Acta Chim. Slov. 2004, 51, 117-126. 117 Scientific Paper MEASUREMENTS AND CORRELATION OF OSMOTIC COEFFICIENTS AND EVALUATION OF VAPOR PRESSURE FOR SOLUTIONS OF KCH3COO AND NaCH3COO IN METHANOL AT 25 °C† Karamat Nasirzadeh,"' * Roland Neueder b Department oj Lhemistry, taculty oj Science, Azerbaijan University oj 1 arbiat Moallem, 1 abriz, Iran Institute of Physical and Theoretical Chemistry, University of Regensburg, 93040 Regensburg, Germany Fax: +49-941-9434532; E-mail: Karamat.nasirzadeh@chemie.uni-regensburg.de This paper is dedicated to Prof. Josef Barthel in honour of his 75 birthday. Received 22-11-2003 Abstract The osmotic coefficients of potassium acetate and sodium acetate in methanol have been measured by the isopiestic method at 25 °C. Sodium iodide was used as isopiestic standard for the calculation of osmotic coefficients. The molality ranges covered in this study correspond to about 0.17-2.51 molkg-1 for potassium acetate and 0.25-1.76 molkg-1 for sodium acetate. The system of equations of Pitzer-Mayorga and MSA-NRTL were used to fit osmotic coefficients. The parameters from the fit were used to calculate the vapor pressures. The osmotic coefficient data are successfully correlated with these models, which provide reliable predictions of vapor pressures. Key words: osmotic coefficient, isopiestic, methanol, models Introduction Thermodvnamic properties of electrolvte solutions are important for a varietv of applications in the chemical processes in industries. Electrolvtes are involved in numerous processes including environmental applications such as chemical waste disposal, separation process and electrochemical process. Osmotic coefficient data of binary electrolvte solutions are required to describe the thermodvnamic behavior of electrolvte solutions with organic solvents. These data are also useful to predict thermodvnamic properties of electrolvtes in mixed solvents. " However; accurate thermodvnamic data are very scare for non-aqueous electrolvte solutions. Barthel and his co-workers " have made accurate vapor pressure-lovvering measurements on a few non-aqueous electrolvte solutions from vvhich osmotic coefficient values may be calculated. There are also other reports on the vapor pressure of some electrolvtes in methanol. ' The reported data for Nal in methanol ' solutions have been used as the isopiestic K. Nasirzadeh, R. Neueder: Measurements and Correlation of Osmotic Coefficients and Evaluation… 118 Acta Chim. Slov. 2004, 51, 117-126. reference standards and the osmotic coefficients of some solutes in methanol have been measured. In this work the osmotic coefficients and vapor pressures of solutions of potassium acetate and sodium acetate in methanol are reported. For solutions of KCH3COO in methanol, a few vapor pressure data has been reported; however, for concentrations lower than 1 mol-kg" , only one data point has been given. Information for the activitv and osmotic coefficients of the solution of sodium acetate in methanol has not been reported. The osmotic coefficients have been measured using an improved isopiestic apparatus. For isopiestic reference sodium iodide in methanol solutions were used as described previouslv. Vapor pressures for the solutions of investigated electrolvtes in methanol have been calculated from the osmotic coefficient data by the relevant thermodvnamic relations. The Pitzer-Mavorga model and MSA-NRTL model were successfullv used to reproduce the experimental osmotic coefficients and to derive the vapor pressures. Experimental Apparatus and procedure. The isopiestic apparatus emploved in this research is essentiallv similar to the one used previouslv. Recentlv, this technique has been used for the measurement of osmotic coefficients of some inorganic salts in methanol. This apparatus consisted of a five-leg manifold attached to round-bottom flasks. The five flasks were typically used as follows. Two flasks contained the standard Nal solutions, two flasks contained either potassium acetate or sodium acetate solutions, and the central flask was used as a methanol reservoir. The apparatus was held in a constant temperature bath for at least 120 hours for equilibration at (298.15 ± 0.005) K. Chemicals. The methanol and salts were obtained from Merck. They were ali supra pure reagents (methanol GR., min. 99.8%; Nal, GR., min. 99.5%; KCH3COO GR., min. 99.5%, NaCHaCOO, GR, min 99.5%). Ali chemicals were used without further purification. The salts were dried in an electrical oven at about 393 K for 24 h prior to use. Results and Discussions Experimental results. Isopiestic equilibrium molalities with reference standard solutions of Nal in methanol as reported in Tables 1 and 2 enabled the calculation of the K. Nasirzadeh, R. Neueder: Measurements and Correlation of Osmotic Coefficients and Evaluation… Acta Chim. Slov. 2004, 51, 117-126. 119 osmotic coefficient, $ of the solutions of potassium acetate and sodium acetate in methanol from 0 = (y*fm*/vm) Q) where v and v are the sum of stoičniometnc numbers of anion and cation,v++v_, in the reference solution and in the solutions of potassium acetate or sodium acetate, respectively, m is the molality of the reference standard in isopiestic equilibrium with these solutions, and (j) is the osmotic coefficient of the isopiestic reference standard, calculated at m . The necessary (j) values at any m were obtained from the fitted Pitzer and Mayorga equation, including the /r ' term, as described by Zafarani-Moattar and Nasirzadeh. It was shown that, using <^i)=2, 0C(2)=1.4, f? =0.40830, fy -1.04430, fy --0.875 and (7=-0.02224, the osmotic coefficients of the isopiestic reference standard solutions, 0, are reproducible with standard deviation of 0.005 for Nal in methanol solutions in the range (0.02 to 4.33) mol-kg" at 25 °C. Table 1. Experimental isopiestic molalities, osmotic coefficients, vapor pressures and activity of methanol for KCH3COO in methanol at 25 °C. mNaI / mKCH3COO/ yexp 0calc Pexp / as (mol-kg") (mol-kg") 0.0000 1.000 1.000 (kPa) 16.958 0.0000 1.0000 0.1720 0.1783 0.799 0.799 16.801 0.9909 0.2762 0.2915 0.793 0.793 16.705 0.9853 0.3760 0.4044 0.794 0.794 16.607 0.9796 0.4719 0.5171 0.798 0.798 16.509 0.9739 0.5640 0.6295 0.802 0.802 16.410 0.9682 0.6527 0.7416 0.807 0.807 16.311 0.9624 0.7383 0.8534 0.812 0.812 16.211 0.9566 0.8210 0.9649 0.817 0.817 16.111 0.9507 0.9010 1.0761 0.821 0.822 16.011 0.9449 1.0540 1.2976 0.830 0.830 15.811 0.9333 1.1634 1.4630 0.836 0.836 15.662 0.9246 1.2688 1.6281 0.842 0.842 15.512 0.9159 1.3371 1.7380 0.845 0.846 15.413 0.9101 1.3708 1.7929 0.847 0.847 15.364 0.9072 1.3641 1.7819 0.847 0.847 15.374 0.9078 1.4699 1.9577 0.852 0.853 15.215 0.8986 1.5345 2.0676 0.856 0.856 15.116 0.8928 1.5981 2.1778 0.859 0.859 15.017 0.8870 1.6296 2.2329 0.861 0.860 14.968 0.8841 1.7227 2.3990 0.865 0.865 14.819 0.8755 1.6918 2.3435 0.863 0.863 14.869 0.8784 1.7533 2.4545 0.866 0.866 14.770 0.8726 1.7838 2.5102 0.868 0.868 14.720 0.8697 K. Nasirzadeh, R. Neueder: Measurements and Correlation of Osmotic Coefficients and Evaluation… 120 Acta Chim. Slov. 2004, 51, 117-126. From the calculated osmotic coefficient data, the activity of methanol in potassium acetate and sodium acetate solutions and the vapor pressure of methanol over these solutions were determined at isopiestic equilibrium molalities, with the help of the following thermodynamic relations: 0 = -\naJvmMs (2) ln(p/p*)+ (Bs - V* )(p - p* )/RT (3) In these equations, as is the activity of solvent, Bs, V s and p are second virial coefficient, molar volume and vapor pressure of pure methanol, respectively. The values of Ms=0.032042, 5,s=-2.075x10" m -mol" , Vs = 4.073x10" m -mol" and p = 16957.7 Pa (taken from Barthel et al. ) were used at 298.15 K. Table 2. Experimental isopiestic molalities, osmotic coefficients, vapor pressures and activity of methanol for NaCH3COO in methanol at 25 °C. mNaI / mNaCH3COO / yexp 0calc ^exp' as (mol-kg") (mol-kg") (kPa) 0.0000 0.0000 1.000 1.000 16.958 1.0000 0.2369 0.2558 0.776 0.774 16.743 0.9874 0.2582 0.2822 0.769 0.773 16.721 0.9862 0.2691 0.2943 0.770 0.772 16.711 0.9854 0.3042 0.3319 0.776 0.772 16.679 0.9836 0.3512 0.3878 0.773 0.773 16.632 0.9810 0.3732 0.4154 0.771 0.773 16.608 0.9797 0.3989 0.4436 0.776 0.774 16.585 0.9782 0.4406 0.4948 0.775 0.776 16.541 0.9757 0.4914 0.5554 0.779 0.778 16.489 0.9726 0.5103 0.5789 0.780 0.779 16.469 0.9715 0.5903 0.6799 0.783 0.783 16.381 0.9664 0.6347 0.7385 0.784 0.786 16.330 0.9636 0.6550 0.7601 0.790 0.787 16.312 0.9622 0.7324 0.8676 0.790 0.791 16.217 0.9570 0.7491 0.8863 0.794 0.792 16.201 0.9559 0.7623 0.9084 0.791 0.793 16.182 0.9550 0.7869 0.9424 0.792 0.794 16.152 0.9533 0.8441 1.0246 0.793 0.797 16.079 0.9492 0.8593 1.0441 0.796 0.797 16.062 0.9481 0.8958 1.0964 0.798 0.799 16.016 0.9455 0.9812 1.2215 0.802 0.803 15.906 0.9391 1.0254 1.2877 0.804 0.805 15.848 0.9358 1.1160 1.4175 0.814 0.808 15.735 0.9287 1.1575 1.4997 0.807 0.810 15.664 0.9254 1.1869 1.5384 0.813 0.810 15.631 0.9230 1.3163 1.7631 0.813 0.815 15.443 0.9123 K. Nasirzadeh, R. Neueder: Measurements and Correlation of Osmotic Coefficients and Evaluation… Acta Chim. Slov. 2004, 51, 117-126. 121 h A comparison of our vapor pressure data to that of Tomasula et al. for KCH3COO in methanol is given in Figure 1. Figure 1 shows that the data of Tomasula et al. are somewhat higher than those obtained in this work. However, close examination of the Tomasula et al. data indicates that these authors have used/?*=17.08 kPa for vapor pressure of the pure methanol which is slightlv higher (0.12 kPa) than the value of 16.96 kPa used in this work. If the Tomasula et al. data were corrected according to this value (0.12 kPa), the obtained values are in a good agreement with those of this work. 17.5 0 This work 17 j ) x Tomasula et al. [11] 16.5 O O O O <* 16 Ox O X 15.5 - 0 X ox Po X 15 - ogx 14.5 0.5 1 1.5 2 molality / mol-kg"1 2.5 Figure 1. Comparison of vapor pressures for KCH3COO in methanol solutions at 25 °C. Correlation of data Pitzer model. Several models are available in the literature for the correlation of osmotic coefficients as a function of molalities. The model of Pitzer and Mavorga has been successfullv used for aqueous and in a few cases, for non-aqueous electrolvte solutions.12"14 The experimental osmotic coefficient data were correlated with the model of Pitzer and Mavorga for solutions of KCH3COO and NaCHaCOO in methanol. This model has the following form 15 (j) -1 = f* + mB'p + m C 2r* (4) o 3 K. Nasirzadeh, R. Neueder: Measurements and Correlation of Osmotic Coefficients and Evaluation… 122 Acta Chim. Slov. 2004, 51, 117-126. where S^ A T1/2 /m , 7 T" 1/2 1 (5) (6) A =(\/3)(27NAdsy[e2 / 471^^(7) B* =^+^Qxp[-a(l)Il/2]+0Qxp[-a(2)I1'2] (7) In these equations p , /? , /? ' and č/ are Pitzer’s ionic-interaction parameters; and b are adjustable parameters, and A§ is the Debye-Hiickel constant for the osmotic coefficient on the molal basis. The remaining symbols have their usual meaning. For methanol solutions A§=\294 kg -mol" was calculated using equation (6). From the analysis of the experimental osmotic coefficient data, we found that the values of 6=3.2 kg .mol" , OC(\)=2.0 kg .mol" and CX(2)=IA kg .mol" were satisfactory at 298.15 K. The ion-interaction parameters obtained from the experimental osmotic coefficient data for the investigated systems are shown in Table 3. Table 3. Pitzer parameters for methanol solutions of KCH3COO and NaCH3COO calculated from osmotic coefficients" at 25 °C. no. of molality 0® 0.D data range KCH3COO 23 0.17- 2.51 0.008128 -0.687219 NaCH3COO 26 0.25-1.76 0.026218 -0.128391 P 2) sd(^) 0.838449 0.004572 0.0001 -2.118794 1.9988 0.002 %=1.294 kg1/2-mol"1/2; 6=3.2 kg1/2-mol"1/2; (\X)=2 kg1/2-mol"1/2; %)=1.4 kg1/2-mol"1/2; * standard deviation of osmotic coefficients. MSA-NRTL model. The MSA-NRTL model has been developed as a semi-empirical model for electrolyte solutions. The model calculates the excess Gibbs energy of the electrolyte solution. The Gibbs energy is divided in two parts. The first long-range part corresponds to the electrostatic contribution of ion charges to the excess Gibbs energy. The second contribution is a short-range contribution corresponding to ali short-range forces existing between ions and solvent molecules. The molar excess Gibbs energy, gex, is written as follows: (8) g = gLr + g Therefore, the activity coefficient for a solvent (s) is written as: \nrs=\n^r+\nfs A (9) K. Nasirzadeh, R. Neueder: Measurements and Correlation of Osmotic Coefficients and Evaluation… Acta Chim. Slov. 2004, 51, 117-126. 123 Kunz et al. have focused on the so-called RPM-MSA (restricted primitive model-mean spherical appimimation) and have used it in plače of the PDH (Pitzer-Debye-Hiickel) equation to calculate the long-range distribution to the activitv coefficient of species (i) in solution. The MSA which was originallv applied to RPM electrolvte solutions is known to account for electrostatic interactions between ions in a better way than the DH model and it has been found quite accurate enough to be of great practical abilitv. Kunz et al. have ignored the electro-neutralitv assumption and have made two modifications in using NRTL which lead them to have four different adjustable parameters, that is, Tam, Tcm, Vc,ac and Va,ca- A combination of ali adopted expressions that have been given by Kunz et al. gave us the following equation for MSA-NRTL model. NaCH3COO>LiCH3COO, which indicates increasing ion–ion interactions from KCH3COO to LiCH3COO. There is, indeed, evidence for a higher ion pairing of K. Nasirzadeh, R. Neueder: Measurements and Correlation of Osmotic Coefficients and Evaluation… Acta Chim. Slov. 2004, 51, 117-126. 125 21 LiCH3COO compared to NaCH3COO in methanol from conductometric studies.21 For instance, Barthel and Neueder21 have reported association constants of 90.6 and 23.2 for LiCH3COO and NaCH3COO, respectively. This result indicates strong ion–ion interaction in LiCH3COO + methanol solutions in compared to NaCH3COO in methanol. There aren’t any conductance data for KCH3COO in literature. 1.05 0.95 0.9 0.85 0.8 0.75 0.7 0.5 1 1.5 1 molality / mol.kg 2.5 Figure 2. Experimental osmotic coefficients of KCH3COO, NaCH3COO and LiCH3COO in methanol solutions at 25 °C. Lines were generated using the Pitzer and MSA-NRTL models. Conclusions Experimental osmotic coefficient measurements have been reported for KCH3COO and NaCHaCOO in methanol solutions by an improved isopiestic method at 25 °C. Experimental data of the investigated systems are satisfactorily correlated using the Pitzer-Mayorga and MSA-NRTL models. Model parameters are obtained and used to calculate vapor pressure for methanol salt systems. The models of MSA-NRTL and Pitzer-Mayorga have been shown to correlate the experimental osmotic coefficient data with very good accuracy. For the Pitzer and Mayorga model, data analysis shows that the values ? (1)=2.0, ?(2)=IA, b=3.2 based on the best representation of some lithium salts in methanol, also give a good overall results 1 K. Nasirzadeh, R. Neueder: Measurements and Correlation of Osmotic Coefficients and Evaluation… 126 Acta Chim. Slov. 2004, 51, 117-126. for KCH3COO and NaCHaCOO in methanol solutions. The fit accuracy obtained with the MSA-NRTL model is the same as that obtained with the Pitzer model. The main advantage of the MSA-NRTL model is the relative physical significance of the parameters, since the