Control of extractive distillation process for separating heterogenerous ternary azeotropic mixture via adjusting the solvent content

Control of extractive distillation process for separating heterogenerous ternary azeotropic mixture via adjusting the solvent content

Accepted Manuscript Control of extractive distillation process for separating heterogenerous ternary azeotropic mixture via adjusting the solvent cont...

3MB Sizes 2 Downloads 123 Views

Accepted Manuscript Control of extractive distillation process for separating heterogenerous ternary azeotropic mixture via adjusting the solvent content Yong Wang, Xia Zhang, Xiaobin Liu, Wenting Bai, Zhaoyou Zhu, Yinglong Wang, Jun Gao PII: DOI: Reference:

S1383-5866(17)32331-6 http://dx.doi.org/10.1016/j.seppur.2017.09.008 SEPPUR 14017

To appear in:

Separation and Purification Technology

Received Date: Revised Date: Accepted Date:

19 July 2017 2 September 2017 3 September 2017

Please cite this article as: Y. Wang, X. Zhang, X. Liu, W. Bai, Z. Zhu, Y. Wang, J. Gao, Control of extractive distillation process for separating heterogenerous ternary azeotropic mixture via adjusting the solvent content, Separation and Purification Technology (2017), doi: http://dx.doi.org/10.1016/j.seppur.2017.09.008

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Control of extractive distillation process for separating heterogenerous ternary azeotropic mixture via adjusting the solvent content Yong Wanga, Xia Zhanga, Xiaobin Liua, Wenting Baia, Zhaoyou Zhua, Yinglong Wanga*, Jun Gaob a

College of Chemical Engineering, Qingdao University of Science and Technology, Qingdao

266042, China b

College of Chemical and Environmental Engineering, Shandong University of Science and

Technology, Qingdao, 266590, China Corresponding Author *E-mail: [email protected] Abstract The existence of heterogeneous azeotrope of toluene-water and homogenous azeotrope of toluene-methanol makes it difficult to separate the toluene-methanol-water mixture. Two methods of three column extractive distillation and two column extractive distillation using decanter were explored to separate the ternary azeotrope. Diethylene glycol and N-methyl-2-pyrrolidone were used as heavy solvent in the two processes, respectively. The UNIQUAC physical model was used in both simulations. Based on the minimum total annual cost, variables of the two processes were optimized and the results indicated the two column extractive distillation using decanter can save 51.4 % of total annual cost than three column extractive distillation. The dynamics of two column extractive distillation using decanter was studied due to its superiority of economics. Several common control schemes were used to investigate the controllability of two column extractive distillation using decanter and all schemes showed poor controllability on feed composition disturbances. An improved control structure was designed to achieve better control of the process. In the improved control scheme, the temperature controller of column C1 was replaced by a proportional controller, and a certain amount of solvent flow rate was increased. The integral squared error was calculated to compare the dynamic performances of the improved control structure with different solvent flow rate, and a suitable amount of solvent was found in view of the controllability and economy. Keywords: Ternary azeotrope; Extractive Distillation; Controllability; Process evaluation 1. Introduction Toluene and methanol are both widely used as solvents and raw materials in chemical industry. 1

Ternary mixture of toluene, methanol, and water is commonly produced from the pharmaceutical or fine chemical processes [1]. Recycling and reusing the organic chemicals from waste water is a hot topic in view of environmental protection and economy. However, separation of the ternary mixture is difficult because two kinds of azeotropes are formed in this mixture. The heterogeneous azeotrope of toluene-water is composed of 52.30-56.41 mol % water at 357.25-358.15 K under 1 atm and the homogenous azeotrope toluene-methanol is formed with 88.20-88.60 mol % methanol at 336.41-336.95 K under 1 atm [2]. Conventional distillation methods cannot separate the azeotropic mixture effectively, so it is necessary to explore other methods such as azeotropic distillation [3-6], extractive distillation [7-12], pressure-swing distillation [13-19], and other new separation techniques [20-23]. Extractive distillation as one of the most widely used technologies is mainly based on the introduction of a third component to change the relative volatility between the raw material components. Li et al. [8] discussed the separation of benzene-cyclohexane mixture using sulfolane as solvent and designed some retrofitted processes for reducing the energy requirement by combining heat-integrated technology and intermediate heating. Luyben [10] designed a method for separation of methanol-toluene binary azeotropic mixture with an intermediate-boiling solvent triethylamine and indicated indirect sequence (toluene-methanol) was more economical than direct sequence (methanol-toluene). An et al. [11] investigated two cases of n-propanol and water with N-methyl-2-pyrrolidone (NMP) as solvent and ethyl acetate-ethanol with furfural as solvent to verify the economic advantage of process by combining pre-concentration column and extractive distillation column. So far there are few research on the separation of ternary system [24, 25]. Raeva et al. [24] discussed the separation of three ternary systems of ethal acetate-ethanol-water, tetrahydrofuran–methanol–water and acetonitrile–methanol–water and indicated the selection of solvent was not limited the relative volatility. The literatures mentioned above investigated extractive distillation process with two-column or triple-column, which is used to separate binary or ternary azeotropic mixtures. A recent paper by Zhao et al. [26] investigated a method of heterogeneous azeotropic distillation with two-column to separate the ternary azeotrope ethanol-toluene-water and toluene which is one component of the system itself was used as solvent for this process. Comparison of two methods was studied based on the reduce of energy cost and total annual cost. Dynamics of the heterogeneous azeotropic distillation, however, was 2

not given in that study. Research on dynamic control performance is of great importance for a distillation process to maintain the purity of products when encountering the feed flow rate and composition disturbances. In recent years, design and control strategies of distillation process have been investigated for the separation of binary process [27-35]. Yang et al. [31] investigated the design and control of extractive distillation system for benzene-acetonitrile separation using dimethyl sulfoxide as solvent. Gil et al. [32] discussed a process of dehydrating ethanol with solvent glycerol and investigated two different control schemes for this process. Li et al. [34] explored three configurations of extractive distillation process to obtain the optimal separation configurations (extractive dividing-wall column process) and designed three control strategies to investigate the controllability of the extractive dividing-wall column process. However, the research on the control performance of ternary system is rare. In an insightful paper, Luyben [35] studied the dynamic control performance of extractive distillation systems for a ternary mixture benzene-cyclohexane-toluene

with

N,N-Dimethylformamide

as

solvent

and

discussed

controllability of conventional and thermally coupled ternary extractive distillation processes. The three column extractive distillation and two column extractive distillation using decanter were compared based on minimum total annual cost (TAC). TAC, which includes annual investment cost and operating cost, is usually used as an economic evaluation criterion of process design, and it was calculated over a three-year payback period. The operating costs mainly consist of consumption of stream for reboilers and cooling water for condensers. The basis of economics and calculated formulas are summarized in Table 1. The controllability of two column extractive distillation using decanter was investigated and an innovative control scheme was designed to achieve better control. 2. The process of three column extractive distillation The selection of solvent is crucial for extractive distillation process. How to choose a suitable solvent has been discussed in many papers [36-40]. The relative volatility between the light and heavy component is one of the important criterion for the choice of solvent. The greater the relative volatility between the light and heavy component, the easier the azeotrope can be separated. Diethylene glycol (DEG) is chosen as a heavy solvent of this system for the three column extractive distillation. Gramajo de Doz [41] and Tamura [42] tested the liquid-liquid 3

equilibrium parameters of toluene-methanol-water using the model of UNIQUAC. Wang [1] regressed the binary interaction parameters and the liquid-liquid equilibrium data of toluene-methanol-water using the model of NRTL and UNIQUAC, and the results showed that the precision of correlation using UNIQUAC is higher than that of NRTL for the mixture of toluene-methanol-water. We compared the azeotropic data calculated by the model of NRTL-RK and UNIQUAC with the experimental data published in Azeotropic Data [2], and found that the model of UNIQUAC is much more precise. Thus, the model of UNIQUAC is used in this paper. The flowrate of DEG and other relevant design variables in this process are optimized to get the minimal TAC. The sequential iterative optimization procedure for the three column extractive distillation was shown in Fig. 1 and all the optimal results were listed in Table 2. The flowsheet with detailed information of three column extractive distillation is shown in Fig. 2. Both column C1 and C2 are extractive distillation columns, and column C3 is solvent recovery column. The solvent DEG and the azeotropic mixture with 37 mol % toluene, 14 mol % water, and 49 mol % methanol are fed into C1, and the key component of toluene is distillated from the system first. The other two components of water and methanol mixed with DEG are fed into C2 in which methanol is distillated from the top of column C2. Water is distillated from C3, where solvent DEG is separated from water and recycled to C1. Because the solvent DEG has a certain losses throughout the process, the recycled stream should be mixed with a solvent make-up stream. The purity of toluene, methanol and water can reach 99.9 mol %. 3 The process of two column extractive distillation using decanter 3.1 Process design and economic analysis Different with the traditional three column extractive distillation, NMP is chosen as the solvent in the process of two column extractive distillation using decanter. The residue curve map [43-46] (RCM) of this system contains four components (toluene, methanol, water, NMP), and it is divided into two sections. The RCM of ternary system is shown in Figs. 3a and 3b. The residue curves with arrows from the unstable node point to pure NMP and there is no presence of distillation boundary in the RCM. This is an ideal situation for this extractive distillation process. The recycle stream of NMP with the feed of F1 was separated into B1 and D1, namely that the lowest pure component, methanol, and a toluene-water-NMP mixture as F2 is fed into the second column. The point of F2 is separated into B2 (pure NMP) and D2, which is entered into the 4

decanter. Then D2 is separated two-phase of toluene and water (L1 and L2). UNIQUAC model is also used as the thermodynamics model of two column extractive distillation using decanter. The flowrate of NMP and other relevant design variables in this process are optimized to get the minimal TAC. The sequential iterative optimization procedure for the process of two column extractive distillation using decanter was shown in Fig. 4 and all the optimal results were listed in Table 3. The flowsheet with detailed information of two column extractive distillation using decanter is shown in Fig. 5. A heat exchanger is added into the flowsheet because the distillation stream from the solvent recovery column needed to be cooled to 283.15 K. The reason is that toluene and water are miscible with each other at higher temperatures. The feed flowrate of this simulation is set at 100 kmol/h and the feed composition is same with the three column extractive distillation. The optimized flowrate of solvent NMP in recycle stream is set at 80 kmol/h. The solvent NMP and the azeotropic mixture methanol-toluene-water are fed into the extractive distillation column C1. High purity (99.9 mol %) of methanol can be obtained in C1 due to the addition of NMP, which makes the increase of relative volatility between methanol and toluene. The bottom stream of C1 which contains NMP, toluene and water is fed into the solvent recovery column C2, in which NMP gathers in the bottom with the purity of approximately 1. The NMP from the solvent recovery column can be recycled back to extractive distillation column for reusing. The distillation stream with toluene-water is cooled to 283.15 K and then is fed into a decanter. In the decanter, the mixture can be split into two phase, one contains high purity of toluene (99.8 mol %) and the other phase contains high purity of water (approximately 1) at the temperature of 283.15K. The comparison between three column extractive distillation and two column extractive distillation using decanter is shown in Table 4. The result indicates that the process of two column extractive distillation using decanter is much more economical compared with three column extractive distillation and the process of two column extractive distillation using decanter can save 51.4 % of TAC than the three column extractive distillation. 3.2 Dynamic control After the study of steady state, it is also essential to discuss the dynamic control performance. Before exporting the stable file to a dynamic file using Aspen Plus software, the sizes of column sumps and reflux drums are designed to provide a 5 min holdup if half-full. Pumps and valves are 5

set suitable pressure drops in order to meet the dynamic operation [47]. According to the "slope criterion" suggested by Luyben [48], stage 35 and 31 in the column C1 and column C2 are selected as control stage to keep their temperature constant, respectively. There are some basic control structures in different dynamic control schemes. These regulatory control structures are added first and the detailed description are represented as below: (1) Feed flow rate is flow-controlled (reverse acting). (2) The total solvent flow rate is in proportion to the feed flow rate. (3) The pressure in each column is controlled by manipulating the heat removal rate in its condenser (reverse acting). (4) The flow rate of distillation (direct acting) is manipulated to control the reflux drum levels in columns. (5) The flow rate of the bottom (direct acting) is manipulated to control the base level in the extractive column. (6) The temperature of sensitive plate is controlled by manipulating the reboiler heat input in the column (reverse acting). (7) The cooler heat duty (reverse acting) is manipulated to hold the temperature of the solvent feed. Flow rate controllers are set at Kc = 0.5 and τ I = 0.3 min. All liquid level controllers are set at Κc = 2 and τI = 9999 min. The pressure controllers use the default values in Aspen Plus Dynamics. All temperature control structures are set 1 min dead time. The relay-feedback test is run and Tyreus-Luyben tuning is used to determine the controller parameters [47]. 3.2.1 Control structure of two column extractive distillation using decanter with fixed reflux ratio (CS1) The control structure CS1 of two column extractive distillation using decanter with fixed reflux ratio is shown in Fig. 6, and Table 5 gives the ultimate data for the temperature control after tuning. Two kinds of feed disturbances were implemented to evaluate the control performance. One kind of feed disturbance is ± 20 % changes in fresh feed flow rate, and the other is ± 20 % changes in the feed composition. The corresponding dynamic responses are shown in Fig.7. The purity of toluene, methanol and water returned to their specified values after approximately 4 h since the feed flow rate was increased to 120 kmol/h and decreased to 80 kmol/h. When faced 6

with ± 20 % feed composition disturbances (58.8 mol % methanol and 39.2 mol % methanol), the purity of toluene and water returned to their specified value quickly, however, the purity of methanol showed a value of 99.88 mol % (+20 % composition disturbance) and 99.84 mol % (-20 % composition disturbance) at the new steady state. There was a large deviation compared with the specified purity of 99.90 mol %. Hence, it can be concluded that the control structure CS1 cannot handle ± 20 % composition disturbances well. 3.2.2 Control structure of two column extractive distillation using decanter with reflux/feed flow rate (R/F) ratio (CS2) Based on the control structure CS1, an improved structure CS2 was proposed. The control structure CS2, in which the reflux is in proportion to the feed flow rate, is shown in Fig. 8. Table 6 gives the ultimate data for the temperature control after tuning. Two kinds of feed disturbances are also introduced to evaluate the control performance of structure CS2. The corresponding dynamic responses are shown in Fig. 9. The purity of toluene, methanol, and water returned to their specified values after approximately 4 h since the ± 20 % feed flow rate disturbances were introduced. When faced with ± 20% feed composition disturbances, the purity of toluene and water went back their specifications quickly, however the purity of methanol showed a value of 99.60 mol % (- 20 % composition disturbance) at the new steady state. There also was a large deviation compared with the specified purity of 99.90 mol %. Hence, it can be concluded that the control structure CS2 cannot handle ± 20 % feed composition disturbances well. 3.2.3 Control structure of two column extractive distillation using decanter with heat duty of reboiler/feed flow rate (Q/F) ratio (CS3) A control scheme with Q/F ratio has been suggested by Luyben [48]. The control structure CS3, in which the reboiler heat duty is in proportion to the feed flow rate, is shown in Fig. 10. Table 7 gives the ultimate data for the temperature control after tuning. The closed-loop responses to ± 20 % changes in the feed flow rate and ± 20 % changes in the feed composition are displayed in Fig. 11. The purity of toluene, methanol, and water returned to their specified values after approximately 4 h since the feed flow rate was increased to 120 kmol/h and decreased to 80 kmol/h. When faced with ± 20 % feed composition disturbances, the purity of toluene and water reverted to their initial value quickly, however the purity of methanol showed a value of 99.60 7

mol % (- 20 % composition disturbance) at the new steady state. Such a control structure also cannot meet the specified purity of 99.90 mol %. Hence, it can be concluded that the control structure CS3 cannot handle ± 20 % feed composition disturbances well. 3.2.4 Improved control structure (CS4) Through a series of explorations, none of the aforementioned control schemes can handle ± 20 % composition disturbances well. In the recent paper, Luyben [35] draw a conclusion that the existence of four components in the extractive column makes the use of temperature to infer composition problematic. For this, an innovative control scheme was designed, which the temperature controller of column C1 was replaced by a proportional controller. The CS4 and control panel of two column extractive distillation using decanter are shown in Fig. 12. Table 8 gives the ultimate data for the temperature control and composition control after tuning. The closed-loop responses to ± 20 % changes in the feed flow rate and ± 20 % changes in the feed composition are shown in Fig. 13. This improved control structure CS4 handled ± 20 % feed flow rate disturbances well, and when faced with ± 20 % feed composition disturbances, the purity of toluene and water went back their specified values quickly, however the purity of methanol showed a value of 99.60 mol % (- 20 % composition disturbance) at the new steady state. Hence, it can be concluded that the control structure CS4 also cannot handle ± 20 % composition disturbances well. 3.2.5 Improved control structure (CS4) with an adjusted solvent flow rate In literatures published, some variables of process were changed to obtain better control performance [47, 49]. Wang et al. [47] explored the effect of solvent flow rate on controllability of extractive distillation for separating binary azeotropic mixture of n-heptane and isobutanol. After discussing some design variables of the economic optimal design, Wang et al. [47] studied a preferable strategy by increasing solvent flow rates with better controllability at the cost of increasing a small TAC. In view of this, the dynamic control performances with increased solvent flow rate from 80 to 90 kmol/h were investigated. Only the improved control structure CS4 can achieve the desired control effect. The corresponding dynamic responses are shown in Fig. 14 (90 kmol/h solvent). The purity of toluene, methanol, and water returned to their specified values after approximately 4 h since ± 20 % feed flow rate disturbances were introduced. When faced with ± 20 % composition disturbances, 8

the purity of methanol, toluene, and water returned to their specifications approximately 8 h. Although the modified control structure needed more time to reach a steady value, it can able to return back to its desired values. Based on the improved control structure, the amount of solvent was studied. Comparison of dynamic performances of the CS4 with increased solvent flow rate of 83, 85, 90 and 100 kmol/h were studied and the corresponding dynamic responses are shown in Fig. 15. When faced with ± 20 % feed flow rate disturbances, the purity of toluene, methanol, and water returned to their specified values approximately. When faced with ± 20% methanol composition disturbances, the purity of toluene and water meet their specifications quickly, however the purity of methanol showed a value of 99.88 mol % (+ 20 % composition disturbance) at the new steady state with solvent 83 kmol/h. Some indexes could be used to evaluate the performance of a control system, such as the integral of squared error (ISE), integral of time multiplied by absolute error (ITAE), integral of absolute error (IAE), and integral of time multiplied by squared error (ITSE) [50]. Integral squared error (ISE) was used to compare the dynamic performances of the CS4 with different solvent flow rates of 85, 90, and 100 kmol/h, and the related formula of ISE was expressed by Eq. (1). t

ISE   ( y  y sp )2 dt

(1)

t0

where ysp is the set value of purity, y is the actual value of purity, t0 is initial time, and t is terminal time. Table 9 gives ISE indexes for dynamic performance of improved control structure CS4 with solvent flow rate of 85, 90, and 100 kmol/h , and the results indicated that the smallest value for the average of ISE was the improved control structure with solvent 85 kmol/h. Considering the controllability and economy, the optimal design flow rate is 85 kmol/h. The other design variables were fixed at the same values as which for the design flow rate of 80 kmol/h. In this section, several control schemes have been explored, almost all of the control structures exhibited better controllability on the interference of feed flow rate, whereas the control effect of feed composition interference were all unsatisfactory. When the amount of solvent was increased from 80 to 85 kmol/h, only the improved control structure CS4 can achieve the desired control effect. The fact is that CS4 can obtain a better control effect with the solvent increased (only 9

6.25 %), besides, two column extractive distillation using decanter is still more economical compared with three column extractive distillation. Conclusions Two processes of three column extractive distillation and two column extractive distillation using decanter were established to separate the ternary azeotrope of toluene-methanol-water. According to the minimization of total annual cost, two processes were optimized and the results indicated that the process of two column extractive distillation using decanter can save 51.4 % TAC than three column extractive distillation. The dynamics of two column extractive distillation using decanter were studied and four control structures were explored. The proposed control structure CS4 in which the temperature controller of column C1 was replaced by a proportional controller and a certain amount solvent flow rate was increased can achieve the desired control performance. Under the condition of the feed flow rate with 100 kmol/h, comparison of dynamic performances of CS4 with different solvents of 83, 85, 90, and 100 kmol/h indicated the purity of methanol cannot return back to the specified purity of 99.90 mol % with solvent of 83 kmol/h. Integral squared error was used to compare the dynamic performances of the improved control structures with solvents of 85, 90, and 100 kmol/h and the results indicated that the smallest value for the average of integral squared error was the improved control structure with solvent of 85 kmol/h. The two column extractive distillation using decanter with the adjusted solvent content is a preferable choice for the separation of toluene-methanol-water mixture.

Acknowledgment Financial supports from National Natural Science Foundation of China (Project 21776145 and Project 21676152) are gratefully acknowledged.

Symbols used TAC = total annual cost [$/y] C = distillation column B = bottom flow rate [kmol/h] D = distillate flow rate [kmol/h] Feed = feed flow rate [kg/h] Rrec = solvent flow rate [kg/h] RCM = residue curve map 10

ID = diameter of the column [m] NF = location of fresh feed tray NR = location of solvent feed tray NT = number of stages P = pressure [atm] RR = reflux ratio PC = pressure controller of the columns TC = temperature controller of the columns KC = gain of the controller τi = integral time of the controller [min]

11

References [1] B. Wang, W. Ran, Y. Li, J. Xiong, S. Li. Ternary and Quaternary Liquid–Liquid Equilibria for Systems of Water+ Methanol+ Methylanthranilate+ Toluene at Different Temperatures. J. Chem. Eng. Data, 57(2012) 3309-3314. [2] J. Gmehling, J. Menke, J. Krafczyk, J. Fischer, S. Pereira Nunes, K. Peinemann, Azeotropic Data. Wiley-VCH , Wancouver 1994. [3] Q.K. Le, I.J. Halvorsen, O. Pajalic, S. Skogestad, Dividing wall columns for heterogeneous azeotropic distillation, Chem. Eng. Res. Des. 99 (2015) 111-119. [4] V.K.P.J. Devi, P.S.T. Sai, A.R. Balakrishnan, Heterogeneous azeotropic distillation for the separation of n-propanol+water mixture using n-propyl acetate as entrainer, Fluid. Phase. Equilib. 447 (2017) 1-11. [5] V. Gomis, R. Pedraza, M.D. Saquete, A. Font, J. García-Cano, Ethanol dehydration via azeotropic distillation with gasoline fraction mixtures as entrainers: A pilot-scale study with industrially produced bioethanol and naphta, Fuel Process. Technol. 140 (2015) 198-204. [6] M.B. Franke, MINLP optimization of a heterogeneous azeotropic distillation process: Separation of ethanol and water with cyclohexane as an entrainer, Comput. Chem. Eng. 89 (2016) 204-221. [7] A.V. Timoshenko, E.A. Anokhina, A.V. Morgunov, D.G. Rudakov, Application of the partially thermally coupled distillation flowsheets for the extractive distillation of ternary azeotropic mixtures, Chem. Eng. Res. Des. 104 (2015) 139–155. [8] L. Li, Y. Tu, L. Sun, Y. Hou, M. Zhu, L. Guo, Q. Li, Y. Tian, Enhanced efficient extractive distillation by combining heat-integrated technology and intermediate heating, Ind. Eng. Chem. Res. 55 (2016) 8837-8847. [9] H. Wang, X. Cui, C. Li, J. Fang, Separation of Ethyl Acetate-Dichloromethane-Ethanol by Extractive Distillation: Simulation and Optimization, Chem. Eng. Technol. 36 (2013) 627-634. [10] W.L. Luyben, Improved design of an extractive distillation system with an intermediate-boiling solvent, Sep. Purif. Technol. 156 (2015) 336-347. [11] Y. An, W. Li, Y. Li, S. Huang, J. Ma, C. Shen, C. Xu, Design/optimization of energy-saving extractive distillation process by combining preconcentration column and extractive distillation column, Chem. Eng. Sci. 135 (2015) 166-178. [12] Z. Bao, W. Zhang, X. Cui, X. Jiao, Design, Optimization and Control of Extractive Distillation for 12

the Separation of Trimethyl Borate–Methanol, Ind. Eng. Chem. Res. 53 (2012) 14802-14814. [13] Y. Wang, P. Cui, Z. Zhang, Heat-Integrated Pressure-Swing-Distillation Process for Separation of Tetrahydrofuran/Methanol with Different Feed Compositions, Ind. Eng. Chem. Res. 53 (2014) 7186–7194. [14] W. Li, L. Shi, B. Yu, M. Xia, J. Luo, H. Shi, C. Xu, New Pressure-Swing Distillation for Separating Pressure-Insensitive Maximum Boiling Azeotrope via Introducing a Heavy Entrainer: Design and Control, Ind. Eng. Chem. Res. 52 (2013) 7836-7853. [15] Z. Zhu, L. Wang, Y. Ma, W. Wang, Y. Wang, Separating an azeotropic mixture of toluene and ethanol via heat integration pressure swing distillation, Comput. Chem. Eng. 76 (2015) 137-149. [16] W.L. Luyben, Pressure-Swing Distillation for Minimum- and Maximum-Boiling Homogeneous Azeotropes, Ind. Eng. Chem. Res. 51 (2012) 10881–10886. [17] Z. Zhu, D. Xu, X. Liu, Z. Zhang, Y. Wang, Separation of acetonitrile/methanol/benzene ternary azeotrope via triple column pressure-swing distillation, Sep. Purif. Technol. 169 (2016) 66-77. [18] G. Modla, Separation of a Chloroform–Acetone–Toluene Mixture by Pressure-Swing Batch Distillation in Different Column Configurations, Ind. Eng. Chem. Res. 50 (2011) 8204-8215. [19] S. Liang, Y. Cao, X. Liu, X. Li, Y. Zhao, Y. Wang, Y. Wang, Insight into pressure-swing distillation from azeotropic phenomenon to dynamic control, Chem. Eng. Res. Des. 117 (2017) 318-335. [20] F. Chen, K. Mo, Q. Zhang, S. Fei, Y. Zu, L. Yang, A novel approach for distillation of paeonol and simultaneous extraction of paeoniflorin by microwave irradiation using an ionic liquid solution as the reaction medium, Sep. Purif. Technol. 183 (2017) 73-82. [21] S. Mrayed, D. Maccioni, G. Leslie, Efficiency of Membrane Distillation to Produce Fresh Water, Chem. Eng. Technol. 7 (2013) 991-995. [22] L. Eykens, K.D. Sitter, C. Dotremont, L. Pinoy, B.V.D. Bruggen, Membrane synthesis for membrane distillation: A review, Sep. Purif. Technol. 182 (2017) 36-51. [23] H. H. Salih, S.A. Dastgheib, Treatment of a hypersaline brine, extracted from a potential CO2 sequestration site, and an industrial wastewater by membrane distillation and forward osmosis, Chem. Eng. J. 325 (2017) 415-423. [24] V.M. Raeva, A.Y. Sazonova, Separation of ternary mixtures by extractive distillation with 1,2-ethandiol and glycerol, Chem. Eng. Res. Des. 99 (2015) 125-131. 13

[25] Y. Zhao, T. Zhao, H. Jia, X. Li, Z. Zhu, Y. Wang, Optimization of the composition of mixed entrainer

for

economic

extractive

distillation

process

in

view

of

the

separation

of

tetrahydrofuran/ethanol/water ternary azeotrope, J. Chem. Technol. Biot. (2017). [26] L. Zhao, X. Lyu, W. Wang, J. Shan, T. Qiu, Comparison of heterogeneous azeotropic distillation and extractive distillation methods for ternary azeotrope ethanol/toluene/water separation, Comput. Chem. Eng. 100 (2017) 27-37. [27] Y. Wang, Z. Zhang, D. Xu, W. Liu, Z. Zhu, Design and control of pressure-swing distillation for azeotropes with different types of boiling behavior at different pressures, J. Process. Contr. 42 (2016) 59-76. [28] E. Ebrahimzadeh, L. L. Baxter, Plant-wide control of coupled distillation columns with partial condensers, Appl. Therm. Eng. 102 (2016) 785-799. [29] B. Luo, H. Feng, D. Sun, X. Zhong, Control of fully heat-integrated pressure swing distillation for separating isobutyl alcohol and isobutyl acetate, Chem. Eng. Process. 110 (2016) 9-20. [30] H. Zheng, Y. Li, C. Xu, Control of Highly Heat-Integrated Energy-Efficient Extractive Distillation Processes, Ind. Eng. Chem. Res. 56 (2017) 5618-5635. [31] S. Yang, Y. Wang, G. Bai, Y. Zhu, Design and Control of an Extractive Distillation System for Benzene/Acetonitrile Separation Using Dimethyl Sulfoxide as an Entrainer, Ind. Eng. Chem. Res. 52 (2013) 13102-13112. [32] I.D. Gil, J.M. Gómez, G. Rodríguez, Control of an extractive distillation process to dehydrate ethanol using glycerol as entrainer, Comput. Chem. Eng. 39 (2012) 129-142. [33] Y. Cao, J. Hu, H. Jia, G. Bu, Z. Zhu, Y. Wang, Comparison of pressure-swing distillation and extractive distillation with varied-diameter column in economics and dynamic control, J. Process. Contr. 49 (2017) 9-25. [34] L. Li, L. Guo, Y. Tu, N. Yu, L. Sun, Y. Tian, Q. Li, Comparison of different extractive distillation processes for 2-methoxyethanol/toluene separation: Design and control, Comput. Chem. Eng. 99 (2017) 117-134. [35] W.L. Luyben, Control Comparison of Conventional and Thermally Coupled Ternary Extractive Distillation Processes, Chemical Engineering Research & Design, 106 (2016) 253-262. [36] M.T.G. Jongmans, A. Londoño, S.B. Mamilla, H.J. Pragt, K.T.J. Aaldering, G. Bargeman, M.R. Nieuwhof, A.T. Kate, P. Verwer, A.A. Kiss, Extractant screening for the separation of dichloroacetic 14

acid from monochloroacetic acid by extractive distillation, Sep. Purif. Technol. 98 (2012) 206-215. [37] A.Y. Sazonova, V.M. Raeva, A.K. Frolkova, Design of extractive distillation process with mixed entrainer, Chem. Pap. 70 (2016) 594-601. [38] H. Matsuda, H. Takahara, S. Fujino, D. Constantinescu, K. Kurihara, K. Tochigi, K. Ochi, J. Gmehling, Selection of entrainers for the separation of the binary azeotropic system methanol + dimethyl carbonate by extractive distillation, Fluid. Phase. Equilib. 310 (2011) 166-181. [39] Z.G. Zhang, D.H. Huang, M. Lv, P. Jia, D.Z. Sun, W.X. Li, Entrainer selection for separating tetrahydrofuran/water azeotropic mixture by extractive distillation, Sep. Purif. Technol. 122 (2014) 73-77. [40] S. Yuan, C. Zou, Y. Hong, Z. Chen, W. Yang, Study on the separation of binary azeotropic mixtures by continuous extractive distillation, Chem. Eng. Res. Des. 93 (2015) 113-119. [41] M.B. Gramajo de Doz, C. M. Bonatti, H. N. Sólimo. Liquid–liquid equilibria of ternary and quaternary systems with two hydrocarbons, an alcohol, and water at 303.15 K: systems containing 2, 2, 4-trimethylpentane, toluene, methanol, and water, or 2, 2, 4-trimethylpentane, toluene, ethanol, and water. Fluid. Phase. Equilib. 205(2003) 53-67. [42] K. Tamura, Y. Chen, T. Yamada. Ternary and quaternary liquid− liquid equilibria for fuel additives of the water+ methanol+ toluene and water+ methanol+ toluene+ methyl tert-butyl ether or tert-amyl methyl ether systems at 298.15 K. J. Chem. Eng. Data, 46(2001) 1381-1386. [43] O.A. Perederic, V. Pleşu, P. Iancu, G. Bumbac, A.E. Bonet-Ruiz, J. Bonet-Ruiz, B. Muchan, Simulation and process integration for tert-amyl-methyl ether (TAME) synthesis, Comput. Chem. Eng. 83 (2015) 79-96. [44] M. Meyer, J. M. Reneaume, J. M. Lelann, A general approach to generate distillation regions for azeotropic and heteroazeotropic multicomponent mixtures, Comput. Chem. Eng. 23 (1999) S101–S104. [45] L.M.C. Silva, S. Mattedi, R. Gonzalez-Olmos, M. Iglesias, Azeotropic behaviour of (benzene+cyclohexane+chlorobenzene) ternary mixture using chlorobenzene as entrainer at 101.3 kPa, J. Chem. Thermodyn. 38 (2006) 1725-1736. [46] V. Gerbaud, J. Xavier, R.D. Ivonne, B. Olivier, R. Olivier, V. Alain, C. Pierre, Practical residue curve map analysis applied to solvent recovery in non-ideal binary mixtures by batch distillation processes, Chem. Eng. Process. 45 (2006) 672-683. 15

[47] Y. Wang, S. Liang, G. Bu, W. Liu, Z. Zhang, Z. Zhu, Effect of Solvent Flow Rates on Controllability of Extractive Distillation for Separating Binary Azeotropic Mixture, Ind. Eng. Chem. Res. 54 (2015) 12908-12919. [48] W.L. Luyben, Distillation Design and Control Using Aspen™ Simulation, John Wiley & Sons, New Jersey 2013. [49] W.B. Ramos, M.F.D. Figueirêdo, K.D. Brito, S. Ciannella, L.G.S. Vasconcelos, R.P. Brito, Effect of Solvent Content and Heat Integration on the Controllability of Extractive Distillation Process for Anhydrous Ethanol Production, Ind. Eng. Chem. Res. 55 (2016) 11315–11328. [50] R.A. Krohling, J. P. Rey, Design of optimal disturbance rejection PID controllers using genetic algorithms, Ieee. Evolut. Comput. 5 (2001) 78-82.

16

Table 1. Basis of economics and calculated formulas.

Items

Formulas and values

column vessel

column diameter (D) = Aspen tray sizing column length (L) = NT trays with 2 ft spacing plus 20% extra length investment cost=17640D1.066L0.802

condensers

where D and L are in m

heat transfer coefficient = 0.852 kW/ (K·m2) differential temperature = log-mean temperature difference of inlet and outlet temperature differences investment cost = 7296A0.65 where A is in m2

utility prices

low pressure steam (160 ℃) = $7.78/GJ medium pressure steam (184 ℃) = $8.22/GJ high pressure steam (254 ℃) = $9.88/GJ cooling water = $0.354/GJ

TAC = (investment cost/payback period) + operating cost payback period

17

Table 2. The optimization results of three column extractive distillation.

Variables

Column 1

Column 2

Column 3

P (atm)

0.1

1

0.5

NT

45

45

29

NF

32

21

25

NR

3

Rrec (kmol/h)

275

RR

0.1

0.9

2.1

ID (m)

1.5221

0.8049

Annual operating cost ($/y)

503649.1

0.8810 971893.9

Total capital investment ($/y)

1443948.3

TAC ($/y)

2201095.1

18

244236.0

Table 3. The optimization results of two column extractive distillation using decanter.

Variables

Column 1

Column 2

P (atm)

1

1

NT

51

39

NF

26

7

NR Rrec (kmol/h)

4 80

RR

1.51

1.01

ID (m)

0.85

0.98

Annual operating cost ($/y)

361806.7

401751.0

Total capital investment ($/y)

920366.3

TAC ($/y)

1070346.5

19

Table 4. The comparison of the two processes. three column extractive distillation

two column extractive distillation using decanter

Purity of toluene (mol %)

99.9

99.8

Purity of methanol (mol %)

99.9

99.9

Purity of water (mol %)

99.9

99.9

Purity of entrainer (mol %)

99.99

99.99

Flowrate of entrainer (kmol/h)

275

80

Total capital investment ($/y)

1443948.3

915264.0

Annual operating cost ($/y)

1719779.0

700858.8

TAC ($/y)

2201095.1

1070346.5

20

Table 5. Transmitter Ranges, Controller Output Ranges, and Tuning Parameters of Two Temperature Controllers of Optimal Design.

Parameter

TC1

TC2

controlled variable

T1,35

T2,31

manipulated variable

R (Fixed)

R (Fixed)

transmitter range (℃)

0−179.8

0−344.3

controller output range

0−11.87 GJ/h

0−10.15 GJ/h

gain K c

2.63

0.29

integral time τI (min)

9.2

15.8

21

Table 6. Transmitter Ranges, Controller Output Ranges, and Tuning Parameters of Two Temperature Controllers of Optimal Design.

Parameter

TC1

TC2

controlled variable

T1,35

T2,31

manipulated variable

R/F

R/F

transmitter range (℃)

0−179.8

0−344.3

controller output range

0−11.87 GJ/h

0−10.15 GJ/kmol

gain K c

2.63

0.29

integral time τI (min)

9.2

15.8

22

Table 7. Transmitter Ranges, Controller Output Ranges, and Tuning Parameters of Two Temperature Controllers of Optimal Design.

Parameter

TC1

TC2

controlled variable

T1,35

T2,31

manipulated variable

Q/F

Q/F

transmitter range (℃)

0−179.8

0−344.3

controller output range

0−0.12 GJ/kmol

0−0.10 GJ/kmol

gain K c

3.03

0.29

integral time τI (min)

9.2

15.8

23

Table 8. Transmitter Ranges, Controller Output Ranges, and Tuning Parameters of Optimal Design.

Parameter

CC1

TC2

controlled variable

X (methanol)1,1

T2,31

manipulated variable

T1,35

R (Fixed)

transmitter range

0−2.00

0−344 (℃)

controller output range

0−3.02

0−10.16 GJ/h

gain K c

25.2

0.27

integral time τI (min)

10.6

13.2

24

Table 9. ISE index for dynamic performance of improved control structure CS4 with solvent flow rate of 85, 90, and 100 kmol/h.

Flowrate (+20 %)

Flowrate (-20 %)

Methanol (+20 mol %)

Methanol (-20 mol %)

85 kmol/h

90 kmol/h

100 kmol/h

ISE

ISE

ISE

Methanol (×10-6)

11.06

236

292

Water (×10-6)

0.0001995

0.0001006

0.0001185

Toluene (×10-6)

1.84

2.09

2.02

Average (×10-6)

4.30

79.39

98.01

Methanol (×10-6)

1.42

12.8

21.8

Water (×10-6)

0.0001995

0.0001105

0.0001055

Toluene (×10-6)

4.76

4.96

4.14

Average (×10-6)

2.06

5.82

8.65

Methanol (×10-6)

1.57

4.22

9.66

Water (×10-6)

2.04

1.08

1.09

Toluene (×10-6)

3.41

3.79

2.93

Average (×10-6)

2.34

3.03

4.56

Methanol (×10-6)

711

805

741

Water (×10-6)

0.198

0.203

0.281

Toluene (×10-6)

1.59

1.51

0.107

Average (×10-6)

237.59

268.91

247.13

25

Figure1. Sequential iterative optimization procedure for three column extractive distillation.

26

Figure 2. The flowsheet of three column extractive distillation.

27

(a)toluene(TOLUE-01)-methanol(METHA-01)-NMP Figure 3. The residue curve map of ternary mixture.

28

(b)NMP-toluene-water

Figure 4. Sequential iterative optimization procedure for two column extractive distillation using decanter. 29

Figure 5. The flowsheet of two column extractive distillation using decanter.

30

Figure 6. Control structure of two column extractive distillation using decanter with fixed ratio (CS1).

31

Figure 7. Dynamic performances of the control structure of two column extractive distillation using decanter with fixed ratio (CS1): (a) feed flow rate disturbances; (b) feed composition disturbances

32

Figure 8. Control structure of two column extractive distillation using decanter with reflux/feed flow rate (R/F) ratio (CS2)

33

Figure 9. Dynamic performances of the control structure of two column extractive distillation using decanter with reflux/feed flow rate (R/F) ratio (CS2): (a) feed flow rate disturbances; (b) feed composition disturbances

34

Figure 10. Control structure of two column extractive distillation using decanter with heat duty of reboiler/feed flow rate (Q/F) ratio (CS3)

35

Figure 11. Dynamic performances of the control structure of two column extractive distillation using decanter with heat duty of reboiler/feed flow rate (Q/F) ratio (CS3): (a) feed flow rate disturbances; (b) feed composition disturbances

36

Figure 12. Improved control structure of two column extractive distillation using decanter (CS4) (80 kmol/h)

37

Figure 13. Dynamic performances of the improved control structure of two column extractive distillation using decanter (CS4) (80 kmol/h): (a) feed flow rate disturbances; (b) feed composition disturbances

38

Figure 14. Dynamic performances of the improved control structure of two column extractive distillation using decanter (CS4) (90 kmol/h): (a) feed flow rate disturbances; (b) feed composition disturbances

39

Figure 15. Comparison of dynamic performances of the improved control structure of two column extractive distillation using decanter (CS4) (83, 85, 90, 100 kmol/h): (a) feed flow rate disturbances; (b) feed composition disturbances 40

Highlights 1) The traditional extractive distillation with triple-column and the extractive distillation process with two-column decanter for separation of methanol-toluene-water mixture was compared based on minimum total annual cost. 2) Several common control schemes were used to investigate the controllability of extractive distillation with two-column decanter. 3) An innovative control scheme was designed for extractive distillation process with two-column decanter to achieve better control.

41