Separation and Purification Technology 169 (2016) 66–77
Contents lists available at ScienceDirect
Separation and Purification Technology journal homepage:
www.elsevier.com/locate/seppur
Separation of acetonitrile/methanol/benzene ternary azeotrope via triple column pressure-swing distillation Zhaoyou Zhu, Dongfang Xu, Xingzhen Liu, Zhen Zhang, Yinglong Wang
⇑
College of Chemical Engineering, Qingdao University of Science and Technology, Qingdao 266042, China
ar ti cl e
in fo
Article history: Received 28 April 2016 Received in revised form 2 June 2016 Accepted 3 June 2016 Available online 4 June 2016 Keywords: Pressure-swing distillation Residue curve maps Separation configuration Ternary azeotrope
a bs tra c
t
Acetonitrile/methanol/benzene mixture forms more than one different azeotropes and its triangular diagram presents several distillation boundaries at atmospheric pressure. A process named as triple column pressure-swing distillation (TCPSD) was proposed to separate this complex ternary system. The feasibility of the process was confirmed using residue curve maps and rigorous steady-s tate simulatio ns were implemented on Aspen Plus. On basis of minimum total annual cost, several operating parameters were optimized by pressure-swing optimization software using the sequential iterative optimizati on procedure and the economics of TCPSD were compared with four different separation configurations. The results demonstrated that the A-M-B separation configurat ion was the most optimal column sequence in global optimizatio n to separate acetonitrile/methanol/benzene azeotropic mixture using TCPSD. TCPSD may arouse the interest of researchers in various fields and can assist engineers to select the optimal separating process. 2016 Elsevier B.V. All rights reserved.
1. Introduction
There is a common and complex ternary system containing acetonitrile, methanol, and benzene in the chemical and pharmaceutical industries. Methanol (CH 3OH), acetonitrile (CH 3CN), and benzene (C6H6) are commonly used as organic solvents in chemical and pharmaceutical industries due to their excellent physicochemical properties [1–3]. Therefore, it is attractive and necessary to separate and reuse acetonitrile, methanol, and benzene in order to protect environment and conserve resource. However, conventional separation process cannot separate ternary azeotropic mixture efficiently due to the complexity of acetonitrile/methanol/ benzene system which has more than one different azeotropes and distillation boundaries in the ternary system . The boiling points of acetonitrile, methanol, and benzene are 354.63, 337.68, and 353.28 K under atmospheric pressure, respectively. The minimum-boiling azeotropes have compositions of 76.49 wt% (mass fraction) methanol at 336.55 K between methanol and acetonitrile, 38.53 wt% methanol at 331.39 K between methanol and benzene, and 31.66 wt% acetonitrile at 345.94 K between acetonitrile and benzene. Hence, the acetonitrile/methanol/benzene system belongs to the Serafimov class 3.0-2 [4,5], and Modla et al.
⇑
Corresponding author. E-mail address:
[email protected] (Y. Wang).
http://dx.doi.org/10.1016/j.seppur.2016.06.009 1383-5866/ 2016 Elsevier B.V. All rights reserved.
[6] proposed some feasible methods using batch pressure-swing distillation for this classification. This paper studied the continuous pressure-swing distillation to separate the complex ternary mixture of acetonitrile/methanol/benzene into desired pure products with the minimum total annual cost (TAC). Many published literatures focused on the separation of ternary mixture in recent years. There are some methods for separation of ternary azeotropic system such as membrane [7], continuous distillation [8,9], and batch distillation [6,10–13]. The concentration of water was improved at 94.9 wt% for the purification of ethyl acetate/ethanol/water ternary azetropic mixtures using membrane by Xia et al. [7]. Membrane separation has broken the azeotropic composition, however, the engineer should balance the economy and products’ purities. Modla [11] introduced a new triple column configuration applied in pressure-swing batch distillation for separation of chloroform/acetone/toluene. Phimister and Seider [12] investigated the operation of a semicontinuous, middle-vessel column to separate a nearly-ideal ternary mixture. Huang et al. [8] studied direct and indirect sequence ideal heart-integrated distillation column to separate a simple close-boiling ternary mixture. As for the continuous distillation, Zong [9] studied the separation of acetonitrile/methanol/benzene system by the analysis of the distillation curves and distillation region characteristics via quintuple columns pressure-swing distillation, and the product purity was improved to 99.0 wt%. The continuous distillation can be
67
Z. Zhu et al. / Separation and Purification Technology 169 (2016) 66–77
Nomenclature
B D Feed ID NT NF NREC P REC RR TAC
bottom flow rate [kg/h] distillate flow rate [kg/h] feed flow rate [kg/h] diameter of the column [m] number of stages number of feed locations number of recycle location pressure [kPa] recycle flow rate [kg/h] reflux ratio total annual cost [$/y]
Acronyms ED extraction distillation PSD pressure-swing distillation PSDOS pressure-swing distillation optimization software RCMs residue curve maps TCPSD triple column pressure-swing distillation Indices 1, 2, 3
column index
competitively alternative in comparison with batch distillation due to the properties of larger throughput. Pressure-swing distillation (PSD) [14–22] avoids the potential problem of introducing the third component and has gained lots of attention from researchers in recent years, by comparison with special distillation such as extractive distillation (ED) [23–28], azeotropic distillation [29–32], and reactive distillation [33–35]. Munoz [36] studied ED and PSD process for isobutyl alcohol/isobutyl acetate separation using the commercial simulator Aspen HYSYS while Lladosa et al. [37] investigated the separation of din-propyl ether and n-propyl alcohol and found that PSD was more attractive than ED using an entrainer. For separating ternary mixture containing azeotropes, Knapp and Doherty [38] introdu ced the method of the triple column pressure-swing distillation (TCPSD). The simulation and economic analysis for the ternary system, however, were not given in Knapp’s paper. For a complex and special
data [9,48] was regressed to accurately calculate the thermodynamic property of acetonitrile/methanol/benzene system. The Wilson model with the regressed interaction parameters fits well with the vapor-liquid equilibrium of the acetonitrile/methanol/ benzene system. The regressed interaction parameters for the Wilson model are shown in Table 1.
ternary azeotropic system which forming more than one different azeotrope and containing several distillation boundaries, one or two columns cannot achieve a desired separation, hence, TCPSD was proposed in this paper. The influence of pressure on azeotropic composition and azeotropic temperature for binary system is always used to prove pressure-sensitivity [39–41] which is one of decisive factors for PSD, and triangular diagrams including residue curve maps (RCMs) [5,38,42–46] are usually used to analyze the feasibility of multicomponent separation. Fien and Liu [47] reviewed the design of separation processes and illustrated the use of ternary diagrams including RCMs for the feasibility analysis, flowsheet development, and preliminary design. Kiva and Krolikowski [4] presented different approaches to determine the product composition region for azeotropic mixture with different shaped distillation lines. Hence, pressure-swing distillation makes the separation possible by moving distillation boundaries that lie between the corresponding purity values of desired products at different pressures. In this paper, a rigorous separation process of TCPSD was explored for the complex ternary system and four different separation sequences were analyzed. The optimum design of TCPSD to obtain more economically competitive process based on the minimization TAC was achieved by a sequential iterative method.
The total capital investment was divided by a five-year payback period and the operating time of designs was set at 8000 h/year. The column and sieve plate parameters were calculated via the ‘‘tray sizing” function in Aspen Plus. The overall heat transfer coefficients recommended by Luyben [49] are 0.852 kW/(K m2) for condensers and 0.568 kW/(Km2) for reboilers, respectively. The basis of the economics, the sizing relationships, and parameters are taken from Douglas [50].
2.2. Economics TAC consists of annual operating cost and capital investment and is used to evaluate different process design. Annual operating cost comprises annual steam and cooling water cost, while annual capital investment mainly includes cost of column vessel, plate, and heat exchangers. The capital investment usually ignores small items such as reflux drums, pumps, valves, and pipes because there are a large gap between major investment and small items costs.
3. TCPSD process
The feed flow rate is 1000 kg/h with 70.0 wt% methanol, 20.0 wt % acetonitrile, and 10.0 wt% benzene. The purity of three products was set as 99.9 wt%.
3.1. Process analysis using RCMs Triangular diagram containing RCMs is used to describe the equilibrium relationships. The RCMs for the acetonitrile/methanol/benzene ternary system at 101.33 and 607.95 kPa are drawn and shown in Figs. 1 a and b. That figures portray that all residue Table 1
The regressed interaction parameters of the Wilson model.
2. Basis of design
2.1. Property method The quality of physical model parameters guarantees the accuracy of simulated results. The interaction parameters using the ‘‘Data Regression” function based on the vapor-liquid equilibrium
CH4O/C2 H3 N Aij Aji Bij Bji Cij Cji
CH4O/C6H6
0 0
5.07 97.08 262.83 0 0
9.70 0 2399.15 1965.61 0 0
C2H3N/C6H6 0
233.59 179.20 0 0
68
Z. Zhu et al. / Separation and Purification Technology 169 (2016) 66–77
Fig. 1. RCMs for A-M-B separation configuration: (a) 101.33 kPa; (b) 607.95 kPa; (c) Summary of the distillation boundaries positions at 607.95–101.33–607.95 kPa.
69
Z. Zhu et al. / Separation and Purification Technology 169 (2016) 66–77
3.2. Pressure determination
curves in a distillation region point from unstable node toward stable node, hence, three pure component points are all stable node and the methanol/benzene azeotrope point is an unstable node. The other two azeotropes are saddle point which has no residue curves coming in or going out. There are two distillation boundaries moving from the methanol/benzene azeotrope to the acetonitrile/benzene and acetonitrile/methanol azeotropes in the RCMs, respectively. Distillation boundaries divide the products into three differen t distillation regions (region 1, region 2, and region 3 as illustrated in Fig. 1 ). The separation principle could be overviewed by overlap of the boundaries as depicted in Fig. 1 c. The gray lines stand for the mass balance lines and the dashed and solid lines are distillation boundaries at 101.33 and 607.95 kPa, respectively. The recycle stream (D 3) assists the feed
P1, the pressure of the first column, was determined by minimizing TAC when P 2 and P 3 were fixed at 101.33 and 709.28 kPa. When the high-pressure column is operated at 709.28 kPa, medium-pressure steam other than high-pressure steam can be used as heating medium in the reboiler in order to reduce the energy operating costs. The pressure should be compatible with the choice of heating steam and the pressure range from 303.98 to 709.28 kPa was considered. The other parameters such as NT, NF, and RR were optimized when the pressure changed. As shown in Fig. 2 a, there is a minimal value of TAC and the corresponding pressure of 607.95 kPa is selected for C 1. The cooling water of 305.15 K can be used when the tempera-
to be separated into D 1 and B 1, namely that the highest pure component, acetonitrile, moves to the bottom and an acetonitrile/ methanol/benzene ternary mixture stream is obtained as the distillate of the first column (C 1) operating at 607.95 kPa. The distillate crossing the distillation boundary from region 1 to region 2 enter the second column (C 2) as a new feed operating at 101.33 kPa. Then D1 splits into D 2 and B 2 so that the highest component, methanol, achieve the high purity in the product stream. The distillate of the distillation column is also a ternary mixture as a feed stream entering the third column (C 3). The mixt ure stre am in region 3 can be separated into the benzene product stream (B 3) and D3 at 607.95 kPa. The acetonitrile/methanol/benzene ternary system can be separated in the TCPSD process with the A-M-B separation configuration.
ture of reflux drum is 325.85 K at 81.06 kPa. The pressure of C 2 was evaluated from 81.06 to 121.59 kPa to find the minimum TAC. As shown in Fig. 2 b, the TAC decreases with the pressure of C2 increases from 81.06 to 101.33 kPa, and increases when the pressure rises from 101.33 to 121.59 kPa. The pressure of C 2 is determined as 101.33 kPa. The pressure of the third column was considered from 303.98 to 709.28 kPa. Fig. 2 c illustrates that the trend of TAC decreases from 303.98 to 607.95 kPa then increases when the pressure rises from 607.95 to 709.28 kPa. The result demonstrates that 607.95 kPa is suitable for the pressure of C 3. The optimal pressure for C 1 and C 2 can be readjusted by minimizing TAC and the corresponding results are shown in Table 2. The results show that the operating condition of case 2 is more
Fig. 2. (a) Effect of pressure for C
1
on TAC; (b) Effect of pressure for C
2
on TAC; (c) Effect of pressure for C
3
on TAC.
Z. Zhu et al. / Separation and Purification Technology 169 (2016) 66–77
70 Table 2
Case studies of A-M-B separation configuration.
Variables (kPa) P1 (kPa) P2 (kPa) P3 NT1 NT2 NT3 NF1/NREC NF2 NF3 RR1 RR2 RR3 (m)ID1
Case1 506.63 101.33 607.95 48 54 14 38/32 25 2 4 2.01 2.60 0.10 0.72
(m)ID2 0.74 (m)ID3 0.39 REC(kg/h) 1314 Total Capital investment ($) 1.825 Annual Operating cost ($/y) 5.928 TAC($/y) 9.577
Case2
Case3
607.95 101.33 607.95 47 50 15
709.28 101.33 607.95 45 46 21
37/28 18 4 2.13 3.19 0.01 0.66 0.69 0.32 106 105 5 10
50 51 16 37/14
11 2
16 3 1.33 2.59 1.28 0.61
0.73 0.52 953.0 1.636 106 4.966 105 8.239 105
Case4
Case5
607.95 101.33 506.63 46 50 13 40/29 18 4 5 1.94 2.76 0.21 0.66
607.95 101.33 709.28 45 50 14 37/28 18 1.85 2.96 0.14 0.65
0.68 0.36
0.70 0.34
1231
1050 1.828 106 5.678 105 9.333 105
1061 1.674 106 5.036 105 8.383 105
Case6
Case7
607.95 91.19 607.95 51 52 14 37/28 17
607.95 111.46 607.95
42/27
1.81 2.64 0.39 0.65
1.83 2.82 0.07 0.65
0.70 0.38
0.70 0.39
1087 1.671 106 5.085 105 8.427 105
1119 1.683 106 5.103 105 8.469 105
1.679 106 5.055 105 8.413 105
suitable for the triple column process. The pressure combination of 607.95–101.33–607.95 kPa is used for the three columns.
3.3. Optimization The optimization of double column PSD via the sequential iterative optimization procedure was studied by researchers [39,51–54]. There is no literature on the optimization procedure of TCPSD because a lot of parameters need to be determined. The optimization procedure of TCPSD following the sequential iterative optimization procedure is given and shown in Fig. 3 . Design variables such as product purities must be specified in the process optimization. The number of stages in C , C , and C columns 1 2 (NT1, NT2, and NT3), the reflux ratios of three columns (3 RR1, RR2, and RR3), the feed locations ( NF1, NF2, and NF3), and recycle location (NREC) need to be optimized to achieve the optim al operating parameters and minimal TAC. High reflux ratio improves product purity but increases heat duties of reboiler, so there is an optimum RR3 as well as RR2 and RR1 that gives the minimum TAC. The optimization process is so extraordinary complicated, so an efficient calculation software is developed to implement the optimization procedures and get more accurate results. This software named as Pressure-Swing Distillation Optimization Software (PSDOS) [55] by calling the Aspen Plus with the help of Visual Basic was developed on the basis of the sequential iterative optimization procedure and was authorized by the copyright protection center of China. The pressure combination was fixed as 607.95–101.33–607. 95 kPa for three columns. There are initially 50 stages in all columns including the condensers and reboilers and the initial reflux ratios are set as 2. The ‘‘Design Spec” and ‘‘Vary” functions of columns in Aspen Plus are utilized to adjust the distillate flow rate to obtain 99.9 wt% products. Feed locations (NF1, NF2, and NF3) and recycle location (NREC) are optimized as the innermost loop to achieve the minimal TAC. Then reflux ratios are determined. Feed locations are also reconsidered during the optimization of RR. Finally, an optimized value of the number of stage ( NT1, NT2, and NT3) is obtained. The number of stages has an impact on the separation efficiency and the equipment investment. The results obtained from PSDOS for the TCPSD process contain the operating parameters of devices, stream information, heat duties, and equipment sizes. The flowsheet of the optimized TCPSD is shown in Fig. 4 . The reboiler and condenser duties of TCPSD
Fig. 3. Sequential iterative optimization procedure for TCPSD.
Z. Zhu et al. / Separation and Purification Technology 169 (2016) 66–77
71
Fig. 4. Flowsheet of the optimal TCPSD process for A-M-B separation configuration.
Fig. 5. Composition profiles of optimized TCPSD process.
process are 2.050 103 and 2.003 103 kW, respectively. Fig. 5 shows the composition profiles of TCPSD process with the stream points and the different sections of the column labeled. The TAC calculated for the optimized system is 8.239 105 $/y. The annual operating cost and total capital investment are 4.966 105 $/y and 1.636 106 $, respectively.
4. Separation configuration
The different separation configurations could be feasible to separate the ternary azeotrope with the different pressure combination. The separation sequence is an important decisive factor for the energy consumption when the feed composition fixed in the triangular diagrams. As seen from Fig. 1 c, feed composition is
located in the region 1 nearing the distillation boundary between regions 1 and 2. Hence, designs and optimizations of different separation configurations such as A-B-M, M-A-B, and M-B-A were analyzed by minimizing TAC. In case of A-B-M separation configuration, feed stream splits into acetonitrile product and the distillate of C 1 should be located in region 3 at second operating pressure to obtain pure benzene component in C 2. Ensuring to extract methanol with purity of 99.9 wt% from the bottom of C 3, distillation boundaries can be shifted with the change of P 3 in order to make D 2 situated in region 2. Material balance lines indicated the feasibility of this separation sequence process as shown in Fig. 6 . Several operating parameters are optimized using PSDOS similarly in A-M-B separation configuration. The results demonstrate that the pressure combination of 303.98–709.27–60.80 kPa is carried out to obtain the optimal
Z. Zhu et al. / Separation and Purification Technology 169 (2016) 66–77
72
Fig. 6. RCMs for A-B-M separat ion configuration: (a) 303.98 kPa; (b) 709.27 kPa; (c) 60.80 kPa; (d) Summary of the distillati on boundaries positions at 303.98–709.27– 60.80 kPa.
Table 3
Case studies of A-B-M separation configuration.
Variables (kPa) P1 (kPa) P2 (kPa) P3 NT1 NT2 NT3 NF1/NREC NF2 NF3 RR1 2 RR RR3 (m) ID1 (m) ID2 (m) ID3 REC(kg/h) Total Capital investment ($) Annual Operating cost ($/y) TAC($/y)
Case1 202.65 709.27 60.80 36 13 44 3/24 2 23 0.60
Case2
Case3
303.98 709.27 60.80 39 18 46
7 21
23
0.01 1.41 0.95 0.61 1.05 3251 2.512 8.092 1.311
48 15 43 4/18 4
2
38 25 46 28/29 2
303.98 810.60 60.80 38 18 47 3/17
24
24
Case6 303.98 709.27 70.93
6/21
2 22
0.89
1.45
1.40
1.47
0.01 1.65 0.83 0.52 0.90
0.15 1.4 0.85 0.62 1.04
2.40 1.63 0.83 0.95 0.90
0.01 1.74 0.82 0.52 0.91
0.01 1.81 0.84 0.53 0.90
2060
Case5
303.98 607.95 60.80 37 14 47 3/19
1.49
106 105 6 10
Case4
405.30 709.27 60.80
3029 2.277 106 6.826 105 1.138 106
2087 2.668 106 8.106 105 1.344 106
2051 2.639 106 9.036 105 1.431 106
2153 2.305 106 6.803 105 1.141 106
2.245 106 7.191 105 1.168 106
Z. Zhu et al. / Separation and Purification Technology 169 (2016) 66–77
73
Fig. 7. Flowsheet of the optimal TCPSD process for A-B-M separation configuration.
operating parameters and minimize the TAC for this separating condition, as is shown in Table 3. In cases, a lower pressure condition of C 3 is ignored because that a reflux drum temperature of 314.35 K occurs at a pressure of 50.66 kPa then cheaper cooling water is not satisfied the condensation task. The optimization results show that the TAC calculated for the optimized A-B-M separation configuration process is 1.138 106 $/y, annual operating cost is 6.826 105 $/y, and total capital investment is 2.277 106 $. The flowsheet of the optimal TCPSD with A-B-M sequence is in detail shown in Fig. 7 . Fig. 8 describes RCMs at different pressures and the material
TAC of the TCPSD process with A-M-B sequence decreases by 27.60% comparing with A-B-M sequence process while annual operating cost and total capital investment decrease by 27.25 and 28.15%, respectively. In M-A-B sequence, the calculated TAC of TCPSD is 1.204 106 $/y, which is much bigger than that of AM-B sequence PSD. The economy of M-B-A sequence is the worst in four different separation configurations, TAC of those increases by 34.61% than the optimized A-M-B sequence. The results indicate that a reasonable global minimal TAC exists in the overall operating pressures for the separation of complex ternary acetonitrile/ methanol/benzene system and A-M-B separation configuration is
balance lines in M-A-B separation configuration which indicate the feasibility of separation configuration using TCPSD. This process, operating at three different pressures, is firstly fed to low pressure column which obtain a 99.9 wt% product of methanol while the distillate is pressurized and fed to the second column at a slightly higher pressure. Hence, pure acetonitrile and benzene component are given as bottom product in C 2 and C3, respectively. The optimum values in each diagram are the optimum parameters in global optimization and the partially results simulated by Aspen Plus is listed in Table 4 . The pressure combination which is selected with 91.19–303.98–506.63 kPa is optimized for the three columns due to the minimum TAC. The flowsheet of the optimal TCPSD with M-A-B sequence is shown in Fig. 9 which in detail offers stream information and heat duties. By comparison with M-A-B sequence, the latter two columns are inverted in M-B-A separation configuration and the feasible material balance lines are described in triangular phase diagram as seen in Fig. 10 . The operating pressures of TCPSD process are shifted, adjusted, and optimized while operating parameters are optimized by sequential iterative optimization procedure. The minimum operating parameters of TCPSD process is listed at Table 5 in detail and the minimum TAC of pressure combination of 101.33–709.28–405.30 kPa is 1.260 106 $/y, hence, the optimization information is portrayed in Fig. 11. Four different separation configurations simulated with different pressures were analyzed and compared to separate a complex ternary mixture of 20 wt% acetonitrile, 70 wt% methanol, and 10 wt% benzene. The product purity of 99.9 wt% can be achieved through the TCPSD in all different separation configurations. The
more attractive in terms of steady-state economics.
5. Conclusion
A TCPSD process for separating complex ternary azeotropic mixture containing acetonitrile, methanol, and benzene was explored on Aspen Plus platform. The feasibility of TCPSD for separating this mixture was analyzed by its triangular diagrams including RCMs and showing the shifts of distillation boundaries as operating pressure changes. Since large amount parameters need to be optimized on basis of total annual cost, the optimal conditions on the basis of minimum TAC were obtained using the sequential iterative optimization procedure by PSDOS that saves the optimization time and improves the accuracy. TCPSD process improves products’ purities to 99.9 wt% and the economics of TCPSD were compared with different separation configurations. The results demonstrated that the optimization of variables denotes that the A-M-B separation configuration was the most optimum column sequence to separate acetonitrile/methanol/benzene ternary system using TCPSD. These studies reveal that TCPSD process is worth considering for the conceptual design stage of the process for separating complex ternary azeotropic system.
Notes
The authors declare no competing financial interest.
74
Z. Zhu et al. / Separation and Purification Technology 169 (2016) 66–77
Fig. 8. RCMs for M-A-B separation configuration: (a) 91.19 kPa; (b) 506.63 kPa; (c) Summary of the distillation boundaries positions at 91.19–303.98–506.63 kPa.
75
Z. Zhu et al. / Separation and Purification Technology 169 (2016) 66–77 Table 4
Case studies of M-A-B separation figuration.
Variables (kPa)P1 (kPa) P2 (kPa) P3 NT1 NT2 NT3 NF1/NREC NF2 NF3 RR1 RR2 RR3 (m)ID1
Case1
Case2
81.06 303.98 506.63 50 55 14 12/34 39 4 2 0.84 2.90 0.19 0.83
Case3
91.19 303.98 506.63 54 53 13
101.33 303.98 506.63 52 51 13
14/34 41 2 1.01 2.39 0.09 0.86
(m)ID2 0.95 (m)ID3 0.53 REC(kg/h) 2515 Total Capital investment ($) 2.333 Annual Operating cost ($/y) 8.001 TAC($/y) 1.267
0.84 0.51 106 105 6 10
56 54 16 21/35
38 3
42 4 1.65 1.81 0.11 0.96
Case4
Case5
91.19 202.65 506.63 54 47 25 20/33 27 3 2 2.08 2.80 0.48 1.24
91.19 405.30 506.63 53 54 24 14/34 41
0.83 1.15 0.54 0.74 2522 2749 3727 2.224 106 2.301 106 7.597 105 8.243 105 1.204 106 1.284 106
Case6
Case7
91.19 303.98 405.30 54 48 12 12/33 37
91.19 303.98 607.95
13/33
0.50 1.80 1.60 0.79
1.00 2.37 1.29 0.83
1.01 2.29 0.05 0.84
0.79 0.80
0.87 0.75
0.89 0.51
4167 3.321 106 1.445 106 2.110 106
2518 2.495 106 9.615 105 1.460 106
2561 2.550 106 8.473 105 1.357 106
2.222 106 7.616 105 1.206 106
Fig. 9. Flowsheet of the optimal TCPSD process for M-A-B separation configuration.
Fig. 10. RCMs for M-B-A separation configuration: (a) 405.30 kPa; (b) Summary of the distillation boundaries positions at 101.33–709.28–405.30 kPa.
Z. Zhu et al. / Separation and Purification Technology 169 (2016) 66–77
76 Table 5
Case studies of M-B-A separation figuration.
Variables (kPa) P1 (kPa) P2 (kPa) P3 NT1 NT2 NT3 NF1/NREC NF2 NF3 RR1 RR2 RR3 (m)ID1
Case1
Case2
101.33 607.95 405.30 52 20 43 23/39 4 5 28 1.40 1.21 4.60 0.86
Case3
101.33 709.28 405.30 54 16 41
51 14 41 19/36 4
26 1.10 0.60 4.40 0.75
(m)ID2 0.63 (m)ID3 0.91 REC(kg/h) 1954 Total Capital investment ($) 2.514 Annual Operating cost ($/y) 9.414 TAC($/y) 1.444
0.58 0.89 106 105 6 10
101.33 810.6 405.30
4 28
55 15 37 23/38 3 26 1.90 0.50 4.70 0.86
Case4
Case5
101.33 709.28 303.98 44 26 33 20/37 5 5 17 1.54 0.49 3.15 1.02
101.33 709.28 506.63 53 16 43 20/27
0.56 0.66 0.91 1.10 1996 1943 2800 2.294 106 2.269 106 8.012 105 9.088 105 1.260 106 1.363 106
28
Case6
Case7
91.19 709.28 405.30 56 16 41 19/38
111.46 709.28 405.30
18/36
1.70 2.32 1.02 1.34
27 1.12 0.59 5.41 0.74
1.01 0.56 4.26 0.76
1.35 0.92
0.56 0.92
0.61 0.93
5438 2.844 106 1.208 106 1.777 106
1811 3.645 106 1.969 106 2.698 106
2165 2.322 106 8.015 105 1.266 106
2.391 106 8.673 105 1.346 106
Fig. 11. Flowsheet of the optimal TCPSD process for M-B-A separation configuration.
[1] G.P. Cunningham, G.A. Vidulich, R.L. Kay, Several properties of acetonitrilewater, acetonitrile-methanol, and ethylene carbonate-water systems, J. Chem. Eng. Data 12 (1967) 336–337. [2] W. EarleáWaghorne, Enthalpies of transfer of some non-electr olytes from acetonitrile to acetonitrile–methanol mixtures, J. Chem. Soc., Faraday Trans. 87
[8] K. Huang, L. Shan, Q. Zhu, J. Qian, Design and control of an ideal heat-integrated distillation column (ideal HIDiC) system separating a close-boiling ternary mixture, Energy 32 (2007) 2148–2156. [9] L. Zong, Study on the Separation Process of Methanol+Acetonitrile+Benzene System Thesis, HeBei University of Technology, 2013. [10] Z. Zhu, X. Li, Y. Cao, X. Liu, Y. Wang, Design and control of a middle vessel batch distillation process for separating the methyl formate/methanol/water ternary system, Ind. Eng. Chem. Res. 55 (2016) 2760–2768. [11] G. Modla, Separation of a chloroform–acetone–toluene mixture by pressureswing batch distillation in different column configurations, Ind. Eng. Chem. Res. 50 (2011) 8204–8215. [12] J.R. Phimister, W.D. Seider, Semicontinuous, middle-vessel distillation of ternary mixtures, AIChE J. 46 (2000) 1508–1520. [13] G.U.B. Babu, E.K. Pal, A.K. Jana, An adaptive vapor recompression scheme for a
(1991) 2443–2446. [3] S. Mandal, V.G. Pangarkar, Separation of methanol–benzene and methanol– toluene mixtures by pervaporation: effects of thermodynamics and structural phenomenon, J. Membr. Sci. 201 (2002) 175–190. [4] V.N. Kiva, L.J. Krolikowski, Feasibility of separation for distillation of azeotropic ternary mixtures: a survey and analysis, Chem. Eng. Res. Des. 95 (2015) 195– 210. [5] L. Hegely, P. Lang, A new algorithm for the determination of product sequences in azeotropic batch distillation, Ind. Eng. Chem. Res. 50 (2011) 12757–12766 . [6] G. Modla, P. Lang, F. Denes, Feasibility of separation of ternary mixtures by pressure swing batch distillation, Chem. Eng. Sci. 65 (2010) 870–881 . [7] S. Xia, W. Wei, G. Liu, X. Dong, W. Jin, Pervaporation properties of polyvinyl alcohol/ceramic composite membrane for separation of ethyl acetate/ethanol/ water ternary mixtures, Korean J. Chem. Eng. 29 (2012) 228–234.
ternary batch distillation with a side withdrawal, Ind. Eng. Chem. Res. 51 (2012) 4990–4997. [14] J. Lee, J. Cho, D.M. Kim, S. Park, Separation of tetrahydrofuran and water using pressure swing distillation: modeling and optimization, Korean J. Chem. Eng. 28 (2011) 591–596. [15] W.L. Luyben, Pressure-swing distillation for minimum- and maximumboiling homogeneous azeotrop es, Ind. Eng. Chem. Res. 51 (2012) 10881– 10886. [16] B. Yu, Q. Wang, C. Xu, Design and control of distillation system for methylal/ methanol separation. Part 2: Pressure swing distillation with full heat integration, Ind. Eng. Chem. Res. 51 (2012) 1293–1310. [17] Y. Wang, Z. Zhang, H. Zhang, Q. Zhang, Control of heat integrated pressureswing-distillation process for separating azeotropic mixture of tetrahydrofuran and methanol, Ind. Eng. Chem. Res. 54 (2015) 1646–1655.
Acknowledgement
Financial support from the National Natural Science Foundation of China (Project 21306093) is gratefully acknowledged. References
Z. Zhu et al. / Separation and Purification Technology 169 (2016) 66–77
77
[18] J.F. Mulia-Soto, A. Flores-Tlacuahuac, Modeling, simulation and control of an internally heat integrated pressure-swing distillation process for bioethanol separation, Comput. Chem. Eng. 35 (2011) 1532–1546. [19] D. Fissore, M. Pin, A.A. Barresi, On the use of detailed models in the MPC algorithm: the pressure-swing distillation case, AIChE J. 52 (2006) 3491–3500 . [20] J.U. Repke, F. Forner, A. Klein, Separation of homogeneous azeotropic mixtures by pressure swing distillation–analysis of the operation performance, Chem. Eng. Technol. 28 (2005) 1151–1157. [21] B. Kiran, A.K. Jana, A hybrid heat integration scheme for bioethanol separation through pressure-swing distillation route, Sep. Purif. Technol. 142 (2015) 307– 315. [22] A.M. Fulgueras, J. Poudel, D.S. Kim, J. Cho, Optimization study of pressureswing distillation for the separation process of a maximum-boiling azeotropic system of water-ethylenediamine, Korean J. Chem. Eng. 33 (2016) 46–56. [23] J. Qin, Q. Ye, X. Xiong, N. Li, Control of benzene-cyclohexane separation system via extractive distillation using sulfolane as entrainer, Ind. Eng. Chem. Res. 52 (2013) 10754–10766. [24] W. Shen, L. Dong, S.a. Wei, J. Li, H. Benyounes , X. You, V. Gerbaud, Systema tic
[37] E. Lladosa, J.B. Montón, M. Burguet, Separation of di-n-propyl ether and npropyl alcohol by extractive distillation and pressure-swing distillation: computer simulation and economic optimization, Chem. Eng. Process. 50 (2011) 1266–1274. [38] J.P. Knapp, M.F. Doherty, A new pressure-swing-distillation process for separating homogeneous azeotropic mixtures, Ind. Eng. Chem. Res. 31 (1992) 346–357. [39] 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. [40] W.L. Luyben, Methanol/trimethoxysilane azeotrope separation using pressureswing distillation, Ind. Eng. Chem. Res. 53 (2014) 5590–5597. [41] W.L. Luyben, Control of a heat-integrated pressure-swing distillation process for the separation of a maximum-boiling azeotrope, Ind. Eng. Chem. Res. 53 (2014) 18042–18053. [42] D.Y.-C. Thong, M. Jobson, Multicomponent homogeneous azeotropic distillation 3. Column sequence synthesis, Chem. Eng. Sci. 56 (2001) 4417– 4432.
design of an extractive tion for maximum-boiling azeotrope s with heavy entrainers, AIChE J.distilla 61 (2015) 3898–3910. [25] K. Benyahia, H. Benyounes, W. Shen, Energy evaluation of ethanol dehydration with glycol mixture as entrainer, Chem. Eng. Technol. 37 (2014) 987–994 . [26] Z. Fan, X. Zhang, W. Cai, F. Wang, Design and control of extraction distillation for dehydration of tetrahydrofuran, Chem. Eng. Technol. 36 (2013) 829–839 . [27] P. Langston, N. Hilal, S. Shingfield, S. Webb, Simulation and optimisation of extractive distillation with water as solvent, Chem. Eng. Process. 44 (2005) 345–351. [28] A.A. Kiss, D.J.P.C. Suszwalak, Enhanced bioethanol dehydration by extractive and azeotropic distillation in dividing-wall columns, Sep. Purif. Technol. 86 (2012) 70–78. [29] W.L. Luyben, Design and control of a fully heat-in tegrated pressure-s wing azeotropic distillation system, Ind. Eng. Chem. Res. 47 (2008) 2681–2695 . [30] S.-J. Wang, D.S. Wong, Online switching of entrainers for acetic acid dehydration by heterogeneous azeotropic distillat ion, J. Process Contr. 23 (2013) 78–88. [31] O.M. Wahnschafft, J. Koehler, E. Blass, A.W. Westerberg, The product composition regions of single-feed azeotropic distillation columns, Ind. Eng. Chem. Res. 31 (1992) 2345–2362. [32] A. Tabari, A. Ahmad, A semicontinuous approach for heterogeneous azeotropic distillation processes, Comput. Chem. Eng. 73 (2015) 183–190. [33] Y. Lin, J. Chen, J. Cheng, H. Huang, C. Yu, Process alternatives for methyl acetate conversion using reactive distillation. 1. Hydrolysis, Chem. Eng. Sci. 63 (2008) 1668–1682. [34] A.K. Mathew, N. Kaistha, M.V.P. Kumar, Control of quaternary ideal endothermic reactive distillation with and without internal heat integration,
[43] L. Laroche, N.separability Bekiaris, H.W. M.synthesis, Morari, Homogeneous azeotropic distillation: andAndersen, flowsheet Ind. Eng. Chem. Res. 31 (1992) 2190–2209. [44] L.J. Krolikowski, Determination of distillation regions for non-ideal ternary mixtures, AIChE J. 52 (2006) 532–544. [45] D.K.M. Seiler, W. Arlt, Hyperbranched polymers: new selective solvents for extractive distillation and solvent extraction, Sep. Purif. Technol. 30 (2003) 179–197. [46] R.B.P.A.M. Springer, R. Krishna, Influence of interphase mass transfer on the composition trajectories and crossing of boundaries in ternary azeotropic distillation, Sep. Purif. Technol. 29 (2002) 1–13 . [47] G.A. Fien, Y. Liu, Heuristic synthesis and shortcut design of separation processes using residue curve maps: a review, Ind. Eng. Chem. Res. 33 (1994) 2505–2522. [48] T. Ohta, I. Nagata, Vapor-liquid equilibriums for the ternary systems acetonitrile-2-butanone-benzene and acetonitrile-methanol-benzene at 328.15 K, J. Chem. Eng. Data 28 (1983) 398–402 . [49] W.L. Luyben, Distillation Design and Control Using Aspen Simulation, John Wiley & Sons, 2013 . [50] J.M. Douglas, Conceptual Design of Chemical Processes, vol. 1110, McGrawHill, New York, 1988 . [51] 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. [52] W.L. Luyben, Comparison of extractive distillation and pressure-swing distillation for acetone/chloroform separation, Comput. Chem. Eng. 50 (2013) 1–7.
Chem. Eng. Technol. 39 A (2016) [35] Y. Tavan, S.H. Hosseini, novel775–785. integrated process to break the ethanol/water azeotrope using reactive distillat ion – Part I: Parametric study, Sep. Purif. Technol. 118 (2013) 455–462. [36] R. Muñoz, J.B. Montón, M.C. Burguet, J. de la Torre, Separa tion of isobutyl alcohol and isobutyl acetate by extractive distillation and pressure-swing distillation: simulation and optimization, Sep. Purif. Technol. 50 (2006) 175– 183.
[53] H. Luo, K. Liang, Li, Y. Li, M.distillation Xia, C. Xu, Comparison of pressure-swing distillation andW.extractive methods for isopropyl alcohol/ diisopropyl ether separation, Ind. Eng. Chem. Res. 53 (2014) 15167–15182. [54] H. Wei, F. Wang, J. Zhang, B. Liao, N. Zhao, F. Xiao, W. Wei, Y. Sun, Design and control of dimethyl carbonate–methanol separation via pressure-swing distillation, Ind. Eng. Chem. Res. 52 (2013) 11463–11478. [55] Qingdao University of Science & Technology. Pressure-Swing Distillation Optimization Software: porpoise V1.0. NO. 2015SR216601.
