Professional Documents
Culture Documents
com
Abstract
Ethyl acrylate (EA) is widely used in industry as a precursor for varnishes, adhesive, and nishes of papers and textiles. This important ester can be produced directly from ethanol (EtOH) and acrylic acid (AA) via esterication reaction with the presence of sulphuric
acid as homogeneous catalyst. The proposed design owsheet of this process includes a CSTR reactor coupled with a rectier and an
overhead decanter. In order to further purify the nal EA product, another stripper is needed with its top vapor recycled back to decanter. The simplest and industrial easily applicable overall control strategy will be investigated with only one tray temperature control loop
in each of the two columns. The nal proposed overall control strategy of this process is found to be dierent than another similar coupled reactor/columns process published earlier [I-L. Chien, Y.P. Teng, H.P. Huang, Y.T. Tang, Design and control of an ethyl acetate
process: coupled reactor/column conguration, J. Proc. Cont. 15 (2005) 435449]. Both EtOH and AA feed ow rates are used as manipulated variables in the overall control strategy with CSTR heat duty left as throughput manipulator for the overall process. The nal EA
product with stringent specications of 0.1 wt% EtOH and 0.005 wt% AA impurities can be achieved with this proposed overall control
strategy despite feed ow rate and feed composition disturbances.
2007 Elsevier Ltd. All rights reserved.
Keywords: Ethyl acrylate; Esterication reaction; Reactive distillation; Coupled reactor/columns; Optimum design; Overall control strategy
1. Introduction
Ethyl acrylate (EA) is widely used in industry as a precursor for varnishes, adhesive, and nishes of papers and
textiles. This important ester can be produced directly from
ethanol (EtOH) and acrylic acid (AA) via esterication
reaction with the presence of sulphuric acid as homogeneous catalyst. Only in a very recent paper [2], the kinetics
of this esterication reaction has been given. There is no
paper in the literature on the subject of the production of
ethyl acrylate, thus the results of this paper should be useful to other researchers.
In this study, overall control strategy of this process
with coupled reactor/columns conguration will be stud-
0959-1524/$ - see front matter 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jprocont.2007.02.006
ied. The principal behind the coupled reactor/columns conguration, similar to reactive distillation, is that the
continuous removal of products from the esterication
reaction mixture by distillation reduces the backward reaction rate. The advantage of the coupled reactor/columns
conguration over reactive distillation according to Yi
and Luyben [3] include: the existing reactor/columns in
the plant can be retrotted for this usage; easy maintenance
of the overall system; larger reactor holdup and dierent
reaction temperature can easily be designed; etc.
In a three-paper series by Yi and Luyben [35], they
studied the design and control of various coupled reactor/column systems. The studied systems include: a binary
reactor/rectier, a binary reactor/stripper, a multicomponent reactor/rectier, a multicomponent reactor/rectier/
stripper, and a more complex process that consists of a
coupled reactor/stripper, two distillation columns and
one recycle stream. Their studied systems are very simple,
216
ideal chemical systems and also no liquidliquid equilibrium is considered. Chiang et al. [6] studied a coupled reactor/column system for the production of amyl acetate.
Their process is much simpler than the studied ethyl acrylate process because amyl acetate has the highest boiling
point in the system. Their system with the conguration
of reactor with rectier on top and stripper on the bottom
produces amyl acetate from the bottom of the stripper and
also produces water through aqueous phase of a decanter.
Chien et al. [1] proposed a coupled reactor/column conguration for the production of ethyl acetate. Their process
owsheet conguration is very similar to the ones in this
paper. However, the overall control strategy is dierent
than the one will be developed for this system.
The organization of this paper is as follows. The thermodynamic properties of this four-component system and
the kinetics of this esterication reaction will be given in
Section 2. The design owsheet of a complete coupled reactor/column system will be proposed in Section 3. A nal
EA product purity of over 99.5 wt% will be obtained in
the proposed design with stringent specications of
0.1 wt% EtOH and 0.005 wt% AA impurities. In Section
4, the overall control strategy of this process will be investigated. Only one tray temperature control loop in each of
the two columns (rectier and stripper) will be used. Feed
ow rate and feed composition disturbances will be used
to test the overall control strategy. Some concluding
remarks will be drawn in the nal section.
EtOH
$ CH2 CHCOOC2 H5 H2 O
EA
C 2EA C 2H2 O
K2
!
mol=dm3 min
where
15900
k1 3:26 10 exp
RT
6490
4
K 2:71 10 exp
RT
6
and R is gas constant (1.987 cal/mol/K) and Ccat is assumed to have value of 0.15 mol/dm3. In the kinetic equation, all concentrations are with unit of (mol/dm3) and
temperature in K.
3. Design owsheet of the complete process
The RCM for the EtOHEAH2O three-component
system and AAEAH2O three-component system can be
seen in Figs. 1 and 2. From these two gures, the highest
boiling point temperature of the whole system including
the pure components and azeotropes is the acrylic acid
(AA) at 141.19 C and the lowest temperature of the whole
system is the EtOH + EA + H2O three-component azeotrope at 75.29 C. The two products (ethyl acrylate and
water) of this esterication reaction are neither the lightest
nor the heaviest component in the system, thus the complete designed process will need to be more complex in
comparison with the other reactive distillation papers in
the literature.
The proposed design of this process including a CSTR
reactor coupled with a rectier (without heat source). The
heat input in the CSTR totally vaporizes the reactor outlet
stream to vapor phase and continuously enter the rectier
from the bottoms to promote the forward reaction further.
The bottom liquid stream from the rectier containing
mostly heavy boiler AA is recycled back to the CSTR.
The composition of the top vapor stream from the rectier
is close to the lightest boiler of ternary azeotrope of EtOH + EA + H2O. This stream after sub-cooling to 40 C
can be naturally separated inside a decanter to form
organic and aqueous phases. Extra water is added in the
decanter to maintain suitable composition inside of the
liquidliquid boundary. The water purity of the aqueous
phase is quite high, thus it can go to a waste water treatment plant for discharge. The organic phase composition
by natural liquidliquid separation has the benet of crossing the distillation boundary into a desirable region to
obtain pure ethyl acylate product (see Fig. 1). This organic
217
EtOH
(78.31 C)
0.9
(78.20 C)
0.4
0.5
0.6
0.7
0.8
(77.87 C)
0.1
0.2
0.3
(75.29 C)
EA
WATER
(99.40 C)
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
(100.02 C)
(81.04 C)
AA
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
(141.19 C)
EA
(99.40 C)
WATER
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
(100.02 C)
(81.04 C)
tion into the nal EA product. The top vapor of this stripper with composition near the top vapor of the rectier is
218
Decanter
Aqueous
Organic
Reflux
AA
EtOH
Steam
CSTR
Reboiler
Steam
EA
5.60e+5
5.58e+5
5.55e+5
5.56e+5
5.50e+5
Profit ($/year)
Profit ($/year)
also condensed and then fed into the decanter. The bottom
stream of the stripper is the nal EA product with stringent
specications of 0.1 wt% EtOH and 0.005 wt% AA impurities in this product stream. This conceptual design of the
overall process owsheet can be seen in Fig. 3.
The design owsheet is selected based on the maximization of Total Annual Prot (TAP) for the overall system.
This TAP includes: the product value minus the costs of
two feed streams, minus annualized capital costs, minus
total utility costs, and minus the waste water treatment
5.45e+5
5.40e+5
5.35e+5
5.30e+5
20
40
60
80
100
5.52e+5
5.50e+5
Water flow rate=50 mole/min
Column 1=24 stages
Column 2=7 stages
5.48e+5
5.46e+5
125
5.25e+5
0
5.54e+5
120
126
127
129
130
131
132
133
5.568e+5
5.568e+5
5.567e+5
Profit ($/year)
5.567e+5
Profit ($/year)
128
5.566e+5
5.565e+5
5.564e+5
Water flow rate=50 mole/min
CSTR volume=126 cum
Column 2=7 stages
5.563e+5
5.566e+5
5.565e+5
5.564e+5
Water flow rate=50 mole/min
CSTR volume=126 cum
Column 1=24 stages
5.563e+5
5.562e+5
5.561e+5
5.562e+5
22
23
24
Tray of Column 1
25
26
Tray of Column 2
6.0e+5
Profit ($/year)
5.9e+5
5.8e+5
5.7e+5
5.6e+5
5.5e+5
94
96
98
100
102
104
106
108
110
219
Table 1
Optimum process design for this system
CSTR holdup (m3)
64
20
7
102
82.2 mol% EtOH
17.8 mol% H2O
66
100 mol% AA
258
90
151.8
100.0
84.05
145.55
70.10
204.1
63.9
99.78 mol% (99.89 wt%) EA
0.0069 mol% (0.005 wt%) AA
0.213 mol% (0.10 wt%) EtOH
1 104 mol% H2O
220
Ratio to a
throughput
manipulator
TC
TC
Decanter
LC
PC
Organic
Reflux
LC
Aqueous
PC
1
FC
X
FC
AA
FC
19
EtOH
6
LC
Steam
7
20
CSTR
Reboiler
Steam
EA
0.5
0.0
-0.5
-1.0
-1.5
1
9 10 11 12 13 14 15 16 17 18 19 20
0.3
221
0.2
0.1
0.0
-0.1
-0.2
1
Fig. 7. Closed-loop sensitivity plot with organic reux ratio and AA feed
xed.
222
0.4
AA-5%
EtOH(H2O)+20%
EtOH(H2O)-20%
throughput+10%
throughput-10%
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
0.10
1.0
0.5
0.05
0.00
-0.05
-0.10
0.0
-0.15
-0.5
-0.20
1
Fig. 9. Closed-loop sensitivity plot with organic reux ratio and CSTR
duty xed.
-1.5
1
9 10 11 12 13 14 15 16 17 18 19 20
AA-5%
EtOH(H2O)+20%
EtOH(H2O)-20%
throughput+10%
throughput-10%
0.3
0.2
0.1
0.0
-0.1
-0.2
-0.3
-0.4
1
Fig. 8. Closed-loop sensitivity plot with organic reux ratio and EtOH
feed xed.
1.0
1.0
0.5
0.5
0.0
-0.5
-1.0
0.0
-0.5
-1.0
-1.5
-1.5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8
9 10 11 12 13 14 15 16 17 18 19 20
AA-5%
EtOH(H2O)+20%
EtOH(H2O)-20%
throughput+10%
throughput-10%
0.2
0.0
-0.2
-0.4
AA-5%
EtOH(H2O)+20%
EtOH(H2O)-20%
throughput+10%
throughput-10%
0.4
0.4
223
0.2
0.0
-0.2
-0.4
-0.6
-0.6
1
Fig. 10. Closed-loop sensitivity plot with CSTR duty and AA feed xed.
Fig. 11. Closed-loop sensitivity plot with CSTR duty and EtOH feed
xed.
224
92
95.4
95.2
95.0
st
st
94
95.6
95.6
T of 1 column (C)
5
T of 1 column (C)
5
96
st
T of 1 column (C)
5
95.8
98
94.8
94.6
AA feed +1%
AA feed -1%
94.4
10
15
20
25
30
95.2
95.0
94.8
Q1 +1%
Q1 -1%
94.6
94.2
0
95.4
94.4
0
10
Time (hr)
15
20
25
30
10
Time (hr)
15
20
25
30
20
25
30
Time (hr)
4.372
4.370
4.368
4.366
4.364
4.362
4.360
4.370
4.368
4.370
4.366
4.364
4.362
4.365
4.360
4.360
ETOH feed +1%
ETOH feed -1%
4.358
AA feed +1%
AA feed -1%
4.358
Q1 +1%
Q1 -1%
4.355
4.356
0
10
15
20
25
30
10
Time (hr)
15
20
25
30
10
Time (hr)
15
Time (hr)
Fig. 12. Open-loop responses with step changes in either of EtOH feed, AA feed, or CSTR duty.
225
120
Temperature ( oC)
110
95.5
95.0
st
T5 of 1 Column (C)
96.0
94.5
90
AA Feed +1%
AA Feed -1%
94.0
100
Base case
EtOH feed +0.01%
EtOH feed -0.01%
93.5
0
10
15
20
25
30
80
Time (hr)
1 2
9 10 11 12 13 14 15 16 17 18 19 20
105
Base case
Q2 +1%
Q2 -1%
98
97
Temperature ( oC)
96
st
T5 of 1 Column (C)
100
95
94
95
90
85
93
80
92
0
10
15
Time (hr)
20
25
30
75
1
Fig. 14. Open-loop sensitivity plot for the rectifying and stripping
columns while CSTR level is controlled by AA feed.
226
6.0e-5
5.5e-5
5.20e-5
5.0e-5
4.5e-5
4.0e-5
3.5e-5
3.0e-5
2.5e-5
50
100
150
200
250
1.015e-3
1.010e-3
1.007e-3
1.005e-3
1.000e-3
9.950e-4
9.900e-4
300
50
100
95
94
93
92
250
300
150
200
Time (hr)
250
300
200
250
300
200
250
300
102
100
98
96
94
50
100
150
200
250
300
50
100
50
100
Time (hr)
99.645
T5 of 2nd Column ( o C)
200
104
T5 of 1st Column (o C)
96
99.640
99.635
99.630
99.625
99.620
99.615
50
100
150
200
250
72
70
68
66
64
62
60
300
4.365
4.364
4.363
4.362
4.361
4.360
50
100
150
150
Time (hr)
Time (hr)
150
Time (hr)
Time (hr)
200
250
300
Time (hr)
154
152
150
148
146
144
142
140
50
100
150
Time (hr)
AA Concentration-5%
Fig. 15. Closed-loop performance for Control Strategy (I) with AA feed concentration changes from 100 mol%AA to 95 mol%AA at t = 2 h.
9e-5
8e-5
7e-5
6e-5
5.03e-5
5e-5
4.90e-5
4e-5
3e-5
2e-5
50
100
150
200
250
1.002e-3
1.000e-3
9.974e-4
9.950e-4
9.900e-4
9.850e-4
300
50
100
99
200
250
300
200
250
300
200
250
300
200
250
300
110
98
97
96
95
94
93
92
91
108
106
104
102
100
98
96
94
50
100
150
200
250
300
50
100
99.65
99.64
99.63
99.62
99.61
99.60
99.59
50
100
150
200
250
74
73
72
71
70
69
68
67
300
50
100
4.370
4.368
4.366
4.364
4.362
4.360
50
100
150
150
Time (hr)
Time (hr)
150
Time (hr)
Time (hr)
150
Time (hr)
Time (hr)
227
200
250
300
Time (hr)
158
156
154
152
150
148
146
50
100
150
Time (hr)
EtOH Feed H2O Concentration +20%
EtOH Feed H2O Concentration -20%
Fig. 16. Closed-loop performance for Control Strategy (I) with EtOH feed concentration 20% changes at t = 2 h.
228
5.04e-5
5.02e-5
5.02e-5
5.00e-5
4.98e-5
4.96e-5
4.94e-5
4.92e-5
4.90e-5
4.88e-5
4.86e-5
4.84e-5
50
100
150
200
250
1.004e-3
1.003e-3
1.003e-3
1.002e-3
1.001e-3
1.000e-3
9.990e-4
9.980e-4
300
50
100
150
200
250
300
200
250
300
200
250
300
200
250
300
Time (hr)
Time (hr)
95.10
95.05
95.00
94.95
94.90
102.0
101.8
101.6
101.4
101.2
101.0
100.8
100.6
100.4
94.85
50
100
150
200
250
300
50
100
99.65
99.64
99.63
99.62
99.61
99.60
99.59
50
100
150
200
250
70.5
70.0
69.5
69.0
68.5
68.0
67.5
67.0
300
50
100
69.5
4.36435
4.36430
150
Time (hr)
Time (hr)
4.36425
4.36420
4.36415
4.36410
4.36405
0
150
Time (hr)
Time (hr)
69.0
68.5
68.0
67.5
67.0
66.5
66.0
65.5
50
100
150
200
250
300
Time (hr)
50
100
150
Time (hr)
AA Concentration-5%
Fig. 17. Closed-loop performance for Control Strategy (II) with AA feed concentration changes from 100 mol%AA to 95 mol%AA at t = 2 h.
problem. For example, with 10% changes in the throughput manipulator (CSTR heat duty), the setpoint of tray #5
temperature at rectier should decrease around 0.3 C
according to Fig. 9 so that AA impurity in the nal product
can return back to 0.005 wt%.
With above 10% changes of the CSTR heat duty, the
product rate is changed from 63.8824 mol/min to
69.66846 mol/min (a +9.1% increase in production) or
from 63.8824 mol/min to 58.0364 mol/min (a 9.1%
6.5e-5
6.0e-5
5.5e-5
4.98e-5
5.0e-5
4.95e-5
4.5e-5
4.0e-5
3.5e-5
0
50
100
150
200
250
1.015e-3
1.010e-3
1.00e-3
1.005e-3
1.000e-3
9.950e-4
9.96e-4
9.900e-4
9.850e-4
0
300
50
100
96.5
200
250
300
200
250
300
200
250
300
200
250
300
112
97.0
96.0
95.5
95.0
94.5
94.0
110
108
106
104
102
100
98
96
94
92
93.5
0
50
100
150
200
250
300
50
100
150
Time (hr)
Reboiler Duty of 2 Column (KW)
Time (hr)
99.640
99.635
99.630
99.625
73
72
71
nd
o
nd
T5 of 2 Column ( C)
150
Time (hr)
Time (hr)
99.620
99.615
99.610
99.605
99.600
0
50
100
150
200
250
70
69
68
67
0
300
50
100
150
Time (hr)
Time (hr)
69
4.3645
4.3644
229
4.3643
4.3642
4.3641
68
67
66
65
64
63
4.3640
0
50
100
150
200
250
300
Time (hr)
50
100
150
Time (hr)
EtOH Feed H2O Concentration +20%
EtOH Feed H2O Concentration -20%
Fig. 18. Closed-loop performance for Control Strategy (II) with EtOH feed concentration 20% changes at t = 2 h.
tion has been investigated. Unlike other paper in the literature with only one reactive distillation column to
produce products like methyl acetate or butyl acetate, this
overall process owsheet is more complex including a
CSTR, a rectifying column, a decanter, and another stripping column. The design procedure is based on the maximization of total annual prot (TAP) for the overall
system. The nal EA product is having very high purity
of 99.78 mol% (or 99.89 wt%) with two impurities of
AA and EtOH meeting stringent specications of
230
6.5e-5
6.0e-5
5.5e-5
5.24e-5
5.0e-5
4.77e-5
4.5e-5
4.0e-5
3.5e-5
0
50
100
150
200
250
1.08e-3
1.06e-3
1.04e-3
1.01e-3
1.02e-3
1.00e-3
9.80e-4
9.89e-4
9.60e-4
9.40e-4
9.20e-4
0
300
50
100
97
96
95
94
93
250
300
200
250
300
200
250
300
200
250
300
115
110
105
100
95
90
85
50
100
150
200
250
300
50
100
99.75
99.70
99.65
99.60
99.55
99.50
0
50
100
150
150
Time (hr)
Time (hr)
200
120
98
200
250
80
78
76
74
72
70
68
66
64
62
60
0
300
50
100
150
Time (hr)
Time (hr)
76
4.368
150
Time (hr)
Time (hr)
4.366
4.364
4.362
74
72
70
68
66
64
62
60
58
56
50
100
150
200
250
300
Time (hr)
50
100
150
Time (hr)
Throughput +10%
Throughput-10%
Fig. 19. Closed-loop performance for Control Strategy (II) with 10% changes in CSTR duty at t = 2 h.
231
Water
FC
TC
TC
X
Decanter
LC
PC
LC
Organic
Reflux
Aqueous
PC
1
FC
TC
X
FC
AA
TC
FC
19
EtOH
LC
Steam
7
20
LC
CSTR
Reboiler
EA
Acknowledgements
The rst author, I-Lung Chien, take this opportunity to
thank Prof. Dale Seborg for his guidance and valuable advices over the years. This paper was prepared while I-Lung
was back visiting Department of Chemical Engineering
UCSB after completing his Ph.D. degree as Dales student
more than 20 years ago. There is an old Chinese saying
which conveys I-Lungs appreciation to Prof. Seborg:
Once becoming a teacher of yours, always respect him
as your father. At this occasion of Dales 65th birthday,
I-Lung wish him happiness always and longevity.
References
[1] I-L. Chien, Y.P. Teng, H.P. Huang, Y.T. Tang, Design and control of
an ethyl acetate process: coupled reactor/column conguration, J.
Proc. Cont. 15 (2005) 435449.
[2] M. Witczak, W. Grzesik, J. Skrzypek, Kinetyka estrykacji kwasu
akrylowego nizszymi alkoholami alifatycznymi, Inz. Chem. Procesowa 25 (2004) 331340.
[3] C.K. Yi, W.L. Luyben, Design and control of coupled reactor/
column systems Part 1. A binary coupled reactor/rectier system,
Comput. Chem. Eng. 21 (1997) 2546.
[4] C.K. Yi, W.L. Luyben, Design and control of coupled reactor/
column systems Part 2. More complex coupled reactor/column
system, Comput. Chem. Eng. 21 (1997) 4767.
[5] C.K. Yi, W.L. Luyben, Design and control of coupled reactor/
column systems Part 3. A reactor/stripper with two columns and
recycle, Comput. Chem. Eng. 21 (1997) 6986.
[6] S.F. Chiang, C.L. Kuo, C.C. Yu, D.S.H. Wong, Design alternatives
for the amyl acetate process: coupled reactor/column and reactive
distillation, Ind. Eng. Chem. Res 41 (2002) 32333246.
[7] U. Weidlich, J. Gmehling, A modied UNIFAC model. 1. Prediction
of VLE, HE and c1, Ind. Eng. Chem. Res. 26 (1987) 13721381.
[8] J. Gmehling, J. Li, M. Schiller, A modied UNIFAC model. 2.
Present parameter matrix and results for dierent thermodynamic
properties, Ind. Eng. Chem. Res. 32 (1993) 178193.
[9] J.G. Hayden, J.P. OConnell, A generalized method for predicting
second virial coecients, Ind. Eng. Chem. Proc. Des. Dev. 14 (1975)
209216.
[10] J.M. Douglas, Conceptual Process Design, McGraw-Hill, New York,
1988.
[11] R. Turton, R.C. Bailie, W.B. Whiting, J.A. Shaeiwitz, Analysis,
Synthesis, and Design of Chemical Processes, Prentice-Hall, Englewood Clis, NJ, 1998.
[12] H.Y. Lee, H.P. Huang, I-L. Chien, Control of reactive distillation
process for production of ethyl acetate, J. Proc. Cont. 17 (2007) 363
377.
[13] S.G. Huang, C.L. Kuo, S.B. Huang, Y.W. Chen, C.C. Yu, Temperature control of heterogeneous reactive distillation, AIChE J. 50
(2004) 22032216.
[14] D.B. Kaymak, W.L. Luyben, Comparison of two types of twotemperature control structures for reactive distillation columns, Ind.
Eng. Chem. Res. 44 (2005) 45254640.
[15] I-L. Chien, P.S. Fruehauf, Consider IMC tuning to improve
controller performance, Chem. Eng. Prog. 86 (1990) 3341.