Design and Control of a Reactive Distillation Process for Synthesizing Propylene Carbonate from Indirect Alcoholysis of Urea

Design and Control of a Reactive Distillation Process for Synthesizing Propylene Carbonate from Indirect Alcoholysis of Urea

Proceedings, 10th IFAC International Symposium on Proceedings, 10th of IFAC International Symposium on Advanced Control Chemical Processes Proceedings...

662KB Sizes 0 Downloads 70 Views

Proceedings, 10th IFAC International Symposium on Proceedings, 10th of IFAC International Symposium on Advanced Control Chemical Processes Proceedings, 10th of IFAC International Symposium on at www.sciencedirect.com Advanced Chemical Processes Available Shenyang,Control Liaoning, China, July 25-27, 2018 online Proceedings, 10th IFAC International Symposium on Proceedings, 10th of IFAC International Symposium on Advanced Control Chemical Processes Shenyang, Liaoning, China, July 25-27, 2018 Advanced Control of Chemical Processes Advanced of China, Chemical Processes Shenyang,Control Liaoning, July 25-27, 2018 Shenyang, Liaoning, China, July 25-27, 2018 Shenyang, Liaoning, China, July 25-27, 2018

ScienceDirect

IFAC PapersOnLine 51-18 (2018) 333–338

Design and Control of a Reactive Distillation Process for Synthesizing Propylene Design and Control of a Reactive Distillation Process for Synthesizing Propylene Design and Control of a Reactive Process for Propylene Carbonate fromDistillation Indirect Alcoholysis ofSynthesizing Urea Design of a Reactive Process for Propylene Carbonate fromDistillation Indirect Alcoholysis ofSynthesizing Urea Design and and Control Control of a Reactive Distillation Process for Synthesizing Propylene Carbonate from Indirect Alcoholysis of Urea Carbonate from Indirect Alcoholysis of Urea San-Jang Wang*, David Shan-Hill Wong**, En-Ko Lee* Carbonate from Indirect Alcoholysis of Urea

San-Jang Wang*, David Shan-Hill Wong**, En-Ko Lee*  San-Jang Wang*, David Shan-Hill Wong**, En-Ko En-Ko Lee*  San-Jang Wang*, David Shan-Hill Wong**, San-Jang Wang*, David Shan-Hill Wong**, En-Ko Lee* Lee*   *Center for Energy and Environmental Research, National Tsing Hua University, Hsinchu 30013, Taiwan (e-mail:  *Center for Energy and Environmental Research, National Hua University, Hsinchu 30013, Taiwan (e-mail: [email protected] andTsing [email protected]) *Center for Energy and Environmental Research, National Tsing Hua University, Hsinchu 30013, Taiwan (e-mail: [email protected] and [email protected]) *Center for Energy and Environmental Research, National Tsing Hua University, Hsinchu 30013, Taiwan (e-mail: **Department Chemical Engineering, National Tsing Hua University, Hsinchu 30013, Taiwan (e-mail: *Center for Energy of and Environmental Research, National Tsing Hua University, Hsinchu 30013, Taiwan (e-mail: and [email protected]) **Department of Chemical [email protected] Engineering, National Tsing Hua University, Hsinchu 30013, Taiwan (e-mail: [email protected] and [email protected]) [email protected]) [email protected] and Hua [email protected]) **Department of Chemical Engineering, National Tsing University, Hsinchu 30013, Taiwan (e-mail: [email protected]) **Department National **Department of of Chemical Chemical Engineering, Engineering, National Tsing Tsing Hua Hua University, University, Hsinchu Hsinchu 30013, 30013, Taiwan Taiwan (e-mail: (e-mail: [email protected]) [email protected]) [email protected]) Abstract: Dimethyl carbonate is a green compound with a broad variety of application. In the study, the Abstract: Dimethyl asynthesizing green compound with acarbonate broad variety application. In the study, the process design and carbonate control ofis propylene (PC)of for the dimethyl carbonate Abstract: Dimethyl carbonate is aasynthesizing green compound with aacarbonate broad variety of application. In the study, the process design and control of propylene (PC) for the dimethyl carbonate Abstract: Dimethyl carbonate is green compound with broad variety of application. In the study, the raw material is investigated indirect alcoholysis of urea. attractive production by using CO 2 as a of Abstract: Dimethyl carbonate is asynthesizing green compound with aby broad variety of for application. InThis the study, the process design and control propylene carbonate (PC) the dimethyl carbonate production by using CO as a raw material is investigated by indirect alcoholysis of urea. This attractive process design and control of synthesizing propylene carbonate (PC) for the dimethyl carbonate 2 2 of urea shows many advantages such as environmentally friendly chemicals, indirect alcoholysis process design and route control of synthesizing propylene by carbonate (PC) for the dimethyl carbonate production by using CO as a raw material is investigated indirect alcoholysis of urea. This attractive 2 indirect alcoholysis route ofmild shows advantages such as environmentally friendly chemicals, production by using CO aaurea raw material is by indirect alcoholysis of This attractive 2 as cheap raw materials, and safemany operation condition. Some reaction distillation (RD)-based production by using route CO rawand material is investigated investigated bysuch indirect alcoholysis of urea. urea. This attractive 2 as indirect alcoholysis of urea shows many advantages as environmentally friendly chemicals, cheap raw materials, and mild and safe operation condition. Some reaction distillation (RD)-based indirect alcoholysis route of urea shows many advantages such as environmentally friendly chemicals, processes for PC synthesis by this route are proposed, designed, and optimized in this work. These indirect alcoholysis route of urea shows many advantages such as environmentally friendly chemicals, cheap raw materials, and mild and safe operation condition. Some reaction distillation (RD)-based processes for PC ofsynthesis by this route are proposed, designed, and optimized in thisoperation. work. These cheap raw materials, and mild and safe operation condition. Some reaction distillation (RD)-based processes consist two operation configurations, near neat operation and excess reactant The cheap raw materials, and mild and safe operation condition. Some reaction distillation (RD)-based processes consist for PC by this route are proposed, designed, and optimized in this work. These processes ofsynthesis two of operation configurations, near to neat operation and excess economic reactant operation. The for PC synthesis by this route are proposed, designed, and optimized in this work. These intensified technologies heat integration in addition RD are used to design PC synthesis processes for PC synthesis by this route are proposed, designed, and optimized in this work. These processes consist of two operation configurations, near neat operation and excess reactant operation. The intensified technologies of heat integration in addition to RD are used to design economic PC synthesis processes of operation configurations, near neat operation and excess reactant operation. processes. Steady-state results indicate thatto the novel intensified process containing a The RD processes consist consist of two two simulation operation configurations, near neat operation and excess reactant operation. The intensified technologies of heat integration in addition RD are used to design economic PC synthesis processes. Steady-state simulation results indicate that the novel intensified process containing a RD intensified technologies of heat integration in addition to RD are used to design economic PC synthesis column and a conventional distillation column with internal vapor compression provides the most intensified technologies simulation of heat integration in addition tothe RDnovel are used to design economic PC synthesis processes. Steady-state results indicate that intensified process containing aamost RD column and a conventional distillation column with internal vapor compression provides the processes. Steady-state simulation results indicate that the novel intensified process containing RD economical design. Thissimulation process is operated under with excess reactant and fully utilizes theprovides special azeotrope processes. Steady-state resultscolumn indicate that internal the novel intensified process containing amost RD column and a conventional distillation vapor compression the economical design. This process is operated under excess reactant and fully utilizes the special azeotrope column and a conventional distillation column with internal vapor compression provides the most characteristic propylene carbonate and propylene glycol pair. This pair forms a homogeneous column anddesign. aofconventional distillation column with internal vapor compression provides the most economical This process is operated under excess reactant and fully utilizes the special azeotrope characteristic of propylene carbonate and propylene glycol pair. This pair forms a homogeneous economical design. This is operated excess reactant and fully utilizes special azeotrope minimum-boiling azeotrope near pure PGunder end under low pressure. However, thisthe vanishes economical design. This process process isthe operated under excessglycol reactant and This fully utilizes theazeotrope special azeotrope characteristic of propylene carbonate and propylene pair. pair forms aa homogeneous characteristic of propylene carbonate and propylene glycol pair. This pair forms homogeneous minimum-boiling azeotrope near the pure PG end under low pressure. However, this azeotrope vanishes under high pressure. Furthermore, steady-state analysis is used to design a simple temperature control characteristic of propylene carbonate and propylene glycol pair. This pair forms a homogeneous minimum-boiling azeotrope near the pure PG end under low pressure. However, this azeotrope vanishes under high pressure. Furthermore, steady-state analysis is used to design a simple temperature control minimum-boiling azeotrope near the pure PG end under low pressure. However, this azeotrope vanishes strategy. Different desired temperature profiles can be found in the RD column under various feed flow minimum-boiling azeotrope near the pure PG end under low pressure. However, this azeotrope vanishes under high pressure. Furthermore, steady-state analysis is used to design aa simple temperature control strategy. Different desired temperature profiles can be found in product the RD purity column under various feed under high pressure. Furthermore, steady-state analysis is used to design simple temperature control rates. Set point of the temperature loop for maintaining bottom of the RD column is flow then under high pressure. Furthermore, steady-state analysis is used to design a simple temperature control strategy. Different desired temperature profiles can be found in the RD column under various feed rates. SetDifferent point of the temperature loopDynamic for maintaining bottom product purity ofunder the RD column is flow then strategy. desired temperature profiles can be found in the RD column various feed flow reset when throughput rate changes. simulation results reveal that the proposed temperature strategy. Different desired temperature profiles can be found in product the RD purity columnofunder various feed flow rates. Set point of the temperature loop for maintaining bottom the RD column is then reset when throughput rate changes. Dynamic simulation results reveal that the proposed temperature rates. Set point of the temperature loop for maintaining bottom product purity of the RD column is control canpoint maintain purities atDynamic their desired values in face of feedpurity flow the disturbances. rates. Set of theproduct temperature loop for maintaining bottom product of the RD column is then then reset when throughput rate changes. simulation results reveal that proposed temperature control can maintain product purities atDynamic their desired values in face of feed that flow the disturbances. reset when throughput rate changes. simulation results reveal proposed temperature reset when throughput rate changes. Dynamic simulation results reveal that the proposed temperature control can maintain product purities at their desired values in face of feed flow disturbances. © 2018,can IFAC (International Federation oftheir Automatic Control) Hosting by Elsevier Ltd. All rights reserved. Keywords: Design, control, propylene carbonate, plant-wide, urea. control maintain product purities values face control can Design, maintaincontrol, productpropylene purities at atcarbonate, their desired desired values in inurea. face of of feed feed flow flow disturbances. disturbances. Keywords: plant-wide, Keywords: Design, control, propylene carbonate, plant-wide, urea.  Keywords: urea. Keywords: Design, Design, control, control, propylene propylene carbonate, carbonate, plant-wide, plant-wide, urea.  DMC are sequentially synthesized from the plant-wide  1. INTRODUCTION DMC are synthesized fromin the  process. Thesequentially most important characteristic this plant-wide plant-wide 1. INTRODUCTION  DMC are sequentially synthesized from the process. The most important characteristic in this plant-wide plant-wide DMC are sequentially synthesized from the 1. INTRODUCTION process is that PG, a byproduct in the transesterification are sequentially synthesized fromin the plant-wide Carbon dioxide (CO1. emission has recently become a global DMC 2) INTRODUCTION process. The most important characteristic this plant-wide 1. INTRODUCTION process. The most important characteristic in this plant-wide process is that PG, a byproduct in the transesterification Carbon dioxide (CO emission hasand recently become arise global 22) significant reaction for that DMC synthesis, can beinused as the reactant for process. The most important characteristic in this plant-wide issue due to the continuous of process is PG, a byproduct the transesterification Carbon dioxide (CO ) emission has recently become a global 2 process is that PG, a byproduct in the transesterification reaction for DMC synthesis, can be used as the reactant for issue due to the significant and continuous rise of Carbon dioxide (CO ) emission has recently become a global PC synthesis and the released ammonia from PC synthesis 2 process is that PG, a byproduct in the transesterification concentration. COand utilization and chemical atmospheric CO 2the 2recently reaction for DMC synthesis, can be used as the reactant for Carbon dioxide (CO emission has become arise global 2) significant issue due to continuous of reaction for DMC synthesis, can be used as the reactant for PC synthesis andback the released ammonia PC synthesis concentration. CO and chemical atmospheric CO issue due to the significant and continuous rise of 22high 2 can be recycled to produce urea by from reacting with CO 2 utilization 2. reaction for DMC synthesis, can be used as the reactant for conversion into value-added products play a key role in PC synthesis and the released ammonia from PC synthesis issue due to the significant and continuous rise of concentration. CO utilization and chemical atmospheric CO 2 2 PC synthesis and the released ammonia from PC synthesis can be recycled back to produce urea by reacting with CO 2 conversion into high value-added products play a key role in 2. concentration. CO utilization and chemical atmospheric CO The overall process forms a green chemical cycle, which 2 2 PC synthesis and the released ammonia from PC synthesis emission. Dimethyl carbonate (DMC) is an minimizing CO can be recycled back to produce urea by reacting with CO concentration. CO and chemical atmospheric CO22high 2. 2 utilization conversion into value-added products play a key role in The overall process forms a green chemical cycle, which can be recycled back to produce urea by reacting with CO emission. Dimethyl carbonate (DMC) is an minimizing CO conversion into high value-added products play a key role in 22 compound and chemical intermediate with increases theprocess utilization ofa raw materials for the DMC can be recycled back forms to produce urea by reacting withwhich CO22.. important organic The overall green chemical cycle, conversion into high value-added products play a key role in emission. Dimethyl carbonate (DMC) is an minimizing CO increases the utilization of raw materials for the DMC 2 emission. The overall process forms aa green chemical cycle, which important organic and There chemical with Dimethyl carbonate (DMC) is an minimizing CO production. Some reports (Hamidipour et al., 2005; Zhang et 2 compound overall process forms greenmaterials chemical cycle, which the label of ‘‘green chemical’’. areintermediate several reaction increases the utilization of raw for the DMC Dimethyl carbonate (DMC) iswith an The minimizing CO 2 emission. important organic compound and chemical intermediate production. Some reports (Hamidipour et al., 2005; Zhang et increases the utilization of raw materials for the DMC the label of ‘‘green chemical’’. There are several reaction important organic compound and chemical intermediate with al., 2005; Holtbruegge et al., 2013; Holtbruegge et al., 2014) increases the utilization of raw materials for the DMC paths to manufacture DMC. Among these paths, DMC production. Some reports (Hamidipour et al., 2005; Zhang et important organic compound and chemical intermediate with the label of ‘‘green chemical’’. There are several reaction al., 2005; Holtbruegge et al., 2013; Holtbruegge et al., 2014) production. Some reports (Hamidipour et al., 2005; Zhang et paths to of manufacture DMC. Among these paths,reaction DMC the label ‘‘green chemical’’. There are several can be found in discussing about the process design of2014) urea production. Some reports (Hamidipour et al., 2005; Zhang et as a raw material attracts much production utilizing CO 2 al., 2005; Holtbruegge et al., 2013; Holtbruegge et al., the label of ‘‘green chemical’’. There are several reaction paths to manufacture DMC. Among these paths, DMC can be found in discussing about the process design of urea al., 2005; Holtbruegge et al., 2013; Holtbruegge et al., 2014) paths to manufacture these paths, DMC a Among raw material attracts much production utilizing CODMC. 2 synthesis and DMC synthesis. However, nowadays, very few 2 asdirect al., 2005; Holtbruegge et al., 2013; Holtbruegge et al., 2014) attention because it supplies environment benefits. The can be found in discussing about the process design of urea paths to manufacture DMC. Among these paths, DMC aa raw material attracts much production utilizing CO2 as synthesis and DMC synthesis. However, nowadays, very few be in about the process design of attention because it supplies environment benefits. The can raw material attracts much production utilizing literatures investigate the process design and control forurea PC 2 asdirect can be found found in discussing discussing about the process designvery of urea methods toCO produce include direct synthesis CO 2-based synthesis and DMC synthesis. However, nowadays, few aDMC raw material attracts much production utilizing CO 2 asdirect attention because it supplies environment benefits. The literatures investigate the process design and control for PC synthesis and DMC synthesis. However, nowadays, very few -based methods to produce DMC include direct synthesis CO attention because it supplies direct environment benefits. The 2 synthesis. In this work, some processes, classified in terms of 2 methanol, synthesis and DMC synthesis. However, nowadays, very few with transesterification of propylene carbonate (PC) investigate the process design and control for PC attention because it supplies direct environment benefits. The literatures -based methods to produce DMC include direct synthesis CO synthesis. In this work, some processes, classified in terms of 2 literatures investigate the process design and control for PC with methanol, transesterification of propylene carbonate (PC) -based methods to produce DMC include direct synthesis CO two operation configurations, near neat operation and excess 2ethylene literatures investigate the process design and control for PC or carbonate with methanol, direct and indirect synthesis. In this work, some processes, classified in terms of methods to produce DMC include direct synthesis CO 2-based with methanol, transesterification of propylene carbonate (PC) two operation configurations, near neat operation and excess synthesis. In this work, some processes, classified in terms of or ethylene carbonate with methanol, direct and indirect with methanol, transesterification of propylene carbonate (PC) reactant operation, for PC synthesis by reacting and PG In this work, some processes, classifiedurea in terms of alcoholysis ofcarbonate urea. In this work, wepropylene willdirect concentrate on(PC) the synthesis. two operation configurations, near neat operation and excess withethylene methanol, transesterification of carbonate or with methanol, and indirect reactant operation, for PC synthesis by reacting urea and PG two operation configurations, near neat operation and excess alcoholysis of urea. In this work, we will concentrate on the or ethylene carbonate with methanol, direct and indirect are proposed, designed, and optimized by minimizing total two operation configurations, near neat operation and excess method of the indirect alcoholysis of urea. This method starts reactant operation, for PC synthesis by reacting urea and PG or ethyleneofcarbonate withwork, methanol, direct and indirect alcoholysis urea. In this we will concentrate on the are proposed, designed, and optimized by(RD) minimizing total reactant operation, for PC synthesis by urea and PG method of the indirect alcoholysis of urea. This method starts alcoholysis of urea. In this work, we will on the cost (TAC). distillation and internal reactant operation, forReactive PCand synthesis by reacting reacting urea andtotal PG from urea and propylene or concentrate ethylene glycol to annual are proposed, designed, optimized by minimizing alcoholysis of indirect urea. In alcoholysis thisglycol work, (PG) we will concentrate on the method of the of urea. This method starts are proposed, designed, and optimized by minimizing total annual cost (TAC). Reactive distillation (RD) and internal from urea and propylene glycol (PG) or ethylene glycol to method of the indirect alcoholysis of urea. This method starts vapor compression technologies are used to intensify the are proposed, designed, and optimized by minimizing total synthesize ammonia and PC or ethylene carbonate. This annual cost (TAC). Reactive distillation (RD) and internal method of the indirect alcoholysis of urea. This method starts from urea and propylene glycol (PG) or ethylene glycol to vapor compression technologies areIn used to intensify the annual cost (TAC). Reactive distillation (RD) and internal from urea and propylene glycol (PG) or ethylene glycol to synthesize ammonia and PC or ethylene carbonate. This plant-wide PC synthesis processes. addition, the strategy annual compression cost (TAC). technologies Reactive distillation (RD) and internal reaction provides many benefits, as cheap and are used to intensify the from urearoute and propylene glycol (PG) orsuch ethylene glycol to vapor synthesize ammonia and PC or ethylene carbonate. This plant-wide PC synthesis processes. In addition, the strategy vapor compression technologies are used to intensify the reaction route provides many benefits, suchcarbonate. as cheap This and vapor synthesize ammonia and PC or ethylene of temperature control instead of composition control is compression technologies are used to intensify the easily available raw materials, mild reaction conditions, safe plant-wide PC synthesis processes. In addition, the strategy synthesize ammonia and PC or ethylene carbonate. This reaction route provides many benefits, such as cheap and of temperature control instead of composition control is plant-wide PC synthesis processes. In addition, the strategy easily available raw materials, mild reaction conditions, safe reaction route provides many benefits, such as cheap and proposed to maintain product purities by steady-state analysis plant-wide PC synthesis processes. In addition, the strategy operations, environmentally friendly chemicals, etc. Fig. 1 temperature control instead of composition control is reaction route provides many mild benefits, suchconditions, as cheap safe and of easily available raw materials, reaction proposed to maintain product purities by steady-state analysis of temperature control instead of composition control is operations, environmentally friendly chemicals, etc. Fig. 1 easily raw materials, mild safe for the most economical plant-wide temperature control instead ofprocess. composition control is shows the concept the plant-wide processconditions, foretc. the Fig. DMC proposed to maintain product purities by steady-state analysis easily available available raw of materials, mild reaction reaction conditions, safe1 of operations, environmentally friendly chemicals, for the most economical plant-wide process. proposed to maintain product purities by steady-state analysis shows the concept of the plant-wide process for the DMC operations, environmentally friendly chemicals, etc. Fig. 1 proposed to maintain product purities by steady-state analysis production by indirect alcoholysis of urea. Urea, PC, and the most economical plant-wide process. operations, environmentally friendly chemicals, etc. Fig. 1 for shows the concept of the plant-wide for the production by indirect alcoholysis of process urea. Urea, PC,DMC and for shows the of plant-wide for DMC for the the most most economical economical plant-wide plant-wide process. process. shows the concept concept of the the plant-wideof process process for the the DMC production by indirect alcoholysis urea. Urea, PC, and production production by by indirect indirect alcoholysis alcoholysis of of urea. urea. Urea, Urea, PC, PC, and and

Copyright 2018 IFAC 327 Hosting by Elsevier Ltd. All rights reserved. 2405-8963 © 2018, IFAC (International Federation of Automatic Control) Copyright 2018 responsibility IFAC 327Control. Peer review©under of International Federation of Automatic Copyright © 2018 IFAC 327 10.1016/j.ifacol.2018.09.322 Copyright © 2018 IFAC 327 Copyright © 2018 IFAC 327

2018 IFAC ADCHEM 334 San-Jang Wang et al. / IFAC PapersOnLine 51-18 (2018) 333–338 Shenyang, Liaoning, China, July 25-27, 2018

software. Fig. 2 shows the VLE relationship for PG-PC pair. It indicates that homogeneous minimum-boiling azeotrope can be found for this pair under low pressure (0.12 atm) (Mathuni et al., 2011). However, there is no azeotrope under relatively high pressure (1 atm). This important characteristic is fully utilized to reduce energy consumption in the following process design for PC synthesis under excess reactant operation.

NH3 CO 2

urea synthesis

urea

PC synthesis

PG

PC

DMC synthesis

MeOH

1.0

Fig. 1. Concept of the plant-wide process for DMC production by indirect alcoholysis of urea. 2. KINETIC AND THERMODYNAMIC MODELS

0.8

0.6 1.00

0.4

0.98

PG vapor mole fraction

DMC

PG vapor mole fraction

H2O

0.2

0.96

0.94

0.92 Col 1 vs Col 2 Col 5 vs Col 6

0.90 0.90

0.92

0.94

0.96

0.98

1.00

PG liquid mole fraction

Three reaction steps given below are proceeded for DMC production using CO2 as the raw material by indirect alcoholysis of urea.

0.0 0.0

0.2

0.4

0.6

0.8

1.0

PG liquid mole fraction

(a)

CO2 + 2 NH3 = Urea + H2O

(1)

1.0

Urea + PG → PC + 2 NH3

(2)

0.8

0.6 1.00

0.4

0.98

PG vapor mole fraction

PG vapor mole fraction

PC + 2 MeOH ⇌ DMC + PG (3) In the first step, urea is produced by reacting CO2 and ammonia. Next, urea is reacted with PG to coproduce PC and ammonia. DMC is synthesized in the final step by the transformation reaction of PC and methanol (MeOH). The byproducts, ammonia and PG produced from steps 2 and 3, respectively, are recycled back to steps 1 and 2, respectively, as reactants. The overall reaction is the one of CO2 and MeOH to produce DMC and water. Ammonia and PG are the intermediates. In this work, we first explore the design of different PC synthesis processes. Some possible RD-based configurations are proposed to synthesize PC. Finally, the control of the most economical process is investigated.

0.2

0.96

0.94

0.92

0.90 0.90

0.92

0.94

0.96

0.98

1.00

PG liquid mole fraction

0.0 0.0

0.2

0.4

0.6

0.8

1.0

PG liquid mole fraction

(b) Fig. 2. VLE relationship for PG-PC pair under pressures (a) 0.12 atm and (b) 1 atm. 3. DESIGN OF PC SYNTHESIS PROCESSES

Wang (2012) presented a kinetic model given below for PC synthesis by using MgO as the heterogeneous catalyst.

In the design of the process to synthesize PC, the objective is to minimize the TAC by adjusting design variables. TAC is calculated by the sum of operating cost and capital cost/payback year. The operating cost (OC) includes the costs of steam, cooling water, catalyst (replaced annually), and electricity. The capital cost (CC) comprises the costs of the column shell, stage, heat exchanger, and compressor. The formulas of these cost estimations are taken from Turton et al. (2012). Here, a payback of eight years is used. An annual operating time is assumed to be 8322 hours. The unit costs of medium pressure steam (at 10 barg and 184 °C), high pressure steam (at 41 barg and 254 °C), and cooling water (at 30 °C) are 14.19 US$/GJ, 17.70 US$/GJ, and 0.354 US$/GJ, respectively. The unit cost of electricity is 16.8 US$/GJ.

562.602  rPC 1.5888  exp(  )  Curea  C PG (4) T where rPC represent the reaction rate of PC synthesis (mol  liter-1  min-1). T and Ci denote temperature (K) and concentration (mol  liter-1) of component i, respectively. The catalyst concentration is assumed to be 2 wt%, used in the experiment study.

In the study, the process simulation is conducted by a rigorous model provided by ChemCad software. The vaporliquid equilibrium (VLE) relationship is described by UNIQUAC activity coefficient model. In the PC synthesis reaction system, only the thermodynamic data of ureaammonia and PG-PC pairs are found in the literatures (Mathuni et al., 2011; Holtbruegge et al., 2013; Voskov and Voronin, 2016). The solubility of ammonia in the reaction mixture is represented by using Henry coefficients. The phase-equilibrium relationships for the other pairs are determined in the study by the ideal model from ChemCad

3.1 Near Neat Operation RD is an intensification technology that integrates reaction and separation operations in a single vessel. The RD configuration for PC synthesis is shown in Fig. 3. The SCDS 328

2018 IFAC ADCHEM Shenyang, Liaoning, China, July 25-27, 2018 San-Jang Wang et al. / IFAC PapersOnLine 51-18 (2018) 333–338

module in ChemCad is used to simulate a RD column. Two reactants (urea and PG) are fed into the first RD column for PC synthesis. Ammonia is removed from the partial condenser. PC is achieved from the column bottom and then fed into the second RD column to react with MeOH for DMC synthesis. In the second RD column, the overhead product mainly contains a mixture of DMC and MeOH. PG product is withdrawn from the column bottom and is recycled back to the first RD column for PC synthesis. PG is converted into the reactant for the PC synthesis by indirect alcoholysis of urea. Therefore, in addition to the first RD column for PC synthesis, the second RD column to synthesize DMC by reacting PC and MeOH is also optimally designed in the study.

335

reactant was fed above MeOH reactant into the RD column and this catalyst was fed along with MeOH stream. In the DMC synthesis column, there are large amounts of MeOH in the reflux stream. Therefore, in our study, this catalyst is selected to be fed into the reflux stream for mixing with MeOH and then returned to the RD column for catalyzing the reaction of PC and MeOH. DMC/MeOH mixture is recovered from the column top while high-purity PG with 99.9 mol% is withdrawn from the column bottom and recycled back to the PC synthesis column. DMC and MeOH can form an azeotrope and the azeotrope can be separated by extractive distillation or pressure swing distillation from our previous study (Wang et al., 2010) and some literatures (Wei et al., 2013; Zhang et al., 2017). Therefore, in the present study, the emphasis is not put on the separation of DMC and MeOH. Fig. 3 also shows the optimal design with the minimal TAC by an optimization procedures. The optimal feed ratio of PG and urea is 1.02. High-purity (99.5 mol%) PC withdrawn from the bottom of the PC synthesis column is fed into the RD column for DMC synthesis. DMC/MeOH mixture is recovered from the top of DMC synthesis column while highpurity PG with 99.9 mol% is withdrawn from the column bottom and recycled back to the RD column for PC synthesis. The minimized TAC for the PC synthesis by the RD configuration, given in Table 1, is 973.4×103 US$/year. OC is almost ten times CC. Here CD denotes conventional distillation.

Fig. 3. Process flowsheet of RD configuration for PC synthesis in conjunction with RD column for DMC synthesis. In the PC synthesis, urea is fed with the flow rate of 40 kmol/h and its reaction conversion is designed to be greater than 99.98 mol% in the following different process configurations. The purity of vapor ammonia removed from the partial condenser is designed at 99 mol% while the PC purity of column bottom is set at 99.5 mol%. There are some design and operation variables in the PC synthesis column. The design variables include the feed ratio of PG and urea reactants, the pressure of the RD column, the number of total stages, the stage numbers of the rectifying and reaction sections, and the locations of urea and PG feed stage. The operation variables include the condenser duty and the reboiler duty. Reboiler duty is varied to satisfy the PC product specification (99.5 mol%) at column bottom while condenser duty is changed to maintain the product purity of vapor ammonia at 99 mol% from column top.

Table 1. TAC comparison of different configurations for PC synthesis Configuration Total CC RD 84.6 RD+CD_HI1 220.2 RD+CD_HI2 196.6 RD+CD_HI3 187.9 3 Cost unit = 10 US$/year

Total OC 888.8 835.3 674.2 673.7

TAC 973.4 1055.5 870.8 861.6

3.2 Excess Reactant Operation

In the DMC synthesis RD column, the product purity of bottom PG is set at 99.9 mol% while the reaction conversion of PC is set at 99.98 mol%. Design variables include feed ratio of MeOH and PC, number of total stages in the RD column, the feed stage of PC, and the feed stage of MeOH. The operation variables consist of reboiler duty and reflux ratio. Reboiler duty and reflux ratio are adjusted to satisfy the requirements of PG composition at column bottom and PC reaction conversion, respectively. The kinetic model and the thermodynamic model from Holtbruegge et al. (2013) are adopted in the RD column design for DMC synthesis. DMC and MeOH form a minimum-boiling homogeneous azeotrope. In the DMC synthesis reaction, sodium methoxide is used as a homogeneous catalyst to catalyze the reaction of PC and MeOH. Holtbruegge et al. (2013) specified that the homogeneous catalyst has a low solubility in the high-boiling reactant PC. In the study of Holtbruegge et al. (2014), PC 329

In the above section, the optimal feed ratio of PG and urea is very close to one. In this section, we explore the process design with excess PG reactant in the RD column for PC synthesis. In this type of configuration, the bottom product, mainly containing PC and excess PG, of the RD column is introduced to a CD column for the separation of PC and PG. Heat integration between the RD and CD columns can then be implemented to reduce the energy consumption of the PC synthesis process by properly adjusting the pressures of RD and CD columns. In this study, three configurations of heat integrated RD+CD are proposed and their optimal designs are given below. Similar optimization procedures given in 3.1 section are also used in these heat integrated RD+CD configurations. Fig. 4 shows the first configuration of heat integrated RD+CD (named as RD+CD_HI1). The overhead temperature of CD column is designed to be greater than the bottom temperature of the RD column. A minimum temperature difference of 10 K is ensured between the overhead vapor of the CD column and the bottom liquid of

2018 IFAC ADCHEM 336 San-Jang Wang et al. / IFAC PapersOnLine 51-18 (2018) 333–338 Shenyang, Liaoning, China, July 25-27, 2018

the RD column. Then the latent heat of the overhead vapor from the CD column can be released to the reboiler of the RD column. In the RD column, design variables contain feed ratio of PG and urea, column pressure, number of total stages, stage numbers of the rectifying and reaction sections, and feed stages of urea and PG. Operation variables consist of condenser duty and reboiler duty in the column, used to satisfy the requirement of overhead ammonia product purity (99 mol%) and urea reaction conversion, respectively. In the CD column, design variables include column pressure, number of total stages and feed stage. Two operation variables, reflux ratio and reboiler duty, are varied to satisfy the product specifications of PG (98 mol%) and PC (99.5 mol%) at the column top and bottom, respectively. The optimal design of RD+CD_HI1 configuration is also provided in Fig. 4. The overhead temperature of the highpressure CD column and base temperature of the lowpressure RD column are 184.5 and 161.4 °C, respectively. The large temperature difference between these two temperatures insures that heat integration is a practical method to save energy consumption. The condenser duty of the CD column and the reboiler duty of the RD column are 1741.7 MJ/h and 5675.5 MJ/h, respectively. The condenser duty in the high-pressure CD column can be utilized to produce vapor in the RD column base, and an auxiliary steam-driven reboiler is needed with a duty of 3933.8 MJ/h. Similar DMC synthesis column to that in Fig. 3 is also given in Fig. 4. The minimized TAC of the RD+CD_HI1 configuration for PC synthesis, also given in Table 1, is 1055.5×103 US$/year. In comparison with the RD configuration, the TAC is increased from 973.4×103 US$/year for the RD configuration with almost neat operation to 1055.5×103 US$/year for the RD+CD_HI1 configuration with excess reactant operation. The increased capital cost due to the addition of CD column is greater than the reduced operating cost due to the heat integration in the RD+CD_HI1 configuration.

economical benefit can be achieved by operating bottom section of the CD column at low pressure and top section of the CD column at high pressure. This configuration is shown in Fig. 5 and is referred to as RD+CD_HI2. The CD column is divided into top and bottom sections. The overhead vapor of the bottom section is compressed by a compressor and used as a heating medium of the base of the top section. Therefore, the top section of the CD column can be operated under high pressure and temperature. The pressure of the top section can be manipulated such that overhead temperature of the CD column is greater than the base temperature of the CD column or the base temperature of the RD column. Therefore, heat integration can be implemented to save energy consumption by two different ways. The first way is to release the latent heat of overhead vapor to the reboiler in the CD column itself. If the condenser duty is greater than the reboiler duty, the additional latent heat can be released to the reboiler of the RD column. For the RD column in Fig. 5, the same design and operation variables as those in the RD+CD_HI1 configuration are used in the RD+CD_HI2 configuration. For the CD column, the pressure and stage number of top and bottom sections, respectively, are new design variables in the RD+CD_HI2 configuration.

Fig. 5. Process flowsheet of RD+CD_HI2 configuration for PC synthesis in conjunction with RD column for DMC synthesis. The optimal design of the RD+CD_HI2 configuration is also given in Fig. 5. The overhead temperature of the highpressure top section and base temperature of the low-pressure bottom section in the CD column are 195.0 °C and 151.2 °C, respectively. The condenser duty of the top section and the reboiler duty of the bottom section are 1100.2 MJ/h and 676.5 MJ/h, respectively. Thus, the latent heat of the top section can not only be released 676.5 MJ/h to the reboiler of the bottom section in the CD column but also released 423.7 MJ/h to the reboiler of the RD column. An auxiliary steamdriven reboiler is required with a duty of 5266.1 MJ/h in the RD column. The TAC of the optimal RD+CD_HI2 configuration for PC synthesis, given in Table 1, is 870.8×103 US$/year. In comparison with the RD configuration, a 10.5 % reduction of TAC can be obtained for the RD+CD_HI2 configuration. In comparison with the RD+CD_HI1 configuration, TAC can be reduced by 17.5 % for the RD+CD_HI2 configuration because it takes full advantage of the characteristic of PC-PG VLE relationship by adding an internal compressor between top and bottom sections of the CD column.

Fig. 4. Process flowsheet of RD+CD_HI1 configuration for PC synthesis in conjunction with RD column for DMC synthesis. In the CD column shown in Fig. 4, column top and bottom are all operated under high temperature. One of the methods to reduce energy consumption of the CD column is to operate it under low pressure. However, from the VLE relationship for PG-PC pair shown in Fig. 2, PC-PG pair can form a homogeneous minimum-boiling azeotrope near the pure PG end under low pressure. This indicates that using the CD column to separate these two components would require many stages or high energy consumption. Nevertheless, the azeotrope vanishes under high pressure. That is, much 330

2018 IFAC ADCHEM Shenyang, Liaoning, China, July 25-27, 2018 San-Jang Wang et al. / IFAC PapersOnLine 51-18 (2018) 333–338

In addition to the heat integration shown in Fig. 5, the overhead vapor of the top section in the CD column can release total latent heat to the reboiler of the RD column. This configuration is named as RD+CD_HI3. The optimization procedures for this configuration are almost the same as those for the RD+CD_HI2 configuration. Heat integration is implemented between CD top and RD bottom in the RD+CD_HI3 configuration. Fig. 6 shows the optimal design of this configuration. The TAC of this configuration, given in Table 1, is 861.6×103 US$/year. In comparison with the RD configuration, TAC can be reduced by 11.5 % for the RD+CD_HI3 configuration. In comparison with the RD+CD_HI1 configuration, a 18.4 % reduction of TAC can be obtained for the RD+CD_HI3 configuration. This RD+CD_HI3 configuration provides the most economic design for the PC synthesis in the plant-wide process to produce the final product DMC. In the following section, the control strategy of this most economic optimal process is investigated under feed flow disturbances.

337

then used in the study. In the Fig. 7, different reflux ratios are necessary to be designed in order to maintain overhead ammonia and bottom PC product purities. However, the top temperature is almost unchanged under different feed flow rates. Therefore, the flow rate of cooling water is manipulated to control the temperature of vapor withdrawn from the partial condenser. Fig. 8 shows the open-loop sensitivity analysis for the RD column of PC synthesis under +0.1% changes of reboiler duty. Temperature at stage 10 gives the most sensitivity except top temperature and is then chosen as the controlled variable. In the RD column, the temperatures located at stage 1 and stage 10 are then controlled by manipulating cooling water and reboiler duty. However, from Fig. 7, the set point of bottom temperature loop needs to be changed under feed flow changes if bottom PC product purity is wanted to be maintained at its desired value. In the inventory control loops, RD column pressure is controlled by changing the overhead vapor flow. The reflux drum level and bottom level are maintained by varying reflux flow and bottom flow rates, respectively. In the control of the CD column, reflux ratio is controlled by changing distillate flow rate to maintain the overhead PG product purity. Fig. 9 gives the open-loop sensitivity analysis for the CD column under +1% changes of reboiler duty. To maintain the bottom PC product purity, the temperature located at stage 7 in the bottom section is chosen as the controlled variable because of its most sensitivity to rebolier duty. In addition, the reflux drum level and bottom level are maintained by changing reflux flow and bottom flow rates, respectively.

Fig. 6. Process flowsheet of RD+CD_HI3 configuration for PC synthesis in conjunction with RD column for DMC synthesis.

170

Temperature (oC)

150

4. CONTROL OF A PC SYNTHESIS PROCESS For a RD column in the kinetic regime, the desired steadystate temperature or composition profile changes when feed rate changes and product purities are kept at their designed values (Wang et al., 2003). Fig. 7 gives the desired temperature profiles of the RD column for PC synthesis in Fig. 6 under +10% feed flow changes. The temperature profile between stages 2 and 11 at nominal state is located below the temperature profiles between stages 2 and 11 under +10% feed flow changes. In industrial applications, temperature control is usually used instead of composition control. The reason is that most product analyzers, such as gas chromatographs, suffer from large measurement delays and high investment and maintenance cost. To improve the control performance and have a faster mean of measuring the changes occurring in column during transient conditions, the feasibility of temperature control is investigated in this work by adding disturbances of changes in reactant feed rates. Hence there are three keys to controlling such a process: (1) to maintain the correct stoichiometric balance between the reactant feeds, (2) to maintain the product quality, (3) to account for possible changes in control objective when throughput rate changes. The product purity and stoichiometric balance between the reactant feeds must be maintained when operating a RD column. Feed ratio control is the simplest way to maintain stoichiometric balance and is

130 110 90

+10% 0% -10%

70 50 30 1

6

11

16

21

26

Number of stage

Fig. 7. The desired temperature profiles of the RD column under different feed rates.

Temperature change (oC)

1.0

+0.1% -0.1%

0.5

0.0

-0.5

-1.0 1

6

11

16

21

26

Number of stage

Fig. 8. Open-loop sensitivity analysis for the RD column under +0.1% changes of reboiler duty. 331

2018 IFAC ADCHEM 338 San-Jang Wang et al. / IFAC PapersOnLine 51-18 (2018) 333–338 Shenyang, Liaoning, China, July 25-27, 2018

feed flow changes, respectively. Controlled stage temperature is quickly increased and can be settled at its new set point. Bottom PC product purity can be operated at its desired value under temperature control. In the CD column, only small variations of overhead temperature and PG composition are observed under constant reflux ratio control. In the bottom temperature loop, the set point of the controlled stage temperature is unnecessary to be reset. The controlled stage temperature returns to its set point and bottom PC purity can be maintained at its designed value.

Temperature change (oC)

1.0 +1% for top section -1% for top section +1% for bottom section -1% for bottom section

0.5

0.0

-0.5

-1.0 1

4

7

10

13

16

19

REFERENCES

Number of stage

Fig. 9. Open-loop sensitivity analysis for the CD column under +1% changes of reboiler duty. Ammonia composition (mole fraction)

36

34

T1 (oC)

+10% -10% 32

30 0

10

20

30

0.995

+10% -10% 0.990

0.985

40

0

10

Time (h)

20

30

40

30

40

30

40

30

40

Time (h) 0.94

PC composition (mole fraction)

90

T10 (oC)

80

70

+10% -10%

60

+10% -10%

0.91

0.88 0

10

20

30

40

0

10

Time (h)

20

Time (h)

(a) 0.985

PG composition (mole fraction)

188.0

+10% -10%

T1 (oC)

187.9

187.8

187.7

0.975

+10% -10%

0.965

0.955 0

10

20

30

40

0

10

Time (h)

20

Time (h) 1.000

143

PC composition (mole fraction)

145

T7 (oC)

+10% -10%

141

139

+10% -10%

0.995

0.990 0

10

20

30

40

0

Time (h)

10

20

Time (h)

(b)

Fig. 10. Temperature and composition profiles of (a) RD and (b) CD columns under +10% feed rate disturbances. Fig. 10 shows the temperature and composition profiles of RD and CD columns under +10% feed rate disturbances. In the RD column, the overhead temperature is controlled at its set point and overhead vapor ammonia composition can be maintained at its desired value. In the bottom temperature loop, the set point of controlled stage temperature is changed from 63.6 oC to 83.3 oC and 75.5 oC under +10% and -10% 332

Hamidipour, M., Mostoufi, N., and Sotudeh-Gharebagh, R. (2005). Modeling the synthesis section of an industrial urea plant. Chem. Eng. J., 106, 249. Holtbruegge, J., Leimbrink, M., Lutze, P., and Górak, A. (2013). Synthesis of dimethyl carbonate and propylene glycol by transesterification of propylene carbonate with methanol: Catalyst screening, chemical equilibrium and reaction kinetics. Chem. Eng. Sci., 104, 347. Holtbruegge, J., Wierschem, M., and Lutze, P. (2014). Synthesis of dimethyl carbonate and propylene glycol in membrane-assisted reactive distillation process: Pilotscale experiments, modeling and process analysis. Chem. Eng. Process., 84, 54. Mathuni, T., Kim, J.I., and Park, S.J. (2011). Phase equilibrium and physical properties for the purification of propylene carbonate (PC) and γ-butyrolactone (GBL). J. Chem. Eng. Data, 56, 89. Turton, R., Bailie, R.C., Whiting, W.,B., and Shaeiwitz, J.A. (2012). Analysis, synthesis, and design of chemical processes, Pearson Education, Boston. Voskov, A.L. and Voronin, G.F. (2016). Thermodynamic model of the urea synthesis process. J. Chem. Eng. Data, 61, 4110. Wang, S.J., Wong, D.S.H., and Lee, E.K. (2003). Control of a reactive distillation column in the kinetic regime for the synthesis of n-butyl acetate. Ind. Eng. Chem. Res., 42, 5182. Wang, S.J., Yu, C.C., and Huang, H.P. (2010). Plant-wide design and control of DMC synthesis process via reactive distillation and thermally coupled extractive distillation. Comput. Chem. Eng., 34, 361. Wang, Y. (2012). Green process development of dimethyl carbonate, Qingdao University of Science and Technology, Master Thesis. Wei, H.M., Wang, F., Zhang, J.L., Liao, B., Zhao, N., Xiao, F.K., Wei, W., and Sun, Y.H. (2013). Design and control of dimethyl carbonate-methanol separation via pressureswing distillation. Ind. Eng. Chem. Res. 52, 11463. Zhang, Q., Peng, J., and Zhang, K. (2017). Separation of an azeotropic mixture of dimethyl carbonate and methanol via partial heat integration pressure swing distillation. Asia-Pac. J. Chem. Eng., 12, 50. Zhang, X., Zhang, S., Yao, P., and Yuan, Y. (2005) Modeling and simulation of high-pressure urea synthesis loop. Comput. Chem. Eng., 29, 983.