Studies on Ethyl Acetate Reactive Distillation and Dehydration of Acetic Acid Having Impurities
Date Issued
2007
Date
2007
Author(s)
Lee, Hao-Yeh
DOI
zh-TW
Abstract
This dissertation consists of two parts: esterification of ethyl acetate (EtAc) via reactive distillation (RD) and the dehydration of acetic acid containing impurities.
In the first part, a homogeneous catalytic reactive distillation is focused. First, the effect of the Murphree plate efficiency (MPE) on the dynamics of such a reactive distillation is investigated. Output multiplicity was found while the organic reflux flow rate is a manipulation variable. However if reflux ratio is chosen as the new manipulating variable, this output multiplicity can be eliminated. Furthermore, the stage numbers in the RD are not inversely proportional to the MPE exactly. Second, alternative ways to improve control of an EtAc reactive distillation process are explored. In order to save energy and maintain the operating point at the original steady-state, closed-loop sensitivity plus open-loop sensitivity analysis is found effective in developing control systems for good control performance. As a result of the analysis, feasible ranges for control are also presented. Third, comparison between the optimal designs and control of homogeneous and heterogeneous catalytic reactive distillations is made. The homogeneous catalytic RD process has higher capital but less energy cost than the heterogeneous one, but, the former has less concern on the issues of catalyst installation and replacement. As for the control, heterogeneous catalytic process has faster response and lower steady-state offset than the homogeneous one.
In the second part, the process flowsheets with and without tiny impurity of (p-xylene) PX or (m-xylene) MX are considered in the optimal design of acetic acid dehydration column via heterogeneous azeotropic distillation. First, the simulation results show that dramatic differences in the base-case conditions occur upon adding 0.1 mol% of PX or MX impurity in the feed. Over 50% more TAC and operating energy are needed to operate this column with side stream due to these impurities in the feed. However, by considering feed tray location as an additional variable in the optimal design, significant savings in total annual cost can be obtained. Second, The bifurcation phenomenon was found in the case of feed with impurity, no matter the column has a side stream r not. this phenomenon was not found for the case of feed without impurity. For the former, the column should be operated at the optimal case, and, a dual-point temperature control strategy is needed. Notice that this optimal case is an open-loop unstable. In that case, PX concentration inside the column needs to be maintained in order to keep the composition profile following the path with more PX in the bottom part of the column. To ensure an energy-efficient operation, a side stream operating strategy is devised to preserve PX inside the column. The proposed overall control strategy is tested under various feed disturbances. High-purity products can still be obtained despite disturbances.
In the first part, a homogeneous catalytic reactive distillation is focused. First, the effect of the Murphree plate efficiency (MPE) on the dynamics of such a reactive distillation is investigated. Output multiplicity was found while the organic reflux flow rate is a manipulation variable. However if reflux ratio is chosen as the new manipulating variable, this output multiplicity can be eliminated. Furthermore, the stage numbers in the RD are not inversely proportional to the MPE exactly. Second, alternative ways to improve control of an EtAc reactive distillation process are explored. In order to save energy and maintain the operating point at the original steady-state, closed-loop sensitivity plus open-loop sensitivity analysis is found effective in developing control systems for good control performance. As a result of the analysis, feasible ranges for control are also presented. Third, comparison between the optimal designs and control of homogeneous and heterogeneous catalytic reactive distillations is made. The homogeneous catalytic RD process has higher capital but less energy cost than the heterogeneous one, but, the former has less concern on the issues of catalyst installation and replacement. As for the control, heterogeneous catalytic process has faster response and lower steady-state offset than the homogeneous one.
In the second part, the process flowsheets with and without tiny impurity of (p-xylene) PX or (m-xylene) MX are considered in the optimal design of acetic acid dehydration column via heterogeneous azeotropic distillation. First, the simulation results show that dramatic differences in the base-case conditions occur upon adding 0.1 mol% of PX or MX impurity in the feed. Over 50% more TAC and operating energy are needed to operate this column with side stream due to these impurities in the feed. However, by considering feed tray location as an additional variable in the optimal design, significant savings in total annual cost can be obtained. Second, The bifurcation phenomenon was found in the case of feed with impurity, no matter the column has a side stream r not. this phenomenon was not found for the case of feed without impurity. For the former, the column should be operated at the optimal case, and, a dual-point temperature control strategy is needed. Notice that this optimal case is an open-loop unstable. In that case, PX concentration inside the column needs to be maintained in order to keep the composition profile following the path with more PX in the bottom part of the column. To ensure an energy-efficient operation, a side stream operating strategy is devised to preserve PX inside the column. The proposed overall control strategy is tested under various feed disturbances. High-purity products can still be obtained despite disturbances.
Subjects
醋酸乙酯
反應蒸餾
板效率
醋酸去水
共沸蒸餾
雜質
ethyl acetate
reactive distillation
plate efficiency
acetic acid dehydration
impurity
Type
thesis
File(s)
Loading...
Name
ntu-96-D91524004-1.pdf
Size
23.53 KB
Format
Adobe PDF
Checksum
(MD5):812375ce9021258dc7c128bb9e635016