Food Chemistry 320 (2020) 126566

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Thermophysical and rheological properties of sorbitol + ([mmim] T

(MeO)2PO2) ionic liquid solutions: Solubility, density and viscosity
⁎
Mohammad Abbasi, Gholamreza Pazuki , Ahmadreza Raisi, Mojhdeh Baghbanbashi
Department of Chemical Engineering, Amirkabir University of Technology (Tehran Polytechnic), Iran

A R T I C LE I N FO A B S T R A C T

Keywords: In this study, the solubility, density and viscosity of sorbitol as a sugar alcohol in the ([mmim](MeO)2PO2) ionic
Thermophysical and rheological properties liquid (IL) were measured. The results indicated that sorbitol is highly soluble in this IL. The Flory-Huggins
Sugar alcohols model with an average value for the χ parameter was successfully applied to predict the solubility of sorbitol in
Sorbitol IL. The thermodynamic properties such as enthalpy, entropy and Gibbs free energy of dissolution were obtained
Ionic liquid
using experimental solubility data, which demonstrated that the dissolution process is endothermic and non-
Solubility
Density
spontaneous and includes an entropy increase. In addition, the apparent molar volume, apparent molar ex-
Viscosity pansion and thermal expansion coefficient were calculated. The study of the rheological behavior revealed that
the sorbitol/IL solution is Newtonian and the Arrhenius, Litovitz, Orrick-Erbar-Type and Vogel-Fulcher-Tamman
models were used to correlate the viscosity data.

1. Introduction responsible for physical properties such as melting point, viscosity and
density, while anions control the chemical properties and interactions.
The use of renewable sources for the production of energy and ILs are described as designer solvents because their properties are ad-
materials has become an important issue in the past few decades and justable for different applications by changing the anion and cation. ILs
leads to intensive growth in use of petrochemical explorations and have very low vapor pressure, so they can be used as an alternative for
environmental issues (Carneiro, Rodríguez, Held, Sadowski, & Macedo, many volatile organic solvents, avoiding solvent losses and environ-
2014; Carneiro, Rodríguez, & Macedo, 2013). Today, the use of biomass mental pollution. These solvents can dissolve a wide range of organic
is one of the best options to solve these problems. Biomass is composed and inorganic substances and have better solubility than common or-
of organic molecules containing carbon and hydrogen elements ganic solvents. ILs have a wide liquid phase range and are stable in the
(Mohan, Goud, & Banerjee, 2015). Cellulose, hemicellulose and lignin air, water and while facing intense interaction and high temperature.
are biopolymers composing biomass. These biopolymers can be pro- They act as catalysts in some operations like hydrolysis and increase
cessed to produce fuels, energy and added-value chemicals yield. These solvents can be used in the selective extraction of poly-
(Baghbanbashi & Pazuki, 2016; Carneiro, Rodríguez, & Macedo, 2013). saccharides (Sheldon, 2001; Plechkovaa and Seddon, 2008). All of these
However, whilst the bio-refining concept is a suitable alternative for properties present ILs as a good solvent for biomass applications and
petrochemical industry, biomass processing has some challenges that bio-refining processes. The solubility of natural polymers present in
have to be considered. Pretreatment is an important process in bio-re- biomass, like cellulose, in ILs can be modified by changing the cation or
finery. Pretreatment processes usually need high temperatures, large anion. Anions forming hydrogen bonds with hydroxyl groups are ef-
amounts of water and use of corrosive chemicals. These standard pro- fective for the dissolution of biopolymers. Small size and alkalinity of
cesses can be undesirable because of the high operating costs and safety anions increase the biopolymers solubility (Fang, 2014).
concerns. On the other hand, the process risks and limitations of ex- Sugar alcohols are valuable additives that can be obtained from bio-
isting common solvents create the need for a new clean solvent refinery (Baghbanbashi & Pazuki, 2014, 2016). The knowledge of vo-
(Karthikeyan, Mehariya, Woon, & Wong, 2017). lumetric and transport properties of binary systems containing sugar
Ionic liquids (ILs) are molten salts that are liquid at temperatures alcohols in ILs is important to select the best solvent for these processes.
below 100 °C. They are compounds composed of ions, an organic cation On the other hand, these properties are important to design bio-refinery
and anion, which can be organic or inorganic. Cations in ILs are processes. Some researchers have experimentally and theoretically

⁎
Corresponding author.
E-mail address: ghpazuki@aut.ac.ir (G. Pazuki).

https://doi.org/10.1016/j.foodchem.2020.126566
Received 7 October 2019; Received in revised form 20 February 2020; Accepted 4 March 2020
Available online 05 March 2020
0308-8146/ © 2020 Elsevier Ltd. All rights reserved.
M. Abbasi, et al. Food Chemistry 320 (2020) 126566

studied the volumetric and transport properties of sugar alcohols in ILs. employed to measure the solubility of sorbitol in [mmim](MeO)2PO2 at
For example, Carneiro et al. (Carneiro, Rodríguez, & Macedo, 2012a, various temperatures. After preliminary experiments, sorbitol was
2012b; Carneiro, Rodríguez, et al., 2013) measured the solubility of added in slight excess to a fixed amount of [mmim](MeO)2PO2 resulting
saccharides (glucose, fructose, xylose and galactose) in different ILs. in a saturated solution of sugar alcohol in IL. The solution was mag-
They used the NRTL and UNIQUAC models for modeling the solubility netically stirred for 48 h. A jacket in which water was flowing from a
of saccharides in ILs. In other works, Carneiro et al. (Carneiro, Held, thermostatic bath (Wisecircu, Germany) was applied to keep the tem-
Rodríguez, Sadowski, & Macedo, 2013; Carneiro et al., 2014) measured perature constant with a precision of, 0.1 K at 303.15, 313.15, 323.15
the solubility and density of sugars (glucose and fructose) and sugar and 333.15 K. After saturation of the solution, the stirring was ceased.
alcohols (sorbitol and xylitol) in different ILs and correlated the ex- After a while, a clear top phase appeared and the excess solid de-
perimental data by the PC-SAFT model. Ji and Held (Ji & Held, 2016) posited. Then, sampling was done by 2 ml syringes, three times. The
studied the effects of temperature and pressure on the density of sys- samples were quantitatively analyzed by high–performance liquid
tems containing ILs using the PC-SAFT model. They used ILs containing chromatography (HPLC) using an Asahipak NH2P-50
five different cations and twelve different anions from 280 to 380 K and 250 mm × 4.6 mm column with acetonitrile/water (75/25% v/v) as a
pressures up to 1200 bar. Schmidt et al. (Schmidt et al., 2012) measured mobile phase flowing at 1.0 ml.min-1. The detection was carried out by
density and viscosity of [emim][EtSO4] from 10 to 140 °C and pressure the refractive index at 303.15 K. Moreover, the injection was made
to 140 MPa. They suggested an equation for viscosity as a function of using PLATIN blue with 10 L volume. Using this apparatus, the cali-
temperature, pressure and density. Moreover, Shakeel et al. (Shakeel, bration curve of sorbitol was plotted and concentrations of samples
Haq, & Siddiqui, 2015) measured the solubility of vanillin as a food were calculated (Zarei, Abdolrahimi, & Pazuki, 2019).
medicine in ten benign solvents at 298–318 K and estimated the value
of thermodynamic functions such as molar enthalpy, entropy and Gibbs 2.3. Density and viscosity
free energy of dissolution. In addition, they obtained vanillin solubility
at 298–318 K and 0.1 MPa in various mixtures of propane-1, 2- After the determination of solubility, the density and viscosity of the
diol + water, where water acted as co-solvent (Shakeel, Haq, Siddiqui, sorbitol/IL solution were measured below the solubility concentration.
Alanazi, & Alsarra, 2015). The volumetric properties and taste behavior To measure the thermophysical properties of the system (density and
of sucrose have been studied in aqueous solutions of ionic liquid (3- viscosity), solutions with various concentrations up to a value less than
hydroxypropylammonium acetate) at 293.15–318.15 K and atmo- the saturated solubility at each temperature were prepared. Then the
spheric pressure by Singh et al. (Singh, Chhotaray, & Gardas, 2015). viscosity and density for each concentration and temperature were
They determined apparent molar volume, apparent specific volume, measured. The density of each sample was measured using a 5 ml
apparent molar isentropic compressibility and apparent specific isen- pycnometer at different temperatures. The viscosity of the samples was
tropic compressibility using measured density and sound velocity. They measured using Brookfield (DV-II + Pro extra viscometer by Wingather
applied these results to investigate solute-cosolute interactions. Omran V3.01 program) in various spindle periods at 30, 40, 50 and 60 °C. The
et al. (Omran, Kiani, Saraei, Koohyar, & Rostami, 2019) measured the precisions of density and dynamic viscosity measurements were
physicochemical properties (viscosities and refractive indices) of three ± 3 × 10−3g . cm−3 and ± 1 × 10−4mPa. s , respectively. All experiments
food additives, such as aqueous solution of citric acid and ternary so- were replicated three times.
lutions of water + D-sorbitol + citric acid and water + glycerol + ci-
tric acid with citric acid mole fractions of 0.03–0.21 at 2.4. FTIR
293.15–323.15 K. Zafarani-Moattar et al. (Zafarani-Moattar, Shekaari,
& Haji Agha, 2019) studied the thermophysical properties and taste The FTIR analysis with a Nicolet Nexus 670 spectrometer (Nicolet
behavior of sucrose in aqueous solutions of ionic liquids (1-octyl-3- Instrument Co., Madison, WI, USA) was used to study the interactions
methyl imidazolium bromide and 1-octyl-3-methyl imidazolium between sorbitol and IL. The FTIR analysis was accomplished at wa-
chloride) by applying volumetric, compressibility and viscometric venumbers in the range of 400 cm−1 to 4000 cm−1 at ambient condi-
properties at 288.15–318.15. They determined the interactions of su- tions and with a spectral resolution of 4 cm−1.
crose and ionic liquids based on the measured properties.
In this work, the solubility of sorbitol in 1,3-dimethylimidazolium 3. Thermodynamic modeling
dimethyl phosphate ([mmim](MeO)2PO2) was measured at tempera-
tures of 303 to 333 K. Then, density and viscosity of the mixture at 3.1. Solubility calculation of sorbitol in IL
concentrations lower than the solubility concentration were measured
at the temperature range of 303 to 333 K. The solubility of sugar al- The solubility of sorbitol in [mmim](MeO)2PO2 was calculated
cohols in different ILs was studied using solubility parameter and Flory- using the Flory-Huggins (FH) model. The IL activity coefficients as the
Huggins model and the solubility values were predicted by determining solvent and the sugar as solute were obtained as follows (Kumar &
the χ12 parameter. In addition, two other thermodynamic properties Gupta, 2003):
including density and viscosity were studied and discussed.
φ1 1
ln γ1 = ln + ⎛1 − ⎞ φ2 + χ12 φ22
2. Materials and methods x1 ⎝ r⎠ (1)
φ
2.1. Materials ln γ2 = ln 2 + φ1 (1 − r ) + χ12 rφ12
x2 (2)

In these experiments, [mmim](MeO)2PO2 IL with purity higher than where φi is volume fraction of component i defined as:
98% and sorbitol with purity higher than 99.5% (biochemistry grade) x i Vi
were used and were supplied by Merck. φi = 2
, i = 1, 2
∑ x i Vi
i=1 (3)
2.2. Solubility
In addition, r is the volume ratio of the solute to the solvent ( V2 ),
V1
The solubility of sorbitol in ionic liquid was measured by an iso- x i and χ12 are mole fraction and Flory-Huggins interaction parameter
thermal technique according to the procedure presented by Carneiro between the solute and solvent, respectively. Applying the solid–liquid
et al. (Carneiro, Held, et al., 2013). The isothermal technique was phase equilibria, the solubility of the sorbitol in [mmim](MeO)2PO2 can

2
M. Abbasi, et al. Food Chemistry 320 (2020) 126566

be obtained as below: 0

1 Δhfus Tm -0.2
x2 = exp ⎧− ⎛ − 1⎞ ⎫
γ2 ⎨
⎩ RTm ⎝ T ⎠⎬⎭ (4)
-0.4

ln(x2)
In this equation, Δhfus and Tm are the enthalpy of fusion and the
-0.6
normal melting point, respectively. The activity coefficient of the sugar
can be calculated from equation (2). The Flory- Huggins parameter ( χ12 )
-0.8
is related to the solubility parameters of solvent and solute as follows
(Rubinstein & Colby, 2003): -1
v
χ12 = 1 (δ1 − δ2)2 -1.2
RT (5)
-0.0002 -0.00015 -0.0001 -0.00005 0 0.00005 0.0001 0.00015 0.0002
where, δi is the Hildebrand solubility parameter of component i 1/T-1/Thm
which is the square root of adhesion energy (Rubinstein & Colby, 2003):

Δu vap 1/2
Δh vap − RT ⎞
1/2
Fig. 1. The changes of ln x2 versus ( 1
T
−
1
Thm ).
δt = ⎛ i
⎜
⎞
⎟ =⎛ i
⎜ ⎟

⎝ vi ⎠ ⎝ vi ⎠ (6) Gibbs free energy of dissolution are obtained from the slope and in-
The Flory -Huggins parameter obtained from equation (5) is always tercept of the diagram of ln x2 versus ( 1
T
−
1
Thm ), which is presented in
positive representing the low solubility of solute in the solvent. How- Fig. 1 (Apelblat & Manzurola, 1999):
ever, this equation is not applicable for a system containing a highly
soluble solute. Therefore, the Flory- Huggins parameter is modified as ⎡ ⎤
∂ ln x2
°
Δhdiss = −R ⎢ ⎥
follows (Rubinstein & Colby, 2003):
v1
⎢∂ 1 − 1
⎣ T Thm ( ) ⎥
⎦ (8)
χ12 = A + (δ1 − δ2 )2
RT (7) °
Δgdiss = −RThm k (9)
where A is an adjustable parameter. In the present study, the parameter J
In these equations k is the intercept of ln x2 line, R = 8.314 mol . K
A was calculated from the available solubility data of the sugar alcohols
and Thm is the harmonic average temperature and can be calculated by
including sorbitol and xylitol in [mmim](MeO)2PO2 IL. By applying the
the following equation:
obtained results, the solubility of sorbitol in [emim][EtSO4], [emim]
[DCA] and [bmim][DSA] ionic liquids was predicted and reported in N
Thm = N
Table S1. Therefore, the proposed model is a predictive model. 1
∑ Ti
i=1 (10)
4. Results and discussion
The entropy of dissolution can be obtained from the calculated
enthalpy and Gibbs free energy of dissolution as follows:
4.1. Solubility of sorbitol in [mmim](MeO)2PO2 IL
° °
°
Δhdiss − Δgdiss
The experimental results of sorbitol solubility in [mmim](MeO)2PO2 Δsdiss =
Thm (11)
at 303–333 K and 0.1 MPa are reported in Table 1. The HPLC results are
reported in Fig. S1. The values of thermodynamic properties for the binary solution of
As can be seen in Table 1, the solubility of sorbitol in the IL in- sorbitol and [mmim](MeO)2PO2 are reported in Table S3. The values of
creases with temperature. From the viscosity results, it is observed that the thermodynamic properties are positive indicating that the process
IL viscosity decreases by increasing temperature. The lower viscosity at of sorbitol dissolution in [mmim](MeO)2PO2 is an endothermic and
high temperatures causes a better mixing and the solubility of sorbitol non-spontaneous process and entropically favorable. The low values of
in the IL increases. In order to study the interactions between sorbitol enthalpy and Gibbs free energy of dissolution represent that a lower
and IL, the FTIR spectrum analysis was studied. The result of FTIR is amount of energy is needed for breaking the intermolecular interactions
presented in Fig. S2. In the FTIR analysis, the FTIR spectra are divided and, therefore, the solubility increases. In addition, the positive entropy
into two regions including the fingerprint region in 1300–4000 indicates that after dissolving, the molecules have more irregularity
cm−1range of wavenumbers and the OH and CH functional groups re- than before dissolution (Carneiro, Rodríguez, et al., 2013).
gion in the range of 2500 − 3500 cm−1 range. The main peaks re- The modeling of the solubility of sugar in IL is discussed by con-
presenting the functional groups are shown in Table S2 (Pavia, sidering the empirical correlations and the Flory-Huggins thermo-
Lampman, Kriz & Vyvyan, 2009). As presented in Table S2, the higher dynamic model. The Flory- Huggins interaction parameter ( χ12 ) is cal-
intensity peak is assignable to hydroxyl band (OH stretching), in- culated using the solubility parameter. The first applied equation is the
dicating the formation of strong hydrogen bonds during the sorbitol Vant-Hoff relation (Shakeel & Anwer, 2015):
dissolution in the IL [mmim](MeO)2PO2. The thermodynamic proper- b
ties of dissolution such as enthalpy, entropy and Gibbs free energy can ln x2 = a +
T (12)
be obtained by applying the solubility data. The modified Van’t Hoff
method was applied to calculate these properties. The enthalpy and the where a and b are constants which are obtained from the linear re-
gression of ln x versus 1 T . Also, the modified Apelblat relation (equation
Table 1
(13)) was used for correlating the solubility data (Apelblat &
Experimental results of the sorbitol solubility in [mmim](MeO)2PO2 IL at var- Manzurola, 1999):
ious temperatures. B
ln x2 = A + + C ln T
T [K] 303 313 323 333 T (13)

Solubility (g sorbitol/g solution) 0.305 0.383 0.476 0.570 where A, B and C are the constants of model. The values of the models’
constants, R2 and RMSD are reported in Table S4. In addition, the Bu-
Standard uncertainties: u (T ) = ± 0.01K , u (w2) = ± 0.07 . chowski-Ksiazaczak (BK) equation that was first proposed for the solid-

3
M. Abbasi, et al. Food Chemistry 320 (2020) 126566

liquid equilibrium, was applied for correlating solubility of sorbitol in Table 2

[mmim](MeO)2PO2. The BK equation is written as (Książczak, 1999): The excess molar volume values (v E ) at various temperatures.

λ (1 − x2 ) ⎫ 1 1 ⎤ x2 303 K 313 K 323 K 333 K

ln ⎧1 + = λh ⎡ −
⎨
⎩ x2 ⎬
⎭ ⎢
⎣ T Thm ⎥
⎦ (14) 0.06 −0.1387 −0.0987 −0.1157 −0.1455
0.12 −0.1950 −0.1043 −0.1790 −0.1571
where λ and h are constants and Tm is the sorbitol′s melting point. 0.18 −0.1741 −0.1534 −0.1534 −0.1884
The solubility of sorbitol in [mmim](MeO)2PO2 IL was calculated by 0.23 −0.3090 −0.3327 −0.2119 −0.2250
applying the Flory- Huggins model as a predictive model. According to 0.29 −0.4158 −0.3080 −0.2607 −0.2922
the wide range of the values of parameter A (-4.309 to −3.94), its value 0.34 −0.4595 −0.3983 −0.3622 −0.3504
0.40 −0.3804 −0.3069 −0.2898
for the proposed model was chosen to be −4. The solubility results of
0.45 −0.3774 −0.2733
the modified Flory–Huggins model are reported in Table S5. The 0.50 −0.3680 −0.3102
average value of AARD% for solubility prediction is 3.924. As can be 0.55 −0.3627
seen from the results, the Flory-Huggins model can successfully predict 0.60 −0.3276
the solubility of sorbitol in [mmim](MeO)2PO2. The calculated solubi-
lity of xylitol in [mmim](MeO)2PO2 is reported in Table S11. For
comparison, the sorbitol solubility in water at various temperatures, volume of the ideal solution (v Id ) (Bhuiyan & Uddin, 2008).
which was reported by van der Sman (2017) are given in Table S12. The The values of the calculated excess molar volume for the solution
high solubility of sorbitol in water is due to the strong hydrogen reported in Table 2 are negative. A negative value of the excess molar
bonding between the solute and solvent. volume of solution can be caused by various factors. The most im-
portant factor observed here is the formation of attractive interactions
4.2. Density and binary volumetric properties such as hydrogen bonds during dissolution, as confirmed by the FTIR
analysis (Table S1) and high dissolution of sorbitol in IL. As the excess
The experimental data of density for binary mixtures of different volume values indicate, the solution has a negative deviation from the
concentrations of sorbitol at 303–333 K are reported in Table S13. The ideality. On the other hand, by increasing the temperature, the negative
results for measuring the density at various temperatures are shown in deviation is reduced. Since the main reason for negative deviation in
Fig. 2. As can be seen, the density data at any given temperature is this solution is the formation of hydrogen bonds, decreasing the de-
linearly fitted with high precision. This indicates the linear relationship viation by increasing the temperature indicates a decrease in the den-
of the solution density with concentration at different temperatures. As sity of hydrogen bonds, which causes expansion or increase in the ex-
expected, the density decreases with increasing temperature and it cess volume of the solution (Bhuiyan & Uddin, 2008; Zarei et al., 2019).
enhances with increasing the mass fractions of sorbitol. In addition, the The coefficients of the Redlich-Kister model were calculated for the
changes in density at lower temperatures are more severe, which is binary solution of sorbitol and [mmim](MeO)2PO2 IL by applying the
completely evident from the slope of the lines. The result of the linear v E data. The model is written as follows (Bhuiyan & Uddin, 2008):
fitting of these data at various temperatures may be seen in Table S14. n

The excess molar volume of a N-component system can be obtained v E = x1 x2 ∑ ai (x2 − x1 )i

i=0 (17)
using the density of components as follows:
N where n is the polynomial power and ai is the polynomial coefficient.
3 1 1⎞
v E ( cm mol ) = ∑ xi Mi ⎛ ρ ⎜ − ⎟
E
The values of v can be plotted in terms of (x2 − x1) and the best fitted
i=1 ⎝ ρi ⎠ (15) x 1 x2
polynomial would be obtained. Accordingly, a fourth degree poly-
The above equation can be written for a binary mixture: nomial was chosen. The model’s coefficients at various temperatures
and the AAD% values are reported in Table S15.
3 (x1 M1 + x2 M2 ) ⎛ x1 M1 ⎞ ⎛ x2 M2 ⎞ ⎤
v E ( cm mol ) = −⎡
⎢⎜ ρ ⎟ + ⎜ ρ ⎟⎥
One of the apparent molar properties is the apparent molar volume,
ρ ⎣⎝ 1 ⎠ ⎝ 2 ⎠⎦ (16) which for a binary solution is expressed as below:
Here, x i , Mi and ρi are the mole fraction, molecular weight, and V = n1 V1 + n2 V2ϕ (18)
density of component i , respectively and ρ is the solution density. The
first part of the right-hand side of equation (16) represents the real where n1, n2 , V1 and V2ϕ
are the number of moles of solvent, the moles of
molar volume of the solution, and the second part indicates the molar solute in solution, the molar volume of pure solvent, and the apparent
molar volume of solute, respectively. The apparent molar volume for
1.36 the solute (sorbitol) in the solute-solvent mixture (sorbitol-IL) can be
1.35 written as (Shekaari, Mansoori, & Sadeghi, 2008):
1.34 M2 (ρ − ρ1 )
V2ϕ = −
1.33 ρ mρρ1 (19)
1.32
Here, ρ and ρ1 are the density of solution and pure solvent, re-
ȡ(g/cm3)

1.31
spectively, m is the solute molality (mol/ kg ), and M2 is molecular
1.3 weight of the solute (Shekaari et al., 2008). The apparent molar volume
1.29 of sorbitol in [mmim](MeO)2PO2 is calculated using the density data
1.28 based on equation (20) at various temperatures. The results of the ap-
1.27
parent molar volume of sorbitol are presented in Table S16.
By fitting the apparent molar volume data, a relationship between
1.26
the apparent molar volume and molality of sorbitol is obtained. The
1.25 value of apparent molar volume at the very low concentrations is called
0 0.1 0.2 0.3 0.4 0.5 0.6
w 2% limited apparent molar volume (V ϕ0 ), which is equal to the partial molar
volume. The value of the apparent molar volume is equal to the inter-
Fig. 2. The measured density data at different temperatures. ♦T = 303 K, cept of fitting the obtained equations (Table S17). As can be seen in
■T = 313 K, ▲ T = 323 K, × T = 333 K. Table S17, by increasing temperature, the value of concentration

4
M. Abbasi, et al. Food Chemistry 320 (2020) 126566

coefficient decreases in the model obtained for apparent molar volume, 9

which indicates a decrease in interactions at higher temperatures and 8
confirms what was seen in additional volume changes. The limited
apparent molar volume (V ϕ0 ) can be fitted with a quadratic polynomial 7
at the studied temperatures. The result is as follows: 6

V ϕ0 = −0.0007T 2 + 0.4716T + 67.189 R2 = 0.9917 (20) 5

ln(Ș)
According to the obtained equation for V ϕ0 , the limited apparent 4
molar expansion can be calculated as follows: 3
0 0.05
0
2 0.1 0.15
Eϕ0 = ⎛ ∂V ϕ
⎜
∂T ⎞ = −0.0014T + 0.4716
⎟
0.2 0.25
⎝ ⎠P (21) 0.3 0.35
1
0.4 0.45
As can be seen, the limited apparent molar expansion (Eϕ0 ) is posi- 0
0.5 0.55
tive at all temperatures. On the other hand, the value of this parameter 0.00295 0.003 0.00305 0.0031 0.00315 0.0032 0.00325 0.0033 0.00335
decreases with increasing temperature. 1/T

Fig. 3. The plot of ln η versus 1/ T for correlating the Arrhenius model.

4.3. Viscosity
4.3.2. Litovitz model
The viscosity and rheological properties of a fluid are one of the The Litovitz model shows the temperature dependency of the so-
most important properties in terms of the effect of material handling lution viscosity as follows (Haj-Kacem et al., 2014):
and equipment design. The results of the viscosity measurement of the
solution of sorbitol in [mmim](MeO)2PO2 at various concentrations and ln η = ln A − E / RT 3 (23)
temperatures are given in Table 3. Unlike the Arrhenius model, the Litovitz model uses 1
instead of
Fig. S3 represents the viscosity changes with temperature at various T3
reversing the absolute temperature. The correlated result is shown in
concentrations of sorbitol in [mmim](MeO)2PO2 IL. It was observed
Fig. S4 and the model’s parameters are reported in Table S19. These
that the solution viscosity decreases with increasing temperature.
data of the Litovitz model results in Table S19 indicate an increase in
However, the rate of viscosity changes is lower at high temperatures.
the activation energy and an increase in the slope of viscosity changes
On the other hand, increasing the concentration of sorbitol in the so-
at higher concentrations, similar to the Arrhenius model.
lution enhances the viscosity and the magnitude of the changes is more
important at higher concentrations. Subsequently, models like Ar-
4.3.3. Orrick-Erbar-type model
rhenius, Litovitz, Orrick-Erbar-type and Vogel-Fulcher-Tamman were
This model combines viscosity and density properties (Poling,
proposed for modeling viscosity data.
Prausnitz, & O’Connell, 2001):
η B
4.3.1. Arrhenius model ln =A+
ρM T (24)
In order to investigate the temperature dependency of viscosity, the
Arrhenius model was developed as follows (Haj-Kacem et al., 2014): Here ρ , A and B are the density and model′s constants (Ahosseini &
Scurto, 2008). In order to correlate the viscosity results using the Or-
ln η = ln A − E / RT (22) η
rick-Erbar-type model, the value of ln ρM is calculated by applying the
where η , A, E and R are the viscosity, constant of model, activation measured density data and plotted versus 1/ T . The results are shown in
energy of shear rate and the gas constant (8.314 J/mol/K), respectively. Fig. S5. The molecular weight of the solution is calculated using the
To calculate the parameters of the Arrhenius model, the plot of ln η average molecular weight of the components based on the following
versus 1/ T is represented in Fig. 3. The values of the coefficient and the equation:
intercept are reported in Table S18. As can be seen from the results in M = x1 M1 + x2 M2 (25)
Table S18, with increasing concentration, the value of activation energy
where x i and Mi are the mole fraction and molecular weight of com-
of the shear rate increases linearly which means the energy needed to
ponent i .
move the fluid layers on each other increases at higher concentrations.
The calculated parameters for the Orrick-Erbar-type model (Table
On the other hand, increasing this parameter causes the slope of visc-
S20) also confirm the previous observations. The viscosity increases by
osity changes to be enhanced at higher concentrations.
increasing the concentration and the rate of viscosity changes is more
important at higher concentrations. Comparing the calculated coeffi-
Table 3
Viscosity values of sorbitol solution in [mmim](MeO)2PO2 (mPa. s ). cients for the Orrick-Erbar-type model with the coefficients of the Ar-
rhenius model reveals that the value of slope in the Orrick-Erbar-type
Mass fraction 303 K 313 K 323 K 333 K
model is lower. This is due to the effect of introducing the density in the
0 65.3 41.15 30.3 20.8 viscosity equation.
0.05 89. 2 54.6 36.4 25.7
0.10 92.6 55.8 37.7 26.57 4.3.4. Vogel-Fulcher-Tamman model:
0.15 119.5 71.68 47.12 32.58
The Vogel-Fulcher-Tamman model (VFT) is one of the most widely
0.20 218.2 104.8 62.05 41.39
0.25 568.5 301.6 145.4 79.68 used models for ILs (Fulcher, 1992; Tammann & Hesse, 1926; Vogel,
0.30 855.8 526.6 258.4 133.8 1921):
0.35 759.9 389.9 170
0.40 499.3 222.4
B
ln η = ln A +
0.45 634.2 294.9 T − T0 (26)
0.50 379.7
0.55 658.6 Here, A , B and T0 are adjustable parameters. By correlating the
experimental values, the parameters of this model were calculated as

5
M. Abbasi, et al. Food Chemistry 320 (2020) 126566

shown in Table S21. Since the VFT model has three adjustable para- model in obtaining phase behavior of aqueous solutions of sugars. Journal of
meters, the measured values were not sufficient at concentrations Molecular Liquids, 195, 47–53. https://doi.org/10.1016/j.molliq.2014.01.028.
Baghbanbashi, M., & Pazuki, G. (2016). Modelling and thermodynamic assessment of
higher than 0.30, the model is not correlated for values above this biofuel production process from biomass with the SAFT-VR approach. International
concentration. As can be seen, the data are well fitted on the VFT Journal of Hydrogen Energy, 41(38), 16797–16806. https://doi.org/10.1016/j.
ijhydene.2016.07.150.
model, whilst at several concentrations, the model values and experi- Baghbanbashi, M., & Pazuki, G. (2016). Application of SAFT‐VR Equation of State for
mental data are exactly equal. Prediction of Thermophysical Properties of Sugar Solutions. Journal of Food Process
Engineering, 39(6), 601–609. https://doi.org/10.1111/jfpe.12253.
Bhuiyan, M., & Uddin, M. (2008). Excess molar volumes and excess viscosities for mix-
4.4. Flow behaviour of sorbitol -[mmim](MeO)2PO2 IL solutions tures of N, N-dimethylformamide with methanol, ethanol and 2-propanol at different
temperatures. Journal of Molecular Liquids, 138(1), 139–146.
Carneiro, A. P., Held, C., Rodríguez, O., Sadowski, G., & Macedo, E.n. A. (2013).
To determine the type of fluid, the shear stress should be de-
Solubility of sugars and sugar alcohols in ionic liquids: Measurement and PC-SAFT
termined at different shear rates. The shear stress diagrams based on modeling. The Journal of Physical Chemistry B, 117(34), 9980–9995.
shear rates at different temperatures and concentrations are presented Carneiro, A. P., Rodríguez, O., Held, C., Sadowski, G., & Macedo, E.n. A. (2014). Density
of mixtures containing sugars and ionic liquids: Experimental data and PC-SAFT
in Table S6. As can be seen, all of the figures are linearly fitted with modeling. Journal of Chemical & Engineering Data, 59(10), 2942–2954.
high precision. The linear fitting of shear stress in terms of shear rate Carneiro, A. P., Rodríguez, O., & Macedo, E. A. (2012b). Solubility of xylitol and sorbitol
indicates that the fluid has Newtonian behavior. in ionic liquids–Experimental data and modeling. The Journal of Chemical
Thermodynamics, 55, 184–192.
Carneiro, A. P., Rodríguez, O., & Macedo, E. A. (2012a). Solubility of monosaccharides in
5. Conclusion ionic liquids–Experimental data and modeling. Fluid Phase Equilibria, 314, 22–28.
Carneiro, A. P., Rodríguez, O., & Macedo, E.n. A. (2013). Fructose and glucose dissolution
in ionic liquids: Solubility and thermodynamic modeling. Industrial & Engineering
In this work, the thermodynamic and thermophysical properties of Chemistry Research, 52(9), 3424–3435.
solutions containing sorbitol and [mmim](MeO)2PO2 IL was measured Fang, Z. (2014). Production of biofuels and chemicals with ionic liquids. Springer.
Fulcher, G. S. (1992). Analysis of Recent Measurements of the Viscosity of Glasses. Journal
at temperatures of 303, 313, 323 and 333 K. The Flory-Huggins para- of the American Ceramic Society, 75(5), 1043–1055.
meter value ( χ ) was obtained for the solubility of sorbitol and xylitol Haj-Kacem, R. B., Ouerfelli, N., Herráez, J. V., Guettari, M., Hamdae, H., & Dallel, M.
sugar alcohols in [emim][EtSO4], [emim][DCA] and [bmim][DSA] ILs. (2014). Contribution to modeling the viscosity Arrhenius-type equation for some
solvents by statistical correlations analysis. Fluid Phase Equilibria, 383, 11–20.
The negative values of χ demonstrate the high solubility of the sugar Ji, X., & Held, C. (2016). Modeling the density of ionic liquids with ePC-SAFT. Fluid Phase
alcohols in the ILs. The solubility of sorbitol in [mmim](MeO)2PO2 was Equilibria, 410, 9–22.
Karthikeyan, O. P., Mehariya, S., Woon, J., & Wong, Ch. (2017). Bio-refining of food
calculated from experimental data. The average Flory-Huggins para- waste for fuel and value products. Energy Procedia, 136, 14–21.
meter and the AARD% showed that the Flory-Huggins model can pre- Książczak, H. B.a. A. (1999). Thermodynamics of Solubility of Solids in Liquids. Polish
dict the sorbitol solubility, precisely. In addition, the apparent molar Journal of Chemistry, 73, 1631–1663.
Kumar, A., & Gupta, R. K. (2003). Fundamentals of polymer engineering, revised and ex-
volume was calculated by applying the experimental density results of panded. CRC Press.
the binary solution of sorbitol and [mmim](MeO)2PO2 and negative Mohan, M., Goud, V. V., & Banerjee, T. (2015). Solubility of glucose, xylose, fructose and
galactose in ionic liquids: Experimental and theoretical studies using a continuum
values were obtained at all temperatures and weight fractions in-
solvation model. Fluid Phase Equilibria, 395, 33–43.
dicating attractive interactions between the sorbitol and [mmim] Omran, M. M., Kiani, F., Saraei, A. G.-H., Koohyar, F., & Rostami, A. M. (2019).
(MeO)2PO2. The flow curve for the sorbitol/IL solution showed that the Determination and study on refractive indices and viscosities of aqueous solutions of
citric acid, (citric acid + glycerol), and (citric acid + d-sorbitol) at T = 293.15
rheological behavior of this system is Newtonian. The viscosity of the K–323.15 K and atmospheric pressure. Food Chemistry, 289, 436–442.
solution was correlated by the Arrhenius, Litovitz, Orrick-Erbar-Type Pavia, D. L., Lampman, G. M., Kriz, G. S., & Vyvyan, J. R. (2009). Introduction to
and Vogel-Fulcher-Tamman models and their parameters were ob- Spectroscopy, 4th ed., Cengage Learning.
Plechkovaa, N. V., & Seddon K. R. Seddon (2008). Applications of ionic liquids in the
tained. The Fulcher-Tamman model was the best model for the viscosity chemical industry, Chemical Society Reviews, 37, 123-150.
prediction due to its low AARD%. Poling, B. E., Prausnitz, J. M., & O’Connell, J. P. (2001). The properties of gases and liquids
(5th ed.). McGraw-Hill.
Rubinstein, M., & Colby, R. H. (2003). Polymer physics. OUP Oxford.
CRediT authorship contribution statement Schmidt, H., Stephan, M., Safarov, J., Kul, I., Nocke, J., Abdulagatov, I. M., & Hassel, E.
(2012). Experimental study of the density and viscosity of 1-ethyl-3-methylimida-
zolium ethyl sulfate. The Journal of Chemical Thermodynamics, 47, 68–75.
Mohammad Abbasi: Conceptualization, Methodology, Software. Shakeel, F., & Anwer, M. K. (2015). Dissolution thermodynamics and solubility of sily-
Gholamreza Pazuki: Writing - original draft, Supervision. Ahmadreza marin in PEG 400-water mixtures at different temperatures. Drug Development and
Raisi: Supervision, Writing - review & editing. Mojhdeh Industrial Pharmacy, 41(11), 1819–1823.
Shakeel, F., Haq, N., & Siddiqui, N. A. (2015). Solubility and thermodynamic function of
Baghbanbashi: Writing - original draft. vanillin in ten different environmentally benign solvents. Food Chemistry, 180,
244–248.
Shakeel, F., Haq, N., Siddiqui, N. A., Alanazi, F. K., & Alsarra, I. A. (2015). Solubility and
Declaration of Competing Interest thermodynamic behavior of vanillin in propane-1,2-diol+water cosolvent mixtures
at different temperatures. Food Chemistry, 188, 57–61.
Shekaari, H., Mansoori, Y., & Sadeghi, R. (2008). Density, speed of sound, and electrical
The authors declare that they have no known competing financial conductance of ionic liquid 1-hexyl-3-methyl-imidazolium bromide in water at dif-
interests or personal relationships that could have appeared to influ- ferent temperatures. The Journal of Chemical Thermodynamics, 40(5), 852–859.
ence the work reported in this paper. Sheldon, R. (2001). Catalytic reactions in ionic liquids. Chemical Communications, 23,
2399–2407.
Singh, V., Chhotaray, P. K., & Gardas, R. L. (2015). Effect of protic ionic liquid on the
Appendix A. Supplementary data volumetric properties and taste behaviour of sucrose. Food chemistry, 169, 478–483.
Tammann, G., & Hesse, W. (1926). Die Abhängigkeit der Viscosität von der Temperatur
bie unterkühlten Flüssigkeiten. Zeitschrift für anorganische und allgemeine Chemie,
Supplementary data to this article can be found online at https:// 156(1), 245–257.
doi.org/10.1016/j.foodchem.2020.126566. van der Sman, R. G. M. (2017). Predicting the solubility of mixtures of sugars and their
replacers using the Flory-Huggins theory. Food & Function, 8, 360–371.
Vogel, H. (1921). Das Temperaturabhangigkeitsgesetz der Viskositat von Flussigkeiten.
References Physikalische Zeitschrift, 22, 645–646.
Zafarani-Moattar, M. T., Shekaari, H., & Mazaher Haji Agha, E. (2019). Effect of ionic
liquids 1-octyl-3-methyl imidazolium bromide or 1-octyl-3-methyl imidazolium
Ahosseini, A., & Scurto, A. M. (2008). Viscosity of imidazolium-based ionic liquids at chloride on thermophysical properties and taste behavior of sucrose in aqueous
elevated pressures: Cation and anion effects. International Journal of Thermophysics, media at different temperatures: Volumetric, compressibility and viscometric prop-
29(4), 1222–1243. erties. Food Chemistry, 295, 662–670.
Apelblat, A., & Manzurola, E. (1999). Solubilities ofo-acetylsalicylic, 4-aminosalicylic, Zarei, S., Abdolrahimi, S., & Pazuki, G. (2019). Thermophysical characterization of sor-
3,5-dinitrosalicylic, andp-toluic acid, and magnesium-DL-aspartate in water fromT= bitol and 1–ethyl–3–methylimidazolium acetate mixtures. Fluid Phase Equilibria, 497,
(278 to 348) K. The Journal of Chemical Thermodynamics, 31(1), 85–91. 140–150.
Baghbanbashi, M., & Pazuki, G. (2014). A new hydrogen bonding local composition based