Contents lists available at ScienceDirect

J. Chem. Thermodynamics

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

Effect of temperature and composition on the surface tension and

surface properties of binary mixtures containing DMSO and short chain

alcohols

Ahmad Bagheri ⇑, Mostafa Fazli, Malihe Bakhshaei

Department of Chemistry, Semnan University, P.O. Box 35131-19111, Semnan, Iran

A b s t r a c t

A r t i c l e i n f o

Surface tension of binary mixtures of methanol, ethanol and isopropanol with DMSO (dimethyl sulfoxide) was measured over the whole range of composition at atmospheric pressure of 82.5 kPa within the temperatures between (298.15 and 328.15) K. The experimental measurements were used to calculate in surface tension deviations (Dr). The sign of Dr for all temperatures is negative (except of methanol/DMSO system) because of the factors of hydrogen bonding and dipole–dipole interactions in the DMSO-alcohol systems. Surface tension values of the binary systems were correlated with FLW, MS, RK and LWW models. The mean standard deviation obtained from the comparison of experimental and calculated surface tension values for binary systems with three models (FLW, MS and RK) at various temperatures is less than 0.83. Also, the results of the LWW model were used to account for the interaction energy between alcohols and DMSO in binary mixtures. The temperature dependence of r (surface tension) at fixed composition of solutions was used to estimate surface enthalpy, Hs, and surface entropy, Ss. The results obtained show that the values of the thermodynamic parameters for alcohol/DMSO mixtures decrease with increasing alkyl chain length of alcohol. Finally, the results are discussed in terms of surface mole fraction and lyophobicity using the extended Langmuir (EL) isotherm.

:Article history

Received 20 January 2016

Received in revised form 28 May 2016

Accepted 9 June 2016

Available online 13 June 2016

:Keywords

Surface tension

Correlation

Surface enthalpy

Lyophobicity

contain polar or polar aprotic solvents [13,11,12]. The experimental values of surface tension of the binary systems consisting of DMSO with methanol, ethanol and isopropanol were determined over the whole range of composition at (298.15, 308.15, 318.15 and 328.15) K.

After surveying the literature, a number of research studies have been located on the surface tensions of DMSO with alkanols binary mixtures at various temperatures. However, no surface tension values were previously reported for DMSO with isopropanol at different temperatures and DMSO with methanol and ethanol at (308.15, 318.15 and 328.15) K [1,2,13–15].

After experimental measurements, the data obtained were
analysed using various methods. In the first section, the concentration dependence of the surface tension of binary DMSO-alcohol mixtures at various temperatures are correlated using Fu et al. (FLW) [16], the Myers-Scott (MS) [17] and Redlich-Kister (RK) [18] and Li et al. (LWW) equations, and then in a new approach, the effect of alcohol structure on the interaction energy values (between alcohols and DMSO) is discussed in the binary systems [19,20].

In the second section of this work, by employing the measured surface tensions at various temperatures, the thermodynamic

1. Introduction

Among the various experimental methods, surface tension is one of the most powerful techniques that provide information about surfaces and intermolecular interaction. The study of the surface properties in liquid mixtures is of great interest from both scientific and industrial points of view [1–3].

Dimethyl sulfoxide (DMSO) is a non-aqueous dipolar aprotic solvent used in pharmaceuticals, surface cleaners, extraction, electrochemistry and as a solvent for polymers. As pure solvent, DMSO has a large dipole moment and high dielectric constant (the relative dielectric constant e = 46.50 and dipole moment l = 4.06 D at 298.15 K) [4–6].

Also, alcohols are considered to be a water-like solvent in terms of hydrogen-bonding ability and are an important solvent due to their use in the chemical, pharmaceutical, fuel cell, cosmetic industries
and others [7–10].

This study is a continuation of our systematic experimental research on surface properties of binary liquid mixtures that

A. Bagheri et al. / J. Chem. Thermodynamics 101 (2016) 236–244

2.2. Apparatus and procedure

The surface tension o each sample liquid (pure or mixture) was measured using the platinum-iridium ring method with a PC controlled KSV Sigma 701 tensiometer. The ring was thoroughly cleaned and flamed before each measurement. The measurements were automatically corrected to the actual values by means of the Huh and Mason compensation for interface distortion [30,31]. The use of boiling distilled water can be a good (alternative) procedure (for checking the performance of the instrument). The uncertainty in the calculated surface tensions (the mean value from 7 points) was estimated to be ±0.04 mNm1. The measurements were carried in the temperature range from 298.15 K to 328.15 K and at pressure of 82.5 kPa. The sample under measurement was kept thermostatted in a double jacketed glass cell by means of a water bath, using a LAUDA circulator (model ECO RE415) equipped with a Pt100 probe, immersed in solution, and able to control the temperature within ±0.01 K. The binary mixtures were prepared by mass using a Sartorius analytical balance (model BP 121S, accurate to ±0.1 mg). The standard
uncertainty of the mole fractions is ±1.000  103.

3. Results and discussion

3.1. Correlation of surface tension and concentration in binary mixtures

The concentration dependence of the surface tension of mixtures can be represented in terms of the surface tension deviation, Dr, defined as:

where r is the surface tension of the mixture, and r i is the surface
tension of the pure component I at the same temperature and pressure of the mixture, xi is the mole fraction of the component I and n is the number of components.
A few empirical and thermodynamic-based equations are available to correlate the surface tension; some of them have recently proposed and are well founded on a thermodynamic basis. Fuet al. (FLW) proposed a two-parameter equation to correlate the surface tension with the composition in the binary systems which is based on the local composition concept due to the Wilson equation [16]. Another surface tension correlation was proposed by Liet al. (LWW) for liquid mixtures, which is based on the Wilson equation for the excess Gibbs energy [19,32]. Other empirical equation such as Myers Scott (MS) has been applied for the correlation of binary surface tension values [17]. The relevant equations for all the models for surface tension (or the surface tension deviation) correlation, in this work, are summarized in Table 3.

In Table 4, the measured surface tension is presented for binary
systems over the range T = (298.15–323.15) K at intervals of 10 K
as a function of mole fraction.

Figs. 1–3 show the behaviour of surface tension of alcohol in DMSO at various temperatures. In all systems, the surface tension, r, decreased with increasing alcohol concentration. This trend is non-linear, with the change in surface tension caused by a given change in alcohol mole fraction being greater at low mole fraction than at high mole fraction. On the other hand, as discussed above, DMSO and alcohol interact strongly and mixtures form non-ideal solutions particularly in the low concentration regime. This behaviour is typically explained by a difference in distribution of molecules between the surface and the bulk of the liquid. In a characteristic case, the compound having a lower surface tension

properties of the surface (Hs and Ss) have been obtained. Finally, a Langmuir type isotherm model (or the extended Langmuir (EL)) is employed to determine the surface concentrations from the knowledge of the bulk mole fractions. The results provide information on the molecular interactions between the unlike molecules that exist at the surface and the bulk at various temperatures [20,21].

2. Experimental

2.1. Materials

All compounds were purchased from Merck and used without any further purification. Purity of each compound was ascertained by gas chromatography (GC type Agilent 7820A Agilent Technologies). The water mass fraction of the all components was determined by Karl Fisher (K.F.) titration, and it was found to be less than 0.0008 for DMSO and less than 0.004 for alcohols. Chemicals were kept in dark bottles. The detailed information of the materials used in the experiment is listed in Table 1.

The purity of components was checked by comparing the measured surface tension with those reported in the literature [13–15,22–29]. The resultant values are in good agreement with values found in the literature and reported in Table 2 [14]. Bi-distilled water has been used for checking the performance of the tensiometer instrument in previous publication [25].

A. Bagheri et al. / J. Chem. Thermodynamics 101 (2016) 236–244

Fig. 2. Experimental surface tension, r, against mole fraction, x2, at various temperatures: (d) 298.15 K, (h) 308.15 K, (N) 318.15 K and (*) 328.15 K, for ethanol (2)/DMSO (1). The continuous dashed curves represent the correlation with the FLW equation.

(2)/DMSO (1) indicate that the components with the strongest molecular interactions in each binary mixture settle down in the bulk liquid phase instead of interacting in the surface phase between liquid and vapour phases, moving the curves to the rich region of these compounds.

However, there is a different trend for the methanol/DMSO mixture, the strong interactions between DMSO and methanol decrease the tendency of methanol to adsorb at the liquid–vapour

(alcohol) is expelled from the bulk to the liquid vapour interface due to the attractive forces between solvent molecules [1,2,20].

Over the whole composition range, the values of Dr for the binary mixtures of ethanol and isopropanol with DMSO are negative but the values of Dr for the methanol/DMSO mixtures are positive. The values of jDrj for mixtures of alcohol (ethanol or isopropanol)/DMSO decrease with increasing temperature and the alkyl chain length of alcohol (see Fig. 4).

In the binary mixtures of DMSO with ethanol and isopropanol, the surface tension deviation (Dr) values increase with increase in the difference of surface tension values of pure components, pass through a minimum at x1  0.25–0.375 (in isopropanol (2)/DMSO (1)) and x2  0.33–0.45 (in ethanol (2)/DMSO (1)). The asymmetry in the curves of isopropanol (2)/DMSO (1) and ethanol

A. Bagheri et al. / J. Chem. Thermodynamics 101 (2016) 236–244

methanol molecules form hydrogen-bonded complexes with DMSO, in which these interactions partially replace methanol– methanol hydrogen bonding [33,36–39].

In Figs. 5–7 we have plotted the experimental and fitted values of surface tension of the binary systems of alcohol/DMSO as a function of the composition with FLW, MS, RK and LWW models:

In Table 5, the adjusted coefficients of the equations used to correlate the binary data are listed as well as the respective standard deviations of the fittings. The adjustable parameters in LWW model (Table 3) represent the following:

Fig. 5. Plot of surface tension, r, against mole fraction at 298.15 K. The symbols refer to the experimental data for alcohol (2)/DMSO (1) mixture: (d) methanol, (N) ethanol and (s)  isopropanol. The continuous dotted curves represent the correlation with the FLW equation.

Fig. 6. Plot of surface tension deviation, Dr, against mole fraction for methanol (2)/ DMSO (1) system. The symbols refer to the experimental data at various temperatures: (d) 298.15 K, (s) 308.15 K, (N) 318.15 K and (j) 328.15 K. The
continuous dotted curves represent the correlation with MS equation and the solid curves represent the correlation with the RK equation.

Fig. 4. Plot of surface tension deviation, Dr, against mole fraction at 308.15 K. The symbols refer to the experimental values for the alcohol (2)/DMSO (1) mixture: (N) methanol, (j) ethanol and (d) isopropanol.

interface, which is in turn responsible for the positive Dr for the methanol/DMSO system. According to results of the ERAS (extended real associated solution) model, the value of cross-association constant between alcohol and DMSO (Kalcohol-DMSO) was reduced from methanol to 1-pentanol at 298.15 K due to strong reduction of hydroge  bonding [33–35]:

The two factors (surface tension and hydrogen bonding) are antagonistic in migration of methanol molecules from the bulk to interface. This is due to the hydrogen-bonding factor between methanol and DMSO, the more abundant methanol prefers the bulk liquid phase (according to the results of ERAS model), but due to the surface tension factor it prefers the interface. From the experimental results, it appears that the hydrogen-bonding factor is more important than the surface tension factor. The strong interactions between DMSO and methanol decrease the tendency of methanol to adsorb at the liquid–vapour interface. Numerous studies reported in the literature have provided evidence that

A. Bagheri et al. / J. Chem. Thermodynamics 101 (2016) 236–244

An iterative method has been used to derive the a and b parameters by inserting experimental values in Eq. (10) [19,20]. Table 7 lists the values of the adjustable parameters a and b obtained by fitting Eq. (10) to the experimental results at 298.15 K.

Fig. 12 shows the experimental surface pressure values together with the fitting curves obtained using the EL model for the alcohol/DMSO systems.

The agreement between the experimental surface pressure values and those calculated from the EL model relation is found to be satisfactory.

The b values for the mixtures of DMSO with ethanol and isopropanol are greater than unity and parameter k is zero (a  1), which indicates that the compound prevalently adsorbed in both cases is the alcohol. Both contributions tend to decrease the surface tension of the mixture and the Dr–xalcohol curves show negative minima. Also, Table 7 also indicates that the lyophobicity (b) in the alcohol/DMSO mixtures increases with the alkyl chain length.

In the methanol/DMSO system, the fittings with EL model provide values of b = 1 and k > 0 (a = 1.37), this shows, the concentration of each compound being the same in bulk and surface phases (due to the lyophobicity is not occurs) and the unlike-pair interactions contribute to decrease the surface tension of the mixture.

In similar cases of methanol (2)/DMSO (1) (a positive maximum in the Dr–x2 curve), several papers proposed that the molecules of methanol because they are much smaller than those of DMSO (v2/v1 = 0.57) and so, many more methanol molecules are needed to cover the same area. Thus, despite the same concentration of components at the interface, the contribution of the DMSO dominates in the Dr–x2 representation [42–44].

The ‘‘non-ideality” of the surface tension of a given system is also reflected using the surface layer concentration. A plot of the surface mole fraction, x2,s, versus the bulk mole fraction, x2, gives us the possibility of establishing the non-ideality of a given system by observing the form of the corresponding curve (see Fig. 13). For ideal systems, this curve tends to be a line very close to the one a

As shown in Fig. 11, surface enthalpy (Hs) is positive and depends remarkably both on the bulk composition and temperature.

3.4. The surface concentration of components in binary mixtures

The surface concentration is another parameter that produces some useful information about the surface behaviour of mixtures. An applied model (Extended Langmuir, EL) was reported, which describes the surface tension of binary liquid mixtures as a function of the bulk composition [21].

If /2;s is the volume fraction of component 2 in the surface phase, and /2 is the volume fraction of component 2 in the bulk phase, then we define them in general as follow:

where ni is the number of moles of the ith component, v1 and v2 are molar volume of the 1 and 2.

Briefly, this model considers the surface of a binary liquid mixture
as a thin but finite layer and the following expression was developed for the relationship between /2;s and /2:

A. Bagheri et al. / J. Chem. Thermodynamics 101 (2016) 236–244

The standard Gibbs energy (DG0) can be used to evaluate the spontaneity of an adsorption process. A negative DG0 indicates the adsorption occurs spontaneous and is thermodynamically stable, whereas the positive DG0 means this process is not favoured and represents a non-spontaneous transfer [46].

The results in Table 7 show that the value of DG0 increases with the increase of the alkyl chain length of the alcohol (due to the lesser interaction between alcohol and DMSO). In the other words, these results indicate that the adsorption process becomes more favourable with increasing alkyl chain length of the alcohol from methanol to isopropanol.

As can be seen in Table 7, the agreement between the experimental results and the values obtained from the Eq. (11) is reasonable.

4. Conclusion

In this paper, surface tension values for binary mixtures of alcohol (methanol, ethanol and isopropanol) with DMSO, are reported over ranges of composition at various temperatures. In all systems, the surface tension of mixtures decreases with increasing temperature. The results of the present study show that by adding alcohol to DMSO the surface tension decreases non-linearly and most surface tension changes occur at a low concentration of alcohol. The surface tension deviation, Dr, is positive for methanol/DMSO, and is negative for DMSO with ethanol and isopropanol. This shows that the interaction is strong, namely formation of hydrogen bonding between DMSO and methanol, and the interaction between DMSO with ethanol and isopropanol is weak. The experimental surface tension values for binary mixtures were correlated by four models (FLW, MS, RK and LWW) and the best results were obtained by the Myers-Scott (MS) model. In a new approach, we describe the alkyl chain length of alcohol on the interaction energy values in binary mixtures by using values of the surface tension. The results obtained show the Ualcohol-DMSO value decreases with increasing alkyl chain length of alcohol due to the lesser interaction between alcohol and DMSO at any temperature. The values of surface mole fraction of alcohol were calculated using the EL model. The extracted b values indicate the greater affinity of alcohol for the surface, and this trend is amplified with the increase of the alkyl chain length of the alcohol.

Acknowledgments

The authors are grateful to Dr. S. Maryam Sajjadi who kindly read the manuscript and made many helpful suggestions. Also, we gratefully acknowledge the financial support received for this research work from the Research Council of Semnan University.