spectral collocation method can improve the efficiency of ePC-SAFT-DFT calculation significantly.