Fluid Phase Equilibria 305 (2011) 101–105

Contents lists available at ScienceDirect

Fluid Phase Equilibria

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

Liquid–liquid equilibria of water + 3-hydroxy-2-butanone + ethyl ethanoate

Yan-Yang Wu a , Kui Chen a,∗ , De-Tao Pan a , Jia-Wen Zhu a , Bin Wu a , Ya-Ling Shen b
a
Chemical Engineering Research Center, East China University of Science & Technology, Shanghai 200237, China
b
State Key Laboratory of Bio-Reactor Engineering, East China University of Science & Technology, Shanghai 200237, China

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

Article history: (Liquid–liquid) equilibrium (LLE) data of the solubility curves and tie-line compositions have been
Received 13 January 2011 determined for mixtures of (water + 3-hydroxy-2-butanone + ethyl ethanoate) at 298.15 K, 308.15 K
Received in revised form 6 March 2011 and 318.15 K and 101.3 kPa. Distribution coefficients and separation factors have been evaluated for
Accepted 8 March 2011
the immiscibility region. The reliability of the experimental tie-lines has been confirmed by using
Available online 21 March 2011
Othmer–Tobias correlation. The LLE data of the ternary systems have been predicted by UNIFAC method.

Keywords:
© 2011 Elsevier B.V. All rights reserved.
Liquid–liquid equilibria
Phase equilibria
3-hydroxy-2-butanone
Ethyl ethanoate

1. Introduction method. No organic solvent has been investigated and reported

3-Hydroxy-2-butanone, also known as acetoin or acetylmethyl-
carbinol, is a valuable flavor existing naturally in corn, kidney beans,
peas, and broccoli [1–3] and also the common intermediate for
diacetyl and 2,3-butanediol synthesis [4]. Acetoin is useful as a food
additive and is an intermediate in chemical synthesis, consumer
demand has increased [5]. Thus, it has become a platform chemical
in recent years and has been included in the top value chemicals
from biomass by the Department of Energy, USA, in 2004 [6].
There are several chemical synthetic methods for 3-hydroxy-
2-butanone preparation [7,8], among which the microbial way
has been attracting much attention nowadays because it is more
environmentally friendly and more cost-effective than petroleum-
based chemical processes. Its microbial preparation has been
observed in several yeasts and bacteria from various genera such
as Klebsiella, Bacillus, Enterobacter aerogenes [9–14]. The separa-
tion and purification of 3-hydroxy-2-butanone from fermentation
broth, which has been little studied, is essential to realize the indus-
trial production for 3-hydroxy-2-butanone.
Since the fermented liquors contain only a few percent of
3-hydroxy-2-butanone along with complicated impurities which
cause difficulty in separation and since 3-hydroxy-2-butanone
has a higher boiling point than that of water and can form an
azeotrope (∼0.03 by weight of 3-hydroxy-2-butanone) with water
and cannot be distilled out directly [15], extraction from the fer-
mentation liquors by a suitable solvent seems to be a feasible
∗ Corresponding author. Tel.: +86 21 64253914; fax: +86 21 64253914.
E-mail address: chenkui@ecust.edu.cn (K. Chen).
for 3-hydroxy-2-butanone extraction in literature until now. Ethyl
ethanoate used in this study may be a suitable solvent for extrac-
tion of 3-hydroxy-2-butanone from water, being capable of forming
azeotropic mixtures with water to take it from 3-hydroxy-2-
butanone.
The aim of this work is to present the phase behavior of LLE
for the (water + 3-hydroxy-2-butanone + ethyl ethanoate) ternary
system at 298.15 K, 308.15 K and 318.15 K and 101.3 kPa. The tie
lines have also been predicted using the UNIFAC method (a group
contribution method) developed by Fredenslund et al. [16], and
compared with the experimental data.
2. Experimental
2.1. Chemicals
All the chemicals used in this study were purchased from
commercial sources. Ethyl ethanoate was provided by Shanghai
Lingfeng Chemical Reagent Co., Ltd. and had a minimum mass
fraction purity of 0.995. It was used directly without further treat-
ment in this study. Water was distilled twice before utilization.
3-hydroxy-2-butanone was supplied by Sigma–Aldrich with a min-
imum mass fraction purity of 0.98. It exists as its crystalline dimer.
Liquor 3-hydroxy-2-butanone and its crystalline dimer can be
transformed each other through changing the temperature. The
transformation between them has been studied in this research as
follows: firstly, put the crystalline dimer into a clean glass cell with
a water jacket to maintain isothermal conditions. Elevate the tem-
perature in the interval of 1 K gradually from 288.15 K and settle at
each temperature for 24 h. Then observe visually whether the crys-

0378-3812/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.fluid.2011.03.004
102 Y.-Y. Wu et al. / Fluid Phase Equilibria 305 (2011) 101–105

Table 1
Refractive indexes (nD ) at T = 293.15 K and boiling points (Tb ) at 101.3 kPa, of the compounds.

Component nD Tb (K)
a
Experimental Literature Experimental Literaturea

3-Hydroxy-2-butanone 1.4198 1.4171 421.53 421.15

Ethyl ethanoate 1.3701 1.3704 350.24 350.21
Water 1.3325 1.3325 373.30 373.26
a
Taken from Ref. [18].

talline dimmer has changed from to liquor 3-hydroxy-2-butanone
or not and record the transformation temperature. It is found that
the crystalline dimer can be decomposed into liquor 3-hydroxy-
2-butanone completely when being heated above 333.15 K for
8 h, which is in accordance with the literature [17], Whereas the
liquor 3-hydroxy-2-butanone has not been found any crystalliza-
tion when being cooled at 293.15 K within 72 h. Therefore, all
the experiments related to 3-hydroxy-2-butanone in this study
have been carried out within 24 h after the liquor 3-hydroxy-2-
butanone being obtained to make sure 3-hydroxy-2-butanone has
not become its dimer during the experiments.
The purity of these materials was checked and assured by gas
chromatography. The normal boiling point and refractive index
values were measured in this study and reported in Table 1 in com-
parison with the literature data so as to demonstrate the purity
of the compounds. Refractive indexes were measured by an Abbe
refractometer (WZS-I model, made in Shanghai, China), with an
accuracy of ±0.0001. The boiling points were determined by an
Ebuillometer (DZBW model, made in Nanjing, China), with an accu-
racy of ±0.01 K.
2.2. Equilibrium measurements
Three different temperatures (298.15 K, 308.15 K and 318.15 K)
at 101.3 kPa were selected to study the ternary equilibrium system
in order to observe the evaluation of the bi-nodal curves and tie-
lines.
The bi-nodal (solubility) curves were determined by the cloud
point method in an equilibrium glass cell with a water jacket to
maintain isothermal conditions. The temperature in the cell was
kept constant by circulating water from a water bath (SUPER-
CONSTANTTEP BATH, Shanghai precision science instrument Co.,
Ltd.), which is equipped with a temperature controller capable of
maintaining the temperature within ±0.1 K. The major central part
of the solubility curves was obtained by titrating heterogeneous
mixtures of water + ethyl ethanoate with 3-hydroxy-2-butanone
until the turbidity had disappeared. For the water-side and solvent-
side, regions in which the curve and the sides of the triangle
are close and exhibit similar slopes, binary mixtures of either
(water + 3-hydroxy-2-butanone) or (ethyl ethanoate + 3-hydroxy-
2-butanone) were titrated against the third component until the
transition from homogeneity to heterogeneity was observed.
All mixtures were prepared by weighing with a Sartorius
scale accurate to within ±10−4 g. Mutual solubility values of the
(water + ethyl ethanoate) binary were measured using the method
based on the detection of the cloud point [19–21]. The transi-
tion point between the homogeneous and heterogeneous zones
was determined visually. The reliability of the method depends on
the precision of the micro burette with an accuracy of ±0.01 cm3 ,
and is limited by the visual inspection of the transition across
the apparatus. The accuracy of the visual inspection of the tran-
sition is achieved by waiting approximately 5 min in the transition
point and observing the heterogeneity. All visual experiments were
repeated at least three times in order to acquire high accuracy. The
uncertainty in mass fractions was less than ±0.002.
End-point determinations of the tie-lines were based upon the
independent analysis of the conjugate phases that were regarded as
being in equilibrium [19–21]. For this purpose, mixtures of known

Table 2
Experimental bi-nodal curve data (mass fraction Wi ) of {water (1) + 3-hydroxy-2-butanone (2) + ethyl ethanoate (3)} at different temperatures.

T (K) W1 W2 W3 W1 W2 W3 W1 W2 W3

298.15 0.9178 0.0000 0.0822 0.4378 0.3466 0.2156 0.1332 0.3155 0.5513
0.8291 0.0833 0.0876 0.3660 0.3660 0.2680 0.1089 0.3035 0.5876
0.7732 0.1339 0.0929 0.2957 0.3661 0.3382 0.0834 0.2650 0.6516
0.7386 0.1672 0.0942 0.2806 0.3604 0.3590 0.0630 0.2445 0.6925
0.7147 0.1863 0.0990 0.2506 0.3587 0.3907 0.0595 0.2035 0.7370
0.6404 0.2472 0.1124 0.2369 0.3565 0.4066 0.0468 0.1846 0.7686
0.5578 0.2997 0.1425 0.2003 0.3521 0.4476 0.0447 0.1398 0.8155
0.5305 0.3130 0.1565 0.1832 0.3410 0.4758 0.0419 0.0892 0.8689
0.5026 0.3274 0.1700 0.1589 0.3382 0.5029 0.0363 0.0000 0.9637
308.15 0.9019 0.0000 0.0981 0.3857 0.3404 0.2739 0.1684 0.3041 0.5275
0.8505 0.0496 0.0999 0.3693 0.3400 0.2907 0.1402 0.2936 0.5662
0.8013 0.0938 0.1049 0.3458 0.3411 0.3131 0.1204 0.2622 0.6174
0.7349 0.1510 0.1141 0.3267 0.3397 0.3336 0.0985 0.2411 0.6604
0.6935 0.1872 0.1193 0.3012 0.3395 0.3593 0.0879 0.2099 0.7022
0.6597 0.2107 0.1296 0.2806 0.3342 0.3852 0.0709 0.1842 0.7449
0.5647 0.2809 0.1544 0.2465 0.3332 0.4203 0.0691 0.1366 0.7943
0.5252 0.3013 0.1735 0.2214 0.3259 0.4527 0.0677 0.0847 0.8476
0.4529 0.3278 0.2193 0.1915 0.3199 0.4886 0.0661 0.0000 0.9339
318.15 0.8776 0.0000 0.1224 0.4497 0.2974 0.2529 0.2051 0.2852 0.5097
0.8400 0.0324 0.1276 0.4249 0.3019 0.2732 0.1896 0.2750 0.5354
0.8027 0.0664 0.1309 0.4135 0.3018 0.2847 0.1654 0.2660 0.5686
0.7552 0.1080 0.1368 0.3763 0.3038 0.3199 0.1256 0.2296 0.6448
0.7185 0.1394 0.1421 0.3157 0.3018 0.3825 0.1123 0.1985 0.6892
0.6877 0.1621 0.1502 0.2900 0.3021 0.4079 0.0958 0.1860 0.7182
0.6138 0.2244 0.1618 0.2728 0.2994 0.4278 0.0877 0.1393 0.7730
0.5464 0.2634 0.1902 0.2435 0.2967 0.4598 0.0635 0.0691 0.8674
0.4989 0.2871 0.2140 0.2308 0.2913 0.4779 0.0596 0.0000 0.9404
 Y.-Y. Wu et al. / Fluid Phase Equilibria 305 (2011) 101–105 103

Table 3 3-Hydroxy-2-butanone
Experimental tie-line data in mass fractions for water (1) + 3-hydroxy-2-butanone
(2) + ethyl ethanoate (3) ternary system. 0.00
1.00
T/K Organic phase Aqueous phase

W13 W23 W33 W11 W21 W31

0.25
298.15 0.0578 0.2313 0.7109 0.6832 0.2132 0.1036 0.75
0.0720 0.2668 0.6612 0.6277 0.2531 0.1192
0.0511 0.2023 0.7466 0.7106 0.1889 0.1005
0.0545 0.1752 0.7703 0.7341 0.1646 0.1013
0.0456 0.0798 0.8746 0.8377 0.0753 0.0870 0.50
0.50
0.0482 0.1244 0.8274 0.7860 0.1198 0.0942
308.15 0.1063 0.2566 0.6371 0.6388 0.2266 0.1346
0.0717 0.1800 0.7483 0.7275 0.1607 0.1118
0.0704 0.1054 0.8242 0.8024 0.0963 0.1013
0.75
0.0692 0.1613 0.7695 0.7481 0.1422 0.1097 0.25
0.0914 0.2181 0.6905 0.6903 0.1880 0.1217
0.0684 0.0910 0.8406 0.8205 0.0788 0.1007
318.15 0.1771 0.2782 0.5447 0.6129 0.2207 0.1664
0.1183 0.2142 0.6675 0.6707 0.1786 0.1507 1.00
0.00
0.0832 0.1344 0.7824 0.7460 0.1098 0.1442
0.00 0.25 0.50 0.75 1.00
0.1242 0.2310 0.6448 0.6501 0.1915 0.1584
0.0786 0.0804 0.8410 0.7977 0.0704 0.1319 Ethyl ethanoate Water
0.0933 0.1822 0.7245 0.7052 0.1532 0.1416
Fig. 1. Ternary diagram for LLE of (water + 3-hydroxy-2-butanone + ethyl
ethanoate) at 298.15 K: , experimental solubility; —, experimental solubility
masses of water, 3-hydroxy-2-butanone, and ethyl ethanoate lying curve; , experimental tie-line data; , calculated (UNIFAC) tie-line data; . . .,
calculated tie-lines.
within the heterogeneous zone were introduced into the equi-
librium cell and were agitated for 3 h with a magnetic stirrer
vigorously, and then left for 4 h to settle down into raffinate equations:
(aqueous) and extract (solvent) layers. The compositions of liq- Wi3
uid samples withdrawn from conjugate phases were analyzed by Di = (1)
Wi1
a gas chromatograph (GC112A) with a thermal conductivity detec-
tor (TCD), after calibration with gravimetrically prepared standard distribution coefficient of 3-hydroxy-2-butanone D2
S= = (2)
solutions. A GDX-102 packed column (3 m × Ø 3 mm × 0.5 mm) was distribution coefficient of water D1
used to separate components. They were all produced by Shang-
where Wi3 and Wi1 are the mass concentrations of component i in
hai Hengping Scientific Instrument Co., Ltd. The oven, injector
solvent-rich and water-rich phases, and D1 and D2 are the distribu-
and detector temperatures were 433.15 K, 453.15 K and 453.15 K,
tion coefficients of water and 3-hydroxy-2-butanone, respectively.
respectively. High-purity hydrogen (0.999999 purity) was used as
The distribution coefficients and separation factors for each
the carrier gas at a constant flow rate of 30 mL min−1 . The detector
temperature are given in Table 4. Separation factors were found to
was connected to a FJ-2003B integrator. Each sample was analyzed
be much greater than 1, for the systems reported here, which means
at least thee times to ensure accuracy. The uncertainty in mass
that the extraction of 3-hydroxy-2-butanone by ethyl ethanoate is
fractions was within ±0.001.
really possible. The separation factor is not constant over the whole
two-phase region. The extracting power of the solvent at each tem-
3. Result and discussion

The LLE measurements were made for the ternary system 3-Hydroxy-2-butanone
of (water + 3-hydroxy-2-butanone + ethyl ethanoate) at 298.15 K, 0.00
308.15 K and 318.15 K and 101.3 kPa. The experimental binodal 1.00
curves for this ternary system at each temperature are listed in
Table 2, for which Wi refers to the mass fraction of ith compo-
nent. The experimental tie-line compositions of the equilibrium 0.25
phases are shown in Table 3, for which Wi1 and Wi3 refer to the 0.75
mass fractions of the ith component in the aqueous and solvent
phases, respectively.
The experimental and predicted equilibrium data through 0.50
0.50
UNIFAC model of the ternary system at T = 298.15 K are plot-
ted in Fig. 1. As can be seen from Fig. 1, the system exhibited
type 1 phase behavior [22,23], having only one liquid pair
of partially miscible (ethyl ethanoate + water) and two pairs 0.75
0.25
of completely miscible (water + 3-hydroxy-2-butanone) and (3-
hydroxy-2-butanone + ethyl ethanoate). Also, similar results are
observed at T = 308.15 K and T = 318.15 K in Figs. 2 and 3.
The effectiveness of 3-hydroxy-2-butanone extraction by ethyl 1.00
0.00
ethanoate is given by its separation factor, which is a measure of the 0.00 0.25 0.50 0.75 1.00
ability of ethyl ethanoate to separate the 3-hydroxy-2-butanone Ethyl ethanoate Water
from water. To show the selectivity and extraction strength of the
Fig. 2. Ternary diagram for LLE of (water + 3-hydroxy-2-butanone + ethyl
solvent to extract 3-hydroxy-2-butanone, the distribution coeffi-
ethanoate) at 308.15 K: , experimental solubility; —, experimental solubility
cients, Di , for water (i = 1) and 3-hydroxy-2-butanone (i = 2), and curve; , experimental tie-line data; , calculated (UNIFAC) tie-line data; . . .,
the separation factors, S, are calculated according to the following calculated tie-lines.
104 Y.-Y. Wu et al. / Fluid Phase Equilibria 305 (2011) 101–105

3-Hydroxy-2-butanone 2.0

0.00
1.00

1.6

0.25
0.75
1.2

D2
0.50
0.50
0.8

0.75 0.4
0.25

1.00 0.0
0.00 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35
0.00 0.25 0.50 0.75 1.00 W21
Ethyl ethanoate Water
Fig. 4. Distribution coefficient D2 of 3-hydroxy-2-butanone as a function of the mass
Fig. 3. Ternary diagram for LLE of (water + 3-hydroxy-2-butanone + ethyl fraction W21 of 3-hydroxy-2-butanone in aqueous phase: , 298.15 K; , 308.15 K;
ethanoate) at 318.15 K: , experimental solubility; —, experimental solubility , 318.15 K.
curve; , experimental tie-line data; , calculated (UNIFAC) tie-line data; . . .,
calculated tie-lines.
relation factor (R2 ) being approximately unity indicates the degree
of consistency of the related data.
perature, plots of D2 versus W21 and S versus W21 are shown in
The experimental equilibrium data have been compared with
Figs. 4 and 5, respectively.
predicted values by UNIFAC using the interaction parameters
The reliability of experimentally measured tie-line data can be
between CH3 , CH2 , CH, OH, CH3 CO, CH3 COO and H2 O functional
ascertained by applying the Othmer–Tobias correlation [24] at each
groups obtained by Magnussen et al. [25] As shown in Fig. 1, LLE
temperature as below:
data predicted by the UNIFAC method cannot be adequately fitted
1 − W  1 − W  with the experimental LLE data at T = 298.15 K. Similar lack of fits
11 33
ln = a + b × ln (3)
W11 W33 are observed at T = 308.15 K and T = 318.15 K.
The root mean square deviations (RSMD) are calculated from the
where W11 is the mass fraction of water in the water-rich phase, difference between the experimental data and the predictions of
W33 is the mass fraction of ethyl ethanoate in the solvent-rich the UNIFAC model at each temperature according to the following
phase, and a and b are the constants. formula:
The parameters of this correlation are listed in Table 5 and the ⎡
N    2 ⎤1/2
correlation is shown in Fig. 6 for the temperatures studied. The cor- exp
cal
Wijk − Wijk
⎢ k j i ⎥
RSMD = ⎣ ⎦ (4)
6N
Table 4
Distribution coefficients (Di ) of water (i = 1) and 3-hydroxy-2-butanone (i = 2) and
separation factors (S).
where Wijk is the composition of component i in phase j on tie-line k.
T (K) D1 D2 S N is the number of the tie-lines. The UNIFAC model predicts the LLE
298.15 0.0845 1.0847 12.83 data for 298.15 K, 308.15 K and 318.15 K with a root mean square
0.1148 1.0542 9.19
0.0719 1.0709 14.89
25
0.0742 1.0641 14.33
0.0545 1.0589 19.44
0.0614 1.0387 16.93
308.15 0.1664 1.1330 6.81 20
0.0985 1.1200 11.37
0.0877 1.0946 12.48
0.0925 1.1341 12.26 15
0.1323 1.1603 8.77
0.0834 1.1552 13.86
318.15 0.2889 1.2603 4.36 10
S

0.1763 1.1992 6.80

0.1116 1.2244 10.98
0.1912 1.2060 6.31
5
0.0986 1.1412 11.58
0.1324 1.1895 8.98

0
Table 5
Constants of Othmer–Tobias equation for the water + 3-hydroxy-2-butanone + ethyl
ethanoate ternary system (R2 : regression coefficient). -5
2
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
T (K) a b R
W21
298.15 0.0311 0.8594 0.9983
308.15 0.1025 1.1759 0.9954
Fig. 5. Separation factor, S, as a function of the mass fraction of 3-hydroxy-2-
318.15 0.4473 1.5785 0.9861
butanone in the aqueous phase, W21 : , 298.15 K; , 308.15 K; , 18.15 K.
 Y.-Y. Wu et al. / Fluid Phase Equilibria 305 (2011) 101–105 105

1.0 Selectivity diagrams on a solvent-free basis are obtained by

plotting W23 /(W23 + W13 ) versus W21 /(W21 + W11 ) for each tem-
0.5
perature in Fig. 7. The selectivity diagram indicated that the
0.0 performance of the solvent decreases with increasing temperature.

-0.5
4. Conclusions
ln((1-W 11)/W 11)

-1.0
The LLE data of the ternary mixtures water + 3-hydroxy-2-
-1.5 butanone + ethyl ethanoate have been presented at 298.15 K,
308.15 K and 318.15 K. The UNIFAC model has been used to predict
-2.0
the LLE data. It has been observed that the UNIFAC predictions do
-2.5 not fit the experimental results quantitatively, but it agrees qual-
itatively. The separation factor is found to be much greater than
-3.0 1 and it is not constant over the whole two-phase region. It is
concluded that ethyl ethanoate may serve as a feasible solvent to
-3.5
-3.0 -2.4 -1.8 -1.2 -0.6 0.0 0.6 extract 3-hydroxy-2-butanone from its aqueous solutions.
ln((1-W33)/W33)
References
Fig. 6. Othmer–Tobias plots of the (water + 3-hydroxy-2-butanone + ethyl
ethanoate) ternary systems: , 298.15 K; , 308.15 K; , 18.15 K; —, Eq. (3). [1] P.A. Buck, M.A. Joslyn, J. Agric. Food Chem. 4 (1956) 548–552.
[2] G.B. Ron, J.O. William, R.T. Gary, N. Youngla, J. Agric. Food Chem. 47 (1999)
4353–4356.
1.0 [3] J.W. Ralls, J. Agric. Food Chem. 7 (1959) 505–507.
[4] P. Romano, G. Suzzi, Appl. Environ. Microbiol. 62 (1996) 309–315.
[5] Y.F. Liu, S.L. Zhang, Y.C. Yong, Z.X. Ji, X. Ma, Z.H. Xu, S.W. Chen, Process Biochem.
46 (2011) 390–394.
0.8 [6] T. Werpy, G. Petersen, Top Value Added Chemicals from Biomass: Volume I.
Results of Screening for Potential Candidates from Sugars and Synthesis Gas,
2004.
[7] T. Fumio, T. Koichi, H. Tange, J. Chem. Soc. 1 (1989) 1555–1556.
W 23/(W 23+W 13)

0.6 [8] R.H. Blom, J. Am. Chem. Soc. 67 (1945) 494–498.

[9] E.K. Yu, N. Levitin, J.N. Saddler, Biotechnol. Lett. 4 (1982) 741–746.
[10] E.K. Yu, J.N. Saddler, Appl. Environ. Microbiol. 46 (1983) 630–635.
[11] N.B. Jansen, M.C. Flickinger, G.T. Tsao, Biotechnol. Bioeng. 26 (1984) 362–369.
0.4 [12] V.M. Laube, D. Groleau, S.M. Martin, Biotechnol. Lett. 6 (1984) 257–262.
[13] B. Dettwiler, I.J. Dunn, E. Heinzle, J.E. Prenosil, Biotechnol. Bioeng. 41 (1993)
791–800.
[14] Z.J. Xiao, P.H. Liu, J.Y. Qin, P. Xu, Appl. Microbiol. Biotechnol. 74 (2007)
0.2 61–68.
[15] P.P. Giovannini, A. Medici, C.M. Bergamini, M. Rippa, Bioorg.Med.Chem. 4 (1996)
1197–1201.
[16] A. Fredenslund, J. Gmehling, P. Rasmussen, Fluid Phase Equilib. 1 (1977) 317.
0.0 [17] Y.Y. Zeng, S. Lin, H. Chen, The relationship between yield of acetoin-dimer and
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
optical rotation of acetoin-monomer, in: Seventh Symposium on China’s F & F,
W21/(W21+W11) 2008.
[18] R.L. David, CRC Handbook of Chemistry and Physics, 87th ed., CRC Press Inc.,
Fig. 7. Selectivity diagram at investigated temperature values (solvent-free basis): Taylor and Francis, Boca Raton, FL, 2007.
, 298.15 K; , 308.15 K; , 18.15 K. [19] Y.Y. Wu, J.W. Zhu, K. Chen, B. Wu, J. Fang, Y.L. Shen, Fluid Phase Equilib. 265
(2008) 1–6.
[20] Y.Y. Wu, J.W. Zhu, K. Chen, B. Wu, J. Fang, Y.L. Shen, J. Chem. Eng. Data 53 (2008)
deviation of 0.0969, 0.0947 and 0.1100, respectively, between the 559–563.
[21] Y.Y. Wu, J.W. Zhu, K. Chen, B. Wu, J. Fang, Y.L. Shen, Fluid Phase Equlib. 266
observed and calculated mass concentrations. This value indicates
(2008) 42–46.
that the LLE data of this ternary system did not predicted well [22] A. Arce, A. Blanco, P. Souza, J. Chem. Eng. Data 38 (1993) 201–203.
with this equilibrium model. As it can be seen from Figs. 1–3, the [23] H. Ghanadzadeh, A.K. Haghi, Fluid Phase Equilib. 243 (2006) 45–50.
predicted tie lines (dashed lines) are relatively in poor agreement [24] E. Isabel, J. Chem. Eng. Data 39 (1994) 834–839.
[25] T. Magnussen, P. Rasmussen, A. Fredenslund, Ind. Eng. Chem. Process Des. Dev.
with the experimental data (solid lines). However, in the lack of 20 (1981) 331–339.
experimental data this model can be used for correlation.