Fluid Phase Equilibria 406 (2015) 55–60

Contents lists available at ScienceDirect

Fluid Phase Equilibria

journal homepage: www.elsevier.com/locate/fluid

Experimental isobaric vapor liquid equilibrium for the ternary

system of sec-butyl alcohol + sec-butyl acetate + dimethyl sulfoxide at
101.3 kPa
Xuemei Zhang a , Yunxian Liu a , Hongpeng Liu a , Jie He a,∗ , Chungui Jian b
a
School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, PR China
b
Tianjin Zhonghao Unitech co., Ltd., Tianjin 300072, PR China

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

Article history: Isobaric vapor–liquid equilibrium (VLE) data were measured for the binary systems sec-butyl
Received 17 April 2015 alcohol + dimethyl sulfoxide (DMSO), sec-butyl acetate + DMSO and the ternary system sec-butyl alco-
Received in revised form 10 July 2015 hol + sec-butyl acetate + DMSO at 101.3 kPa using a vapor–liquid equilibrium still. The thermodynamic
Accepted 13 July 2015
consistency test for the binary experimental data was made by Herington method, Wisniak’s L–W method
Available online 18 July 2015
and a point to point method of Van Ness test described by Fredenslund et al. And the thermodynamic
consistency of the ternary system was checked by Van Ness test. The VLE data for the binary systems were
Keywords:
correlated with the NRTL, Wilson and UNIQUAC activity coefficient models. And the binary and ternary
Vapor–liquid equilibrium
sec-butyl acetate
VLE data predicted by the obtained binary interaction parameters agree well with the experimental data.
sec-Butyl alcohol The experimental results show that the relative volatility between sec-butyl alcohol and sec-butyl acetate
Dimethyl sulfoxide is changed obviously when the mass ratio of DMSO to the pseudo-azeotrope mixture is 6:5. And DMSO is
a promising extractive agent for the separation of the mixture of sec-butyl alcohol and sec-butyl acetate
in extractive distillation.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction (2) + DMSO (3) have not been reported in the literature. VLE data of
the sec-butyl alcohol (1) + sec-butyl acetate (2) + DMSO (3) system
sec-Butyl alcohol and sec-butyl acetate are important solvents are vital for the simulation and design of an extractive distillation
and fine chemicals. The difference between vapor and liquid com- process. To provide fundamental data for the separation of sec-
position is nearly negligible. The mixture of sec-butyl alcohol and butyl alcohol and sec-butyl acetate by extractive distillation using
sec-butyl acetate is difficult to be separated by a conventional distil- DMSO as extractive agent, isobaric VLE data for these systems were
lation process. Extractive distillation is a practical way of separating explored in this work.
the difficult separated mixture, in which extractive agent is added
to alter the vapor liquid equilibrium (VLE) of the binary system 2. Experimental
sec-butyl alcohol and sec-butyl acetate and make the separation
feasible. N,N-dimethyl formamide and N,N-dimethyl acetamide 2.1. Experimental chemicals
were proved to be promising solvent in separation of sec-butyl alco-
hol and sec-butyl acetate mixture by extractive distillation in our The analytical reagents from Tianjin Jiangtian Co., Ltd., China,
previous work [1,2]. Dimethyl sulfoxide (DMSO) is also indicated sec-butyl alcohol, sec-butyl acetate and DMSO, were used in this
as a potential extractive agent for the separation of the discussed work. Their molecular formula, CASRN and mass fraction are listed
binary mixture [1,3]. So far, only the isobaric VLE data of sec-butyl in Table 1. The purities of the chemicals were determined by a
alcohol (1) + sec-butyl acetate (2) can be found in the open litera- gas chromatograph (GC) equipped with a flame ionization detector
ture [4]. The VLE data for sec-butyl alcohol (1) + DMSO (3), sec-butyl (FID). The physical properties of the pure components are shown
acetate (2) + DMSO (3) and sec-butyl alcohol (1) + sec-butyl acetate in Table 2. The saturated vapor pressure pi s of the chemicals were
calculated by the following extended Antoine equation

ln(psi /kPa) = C1,i + C2,i /[(T/K) + C3,i ] + C4,i (T/K) + C5,i ln(T/K)
∗ Corresponding author.
C7,i
E-mail address: hejie tju@126.com (J. He). + C6,i (T/K) (1)

http://dx.doi.org/10.1016/j.fluid.2015.07.029
0378-3812/© 2015 Elsevier B.V. All rights reserved.
56 X. Zhang et al. / Fluid Phase Equilibria 406 (2015) 55–60

Table 1 Table 4
Experimental reagents list. Isobaric VLE data and activity coefficients for the binary system of sec-butyl alcohol
(1) + DMSO (3) at 101.3 kPa.a
Reagent sec-Butyl alcohol sec-Butyl acetate DMSO
T (K) x1 y1 1 2
Molecular formula C4 H10 O C6 H12 O2 C2 H6OS
CASRN 78-92-2 105-46-4 67-68-5 372.90 1.000 1.000 0.957 –
Purity (mass fraction) 0.999 0.997 0.998 373.50 0.971 0.999 0.963 0.677
375.45 0.904 0.996 0.962 0.734
378.50 0.815 0.991 0.952 0.775
Table 2 383.00 0.720 0.983 0.915 0.825
Physical properties of the components.a , b 388.45 0.605 0.966 0.893 0.922
394.20 0.513 0.946 0.858 0.954
Component sec-Butyl alcohol sec-Butyl acetate DMSO 399.30 0.436 0.923 0.841 0.973
M (g mol ) −1
74.1228 116.1601 78.1350 404.25 0.365 0.894 0.840 0.996
 (kg m−3 ) 802.3157 866.4043 1096.2170 410.65 0.288 0.851 0.842 1.000
Tb (K) 372.90 383.15 464.00 414.80 0.246 0.816 0.842 1.014
Tc (K) 535.90 561.00 729.00 420.40 0.197 0.762 0.843 1.025
pc (kPa) 4188.50 3170.00 5650.00 425.25 0.164 0.711 0.831 1.024
ω 0.5808 0.3954 0.2806 429.55 0.137 0.659 0.825 1.024
435.05 0.106 0.581 0.818 1.028
a
Taken from Aspen Plus V7.2 physical properties databanks. 441.20 0.078 0.490 0.804 1.012
b
The density  was obtained at 298.15 K. 444.95 0.062 0.422 0.795 1.013
449.90 0.040 0.305 0.789 1.036
Table 3 456.65 0.017 0.149 0.767 1.031
Antoine constants of the components.a 458.55 0.012 0.108 0.751 1.023
460.10 0.007 0.065 0.745 1.024
Component sec-Butyl alcohol sec-Butyl acetate DMSO 464.00 0.000 0.000 – 0.983
a
C1 115.6422 45.6932 49.3651 Uncertainties: u(x1 ) = 0.004, u(y1 ) = 0.004, u(T) = 0.02 K, u(p) = 0.25 kPa.
C2 −10,236.00 −6097.90 −7620.60
C3 0.0 0.0 0.0
C4 0.0 0.0 0.0 3. Experimental data and discussion
C5 −14.1250 −4.2398 −4.6279
C6 2.3559 × 10−17 2.1506 × 10−18 4.3819 × 10−7
3.1. Experimental data
C7 6.00 6.00 2.00
Tmin (K) 158.45 174.15 291.67
Tmax (K) 535.90 561.00 729.00 The reliability of the operation procedures and the experimen-
a
Taken from Aspen Plus V7.2 physical properties databanks.
tal apparatus were confirmed in our previous work [8]. The isobaric
VLE data of the binary systems sec-butyl alcohol (1) + DMSO (3), sec-
butyl acetate (2) + DMSO (3) and ternary system sec-butyl alcohol
(1) + sec-butyl acetate (2) + DMSO (3) were measured at 101.3 kPa
The value of constants C1 ,i ∼ C7 ,i , Tmin and Tmax for the pure with the modified Othmer still. The experimental data T, x1 , y1 and
components taken from Aspen Plus V7.2 are shown in Table 3.

Table 5
2.2. Apparatus and procedure
Isobaric VLE data and activity coefficients for the binary system of sec-butyl acetate
(2) + DMSO (3) at 101.3 kPa.a
A double circulation vapor–liquid equilibrium still (a modified
Othmer still) was used in the measurement of the isobaric VLE data. T (K) x1 y1 1 2

It includes a liquid-phase sampling port, a vapor-phase sampling 383.15 1.000 1.000 1.013 –
port, an equilibrium chamber, a condenser, a pressure control sys- 385.20 0.968 0.993 0.977 2.623
386.10 0.922 0.985 0.990 2.226
tem and a heating bar. The details of the vapor–liquid equilibrium
387.50 0.864 0.976 1.004 1.934
still were described by Yuan et al. [5] and Li et al. [6]. The exper- 389.10 0.797 0.963 1.025 1.877
imental procedure was described in our previous work [1,2,7]. In 390.55 0.714 0.951 1.083 1.669
this process, both the vapor and the liquid phases were contin- 391.15 0.682 0.945 1.107 1.647
uously circulating to provide intimate contact of the two phases. 392.10 0.650 0.941 1.125 1.548
392.95 0.581 0.931 1.215 1.465
Equilibrium state were assumed when the fluctuation in vapor tem- 393.65 0.559 0.928 1.234 1.415
perature was within 0.1 K in a about (0.5–1) h. Then the liquid and 393.90 0.527 0.921 1.290 1.434
vapor condensate samples were taken simultaneously. To decrease 395.35 0.475 0.912 1.360 1.363
the influence of sample to the equilibrium of the system, the sam- 397.85 0.372 0.901 1.599 1.169
399.05 0.336 0.892 1.695 1.155
ple amount is determined as 0.1 mL. And three analyses were made
399.35 0.328 0.890 1.719 1.149
for each sample to reduce errors. 400.55 0.288 0.881 1.874 1.124
401.65 0.270 0.876 1.929 1.098
2.3. Sample analysis 402.75 0.248 0.866 2.015 1.107
403.35 0.238 0.862 2.056 1.102
405.65 0.202 0.845 2.233 1.089
All the samples were analyzed with a gas chromatograph (GC) 410.60 0.152 0.815 2.512 1.031
(SP-2100, supplied by Beijing Beifen Ruili Analysis Instrument Ltd.). 423.10 0.083 0.701 2.889 1.021
It was equipped with a flame ionization detector (FID) and a sil- 426.30 0.072 0.664 2.920 1.025
ica capillary column (0.32 mm (i.d.) × 30 m × 0.25 ␮m) packed with 441.20 0.036 0.466 2.906 1.002
446.40 0.026 0.376 2.895 0.999
PEG-20 M. Nitrogen was used as the carrier gas at a constant flow 450.05 0.019 0.298 2.902 1.008
of 40 mL/min. The temperatures of vaporizer and detector were 453.90 0.012 0.207 2.941 1.018
503.15 K and 513.15 K, respectively. The oven temperature was pro- 456.75 0.008 0.148 2.972 1.009
grammed at 363.15 K (for 4.5 min), then increased to 463.15 K at 464.00 0.000 0.000 – 0.972
20 K min−1 , finally was maintained at 463.15 K for 5 min. a
Uncertainties: u(x1 ) = 0.004, u(y1 ) = 0.004, u(T) = 0.02 K, u(p) = 0.25 kPa.
 X. Zhang et al. / Fluid Phase Equilibria 406 (2015) 55–60 57

Table 6
Isobaric VLE data and activity coefficients for the ternary system of sec-butyl alcohol
(1) + sec-butyl acetate (2) + DMSO (3) at 101.3 kPa.a

T/K x1 x2 y1 y2 1 2 3

402.16 0.364 0.023 0.845 0.068 0.856 1.733 0.913

402.96 0.341 0.035 0.795 0.108 0.839 1.769 0.972
402.70 0.338 0.060 0.743 0.166 0.839 1.681 0.924
403.16 0.295 0.074 0.675 0.226 0.817 1.741 0.975
403.31 0.270 0.095 0.608 0.290 0.800 1.733 0.994
403.30 0.239 0.122 0.519 0.377 0.771 1.755 0.998
402.55 0.223 0.147 0.458 0.439 0.745 1.731 1.041
403.33 0.187 0.164 0.375 0.514 0.710 1.778 1.050
402.83 0.163 0.189 0.311 0.576 0.685 1.753 1.101
402.81 0.143 0.206 0.261 0.627 0.655 1.751 1.088
402.51 0.113 0.229 0.194 0.698 0.621 1.768 1.050
401.95 0.082 0.254 0.118 0.757 0.529 1.755 1.150
400.66 0.048 0.291 0.060 0.833 0.478 1.748 1.109
a
Uncertainties: u(x1 ) = 0.004, u(y1 ) = 0.004, u(T) = 0.02 K, u(p) = 0.25 kPa.

calculated activity coefficients  1 ,  2 are listed in Tables 4–6. The Fig. 1. Thermodynamic consistency check results by Herington method (䊏, sec-
determination method of the uncertainties for the vapor and liq- butyl alcohol (1) + DMSO (3); , sec-butyl acetate (2) + DMSO (3)).
uid compositions is as follows: Each sample was measured three
times. Then, the mean value was calculated for each sample. And Wisniak’s L–W test can be expressed as following [10,11]
the uncertainty of the vapor and liquid compositions is the maxi-  1 
 1 
mum difference between the experimental value and mean value,  0 Lk dx1 − 0 Wk dx1 
which is given in Tables 4–6. F = 100 ×  1 1 (5)
0
Lk dx1 + 0
Wk dx1
 0
3.2. Data processing T xi s0
Lk =  i 0 i (6)
xi si − T
The relationship of vapor liquid equilibrium system at low or
RT
  yi

medium pressure is expressed as follows [9] Wk =  xi ln i − xi ln (7)
xi si0 xi
ˆ i yi p = xi i s ps ,
 (i = 1, 2, . . ., N) (2)
i i where Ti0 is the boiling point of pure component i at 101.3 kPa; si0
is the molar entropy of vaporization of component i, which is the
where p is the pressure of the vapor–liquid equilibrium system; pi s
difference of molar entropy between the vapor and liquid phase
is the saturation vapor pressure of component i at the equilibrium
at the boiling point Ti 0 . The values of molar entropy for each com-
temperature T and was calculated from the extended Antoine equa-
ponent taken from Aspen PlusV7.2 physical properties databanks
tion; xi , yi are the liquid and vapor mole fractions of component i,
are listed in Table 7. When the value of F is less than 3–5, the iso-
respectively; is and  ˆ i are the fugacity coefficient of pure vapor i
baric VLE data are thermodynamically consistent. The results of
and component i in vapor phase, respectively, which were obtained
Wisniak’s L–W test are presented in Fig. 2 and Table 8.
by the Soave–Redlich–Kwong (SRK) equation;  i is the activity coef-
Van Ness test is regarded as a modeling capability test [14].
ficient of component i in liquid phase. The activity coefficients for
In this work, the NRTL activity coefficient model was chosen to
each component in the liquid phase  i can be determined by Eq.
check the thermodynamic consistency of the binary systems and
(2). The experimental activity coefficients calculated by Eq. (2) are
the ternary system. The criterion is described as following equation
listed in Tables 4–6.
[15]
The thermodynamic consistency for the binary experimental
1  exp 
data was checked by three methods, Herington method [11], Wis- N

niak’s L–W method [10,11] and a point to point method of Van Ness y = 100 yi − yical  (8)
N
test described by Fredenslund et al. [12–15]. And the thermody- i=1
namic consistency of the ternary system was checked by the Van
where N is the number of experimental points; the superscript
Ness test.
exp means experimental data; the superscript cal means calculated
Herington method [9] is expressed as
value. If the mean absolute deviation y is less than 1, the exper-
  imental data are considered to be thermodynamically consistent.
   x1 =1 1 
 S+ − S−   x1 =0
ln 2 dx 1 The checking results for the binary systems are given in Table 8,
D = 100 ×   = 100 ×  x =1   (3)
S+ + S− 1 ln 1  dx1
x1 =0 2 Table 7
The molar entropy of vaporization of component.a , b
Tmax − Tmin
J = 150 × (4) Component sec-Butyl sec-Butyl DMSO
Tmin alcohol acetate

sV (J kmol−1 K−1 ) −394,870 −540,430 −185,880

where S+ is the area of figure ln( 1 / 2 ) vs x1 above the X-axis, S− is sL (J kmol−1 K−1 ) −504,140 −630,990 −280,410
the area of figure ln( 1 / 2 ) vs x1 below the X-axis. When the value s0 (J kmol−1 K−1 ) 109,270 90,560 94,530
of D−J is less than 10, the isobaric VLE data are considered ther- a
sV and sL are the molar entropy of the pure components for vapor phase and
modynamically consistent. The thermodynamic consistency check liquid phase, which are taken from Aspen Plus V7.2 physical properties databanks.
results are presented in Fig. 1 and Table 8. b
The molar entropy of vaporization s0 = sV − sL .
58 X. Zhang et al. / Fluid Phase Equilibria 406 (2015) 55–60

Table 8
Thermodynamic consistency check results for the binary systems of sec-butyl alcohol (1) + DMSO (3) and sec-butyl acetate (2) + DMSO (3).

System D J D–J Result

Herington method
sec-Butyl alcohol (1) + DMSO (3) 18.711 36.645 −17.934 Pass
sec-Butyl acetate (2) + DMSO (3) 7.927 27.862 −19.935 Pass

System L W F Result

Wisniak’s L–W test

sec-Butyl alcohol (1) + DMSO (3) 15.703 16.444 2.305 Pass
sec-Butyl acetate (2) + DMSO (3) 24.861 26.233 2.685 Pass

System y1 y2 y3 Result

Van Ness test

sec-Butyl alcohol (1) + DMSO (3) 0.395 0.395 – Pass
sec-Butyl acetate (2) + DMSO (3) 0.299 0.299 – Pass
sec-Butyl alcohol (1) + sec-butyl acetate (2) + DMSO (3) 0.508 0.286 0.649 Pass

which indicate that all the experimental data pass the thermody-
namic consistency test.
The experimental data for the binary systems were corre-
lated with the Wilson, NRTL, and UNIQUAC models by Aspen Plus
V7.2 [16]. The binary interaction parameters of each model were
obtained by minimizing the maximum likelihood objective func-
tion, which is expressed
⎡ 2

N
Ti
exp
− Ti
exp 2 exp
pi − pcal
2 exp cal
x1,i − x1,i
Q = ⎣ + i
+
ıT ıp ıx1
i=1

2
⎤
exp cal
y1,i − y1,i
+ ⎦ (9)
ıy1

where N is the number of experimental points; T and p are the equi-

librium temperature and pressure; respectively; x1 and y1 are the
liquid and vapor phase mole fraction of light component, respec-
tively; ı is the standard deviation of the indicated physical property. Fig. 3. T vs x1 , y1 diagram for the sec-butyl alcohol (1) + DMSO (3) system at 101.3 kPa
The binary interaction parameters of Wilson, NRTL, and (䊏, experimental data; , calculated data with NRTL model; , calculated data with
UNIQUAC models are listed in Table 9, together with the root- UNIQUAC model; 䊐, calculated data with Wilson model).

mean-square deviations (RMSD) of the equilibrium temperature

Fig. 4. T vs x1 , y1 diagram for the sec-butyl acetate (1) + DMSO (3) system at 101.3 kPa
Fig. 2. Thermodynamic consistency check results by Wisniak’s L–W test (䊏, L for (䊏, experimental data; , calculated data with NRTL model; , calculated data with
sec-butyl alcohol (1) + DMSO (3) system; 䊉, W for sec-butyl alcohol (1) + DMSO (3) UNIQUAC model; 䊐, calculated data with Wilson model).
system; 䊐, L for sec-butyl acetate (2) + DMSO (3) system; , W for sec-butyl acetate
(2) + DMSO (3) system).
 X. Zhang et al. / Fluid Phase Equilibria 406 (2015) 55–60 59

Table 9
The binary interaction parameters and root-mean-square deviations (RMSD) for binary systems.

Model Correlation parameters RMSD

aij aji bij bji ıy1 a ıT b

sec-Butyl alcohol (1) + sec-butyl acetate (2) [1]

NRTL 5.2706 0.3259 −1586.70 −356.75 0.0107 0.3439
Wilson −0.1245 −4.9766 180.06 1601.26 0.0111 0.3447
UNIQUAC −0.6831 −0.4083 160.88 214.75 0.0111 0.3460

sec-Butyl alcohol (1) + DMSO (3)

NRTLc 0.0171 −3.3345 −624.75 2373.26 0.0035 0.4452
Wilsond 7.2661 −0.7758 −3970.54 788.87 0.0043 0.6931
UNIQUACe −0.4916 3.0451 560.08 −1865.71 0.0043 0.6053

sec-Butyl acetate (2) + DMSO (3)

NRTLc 3.1645 −2.7466 −1108.75 1508.29 0.0049 0.3735
Wilsond 3.3933 −6.8786 −1711.57 2494.34 0.0036 0.3372
UNIQUACe −0.2267 0.1710 −149.92 −32.93 0.0072 0.5269
exp 2 1/2
a
ıy1 = [(1/n)˙i=1
n
(y1,i − y1,i
cal
) ] .
2 1/2
exp
b
ıT = n
[(1/n)˙i=1 (Ti −
Tical ) ] .
c
NRTL,  ij = aij + bij /T, the value of ˛ij was fixed at 0.3 for the binary systems.
d
Wilson, lnAij = aij + bij /T.
e
UNIQUAC,  ij = exp(aij + bij /T).

Table 10
Mean absolute deviations of equilibrium temperature and vapor-phase mole fraction for the ternary system of sec-butyl alcohol (1) + sec-butyl acetate (2) + DMSO (3).

Model Maximum absolute deviations Mean absolute deviations

c c c c
max T (K) max y1 max y2 max y3 ıT (K)a ıy1 b ıy2 b ıy3 b

Wilson 0.55 0.011 0.011 0.011 0.36 0.005 0.006 0.005

NRTL 0.18 0.013 0.010 0.012 0.07 0.005 0.003 0.006
UNIQUAC 0.55 0.011 0.011 0.010 0.36 0.005 0.006 0.004
 exp 
n 
a
ıT = (1/n)˙i=1 Ti − T cal , where n is the number of data points for the ternary system.
 exp i 
n 
b
ıy = (1/n)˙i=1 yi − yical , where n is the number of data points for the ternary system.
c
max T and max y are the maximum deviations of temperature and vapor composition, respectively.

and vapor phase mole fraction. The experimental VLE data and the the maximum absolute deviations and mean absolute deviations
calculated data with the correlated binary interaction parameters of equilibrium temperature and vapor mole fraction are listed in
are shown in Figs. 3 and 4. Table 10. The results show that the models predict the VLE data of
The binary interaction parameters obtained with NRTL, Wil- the ternary system with reasonable deviations.
son and UNIQUAC models were used to predict the ternary VLE The residue curves of the ternary system predicted by NRTL
data of sec-butyl alcohol (1) + sec-butyl acetate (2) + DMSO (3). And model and the experimental data were given in Fig. 5. The figure
shows that the lines connecting the liquid points and their cor-
responding vapor points are tangent with the residue curves at
the liquid points, which implies that the predicted residue curves

Fig. 5. Residue curve of the ternary system sec-butyl alcohol (1) + sec-butyl acetate
(2) + DMSO (3) (䊏, experimental liquid phase composition taken from Table 6; ,
experimental vapor phase composition taken from Table 6; solid lines, residue
curves, which are calculated by NRTL model with its parameters in Table 9 and Fig. 6. y1 –x1 vs x1 diagram for the comparison of VLE behavior of binary system sec-
gas-phase model is Soave–Redlich–Kwong (SRK) equation; dash lines, pairs of VLE butyl alcohol (1) + sec-butyl acetate (2), with and without DMSO (䊏, experimental
data). data with DMSO; , literature data without DMSO [4]).
60 X. Zhang et al. / Fluid Phase Equilibria 406 (2015) 55–60

can represent the experimental data well [17]. The VLE data of the References
binary system sec-butyl alcohol (1) + sec-butyl acetate (2) in the lit-
erature [4] and that in the ternary system excluding DMSO are [1] X.M. Zhang, Y.X. Liu, C. Jian, F. Wei, H.P. Liu, Fluid Phase Equilib. 383 (2014)
5–10.
shown in Fig. 6. It indicates that the relative volatility of binary [2] X.M. Zhang, Y.X. Liu, J. He, W. Yang, H.P. Liu, C.H.G. Jian, Fluid Phase Equilib.
mixture sec-butyl alcohol + sec-butyl acetate is changed obviously 394 (2015) 83–87.
when the mass ratio of DMSO to the binary mixture is 6:5. [3] L. Berg, U.S. Patent 4,718,988, January 12, 1988.
[4] H.X. Wang, J.J. Xiao, Y.Y. Shen, C.S. Ye, L. Li, T. Qiu, J. Chem. Eng. Data 58 (2013)
1827–1832.
4. Conclusion [5] L. Yuan, L. Wang, P. Bai, J. Chem. Eng. Data 58 (2013) 3216–3221.
[6] Q. Li, F. Xing, Z. Lei, B. Wang, Q. Chang, J. Chem. Eng. Data 53 (2007) 275–279.
[7] X.M. Zhang, J. Zuo, C.G. Jian, J. Chem. Eng. Data 55 (2010) 4896–4902.
The isobaric VLE data for the binary systems sec-butyl alcohol [8] K. Kurihara, M. Nakamichi, K. Kojima, J. Chem. Eng. Data 38 (1993) 446–449.
(1) + DMSO (3), sec-butyl acetate (2) + DMSO (3) and ternary system [9] M.H. Chen, D.Z. Cong, T.N. Fang, M.Z. Qi, Principles of Chemical Engineering,
sec-butyl alcohol(1) + sec-butyl acetate (2) + DMSO (3) were mea- 2nd ed., Chemical Industry Press, Beijing, 2009.
[10] R.R. Chen, H. Ding, M.C. Liu, J.L. Qi, H. Zhou, N. Chen, Fluid Phase Equilib. 382
sured at 101.3 kPa using a modified Othmer still. All of the binary (2014) 133–138.
VLE data passed the Herington test, Wisniak’s L–W method and a [11] J. Wisniak, A new test for the thermodynamic consistency of vapor–liquid
point to point method of Van Ness test described by Fredenslund equilibrium, Ind. Eng. Chem. Res. 32 (1993) 1531–1533.
[12] H.C. Van Ness, S.M. Byer, R.E. Gibbs, AIChE J. 19 (1973) 238–244.
et al. And the ternary VLE data passed the Van Ness test. The iso-
[13] A. Fredenslund, J. Gmehling, P. Rasmussen, Vapor–liquid Equilibria Using
baric VLE data of the binary systems were correlated with Wilson, UNIFAC: A Group Contribution Method, Elsevier, Amsterdam, 1977.
NRTL, and UNIQUAC models and binary interaction parameters [14] P.L. Jackson, R.A. Wilsak, Thermodynamic consistency tests based on the
were obtained. And the VLE data for the binary and ternary sys- Gibbs–Duhem equation applied to isothermal, binary vapor–liquid
equilibrium data: data evaluation and model testing, Fluid Phase Equilib. 103
tems predicted by the obtained interaction parameters are in good (1995) 155–197.
agreement with the experimental data. The relative volatility of [15] J.W. Kang, V. Diky, R.D. Chirico, J.W. Magee, C.D. Muzny, I. Abdulagatov, A.F.
binary mixture sec-butyl alcohol + sec-butyl acetate can be changed Kazakov, M. Frenkel, Quality assessment algorithm for vapor–liquid
equilibrium data, J. Chem. Eng. Data 55 (2010) 3631–3640.
obviously when the mass ratio of DMSO to the binary mixture is 6:5. [16] Aspentech, Aspen Property System: Physical Property Methods and Models
Thus, DMSO is a promising extractive agent for separation of the 11.1, Aspen Technology, Inc., Burlington, MA, 2001.
binary mixture of sec-butyl alcohol + sec-butyl acetate in extractive [17] S. Widagdo, W.D. Seider, AIChE J. 42 (1996) 96–130.
distillation.