Thermodynamics of solvation in propylene glycol and methyl cellosolve

J. Chem. Thermodynamics 78 (2014) 32–36

Contents lists available at ScienceDirect

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

Thermodynamics of solvation in propylene glycol and methyl cellosolve I.A. Sedov ⇑, M.A. Stolov, B.N. Solomonov Chemical Institute, Kazan Federal University, Kremlevskaya 18, Kazan 420008, Russia

a r t i c l e

i n f o

Article history: Received 7 February 2014 Received in revised form 2 June 2014 Accepted 3 June 2014 Available online 16 June 2014 Keywords: Propylene glycol Methyl cellosolve Ethylene glycol Gibbs free energy Enthalpy Solvation Solvophobic effect

a b s t r a c t Limiting activity coefficients of low-polar substances: aliphatic and aromatic hydrocarbons, including alkanes, cycloalkanes, alkylbenzenes, and halobenzenes in two solvents, propylene glycol and methyl cellosolve, were measured at temperature T = 298.15 K using gas chromatographic headspace analysis technique. The Gibbs free energies of solvation were calculated from these data and analyzed together with the enthalpies of solvation for the same systems. It was shown that the Gibbs free energies of solvation in propylene glycol are significantly lower than in its homologue ethylene glycol, and in methyl cellosolve they are lower than in propylene glycol. This difference is mainly due to the solvophobic effect, which strength is decreasing in the same order: ethylene glycol > propylene glycol > methyl cellosolve. The contribution of the solvophobic effect into the Gibbs free energies of solvation can be determined using a Gibbs free energy versus enthalpy of solvation plot. This contribution is shown to grow up linearly with the molecular volume of a solute in propylene glycol and methyl cellosolve, as well as in ethylene glycol and in monohydric alcohols. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Methyl cellosolve (2-methoxyethanol, CH3OCH2CH2OH, further MC) and propylene glycol (1,2-propanediol, CH3CH(OH)CH2OH, further PG) are important industrial solvents. They are also structural isomers differing from a molecule of another industrial solvent, ethylene glycol (1,2-ethanediol, HOCH2CH2OH, further EG), by one CH2 group. The process of solvation of a molecule by transferring it from the gas phase into solvent can be described in terms of thermodynamic functions. The standard molar Gibbs free energy of solvation Dsolv G is related to the Ostwald solubility coefficient L of a gaseous solute through equation Dsolv G ¼ RT lnðRT=LP  V M Þ, where V M is the molar volume of solvent, P  is the standard pressure, and the molar fraction based standard state is used for solution. For solid and liquid solutes, it is the difference of the standard molar Gibbs free energies of dissolution at infinite dilution and vaporization of a solute: Dsolv G ¼ Dsoln G  Dv ap G. In turn, Dsoln G can be related to the limiting activity coefficient c1 of a given solute in a given solvent: Dsoln G ¼ RT ln c1 . Knowledge of the values of solvation properties is useful for optimization of industrial processes and for the development of models of solutions. However, the Gibbs free energies of solvation in PG and MC have been unknown for almost any compound. ⇑ Corresponding author. Tel.: +7 9600503916; fax: +7 8432315346. E-mail address: [email protected] (I.A. Sedov). http://dx.doi.org/10.1016/j.jct.2014.06.006 0021-9614/Ó 2014 Elsevier Ltd. All rights reserved.

In the bulk liquid phase, both MC and PG molecules are able to form hydrogen bonds with each other. The molecules of PG have two hydroxyl groups and can form a branched three-dimensional network of hydrogen bonds similar to that of ethylene glycol. The molecule of MC has only one hydroxyl group and is unable to form similar structures. Though not studied yet, the mechanism of self-association of liquid MC should be similar to that of monohydric alcohols forming chain linear and cyclic hydrogen-bonded aggregates. Such difference in the structure of the liquid phase should lead to a significant difference in solvation properties of these two isomeric solvents. An interesting feature of self-associating solvents is the solvophobic effect. Compounds which are unable to be a donor neither an acceptor of hydrogen bonds have an increased Gibbs free energy of solvation and, therefore, a decreased solubility in solvents that form intermolecular hydrogen bonds. Such solutions are also characterized with low entropies of solvation and increased heat capacity in comparison with solutions of the same compounds in aprotic organic solvents. Such behavior is known as the solvophobic effect [1,2] and is similar to the hydrophobic effect in water. The solvophobic effect can stabilize micelles, vesicles or other noncovalently bonded structures in non-aqueous media. Formation of micelles in PG has been studied, in particular, on the sample of nonylphenoxypolyethoxyethanol [3]. However, the solvophobic effect induced by PG is much weaker than the hydrophobic effect in water, and PG is able to denaturate or weaken the hydrophobically bonded structure of proteins [4] like other organic solvents.

33

I.A. Sedov et al. / J. Chem. Thermodynamics 78 (2014) 32–36

In our recent papers [1,5–7] we developed a methodology to describe the solvophobic effect in various solvents on the basis of the values of Gibbs free energy and enthalpy of solvation for low-polar solutes. Now we report new experimental values of the Gibbs free energy of solution and solvation in PG and MC and apply the above mentioned method to compare the strength of the solvophobic effect in these two and other solvents. 2. Experimental 2.1. Available experimental data on the thermodynamic functions of diluted solutions In our previous paper [8], we have studied the enthalpies of solution in PG and MC for a set of low-polar compounds. No other values of the enthalpies of solution of hydrocarbons and their halogenated derivatives in these solvents were found in literature. The Gibbs free energy of solvation in PG can be calculated for several low polar solutes from the experimental data reported in literature, though this solvent is much less studied than its homologue, ethylene glycol. Even less data are available for MC. Leitman and Gaile published [9] the values of limiting activity coefficients of several C6 to C9 hydrocarbons in PG at T = (316 and 324) K. In the work [10], the temperature dependences of the solubility of n-heptane in PG and vice versa were studied. (Despite this is not a strict formula, the (liquid + liquid) equilibrium data for poorly soluble compounds expressed in molar fractions can be used to estimate the limiting activity coefficient through equation c1 ¼ x22 =x21 , where x22 and x21 are the molar fractions of compound 2 in two equilibrium phases, and then we can calculate the Gibbs free energy of solvation). Later, the solubilities of n-hexane, n-heptane, n-octane, benzene, toluene, ethylbenzene, o-, m-, and p-xylene in PG were reported [11] at various temperatures. The solubility of phenothiazine was measured in PG [12] and the solubility of pyrene in MC [13]. Limiting activity coefficients of benzene, cyclohexene and cyclohexane in PG were also measured [14] at temperatures T > 343 K. Rohrshneider reported [15] (gas + liquid) partition coefficients at T = 298 K for a large set of (solute + solvent) systems including octane and toluene dissolved in MC. Later, Park et al. [16] investigated the same set of solutes and solvents using an improved methodology. There are several works on the solubilities of saturated hydrocarbons [17,18] in MC at various temperatures and pressures, but the concentrations at saturation are rather high and these data cannot be used to calculate the Gibbs free energy of solvation related to infinitely diluted solutions. For hydrocarbon-MC systems, VLE data have also been obtained [19,20], but not at T = 298 K. In the present work, novel data for solutions of hydrocarbons and their halogenated derivatives in PG and MC at infinite dilution and T = 298 K are obtained using GC headspace techniques. These results are used for both qualitative and quantitative interpretation of solvophobic effects in PG and MC. 2.2. Materials and methods Propylene glycol and methyl cellosolve with purity > 0.99 were purchased from Acros Organics. All the solutes were at least 0.99 pure grade from Sigma–Aldrich, Acros and Fluka. They have been used without further purification. The absence of significant amounts of impurities has been confirmed by gas chromatography (see table 1).

TABLE 1 Source and purity of the chemicals. Chemical name

Mass fraction purity

Source

Propylene glycol Methyl cellosolve n-Hexane n-Heptane n-Octane n-Nonane n-Decane n-Undecane Methylcyclopentane Cyclohexane Methylcyclohexane Cyclooctane Cyclohexene 1,7-Octadiene 4-Vinyl-1-cyclohexene Benzene Toluene Ethylbenzene o-Xylene m-Xylene p-Xylene p-Cymene Fluorobenzene Chlorobenzene Bromobenzene Naphthalene

0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.995 0.995 0.99 0.99 0.99 0.99 0.995 0.99 0.998 0.99 0.99 0.99 0.99 0.99 0.99 0.996 0.99 0.99

Acros Organics Sigma–Aldrich Sigma–Aldrich Acros Organics Sigma–Aldrich Acros Organics Acros Organics Acros Organics Sigma–Aldrich Acros Organics Acros Organics Sigma–Aldrich Acros Organics Acros Organics Fluka Sigma–Aldrich Sigma–Aldrich Fluka Sigma–Aldrich Sigma–Aldrich Sigma–Aldrich Acros Organics Acros Organics Acros Organics Acros Organics Acros Organics

580 chromatograph with a headspace autosampler. The samples of equilibrium vapor phase were taken from thermostated (T = 298 K) 22 ml vials containing 5 ml of solution or pure substance and transferred to the gas chromatograph. The ratio of areas of chromatographic peaks in experiments with a solution and with pure solute (denoted as A) is equal to the ratio of vapor pressures pA=S =pAsat of solute A over its dilute solution in solvent S and over pure A. The limiting activity coefficient A=S A=S cA=S =ðpAsat  xA=S Þ, where xA=S is the molar frac1 is given by: c1 ¼ p tion of A in solution. We made a correction of initial molar fraction of solute for the quantity of evaporated solute. Measurements were carried out at 3 to 4 different concentrations of every solute in the range (0.1 to 1.5) vol% and repeated 2 times for each concentration. No significant difference between the values of the activity coefficients at different concentrations in this range was observed. An average value of the Gibbs free energy of solution Dsoln GA=S ¼ RT ln cA=S from all measurements with the same dis1 solved compound was taken. The Gibbs free energy of solvation can be calculated using a formula Dsolv GA=S ¼ Dsoln GA=S  Dv ap GA , where the Gibbs free energy of vaporization Dv ap GA ¼ RT ln pAsat . The values of pAsat were taken from EPA database [21]. Results are presented in tables 2 and 3. The standard pressure is 1 bar and the standard state for solutions is a hypothetical ideal solution with unit molar fraction of a solute. Relatively large uncertainties were observed in the case of solvation of alkanes in PG (up to 16% for hexane, or ±0.4 kJ  mol1 in the Gibbs energies scale). This is apparently because of a low speed of dissolution of poorly soluble alkanes in viscous PG. It takes a long time to reach the vapor–liquid equilibrium in these systems. In our experiments, we subjected the solutions to vigorous shaking in a vortex. Without this step, alkanes almost did not dissolve, which can be judged by the chromatographic peak areas.

2.3. Measurement of limiting activity coefficients

3. Discussion

Limiting activity coefficients in PG and MC solutions were determined by GC headspace analysis using PerkinElmer Clarus

A comparison of the obtained results with literature values can be made for some systems. The data for solvation of octane and

34

I.A. Sedov et al. / J. Chem. Thermodynamics 78 (2014) 32–36

TABLE 2 and Gibbs free energies of solution Dsoln GA=PG and solvation from the gas phase Dsolv GA=PG in PG at T = 298 Ka. Experimental values of limiting activity coefficients cA=PG 1 Solute (A)

cA=PG 1

u(cA=PG )b 1

Dsoln GA=PG /(kJ  mol–1)

Dv ap GA /(kJ  mol–1)

Dsolv GA=PG /(kJ  mol–1)

n-Hexane n-Heptane n-Octane Benzene Toluene Ethylbenzene o-Xylene m-Xylene p-Xylene Fluorobenzene Chlorobenzene Naphthalene

63.0 109.0 150.0 11.8 17.0 26.2 26.3 28.4 26.7 8.5 13.6 136

10.0 15.0 20.0 0.5 1.0 1.5 1.5 1.0 1.5 0.2 0.5 10

10.3 11.6 12.4 6.1 7.0 8.1 8.1 8.3 8.1 5.3 6.5 12.2

4.0 6.9 9.9 5.1 8.1 10.8 11.7 11.2 11.0 5.6 10.2 22.5

6.3 4.7 2.6 1.0 –1.1 –2.7 –3.6 –2.9 –2.9 –0.3 –3.8 –10.4

a The total pressure inside the vials while measuring activity coefficients was p = 2.38 bar, standard uncertainty for pressure u(p) = 0.01 bar, standard uncertainty for temperature u(T) = 0.2 K. The standard state for liquid (at T = 298 K) substances is the pure liquid and for solids at T = 298 K it is the pure solid substance. b Standard uncertainty of the limiting activity coefficient.

TABLE 3 Experimental values of limiting activity coefficients cA=MC and Gibbs free energies of solution Dsoln GA=MC and solvation from the gas phase Dsolv GA=MC in MC at T = 298 Ka. 1 Solute (A)

cA=MC 1

u(cA=MC )b 1

Dsoln GA=MC /(kJ  mol1)

Dv ap GA /(kJ  mol1)

Dsolv GA=MC /(kJ  mol1)

n-Hexane n-Heptane n-Octane n-Nonane n-Decane n-Undecane Methylcyclopentane Cyclohexane Methylcyclohexane Cyclooctane Cyclohexene 1,7-Octadiene 4-Vinyl-1-cyclohexene Benzene Toluene Ethylbenzene o-Xylene m-Xylene p-Xylene p-Cymene Fluorobenzene Chlorobenzene Bromobenzene Naphthalene

16.3 22.4 27.9 41.4 56.8 94.6 11.0 11.4 14.7 20.1 7.4 10.6 8.6 2.4 3.6 4.9 5.0 5.2 5.2 10.0 1.8 2.9 3.5 11.0

0.3 0.3 0.2 0.5 0.6 1.0 0.3 0.3 0.4 0.3 0.1 0.2 0.1 0.2 0.2 0.1 0.1 0.1 0.1 0.2 0.1 0.2 0.1 1.0

6.9 7.7 8.3 9.2 10.0 11.3 5.9 6.0 6.7 7.4 5.0 5.8 5.3 2.2 3.2 3.9 4.0 4.1 4.1 5.7 1.4 2.6 3.1 5.9

4.0 6.9 9.9 12.7 15.5 18.6 4.2 5.1 6.9 12.2 5.3 8.7 9.6 5.1 8.1 10.8 11.7 11.2 11.0 15.5 5.6 10.2 12.9 22.5

2.9 0.8 1.6 3.5 5.5 7.3 1.7 1.0 0.3 4.7 0.3 2.8 4.2 2.9 4.9 6.9 7.8 7.1 6.9 9.8 4.2 7.6 9.8 16.6

a The total pressure inside the vials while measuring activity coefficients was p = 2.38 bar, standard uncertainty for pressure u(p) = 0.01 bar, standard uncertainty for temperature u(T) = 0.2 K. The standard state for liquid (at T = 298 K) substances is the pure liquid and for solids at T = 298 K it is the pure solid substance. b Standard uncertainty of the limiting activity coefficient.

toluene in MC are in a very good agreement with the (gas + liquid) partition coefficients reported by Park et al. [16]. If their values are recalculated into the Gibbs free energies of solvation, we obtain the same 1.6 and 4.9 kJ  mol1 for octane and toluene respectively. The value of lnc1 for cyclohexane in MC at T = 303.15 K was found [19] to be 2.5171, or Dsoln GA=MC = 6.2 kJ  mol1, which is different from our result at T = 298 K by 0.2 kJ  mol1. The limiting activity coefficients of the hydrocarbons (n-hexane to n-octane, benzene, and toluene) in PG at T = (316 and 324) K from the 1971 work of Leitman and Gaile [9] are 1.5 to 2 times higher than our values at T = 298 K and moreover, are growing up with decreasing temperature. In more recent articles, the data for the same systems are in better agreement with our results. For heptane in propylene glycol, the limiting activity coefficient estimated from LLE data [10] is 105 at T = 313 K and 120 at T = 308 K, while in [9] it is as large as 229 at T = 316 K. Using another work with LLE data [11], one can expect the same coefficient to be about 100 at T = 298 K against our value of 109. The magnitudes of Dsolv GA=PG calculated from results of the work [11]

are equal 6.6, 4.8, and 3.5 kJ  mol1 for hexane, heptane, and octane respectively. Our results (table 2) lie within 0.3 kJ  mol1 for hexane and heptane and within 1 kJ  mol1 for octane. In contrast, calculations of the Gibbs free energies of solvation for benzene, ethylbenzene, o-xylene and p-xylene from LLE data [11] give the values 2.2, 3.6, 2.9 and 1.4 kJ  mol1, while our measurements gave much lower values (1.0, 2.7, 3.6 and 2.9 kJ  mol1 respectively). However, these aromatic compounds are better soluble in PG and their activity coefficients at saturation may be significantly different from c1 . We also cannot exclude the uncertainties in the literature data [11], which can explain a very large difference in the reported solubilities of o-xylene and p-xylene in PG. In figure 1, the values of Dsolv GA=PG (plus signs) and Dsolv GA=MC (triangles) are plotted against the values of Dsolv GA=EG [5] for the same solute. A straight line corresponds to equation y = x. For all considered solutes in both solvents, the Gibbs free energies of solvation are lower than in EG. The relationships between them can be described by the following linear correlations:

35

I.A. Sedov et al. / J. Chem. Thermodynamics 78 (2014) 32–36

Dsolv GA=PG ¼ 0:780Dsolv GA=EG  3:21; 1 n ¼ 12; r ¼ 0:64kJ  mol ; R2 ¼ 0:9809;

ð1Þ

15 5

Dsolv GA=MC ¼ 0:865Dsolv GA=EG  7:59; 1 n ¼ 18; r ¼ 1:01kJ  mol ; R2 ¼ 0:9559:

ð2Þ

-5

Here, addition of a methyl group to a solvent molecule leads to a decrease of Dsolv G and, therefore, to an increase in solubility of low polar compounds. This is similar to the fact that methanol dissolves apolar compounds better than water, and ethanol better than methanol. The solubility of hydrocarbons in MC is better than in PG, which is obviously a consequence of different hydrogen bonding patterns of these two solvents. As mentioned above, MC is unable to form a branched network of hydrogen bonds like PG and EG. The solvophobic effect is weaker in MC than in PG. We can confirm this using our approach [1] for recognition and quantitative description of the solvophobic effects. In figure 2, we plot the Gibbs free energies against the enthalpies of solvation in PG, MC, and EG. The relationship between the Gibbs free energies and enthalpies of solvation is linear for many compounds with different structures dissolved in various nonassociated solvents without intermolecular hydrogen bonds in their bulk phase. We have shown [2] that 978 (solute + solvent) pairs consisting of solutes and solvents with different structure, polarity, molecular size and shape follow the correlation

Dsolv GA=s ¼ 0:627Dsolv HA=s þ 16:3; 1 n ¼ 978; r ¼ 1:6kJ  mol ; R2 ¼ 0:9768:

ð3Þ

The accuracy of a correlation for a group of solutes that belong to a single class of organic compounds is higher, while the coefficients are almost the same. For linear alkanes C1 to C8 dissolved in 34 different solvents (101 data points) [2], such correlation is given by: 1

Dsolv GA=s ¼ 0:632Dsolv HA=s þ 15:6; r ¼ 0:75kJ  mol ; R2 ¼ 0:9890: ð4Þ For solutions in self-associated solvents such as water, formamide, monohydric alcohols, ethylene glycol, the Gibbs free energy of solvation is always higher than for a solution in non-associated solvent, if the enthalpy of solvation keeps the same. Thus, the data points on DG vs DH plot lie above the line corresponding to equation (3) or (4). This line is shown in figure 2. Increased values of the Gibbs free energies leading to a decreased solubility are

-15 -25

-60

-50

-40

-30

-20

FIGURE 2. Gibbs free energy versus enthalpy of solvation in PG (pluses), MC (circles), and EG (crosses) at T = 298 K.

interpreted as a result of the solvophobic effect, and the deviation from the correlation line for hydrocarbons (equation (4)) can be used as its quantitative measure [1,2]. It should be understood that the solvophobic effect may influence both the entropic and enthalpic components of the Gibbs free energy of solvation. However, we have concluded that while in aqueous [22] and formamide [7] solutions the enthalpy is significantly affected by the solvophobic effect, in alcohols including EG [5] it has a rather small influence on the enthalpies. Therefore we view the magnitude of deviation of data points in figure 2 from equation (4) as a contribution of the solvophobic effect into the Gibbs free energy of solvation denoted as Ds:e: G:

Ds:e: GA=S ¼ Dsolv GA=S  0:632Dsolv HA=S  15:6:

ð5Þ

In figure 3, the values of Ds:e: G for solutions of low polar compounds in PG, MC, and EG determined using equation (5) are plotted against a measure of molecular size – the characteristic molecular volume of a solute V Ax , which is calculated from atomic contributions [23]. It is clear that in all three solvents the solvophobic effect is strengthening with the growing solute molecular volume. A linear regression of Ds:e: GA=EG calculated by equation (5) on V Ax gives the following equation:

Ds:e: GA=EG ¼ 6:96V Ax þ 3:01; 1

n ¼ 17; r ¼ 0:85 kJ  mol ; R2 ¼ 0:7133:

ð6Þ

In our previous paper [5], we determined the Gibbs solvophobic effect energies in EG using a different approach: by calculating the part of the solvation Gibbs free energy that is not connected with the solvophobic effects using extrathermodynamic equations. These calculations can be more precise because the values of the enthalpies of solvation, which have a relatively large experimental

15 10

5 0 -5 -10 -15 -20 -10

-5

0

5

10

15

FIGURE 1. Comparison of the Gibbs free energies of solvation in PG (pluses) and MC (circles) for low polar solutes with the Gibbs free energies of solvation for the same compounds in EG at T = 298 K.

FIGURE 3. Gibbs free energy of the solvophobic effect for various solutes in PG (pluses), MC (circles), and EG (crosses) versus the characteristic molecular volume of a solute.

36

I.A. Sedov et al. / J. Chem. Thermodynamics 78 (2014) 32–36

10

0

1-Propanol 1-Butanol

carbon and the strength of the solvophobic effects are dependent on the solvent nature.

Ethylene glycol Propylene glycol Methyl Methanol cellosolve Ethanol

4. Conclusion

2-Propanol

1-Octanol

-10

-35

-30

-25

-20

FIGURE 4. Gibbs free energy versus enthalpy of solvation of n-hexane in various alcohols at T = 298 K.

uncertainty, are not used. Moreover, we have calculated Ds:e: GA=EG for many other solutes for which the enthalpies of solvation are unknown, and the quality of correlation as measured by R2 was also better. However, the slope and intercept of the regression equation [5] were very similar to equation (6): A=EG

Ds:e: G

¼

6:92V Ax

þ 3:24; 1

n ¼ 37; r ¼ 0:70 kJ  mol ; R2 ¼ 0:9398:

ð7Þ

For solutions in MC, a correlation similar to equation (6) is given by:

Ds:e: GA=MC ¼ 2:87V Ax þ 0:12; 1

n ¼ 18; r ¼ 0:36 kJ  mol ; R2 ¼ 0:8130:

ð8Þ

Due to the low variance of the Gibbs energies of solvation in PG and V Ax values of the studied solutes, the slope and intercept determined from a regression are not precise and have a large uncertainty. If we fix the intercept to be 1.50 (roughly an average of the intercepts of equations (6) and (8)), then the correlation for PG is the following:

Ds:e: GA=PG ¼ 4:91V Ax þ 1:50; 1

n ¼ 12; r ¼ 0:81 kJmol ; R2 ¼ 0:6408:

ð9Þ

The correlation coefficient is still low because of the low variance. The slope appears to be a little lower than in a similar correlation [1] for the solvophobic effect contribution in solutions in methanol (5.17), but the intercept is higher (it is 0.23 in the correlation for methanol). The Gibbs solvophobic effect energies in methanol and PG have close values in most cases. For solutions in MC (equation (8)), the slope is between those for butanolic and pentanolic solutions (3.06 and 2.64) [1]. In figure 4, the Gibbs free energy and enthalpy of solvation of nhexane in various alcohols including PG, MC, and EG are plotted against each other. The line corresponding to equation (4) is also shown. This graph shows how the solvation properties of a hydro-

Thermodynamic functions of solvation of low polar solutes in PG and MC have been determined and analyzed. We found another two examples of a general tendency of the solvophobic effect to become stronger with the growing molecular volume of a solute. The strength of the solvophobic effect in various (solute + solvent) systems can be determined using a plot of the Gibbs free energy against the enthalpy of solvation. In PG, the solvophobic effect is almost as strong, as in methanol, while in EG it is significantly stronger, and in MC – significantly weaker than in PG. The obtained results are important for understanding the connection between the nature of a solvent, the strength of the solvophobic effects, and their influence on the physico-chemical processes in solutions. Acknowledgement The work is performed according to the Russian Government Program of Competitive Growth of Kazan Federal University and supported by RFBR (Project No. 14-03-31990). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]

[22] [23]

I.A. Sedov, M.A. Stolov, B.N. Solomonov, J. Phys. Org. Chem. 24 (2011) 1088. I.A. Sedov, B.N. Solomonov, J. Struct. Chem. 54 (2013) 262. A. Ray, Nature 231 (1971) 313. K. Gekko, S. Koga, Biochim. Biophys. Acta Bba Protein Struct. Mol. Enzym. 786 (1984) 151. I.A. Sedov, M.A. Stolov, B.N. Solomonov, Fluid Phase Equilib. 354 (2013) 95. I.A. Sedov, M.A. Stolov, B.N. Solomonov, Russ. J. Phys. Chem. 85 (2011) 621. I.A. Sedov, M.A. Stolov, B.N. Solomonov, J. Chem. Thermodyn. 64 (2013) 120. I.A. Sedov, M.A. Stolov, B.N. Solomonov, J. Chem. Thermodyn. 589 (2014) 247. Y.I. Leitman, A.A. Gaile, Russ. J. Phys. Chem. 45 (1971) 39. S.O. Derawi, G.M. Kontogeorgis, E.H. Stenby, T. Haugum, A.O. Fredheim, J. Chem. Eng. Data 47 (2002) 169. F. Oprea, O. Gutu, Pet. - Gas Univ. Ploiesti Bull. Tech. Ser. 60 (2008) 18. S. Ahmadian, V. Panahi-Azar, M.A.A. Fakhree, W.E. Acree, A. Jouyban, J. Chem. Eng. Data 56 (2011) 4352. M.E.R. McHale, K.S. Coym, K.A. Fletcher, W.E. Acree, J. Chem. Eng. Data 42 (1997) 511. A. Vega, F. Díez, R. Esteban, J. Coca, J. Chem. Eng. Data 46 (2001) 98. L. Rohrschneider, Anal. Chem. 45 (1973) 1241. J.H. Park, A. Hussam, P. Couasnon, D. Fritz, P.W. Carr, Anal. Chem. 59 (1987) 1970. A. Landauer, R.N. Lichtenthaler, J.M. Prausnitz, J. Chem. Eng. Data 25 (1980) 152. H. Bijl, T.W. de Loos, R.N. Lichtenthaler, Fluid Phase Equilibr 22 (1985) 321. F. Carmona, J. Chem. Thermodyn. 33 (2001) 47. B. Marrufo, S. Loras, M. Sanchotello, J. Chem. Eng. Data 55 (2010) 62. US EPA. 2011. Estimation Programs Interface SuiteTM for MicrosoftÒ Windows. United States Environmental Protection Agency, Washington, DC, USA, n.d. I.A. Sedov, B.N. Solomonov, J. Chem. Thermodyn. 42 (2010) 1126. M.H. Abraham, J.C. McGowan, Chromatographia 23 (1987) 243.

JCT 14-81