Fluid Phase Equilibria 367 (2014) 135–142

Contents lists available at ScienceDirect

Fluid Phase Equilibria

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

Densities and dynamic viscosities of (choline chloride + glycerol) deep

eutectic solvent and its aqueous mixtures in the temperature range
(283.15–363.15) K
Anita Yadav, Shruti Trivedi, Rewa Rai, Siddharth Pandey ∗
Department of Chemistry, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016, India

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

Article history: Deep eutectic solvents (DESs) are emerging as one of the most promising environmentally benign and
Received 23 September 2013 cost-effective alternatives to the conventional organic solvents. Many common DESs are readily miscible
Received in revised form 10 January 2014 with water. Aqueous mixtures of DESs have potential to afford modified properties for specific appli-
Accepted 23 January 2014
cations. Densities and dynamic viscosities of a common and popular DES composed of choline chloride
Available online 4 February 2014
and glycerol in 1:2 mole ratio, named glyceline, and its aqueous mixtures in the temperature range
283.15–363.15 K are reported. Decrease in density with increasing temperature is found to follow a
Keywords:
quadratic expression. Excess molar volumes of the aqueous mixtures of glyceline are found to be neg-
Deep eutectic solvents
Density
ative and significant at all temperatures and compositions. Absolute excess molar volume is found to
Excess molar volume decrease as the temperature is increased from 283.15 K to 323.15 K. For temperature above 323.15 K,
Dynamic viscosity the excess molar volume does not change much with further increase in temperature till 363.15 K. Tem-
Logarithmic excess viscosity perature dependence of dynamic viscosity of aqueous mixtures of glyceline in the temperature range
283.15–363.15 K at all compositions is found to be better described by a Vogel–Fulcher–Tamman (VFT)
model. Excess logarithmic viscosities for aqueous mixtures of glyceline are found to be positive at all tem-
peratures and compositions investigated. The results highlight relatively stronger interactions, preferably
H-bonding type, between water and glyceline, as compared to those among water and among glyceline
molecules, respectively. Facile interstitial accommodation of water within H-bonded glyceline network
also appears to contribute to the experimental observations.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction impact of ionic liquids on environment is still the matter of dispute

and these can neither be generalized as toxic nor as green. Here,
Over the last two decades, applications of ionic liquids in sci- type of ions, i.e., cations and anions constituting the ionic liquid,
ence and technology have increased manifold due to their unique plays a significant role in determining the impact of the ionic liq-
and tunable physicochemical properties, such as, negligible vapor uid on environment [8]. Subsequently, reaction media composed
pressure, good thermal stability, tunable viscosity and miscibility of biomaterials are becoming substance of major importance that
with water [1–4]. Apart from reports on better performance of accomplish the unique properties of ionic liquids coupled with the
ionic liquids in many of the applications, several of these inves- advantages associated with biodegradable and non-toxic starting
tigations also reveal outcomes that are unexpected or unusual materials. Recently discovered deep eutectic solvents (DESs) are
[5–7]. Many drawbacks, however, associated with ionic liquids are emerging as a remarkable group of solvents in the area of green
also realized during these investigations. Among others, restricted chemistry and are extensively investigated class of reaction media
physicochemical properties, especially solute solubility and vis- alternative to conventional ionic liquids.
cosity, are perhaps the most noteworthy. Presently, they are the Several applications of DESs in the field of electrodeposition,
cause of major concern to the scientific community promoting ionic electrochemistry, inorganic syntheses, and as a media for several
liquids. Moreover, large scale application of ionic liquids is disad- organic reactions have already been published [9–17]. A com-
vantageous from economic point of view. Most importantly, the mon and popular DES can simply be a mixture of a substituted
quaternary ammonium salt and a hydrogen bond donor which
produces a liquid such that its physicochemical properties resem-
ble those of ionic liquids with several advantages over the latter.
∗ Corresponding author. Tel.: +91 11 26596503; fax: +91 11 26581102.
The most studied DESs perhaps are based on choline chloride
E-mail addresses: sipandey@chemistry.iitd.ac.in,
chempandey@gmail.com (S. Pandey). [ChCl: (2-hydroxyethyl)trimethylammonium chloride]. ChCl is the

0378-3812/$ – see front matter © 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.fluid.2014.01.028
136 A. Yadav et al. / Fluid Phase Equilibria 367 (2014) 135–142

quaternary ammonium salt which is non-toxic, biodegradable, and Table 1

Description of the chemicals used in this work.
relatively cheaper. Choline cation belongs to an important class of
vitamin B family and it is considered to play an essential role in Chemical Source Purification
our daily life by assisting in various metabolic mechanisms [18,19]. done
ChCl also serves as a nutritional supplement of animal feedstock Glyceline 202 (1 mol choline Scionix Ltd. None
[20,21]. The other constituent of a common DES, an H-bond donor, chloride + 2 mol glycerol)
can usually be an alcohol, amide or a carboxylic acid. The prepara- Doubly distilled deionized Merck None
water (HPLC grade)
tion of DES is simple, safe, and cost effective and does not involve
any purification or waste disposal problem. For these reasons DESs
are showing growing interest both at academic and industrial lev- 1:2 and stirred under heating (∼80 ◦ C) until a homogeneous, color-
els. less liquid has been formed [23]. The density and dynamic viscosity
Despite several investigations on these ‘benign/non-toxic’ measurements of glyceline purchased from Scionix Ltd. and that
media, information on physicochemical properties of DESs is still prepared using choline chloride (≥99% from Sigma–Aldrich) and
incomplete and limited. Among the key physicochemical proper- glycerol (≥99.5%, Spectrophotometric grade from Sigma–Aldrich)
ties, density and viscosity have enormous importance as far as were found to be statistically similar. Doubly distilled deionized
various industrial applications are concerned. Water mixtures of water of HPLC grade was obtained from Merck. The description
DES rather than neat DESs may provide substantially improved of the chemicals used together with their sources is provided in
physicochemical properties for many academic and industrial Table 1.
applications [22]. Importantly, the key physicochemical proper-
ties of (DES + water) mixtures can be tuned with the amount of
2.2. Methods
water present. In the current work, we have chosen a common
and popular DES, named glyceline, which is composed of (choline
The aqueous mixtures of glyceline were prepared by mass using
chloride + glycerol) combined in the mole ratio 1:2. When choline
a Denver Instrument balance having a precision of ±0.1 mg. Den-
chloride (m.p. = 302 ◦ C) and glycerol (m.p. = 18 ◦ C) are mixed in 1:2
sities () of the aqueous mixtures of glyceline were measured
mole ratio, the melting point of the resulting mixture is reduced
using a Mettler Toledo, DE45 delta range density meter. The den-
to −40 ◦ C [23,24]. Glyceline is non-flammable and non-toxic. In
sity measurement with the above-mentioned density meter was
the current study, we report temperature-dependent densities and
based on electromagnetically induced oscillations of a U-shaped
viscosities of glyceline and its aqueous mixtures over the entire
glass tube. The uncertainties associated with the density measure-
composition range in the temperature range 283.15–363.15 K. Den-
ments are ≤0.00005 g cm−3 . The measurements were performed at
sities of aqueous mixtures of glyceline have been measured by
10◦ intervals in the temperature range 283.15–363.15 K. Table 2
other researchers. In a recent study, Li et al. reported the densities
presents comparison of densities of glyceline and their aqueous
of (choline chloride + ethylene glycol or glycerol) and their aque-
mixtures measured with our instrumentation with those avail-
ous mixtures, however the temperature range (298.15–333.15 K)
able in literature [25,27,31]. Our values are in excellent agreement
of this report was significantly narrower [25]. In another study, the
with those reported in the literature. The dynamic viscosities ()
same group reported the pressure dependent densities of glyce-
were measured with a Peltier-based (resolution of 0.01 ◦ C and
line and their aqueous mixtures at different temperatures [26]. The
accuracy < 0.05 ◦ C) automated Anton Paar microviscometer (model
temperature range in this study again is rather restricted. Another
AMVn) having calibrated glass capillaries of different diameters
recent contribution toward temperature-dependent density mea-
(1.6, 1.8, 3.0, and 4.0 mm). This instrument is based on the rolling
surement of ammonium and phosphonium based DESs was from
ball principle, where the steel ball rolls down the inside of inclined,
AlNashef et al. in which the densities were predicted using arti-
sample-filled calibrated glass capillaries. The uncertainties associ-
ficial intelligence and group contribution techniques [27]. Li and
ated with the viscosity measurements are ≤0.5%.
coworkers reported the densities, viscosities, and refractive indices
of (N,N-diethylethanol ammonium chloride + glycerol) and (N,N-
diethylethanol ammonium chloride + ethylene glycol) and their 3. Results and discussion
aqueous mixtures from 298.15 to 343.15 K [28].
Viscosity, as a function of temperature, is one of the most impor- 3.1. Temperature dependence of density of glyceline and its water
tant properties that usually govern the usage of DESs in most mixtures
industrial, pharmaceutical, chemical, and biotechnological appli-
cations. Furthermore, viscosity is an essential parameter that helps Experimentally measured densities of glyceline and (glyce-
develop a solubilizing milieu and plays a major part in chemical line + water) mixtures as a function of temperature in the range
transformations in solution [29,30]. Few attempts were made by 283.15–363.15 K over the entire composition range are reported in
Mantle and coworkers to study the viscosity of neat glyceline and Table 3. As expected, densities of glyceline, water, and their mix-
other DESs at different temperatures [9,31]. However, we could not ture are found to decrease with increase in temperature. Thermal
find any report on temperature dependent viscosities of aqueous expansion usually results in decreased density of a substance as the
mixtures of glyceline. temperature is increased. The experimentally measured densities
are found to vary quadratically with the absolute temperature and
were fitted to the following equation:
2. Experimental  = 0 + aT + bT 2 (1)

2.1. Materials where  (g cm−3 ) is the density of the aqueous mixture of glyce-
line. The values of the parameters 0 , a, and b along with the
Glyceline 202 (mol wt. 107.95 g/mol), a mixture of choline chlo- standard deviation of the fits are listed in Table 4. Measured densi-
ride + glycerol (1:2 mole ratio) was purchased from Scionix Ltd. and ties of (glyceline + water) mixtures along with the fits to equation
used as received. Alternatively, glyceline is also prepared by mixing [1] are presented in Fig. 1. It is clear from the recovered values of
choline chloride (≥99% from Sigma–Aldrich) and glycerol (≥99.5%, r2 listed in Table 4 as well as from careful examination of Fig. 1,
Spectrophotometric grade from Sigma–Aldrich) in a mole ratio of where the fits of the measured density against the temperature
 A. Yadav et al. / Fluid Phase Equilibria 367 (2014) 135–142 137

Table 2
Comparison of experimental densities (d ) of water (1) + glyceline (2) mixtures with the literature values at different temperatures (x1 e is the mole fraction of water).

T (K)f x1  (g cm−3 )

0.0000 0.5978 0.9007 1.0000

Exp. Lit. Exp. Lit. Exp. Lit. Exp. Lit.

293.15 1.1943 1.181a

303.15 1.1885 1.1895b 1.1523 1.1525c 1.0715 1.0719c 0.9957 0.9957c
1.1885c
313.15 1.1827 1.1838b 1.1468 1.1470c 1.0668 1.0672c 0.9922 0.9923c
1.1823c
323.15 1.1770 1.1776b 1.1413 1.1414c 1.0617 1.0621c 0.9881 0.9881c
1.1775c
333.15 1.1713 1.1741b 1.1357 1.1358c 1.0566 1.0567c 0.9833 0.9832c
1.1719c
343.15 1.1648 1.1674b
353.15 1.1575 1.1635b
363.15 1.1495 1.1571b
AAD (%) 1.13a 6.12b 0.02c 0.01c 0.03c 0.01c
n
Measurements were performed at atmospheric pressure. AAD% = (100/n) (|i,Literature − i,Experimental |)/i,Literature where n is the number of data points.
i=1
a
Ref. [31].
b
Ref. [27].
c
Ref. [25].
d
Uncertainty: u() = ±0.00005 g/cm3 .
e
Uncertainty: u(x1 ) = ±10−4 .
f
Uncertainty: u(T) = ±0.05 K.

Table 3
Densities (a (g cm−3 )) of water (1) + glyceline (2) mixtures from T = (283.15–363.15) K as a function of mole fraction of water (x1 ).

x1 b wt% (water) T (K)c

283.15 293.15 303.15 313.15 323.15 333.15 343.15 353.15 363.15

0.0000 0 (Glyceline) 1.2001 1.1943 1.1885 1.1827 1.1770 1.1713 1.1648 1.1575 1.1495
0.1080 2 1.1969 1.1911 1.1853 1.1795 1.1739 1.1682 1.1617 1.1545 1.1465
0.2381 5 1.1916 1.1858 1.1800 1.1744 1.1688 1.1631 1.1566 1.1494 1.1414
0.3425 8 1.1860 1.1802 1.1746 1.1690 1.1634 1.1577 1.1513 1.1441 1.1362
0.4022 10 1.1822 1.1765 1.1709 1.1653 1.1597 1.1541 1.1477 1.1406 1.1327
0.5978 20 1.1631 1.1577 1.1523 1.1468 1.1413 1.1357 1.1294 1.1224 1.1146
0.7199 30 1.1429 1.1377 1.1325 1.1271 1.1216 1.1161 1.1099 1.1030 1.0954
0.7997 40 1.1223 1.1174 1.1124 1.1072 1.1018 1.0964 1.0904 1.0836 1.0761
0.8574 50 1.1013 1.0969 1.0922 1.0872 1.0819 1.0767 1.0707 1.0640 1.0567
0.9007 60 1.0798 1.0758 1.0715 1.0668 1.0617 1.0566 1.0507 1.0441 1.0369
0.9335 70 1.0593 1.0559 1.0520 1.0476 1.0427 1.0376 1.0319 1.0255 1.0183
0.9602 80 1.0390 1.0362 1.0328 1.0287 1.0240 1.0191 1.0134 1.0071 1.0001
0.9819 90 1.0192 1.0171 1.0140 1.0102 1.0055 1.0006 0.9951 0.9889 0.9820
1.0000 100 (Water) 0.9997 0.9982 0.9957 0.9922 0.9881 0.9833 0.9778 0.9717 0.9650

Measurements were performed at atmospheric pressure.

a
Uncertainty: u() = ±0.00005 g/cm3 .
b
Uncertainty: u(x1 ) = ±10−4 .
c
Uncertainty: u(T) = ±0.05 K.

according to equation [1] for (glyceline + water) mixtures over the relation in the temperature range (283.15–363.15 K). It is important
entire composition range are presented as solid curves, that the to mention that Li group has reported a linear decrease in density
temperature-dependence of the densities of glyceline and its aque- with temperature for glyceline and its aqueous mixtures [25]. It is
ous mixtures can be conveniently expressed by a simple quadratic attributed to the fact that the temperature range for density data of

Table 4
Result of the regression analysis of density ( (g cm−3 )) versus temperature (T (K)) data according to equation:  (g cm−3 ) = 0 (g cm−3 ) + a(T (K)) + b(T (K))2 for (water + glyceline)
over the temperature range (283.15–363.15) K. Standard deviations are given as ± in parentheses.

x1 wt% (Water) 0 (g cm−3 ) a × 10−3 b × 10−6 r2

0.0000 0 (Glyceline) 1.227 (±0.031) 0.309 (±0.195) −1.435 (±0.303) 0.9993

0.1080 2 1.225 (±0.032) 0.301 (±0.202) −1.421 (±0.312) 0.9992
0.2381 5 1.216 (±0.032) 0.321 (±0.203) −1.445 (±0.313) 0.9992
0.3425 8 1.206 (±0.031) 0.344 (±0.195) −1.474 (±0.301) 0.9992
0.4022 10 1.199 (±0.030) 0.361 (±0.190) −1.497 (±0.294) 0.9993
0.5978 20 1.170 (±0.027) 0.417 (±0.169) −1.566 (±0.261) 0.9994
0.7199 30 1.141 (±0.024) 0.462 (±0.151) −1.617 (±0.234) 0.9995
0.7997 40 1.105 (±0.022) 0.549 (±0.135) −1.729 (±0.208) 0.9996
0.8574 50 1.059 (±0.018) 0.694 (±0.112) −1.926 (±0.173) 0.9997
0.9007 60 1.006 (±0.014) 0.878 (±0.086) −2.178 (±0.133) 0.9998
0.9335 70 0.948 (±0.009) 1.093 (±0.058) −2.478 (±0.090) 0.9999
0.9602 80 0.884 (±0.005) 1.351 (±0.031) −2.842 (±0.047) 0.9999
0.9819 90 0.827 (±0.010) 1.572 (±0.065) −3.157 (±0.101) 0.9998
1.0000 100 (Water) 0.755 (±0.009) 1.880 (±0.062) −3.588 (±0.096) 0.9998
138 A. Yadav et al. / Fluid Phase Equilibria 367 (2014) 135–142

Fig. 1. Variation of density with temperature for water (1) + glyceline (2) mixtures at
different mole fractions of water (x1 ) (: 1.0000; 䊉: 0.9819; : 0.9602; : 0.9335;
夽: 0.9007; : 0.8574; : 0.7997; : 0.7199; : 0.5978; : 0.4022; ♦: 0.3425; :
0.2381; : 0.1080; : 0.0000). The solid lines represent a fit to the equation
 = 0 + aT + bT 2 . Parameters 0 , a, and b along with the goodness-of-fit in terms
2
of r are reported in Table 4.

Li group was much narrower (298.15–333.15 K only). It is clear that

measurements at higher temperatures (363.15 K ≥ T > 333.15 K)
reveal a more complex dependence of density on temperature of
glyceline and its aqueous mixtures. This is manifested in changed
molecular interactions between glyceline and water in this tem-
perature region (vide infra).
As the DES glyceline is constituted of choline chloride and glyc-
erol, it is imperative to compare the temperature dependence of Fig. 2. Variation of excess molar volume (VE (cm3 mol−1 )) with mole fraction of
density of aqueous mixtures of glyceline with those of aqueous water (x1 ) for (water + glyceline) mixtures as the temperature is increased from (A)
(283.15–323.15) K [䊉: 283.15 K; : 293.15 K; : 303.15 K; ♦: 313.15 K; : 323.15 K].
mixtures of glycerol and aqueous mixtures of choline chloride,
(B) (333.15 to 363.15) K [: 333.15 K; : 343.15 K; : 353.15 K; : 363.15 K]. Solid
respectively. The difference in the temperature dependence of the curves show fits according to the Redlich–Kister equation [Eq. (4)] with parameters
density of aqueous mixtures of glyceline from that of aqueous (Aj ) reported in Table 5.
mixtures of glycerol is revealed by the measurements of Jouy-
ban and coworkers [32]. They report an almost linear decrease
in density with temperature for (glycerol + water) system; how- The molecular weight of the glyceline was calculated from their
ever, the temperature range of their investigation was only from individual components according to the equation [23]:
288.15 K to 303.15 K. Adamenko et al. have reported the densi-
M2 = xChCl MChCl + xglycerol Mglycerol (3)
ties of glycerol–water solutions to be a monotonic function of
temperature in the range 290.35–355.95 K [33]. It is noteworthy The excess molar volumes at each temperature for (glyce-
that, for (choline chloride + water) system on the other hand, the line + water) mixtures are presented as a function of x1 , the mole
dependence of density on temperature is quadratic in the temper- fraction of water, in Fig. 2. It is clear that the excess molar vol-
ature range 278.15–318.15 K [34]. It appears that the temperature umes are negative and have significant values at each temperature
dependence of density of (glyceline + water) resembles more that throughout the entire composition range for all (glyceline + water)
of (choline chloride + water) than (glycerol + water) although glyce- mixtures. Interestingly, the maximum absolute excess molar
line is constituted of 2 mol of glycerol and 1 mol of choline chloride. volume for a (glyceline + water) mixture is observed in the water-
rich region (xwater ∼0.6–0.7). Importantly, the absolute values of
3.2. Composition dependence of density of aqueous mixtures of excess molar volumes for (glyceline + water) mixtures appear to
glyceline decrease with increasing temperature in the temperature ranges
283.15–323.15 K (Fig. 2A). However, as temperature is increased
Density of glyceline decreases monotonically as water is added for T > 323.15 K, the excess molar volume appears to change less
as the density of neat glyceline is higher than that of water at significantly; in fact the variation is reversed at higher tempera-
a given temperature (Table 3). The decrease with mole fraction tures (333.15–363.15 K) (see Fig. 2B). While, Li and coworkers also
of water is found to be nonlinear. It is noteworthy that variation report excess molar volume of (glyceline + water) to decrease with
of density with concentration of water at a given temperature is increasing temperature in the range 298.15–323.15 K; the VE at
reported to be nonlinear for (glycerol + water) [33] as well as for 323.15 K was reported by them to be not too different from VE at
(choline chloride + water) [34] systems. In order to afford the extent 333.15 K [25]. 333.15 K was the highest temperature in their inves-
of molecular-level interactions within (glyceline + water) mixtures, tigation. Similarly, VE at 333.15 K was estimated to be very similar to
we estimated excess molar volume (VE ) from experimental density VE at 343.15 K for system composed of (DES + water), where the DES
data using the relationship is constituted of N,N-diethylethanolammonium chloride instead of
x1 M1 + x2 M2
x M x2 M2
 choline chloride [28]. It is important to mention that excess molar
1 1
VE = − + (2) volumes of (glycerol + water) system are also estimated to be nega-
m 1 2
tive by several researchers [32,35,36]. Further, the VE is reported to
Here, x1 , x2 , and 1 , 2 refer to the mole fractions and densi- decrease with temperature in the range 283.15–313.15 K by these
ties, respectively, of water and glyceline at a given temperature, researchers.
and m is the density of the mixture. M1 and M2 are the molecular The excess molar volumes were next fitted to the Redlich–Kister
weights of components 1 (water) and 2 (glyceline), respectively. type polynomial expressions [37]. According to combined nearly
 A. Yadav et al. / Fluid Phase Equilibria 367 (2014) 135–142 139

Table 5
Fit parameters (Ajs ) and correlation coefficient (r2 ) in the Redlich–Kister equation
[Eq. (4)] for excess molar volume (VE ) of water (1) + glyceline (2) mixtures.

T (K) A0 A1 A2 A3 r2

283.15 −1.3557 −0.6776 −0.5943 −0.1750 0.9964

293.15 −1.2571 −0.6512 −0.4568 −0.0489 0.9946
303.15 −1.2158 −0.5987 −0.3408 −0.0101 0.9934
313.15 −1.1958 −0.5353 −0.2851 0.0077 0.9924
323.15 −1.1966 −0.6470 −0.1956 0.3602 0.9414
333.15 −1.2001 −0.6371 −0.2110 0.3430 0.9539
343.15 −1.2076 −0.6443 −0.2081 0.3469 0.9528
353.15 −1.2152 −0.6524 −0.2129 0.3667 0.9495
363.15 −1.2249 −0.6595 −0.2056 0.3656 0.9499

Fig. 3. Dynamic viscosity () as a function of mole fraction of water (x1 ) for
ideal binary solvent/Redlich–Kister (CNIBS/R-K) model, the VE in a (water + glyceline) mixtures at different temperatures. [䊉: 283.15 K; : 293.15 K;
binary solvent mixture at a constant temperature can be expressed : 303.15 K; : 313.15 K; : 323.15 K; : 333.15 K; : 343.15 K; ♦: 353.15 K; :
as: 363.15 K]. Solid curves at each temperature are obtained using the best fit param-
eters of the Redlich–Kister polynomial [Eq. (8)] reported in Table 8. The error in

k measured  is ≤0.5%.
V E = xwater xglyceline Aj (xwater − xglyceline )j (4)
j=0 negative VE for our system. Upon heating, further contraction in
the mixture volume, albeit small, could be tentatively attributed
where Aj and j are the equation coefficients and the degree of the
to the weakening of the H-bond among glycerol and choline chlo-
polynomial expansion, respectively. The numerical values of j can
ride (constituents of the glyceline) and/or among water molecules
be varied to find an accurate mathematical representation of the
with subsequent strengthening of the H-bonding between glyce-
experimental data. Regression analysis was performed to fit the
line and water. The interstitial exclusion of water molecules from
polynomials [i.e., Eq. (4)] to our experimental density data and the
(glycerol + choline chloride) H-bonded network in the temperature
results of the fit are reported in Table 5. It is convenient to use a
range 333.15–363.15 K may be a more important factor in increas-
cross-validation method which is a practical and reliable method
ing absolute value of VE with increasing temperature of the mixture.
to test the predictive significance when only little data are available
[38]. The solid lines connecting each of the VE in Fig. 2 at all pro-
3.3. Dynamic viscosity of glyceline and its aqueous mixtures:
portions are obtained from the CNIBS/R-K model fit (with j = 3) as
temperature and composition dependence
reported in Table 5 which also suggests a fair-to-good correlation
between the predicted and the experimentally obtained values. It is
Experimentally measured dynamic viscosities () of (glyce-
noteworthy from the r2 values that while the fits are satisfactory in
line + water) mixtures over complete composition range at 10◦
the temperature range 283.15–313.15 K; they are not so at higher
interval in the temperature range 283.15–363.15 K are reported
temperatures in the range 323.15–363.15 K. This is in line with the
in Table 6. It is evident from the entries in Table 6 that, as
earlier observations of decrease in VE with increase in temperature
expected, monotonic decrease in dynamic viscosity is observed as
till 323.15 K followed by a slight increase in VE with increase in
the temperature is increased from 283.15 to 363.15 K for a given
temperature for 363.15 ≥ T > 323.15 K (vide supra).
composition of (glyceline + water) mixture. Plots of  versus x1
The negative excess molar volume generally points to con-
(mole fraction of water) for aqueous mixtures of glyceline at differ-
traction in volume upon mixing. The interactions within the
ent temperatures are shown in Fig. 3. Within a (glyceline + water)
mixture are affected by the temperature and the composition
mixture at fixed temperature, dynamic viscosity decreases mono-
of the mixture. In liquid mixtures, the absolute value of molar
tonically upon increasing the mole fraction of water (x1 ). This is
excess volume usually decreases with increasing temperature
expected as  of water is significantly lower than that of neat glyce-
in systems where interactions are present. The negative VE of
line at all investigated temperatures.
(glyceline + water) mixtures at all compositions in the temperature
Abbott and coworkers found that the viscosity-temperature
range 283.15–363.15 K hints at stronger interspecies interactions
profile for neat glyceline follows an Arrhenius-like behavior [23].
(between water and glyceline) than intraspecies interactions (i.e.,
Arrhenius-like behavior is usually considered the most simplistic as
among water molecules or among glyceline molecules) along with
far as temperature dependence of dynamic viscosity is concerned.
facile interstitial accommodation of water within H-bonded glyce-
Although the measured viscosities of glyceline and its aqueous
line network. The interactions between water and glyceline could
mixtures may be described by an Arrhenius model over the temper-
be of the H-bonding type, among others, involving glycerol and
ature range studied (results not shown), Vogel–Fulcher–Tamman
choline chloride – the constituents of the DES. Both OH group
(VFT) model (a model which especially describes ionic liquids
and chloride ion of choline chloride and OH groups of glycerol
having small, symmetric cations [39–41]), appears to more satisfac-
are capable of H-bonding with water. The interspecies interactions
torily describe the viscosities of glyceline and its aqueous mixtures:
usually decrease more as compared to intraspecies interactions
with increasing temperature. As a result, we observe a decrease B
ln  = A + (5)
in absolute value of VE in the temperature range 283.15–323.15 K. T − T0
However, a slight increase in the absolute value of VE as the tem- Here A, B, and T0 are empirical constants (T0 generally corre-
perature is increased from 333.15 to 363.15 K could be attributed sponds to the temperature of divergence where the configurational
to the inherent complexity of the (glyceline + water) mixtures as entropy of the system vanishes). The fits to this model returned cor-
far as interactions within the system are concerned. As mentioned relation coefficients close to unity (r2 > 0.9984) (Table 7). From the
earlier, the H-bonding and other interactions between the unlike recovered parameters, Ea at 298 K was also determined using [42]:
components (i.e., glyceline and water) within the mixtures along  T
2
with interstitial accommodation of water within H-bonded glyce- Ea,␩ = RB (6)
line network would result in volume contraction and thus the T − T0
140 A. Yadav et al. / Fluid Phase Equilibria 367 (2014) 135–142

Table 6
Viscosities (a (mPa s)) for water (1) + glyceline (2) mixtures from T = (283.15–363.15) K as a function of mole fraction of water (x1 ).

x1 b (wt%) T (K)c

283.15 293.15 303.15 313.15 323.15 333.15 343.15 353.15 363.15

0.0000 (0) 1003.9428 472.9671 246.7928 133.3749 81.5895 53.8117 36.7409 26.3331 19.5866
0.1080 (2) 669.9038 322.6054 172.8659 97.6070 60.8033 39.3459 27.4970 20.0963 15.2351
0.2381 (5) 397.1261 200.4529 105.3949 63.6328 41.6544 28.2346 20.1102 14.9828 11.4031
0.3425 (8) 239.0806 125.9387 72.1121 44.9472 29.7991 20.8106 15.2347 11.5860 9.0466
0.4022 (10) 177.9621 95.9405 56.4675 35.8177 24.0906 17.1371 12.6879 9.5736 7.5048
0.5978 (20) 54.6306 32.8956 21.3567 14.7350 10.6386 7.7082 5.6748 4.4946 3.6385
0.7199 (30) 21.8137 14.3726 10.0548 7.0999 5.0760 3.9458 3.1519 2.5717 2.1426
0.7997 (40) 11.3182 7.6483 5.6483 3.8621 3.0075 2.4063 1.9711 1.6472 1.3986
0.8574 (50) 6.0668 4.3516 3.2576 2.5277 2.0196 1.6547 1.3842 1.1855 1.0262
0.9007 (60) 3.8894 2.8657 2.1977 1.7433 1.4169 1.1824 1.0040 0.8822 0.7702
0.9335 (70) 2.6941 2.0191 1.5742 1.2652 1.0444 0.8829 0.7359 0.6654 0.6145
0.9602 (80) 2.0141 1.5294 1.2059 0.9808 0.8193 0.6999 0.6006 0.5212 0.4654
0.9819 (90) 1.5754 1.2065 0.9611 0.7902 0.6578 0.5652 0.4876 0.4353 0.3862
1.0000 (100) 1.2739 0.9836 0.7914 0.6492 0.5487 0.4770 0.4354 0.3860 0.3306

Measurements were performed at atmospheric pressure.

a
Uncertainty: u() = ±0.5%.
b
Uncertainty: u(x1 ) = ±10−4 .
c
Uncertainty: u(T) = ±0.05 K.

Table 7
Summary of parameters associated with viscosity according to the VFT model.

x1 wt% (water) A B T0 (K) r2 Ea,␩ (kJ mol−1 )

0.0000 0 −2.4846 1036.2157 173.0462 0.9999 48.93

0.1080 2 −2.5956 1014.1845 171.8824 0.9999 47.01
0.2381 5 −2.1974 854.1548 178.8536 0.9999 44.36
0.3425 8 −2.2352 833.8173 175.1009 0.9999 40.70
0.4022 10 −2.4514 862.1473 170.2100 0.9999 38.93
0.5978 20 −3.6325 1104.2325 138.3825 0.9998 31.97
0.7199 30 −3.1796 840.0494 149.2902 0.9995 28.02
0.7997 40 −2.3852 502.6864 178.8015 0.9997 26.08
0.8574 50 −2.8367 596.9603 154.5178 0.9999 21.39
0.9007 60 −2.7514 504.9861 160.3426 0.9999 19.65
0.9335 70 −2.5719 391.0429 173.5879 0.9995 18.63
0.9602 80 −3.2428 533.7623 147.6915 0.9999 17.43
0.9819 90 −3.1887 464.3495 155.6861 0.9999 16.91
1.0000 100 −2.9727 369.8629 168.0219 0.9984 16.14

At 298.15 K, Ea,␩ is the activation energy for viscous flow.

Plots of ln  versus 1/T along with the best fit lines according to It is noteworthy that temperature dependence of dynamic
VFT model for (glyceline + water) mixtures are presented in Fig. 4. viscosities of (glycerol + water) system in the temperature range
Goodness-of-fit in terms of r2 along with the recovered parameters 273.15–373.15 K is also be better described based on VFT model
are presented in Table 7. As expected, Ea,␩ decreases monotoni- [43]. The dynamic viscosities of (choline chloride + water) mixtures,
cally as the concentration of water is increased in the mixture. It on the other hand, vary with temperature that can be satisfac-
was found that our viscosity values of neat glyceline at different torily described by the Arrhenius relationship [34]. However, at
temperatures are slightly higher than that of the values of Abbott lower concentrations of chloline chloride, VFT model appears to
et al. [23]. This positive deviation in viscosity could be caused by describe better the temperature variation of dynamic viscosity of
the lesser moisture contents in our samples. the (choline chloride + water) mixtures. For aqueous mixtures of
DES, where the DES is constituted of N,N-diethylethanolammonium
chloride instead of choline chloride, the dynamic viscosity is
reported to vary with temperature that follows VFT equation better
than Arrhenius equation [28].

3.4. Excess logarithmic viscosity and Redlich–Kister treatment

In order to assess the interactions within (glyceline + water)

mixtures, excess logarithmic viscosities, (ln )E , of (glyce-
line + water) mixtures were calculated from the equation
[44]:

(ln )E = ln  − [xwater ln water + xglyceline ln glyceline ] (7)

Plots of (ln E )
versus x1 (mole fraction of water) for (glyce-
Fig. 4. Variation of ln  with 1/T for water (1) + glyceline (2) mixtures at differ-
line + water) mixtures in the temperature range 283.15–363.15 K
ent mole fractions of water. (: 1.0000; 䊉: 0.9819; : 0.9602; : 0.9335; 夽:
0.9007; : 0.8574; : 0.7997; : 0.7199; : 0.5978; : 0.4022; ♦: 0.3425; : are presented in Fig. 5. (ln )E are positive and have significant
0.2381; : 0.1080; : 0.0000). The solid lines represent the best fit for the VFT values at all temperatures over the entire composition range.
(Vogel–Fulcher–Tamman) model [Eq. (5)]. Further, similar to what was observed for excess molar volume,
 A. Yadav et al. / Fluid Phase Equilibria 367 (2014) 135–142 141

Table 8
Parameters (Ajs ) and correlation coefficient (r2 ) recovered from the best fit of
(ln )E according to Redlich–Kister polynomial [Eq. (8)] for water (1) + glyceline (2)
mixtures.

T (K) A0 A1 A2 A3 r2

283.15 4.2754 1.5272 −0.1202 −0.6543 0.9707

293.15 4.0429 1.6321 −0.0197 −0.5130 0.9884
303.15 3.7615 1.7143 0.0811 −0.4808 0.9880
313.15 3.6942 1.3799 0.2558 0.1988 0.9862
323.15 3.5558 1.3814 0.2487 0.1615 0.9878
333.15 3.4109 1.4162 −0.0639 0.3654 0.9798
343.15 3.3416 2.2943 −0.5951 −2.0551 0.5960
353.15 3.2024 2.1213 −0.3802 −1.6821 0.6784
363.15 3.0386 1.4212 0.4327 0.3309 0.9138

Fig. 5. Variation of (ln )E with mole fraction of water (x1 ) for (water + glyceline)
mixtures as the temperature is increased from (283.15–363.15) K [䊉: 283.15 K;
: 293.15 K; : 303.15 K; : 313.15 K; : 323.15 K; : 333.15 K; : 343.15 K; ♦: of interactions within aqueous mixtures of glyceline at relatively
353.15 K; : 363.15 K]. Solid curves show fits according to the Redlich–Kister higher temperatures are re-emphasized nonetheless.
equation [Eq. (8)] with parameters (Aj ) reported in Table 8.
4. Conclusions

the maximum value of (ln )E is again observed to be in the Densities and dynamic viscosities and their temperature and
water-rich region (xwater ∼0.6–0.7). Also, in general, (ln )E for composition dependence of aqueous mixtures of glyceline in the
our system is found to decrease with increasing temperature. It temperature range 283.15–363.15 K reveal interesting informa-
is worth mentioning that the positive (ln )E for our aqueous DES tion on interactions present within these systems. Negative and
mixture is found to be similar to the positive (ln )E observed significant values of excess molar volume and positive values of
earlier when ionic liquid 1-butyl-3-methylimidazolium hexafluo- logarithmic excess viscosity at all temperature and composition
rophosphate was mixed with acetone, 2-butanone, 3-pentanone, hint at (i) presence of stronger interactions between water and
cyclopentanone, and ethyl acetate, respectively [45]. glyceline as compared to those among water or among glyceline
For aqueous mixtures of glycerol, (ln )E are also reported to be molecules, respectively, and (ii) facile interstitial accommodation
positive with significant values [43]. Further, (ln )E was observed of water within H-bonded network of glyceline. As expected, abso-
to decrease with increasing temperature for (glycerol + water) lute value of excess molar volume and excess logarithmic viscosity
mixtures in the temperature range 273.15–373.15 K [43]. In this decrease as temperature is increased up to 323.15 K. However, a
context, behavior of (glycerol + water) is very similar to that of complex interplay of these factors results in almost no change in
(glyceline + water). This T-dependence hints at H-bonding among excess molar volume as the temperature is raised above 323.15 K.
others to be the most important interaction within our system. Complexity of the aqueous glyceline mixtures at higher tempera-
Stronger interspecies H-bonding between water and glyceline (or tures are highlighted by our measurements.
glycerol and choline chloride) as compared to the intraspecies
H-bonding among water or glyceline may give rise to positive Acknowledgements
(ln )E . Increasing the temperature may weaken the interspecies
H-bonding more than the intraspecies H-bonding thus resulting in This work is generously supported by the Department of Science
decreased values of (ln )E as the temperature is increased. Major and Technology (DST), Government of India through a grant to SP
discrepancy in size between glyceline and water must also con- [grant number SB/S1/PC-80/2012].
tribute to (ln )E being positive.
Finally, our (ln )E were correlated by the Redlich–Kister poly- References
nomial equation [37]. According to combined nearly ideal binary
solvent/Redlich–Kister (CNIBS/R-K) model, the (ln )E in a binary [1] R.D. Rogers, K.R. Seddon, Science 302 (2003) 792–793.
mixture at a constant temperature can be expressed as: [2] L.E. Ficke, J.F. Brennecke, J. Phys. Chem. B 114 (2010) 10496–10501.
[3] G.A. Baker, S.N. Baker, S. Pandey, F.V. Bright, Analyst 130 (2005) 800–808.
[4] Ionic Liquids as Green Solvents: Progress and Prospects, in: R.D. Rogers, K.R.

k
Seddon (Eds.), ACS Symposium Series 856, American Chemical Society, Wash-
(ln )E = xwater xglyceline Aj (xwater − xglyceline )j (8) ington, DC, 2003.
[5] R.P. White, J.E.G. Lipson, Macromolecules 46 (2013) 5714–5723.
j=0
[6] H.N. Lee, N. Newell, Z. Bai, T.P. Lodge, Macromolecules 45 (2012) 3627–3633.
[7] A.J.L. Costa, M.R.C. Soromenho, K. Shimizu, J.M.S.S. Esperança, J.N.C. Lopes, L.P.N.
where Aj and j are the equation coefficients and the degree of the Rebelo, RSC Adv. 3 (2013) 10262–10271.
polynomial expansion, respectively. The numerical values of j can [8] C. Ru␤, B. König, Green Chem. 14 (2012) 2969–2982.
[9] M. Figueiredo, C. Gomes, R. Costa, A. Martins, C.M. Pereira, F. Silva, Electrochim.
be varied to find an accurate mathematical representation of the
Acta 54 (2009) 2630–2634.
experimental data. Regression analysis was performed to fit the [10] A.H. Whitehead, M. Pölzler, B. Gollas, J. Electrochem. Soc. 157 (2010)
polynomials [Eq. (8)] to our experimental data with j = 3 and the D328–D334.
results of the fit are reported in Table 8. It is observed from the r2 [11] K. Haerens, E. Matthijs, A. Chmielarz, B. van der Bruggen, J. Environ. Manage.
90 (2009) 3245–3252.
values that the data shows fair-to-good agreement for many (glyce- [12] J.T. Gorke, F. Srienc, R.J. Kazlauskas, Chem. Commun. 10 (2008) 1235–1237.
line + water) mixtures at lower temperatures; however, the fitting [13] P. Domı nguez de Marı a, Z. Maugeri, Curr. Opin. Chem. Biol. 15 (2011) 220–225.
is not satisfactory at higher temperatures. The solid curves con- [14] A.P. Abbott, T.J. Bell, S. Handa, B. Stoddart, Green Chem. 7 (2005) 705–707.
[15] A.P. Abbott, G. Capper, D.L. Davies, K.J. McKenzie, S.U. Obi, J. Chem. Eng. Data
necting (ln )E in Fig. 5 are obtained from the CNIBS/R-K model fit 51 (2006) 1280–1282.
(with j = 3) as reported in Table 8 which further highlights the cor- [16] D. Lindberg, M. de la Fuente Revenga, M. Widersten, J. Biotechnol. 147 (2010)
relation between the predicted and the experimentally obtained 169–171.
[17] J.T. Gorke, R.J. Kazlauskas, F. Srienc, US2009/0117628 (2009).
values. It is convenient to use a cross-validation method which is [18] Food and Nutrition Board, Institute of Medicine. Dietary Reference Intakes for
a practical and reliable method to test the predictive significance Thiamin, Riboflavin, Niacin, Vitamin B6, Folate, VitaminB12, Pantothenic Acid,
when only little data are available [38]. The more complex nature Biotin, and Choline, National Academic Press, Washington, DC, 1998.
142 A. Yadav et al. / Fluid Phase Equilibria 367 (2014) 135–142

[19] S.H. Zeisel, J.K. Blusztajn, Ann. Rev. 14 (1994) 269–296. [31] C. D’Agostino, R.C. Harris, A.P. Abbott, L.F. Gladden, M.D. Mantle, Phys. Chem.
[20] D. Gultom, A. Songsang, U.T. Meulen, J. Anim. Physiol. Anim. Nutr. 85 (2001) Chem. Phys. 13 (2011) 21383–21391.
1–8. [32] D.M. Cristancho, D.R. Delgado, F. Martinez, M.A.A. Fakhree, A. Jouyban, Rev.
[21] Atwater, Ullmann’s Encyclopedia of Industrial Chemistry, Wiley-VCH Verlag Colomb. Cienc. Quim. Farm. 40 (2011) 92–115.
GmbH & Co. KGaA, Weinheim, Germany, 2000. [33] I.I. Adamenko, L.A. Bulavin, V. Ilyin, S.A. Zelinsky, K.O. Moroz, J. Mol. Liq. 127
[22] R. Esquembre, J.M. Sanz, J.G. Wall, F. del Monte, C.R. Mateo, M.L. Ferrer, Phys. (2006) 90–92.
Chem. Chem. Phys. 15 (2013) 11248–11256. [34] S. Shaukat, R. Buchner, J. Chem. Eng. Data 56 (2011) 4944–4949.
[23] A.P. Abbott, R.C. Harris, K.S. Ryder, C. D’Agostino, L.F. Gladden, M.D. Mantle, [35] F. Koohyar, A.A. Rostami, M.J. Chaichi, F. Kiani, J. Chem. 2013 (2013) 1–10.
Green Chem. 13 (2011) 82–90. [36] C.M. Romero, M.S. Páez, D. Pérez, J. Chem. Thermodyn. 40 (2008) 1645–1653.
[24] Q. Zhang, K.D.O. Vigier, S. Royer, F. Je rôme, Chem. Soc. Rev. 41 (2012) [37] O. Redlich, A.T. Kister, Ind. Eng. Chem. 40 (1948) 341–345.
7108–7146. [38] S. Wold, M. Sjöstrom, L. Eriksson, PLS-regression: A Basic Tool of Chemometrics,
[25] R.B. Leron, A.N. Soriano, M.-H. Li, J. Taiwan Inst. Chem. Eng. 43 (2012) vol. 58, Chemom. Intell. Lab Syst., Netherlands, 2001, pp. 109–130.
551–557. [39] H. Jin, B. O’Hare, J. Dong, S. Arzhantsev, G.A. Baker, J.F. Wishart, A.J. Benesi, M.
[26] R.B. Leron, D.S.H. Wong, M.-H. Li, Fluid Phase Equilib. 335 (2012) 32–38. Maroncelli, J. Phys. Chem. B 112 (2007) 81–92.
[27] K. Shahbaz, S. Baroutian, F.S. Mjalli, M.A. Hashim, I.M. AlNashef, Thermochim. [40] O.O. Okoturo, T.J. VanderNoot, J. Electroanal. Chem. 568 (2004) 167–181.
Acta 527 (2012) 59–66. [41] M.S. Miran, H. Kinoshita, T. Yasuda, M.A.B.H. Susan, M. Watanabe, Phys. Chem.
[28] K.R. Siongco, R.B. Leron, M.-H. Li, J. Chem. Thermodyn. 65 (2013) 65–72. Chem. Phys. 14 (2012) 5178–5186.
[29] N.P. Cheremisinoff, Industrial Solvents Handbook, 2nd ed., Marcel Dekker, USA, [42] H. Jin, B. O’Hare, J. Dong, S. Arzhantsev, G.A. Baker, J.F. Wishart, A.J. Benesi, M.
2003. Maroncelli, J. Phys. Chem. B 115 (2011) 1333.
[30] J.A. Riddick, W.B. Bunger, T.K. Sakano, Techniques of Chemistry, Vol. II, Organic [43] J.B. Segur, H.E. Oberstar, Ind. Eng. Chem. 43 (1951) 2117–2120.
Solvents: Physical Properties and Methods of Purification, John Wiley & Sons, [44] A. Heintz, D. Klasen, J.K. Lehmann, J. Solution Chem. 31 (2002) 467–476.
New York, 1986. [45] J. Wang, A. Zhu, Y. Zhao, K. Zhuo, J. Solution Chem. 34 (2005) 585–596.