Physicochemical properties of fatty acid based ionic liquids

Physicochemical properties of fatty acid based ionic liquids

J. Chem. Thermodynamics 100 (2016) 156–164 Contents lists available at ScienceDirect J. Chem. Thermodynamics journal homepage: www.elsevier.com/loca...

3MB Sizes 10 Downloads 99 Views

J. Chem. Thermodynamics 100 (2016) 156–164

Contents lists available at ScienceDirect

J. Chem. Thermodynamics journal homepage: www.elsevier.com/locate/jct

Physicochemical properties of fatty acid based ionic liquids Marisa A.A. Rocha a,b,⇑, Adriaan van den Bruinhorst a, Wolffram Schröer c, Bernd Rathke b, Maaike C. Kroon a,d,⇑ a

Separation Technology Group, Department of Chemical Engineering and Chemistry, Eindhoven University of Technology, Groene Loper 5, 5612 AE Eindhoven, The Netherlands Technische Thermodynamik, Universität Bremen, Badgasteiner Str. 1, D-28359 Bremen, Germany c Institut für Anorganische und Physikalische Chemie, Universität Bremen, Leobener Str. NW II, D-28359 Bremen, Germany d The Petroleum Institute, Dept. Chemical Engineering, P.O. Box 2533, Abu Dhabi, United Arab Emirates b

a r t i c l e

i n f o

Article history: Received 19 October 2015 Received in revised form 20 April 2016 Accepted 29 April 2016 Available online 30 April 2016 Keywords: Ionic liquids Fatty acid Density Viscosity Thermogravimetric analysis Thermal phase behavior

a b s t r a c t In this work a series of fatty acid based ionic liquids has been synthesized and characterized. Densities and viscosities at different temperatures have been measured in the temperature range from (293.15 to 363.15) K. The thermal operating window and thermal phase behavior have been evaluated. The effects of a branched anion and a mono-unsaturated anion on the physicochemical properties have been explored. It has been observed that the density (T = 298.15 K) decreases with the following sequence: methyltrioctylammonium 4-ethyloctanoate > methyltrioctylammonium oleate  tetrahexylammonium oleate > tetraoctylammonium oleate, with no detectable dependency of the thermal expansion coefficients on the total number of carbons in the ionic liquid. An almost linear correlation between the molar volumes and the total number of carbons of the alkanes together with the studied ionic liquids was found. The experimental viscosity data were correlated using the Vogel–Fulcher–Tammann (VFT) equation, where a maximum relative deviation of 1.4% was achieved. The ionic liquid with branched alkyl chains on the anion presents the highest viscosity, and methyltrioctylammonium oleate has the highest viscosity compared to the rest of the oleate based ionic liquids. The short and long-term stability were evaluated for all ionic liquids, their long-term decomposition temperatures were found to be significantly lower than their short-term decomposition temperatures. From the long-term thermal analysis was concluded that the highest temperature at which these ionic liquids can be kept is 363 K. In addition, the thermal behavior, glass transition temperature, crystallization behavior and melting temperatures of the studied ionic liquids are presented. Ó 2016 Elsevier Ltd.

1. Introduction Ionic liquids are salts composed of ions, liquid at room temperature, with remarkable properties, such as low vapour pressure, stable liquid phase over a wide temperature range, low flammability and thermal stability at high temperatures [1–6]. The possibility of tuning these properties over a wide range by adjusting the structure and chemical composition of the constituting ions is one of the main advantages of these systems, which have contributed to intense research in chemical [7–16], biological [17,18], and material science fields [4,5,19,20]. The tuning of the physicochemical properties of ionic liquids goes beyond the

⇑ Corresponding authors at: Separation Technology Group, Department of Chemical Engineering and Chemistry, Eindhoven University of Technology, P.O. Box 513, 5600 MB Eindhoven, The Netherlands. E-mail addresses: [email protected] (M.A.A. Rocha), M.C. [email protected] (M.C. Kroon). http://dx.doi.org/10.1016/j.jct.2016.04.021 0021-9614/Ó 2016 Elsevier Ltd.

different combinations of cations and anions, and nowadays, the research is also focusing on binary and ternary mixtures of ionic liquids [20–24]. The development of industrial processes and design of new ionic liquids needs accurate thermophysical data such as heat capacities, viscosities, densities and vapour pressures. Obviously, these properties cannot be measured for all possible ionic liquids. To overcome experimental limitations, the development of different simulation techniques (Quantum and Molecular Mechanics, Molecular Dynamics, Monte Carlo) [25–36] and group contribution methods [37–40] has increased. Since the unique properties of ionic liquids are governed by the structure and interaction between the ions, a better understanding of property–structure relationships is required. Additionally, the study of this relationship provides qualitative information concerning the interactions at a molecular level of the ionic liquids. Despite of the high number of ionic liquids, 1,3-dialkylimidazolium based ionic liquids are the most studied in the literature, and

157

M.A.A. Rocha et al. / J. Chem. Thermodynamics 100 (2016) 156–164

reagents were used without any further purification. Bromide content of the ionic liquids was determined by AgNO3 test. All the synthesized ionic liquids have a bromide content below 1.4 ppm (solubility of AgBr in water) [54,55]. The ionic liquids were dried and purified under vacuum (<1 Pa) for a minimum of 48 h, under constant stirring, in order to reduce the presence of water or other volatile contents. This process was performed systematically before and during the thermophysical properties measurements. The water content of the ionic liquids was determined using a Metrohm 795 Karl Fischer device (accuracy 50 ppm), being below 200 ppm. The structures of the synthesized ionic liquids were confirmed by NMR on a Bruker 400 MHz spectrometer. All ionic liquids were dissolved in chloroform-d, for 1H–NMR 16 scans and 5 s relaxation delay were applied and for 13C–NMR 256 scans and 1 s relaxation delay. A schematic representation of the synthesized ionic liquids is shown in Fig. 1.

several trends for their thermophysical properties [41–46] and solubility [47,48] have been observed. The natural availability and biodegradability of fatty acids, makes them interesting alternatives to be explored for the design of new ionic liquids [49,50]. Fatty acids have been investigated for the design of new low toxicity ionic liquids [51,52] and for selective metal extractions recently [12,53]. However, there is a lack of physicochemical properties concerning these kind of systems. Therefore, in this work, four fatty acid based ionic liquids were synthesized and characterized. The physicochemical properties, namely density and viscosity were measured in the temperature range (293.15–363.15) K, and the thermal stability and thermal phase behavior have been investigated. 2. Experimental details 2.1. Synthesis, purification and characterization

2.2. Densities and viscosities The ionic liquids used in this work were synthesized by a twostep procedure using the same experimental conditions recently described in the literature [12]. Information on the used reagents and synthesized ionic liquids is summarized in Table 1. The

Density (q) and dynamic viscosity (g) measurements were performed using an automated SVM 3000 Anton Paar rotational Stabinger viscometer – densimeter (Anton Paar, Graz, Austria).

Table 1 Sample description table.

a b c

Name

M/gmol1

CAS RN

Source

Mole fraction purity/%

Purification method

Water contentc/ppm

Tetraoctylammonium bromide Tetrahexylammonium bromide Methyltrioctylammonium bromide Oleic acid 4-Ethyloctanoic acid Sodium hydroxide Tetraoctylammonium oleate Tetrahexylammonium oleate Methyltrioctylammonium oleate Methyltrioctylammonium 4-ethyloctanoate

546.7962 434.5828 448.6095 282.4631 172.2656 39.9972 748.3472 636.1338 650.1605 539.9630

14866-33-2 4328-13-6 35675-80-0 112-80-1 16493-80-4 1310-73-2 1416991-41-7  71125-11-6 

Sigma–Aldrich Sigma–Aldrich Sigma–Aldrich Sigma–Aldrich Sigma–Aldrich Sigma–Aldrich Synthesized Synthesized Synthesized Synthesized

98a 99a 97a 99a 99a 97a 96b 97b 96b 92b

      Drying Drying Drying Drying

      110 170 130 170

under under under under

vacuum vacuum vacuum vacuum

The purities of all compounds were provided by manufacturer in the certificate of analysis. The purities of the synthesized ionic liquids were estimated using 1H-NMR, taking into account all deviations from a 1:1 ratio between the cation and the anion. The water content was determined using Karl Fischer titration. The relative atomic masses used were those recommended by the IUPAC Commission in 2011 [56].

O O

N+

N+

O-

O-

O

N+

O-

tetraoctylammonium oleate, [N8888][C18:1]

tetrahexylammonium oleate, [N6666][C18:1]

methyltrioctylammonium oleate, [N8881][C18:1]

Ctotal = 50

Ctotal = 42

Ctotal = 43

O O-

N+

methyltrioctylammonium 4-ethyloctanoate, [N8881][4C2C8:0]

Ctotal = 35 Fig. 1. Schematic representation, abbreviation and total number of carbons for the ionic liquids under study. Ctotal corresponds to the total number of carbons in the ionic liquid.

158

M.A.A. Rocha et al. / J. Chem. Thermodynamics 100 (2016) 156–164

Three certified density and viscosity reference (ISO 17025/ISO guide 34) standards were used for calibration. They were acquired from Anton Paar and have the product/batch codes APN7.5, APN26 and APN415. The measurements were carried out in the temperature range from (293.15 to 363.15) K, and at atmospheric pressure. For each ionic liquid, at least two independent measurements of the density and the viscosity were performed, using the same experimental conditions. The reproducibility of the dynamic viscosity and density measurements is, according to the manufacturer, 0.45% and ±0.5 kgm3, respectively, in the temperature range of (288.15–378.15) K and the uncertainty of temperature is within ±0.02 K. 2.3. Thermal stability The decomposition temperatures were measured with a thermal analyzer, Perkin Elmer TGA 4000, using a ceramic crucible under a continuous nitrogen flow (20 mLmin1). The short-term stability was determined by scanning the temperature range of (303–673) K at a heating rate of 10 Kmin1 and the long-term performance of the ionic liquids was evaluated by holding the ionic liquid samples during 12 h at a fixed temperature (323, 363 and 403 K). Sample masses were typically 15–20 mg and the weighing precision and sensitivity of the balance are ±0.01% and 1 lg, respectively.

Table 2 Experimental density results at p = (0.10 ± 0.01) MPa, q, for the ionic liquid series as a function of temperature.

q/(kgm3)

T/K

[N8888][C18:1]

[N6666][C18:1]

[N8881][C18:1]

[N8881][4C2C8:0]

293.15 298.15 303.15 308.15 313.15 318.15 323.15 328.15 333.15 338.15 343.15 348.15 353.15 358.15 363.15

878.4 875.3 872.3 869.2 866.1 863.1 860.1 857.1 854.1 851.1 848.2 845.2 842.2 839.2 836.1

889.9 886.9 883.9 880.9 877.8 874.8 871.8 868.8 865.8 862.9 859.9 857.0 854.0 851.0 847.9

890.5 887.5 884.5 881.5 878.5 875.5 872.5 869.5 866.6 863.6 860.7 857.8 854.9 852.0 849.1

893.1 890.1 887.1 884.0 881.0 877.9 874.9 871.8 868.8 865.8 862.8 859.8 856.8 853.9 850.9

ur(q)

±0.004

±0.003

±0.004

±0.008

Standard uncertainties, u, are: u(T) = 0.02 K. The relative standard uncertainty (ur) of the density was estimated considering the purity of each ionic liquid and using the method proposed in the literature [57].

2.4. Thermal phase behavior The thermal behavior was analyzed using a power compensation differential scanning calorimeter (Pyris Diamond DSC, PerkinElmer). The phase behavior was explored in the temperature range of (163–343) K using the continuous method with a heating rate of 5 K min1 and sealed aluminium crucibles. The temperature and heat flux scales of the calorimeter were calibrated by measuring the temperature and the enthalpies of fusion for reference materials indium [CAS Number 7440-74-6, Perkin-Elmer P/N 0319-0033] and lead [CAS Number 7439-92-1, Perkin-Elmer P/N 0319-0036]. During the thermal analysis of the ionic liquids, two isothermal runs, one at the start and one at the end of each measurement, were performed to remove possible memory effects. During the first run, an initial isotherm was performed at 163 K, to observe the initial crystalline phase before the first melting, followed by the standard 5 Kmin1 heating and a final isothermal scan at 343 K to ensure sample melting and stabilization. The crucibles were weighted on a Sartorius CP2P microbalance with a repeatability of ±0.003 mg and the typical sample load was about 10 mg. 3. Results and discussion The structural characterization of the fatty acid based ionic liquids has been performed using 1H–NMR and 13C–NMR, and the results are available as supporting information. Some deviations were found for the integrated areas of the 1H-NMR peaks, being related with residual impurities (mostly IL reactants) which were not removed during synthesis or purification. 3.1. Densities Densities and viscosities of the ionic liquids were determined at atmospheric pressure and in the temperature range of (293.15– 363.15) K. The experimental density results are given in Table 2. The graphical representations of the density against the temperature for the synthesized ionic liquids are presented in Fig. 2. To

Fig. 2. Density as a function of temperature for the studied ionic liquids. The thin lines results from the fitting of the experimental results using Eq. (1). This work: – [N6666][C18:1]; – [N8888][C18:1]; – [N8881][C18:1]; – [N8881][4C2C8:0]. Literature: – [N8881][NTf2] [58].

the best of our knowledge, no data was found in the literature for these systems. The density data were correlated with the temperature using a second order polynomial equation: 3

ln ðq=kg  m Þ ¼ a þ b  T þ c  T 2 ;

ð1Þ

where a, b and c are constants obtained from the fitting. Based on the density data, the molecular volumes, V, for the ionic liquids were calculated by the following Eq. (2):



M

q  NA

;

ð2Þ

where M indicates the molar mass, NA the Avogadro’s number (NA = 6.02214129  1023 mol1) [59], and q the density. The lattice potential energy, UPOT, of the studied ionic liquids was estimated based on the relationship developed by Mallouk et al. [60] and later expanded [61–63], using Eq. (3):

  a ffiffiffi ffi þ b ; U POT ¼ 2  I  p 3 V

ð3Þ

159

M.A.A. Rocha et al. / J. Chem. Thermodynamics 100 (2016) 156–164

which has been applied to a wide range of simple salts MX (1:1),

a and b are fitted coefficients of this type of salt (a = 117.3

kJmol1nm and b = 51.9 kJmol1), V is the molecular volume in nm3 (V = Vcation + Vanion) and I is the ionic strength (for simple salts MX (1:1), I = 1) [64]. The fitting parameters of Eq. (1), the molecular volume and the lattice potential energy at 298.15 K and 0.1 MPa are presented in Table 3 for the four fatty acid based ionic liquids. The graphical representation of the molecular volume and lattice potential energy at 298.15 K and 0.1 MPa as a function of the total number of carbons in the ionic liquid, Ctotal, is presented in Fig. 3 for each ionic liquid. As expected, the densities decrease with increasing temperature (Fig. 2). For a fixed temperature, the density decreases with the following sequence of ionic liquids: [N8881][4C2C8:0] > [N8881] [C18:1]  [N6666][C18:1] > [N8888][C18:1]. The volumetric behavior of the ionic liquids is expressed in terms of the molecular volume, which reflects the size of single ions. The molar mass rules the molecular volume tendency, where the ionic liquid [N8888][C18:1] displays the highest molecular volume (lowest density) among the studied ionic liquids. The molecular volume increases almost linearly as the total number of carbons (in both cation and anion) of the ionic liquid increases, with a contribution of (0.0281 ± 0.0006) nm3 per methylene group (–CH2–), which is comparable with the increment of (0.0269 ± 0.0001) nm3 observed for alkanes (CnH2n+2, n = 5–12, 16) [65]. This value is in good agreement with the literature work found for several families of ionic liquids, where a linear relationship is observed between the molecular volumes and the number of carbons in the alkyl chain of the cation or anion [66–68]. The estimated lattice potential energy decreases with an increase of the total number of carbons in the alkyl chain, reflecting the increase of the molecular volume of the ionic liquid (Eq. (3)). One of the limitations of this approach is the fact that only the Coulombic interactions are taken into account and the dispersion interactions such as van der Waals interactions, hydrogen bonding, p    p stacking interactions and the nanostructuration in the ionic liquids are ignored. This will lead to an underestimation of the lattice potential energy. Nevertheless, it is an approach that can be successfully used to give some insight in terms of Coulombic interactions in the bulk of the ionic liquid. The increase of the alkyl chains of the ionic liquid enhances the steric hindrance, which causes an increase in the distance between the ions, and consequently to a decrease in the cation–anion interactions [69]. Lattice potential energy for [C4mim] based ionic liquids with the anions [Cl]–, [BF4]–, [PF6]–, [TfO]–, and [NTf2]– were calculated based on the literature density data [70–74] and related to the cation–anion interaction energy reported in the literature [69]. This analysis is reported as supporting information. It was found that the lattice potential energies and the dissociation energies decrease with an increase of the molecular volume of the ionic liquid. Within this work, it was observed that the UPOT decreases with increasing alkyl chain length of the ionic liquid. This indicates a lower structural organization efficiency [63,66–68] and thus a decrease of the density. [N8881][NTf2] shows much higher densities [58] compared to the ionic liquids studied in this work. Moreover, [N8881][NTf2] presents a different behavior from the set of the ionic liquids reported here, concerning the molecular volume and lattice

Fig. 3. Molecular volumes and lattice potential energy, at T = 298.15 K, as a function of the total number of carbons of the systems. V: – this work; – [N8881][NTf2] [58]. UPOT: – this work; – [N8881][NTf2] [58]. The lines are intended as a guide to the eye.

potential energy. This behavior is related to the fact that the [NTf2]– anion is composed of more heterogeneous and heavy atoms. A good linear correlation was observed between the molecular volumes and the total number of carbons of the alkanes together with the ionic liquids studied so far (Fig. 4), which emphasizes the additive nature of the methylene group contribution of both cation and anion. The temperature dependence of the isobaric thermal expansion coefficients, ap, was calculated using Eq. (4):



1 @q ap ¼   q @T

 ¼ p

  @lnq ¼ ðb þ 2  c  TÞ @T p

ð4Þ

where q is the density in kg m3, T is the temperature in K and p is the standard pressure (0.1 MPa), while b and c are the fitting parameters of Eq. (1). The thermal expansion coefficients, at 298.15 K and 0.1 MPa, are listed in Table 4. Considering the associated uncertainty, the thermal expansion coefficients are independent of the total number of carbons in the ionic liquid. This behavior was already observed for several other ionic liquids in literature [73,74]. The ap values for the systems vary between (0.666 and 0.698)  103 K1, which are in the same range for the [N8881][NTf2] (0.678  103 K1) [58] and imidazolium based ionic liquids {(0.663–0.697)  103 K1} [74]. 3.2. Viscosities Viscosity relates the internal resistance of a fluid to a shear stress, which is related to intermolecular interactions. The high viscosities of ionic liquids are dependent on their configurational arrangement and charge distribution, which involves not only Coulombic interactions, but also van der Waals interactions, p    p interactions and hydrogen bonding, leading to higher viscosities than the ones found for typical simple molecular liquids. The experimental viscosity data are reported in Table 5 and are summarized in Fig. 5.

Table 3 Fitting parameters of the Eq. (1), molecular volume and lattice potential energy at 298.15 K and 0.1 MPa for the studied ionic liquids. Ionic liquid

a

(104  b)/K1

(107  c)/K2

V/nm3

UPOT/kJmol–1

[N8888][C18:1] [N6666][C18:1] [N8881][C18:1] [N8881][4C2C8:0]

6.975 ± 0.005 6.978 ± 0.004 6.995 ± 0.002 6.991 ± 0.003

6.49 ± 0.28 5.98 ± 0.23 7.00 ± 0.12 6.47 ± 0.19

0.82 ± 0.43 1.39 ± 0.34 0.30 ± 0.19 0.69 ± 0.29

1.420 ± 0.001 1.191 ± 0.001 1.216 ± 0.001 1.007 ± 0.001

312.5 325.1 323.6 337.9

The uncertainties of the fitting parameters correspond to the standard deviation. The uncertainty of the molecular volume was calculated using uncertainty propagation.

160

M.A.A. Rocha et al. / J. Chem. Thermodynamics 100 (2016) 156–164

Fig. 4. Molecular volumes at T = 298.15 K and p = 0.1 MPa plotted in terms of the total number of carbons, Ctotal. – studied ionic liquids; – alkanes (CnH2n+2, n = 5–12, 16) [65]. The red dashed line represents the linear fit of the molecular volumes of alkanes against Ctotal. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 4 Thermal expansion coefficients, at T = 298.15 K and 0.1 MPa for the studied ionic liquids. Ionic liquid

[103  ap (298.15 K)]/K1

[N8888][C18:1] [N6666][C18:1] [N8881][C18:1] [N8881][4C2C8:0]

0.698 ± 0.039 0.681 ± 0.032 0.682 ± 0.017 0.688 ± 0.027

Table 5 Experimental dynamic viscosity results at p = (0.10 ± 0.01) MPa, g, for the ionic liquid series as a function of temperature. T/K

g/(mPas) [N8888][C18:1]

[N6666][C18:1]

[N8881][C18:1]

[N8881][4C2C8:0]

293.15 298.15 303.15 308.15 313.15 318.15 323.15 328.15 333.15 338.15 343.15 348.15 353.15 358.15 363.15

1689.3 1153.4 804.8 575.4 420.8 314.4 239.4 185.4 146.2 117.0 94.9 77.9 64.1 53.5 44.3

2509.9 1623.7 1086.6 749.0 530.9 384.9 285.3 215.5 165.4 129.5 102.9 82.8 67.5 55.5 45.5

2934.0 1971.4 1354.6 957.7 692.7 511.3 384.4 293.8 227.9 179.2 142.7 114.3 92.3 74.9 61.8

5743.4 3733.4 2468.1 1672.3 1159.3 820.4 592.1 434.7 324.9 246.3 189.2 147.4 116.1 92.4 74.1

ur(g)

±0.03

±0.03

±0.03

±0.06

The viscosity of the liquids as a function of temperature and pressure can be correlated with different theoretical models and empirical or semi-empirical expressions [75]. The most commonly used equations to correlate the variation of viscosity with temperature are either Arrhenius (Eq. (5)) or the Vogel–Fulcher– Tammann (VFT) model (Eq. (6)):



Ea RT

g ðTÞ ¼ Ag  exp



Bg ðT  C g Þ

 ð6Þ

where g1 is the viscosity at infinite temperature and Ea is the activation energy. g(T) is the viscosity in mPas, T is the temperature in K, and Ag, Bg and Cg are adjustable parameters. The viscosity data were fitted using Eq. (5) and the derived g1 and Ea are presented as supporting information. A maximum relative deviation of 7.3% was achieved for the correlated values of the ionic liquids. Using the VFT equation leads to a significant improvement. The correlated viscosities are in good agreement with the experimental data and a maximum relative deviation of 1.4% was achieved for the correlated values. The derived coefficients of VFT equation obtained in this work are listed in Table 6. A decrease in viscosity with an increase of temperature is observed (Fig. 5). This decrease is more extreme for ionic liquids with higher viscosity. For the set of ionic liquids described within the report, [N8881][4C2C8:0] presents the highest viscosity and [N8888][C18:1] shows the lowest viscosity, at 298.15 K, as can be observed in Fig. 6. For a selected temperature, the viscosity decreases with the following sequence:

½N8881 ½4 C2 C8:0  > ½N8881 ½C18:1  > ½N6666 ½C18:1  > ½N8888 ½C18:1 

Standard uncertainties, u, are: u(T) = 0.02 K. The relative standard uncertainty (ur) for the viscosity was estimated considering the purity of each ionic liquid.

g ðTÞ ¼ g1  exp

Fig. 5. Viscosity as a function of temperature for the studied ionic liquids. The solid lines represent the correlative values calculated using the Vogel–Fulcher–Tammann fitting parameters (Eq. (6)). This work: – [N6666][C18:1]; – [N8888][C18:1]; – [N8881][C18:1]; – [N8881][4C2C8:0]. Literature: – [N8881][NTf2] [58].



ð5Þ

The ionic liquid with branched alkyl chains on the anion, [N8881] [4C2C8:0], presents the highest viscosity. This is in agreement with the work recently published by Andresova et al. [76], it was shown that the viscosity increases with the branching of the alkyl chain in the cation. Concerning the tetraalkylammonium oleate ionic liquids, [N6666][C18:1] shows a higher viscosity than [N8888][C18:1], contrary to the observed increase of viscosity with increasing alkyl chain [4,45]. Usually, the increase of the aliphatic length in ionic liquids leads to an increase of the viscosity due to the increase of the van der Waals interactions. This is observed if the van der Waals interactions are dominant over the electrostatic interactions, of which the latter are reduced with the increase of the size of the ions [4,45]. The opposite trend that was observed for [N6666] [C18:1] and [N8888][C18:1], could be related to the increase of the steric hindrance due to the increase of the alkyl chain length of the cation together with the large size of the anion [C18:1]–. This results in a decrease of the electrostatic interactions between ion pairs, leading to a lower viscosity. However, this can only be confirmed if other alkyl chain lengths are explored. [N8881][C18:1] has the highest viscosity compared to the rest of the oleate based ionic liquids.

161

M.A.A. Rocha et al. / J. Chem. Thermodynamics 100 (2016) 156–164 Table 6 Fitting coefficients of the VFT equation for the viscosity data of the studied ionic liquids and the derived energy barrier at T = 298.15 K. Ionic liquid

Ag/(mPas)

Bg/K

Cg/K

E (T = 298.15 K)/(kJmol1)

[N8888][C18:1] [N6666][C18:1] [N8881][C18:1] [N8881][4C2C8:0]

0.062 ± 0.006 0.050 ± 0.003 0.022 ± 0.003 0.0057 ± 0.0003

1298 ± 29 1295 ± 16 1716 ± 48 2105 ± 19

166.1 ± 1.7 173.6 ± 0.9 147.9 ± 2.5 140.9 ± 0.8

55.0 ± 3.4 61.7 ± 2.3 56.2 ± 4.0 62.9 ± 1.3

The uncertainties of the fitting parameters corresponds to the standard deviation obtained from the fitting routine. The uncertainty of the derived energy barrier was calculated using propagation of the uncertainties.

Fig. 6. Viscosity (mPas) at T = 298.15 K as a function of the total number of carbon atoms in the ionic liquid, Ctotal. This work: – [N6666][C18:1]; – [N8888][C18:1]; – [N8881][C18:1]; – [N8881][4C2C8:0]. Literature: – [N8881][NTf2] [58].

The energy barrier of the fluid to a shear stress, E, can be evaluated based on the viscosity dependence with the temperature using the following equation [75,77]:

0

1

Bg @ðln½gðTÞÞ B C ¼ R  @ 2 E¼R A Cg 2C g @ð1=TÞ  þ1 T2

ð7Þ

T

The plots of the pre-exponential coefficient, Ag, of the VFT equation (Eq. (6)) and the energy barrier (Eq. (7)) at 298.15 K against the Ctotal in the ionic liquid, are shown in Fig. 7. [N8881][4C2C8:0] presents the lowest pre-exponential coefficient of the VFT equation, Ag, as shown in Fig. 7(I). The branched anion causes a reduction of the molecular motion and consequently a decrease of the surface-to-volume ratio of the ion pairs (low Ag values). Changing from a branched anion to the oleate ([C18:1]–), an increase of Ag can be observed, where the ionic liquids [N6666] [C18:1] and [N8888][C18:1] present the highest values. The higher viscosity of [N6666][C18:1] compared to [N8888][C18:1] is driven by the lower Ag and higher energy barrier. [N8881][C18:1] presents a value of Ag between the ionic liquid with branched anion and tetraalkylammonium cations. The energy barrier at 298.15 K, E, decreases with the increase of total number of carbons on the ionic liquid. 3.3. Thermal stability The short-term thermal stability of all the studied ionic liquids was determined using thermogravimetric analysis. Also, the longterm stability was determined for [N6666][C18:1], [N8881][4C2C8:0]. The decomposition temperature of the ionic liquids [N8888][C18:1] and [N8881][C18:1] was reported in a prior study, using the same experimental conditions [12]. The thermogravimetric analyses (TGA) that were performed for these two ionic liquids are in good

Fig. 7. (I) Pre-exponential coefficient of the Vogel–Fulcher–Tammann equation (Ag/mPas), and (II) energy barrier (E/kJmol1) at T = 298.15 K as a function of the total number of carbon atoms in the ionic liquid, Ctotal. This work: – [N6666][C18:1]; – [N8888][C18:1]; – [N8881][C18:1]; – [N8881][4C2C8:0]. Literature: – [N8881] [NTf2] [58].

agreement with the reported values [12]. The TGA results for the ammonium based ionic liquids are listed in Table 7. The temperature of decomposition, Td, was calculated as the temperature onset. The performance of ionic liquids at a certain temperature for a longer period of time, is important information from an industrial application point of view. In that case, the short-term stability measurements are not applicable. Thus, the long-term performance of all the ionic liquids was evaluated by holding the ionic liquid samples at isothermal conditions (323, 363 and 403 K) for a period of 12 h. The experimental results are presented in Table 7.

162

M.A.A. Rocha et al. / J. Chem. Thermodynamics 100 (2016) 156–164

Table 7 Thermogravimetric analysis results for the set of ionic liquids. Ionic liquids

[N8888][C18:1] [N6666][C18:1] [N8881][C18:1] [N8881][4C2C8:0]

Short-term stability

Long-term stability (isothermal conditions, 12 h)

Td/K

Mass loss (323 K)/%

Mass loss (363 K)/%

Mass loss (403 K)/%

448 [12] 458 442 [12] 450

1.32 0.0 0.54 3.78

4.65 14.85 1.91 6.81

49.57 90.28 61.47 47.49

Standard uncertainties, u, are: u(T) = 1 K and u(mass loss) = 0.02%. Standard pressure, p = (0.10 ± 0.01) MPa.

Fig. 8. Short-term stability TGA curves of [N6666][C18:1] ( ( ) (heating rate of 10 Kmin1).

) and [N8881][4C2C8:0]

The short thermal stability of the studied ionic liquids was found to follow this sequence: [N6666][C18:1]  [N8881][4C2C8:0] > [N8888][C18:1] > [N8881][C18:1]. Fig. 8 shows the short-term stability TGA curves of [N6666][C18:1] and [N8881][4C2C8:0]. It is well known that Td can be an overestimation of the real thermal stability, when it is calculated as the onset temperature obtained by fast heating [78–80]. The graphical representations of the isothermal behaviour of [N8888][C18:1] and [N6666][C18:1] are presented in Fig. 9. The data obtained from the isothermal TGA analysis of the other ionic liquids, [N8881][C18:1] and [N8881] [4C2C8:0], are presented in the supporting information. From the isothermal analysis (Fig. 9), a fast degradation (higher than 45%) can be observed at 403.15 K for all ionic liquids. All the studied ionic liquids present a decomposition temperature significantly lower (more than 50 K) than the presented Td obtained by the short-term stability study (dynamic experiments at scanning rate 10 Kmin1). Instead, the highest stability accessible temperature for these ionic liquids is only 363 K. 3.4. Thermal behavior measurements The thermal behavior of the ionic liquids was determined by differential scanning calorimetry. The melting temperatures (Tm) were taken as the onset temperature of an endothermic peak on heating. Cold crystallization temperature (Tcc) was determined as the onset temperature of an exothermic peak on heating from a subcooled liquid to a crystalline solid. The glass transition temperature (Tg) was obtained by taking the midpoint of the heat capacity change on heating from a glass to a liquid. Crystallization temperature (Tc) was taken as the onset temperature of an exothermic peak on cooling. The results of Tg, Tc, Tcc and Tm are listed in Table 8

Fig. 9. Isothermal TGA analysis of [N8888][C18:1] (I) and [N6666][C18:1] (II).

for each studied ionic liquid. The thermograms of the studied ionic liquids are attached as supporting information. Due to the complexity of the thermal behavior obtained for [N8888][C18:1] and [N8881][C18:1], the peak temperatures are presented as well, in order to identify each transition. Nevertheless, the onset temperature is more suitable to characterize the transition due to the relative independence of the experimental procedure [78]. The studied ionic liquids show a glass transition between 189 and 217 K. This is in agreement with the behavior observed for other ionic liquids, that typically show glass transitions around 173–205 K [81]. Glass transitions on heating, Tg, were found at low

163

M.A.A. Rocha et al. / J. Chem. Thermodynamics 100 (2016) 156–164 Table 8 Melting (Tm), glass transition (Tg), crystallization (Tc) and cold crystallization (Tcc) for the studied ionic liquids. Ionic liquid

Cooling Tc

Heating Tg

Tcc

Tm

226.3 ± 1.6a 223.3 ± 1.0b

248.4 ± 0.1a 243.6 ± 0.6b

K [N8888][C18:1]



193.3 ± 1.3

[N6666][C18:1]



189.9 ± 0.3

[N8881][C18:1]

224.5 ± 0.2a

216.8 ± 1.0

229.8 ± 1.5b [N8881][4C2C8:0]

231.7 ± 0.1a 232.2 ± 0.5b

196.5 ± 0.7





252.6 ± 0.1a 264.0 ± 0.1a 249.3 ± 0.4b 261.6 ± 0.6b

272.2 ± 0.3a





267.2 ± 0.3b

The uncertainties are the twice of the standard deviation of the mean. a peak temperature. b onset temperature. Melting temperature is taken as the onset temperature. Standard pressure, p = (0.10 ± 0.01) MPa.

temperatures with inhibition of crystallization for [N6666][C18:1] and for the branched anion based ionic liquid, [N8881][4C2C8:0]. A crystallization on cooling was found for [N8881][4C2C8:0] at 232 K. [N8888][C18:1] could only be crystallized during the heating step, presenting a cold crystallization at 223 K. The formation of the crystal is obtained when the subcooled liquid is heated from low temperatures, around 30 K above the glass transition temperature. The melting peak appears right after the cold crystallization transition with a Tm of 244 K. The symmetry of ions affects the melting temperature of the ionic liquid. A lower symmetry leads to weaker interactions and to a decrease on efficient packing in the crystal cell, resulting in lower melting temperatures [82]. This behavior was observed for the tetraalkylammonium oleate ionic liquids, in which [N6666][C18:1], a less symmetric ionic liquid compared to [N8888][C18:1], only features a glass transition. As the alkyl chain length increases the symmetry increases, but the dispersive interactions also become stronger, contributing to an increase of the packing efficiency. The [N8881][C18:1] thermogram displays a cold crystallization on two steps (249 and 262 K) followed by the melting transition at 267 K. This ionic liquid presents a 24 K higher melting temperature than the one measured for [N8888][C18:1]. A crystallization on cooling, Tc, was found for [N8881][C18:1] and [N8881][4C2C8:0]. [N8881][C18:1] was studied further in the same temperature range using a lower heating scanning rate of 2 Kmin1 (thermogram available as SI). Using these experimental conditions an inhibition of the glass transition was found. A partial crystallization on the cooling step was detected at (244.9 ± 0.1) K. During the heating step, a cold crystallization at (259.7 ± 0.3) K followed by the melting transition at (274.98 ± 0.02) K was observed. 4. Final remarks The synthesis and the interpretation of the physicochemical properties, such as densities and viscosities at different temperatures, thermal stability and thermal behavior, of four fatty acids based ionic liquids are reported in this work. The effect of the cation and anion on the studied properties is analyzed and discussed. For a fixed temperature, the density decreases with decreasing molecular weight of the ionic liquids. The molecular volume increases linearly with an increase in the total number of carbons on the ionic liquid, with a contribution per –CH2– comparable to the one observed for alkanes. A good linear correlation between the molar volumes and the total number of carbons of the alkanes together with the studied ionic liquids studied so far was observed, emphasizing the additive nature of the –CH2– contribution to the

molecular volume. The thermal expansion coefficients are independent on the total number of carbons on the ionic liquid and are within the range for the [N8881][NTf2] and imidazolium based ionic liquids. The ionic liquid with branched alkyl chains on the anion, [N8881] [4C2C8:0], present the highest viscosity. The higher viscosity detected for [N6666][C18:1] compared to [N8888][C18:1] is in agreement with the lower pre-exponential coefficient and higher energy barrier found. [N8881][C18:1] presents the highest viscosity within the oleate based ionic liquids governed by a higher energy barrier. All the ionic liquids presented a significantly lower decomposition temperature from the one measured using dynamic experiments at a scanning rate of 10 Kmin1. The highest temperature at which the studied ionic liquids can be kept is 363 K. The thermal phase behavior of the synthesized ionic liquids was determined using differential scanning calorimetry. All the ionic liquids presented a melting temperature below 280 K, being liquid at room temperature. A glass transition on heating with inhibition of the crystallization was observed for [N6666][C18:1] and [N8881] [4C2C8:0]. The thermal phase behavior displayed a cold crystallization and melting transition besides the glass transition, while increasing the alkyl chain length from [N6666][C18:1] to [N8888] [C18:1]. [N8881][C18:1] ionic liquids presented complex thermal behavior. The thermal phase behaviors of studied ionic liquids strongly depend on the nature of ions, symmetry and branching of the anion. Acknowledgments Financial support from the Dutch Technology Foundation STW and the company Paques (project nr. STW-Paques 12999) is gratefully acknowledged. M.A.A. Rocha acknowledges the financial support for the short term scientific mission within the COST Action CM1206. We are thankful to Lawien F. Zubeir for the valuable discussions. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jct.2016.04.021. References [1] M.A.A. Rocha, C.F.R.A.C. Lima, L.R. Gomes, B. Schröder, J.A.P. Coutinho, I.M. Marrucho, J.M.S.S. Esperança, L.P.N. Rebelo, K. Shimizu, J.N.C. Lopes, L.M.N.B.F. Santos, J. Phys. Chem. B 115 (2011) 10919–10926. [2] K.R.J. Lovelock, A. Deyko, P. Licence, R.G. Jones, Phys. Chem. Chem. Phys. 12 (2010) 8893–8901.

164

M.A.A. Rocha et al. / J. Chem. Thermodynamics 100 (2016) 156–164

[3] J.M.S.S. Esperança, J.N. Canongia Lopes, M. Tariq, L.M.N.B.F. Santos, J.W. Magee, L.P.N. Rebelo, J. Chem. Eng. Data 55 (2010) 3–12. [4] P. Wasserscheid, T. Welton (Eds.), Ionic Liquids in Synthesis, Second ed., WileyVCH Verlag GmbH & Co. KGaA, Weinheim, Germany, 2008. [5] B. Kirchner (Ed.), Ionic Liquids, Springer Berlin Heidelberg, Berlin, Heidelberg, 2010. [6] M.J. Earle, J.M.S.S. Esperança, M.A. Gilea, J.N.C. Lopes, L.P.N. Rebelo, J.W. Magee, K.R. Seddon, J.A. Widegren, Nature 439 (2006) 831–834. [7] H. Niedermeyer, J.P. Hallett, I.J. Villar-Garcia, P.A. Hunt, T. Welton, Chem. Soc. Rev. 41 (2012) 7780–7802. [8] C. Jork, C. Kristen, D. Pieraccini, A. Stark, C. Chiappe, Y.A. Beste, W. Arlt, J. Chem. Thermodyn. 37 (2005) 537–558. [9] B. Clare, A. Sirwardana, D.R. Macfarlane, Top. Curr. Chem. 290 (2010) 1–40. [10] A. Aggarwal, N.L. Lancaster, A.R. Sethi, T. Welton, Green Chem. 4 (2002) 517– 520. [11] B. Tang, W. Bi, M. Tian, K.H. Row, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 904 (2012) 1–21. [12] D. Parmentier, S.J. Metz, M.C. Kroon, Green Chem. 15 (2013) 205–209. [13] L.F. Zubeir, G.E. Romanos, W.M.A. Weggemans, B. Iliev, T.J.S. Schubert, M.C. Kroon, J. Chem. Eng. Data 60 (2015) 1544–1562. [14] V. Govinda, T. Vasantha, I. Khan, P. Venkatesu, Ind. Eng. Chem. Res. 54 (2015) 9013–9026. [15] V. Govinda, P. Attri, P. Venkatesu, P. Venkateswarlu, J. Phys. Chem. B 117 (2013) 12535–12548. [16] R. Zarrougui, M. Dhahbi, D. Lemordant, J. Solution Chem. 44 (2015) 686–702. [17] N. Jain, A. Kumar, S. Chauhan, S.M.S. Chauhan, Tetrahedron 61 (2005) 1015– 1060. [18] J.K. Lee, M.-J. Kim, J. Org. Chem. 67 (2002) 6845–6847. [19] C. Wang, G. Cui, X. Luo, Y. Xu, H. Li, S. Dai, J. Am. Chem. Soc. 133 (2011) (1919) 11916–11919. [20] M. Bidikoudi, L.F. Zubeir, P. Falaras, J. Mater. Chem. A 2 (2014) 15326–15336. [21] G. Chatel, J.F.B. Pereira, V. Debbeti, H. Wang, R.D. Rogers, Green Chem. 16 (2014) 2051–2083. [22] M. Taige, D. Hilbert, T.J.S. Schubert, Zeitschrift Für Phys. Chem. 226 (2012) 129–139. [23] Q.Q. Baltazar, S.K. Leininger, J.L. Anderson, J. Chromatogr. A 1182 (2008) 119– 127. [24] S. Katsuta, Y. Yoshimoto, M. Okai, Y. Takeda, K. Bessho, Ind. Eng. Chem. Res. 50 (2011) 12735–12740. [25] R. Singh, E. Marin-Rimoldi, E.J. Maginn, Ind. Eng. Chem. Res. 54 (2015) 4385– 4395. [26] C.G. Aimoli, E.J. Maginn, C.R.A. Abreu, J. Chem. Eng. Data 59 (2014) 3041–3054. [27] C.G. Aimoli, E.J. Maginn, C.R.A. Abreu, J. Chem. Phys. 141 (2014) 134101. [28] K.S. Rane, J.R. Errington, J. Phys. Chem. B 118 (2014) 8734–8743. [29] T. Koller, J. Ramos, N.M. Garrido, A.P. Fröba, I.G. Economou, Mol. Phys. 110 (2012) 1115–1126. [30] A.J.L. Costa, M.R.C. Soromenho, K. Shimizu, I.M. Marrucho, J.M.S.S. Esperança, J. N. Canongia Lopes, L.P.N. Rebelo, J. Phys. Chem. B 116 (2012) 9186–9195. [31] B. Schröder, J.A.P. Coutinho, Fluid Phase Equilib. 370 (2014) 24–33. [32] Q. Chen, S.P. Balaji, M. Ramdin, J.J. Gutiérrez-Sevillano, A. Bardow, E. Goetheer, T.J.H. Vlugt, Ind. Eng. Chem. Res. 53 (2014) 18081–18090. [33] M. Ramdin, S.P. Balaji, J.M. Vicent-Luna, J.J. Gutiérrez-Sevillano, S. Calero, T.W. de Loos, T.J.H. Vlugt, J. Phys. Chem. C 118 (2014) 23599–23604. [34] D. Kerlé, R. Ludwig, D. Paschek, Zeitschrift Für Phys. Chem. 227 (2013) 167– 176. [35] D. Kerlé, R. Ludwig, A. Geiger, D. Paschek, J. Phys. Chem. B 113 (2009) 12727– 12735. [36] Y. Zhang, E.J. Maginn, Phys. Chem. Chem. Phys. 16 (2014) 13489–13499. [37] J. Albert, K. Müller, Ind. Eng. Chem. Res. 53 (2014) 17522–17526. [38] R.L. Gardas, J.A.P. Coutinho, AIChE J. 55 (2009) 1274–1290. [39] R.L. Gardas, J.A.P. Coutinho, Fluid Phase Equilib. 266 (2008) 195–201. [40] R.L. Gardas, J.A.P. Coutinho, Ind. Eng. Chem. Res. 47 (2008) 5751–5757. [41] M.A.A. Rocha, J.A.P. Coutinho, L.M.N.B.F. Santos, J. Chem. Phys. 139 (2013) 104502. [42] M.A.A. Rocha, J.A.P. Coutinho, L.M.N.B.F. Santos, J. Phys. Chem. B 116 (2012) 10922–10927. [43] M.A.A. Rocha, J.A.P. Coutinho, L.M.N.B.F. Santos, J. Chem. Phys. (2014) 134502. [44] S. Aparicio, M. Atilhan, F. Karadas, Ind. Eng. Chem. Res. 49 (2010) 9580–9595. [45] D. Rooney, J. Jacquemin, R. Gardas, Top. Curr. Chem. 290 (2010) 185–212. [46] Y. Cao, T. Mu, Ind. Eng. Chem. Res. 53 (2014) 8651–8664. [47] V.R. Vale, S. Will, W. Schröer, B. Rathke, ChemPhysChem 13 (2012) 1860–1867.

[48] X. Shao, W. Schröer, B. Rathke, J. Chem. Eng. Data 59 (2014) 225–233. [49] Fatty Acid Salts (Soap), Environmental Risk Assessment, 2004. [50] M. Sokolsky-Papkov, A. Shikanov, A. Ezra, B. Vaisman, A.J. Domb, Fatty acidbased biodegradable polymers: synthesis and applications, in: M.C. Celina, J.S. Wiggins, N.C. Billingham (Eds.), Polymer Degradation and Performance, ACS Symposium Series, American Chemical Society, Washington, DC, 2009, pp. 60– 69. [51] J. Pernak, B. Łe˛gosz, F. Walkiewicz, T. Klejdysz, A. Borkowski, Ł. Chrzanowski, RSC Adv. 5 (2015) 65471–65480. [52] Q. Yang, D. Xu, J. Zhang, Y. Zhu, Z. Zhang, C. Qian, Q. Ren, H. Xing, A.C.S. Sustain, Chem. Eng. 3 (2015) 309–316. [53] D. Parmentier, T. Vander Hoogerstraete, S.J. Metz, K. Binnemans, M.C. Kroon, Ind. Eng. Chem. Res. 54 (2015) 5149–5158. [54] L. Cammarata, S.G. Kazarian, P.A. Salter, T. Welton, Phys. Chem. Chem. Phys. 3 (2001) 5192–5200. [55] D.R. Lide (Ed.), CRC Handbook of Chemistry and Physics, 84th ed., CRC Press, 2003. [56] M.E. Wieser, N. Holden, T.B. Coplen, J.K. Böhlke, M. Berglund, W.A. Brand, P. De Bièvre, M. Gröning, R.D. Loss, J. Meija, T. Hirata, T. Prohaska, R. Schoenberg, G. O’Conner, T. Walczyk, S. Yoneda, X.-K. Zhu, Pure Appl. Chem. 85 (2013) 1047– 1078. [57] R.D. Chirico, M. Frenkel, J.W. Magee, V. Diky, C.D. Muzny, A.F. Kazakov, K. Kroenlein, I.L. Abdulagatov, G.R. Hardin, W.E. Acree Jr., J.F. Brenneke, P.L. Brown, P.T. Cummings, T.W. de Loos, D.G. Friend, A.R.H. Goodwin, L.D. Hansen, W.M. Haynes, N. Koga, A. Mandelis, K.N. Marsh, P.M. Mathias, C. McCabe, J.P. O’Connell, A. Pádua, V. Rives, C. Schick, J.P. Martin Trusler, S. Vyazovkin, R.D. Weir, J. Wu, J. Chem. Eng. Data 58 (2013) 2699–2716. [58] P.J. Carvalho, S.P.M. Ventura, M.L.S. Batista, B. Schröder, F. Gonçalves, J. Esperança, F. Mutelet, J.A.P. Coutinho, J. Chem. Phys. 140 (2014) 064505. [59] P.J. Mohr, B.N. Taylor, D.B. Newell, Rev. Mod. Phys. 84 (2012) 1527–1605. [60] T.E. Mallouk, G.L. Rosenthal, G. Mueller, R. Brusasco, N. Bartlett, Inorg. Chem. 23 (1984) 3167–3173. [61] H.D.B. Jenkins, H.K. Roobottom, J. Passmore, L. Glasser, Inorg. Chem. 38 (1999) 3609–3620. [62] H.D.B. Jenkins, D. Tudela, L. Glasser, Inorg. Chem. 41 (2002) 2364–2367. [63] L. Glasser, Thermochim. Acta 421 (2004) 87–93. [64] L. Glasser, J. Solid State Chem. 206 (2013) 139–144. [65] A. Aucejo, M.C. Burguet, R. Munoz, J.L. Marques, J. Chem. Eng. Data 40 (1995) 141–147. [66] S. Panda, R.L. Gardas, Fluid Phase Equilib. 386 (2015) 65–74. [67] A. Xu, J. Wang, Y. Zhang, Q. Chen, Ind. Eng. Chem. Res. 51 (2012) 3458–3465. [68] D.-W. Fang, W. Guan, J. Tong, Z.-W. Wang, J.-Z. Yang, J. Phys. Chem. B 112 (2008) 7499–7505. [69] A.M. Fernandes, M.A.A. Rocha, M.G. Freire, I.M. Marrucho, J.A.P. Coutinho, L.M. N.B.F. Santos, J. Phys. Chem. B 115 (2011) 4033–4041. [70] H. Machida, R. Taguchi, Y. Sato, R.L. Smith Jr., J. Chem. Eng. Data 56 (2011) 923– 928. [71] R.L. Gardas, M.G. Freire, P.J. Carvalho, I.M. Marrucho, I.M.A. Fonseca, A.G.M. Ferreira, J.A.P. Coutinho, J. Chem. Eng. Data 52 (2007) 80–88. [72] D. Matkowska, T. Hofman, J. Mol. Liq. 165 (2012) 161–167. [73] M.A.A. Rocha, F.M.S. Ribeiro, A.I.M.C.L. Ferreira, J.A.P. Coutinho, L.M.N.B.F. Santos, J. Mol. Liq. 188 (2013) 196–202. [74] M.A.A. Rocha, C.M.S.S. Neves, M.G. Freire, O. Russina, A. Triolo, J.A.P. Coutinho, L.M.N.B.F. Santos, J. Phys. Chem. B 117 (2013) 10889–10897. [75] K.Y. Viswanath, D.S. Ghosh, T.K. Prasad, D.H.L. Dutt, N.V.K. Rani, Viscosity of Liquids: Theory, Estimation, Experiment, and Data, Springer, 2007. [76] A. Andresova, J. Storch, M. Traïkia, Z. Wagner, M. Bendova, P. Husson, Fluid Phase Equilib. 371 (2014) 41–49. [77] P.G. Debenedetti, F.H. Stillinger, Nature 410 (2001) 259–267. [78] G.W.H. Höhne, W.F. Hemminger, H.-J. Flammersheim, Differential Scanning Calorimetry, Springer-Verlag, Berlin Heidelberg, New York, 1996. [79] C. Maton, N. De Vos, C.V. Stevens, Chem. Soc. Rev. 42 (2013) 5963–5977. [80] A. Seeberger, A.-K. Andresen, A. Jess, Phys. Chem. Chem. Phys. 11 (2009) 9375– 9381. [81] S.A. Mirkhani, F. Gharagheizi, P. Ilani-Kashkouli, N. Farahani, Thermochim. Acta 543 (2012) 88–95. [82] S.G. Raju, S. Balasubramanian, J. Phys. Chem. B 114 (2010) 6455–6463.

JCT 15-728