Predictive Thermodynamic Models For Ionic Liquid Systems And Their Applications In Gas Dehydration Processes

As the most important predictive thermodynamic models, COSMO-RS(conductor-like screening model for real solvents) and UNIFAC-Lei are applied in the field of predicting the thermodynamic properties of ionic liquids(ILs). Choosing the suitable ionic liquid (IL) absorbent is the key of the gas dehydration process. Considering the great variety of ILs, it requires much time and effort to choose the suitable IL by experiment methods. The COSMO-RS model was used to screen the suitable IL for the gas dehydration process; moreover, gas solubility, selectivity, and other physical properties and chemical stabilities of ILs are considered in this work. For some gas-IL systems, their UNIFAC-Lei binary group interaction parameters are not exit.The UNIFAC-Lei binary group interaction parameters are obtained by fitting the experiment data of gas solubility in ILs in order to extend the current UNIFAC-Lei parameter matrix for ILs. In addition, the UNIFAC-Lei model was used to predict the gas-liquid equilibrium of the gas-IL systems in order to establish the rigorous equilibrium stage model for the gas dehydration processes and the mathematical model for the continuous gas dehydration processes.Firstly, in order to get the suitable COSMO-RS parameters for the IL systems, the misfit energy constant, hydrogen bond coefficient, effective contact surface area of a segment in the COSMO-RS model (ADF version)were adjusted to make the average relative deviation of the calculated value and experiment data to the minimum using the Newton-Raphson iteration method. Using the improved parameters in COSMO-RS model to predict the activity coefficients at infinite dilution of organic solutes in ILs, solubility of CO2 in pure ILs and binary mixtures of ILs at high and low temperatures, and VLE of binary systems of alkanes/alkenes and ILs, the average relative deviation of the calculated value and experiment data reduced obviously. In addition, screening the suitable IL (as the additive of extractant) for the extractive distillation of benzene and thiophene with DMF is taken as an example to verify the screening ability of the COSMO-RS model using the improved parameters. As an effective predictive thermodynamics model,UNIFAC-Lei is used to predict gas solubility in ILs. The UNIFAC-Lei binary group interaction parameters of CO2,CO,H2 and ILs can be obtained from literature. However, the UNIFAC-Lei binary group interaction parameters of some gases (CH4, C2H4, O2, H2S,N2O) and ILs are not exit. The UNIFAC-Lei binary group interaction parameters are obtained by fitting the experiment data of gas solubility in ILs to make the average relative deviation of the calculated value and experiment data to the minimum in order to extend the current UNIFAC-Lei parameter matrix for ILs.Secondly, IL is used in gas dehydration process, and the COSMO-RS model was used to screen the suitable IL for the gas dehydration process, and the gas dehydration technology with ILs was systematically studied ranging from the molecular level to industrial scale. The IL [EMIM][Tf2N] was selected to be the most suitable absorbent for CO2 gas dehydration from 285 ILs (15 cations and 19 anions) when considering gas solubility, selectivity, and other physical properties and chemical stabilities together. The corrosion effect of [EMIM][Tf2N] on engineering materials was also studied, and the results show that the engineering materials show a good corrosion resistance in pure[EMIM][Tf2N] and the mixture of [EMIM][Tf2N] + water. In addition, the molecular surface charge density, excess enthalpy and intermolecular forces were analyzed to research the mechanism of CO2 gas drying with[EMIM][Tf2N], and the results show that the hydrogen bond interaction energies and van der Waals dominate for the H2O + [EMIM][Tf2N] and CO2 +[EMIM][Tf2N] systems, respectively, and the binding energy between H2O and [EMIM][Tf2N] is significantly larger than that between CO2 and[EMIM][Tf2N].Thirdly, the solubility data of CO2 in pure [EMIM][Tf2N] and in the[EMIM] [Tf2N] + H2O mixture were measured to research the effect of H2O on the solubility of CO2 in [EMIM][Tf2N]. The results show that CO2 solubility in the [EMIM] [Tf2N] + H2O mixture is less than that in pure [EMIM] [Tf2N]under the same temperature and pressure. This is favorable for the gas dehydration process with ILs because the absorption loss of CO2 as gas product can be reduced in the presence of water. The vapor pressure for the binary mixture of [EMIM][Tf2N] and H2O at different temperatures was measured to research the effect of IL on the vapor pressure of water. It is seen that with the increase of mole fraction of water, the vapor pressure first increases, and then levels off at a certain value (close to the saturated vapor pressure of pure H2O at the same temperature), which is attributed to the demixing of IL and H2O at high H2O concentration. The reduction of vapor pressure of H2O with the addition of IL indicates the stronger affinity between[EMIM][Tf2N] and H2O, which disrupts the associating interaction between H2O and H2O molecules. In addition, The effect of IL flow rate and the water content in the entry IL on the water content in gas product is illustrated, and the results show that with the increase of IL volume flow rate, the water content in gas product first decreases, and then almost remains stable; low water content in the entry IL is beneficial to improving product quality. With the increase of water content in the entry IL, the water content in gas product also increases, exhibiting a linear relationship with the water content in the entry IL. The calculated results by Equilibrium Stage model with UNIFAC-Lei property model under the same experimental conditions are compared, and the results show that the experimental and simulated results agree well with the ARD 11.52%, indicating the validation of the simulation and the UNIFAC-Lei property model for predicting the thermophysical and equilibrium properties for the mixtures containing IL.Finally, the conceptual design of a continuous gas dehydration process with IL was conducted and optimized. The effect of operating parameters, i.e.,the number of theoretical stages (Nt) and operating temperature (T1 of absorption column, mass flow rate (m1L) of the entry IL, flash temperature (T2),and flash pressure (P2),on the separation performance (H2O content in the gas product) with [EMIM][Tf2N] as absorbent were investigated; in addition, the gas dehydration process with TEG was also simulated and optimized for comparison. The effect of operating parameters, i.e., the theoretical stage number of absorption column, mass flow rate of the entry TEG, reboiler temperature, and theoretical stage number of the desorption column on the separation performance (H20 content in the gas product and recovery ratio of CO2) and the mass fraction of lean TEG was investigated. It is seen that in the case of the same water content 489 ppm (mole fraction) in gas product, the flow rates of [EMIM][Tf2N] and TEG are 10000 and 14024 kg·h-1,respectively, and the recovery ratios of CO2 are 98.34% and 93.61%,respectively. It was found that for the gas dehydration process with the IL[EMIM][Tf2N], the total heating and cooling duties can decrease by as high as 80.57% and 80.06%, respectively, when compared with the conventional TEG.