J. CHEM. SOC. FARADAY TRANS., 1994, 90(4), 575-577 Limiting Partial Molar Volumes of Electrolytes in Dimethylformamide-Water Mixtures at 298.15 K Eugenio Garcia-Pafieda, Cayetano Yanes, Juan J. Calvente and Alfred0 Maestre* Departamento de Quimica Fisica , Facultad de Quimica , Universidad de Sevilla , 4 1012Sevilla, Spain Partial molar volumes at infinite dilution, V:, are reported for some 1 :1 electrolytes in dimethylformamide (DMF)-water mixtures covering the whole fraction range at 298.15 K. The results obtained show that the behav- iour of V; both for alkali-metal halides and for hydrophobic electrolytes is dependent on the added DMF. In the water-rich region, the alkali-metal halides exhibit small variations of VT, but two extrema, a minimum and a maximum, in the case of the hydrophobic electrolytes, are observed.In the DMF-rich region, a drastic decrease of V: for all electrolytes is exhibited. The results are discussed on the basis of ion-solvent and solvent-solvent inte ractions. The properties of aqueous solutions containing amides have solvent, respectively. A compilation of p, po and 4, are avail- received considerable attention since they have the peptide able as a supplementary publication.? linkage elements''2 and they are used as model compounds The application of the Redlich-Meyer equation24 was not to obtain information on biochemical systems. In relation to possible owing to the lack of values of the theoretical limiting aqueous mixtures of N,N-dimet hylformamide (DMF), several slopes, S,, in these mixtures.However, 4" were found to vary studies have been made in order to elucidate the mechanism linearly with m1/2over the concentration range investigated of the interaction of DMF with wziter (relative permit- (0-0.37 mol kg-'). The limiting partial molar volume of the cryoscopic and calorimetric mea~urements,~ ti~ities,~,~ ultra-electrolytes, VF =$:,was obtained by least-squares fitting sonic velocities,6 volumes and heat capacities,' NMR,8-'' viscosities",' ' and enthalpies of dilutionI2) where strong hydrogen bonding to the carbonyl group of DMF produces associates of the DMF-(H20), type. On the other hand, studies of electrolytes in aqueous mixtures of DMF (NMR,13 enthalpies of solution, transfer Gibbs energies' 7-19 and viscosities20.2I) have also been published, but limiting partial molar volumes of electrolytes in these mixtures have been restricted to tetraalkylammonium bromides.2 1.22 In this paper we report partial molar volumes at infinite dilution of 1 :1 electrolytes in aqueous mixtures of DMF covering the whole mole fraction range in order to obtain a better insight into ion-solvent and solvent-solvent inter-actions in mixed aqueous solvents.Experimental DMF (Merck, stated purity 99.8%, maximum content H20 0.05%) was dried over a thermally activated 4A molecular sieve prior to use. Ph,PC1 (Janssen Chimica, G.R., stated purity 99Y0) and NaBPh, (Merck, G.R., stated purity 299.5%) were dried for three days at 343 K in a vacuum desiccator.LiC1, NaCl, KC1, NaBr, KBr and KI were reagent grade (Merck) and were used after drying them overnight in an oven at 393 K. All salts were kept in a vacuum desiccator prior to use. DMF and water were degassed prior to making up solutions by weight. Measurements of densities were made using the apparatus and procedures described previ~usly.~~ Densities have an uncertainty (95% confidence limits) of +7 x g ~m-~. Results and Discussion Apparent molar volumes, 4,/cm3 mol-were calculated from solution densities using the standard expression M2 Po-P$,=-+-P WPOP where M, is the molecular weight of the electrolyte, m its molality; p and po represent the density of solution and of the results to the Masson equation: where S; is the experimental slope. Table 1 shows values of VF together with their 95% confidence limits (in parentheses).As can be observed, the experimental results obey the additivity rule within & 1.8 cm3 mol- '. In Fig. 1 we have plotted trends of VF with the mixed solvent composition for alkali-metal halides. At low DMF composition, up to xDMF= 0.3, the influence of added DMF does not produce significant changes on VF, except for LiCl where the effect of DMF begins to be noteworthy from xDMFz 0.17. On the other hand, as the DMF content increases, the effect of DMF produces a drastic decrease in VT up to pure DMF, taking even negative values for LiCl in the DMF-rich region. It is known that interactions between alkali-metal ions and dipolar aprotic solvents occur at the negative pole of the DMF dipole while the positive end of the dipole is sterically hindered from interacting with halide ions.In principle, this fact makes interaction of alkali-metal ions with both water and DMF possible, but that of halide ions is restricted to water only. However, on the basis of the Jones-Dole coeffi- cients of alkali-metal halides in water-amide mixtures, Woldan" has noted that in the water-rich region both alkali- metal ions and halide ions are solvated selectively by water, irrespective of the amide used. However, there is little evi- dence from standard functions of transfer for the preferential solvation of alkali-metal halides in dipolar aprotic aqueous solvents. On the other hand, Gallardo-Jimenez and Lille~~~ have studied enthalpies of interaction of alkali-metal halides with DMF in water in terms of the enthalpic virial second coefficient, hsaltPDMF,which is a measure of the water-mediated salt-DMF interactions.Note that in the case of our electrolytes, hsaltPDMFis negative both in the anionic and cationic series and becomes more negative as the ion size increases, except for LiCl, which seems to indicate that from a t Supplementary publication no. SUP 56985 (29 pp.), deposited with the British Library. Details are available from the Editorial Office. J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 Table 1 Partial molar volumes at infinite dilution, Vr/cm3 mol- ',for some electrolytes in xDMF-(1-x)water mixtures at 298.15 K X LiCl NaCl KCl NaBr KBr KI Ph,PCl NaPh,B 0.0 16.91" 16.62" 26.87" 23.48" 33.73" 45.6" 310b 276.4b 0.0196 0.0417 0.0955 0.1241 0.1 678 0.2227 0.2699 17.4 (0.4) 16.6 (0.4) 16.9 (0.4) 14.9 (0.4) 13.2 (0.3) 15.6 (0.3) 17.4 (0.3) 16.7 (0.3) 16.9 (0.9) 16.9 (0.5) 27.9 (0.3) 26.6 (0.1) 27.3 (0.1) 28.1 (0.3) 27.0 (0.4) 24.3 (0.1) 23.8 (0.3) 24.2 (0.3) 24.6 (0.5) 23.1 (0.7) 34.9 (0.7) 33.2 (0.3) 33.8 (0.3) 34.7 (0.2) 34.5 (0.6) 45.1(0.7) 46.4 (0.3) 47.2 (0.2) 48.1 (0.8) 47.5 (1.0) 310.1 (0.7) 309.5 (0.8) 307.3 (0.2) 306.9 (0.5) 305.9 (0.4) 307.9 (0.4) 307.7 (0.8) 274.5 (0.4) 278.5 (0.6) 288.9 (0.5) 296.9 (0.3) 299.4 (0.3) 0.3003 0.34 15 0.4232 0.4965 0.6598 0.6705 0.8299 1.o 8.6 (0.3) 5.4 (0.9) 1.7 (0.5) -4.4' 14.5 (0.4) 11.3 (1.0) C-5.9' 23.2 (0.4) 21.0 (0.9) 13.0' C- 21.5 (0.5) 17.0 (0.4) 14.0 (1.0) 6.6' 31.3 (0.4) 27.2 (1.0) 25.2 (1.2) 14.1' 44.1 (1.0) 41.7 (1.1) 35.0 (0.9) 30.5' 306.2 (0.3) 304.0 (1.O) 302.1 (0.6) 301.2 (0.2) 297.3 (1.1) 292.6 (0.8) 282.6 (1.0) 297.9 (1.0) 295.6 (0.4) 294.6 (0.6) 288.3 (0.3) 282.9 (1.0) 278.0 (0.5) 280' " Ref.24. Ref. 25. The solubility is too low to obtain accurate V?.'Ref. 26. Ref. 27. thermochemical point of view a favourable interaction obtained ionic partial molar volumes at infinite dilution, Vy, between the ions and DMF occurs. This is not in accordance of alkali-metal ions and halide ions in pure DMF from the with the idea that halide ions cannot interact with DMF ultrasonic vibration potentials method.Note that V? values because of steric hindrance at the positive pole of the DMF in water and in DMF are very similar for the halide ions, dipole. From the above considerations we could assume that except for the Br- ion, but those corresponding to alkali- alkali-metal ions and halide ions interact with DMF through metal ions differ markedly. This means that the difference their respective surrounding water cospheres in a water-rich observed in VT between water and DMF for alkali-metal region. The observed trends for VT could be attributed to halides in pure DMF is chiefly a consequence of cation- modifications in DMF-water interactions made by the pres- DMF interactions. ence of the ions in the mixed aqueous solvent. After this Fig. 2 shows trends of VT with mole fraction of DMF for water-rich region the water around the ions is progressively Ph,PCl, NaBPh, and Bu;NBr," the latter being included being replaced by DMF in the mixture and less water for comparison. In contrast to alkali-metal halides, the three remains to solvate the ions.This fact makes the interactions hydrophobic electrolytes exhibit significant changes of VT in predominantly of cation-DMF type, because halide ions are the water-rich region, with a minimum and a maximum sterically hindered from interacting with DMF molecules whose positions seem to be dependent on the type of electro- when the DMF content is high and, therefore, Vy behaviour lyte. Thus, the maximum is located at approximately the is mainly determined by the electrostriction effect of the same mole fraction of DMF for Ph,PCl and NaBPh,, alkali-metal ions.In this sense, Zana and Yeager3' have whereas for BuiNBr it extends up to the intermediate com- position region. From this extrema region the effect of DMF "I A '* 270 t 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0XDMF Fig. 1 Variation of limiting partial molar volumes, V?, of electro- XDMF lytes with DMF composition for: KI (O),KBr (W), NaBr (A),KCl Fig. 2 Variation of limiting partial molar volumes, V?, with DMF (El), NaCl(A), LiCl(0) composition for: Ph,PCl (O),NaBPh, (A),BuaNBr (m) J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 effect with Ph,PCI and Bu2NBr electrolytes, and by the com- bined effects of the cation and the anion with NaBPh, elec-trolyte.I 550 -I I I I I produces a marked decrease of VF,similar to that for alkali- metal halides.The existence of a minimum and a maximum suggests that combined opposing effects between hydrophilic ion-solvent and hydrophobic ion-solvent interactions occur. Since the maximum is located at xDMFzz 0.27 for Ph,PCl and NaBPh, and considering that the Ph4P+ and BPh, ions have a similar size and comparable hydrophobic characteristics and, moreover, with negligible electrostriction effects due to their weak electric field, we could assign the behaviour of VT in the maximum as due to hydrophobic ion-solvent inter-actions. In contrast, the minima must predominantly be determined by the corresponding counter-ion-solvent inter-actions.In order to confirm that the assignments of both extrema are correct, we can cancel the respective contribu- tions of the C1- and Na' ions to VF of Ph,PCI and NaBPh, ,respectively, by considering the Ph,PBPh, electro-lyte, which can be achieved using the additivity rule in Table 1 according to the expression V?(Ph,PBPh,) = V,"(Ph,PCl) + V,"(NaBPh,) -V,"(NaCI) (3) In Fig. 3 we have plotted the dependence of VT on the mole fraction of DMF for Ph,PBPh,. The plot shows that the minimum has disappeared and only the maximum at xDMF z 0.27 prevails. On the other hand, the drastic decrease of Vr with increasing content of DMF for the three hydrophobic electrolytes (Fig. 2) has been attributed in the case of Bu:NBr22 to structural interactions between this and the DMF.Nevertheless, on the basis of the above considerations for alkali-metal halides, the behaviour of Vr in the DMF-rich region must be determined by the hydrophobic cation We wish to thank DGICYT of Spain for financial support (Proyecto no. PB91-0605). References 1 G. Somsen, Pure Appl. Chem., 1991.63, 1687. 2 G. R. Hedwig, T. H. Lilley and H. Linsdell, J. Chem. SOC., Faraday Trans., 1991,87, 2987, and references therein. 3 G. Douheret and M. Morenas, C. R. Acad. Sci., Ser. C, 1967, 2, 729. 4 R. Reynaud, C. R. Acad. Sci., Ser. C, 1968,266,489. 5 J. Bougard and R. Jadot, J. Chem. Thermodyn., 1975,7, 185. 6 F. Kawaizumi, M. Ohno and Y. Miyahara, Bull. Chem. SOC. Jpn., 1977, 50,2229. 7 C.De Visser, G. Perron and J. E. Desnoyers, J. Chem. Eng. Data, 1977, 22, 74. 8 Y. I. Mitchenko, V. A. Fenin and E. P. Krasnov, Russ. J. Chem., 1978,52, 163. 9 V. Zelano, 2.Phys. Chem. N. F., 1983,138,31. 10 C. M. Kinart, W. J. Kinart and L. Shulski, Pol. J. Chem., 1985, 59, 597. 11 C. Della Volpe, G. Guarino, R. Sartorio and V. Vitagliano, J. Chem. Eng. Data, 1986, 31, 37. 12 R. H. Wood and L. H. Hiltzik, J. Solution Chem., 1980,9,45. 13 A. Fratiello, R. E. Lee, D. P. Miller and V. M. Nishida, Mol. Phys., 1967, 13, 349. 14 S. Taniewska-Osinska and A. Piekarska, Bull. Acad. Pol. Sci., Sir. Sci. Chim., 1978, 26, 613. 15 L. Thakur and R. Prasad, Indian J. Chem., 1980, 19A, 520. 16 W. J. M. Heusvelsland, C. de Visser and G. Somsen, J.Chem. SOC., Faraday Trans. I, 1981,77, 1191. 17 S. P. Rudra, B. P. Chakravarty, K. Kundu and I. N. Basu-Mallick, Z. Phys. Chem. N. F., 1986,150,211. 18 E. A. Gomaa, Thermochim. Acta, 1989, 142, 19. 19 K. Das, K. Bose and K. Kundu, Electrochim. Acta, 1981,26,479. 20 B. N. Prasad, N. P. Singh and M. M. Singh, Indian J. Chem., 1976,14A, 332. 21 N. C. Dey, G. Kumar, B. K. Saikia and I. Haque, J. Solution Chem., 1985,14,49. 22 W. J. M. Heuvelsland and G. Somsen, J. Chem. Thermodyn., 1977,9, 231. 23 C. Yanes, P. Perez-Tejeda, E. Garcia-Paiieda and A. Maestre, J. Chem. SOC., Faraday Trans., 1992,88,223. 24 F. J. Millero, in Water and Aqueous Solutions. Structure, Ther- modynamics and Transport Processes, ed. E. A. Home, Wiley, New York, 1972, ch. 13. 25 C. Jolicoeur, P. R. Philip, G. Perron, P. A. Leduc and J. E. Des-noyers, Can. J. Chem., 1972,50,3167. 26 F. Kawaizumi and R. Zana, J. Phys. Chem., 1974,78, 1099. 27 M. R. J. Dack, K. J. Bird and A. J. Parker, Aust. J. Chem., 1975, 28, 955. 28 M. Woldan, Z. Phys. Chem. N. F., 1986, 150,201. 29 M. A. Gallardo-Jimenez and T. H. Lilley, J. Chem. SOC., Faraday Trans. I, 1989,85,2909. 30 R. Zana and E. B. Yeager, in Modern Aspects of Electrochem-istry, ed. J. O'M. Bockris, B. E. Conway and R. E. White, Plenum Press, New York, 1982, vol. 14, ch. 1. Paper 3/02039H; Received 8th April, 1993