The usual practice for the separation of Isopropyl alcohol (IPA) and water mixture in industry is to add cyclohexane (CyH) as an entrainer via heterogeneous azeotropic distillation. However, it is well-known that heterogeneous azeotropic distillation can exhibit high parametric sensitivity, multiple steady-state, and long transient, and nonlinear dynamics, which can limit the operating range of this IPA dehydration system under feed disturbances. An alternative way for the separation of this mixture with minimum-boiling azeotrope is to use homogeneous azeotropic distillation by adding an extractive agent. According to Gmehling and Mollmann (1998), a suitable extractive agent for this system with high separation factor is dimethyl sulfoxide (DMSO), thus in this work an overall process system for IPA dehydration using DMSO as extractive agent is studied.

IPA is widely used in semi-conductor industry as a cleaning agent, thus the recovery of IPA from waste solvent stream is an important issue worthy of detail study. This waste liquid stream contains mainly isopropyl alcohol and water. In this study, feed stream with equal molar of IPA and water was assumed. The conceptual design of the overall homogenous azeotropic distillation column system contains two columns and a recycle stream. The first column is the homogeneous azeotropic distillation column with two feed streams including the fresh feed stream and the stream to add extractive agent into this column. The top product of this column is pure IPA with the same ultra-high purity of 99.9999 (mol % IPA) as in Chien, et al. (2004) for re-use in the semi-conductor industry. The bottom product contains water and DMSO is sent to the recovery column. In this regular distillation column, these two components are separated with top water product at purity specification of 99.9% and bottom DMSO product recycle back to the first column, thus only a very small DMSO make-up stream is needed in the overall process.

There are quite a few design variables need to be optimized. They are: the feed ratio, total number of stages, and two feed stages of the homogeneous azeotropic column; and the total number of stages and the feed stage of the recovery column. Another design decision is to either keep the recycle stream at the same bottom temperature of the recovery column or an additional cooler is necessary at the recycle stream. Total annualized cost (TAC) will be used for the determination of the above design variables. The final optimized flowsheet is shown in Figure 1 below which including cooling of recycle stream. The TAC of this optimized flowsheet will be compared to the optimized flowsheet via heterogeneous azeotropic column system.

Because the feed composition of a waste IPA stream in semi-conductor industry can vary widely, thus the overall control strategy should be able to handle large feed composition disturbances. The operability range of this homogeneous azeotropic column system under proposed overall control strategy will be compared to that of the heterogeneous azeotropic column system while still keeping the products (IPA and water) purity at their specifications.

Literature Cited

Gmehling, J.; Mollmann, C. Synthesis of Distillation Process Using Thermodynamic Models and the Dortmund Data Bank. Ind. Eng. Chem. Res. 1998, 37, 8, 3112-3123.

Chien, I. L.; Zeng, K. L.; Chao, H. Y. Design and Control of a Complete Heterogeneous Azeotropic Distillation Column System. Ind. Eng. Chem. Res. 2004, 43, 9, 2160-2174.

Keywords: Isopropyl Alcohol Dehydration, Homogeneous Azeotropic Distillation, Optimal Design, Design and Control.