In this study, modeling and optimization works were completed for the separation of 99.9 mol% of methyl ethyl ketone from water through a pressure-swing distillation process since the azeotropic composition varies very sensitively with the change of system pressure. PRO/II with PROVISION release 9.1 was used for the computer simulation and Wilson activity coefficient model was chosen as a modeling equation. A pressure-swing distillation process can be classified into a low-high pressure columns configuration and a high-low pressure columns configuration. In this work, each configurations were optimized for the minimization of steam consumptions, respectively and were compared.

In this study, we compared with results simulated by EOS(Equation of State) using Peng-Robinson model and GERG-2004 model for estimating vapor pressure, latent heat of vaporation, liquid density, and binary isotherm vapor-liquid equilibrium on pure components composing natural gases. We obtained the simulated results that while EOS using GERG-2004 model is more accurate than EOS using Peng-Robinson model for estimating liquid density, but rather it is less accurate for estimating binary isotherm vapor-liquid equilibrium. On the other hand, the use of Costald model in EOS using Peng-Robinson model for increasing more accuracy to calculate liquid density is almost same as EOS using GERG-2004 model within the error of 1 % compared with experimental data. Also, we confirmed that on the estimation of binary isotherm vapor-liquid equilibrium, EOS using GERG-2004 model is more accurate than EOS using Peng-Robinson model, but they are almost same.

요 약 본 연구에서는 Dimethyl ether (DME) 생산 공정 중에 포함되어 있는 이산화탄소 제거를 위한 용매로써 메탄 올 수용액을 사용하는 공정에 대한 전산모사를 수행하였다. 공정모사를 위하여 Aspen tech 사의 Aspen Plus release 7.3을 사용하였으며, 열역학 모델식으로는 PC-SAFT 모델식을 사용하였다. PC-SAFT 모델식에서 필요한 이성분계 상 호작용 매개변수를 결정하기 위하여 실험 데이터를 수집하고 회귀분석을 통해 새롭게 결정하였으며, 결정한 매개변수 의 정확성은 실험 데이타와의 비교를 통해 검증하였다. 한편, 이러한 모델식과 검증한 매개변수를 사용하여 공정을 모델링 하였으며 최적 순환유량과 운전압력 그리고 원료 주입단 등을 결정하여 공정 최적화를 수행하였다.Abstract In this study, simulation works have been performed for the modeling of CO2 removal process contained in the DME production process through an absorber-stripper system using methanol aqueous solution. Aspen Plus release 7.3 in AspenTech company was utilized as a simulation tool and PC-SAFT modeling equation of state was used as a thermodynamic model. Fitting parameters built-in PC-SAFT model was determined by regressing experimental data, predicted results using PC-SAFT model were compared with experimental data in order to verify the exactness of the thermodynamic model. Optimization works have been performed to reduce the utility consumptions using solvent circulation rate, column operating pressure and feed stage location as manipulated variables.