Measurement and Correlation of Solubility of Benzothiazolium Ionic Liquids in Ethanol + Ethyl Benzoate

Article abstract of DOI:10.1021/acs.jced.9b00997

The solubilities of benzothazolium-based ionic liquids (ILs) [HBth][BF₄], [HBth][CH₃SO₃], and [HBth][p-TSA] in binary ethanol + ethyl benzoate solvents were measured using the equilibrium method with the temperature range

Full text of DOI:10.1021/acs.jced.9b00997

pubs.acs.org/jced
Article
Measurement and Correlation of Solubility of Benzothiazolium Ionic
Liquids in Ethanol + Ethyl Benzoate
Tian Yao and Kaifeng Du*
Cite This: https://dx.doi.org/10.1021/acs.jced.9b00997
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: The solubilities of benzothazolium-based ionic liquids (ILs) [HBth][BF4], [HBth][CH3SO3], and [HBth][p-TSA] in
binary ethanol + ethyl benzoate solvents were measured using the equilibrium method with the temperature range from 233.2 to
293.7 K at atmospheric pressure. Results of these measurements were correlated by both λh equation and the modified Apelblat
equation. It was found that both models gave satisfactory correlation results. In the λh model, the average relative deviations were
0.7145, 0.7845, and 0.8119% for [HBth][BF4], [HBth][CH3SO3], and [HBth][p-TSA], respectively. Interestingly, the parameters of
λ and h for the λh model and A, B, and C for the modified Apelblat model were simultaneously expressed as functions of solvent
composition (xsol) for the first time. Furthermore, solubilities of these three ILs in the binary system were predicted through these
two sets of self-created correlation equations with ideal performances. The total average relative deviations were less than 2.2% for
both models. These successful in-depth correlations and validated prediction results have guaranteed the accuracy of experimental
solubility data and provide us a reliable method for accurate solubility prediction at any temperature with any solvent composition.
This developed calculation method in this report lays a foundation for accurate industrial recovery of ILs and gives us valuable
guidance for the design of the specific purification process to acquire desired purity of the product in industrial production.
Song et al.18 reported the first esterification of aromatic acid with
alcohols catalyzed by “temperature-sensitive” ILs. These
INTRODUCTION
■
Aromatic esters are very important chemical intermediates with
functionalized ILs including [HBth][BF4], [HBth][CH3SO3],
wide applications in the chemical industry and are usually
synthesized through acid-catalyzed esterification reactions1 or
and [HBth][p-TSA] had very large solubility differences with
only small-scale changes of temperature in many alcohols.19
synthesized through enzymatic esterification reactions2−4 with
alcohols and benzoic acids as raw materials. Inorganic acids such
They exhibited excellent catalytic performances and ideal
recoveries. Among them, esterification yield of benzoic acid
could reach up to 97.3% in ethanol with [HBth][CH3SO3] as
the catalyst, which is showing great industrial application
as H3PO4, H2SO4, and so forth, have been widely applied as
catalysts for the esterification.5−9 Although high yields were
acquired, usage of a large amount of organic solvents and
difficulty in the catalyst recovery could result in environmental
potential. Although this temperature-sensitive IL-catalyzed
esterification of benzoic acid in ethanol had excellent perform-
ances, there still exists some amount of ILs dissolved in the
ethanol + ethyl benzoate solution after cooling down to room
temperature at the end of the reaction. Due to the high cost of IL
and high requirement of product purity,20 it is vitally important
contamination. Thus, it is necessary to develop novel environ-
mentally friendly, efficient, and recyclable catalysts.
Ionic liquids (ILs), a class of nonmolecular solvents with low
melting points, negligible volatility, wide liquid range, and good
chemical stability and solubility,10,11 have drawn widespread
attention of researchers. In the field of synthetic chemistry, they
often replace hazardous organic solvents and serve as an
environmentally friendly, nonvolatile, and recyclable reaction
medium.12 Many researchers have already successfully applied
ILs in the esterification with reduced reaction time and
temperature as well as satisfactory recyclability.13−17 Later,
Received: October 18, 2019
Accepted: December 27, 2019
https://dx.doi.org/10.1021/acs.jced.9b00997
J. Chem. Eng. Data XXXX, XXX, XXX−XXX
© XXXX American Chemical Society
A

Journal of Chemical & Engineering Data
pubs.acs.org/jced
Article
Figure 1. Synthetic routes and molecular structures of [HBth][BF4], [HBth][CH3SO3], and [HBth][p-TSA].
Table 1. Provenance and Purity of the Chemicals Used in This Work
chemical name
source
prepared in lab
prepared in lab
initial mole fraction purity
purification method
final mole fraction purity analysis method
[HBth][BF4]
[HBth][CH3SO3]
[HBth][p-TSA]
ethyl benzoate
ethanol
0.95
0.96
0.95
≥0.995
≥0.995
≥0.995
developed in our lab
developed in our lab
developed in our lab
no further purification
no further purification
no further purification
≥0.99
≥0.99
≥0.99
HPLC
HPLC
HPLC
prepared in lab
Kelon Chemical Reagent Co., Ltd.
Kelon Chemical Reagent Co., Ltd.
Kelon Chemical Reagent Co., Ltd.
≥0.995
≥0.995
≥0.995
ethyl acetate
to further crystallize IL out for recycle by cooling the system
down to very low temperature. Therefore, accurate solubility
data of these ILs in this binary system at low temperature is of
vital importance for the guidance of industrial separation.
Herein, we report the solubility data of three ILs ([HBth]-
[BF4], [HBth][CH3SO3], and [HBth][p-TSA]) in the binary
ethanol + ethyl benzoate solvents at low temperature ranging
from 233.2 to 293.7 K at atmospheric pressure. These data were
ideally fitted by the λh equation and the modified Apelblat
equation, respectively. Surprisingly, for the first time, parameters
of λ and h for the λh model and A, B, and C for the modified
Apelblat model were simultaneously expressed as functions of
solvent composition (xsol) with satisfactory correlation co-
efficients (R2). Furthermore, solubilities of these three ILs at any
solvent composition and any temperature could be accurately
predicted through both models.
Corporation, Germany) in DMSO-d6 with tetramethylsilane
(TMS) as an internal standard. 1H NMR spectra were recorded
at 400 MHz, and 13C NMR spectra were recorded at 100 MHz.
The electrospray ionization mass spectrometry (ESI-MS)
spectra were recorded on a ZQ 4000 mass spectrometer
(Waters Corporation, USA). An LC20-AT high-performance
liquid chromatography (HPLC) instrument (Shimadzu Co.,
Japan) consisting of a diode array detector was used for the
analysis and quantitation of targeted IL. A Welchrom C18
column (4.6 × 250 mm, 5 μm) was used as the HPLC analytical
column.
Synthesis and Purification of ILs. The synthesis of ILs was
according to the reported literature.19 Briefly, 0.5 mol accurate
amount of benzothiazole was dissolved in 100 mL of anhydrous
ethanol and placed in a round-bottom flask equipped with a
magnetic stirrer in an ice bath. Equal molar of fluoroboric acid
(methanesulfonic acid or p-toluenesulfonic acid) dissolved in
about 40 mL of ethanol was added dropwise into the round-
bottom flask under adequate stirring for about 1 h. Afterward,
the mixture was continuously stirred for 12 h at room
temperature. Ethanol was removed by evaporation under
vacuum to give the crude product as a white solid. It was then
washed three or four times by ethyl acetate and recrystallized in
anhydrous ethanol to give the final pure product, which was
dried for more than 12 h under vacuum at 343 K. The ILs were
characterized by 1H NMR, 13C NMR, and ESI-MS (the detail is
shown in the Supporting Information). The synthetic route of
ILs is shown in Figure 1. Melting points were determined as 398
K for benzothiazolium tetrafluoroborate ([HBth][BF4]), 378 K
for benzothiazolium methanesulfonate ([HBth][CH3SO3]),
and 393 K for benzothiazolium p-toluene sulfonate ([HBth]-
[p-TSA]).19 All these three purified ILs have mole fraction
purities higher than 99% determined by HPLC. The provenance
of main used chemicals and their purities are given in Table 1.
Solubility Measurement. The solubility of ILs was
measured by a static equilibrium method as presented in Figure
2. The concentration determination of ILs was conducted by the
EXPERIMENTAL SECTION
■
Chemicals and Apparatus. Benzothiazole (CAS number
95-16-9) of analytical grade was supplied by Xilong Chemical
Factory in Shantou City, Guangdong. Tetrafluoroboric acid
(CAS number 16872-11-0), methanesulfonic acid (CAS
number 75-75-2), ethyl benzoate (CAS number 93-89-0), p-
toluenesulfonic acid (CAS number 104-15-4), ethyl acetate
(CAS number 141-78-6), anhydrous ethanol, and so forth, were
purchased from Chengdu Kelon Chemical Reagent Co., Ltd.,
with analytical grade or above. All used water was ultrapure
water obtained from an ultrapure water purification system (0.4
mm filter) manufactured by Millipore Co., Ltd. (Bedford, MA,
USA). A RE-3000B rotary evaporator (Shanghai Yarong
Biochemical Instrument Co., Ltd., China) was applied for
vacuum evaporation. Melting points were measured by an XRC-
1 melting point apparatus (Sichuan University Instrument Co.,
Ltd.). Mass values were measured by an electronic balance
(ESJ200-4A, Longteng Electronics Co., Ltd., Shenyang, China)
1
with an uncertainty of 0.0001 g. H NMR and 13C NMR
spectra were recorded on an AV-400 spectrometer (Bruker
B
https://dx.doi.org/10.1021/acs.jced.9b00997
J. Chem. Eng. Data XXXX, XXX, XXX−XXX

Journal of Chemical & Engineering Data
pubs.acs.org/jced
Article
where MA, M1, and M2 represent the molecular weights of the
solute, ethyl benzoate, and ethanol, respectively. xsol is the molar
fraction of ethyl benzoate in the binary solvent system. x is the
molar fraction solubility of IL.
The solubilities of three benzothiazolium-based ILs were
determined in the temperature range of 233.2 to 293.7 K. First of
all, we determined the solubilities in ethanol and compared these
results with the previous literature19 in the Supporting
Information. Then, the solvent compositions (xsol) were defined
as 0.25, 0.33, 0.50, 0.67, and 1.00 in the main text. As
demonstrated in Tables 2−4, solubility data of these ILs in
different solvent compositions were measured and correlated by
the λh equation and the modified Apelblat equation. The λh
model is a semiempirical equation proposed by Buchowski et
al.21 and is given as eq 4.
Figure 2. Apparatus for determination of solubilities of [HBth][BF4],
[HBth][CH3SO3], and [HBth][p-TSA]. (1) Preprogrammed auto-
matic temperature controller, (2) thermostat with ethanol as medium,
(3) rubber plug, (4) mercury thermometer, (5) sample connection, (6)
crystallizer, and (7) magnetic stirrer.
Ä
Å
Å
Å
Å
É
Ñ
Ñ
Ñ
Ñ
i
j
j
y
z
z
z
1
x
1
T
1
Tm
i
j
j
j
y
z
z
ln 1 + λ
− 1 = λhj
−
Å
Ñ
Ñ
Ñ
z
Å
Å
j
j
z
z
Å
Å
Ç
Ñ
Ñ
Ö
k
{
(4)
k
{
HPLC method. First, HPLC separation conditions were
investigated and determined. Then, a certain amount of pure
IL was accurately weighed and dissolved in certain solvent in a
100 mL volumetric flask. Afterward, gradient solutions were
prepared by diluting the original one and injected into the
HPLC in turn to generate the chromatogram. The external
standard curve was obtained by linear regression with the peak
area as the vertical coordinate and concentration of IL as the
horizontal coordinate. Afterward, solvents with different
compositions were carefully prepared and added into a
crystallizer (self-made). A thermostatic refrigeration recycler
(Gongyi Yuhua Instrument Co., Ltd., China) with an
uncertainty of 0.05 K was started to keep the system at a
desired constant temperature. An excess amount of pure IL was
subsequently added in and stirred for at least 4 h to reach the
equilibrium. The solution was then settled for more than 2 h to
guarantee that extra IL was completely precipitated at the
bottom of the solution. A little drop of upper clear liquid was
carefully withdrawn by a pipette into a previously weighed EP
tube with the main body of liquid not disturbed as possible as we
could. Then, the solution-containing EP tube was accurately
weighed. The solution mass (m) equaled to the total mass minus
the mass value of the empty EP tube. The mass values of solvents
ethyl benzoate and ethanol were defined as m1 and m2,
respectively. The solution was then diluted into a certain
volume and analyzed by HPLC to acquire the solute mass (mA)
from an external standard method. The optimal HPLC
conditions on the C18 column were determined as the mobile
phase of water−methanol (50:50, v/v) at a flow rate of 1.0 mL
min−1 with 20 μL injection volume. The column effluents were
monitored at 284 nm for [HBth][p-TSA] and 250 nm for
[HBth][BF4] and [HBth][CH3SO3], which were the maximum
UV adsorption wavelengths of them.
where x is the molar fraction solubility of IL, T represents the
absolute temperature of the system, and Tm is the melting point
of the solute, which was previously experimentally determined. λ
and h are two adjustable parameters obtained from correlation of
the experimental solubility data. The value of λ can reflect the
nonideal feature of the system and could be interpreted as the
average association degree of the solute molecule in the solution
system, whereas h is related to the enthalpy of solution.22
The modified Apelblat equation23−25 is shown as eq 5.
B
T
ln x = A +
+ C ln T
(5)
where x and T similarly represent molar fraction solubility of IL
and absolute temperature of the system, respectively. Parameters
A, B, and C were empirical constants obtained from correlation
of experimental solubility data. The values of A and B reflect the
variation in the solution activity coefficient and provide an
indication of the effect of solution nonideality on the solute
solubility.26
The experimental solubility data shown in Tables 2−4 could
be ideally correlated by both λh model and Apelblat model. The
calculated solubility data and the relative deviations (RDs)
defined by eq 6 are also presented in Tables 1−3. The average
relative deviation (ARD) is defined by eq 7. For the λh model,
the ARDs were 0.7145% for [HBth][BF4], 0.7845% for
[HBth][CH3SO3], and 0.8119% for [HBth][p-TSA]. Intui-
tional comparisons between experimental data and the
calculated ones are shown in Figures 3−5. All model parameters,
correlation coefficients (R2), and root-mean-square deviations
(rmsd) defined by eq 8 were obtained and are listed in Table 5.
|xexp − xcal|
RD =
× 100%
xexp
(6)
(7)
RESULTS AND DISCUSSION
The molar fraction solubility was determined by eqs 1−3.
N
■
|xiexp − xical|
1
N
ARD =
× 100%
∑
xiexp
i
m − mA = m1 + m2
(1)
Ä
É
Ñ
0.5
m1
Å
N
(xexp − xcal)2 Ñ
Å
Å
Å
Ñ
Ñ
M1
Å
Ñ
rmsd = Å
Ñ
Ñ
Ñ
Ñ
Ñ
∑
xsol
=
Å
Å
Å
Å
m1
M1
m2
M2
N
+
(8)
i=1
Å
Ç
Ñ
Ö
(2)
(3)
As can be seen from the curves in Figures 3−5, [HBth][BF4],
[HBth][CH3SO3], and [HBth][p-TSA] had similar solubility
values in the binary ethanol + ethyl benzoate systems. Generally,
these three ILs were slightly soluble in ethanol and almost
mA
MA
m1
x =
mA
MA
m2
M2
+
+
M1
C
https://dx.doi.org/10.1021/acs.jced.9b00997
J. Chem. Eng. Data XXXX, XXX, XXX−XXX

Journal of Chemical & Engineering Data
pubs.acs.org/jced
Article
D
https://dx.doi.org/10.1021/acs.jced.9b00997
J. Chem. Eng. Data XXXX, XXX, XXX−XXX

Journal of Chemical & Engineering Data
pubs.acs.org/jced
Article
E
https://dx.doi.org/10.1021/acs.jced.9b00997
J. Chem. Eng. Data XXXX, XXX, XXX−XXX

Journal of Chemical & Engineering Data
pubs.acs.org/jced
Article
F
https://dx.doi.org/10.1021/acs.jced.9b00997
J. Chem. Eng. Data XXXX, XXX, XXX−XXX

Journal of Chemical & Engineering Data
pubs.acs.org/jced
Article
insoluble in ethyl benzoate. The solubilities of them gradually
increased when the molar ratio of ethanol to ethyl benzoate
increased. For all cases, the solubility values have a decreasing
trend in the temperature dropping process. Considering this
report and the previous literature,19 it is obvious that the
solubilities of all the three ILs decreased more sharply and were
more temperature-sensitive with the drop of temperature when
the molar ratio of ethanol in the binary solvents was higher. If
pure ethyl benzoate was served as the solvent, solubility values
were low and not sensitive to temperature.
From an industrial point of view, the system temperature of
the esterification reaction was necessarily required to drop below
room temperature and even further to crystallize out more
considerable amount of IL. Also, higher purity of the ethyl
benzoate product was achieved with more recovery of IL
catalysts. These experimental solubility data could give us
valuable guidance in the practical purification process.
Interestingly, parameters of λ and h (λh model) can be
described as a function of solvent composition (xsol) for
[HBth][BF4], [HBth][CH3SO3], and [HBth][p-TSA]. This is a
wonderful linear relationship between the logarithm value of λ
(or h) and xsol. For all three ILs (Figure 6), which are
demonstrated as eqs 9 and 10 for [HBth][BF4], eqs 14 and 15
for [HBth][CH3SO3], and eqs 19 and 20 for [HBth][p-TSA]
with all the correlation coefficients (R2) higher than 0.999.
Surprisingly, parameters of A, B, and C (modified Apelblat
model) can also be expressed as a function of xsol for these three
ILs (Figure 7). A, B, and C were quadratically correlated with w
with satisfactory performances. The functional expressions are
presented as eqs 11−13 for [HBth][BF4], eqs 16−18 for
[HBth][CH3SO3], and eqs 21−23 for [HBth][p-TSA].
Similarly, excellent quadratic fittings are acquired with all the
correlation coefficients (R2) higher than 0.995. To the best of
our knowledge, this is the first time to simultaneously express the
parameters of λ and h (λh model) and the parameters of A, B,
and C (modified Apelblat model) as functions of solvent
compositions (xsol) with ideal performances. With these self-
created in-depth correlations in hand, we were able to predict
the solubility of IL in these binary solvents at any temperature
with any solvent composition.
Figure 3. Experimental molar fraction solubilities of [HBth][BF4] in
ethanol + ethyl benzoate (xsol = 0.25, 0.33, 0.50, 0.67, and 1.00). Dashed
lines represent the correlation results based on the λh model.
Figure 4. Experimental molar fraction solubilities of [HBth][CH3SO3]
in ethanol + ethyl benzoate (xsol = 0.25, 0.33, 0.50, 0.67, and 1.00).
Dashed lines represent the correlation results based on the λh model.
The relationships between model parameters and xsol for
[HBth][BF4] are listed as follows:
ln λ = −7.1596xsol − 0.72113 (R2 = 0.9995)
(9)
ln h = 5.5425xsol + 8.8223 (R2 = 0.9992)
(10)
A = 51.494xsol − 106.72xsol − 5.6065 (R2 = 0.9985)
2
(11)
2
B = −4108.9xsol + 8054.4xsol − 2830.7 (R2 = 0.9960)
(12)
2
C = −6.8657xsol + 13.511xsol + 2.0395 (R2 = 0.9986)
(13)
The relationships between model parameters and xsol for
[HBth][CH3SO3] are listed as follows:
ln λ = −3.8343xsol − 3.0324 (R2 = 0.9994)
(14)
ln h = 4.3737xsol + 10.046 (R2 = 0.9999)
Figure 5. Experimental molar fraction solubilities of [HBth][p-TSA] in
ethanol + ethyl benzoate (xsol = 0.25, 0.33, 0.50, 0.67, and 1.00). Dashed
lines represent the correlation results based on the λh model.
(15)
2
A = 14.077xsol − 9.7048xsol − 61.728 (R2 = 0.9993)
(16)
G
https://dx.doi.org/10.1021/acs.jced.9b00997
J. Chem. Eng. Data XXXX, XXX, XXX−XXX

Journal of Chemical & Engineering Data
pubs.acs.org/jced
Article
a c
−
Table 5. Model Parameters for Three ILs at Different Solvent Compositions (xsol)
λh equation
modified Apelblat equation
xsol
λ
h
R2
105 rmsd
[HBth][BF4]
A
B
C
R2
105 rmsd
0.25
0.33
0.5
0.67
1
0.08610
0.04300
0.01350
0.004014
0.0003810
26,600
44,800
103,140
278,100
1,750,000
0.9996
0.9993
0.9996
0.9996
0.9995
3.716
4.316
1.471
1.309
0.2433
−28.670
−35.763
−46.096
−53.775
−60.890
−1114.11
−576.339
200.990
675.790
1125.098
4.943
5.815
7.073
7.990
8.691
0.9996
0.9994
0.9997
0.9996
0.9995
3.516
3.894
1.121
1.338
0.2562
[HBth][CH3SO3]
0.25
0.33
0.5
0.67
1
0.01811
0.01390
0.007242
0.003581
0.001050
69,410
96,920
203,100
439,000
1,823,000
0.9995
0.9994
0.9994
0.9996
0.9995
2.95
−63.245
−63.414
−63.123
−61.848
−57.368
921.129
888.857
744.317
564.762
100.006
9.787
9.769
9.666
9.408
8.619
0.9997
0.9994
0.9995
0.9996
0.9995
1.858
2.875
2.166
1.009
0.193
2.973
1.751
1.084
0.1805
[HBth][p-TSA]
0.25
0.33
0.5
0.67
1
0.0186
0.0122
0.00453
0.00164
0.000272
64,000
95,900
241,200
602,700
3,397,000
0.9993
0.9993
0.9994
0.9994
0.9995
4.464
3.034
2.072
0.7000
0.1147
−52.554
−54.055
−56.588
−58.411
−60.858
592.358
653.376
755.976
819.657
919.133
8.089
8.246
8.477
8.610
0.9992
0.9992
0.9994
0.9994
0.9995
4.705
3.348
2.006
0.6840
0.1153
8.680
a
c
b
λ and h are parameters of the λh equation, and A, B, and C are parameters of the modified Apelblat equation. R2 is the correlation coefficient.
Ä
Å
Å
É
0.5
Ñ
Ñ
2
N
(xexp − xcal
)
Å
Å
Ñ
Ñ
Ñ
Ñ
rmsd = ∑
Å
Å
i=1
N
Å
Ç
Ñ
Ö
Figure 6. Linear fitting of ln λ and ln h with xsol.
H
https://dx.doi.org/10.1021/acs.jced.9b00997
J. Chem. Eng. Data XXXX, XXX, XXX−XXX

Journal of Chemical & Engineering Data
pubs.acs.org/jced
Article
Figure 7. Quadratic fitting of modified Apelblat model parameters (A, B, and C) with xsol.
2
B = −749.67xsol − 167.64xsol + 1016.4 (R2 = 0.9994)
and T were selected to conduct the solubility measurement and
calculate the prediction values through two different models for
all three ILs. The prediction results are presented in Table 6. It
can be seen that the prediction values obtained from two
different ways of estimation are close and both have satisfactory
accuracy. The relative deviations were in acceptable ranges of
0.2541−3.309% for [HBth][BF4], 0.4232−1.234% for [HBth]-
[CH3SO3], and 0.1941−1.582% for [HBth][p-TSA] for both
models. This prediction accuracy could meet the demand of
industrial crystallization of [HBth][BF4], [HBth][CH3SO3],
and [HBth][p-TSA].
(17)
2
C = −2.0553xsol + 1.009xsol + 9.6636 (R2 = 0.9993)
(18)
The relationships between model parameters and xsol for
[HBth][p-TSA] are listed as follows:
ln λ = −5.6717xsol − 2.5627 (R2 = 0.9995)
ln h = 5.3133xsol + 9.7334 (R2 = 0.9999)
(19)
(20)
To summarize, excellent in-depth correlations and prediction
performances were obtained for three benzothiazolium ionic
liquids [HBth][BF4], [HBth][CH3SO3], and [HBth][p-TSA].
As far as we know, this is the first report to simultaneously
correlate two sets of model parameters with solvent
composition. Consequently, the same set of physical property
data were satisfactorily predicted by two different methods.
Results obtained from two different models were highly
consistent with each other. They greatly reduced the uncertainty
of estimation and increased the credibility of the prediction
value from a theoretical and logical point of view. Previously, the
λh equation and the modified Apelblat equation were also
applied to correlate the solubility data of easy-made imidazolium
ILs20,27 and pyridinium ILs28 in many solvents, but none of them
could generate further correlation for widespread prediction.
2
A = 9.2364xsol − 22.503xsol − 47.571 (R2 = 0.9990)
(21)
B = −365.60xsol + 885.27xsol + 397.96 (R2 = 0.9962)
2
(22)
C = −1.4271xsol + 2.5568xsol + 7.5480 (R2 = 0.9972)
2
(23)
To validate our calculation method, solubilities of [HBth]-
[BF4], [HBth][CH3SO3], and [HBth][p-TSA] at certain
temperature with certain solvent composition (xsol) in the
range of 233.2 to 293.7 K were predicted through in-depth
correlations of both λh model (eqs 9, 10, 14, 15, 19, and 20) and
Apelblat model (eqs 11−13, 16−18, and 21−23) at the same
time. Seven solubility-unknown points with given values of xsol
I
https://dx.doi.org/10.1021/acs.jced.9b00997
J. Chem. Eng. Data XXXX, XXX, XXX−XXX

Journal of Chemical & Engineering Data
pubs.acs.org/jced
Article
a c
−
for [HBth][BF4], 0.4232−1.234% for [HBth][CH3SO3], and
0.1941−1.582% for [HBth][p-TSA] for both models. The total
average relative deviations were less than 2.2% for both models.
These successful in-depth correlations and validated prediction
results have guaranteed the accuracy of experimental solubility
data and provide us a reliable method for accurate solubility
prediction at any temperature with any solvent composition.
The developed calculation method in this report could give us
valuable guidance for the design of the specific purification
process to acquire desired purity of the product in industrial
production.
Table 6. Solubility Predictions for ILs at 100 kPa
modified Apelblat
λh equation
equation
x
sol (%) T (K) xexp × 103 xcal × 103 RD (%) xcal × 103 RD (%)
[HBth][BF4]
0.30
0.40
0.50
0.60
0.70
0.80
0.90
235.0
245.0
255.0
265.0
275.0
285.0
295.0
1.657
1.968
1.959
1.783
1.452
1.094
0.7986
1.695
1.973
1.953
1.724
1.404
1.081
0.8014
2.293
0.2541
0.3063
3.309
3.306
1.188
0.3506
1.605
1.960
2.019
1.824
1.489
1.126
0.8054
3.138
0.4065
3.063
2.300
2.548
2.925
0.8515
ASSOCIATED CONTENT
* Supporting Information
■
[HBth][CH3SO3]
sı
0.30
0.40
0.50
0.60
0.70
0.80
0.90
235.0
245.0
255.0
265.0
275.0
285.0
295.0
2.129
1.654
1.296
1.041
0.8186
0.6567
0.5429
2.115
1.661
1.308
1.036
0.8249
0.6629
0.5387
0.6576
2.1409
1.663
1.304
0.5589
0.5441
0.6173
0.8646
0.6474
0.8832
1.234
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.jced.9b00997.
0.4232
0.9259
0.4803
0.7696
0.9441
0.7736
Profiles of IL characterization (1H NMR, 13C NMR, and
1.032
ESI-MS spectra) (PDF)
0.8239
0.6625
0.5362
AUTHOR INFORMATION
Corresponding Author
■
[HBth][p-TSA]
0.30
0.40
0.50
0.60
0.70
0.80
0.90
235.0
245.0
255.0
265.0
275.0
285.0
295.0
2.228
1.705
1.313
0.9691
0.7213
0.5246
0.3821
2.200
1.711
1.302
0.9742
0.7199
0.5272
0.3842
1.257
2.217
1.713
1.303
0.9789
0.7263
0.5329
0.3867
0.4937
0.4692
0.7616
1.011
0.6932
1.582
Kaifeng Du − Sichuan University, Chengdu, P. R. China;
orcid.org/0000-0002-7402-4334; Phone: +86-28-
85405221; Email: kfdu@scu.edu.cn; Fax: +86-28-
85405221
0.3519
0.8378
0.5263
0.1941
0.4956
0.5496
Other Author
1.204
Tian Yao − Sichuan University, Chengdu, P. R. China;
orcid.org/0000-0002-4961-253X
a
Standard uncertainties u(T) = 0.05 K and u(P) = 0.05 kPa and
relative standard uncertainties ur(xi) = 0.031 and ur(xsol) = 0.015. xi
is the mole fraction solubility of ILs at the system temperature T.
b
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.jced.9b00997
c
The solvents are binary ethanol + ethyl benzoate systems, and xsol is
the molar fraction of ethyl benzoate in the binary solvent.
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Therefore, excellent performances for these three ILs in this
report are probably due to the particular molecular properties of
benzothiazolium portion in the IL structures, which may also
have intrinsic relationship with their temperature-sensitive
properties. Further research of molecular simulation is around
the corner in the near future.
Funding
This work was supported by the National Natural Science
Foundation of China (grant nos. 81803486 and 21676170),
China Postdoctoral Science Foundation (grant no.
2018M633385), and Sichuan Science and Technology Program
(grant no. 2019YJ0030).
CONCLUSIONS
■
The solubilities of benzothazolium-based ionic liquids [HBth]-
[BF4], [HBth][CH3SO3], and [HBth][p-TSA] in binary
ethanol + ethyl benzoate solvents were measured using the
equilibrium method with the temperature range from 233.2 to
293.7 K at atmospheric pressure. The experimental solubility
data were correlated by both λh equation and the modified
Apelblat equation with ideal correlation results (R2 > 0.999).
Particularly, in the λh model, the average relative deviations were
satisfactorily obtained as 0.7145, 0.7845, and 0.8119% for
[HBth][BF4], [HBth][CH3SO3], and [HBth][p-TSA], respec-
tively. All of these three ILs were showing stronger temperature-
sensitive property when the molar ratio of ethanol to ethyl
benzoate was higher. To our surprise, the parameters of λ and h
for the λh model and A, B, and C for the Apelblat model were
simultaneously expressed as functions of solvent composition
(xsol) for the first time. Furthermore, these two sets of self-
created in-depth correlation equations were successfully applied
to predict the solubilities of these three ILs in the binary system
with satisfactory performances. The relative deviations of the
prediction results were in acceptable ranges of 0.2541−3.309%
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the National Natural Science
Foundation of China (grant nos. 81803486 and 21676170),
China Postdoctoral Science Foundation (grant no.
2018M633385), and Sichuan Science and Technology Program
(grant no. 2019YJ0030).
REFERENCES
■
(1) Li, X.; Eli, W.; Li, G. Solvent-free synthesis of benzoic esters and
benzyl esters in novel Bro
̷
nsted acidic ionic liquids under microwave
irradiation. Catal. Commun. 2008, 9, 2264−2268.
(2) Wang, Y.; Zhang, D.-H.; Chen, N.; Zhi, G.-Y. Synthesis of benzyl
cinnamate by enzymatic esterification of cinnamic acid. Bioresour.
Technol. 2015, 198, 256−261.
(3) Wang, Y.; Zhang, D.-H.; Zhang, J.-Y.; Chen, N.; Zhi, G.-Y. High-
yield synthesis of bioactive ethyl cinnamate by enzymatic esterification
of cinnamic acid. Food Chem. 2016, 190, 629−633.
J
https://dx.doi.org/10.1021/acs.jced.9b00997
J. Chem. Eng. Data XXXX, XXX, XXX−XXX

Journal of Chemical & Engineering Data
pubs.acs.org/jced
Article
(4) Zhang, D.-H.; Zhang, J.-Y.; Che, W.-C.; Wang, Y. A new approach
to synthesis of benzyl cinnamate: Optimization by response surface
methodology. Food Chem. 2016, 206, 44−49.
(25) Wang, L.-H.; Song, Y.-T.; Chen, Y.; Cheng, Y.-Y. Solubility of
Artemisinin in Ethanol + Water from (278.2 to 343.2) K. J. Chem. Eng.
Data 2007, 52, 757−758.
̈
(26) Wei, D.; Pei, Y. Solubility of Diphenyl Carbonate in Pure
Alcohols from (283 to 333) K. J. Chem. Eng. Data 2008, 53, 2710−
2711.
(5) Maki-Arvela, P.; Salmi, T.; Sundell, M.; Ekman, K.; Peltonen, R.;
Lehtonen, J. Comparison of polyvinylbenzene and polyolefin supported
sulphonic acid catalysts in the esterification of acetic acid. Appl. Catal., A
1999, 184, 25−32.
(6) Heidekum, A.; Harmer, M. A.; Hoelderich, W. F. Addition of
carboxylic acids to cyclic olefins catalyzed by strong acidic ion-exchange
resins. J. Catal. 1999, 181, 217−222.
(7) Dijs, I. J.; van Ochten, H. L. F.; van Walree, C. A.; Geus, J. W.;
Jenneskens, L. W. Alkyl sulphonic acid surface-functionalised silica as
heterogeneous acid catalyst in the solvent-free liquid-phase addition of
acetic acid to camphene. J. Mol. Catal. A: Chem. 2002, 188, 209−224.
(8) Kozhevnikov, I. V. Catalysis by heteropoly acids and multi-
component polyoxometalates in liquid-phase reactions. Chem. Rev.
1998, 98, 171−198.
(9) Sheldon, R. A.; Downing, R. S. Heterogeneous catalytic
transformations for environmentally friendly production. Appl. Catal.,
A 1999, 189, 163−183.
́
(27) Domanska, U.; Bogel-L̷ukasik, E. Measurements and Correlation
of the (Solid + Liquid) Equilibria of [1-Decyl-3-methylimidazolium
Chloride + Alcohols (C2-C12 )]. Ind. Eng. Chem. Res. 2003, 42, 6986−
6992.
(28) Wang, J.; Wu, S.-D.; Yang, X.-Z. Solubility of 1,1′-(Butane-1,4-
diyl)-bis(3-methylpyridinium) Dihexafluorophosphate in Different
Solvents from (288.15 to 333.15) K. J. Chem. Eng. Data 2010, 55,
3942−3945.
(10) Wasserscheid, P.; Keim, W. Ionic liquids - new “solutions” for
transition metal catalysis. Angew. Chem., Int. Ed. 2000, 39, 3772−3789.
(11) Rogers, R. D.; Seddon, K. R. Ionic liquids–solvents of the future?
Science 2003, 302, 792−793.
(12) Yao, T.; Du, K. Temperature-Controlled Mono- and
Diolefination of Arene Using Rh(III)/RTIL as an Efficient and
Recyclable Catalytic System. ACS Sustainable Chem. Eng. 2019, 7,
6068−6077.
(13) Zhao, Y.; Long, J.; Deng, F.; Liu, X.; Li, Z.; Xia, C.; Peng, J.
Catalytic amounts of Bro̷nsted acidic ionic liquids promoted
esterification: study of acidity−activity relationship. Catal. Commun.
2009, 10, 732−736.
(14) Shi, H.; Zhu, W.; Li, H.; Liu, H.; Zhang, M.; Yan, Y.; Wang, Z.
̈
Microwave-accelerated esterification of salicylic acid using Bronsted
acidic ionic liquids as catalysts. Catal. Commun. 2010, 11, 588−591.
(15) Li, X.; Lin, Q.; Ma, L. Ultrasound-assisted solvent-free synthesis
of lactic acid esters in novel SO3H-functionalized Bro
liquids. Ultrason. Sonochem. 2010, 17, 752−755.
(16) Cole, A. C.; Jensen, J. L.; Ntai, I.; Tran, K. L. T.; Weaver, K. J.;
Forbes, D. C.; Davis, J. H. Novel Bronsted acidic ionic liquids and their
̷nsted acidic ionic
̷
use as dual solvent−catalysts. J. Am. Chem. Soc. 2002, 124, 5962−5963.
(17) Zhang, W.; Leng, Y.; Zhu, D.; Wu, Y.; Wang, J. Phosphotungstic
acid salt of triphenyl(3-sulfopropyl)phosphonium: An efficient and
reusable solid catalyst for esterification. Catal. Commun. 2009, 11, 151−
154.
(18) Zhou, X. S.; Liu, J. B.; Luo, W. F.; Zhang, Y. W.; Song, H. Novel
Bro̷nsted-acidic ionic liquids based on benzothiazolium cations as
catalysts for esterification reactions. J. Serb. Chem. Soc. 2011, 76, 1607−
1615.
(19) Wang, L.; Wang, X.; Zuo, T.; Sun, Q.; Liu, P.; Yao, S.; Song, H.
Determination and correlation of the solubility of four Bro̷nsted-acidic
ionic liquids based on benzothiazolium cations in six alcohols. J. Chem.
Thermodyn. 2014, 72, 48−53.
(20) Li, Y.; Wang, L.-S.; Cai, S.-F. Mutual Solubility of Alkyl
Imidazolium Hexafluorophosphate Ionic Liquids and Water. J. Chem.
Eng. Data 2010, 55, 5289−5293.
(21) Buchowski, H.; Ksiazczak, A.; Pietrzyk, S. Solvent activity along a
saturation line and solubility of hydrogen-bonding solids. J. Phys. Chem.
1980, 84, 975−979.
(22) Wei, D.; Li, H.; Liu, C.; Wang, B. The Effect of Temperature on
the Solubility of 11-Cyanoundecanoic Acid in Cyclohexane, n-Hexane,
and Water. Ind. Eng. Chem. Res. 2011, 50, 2473−2477.
(23) Wang, Z.; Wang, J.; Zhang, M.; Dang, L. Solubility of
Erythromycin A Dihydrate in Different Pure Solvents and Acetone +
Water Binary Mixtures between 293 K and 323 K. J. Chem. Eng. Data
2006, 51, 1062−1065.
(24) Li, X.; Yin, Q.; Chen, W.; Wang, J. Solubility of Hydroquinone in
Different Solvents from 276.65 K to 345.10 K. J. Chem. Eng. Data 2006,
51, 127−129.
K
https://dx.doi.org/10.1021/acs.jced.9b00997
J. Chem. Eng. Data XXXX, XXX, XXX−XXX