Fluid Phase Equilibria 432 (2017) 70e75

Contents lists available at ScienceDirect

Fluid Phase Equilibria

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / fl u i d

Liquid - liquid equilibrium for the quaternary reaction system

water þ sec-butyl alcohol þ sec-butyl acetate þ acetic acid
Ling Li, Ting Zeng, Xiaoda Wang, Changshen Ye, Ting Qiu, Zhixian Huang*
School of Chemical Engineering, Fuzhou University, Fuzhou, 350116, Fujian, China

a r t i c l e i n f o a b s t r a c t

Article history: There is a reversible reaction (sec-butyl acetate þ water ⇔ sec-butyl alcohol þ acetic acid) catalyzed by
Received 20 July 2016 acetic acid in the quaternary system water þ sec-butyl alcohol þ sec-butyl acetate þ acetic acid.
Received in revised form In this work, the liquid-liquid equilibrium (LLE) data for this quaternary system were determined at
20 October 2016
303.15e333.15 K at atmosphere pressure. The LLE model of this quaternary reaction system was
Accepted 22 October 2016
established with consideration of the chemical equilibrium of the reversible reaction. The measured LLE
Available online 24 October 2016
data were used to correlate the binary interaction parameters of the NRTL and UNIQUAC models. The
NRTL model is more suitable than UNIQUAC model for describing the LLE of this quaternary reaction
Keywords:
LLE
system.
Quaternary reaction system © 2016 Elsevier B.V. All rights reserved.
Sec-butyl acetate
NRTL model
UNIQUAC model

1. Introduction and low price and toxicity. So the basic research of sec-butyl acetate
has attracted more attention. But sec-butyl acetate as entrainer has
The separation of acetic acid from aqueous solutions is not been reported.
extremely important in industry. Azeotropic distillation and Up to now, the determination of the ternary LLE data for this
extraction distillation were often used to separate the solution of system has been carried out. The experimental LLE date for the
acetic acid-water in industry. And the entrainers were mainly iso- ternary system water þ sec-butyl acetate þ acetic acid was
propyl acetate, butyl acetate, isobutyl acetate, ethyl acetate, ben- measured by Hu et al. [5] at 298.15 K, 303.15 K, 308.15 K and
zene [1e4], and so on. As an isomer of isobutyl acetate, sec-butyl 313.15 K. The NRTL and UNIQUAC models were applied to fit the
acetate has the same water carrying capacity as isobutyl acetate. experimental data for the ternary system. In addition, the LLE date
But it has larger water carrying capacity than isopropyl acetate and for the two ternary systems water þ butyl acetate þ acetic acid and
ethyl acetate. Sec-butyl acetate was produced by the addition re- water þ isobutyl acetate þ acetic acid were determined by Wang
action of acetic acid and 1-butene or 2-butene from liquefied pe- et al. [6] at 304.15 K, 332.15 K and 366.15 K, and the experimental
troleum gas, which has been developed on a scale of 400 000 tons/ data were used to estimate the LLE interaction parameters in the
year by Hunan Zhongchuang Chemical Co., Ltd. since 2006. This NRTL model.
production route exhibits high selectivity and low raw material However, the LLE data for the quaternary system water þ sec-
costs, and thereby significantly reduces the cost of sec-butyl ace- butyl alcohol þ sec-butyl acetate þ acetic acid have not been re-
tate. We did hydrolysis experiment of n-butyl acetate, isobutyl ported in literature. In this quaternary system, there is a reversible
acetate and sec-butyl acetate which catalyzed by acetic acid at the reaction between sec-butyl acetate þ water and sec-butyl
same condition. The Hydrolysis rates were 25.01%, 21.90% and 13.3% alcohol þ acetic acid, where acetic acid acts not only as catalyst
respectively. The prices are 7850 yuan, 12000 yuan, 6000 yuan per but also as reactant. There exist both chemical reaction equilibrium
ton respectively. The LD50 are 10768 mg/kg, 15400 mg/kg, between hydrolysis reaction and esterification and liquid liquid
13400 mg/kg respectively. So it has more stable chemical property equilibrium in this quaternary system water þ sec-butyl
alcohol þ sec-butyl acetate þ acetic acid. In the determination of
the experimental data, it is not only to achieve the chemical equi-
librium but also the liquid-liquid equilibrium. In the mathematical
* Corresponding author.
E-mail address: zxhuang@fzu.edu.cn (Z. Huang).
model of this phase equilibrium with reaction the liquid-liquid

http://dx.doi.org/10.1016/j.fluid.2016.10.027
0378-3812/© 2016 Elsevier B.V. All rights reserved.
 L. Li et al. / Fluid Phase Equilibria 432 (2017) 70e75 71

equilibrium and the reaction equilibrium should be both consid- Table 2

ered. Due to the effect of this reversible reaction, the LLE for this Densities r, refractive indexes nD of pure compounds at P ¼ 101.3 kPa.

quaternary system is more complicated than other systems without Compound r(g/cm3) nD
chemical reaction. In presented paper, the LLE date for the qua- Lit. Exp. Lit. Exp
ternary reaction system of water þ sec-butyl alcohol þ sec-butyl
water(298.15 K) 0.9970 [7,8] 0.9970 1.3325 [8] 1.3320
acetate þ acetic were determined. Additionally, the LLE model for
acetic acid(298.15 K) 1.0437 [8] 1.0446 1.3698 [8] 1.3703
this system was established by considering the hydrolysis of sec- sec-butyl acetate(293.15 K) 0.8720 [8] 0.8710 1.3894 [8] 1.3876
butyl acetate catalyzed by acetic acid. Finally, the experimental sec-butyl alcohol(298.15 K) 0.8080 [8] 0.8040 1.3969 [7] 1.3962
data were correlated with the NTRL and UNIQUAC models. The measurement uncertainties for r, nD and T are u(r) ¼ ±0.001 g cm3,
u(nD) ¼ ±0.0002 and u(T) ¼ ±0.01 K, respectively.
2. Experimental

2.1. Materials

In this paper, all the chemicals were high purity grade. Sec-butyl
5
acetate was distilled in a glass column in order to eliminate the
organic impurities. The sources of the materials, purification
methods, and final purities were listed in Table 1. The refractive 2
indexes of these pure components were measured by a WAY
refractometer with uncertainty of ±0.0002. The densities of the
pure liquids were measured by an densimeter(Anton Paar DMA-58) 3
with uncertainty of ±0.0001 g cm3. The experimental values of
these properties at P¼101.3 kPa were compared with their coun- 6
4
terparts in literature, as shown in Table 2.
7
1
2.2. Sample analysis

Sec-butyl alcohol and sec-butyl acetate were determined by a

Gas Chromatograph (GC-2010, Japan Shimadzu) equipped with a
flame ionization detector (FID) and a free fatty acid phase (FFAP)
column (30 m  0.32 mm  0.25 mm), and their compositions were
analyzed by internal standard method with isopropanol as internal
standard substance. Nitrogen was used as carrier gas at the flow- Fig. 1. Apparatus to measure reaction equilibrium constant. 1. thermostatic oil bath; 2.
rate of 3.0 mL/min. The temperatures of the injection port and thermometer; 3. sampling (or feeding) hole; 4. reactor (three neck flask); 5. condenser
the detector were held at 520.15 K and 550.15 K, respectively. The pipe; 6. thermocouple; 7. magnet stirrer.

column temperature was first kept at 323.15 K for 1 min, then

increasing at the rate of 15 K/min to 423.15 K, finally keeping at
was placed above the three-neck flask to avoid the volatilization of
423.15 K for 1 min. The chromatographic peaks for all the organics
the reactants. Prior to experiment, sec-butyl acetate and acetic acid
involved in the experiment could be completely separated with this
were charged into the three-neck flask, and heated to desired
analysis method. The concentration of acetic acid was determined
temperature by the thermostatic oil bath. In the following, the
by acid base titration with sodium hydroxide (NaOH) as basic
water preheated to the same temperature was added into the
titrant and phenolphthalein as indicator. The water content was
three-neck flask. Meanwhile, this moment was taken as the initial
tested by Switzerland Metrohm 756Karl-Fischer Coulometer with
moment of reaction. During the process of reaction, sample was
uncertainty of ±0.0001mass fraction.
taken through the sampling site every minutes to monitor the
reactant concentration in the reactor. The sample composition was
2.3. Apparatus and procedure
analyzed by the method in section 2.2. When the concentration of
each reactant in the reactor was almost unchanged, the reversible
The reaction equilibrium constant for the hydrolysis of sec-butyl
reaction was considered to arrive its chemical equilibrium.
acetate was measured by the apparatus shown in Fig. 1. The reac-
The most widely used equilibrium still method was employed to
tion was carried out in a 500 mL three-neck flask at atmosphere
measure the LLE data in this paper. The schematic of the experi-
pressure. The temperature inside the three neck flask was
mental apparatus is shown in Fig. 2. The temperature in the liquid-
controlled by a thermostatic oil bath with uncertainty of ±0.5 K.
liquid cell was controlled by the concatenation of a thermostatic
This thermostatic oil bath could provide a rotating magnetic field.
water bath (DBK-501A, Shanghai Laboratory Instrument Works
By means of this magnetic field, a cylindrical magnet was put into
Co.Ltd) and a cooling water bath (DC-3006, Shanghai Hengping
the three-neck flask to stir the liquid to ensure the uniform dis-
Instrument and Meter Factory). The uncertainty of this
tribution of temperature and reactant concentration. A condenser

Table 1
.Mass fraction purities of chemical samples.

Compound Source Initial mass fraction Purification method Final mass fraction Analysis method

acetic acid Sinopharm Chemical Reagent Co., Ltd 0.998 none 0.998 acid base titration
sec-butyl acetate Hunan Zhongchuang Chemical Co., Ltd. 0.987 distillation 0.997 GCa
sec-butyl alcohol Sinopharm Chemical Reagent Co., Ltd 0.998 none 0.998 GCa
a
GC: gas chromatography.
72 L. Li et al. / Fluid Phase Equilibria 432 (2017) 70e75

4
3

Fig. 2. Apparatus to measure LLE data. 1 liquid-liquid equilibrium cell; 2 magnetic

stirrer; 3 thermostatic water bath; 4 cooling water bath; 5 condenser.

temperature-controlling system is ±0.01 K. Before the experiment,

Fig. 3. Correlation of reaction equilibrium constant and temperature.
about 25 mL of mixture sample was put into the equilibrium cell.
The concentration of each chemical was adjusted to ensure the
liquid-liquid phase separation. The mixture in the equilibrium cell water þ sec-butyl alcohol þ sec-butyl acetate þ acetic acid with
was vigorously stirred for at least 4 h, and then left to settle for at reversible reaction were determined at 303.15, 308.15, 313.15,
least 12 h. Finally, a sample of each phase was drown out by a 318.15, 323.15 K and 333.15 K at P¼101.3 kPa, as shown in Table 3.
dropping pipette. The composition of each chemical in the sample
was analyzed by the method in Section 2.2.

3. Results and discussion 3.2. Thermodynamic model of LLE with reversible reaction

3.1. Experimental data If the liquid-liquid phase equilibrium was established, the two
phases are equal not only in temperature and pressure but also in
3.1.1. Reaction equilibrium constant for hydrolysis reaction of sec- fugacity coefficient of each component [9]:
butyl acetate
The ternary system sec-butyl acetate þ water þ acetate acid is xIi gIi ¼ xIIi gIIi (3)
partially miscible. In order to avoid the liquid-liquid separation of
the reaction system, the initial concentration of each chemical was Here, xІi , xІІ
i are the mole fractions of component i in each phase
chosen by referring to the LLE data for the ternary system of the LLE system, respectively. Correspondingly, gІi , gІi І are the
water þ sec-butyl acetate þ acetic acid measured by Hu et al. [5]. fugacity coefficients of component i, which could be calculated by
The equilibrium constant Ke for hydrolysis reaction of sec-butyl the NRTL [10,11] or UNIQUAC [12] model. The superscripts I and II
acetate was measured at five different temperatures (348, 353, indicate organic phase and water phase, respectively. Additionally,
358, 363 and 368 K). The effect of temperature on Ke is shown in the sum of morality of each component must satisfy the normali-
Fig. 3. The increase of temperature with Ke indicates that the hy- zation equation in the each phase of the system:
drolysis of sec-butyl acetate is an endothermic reaction. Generally,
there is a linear relationship between lnKe and 1/T: X
4
ðxi ÞI ¼ 1 (4)

 i¼1
A
Ke ¼ exp þB (1)
T
X
4
Here, both A and B are the parameters of Eq. (1). The experi- ðxi ÞII ¼ 1 (5)
mental data in Fig. 3 were used to correlate the two coefficients of i¼1
Eq. (1) by the least square method: A ¼ 728.21; B ¼ 0.9485. The
correlation coefficient for Eq. (2) is 0.985. The chemical equilibrium For the LLE of the quaternary system water þ sec-butyl
constant Ke for hydrolysis reaction of sec-butyl acetate catalyzed by alcohol þ sec-butyl acetate þ acetic acid, there is a reversible re-
acetic acid was calculated with the following equation: action between sec-butyl acetate þ water and sec-butyl
alcohol þ acetic acid due to the catalysis of acetic acid. As a
728:21 result, the chemical equilibrium of this reversible reaction should
In Ke ¼   0:9485 (2) be taken into consideration when constructing the LLE model of
T
this quaternary system. Obviously, once the LLE was established,
the reversible reaction had reached its chemical equilibrium.
Therefore, the chemical equilibrium constant Ke for each phase of
3.1.2. LLE data for the quaternary system the LLE system with reversible reaction must be equal to each
The experimental LLE data for the quaternary system other:
 L. Li et al. / Fluid Phase Equilibria 432 (2017) 70e75 73

Table 3
LLE data for the system water(1) þ sec-butyl alcohol(2) þ sec-butyl acetate(3) þ acetic acid(4) at P ¼ 101.3 kPa.

T/K No. Organic phase Aqueous phase

xI1 xI2 xI3 xI4 xII1 xII2 xII3 xII4

303.15 1 0.2078 0.1881 0.4773 0.1267 0.9514 0.0108 0.0019 0.0358

2 0.4503 0.0602 0.3304 0.1591 0.9146 0.0054 0.0049 0.0752
3 0.0979 0.1645 0.6202 0.1173 0.9588 0.0082 0.0017 0.0314
4 0.5233 0.0477 0.3125 0.1165 0.9315 0.0056 0.0027 0.0602
5 0.2654 0.1092 0.3962 0.2292 0.9002 0.0089 0.0046 0.0863
6 0.4941 0.0452 0.2648 0.1958 0.8779 0.0064 0.0068 0.1088
7 0.4413 0.0868 0.3278 0.1442 0.9246 0.0079 0.0027 0.0648
8 0.0324 0.0941 0.6179 0.2557 0.9115 0.0034 0.0023 0.0828
308.15 1 0.4978 0.2107 0.1421 0.1525 0.9032 0.0219 0.0058 0.0635
2 0.594 0.1587 0.1075 0.1075 0.9408 0.0159 0.0021 0.0412
3 0.5886 0.2232 0.1178 0.1178 0.9041 0.0377 0.0039 0.0542
4 0.4419 0.2094 0.1068 0.1068 0.9476 0.0148 0.0018 0.0358
5 0.4917 0.2497 0.1296 0.1296 0.9205 0.0273 0.0031 0.0490
6 0.5022 0.1972 0.1428 0.1594 0.9162 0.0192 0.0037 0.0612
7 0.4385 0.1123 0.2052 0.2052 0.8866 0.0136 0.0059 0.0939
8 0.4635 0.1105 0.1395 0.2332 0.8720 0.0116 0.0076 0.1087
313.15 1 0.5395 0.2401 0.1013 0.1191 0.9211 0.0287 0.0028 0.0474
2 0.6144 0.2357 0.0527 0.0972 0.9334 0.0316 0.0020 0.0363
3 0.5250 0.2546 0.0830 0.1087 0.9311 0.0258 0.0021 0.0410
4 0.5436 0.2880 0.0702 0.0982 0.9307 0.0358 0.0019 0.0316
5 0.5581 0.2686 0.0767 0.1037 0.9367 0.0247 0.0019 0.0368
6 0.8348 0.1029 0.0239 0.0399 0.9297 0.0347 0.0017 0.0294
7 0.6107 0.2706 0.0378 0.0896 0.9388 0.0270 0.0014 0.0328
8 0.5244 0.1707 0.1978 0.1068 0.9298 0.0247 0.0031 0.0424
318.15 1 0.4640 0.2574 0.1580 0.1206 0.9334 0.0216 0.0025 0.0425
2 0.4662 0.2177 0.1582 0.1578 0.9085 0.0232 0.0034 0.0649
3 0.3454 0.2280 0.2677 0.1588 0.9300 0.0156 0.0026 0.0518
4 0.3408 0.2031 0.3158 0.1404 0.9378 0.0130 0.0027 0.0465
5 0.5313 0.2449 0.1100 0.1138 0.9269 0.0265 0.0028 0.0437
6 0.4839 0.0986 0.2232 0.1943 0.8842 0.0132 0.0068 0.0958
7 0.5166 0.2473 0.1148 0.1213 0.9318 0.0251 0.0025 0.0407
8 0.4198 0.0974 0.2876 0.1952 0.8939 0.0102 0.0058 0.0901
323.15 1 0.5557 0.3454 0.0526 0.0463 0.9543 0.0299 0.0008 0.015
2 0.5433 0.1001 0.175 0.1816 0.8850 0.0149 0.0063 0.0939
3 0.5382 0.2186 0.1210 0.1222 0.9239 0.0226 0.0025 0.0509
4 0.5670 0.2908 0.0754 0.0668 0.9478 0.0272 0.0013 0.0237
5 0.4137 0.2456 0.1985 0.1422 0.9303 0.0184 0.0024 0.0488
6 0.5409 0.0737 0.1885 0.1969 0.8647 0.0128 0.0082 0.1142
7 0.4767 0.2242 0.1819 0.1171 0.9387 0.0172 0.0018 0.0422
8 0.5604 0.2113 0.1108 0.1174 0.9254 0.0236 0.0025 0.0485
333.15 1 0.4423 0.1704 0.2893 0.0980 0.9527 0.0106 0.0015 0.0353
2 0.5184 0.0664 0.2503 0.1649 0.9001 0.0078 0.0043 0.0877
3 0.3556 0.1949 0.3499 0.0997 0.9559 0.0100 0.0015 0.0326
4 0.5856 0.0442 0.1482 0.2220 0.7980 0.0163 0.0235 0.1622
5 0.3423 0.1818 0.3785 0.0975 0.9580 0.0088 0.0015 0.0317
6 0.4715 0.1383 0.2390 0.1512 0.9208 0.0119 0.0028 0.0644
7 0.4948 0.0529 0.2075 0.2448 0.8252 0.0120 0.0149 0.1479
8 0.4690 0.1454 0.2460 0.1396 0.9271 0.0114 0.0026 0.0588

The measurement uncertainties for T and x are u(T) ¼ ±0.01 K and u(x) ¼ ±0.0002, respectively.

Y vi 2 !2 !2 3
Ke¼ gIi xIi ði ¼ 1; 2; 3; 4Þ (6) XN X n x
I;exp
 xI;cal
II;exp
xik  xII;cal
F ¼ min 4 ik ik
þ ik 5 (8)
k¼1 i¼1 xI;exp
ik
xI;exp
ik
Y vi
Ke¼ gIIi xIIi ði ¼ 1; 2; 3; 4Þ (7) Where i and k are the numbers of components and sets of experi-
mental data, respectively, while superscript exp and cal respect the
experimental and calculated values, respectively. The correlated
binary interaction parameters of the NRTL and UNIQUAC models
are listed in Tables 4 and 5, respectively. It should be pointed out
that the non-randomness parameters of the NRTL model were set
3.3. Correlation of LLE data
as 0.3 to simply the optimizing calculation.
The experimental LLE data for the quaternary system
water þ sec-butyl alcohol þ sec-butyl acetate þ acetic acid were 3.4. Prediction
correlated with NRTL and UNIQUAC models. The binary interaction
parameters of these two models were obtained by minimizing the The NRTL and UNIQUAC models can be used to calculate the LLE
following objective function F through the simplex method [9]: data on the basis of the correlated binary interaction parameters.
74 L. Li et al. / Fluid Phase Equilibria 432 (2017) 70e75

Table 4
Parameters of the NRTL model for the LLE system water(1) þ sec-butyl alcohol(2) þ sec-butyl acetate(3) þ acetic acid(4).

T/K Model parameters/(J mol1)

303.15 Dg12 ¼ 8957.42 Dg21 ¼ 811.342 Dg23 ¼ 1707.38 Dg32 ¼ 15391.82

Dg13 ¼ 18322.13 Dg31 ¼ 18076.74 Dg24 ¼ 20959.37 Dg42 ¼ 915.872
Dg14 ¼ 1165.16 Dg41 ¼ 8163.25 Dg34 ¼ 896.392 Dg43 ¼ 2764.21
308.15 Dg12 ¼ 10396.75 Dg21 ¼ 2156.50 Dg23 ¼ 1207.48 Dg32 ¼ 9712.77
Dg13 ¼ 17900.77 Dg31 ¼ 20999.41 Dg24 ¼ 8057.64 Dg42 ¼ 15151.35
Dg14 ¼ 844.691 Dg41 ¼ 5444.02 Dg34 ¼ 4464.99 Dg43 ¼ 2005.55
313.15 Dg12 ¼ 9660.40 Dg21 ¼ 20998.16 Dg23 ¼ 3423.16 Dg32 ¼ 6424.80
Dg13 ¼ 17525.49 Dg31 ¼ 174.908 Dg24 ¼ 2161.24 Dg42 ¼ 2647.81
Dg14 ¼ 3487.05 Dg41 ¼ 4948.51 Dg34 ¼ 2308.71 Dg43 ¼ 4882.78
318.15 Dg12 ¼ 7979.81 Dg21 ¼ 2068.39 Dg23 ¼ 5886.27 Dg32 ¼ 396.247
Dg13 ¼ 14458.71 Dg31 ¼ 8478.43 Dg24 ¼ 4167.06 Dg42 ¼ 2028.69
Dg14 ¼ 2982.21 Dg41 ¼ 8755.30 Dg34 ¼ 4760.49 Dg43 ¼ 2035.73
323.15 Dg12 ¼ 8053.53 Dg21 ¼ 798.621 Dg23 ¼ 546.330 Dg32 ¼ 9411.70
Dg13 ¼ 17648.30 Dg31 ¼ 8556.24 Dg24 ¼ 7344.26 Dg42 ¼ 12796.06
Dg14 ¼ 803.111 Dg41 ¼ 14789.29 Dg34 ¼ 1821.83 Dg43 ¼ 1956.77
333.15 Dg12 ¼ 6656.17 Dg21 ¼ 789.514 Dg23 ¼ 3637.94 Dg32 ¼ 1716.41
Dg13 ¼ 20043.89 Dg31 ¼ 8750.20 Dg24 ¼ 1106.44 Dg42 ¼ 587.626
Dg14 ¼ 2499.52 Dg41 ¼ 15234.68 Dg34 ¼ 1631.82 Dg43 ¼ 217.870

Table 5
Parameters of the UNIQUAC model for the LLE system water(1) þ sec-butyl alcohol(2) þ sec-butyl acetate(3) þ acetic acid(4).

T/K Model parameters/(J mol1)

303.15 DU12 ¼ 1605.56 DU21 ¼ 80.096 DU23 ¼ 2603.51 DU32 ¼ 1891.51

DU13 ¼ 519.693 DU31 ¼ 19133.08 DU24 ¼ 2056.19 DU42 ¼ 19988.14
DU14 ¼ 255.300 DU41 ¼ 19972.24 DU34 ¼ 286.050 DU43 ¼ 18201.75
308.15 DU12 ¼ 258.458 DU21 ¼ 14978.93 DU23 ¼ 2119.38 DU32 ¼ 15469.60
DU13 ¼ 2953.93 DU31 ¼ 137.400 DU24 ¼ 6092.92 DU42 ¼ 2452.67
DU14 ¼ 5015.01 DU41 ¼ 5142.07 DU34 ¼ 134.368 DU43 ¼ 593.276
313.15 DU12 ¼ 699.556 DU21 ¼ 2171.98 DU23 ¼ 813.819 DU32 ¼ 2155.85
DU13 ¼ 4411.05 DU31 ¼ 935.551 DU24 ¼ 2578.56 DU42 ¼ 63.4043
DU14 ¼ 3313.36 DU41 ¼ 5173.45 DU34 ¼ 3390.17 DU43 ¼ 7256.22
318.15 DU12 ¼ 268.156 DU21 ¼ 7723.87 DU23 ¼ 1049.29 DU32 ¼ 1200.24
DU13 ¼ 1818.96 DU31 ¼ 3648.23 DU24 ¼ 19999.04 DU42 ¼ 1751.35
DU14 ¼ 2266.65 DU41 ¼ 2351.70 DU34 ¼ 2072.75 DU43 ¼ 19982.47
323.15 DU12 ¼ 603.857 DU21 ¼ 2129.38 DU23 ¼ 49.3267 DU32 ¼ 1023.47
DU13 ¼ 2608.53 DU31 ¼ 1067.00 DU24 ¼ 3397.48 DU42 ¼ 2830.83
DU14 ¼ 3714.52 DU41 ¼ 19985.33 DU34 ¼ 19935.91 DU43 ¼ 1421.91
333.15 DU12 ¼ 842.272 DU21 ¼ 3027.32 DU23 ¼ 19997.19 DU32 ¼ 2292.30
DU13 ¼ 1730.28 DU31 ¼ 3388.98 DU24 ¼ 19807.33 DU42 ¼ 1533.59
DU14 ¼ 2787.90 DU41 ¼ 19819.08 DU34 ¼ 19673.69 DU43 ¼ 1547.60

Fig. 4. The relationship between experiment values xexp and calculated values xcal by Fig. 5. The relationship between experiment values xexp and calculated values xcal by
the NRTL model. the UNIQUAC model.
 L. Li et al. / Fluid Phase Equilibria 432 (2017) 70e75 75

The relationship between experiment values xexp and calculated suitable for describing the LLE of the quaternary systems
values xcal were shown in Fig. 4 and Fig. 5. water þ sec-butyl alcohol þ sec-butyl acetate þ acetic acid.
For the NRTL model the average relative deviations of each
component are 7.28%, 6.59%, 9.77%, 5.62% for water, sec-butyl Acknowledgements
alcohol, sec-butyl acetate and acetic acid respectively in organic
phase. And values are 0.77%, 12.66%, 8.92%, 9.26% respectively in We acknowledge the financial support for this work from the
water phase. For the UNIQUAC model the average relative de- National Natural Science Foundation of China (Nos. 21576053 and
viations of each component are 10.97%, 12.54%, 21.71%, 8.17% for 91534106), the Science Foundation for Distinguished Young
water, sec-butyl alcohol, sec-butyl acetate and acetic acid respec- Scholars of Fujian (No. 2014J06004), the Natural Science Founda-
tively in organic phase. And values are 0.84%, 16.76%, 16.15%, 9.92% tion of Fujian Province (No. 2016J01689), the International S&T
respectively in water phase. It suggests that NRTL model is more Cooperation Program of China (No. 2013DFR90540), the Key Project
suitable than UNIQUAC model to describe the LLE for the quater- of Fujian Provincial Department of Science and Technology (No.
nary system water þ sec-butyl alcohol þ sec-butyl acetate þ acetic 2014Y0066).
acid at the temperature from 303.15 K to 333.15 K.
References
4. Conclusion
[1] H.J. Huang, I.L. Chien, J. Chin, Ins. Chem. Eng. 39 (2008) 503e517.
The LLE data for the quaternary systems water þ sec-butyl [2] S.J. Wang, D.S.H. Wong, J. Process Contr. 23 (2013) 78e88.
[3] I.L. Chien, H.P. Huang, T.K. Gau, C.H. Wang, Ind. Eng. Chem. Res. 44 (2005)
alcohol þ sec-butyl acetate þ acetic acid at the temperature range 3510e3521.
of 303.15 Ke333.15 K were measured. The LLE thermodynamic [4] S.J. Li, Ind. Eng. Chem. Res. 48 (2009) 6358e6362.
model of the quaternary system was established with consideration [5] S. Hu, Q.L. Chen, B.J. Zhang, Y.L. Liang, X.N. Gao, Fluid Phase Equilib. 293 (2010)
73e78.
of the chemical equilibrium of the reversible reaction in this sys-
[6] L.J. Wang, Y.W. Cheng, X.J. Xiao, X. Li, J. Chem. Eng. Data 52 (2007) 1255e1257.
tem. The binary interaction parameters of the NRTL and UNIQUAC [7] N.L. Chen, Chem. Ind. Press, 2008.
models were correlated with the LLE data by the Simplex Method. [8] G.J Liu, L.X. Ma, S.G. Xiang, Chem Ind Press, 2002.
The correlated binary interaction parameters were used to predict [9] J.M. Smith, H.C. Van Ness, M.M. Abbott, McGraw - Hill, NewYork, 2005.
[10] H. Renon, J.M. Prausnitz, AIChE J. 14 (1968) 135e144.
the LLE data. The small deviation between the predicted and [11] R.L. Scott, J. Chem. Phys. 25 (1956) 193.
experimental values indicates that both these two models are [12] D.S. Abrams, J.M. Prausnitz, AIChE J. 1 (1975) 116e128.