J. CHEM. SOC. FARADAY TRANS., 1994, 90(18), 2683-2689 Thermodynamic Study on the Transfer of the Tin@), Lead(l1) and Alkaline-earth-metal Ions from Water to Methanol, Dimethyl Sulfoxide, Acetonitrile, Pyridine and N,N-Dimethylthioformamide Monika Chaudhry, Yoshiaki Kinjot and lngmar Persson* Department of Chemistry, Swedish University of Agricultural Sciences, P.O.Box 7015,S-75007 Uppsala, Sweden The thermodynamic functions for the transfer reactions of the tin(ri), lead@) and alkaline-earth-metal ions from water to methanol, acetonitrile, dimethyl sulfoxide and pyridine, and of the tin(ir) and lead(ii) ions to N,N-dimethylthioformamide, are reported. The Gibbs energies of transfer, AtG", have been calculated from the standard electrode potentials of the Sn2+/Sn(s) and Pb2+/Pb(s) couples, which have been determined poten- tiometrically with the Ag(s)/Ag+ electrode as reference in the title solvents, and the Gibbs energies of transfer of the silver ion.The Gibbs energies of transfer of the alkaline-earth-metal ions were calculated from electrode or halfwave potentials of the M2+/M(am) couples in the title solvents and the difference in standard electrode potential between the M2+/M(am) and M2'/M(s) couples reported in the literature. The enthalpies of transfer, A,H*, have been obtained from calorimetrically determined enthalpies of solution of the anhydrous metal tri- fluoromethylsulfonates. The entropies of transfer, A$, have been calculated from the experimentally deter- mined AtG" and AtHe values. All measurements have been carried out at 25°C.The extrathermodynamic tetraphenylarsonium tetraphenylborate (TATB) assumption has been applied in order to calculate the contribu- tions from the single ions. The tin(it) and lead@) ions are solvated more strongly in dimethyl sulfoxide and N,N-dimethylthioformamide than in water, while they are solvated more weakly in methanol and acetonitrile. The small tin(ii) ion is solvated more weakly and the larger lead(ti) ion is solvated more strongly in pyridine than in water. The alkaline-earth-metal ions are solvated more strongly in dimethyl sulfoxide than in water, while meth- anol and the nitrogen donor solvents acetonitrile and pyridine solvate these ions more weakly than water. The enthalpies of transfer for these ions to the solvents studied are exothermic except for the tin(ii), calcium and strontium ions to acetonitrile.The entropies of transfer to all solvents are markedly negative except for N,N-dimethylthioformamide where the TAtS" values are close to zero. The difference in solvation of an ion between two solvents is mainly dependent on the bonding character of the ion-solvate bonds and on the way in which the ion affects the solvent bulk. The transfer thermodynamic functions for the tin(@, lead(@ and alkaline-earth-metal ions from water to methanol, ace- tonitrile, dimethyl sulfoxide and pyridine and for the tin@) and lead@) ions to N,N-dimethylthioformamide, are reported in this paper as part of a series of transfer thermodynamic studies of single cations and anions.'-5 The extra-thermodynamic tetraphenylarsonium tetraphenylborate (TATB) assumption, which states that the AS(C6H5)4+ and B(C6H5),- ions are equally solvated in every solvent,6-8 has been applied in this study in order to calculate the contribu- tions from the individual ions. The choice of the TATB assumption has been discussed previously.1,3 The solvation of metal ions depends mainly on the bonding characteristics of the solvent and the intermolecular forces in the solvent b~lk.~.~ Water and methanol are protic oxygen donor solvents which have mainly electrostatic inter- actions with metal ions, thus they solvate hard electron-pair acceptors well. Both hydrogens in the water molecule can participate in hydrogen bonding while only the hydrogen on the oxygen atom in methanol is able to form hydrogen bonds.This suggests that the aqueous bulk structure is sub- stantially more rigid and well ordered than the methanol one. Acetonitrile is an aprotic nitrogen donor solvent with weak and irregular solvating properties. Acetonitrile solvates the monovalent coinage metal ions well, 1-3*9* O while the copper(@, the alkali and the divalent d" metal ions are very weakly solvated in acet~nitrile.~-' Acetonitrile forms dimers and higher aggregates and the bulk is therefore fairly well f Present address : Chemical Laboratory, College of Education, University of the Ryukyus, Okinawa 903-1, Japan. ordered.' Dimethyl sulfoxide is an aprotic solvent with the possibility to coordinate either through its oxygen or sulfur atom.Coordination through the sulfur atom takes place only to the most soft electron-pair acceptors, e.g. palladium(I1) and platinum(n).' Dimethyl sulfoxide, as an oxygen donor, solvates both hard and soft metal ions ~e11,~-'*'~9'and is a markedly stronger electron-pair donor, D, = 27, than water and methanol with D, values of 18 and 17, respectively.'6*1 Pyridine is an aprotic solvent coordi- nating through its nitrogen atom. Pyridine has strong and typically soft electron-pair donor properties ; the D, value of pyridine is 38.16,17 Pyridine has no well ordered bulk structure18 as the intermolecular forces in pyridine are quite weak. The low relative permittivity of pyridine, E, = 12.3, indicate that long-range coulombic interactions are not effec- tively reduced in pyridine, and that the dipoles of the pyri- dine molecules are ordered around the ions in order to reduce the electric field as effectively as po~sible.~N,N-dimethylthioformamide is a sulfur donor solvent with typi- cally soft electron-pair donor properties, Ds = 52.16p1 N,N-dimethylthioformamide has a high relative permittivity, E, = 47.5,3and many salts are therefore very soluble and completely dissociated.A recent LAXS study has indicated intermolecular sulfur-sulfur interactions in the N,N-dimethylthioformamide bulk at about 3.1 Pyridine and N,N-dimethylthioformamide solvate soft electron-pair accep- tors strongly while their ability to solvate hard ones is poor.3,4 Gritzner et al.have previously reported Gibbs energies of transfer for the lead(I1) ion from acetonitrile to many solvents including the soft electron-pair donor solvents pyridine, J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 tetrahydrothiophene and N,N-dimethylthiof~rmamide.~~-~~The standard electrode and halfwave potentials of the They used the bis(biphenyl)chromium(O)/(r) (BCr) extra-thermodynamic assumption2' to calculate the contribution from the individual ions. The hydration thermodynamics of the alkaline-earth-metal ions have been studied extensively.26-28 The divalent alkaline-earth-metal ions are significantly more weakly hydrated than the divalent transition metal ions. This is because the alkaline-earth-metal ions are larger and the elec- trostatic interactions are therefore weaker, and the ability of the alkaline-earth-metal ions, with filled electron shells, to form covalent interactions is weaker than that of the tran- sition metal ions.This suggests that the alkaline-earth-metal ions are solvated strongly by solvents able to form strong electrostatic interactions, i.e. typical hard donor solvents, while they are expected to be solvated weakly by typical soft donor solvents. The low solubility of the alkaline-earth-metal trifluoromethylsulfonates, most probably due to weak solva- tion, has prevented us from performing calorimetric studies in N,N-dimethylthioformamide. The amount of thermodynamic data available for the alkaline-earth-metal ions in non-aqueous solvents is limited14,''because electrochemical studies on the alkaline- earth-metal couples are difficult to perform.Gibbs energies of transfer of the barium ion have previously been reported from acetonitrile to a large number of solvents using polaro- graphic methods and the BCr extrathermodynamic assump- and enthalpies of transfer of the barium ion using TATB assumption are reported from water to methanol, dimethyl sulfoxide and a~etonitrile.~'.~' Various techniques have been developed to prepare alkaline-earth-metal amalgams as they cannot be prepared by mixing the metal^.^^^^^ The magnesium electrode is known to behave irregularly in aprotic solvent^,^'-^^ which complicates direct measurements.The basic problem with most of the alkaline-earth-metal amalgams is rapid corrosion and passi- vation both in water and non-aqueous solvents.37 The elec- trode and halfwave potentials for the magnesium, calcium and strontium couples in methanol, acetonitrile, dimethyl sul- foxide, pyridine and liquid ammonia, referred to the saturat- ed calomel electrode (SCE) or the silver electrode, have been determined mainly by means of polarographic methods.29v31v38-42 Several authors have stressed the impor- tance of using acetonitrile as reference solvent rather than water as more well defined polarographic waves are obtained. In the case of water, the potentials measured by polaro- graphic methods tend to decrease with time.43 All electrode potentials reported in this study have been recalculated through the Gibbs energies of transfer of the silver ion so they are referred to the NHE in water.Halfwave potentials from polarographic measurements were used as they are a function of the standard potential of M2+/M(am) couples, the solubility of the metal in mercury and its Gibbs energy of amalgamati~n.~~ As E" of a metal amalgam electrode in a solution containing the correspond- ing metal ions and the polarographic halfwave potentials El,, is given by E" = -(RT/nF)ln([(DR/Do)'/'(fo/fR)],where D, and Do are the diffusion coefficients of the reduced and oxidized forms, respectively, and fR and fo are the activity coefficients of the reduced and oxidized forms.45 As the ratios of the diffusion and activity coefficients in the equation above are nearly unity in almost all cases, is usually a very good approximation to E" for a reversible couple.45 The E,/, values were taken to be equal to E", and the Gibbs energes of transfer were calculated from these values. The Gibbs ener- gies of transfer of the tin(@ and lead@) ions have been obtained from potentiometrically determined standard elec- trode potentials of the Sn2 +/Sn(am) and Pb2 +/Pb(am) couples in the studied solvents.alkaline-earth-metal couples are indeed very difficult to measure and the accuracy is sometimes low. Consequently, the accuracy of the Gibbs energies of transfer of the alkaline- earth-metal ions is much lower than for the previously reported ions in this series.'-' The enthalpies of transfer, AtHe, for the metal ions from water to methanol, acetonitrile, dimethyl sulfoxide, pyridine and N,N-dimethylthioformamide were obtained from calori- metrically determined enthalpies of soluion of the anhydrous metal trifluoromethylsulfonates. The entropies of transfer have been calculated from the experimentally determined AtG" and AtH* values.All measurements have been per- formed at 25 "C. Calculations The basis for the TATB assumption is given elsewhere., The Gibbs energies of transfer are calculated from standard elec- trode potentials, and the enthalpies of transfer are calculated from enthalpies of solution, as described elsewhere.' Experimental Materials Solvents Methanol was distilled over calcium hydride.' Acetonitrile (Fluka, analytical grade) and dimethyl sulfoxide (Merck) were purified as described else~here.~~,~~Pyridine (Riedel-deHaen, analytical grade) was kept water-free by storage over molecular sieves of 3 8, pore size, and it was used without further purification.N,N-dimethylthioformamide was prepared by reacting N,N-dimethylformamide (Fluka) and phosphorus pentasul- fide (Merck) in benzene as described previou~ly.~?~~ Preparation of Salts Lead@) trifluoromethylsulfonate was prepared as described el~ewhere~~~~~and tin@) trifluoromethylsulfonate was pre- pared in a similar way. The salts were stored in a desiccator under reduced pressure at 120"C. Elemental analysis of Sn(CF,SO,), gave: Found: Sn, 27.5%; C, 5.7%; S, 15.0%; Calc.: Sn, 28.5%; C, 5.7%; S, 15.4%. Elemental analysis of Pb(CF,SO,), gave: Found: Pb, 40.4%; C, 4.8%; S, 12.5%; Calc.: Pb, 41.0%; C, 4.8%; S, 12.7%.Barium trifluoromethylsulfonate was prepared as described elsewhere,32 and magnesium, calcium and strontium tri-fluoromethylsulfonate were prepared according to the same procedure. The metal content of the salts was checked by EDTA titration.'' The prepared salts were stored in a desic- cator under reduced pressure at 120 "C. Amalgams Tin amalgam containing 0.6 wt.% Sn and lead amalgam con- taining 1.476 wt.% Pb were prepared as described else- here.'^.^^ The tin and lead amalgams were washed with dilute perchloric acid and water to remove impurities.The amalgams were stored under dry nitrogen. Freshly prepared amalgams were used for each experiment as the surface of these amalgams is easily oxidized. Potentiometric Measurements The apparatus and procedure used have been described pre- viou~ly.~Tetrabutylammonium perchlorate, (C,H,),NCIO, , 0.1 mol dm-', was used as ionic medium. The emfs of the cells Ag(s) I Ag+ (10 mmol dm-,) 11 M2+ (1-30 mmol dm-3) I M(am), M = Sn and Pb, were determined in all solvents. The J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 w M AN DMSO Py DMTF \0.8 \ \ \ \0.6 0,4 > 0.2 /0 -0.2 -0.4 -0.6 -0.8 Fig. 1 Diagram of standard electrode potentials of the couples Sn2+/Sn(s)(-----) and PbZ+/Pb(s) (-) in water (W), methanol (M), acetonitrile (AN), dimethyl sulfoxide (DMSO), pyridine (Py) and N,N-dimethylthioformamide(DMTF) at 25 "C.The standard elec- trode potentials for the Ag+/Ag(s) couple (---) are given for com- parison. resistance of the experimental cell was less than 1 MQ. The emfs were measured by a digital voltmeter, Keithley 197, autoranging microvolt DMM, with an internal resistance of 2 GQ. The difference in potential between Pb(s) and Pb(am), 5.7 mV,s4 was corrected for, while there is no potential differ- ence between Sn(s) and Sn(am). The electrodes obeyed Nernst's law within 0.5 mV. The experiments were performed at 25.0 "C by passing thermostatted water through the outer jackets of Ingold vessels. Calorimetric Measurements The enthalpies of solution, AsolHe, were measured with an ampoule calorimeter described previously. s,56 A gold vessel containing 80 ml of solvent without supporting electrolyte was used in the experiments.Two or three ampoules were smashed without changing the solvent. The salt concentra- tion never exceeded 1.5 mmol dm-3. The ampoules were sealed off with an oxygen-propane flame and they were cooled during the sealing to avoid decomposition of the salt. Six experiments in agreement were performed for each salt and solvent. The measurements were performed at 25.000 & 0.002 "C. Results The electrode potentials of the M2'/M(am), M = Sn and Pb, couples have been determined experimentally vs. the reference Ag+/Ag(s) couple (Fig.1). By combining the expe$mental electrode potentials vs. the silver couple and theAtG values of the silver ion, all standard electrode potentials of these couples have been calculated relative to the normal hydrogen electrode in water, see Table 1. The Gibbs energies of transfer have then been calculated from these standard electrode potentials, see Table 2; the AtG" value for the lead@) ion to liquid ammonia is included for comparison. The electrode potentials of the alkaline-earth-metal couples, M2+/M(am), taken from the literature, have been used to calculate the electrode potentials of the M2+/M(s) couples by adding the difference in potential between the M2+/M(am) and M2'/M(s) couples which is 0.400 V, 0.872 V, 1.10 V and 1.34 V, for magnesium,60 calcium,38 strontiums4 and barium,54 respectively, see Table 3.The Gibbs energies of transfer calculated from the standard elec- trode potentials in water and the solvent under study are summarized in Table 2. The enthalpies of solution, Aso,He, of anhydrous tin(@, lead@), magnesium, calcium, strontium and barium tri-fluoromethylsulfonate in water, methanol, dimethyl sulfoxide, acetonitrile, pyridine and N,N-dimethylthioformamide and the enthalpies of transfer of the trifluoromethylsulfonate ion from water to the other solvents3 are given in Table 4.The complete transfer thermodynamics for the tin(rI), lead@) and alkaline-earth-metal ions are summarized in Table 2 and Fig. 2. The Gibbs energies of transfer for the lead@) and barium ions have been reported previously by Gritzner et ~1.~~9~ who applied the BCr assumption to polarographic data. These data are in some cases in good agreement with the values obtained in this study, using the TATB assumption, but in other instances large deviations are observed, see Table 5.These large differences do not seem to be caused by the extrathermodynamic assumptions used.3 Discussion The transfer thermodynamic functions are affected by several processes taking place at the solvation of an ion. These pro- cesses have been discussed previously and include as main Table 1 (a) Experimentally determined cell potentials (in V), where the silver electrode has been used as the reference cell, in methanol (M), acetonitrile (AN), dimethyl sulfoxide (DMSO), pyridine (Py), N,N-dimethylthioformamide (DMTF) at 25 "C and in 0.1 mol dm- tetra-butylammonium perchlorate ionic medium, the given values have been extrapolated to 1 mol dmP3 concentrations of all species in the cell reaction; (b) electrode potentials based on the Gibbs energies of transfer of the silver ion, E"/V,of the silver, lead@) and tin@) couples in water (W), M, AN, Py and DMTF at 25 "C (a) cell M AN DMSO Py DMTF -0.8960 -0.3975 -0.7626 -0.1950 -0.2615 -0.8870 -0.4697 -0.7032 -0.4401 -0.3702 (b) system W M AN DMSO PY DMTF Ag +/Ag(s)"Sn2+/Sn(s) Pb2 +/Pb(s) +0.7997 -0.1375 -0.1205 +0.876 -0.020 -0.017 +0.568 +0.171 +0.093 +0.445 -0.318 -0.264 +0.190 -0.005 -0.256 -0.224 -0.486 -0.600 Ref.3. 2686 J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 Table 2 Transfer thermodynamics for the magnesium, calcium, strontium, barium, lead@) and tin(@ ions from water to methanol (M), acetoni- trile (AN), dimethyl sulfoxide (DMSO), pyridine (Py) N,N-dimethylthioformamide(DMTF) and liquid ammonia mH,(l)] ~ ~~ M AN DMSO A,G~ A,H~ TA$ A,G~ A,Hg TA,~ A,G~ A,H~ TA$ +Sn2 +22.8 -15.6 -38.4 +59.6 +5.1 -54.5 -34.7 -83.4 -48.7 Pb2+ +20.1 -15.9 -36.0 +41.3 -5.2 -46.5 -27.6 -85.3 -57.7 Mg2+ -43.4 +llO -13 +Ca2 +36 -52.9 -89 +78 + 18.7 -59 -69 +Sr2 -53.8 +52 +7.9 -44 -14 -85.7 -72 BaZ+ + 18.4 -59.2 -77.6 +37 -6.2 -48 -25.1 -82.8 -57.5 ~ ~~~~ +Sn2 +25.5 -109.6 -135.6 -67.2 -75.4 -8.2 Pb2+ -26.1 -75.4 -49.3 -92.4 -85.5 +6.9 -114" -162' -48 Mg2+ +Ca2 -21.3 -58" -1 13' -46 +Sr2 -18.1 -79" -1 16' -37 Ba2+ +71 -26.0 -97 -58" -113b -55 ~ ~ ~ ~~~~ " Calculated from the standard electrode potentials reported in ref.57, and the Gibbs energy of transfer for the proton from water to liquid ammonia, -96 kJ mol-', ref. 14. Average value from ref. 58 (based on the enthalpy of formation of the ammoniated electron), and ref. 59 (based on the enthalpy of transfer of the sodium ion). contributions the bonding characteristics of the solvate bonds the endergonic Gibbs energies of transfer, see Table 2. The and the effect of the solvent bulk struct~re.~~~ enthalpies of transfer for these ions are exothermic, and the entropies of transfer are negative, which is especially pro- Methanol nounced for the alkaline-earth-metal ions, see Table 2. This The tin(II), lead(I1) and alkaline-earth-metal ions are all sol- transfer thermodynamic pattern with positive At G" and nega- vated more weakly in methanol than in water as shown by tive &He and T&Sg Values iS also found for the previously Table 3 Standard electrode or halfwave potentials based on the Gibbs energies of transfer of the silver ions, E"/V,of the magnesium, calcium, strontium, barium couples in water (W), methanol (M), acetonitrile (AN), dimethyl sulfoxide (DMSO), pyridine (Py) and liquid ammonia NH,(l) at 25 "C us.NHE in aqueous solution system W M AN DMSO Py NH 30)" Mg2+/Mds) -2.372' -1.W -2.44' -2.74 Ca2 +/Ca(s) -2.868' -2.68/ -2.46' -2.93* -3.17 Sr2 +/Sr(s) -2.89' -2.62' -2.96' -3.3 Ba2 +/Ba(s) -2.912b -2.817' -2.62; -2.72' -3.042' -2.54h -3.2 " Ref.42, based on the difference in standard electrode potential of the hydrogen electrode between water and liquid ammonia being 1.00 V; A,Ge = -96 kJ mol-', ref. 14. Ref. 38, 0.006-0.10 mol dm-, CaCl,. 'Ref. 40, 0.1 mol dm-3 (C,H9),NC10,. Ref. 41, 0.05 rnol dm-, (C,H,),NCF,SO, . Ref. 37,O-0.06 mol dm-3 CaCI, . Ref. 39, ionic medium not reported. Ref. 31,O.l mol dm-3 (C2H5),NC10,. Ref. 29, 0.1 mol dmP3 (C,H,)NClO, . Table 4 Enthalpies of solution, A,,,He/kJ mol -of magnesium, calcium, strontium, barium, lead(I1) and tin@) trifluoromethylsulfonate in water (W), methanol (M), acetonitrile (AN), pyridine (Py) and N,N-dimethylthioformamide(DMTF), the enthalpies of transfer, A,H*/kJ mol-', for the trifluoromethylsulfonate ion from water to other solvents are given as they are used to calculate the enthalpies of transfer for the metal ions, no supporting electrolyte has been used W M AN DMSO Py DMTF +Sn2 -53.1 f1.6 -58.7 1.0 -43.8 f 0.5 -132.9 f1.1 -171.3 f3.3 -106.9 f 1.4 Pb2+ -29.6 & 0.9 -35.5 f1.4 -30.9 f0.9 -111.3 k3.0 -113.6 f 0.8 -93.5 f1.2 Mg2+ -53.6 f 0.4 -87.0 f0.1 ins." ins." ins." ins." Ca2+ -33.6 f0.4 -76.5 f0.8 -9.7 f0.6 -111.3 f1.5 -63.5 f0.6 ins." +Sr2 -22.4 f0.6 -66.2 f1.4 -10.3 f0.3 -104.5 _+ 0.3 -49.1 f1.4 ins." Ba2+ -8.1 f0.9 ins." -10.1 & 0.3 -87.3 f0.3 -42.7 f0.4 ins." CF,SO,-+5.0 +2.1 + 1.8 -4.3 + 10.8 " The compound is not soluble enough for calorimetric measurements. Table 5 Comparison of the AIGe values of the lead@) and barium ions from acetonitrile (AN) to methanol (M), dimethyl sulfoxide (DMSO), pyridine (Py) and N,N-dimethylthioformamide (DMTF) obtained by the TATB and BCr extrathermodynamic assumptions M DMSO Py DMTF TATB BCr TATB BCr TATB BCr TATB BCr Pb2+ -21.2 -42.3" -68.9 -97.9" -67.4 -67.4' -133.7 -77.8" Ba2+ -19 -34.2' -62 -92.1" +34 -29.5' -36.5' " Ref.20. 'Ref. 21. 'Ref. 29. 2687J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 120 more order when the ion leaves than the methanol bulk loses 100 at the introduction of the same ion. This is expected as the aqueous bulk is more ordered than the methanol one. It can 80 be assumed that the energy in the ion-water and ion-methanol bonds is of the same order of magnitude as the bonding characteristics of water and methanol are similar.The weaker solvation of metal ions in methanol than in water is thus more an effect of solvent bulk perturbation than dif- ference in strength of the ion-solvent bond. Acetonitrile-".(.il20 0 The tin@), lead@) and alkaline-earth-metal ions are much -20 more weakly solvated by acetonitrile than by water. The -40 small and hard magnesium ion is solvated very poorly by acetonitrile, A,G" = +110 kJ mol- ',but the endergonicity -60 -80100-1decreases with increasing size and polarizability of the alkaline-earth-metal ions, see Table 2. The enthalpies of transfer of the magnesium ion could not be determined due to the low solubility of the anhydrous magnesium tri-fluoromethylsulfonate and perchlorate salts.The enthalpies of transfer of the other alkaline-earth-metal ions and the tin(I1) and lead(@ ions change from endothermic to exother- mic with increasing size of the ion, see Table 2. The entropies of transfer of the studied divalent metal ions to acetonitrile are slightly more negative than the corresponding values for the copper(II), zinc, cadmium and mercury(I1) ion^.^.^ The weak solvation of the magnesium ion in acetonitrile is also reflected in the Mg-0 and Mg-N bond distances in the hydrate and the acetonitrile solvate. The Mg-0 bond 20 0 -20 -40 -60 -80 -1 00 -1 20 -1 40 60 40 20 0 Fig. 2 The changes of Gibbs energy, -AtGe, (black), enthalpy, -AtHe, (white) and entropy, TAtSe, (hatched) in kJ mol-' for the transfer of the tin(r1) and lead@) ions from water to (a)methanol, (b) acetonitrile, (c) dimethyl sulfoxide, (d)pyridine and (e)N,N-dimethyl-thioformamide at 25 "C studied metal ions in The entropy of transfer term indicates that the difference in the effect of the methanol and aqueous bulks at the introduction of alkaline-earth- metal ions is about the same as for the other divalent metal ions studied, while this difference seems to be smaller for the tin@) and lead@) At the introduction of an ion into a solution solvent molecules are released from the solvent bulk structure for the solvation, and the internal bulk structure is affected when the solvated ions enter.These processes are endothermic and the absolute values increase with increasing strength of the intermolecular forces in the solvent. The nega- tive entropies of transfer show that the aqueous bulk gains distance in the hydrated magnesium ion, Mg(OH2)62+, is 2.11 A,61*62and the average Mg-N bond distance in the Mg(NCCH&,+ ion is 2.15 A.63 The M-N bond distance in the acetonitrile solvate is normally shorter than the M-0 bond distance in the corresponding h~drate.~~-~~ The only case where the M-0 and M--N bond distances are equal is the COPP~X(II) ion.64 The copper(r1) ion is substantially more weakly solvated in acetonitrile than in water, A,G" = +50 kJ mol-'.' The observation of longer Mg-NCCH, than Mg-OH, bond distances strongly supports the very weak solvation of the magnesium ion in acetonitrile.Dimethyl Sulfoxide As expected, dimethyl sulfoxide solvates, the tin(n), lead(I1) and alkaline-earth-metal ions well. The Gibbs energies of transfer of the magnesium, calcium and strontium ions are very similar, -13 kJ mol-', while the AtGe values for the barium, tin(r1) and lead(@ ions are more negative, -25, -35 and -26 kJ mol-', respectively. The enthalpies of transfer of the tin@), lead(@ and alkaline-earth-metal ions are markedly exothermic, about -85 kJ mol-'. This means that the entropies of transfer are markedly negative, -50 to -70 kJ mol-'. These values are much more negative than those found for the zinc, cadmium, mercury(@ and copper(I1) ions (-14.5, -13.8, -14.5 and -6.3 kJ mol- ', re~pectively).~,~ One reason for the large entropies of transfer of the alkaline- earth-metal ions may be that the hydrated and dimethyl sul- foxide solvated strontium and barium ions, and maybe also the calcium ion, have different coordination numbers.The hydrated strontium and barium ions are eight-coordinated in a square antiprismatic fashion, while the dimethyl sulfoxide solvated strontium and barium ions are ~ctahedral.~~ The coordination number of the dimethyl sulfoxide solvated calcium ion is most probably six, while it has been difficult to establish if the hydrated calcium ion is six-or eight-~oordinated.~~The structure of the hydrated tin(r1) ion has been reported to have two short and two long Sn-0 bond distances at 2.3 and 2.8 A, re~pectively.~~.~~The structure of the dimethyl sulfoxide solvated tin(@ ion has a similar struc- ture to that of the hydrate with Sn-0 bond distances at 2.18 and 2.7 A.'" The hydrated and dimethyl sulfoxide solvated lead@) ions are octahedral in solution.70 Pyridine The At G" value to pyridine is positive for the tin@) ion, while it is negative for the lead@) ion.The enthalpies of transfer are very exothermic for both ions, and the entropy of transfer is very negative for the tin@) ion, while the TAtS" value for the lead@) ion is less negative than found for other divalent metal ion^^,^ and in the same order of magnitude as the mono- valent metal ions,3 see Table 2.This may be explained by the fact that the lead@) ion is large and its charge density is lower than for previously studied divalent metal ions. The pyridine-solvated lead@) ion is octahedral in pyridine solu- ti~n.~'The Gibbs energy of transfer of the barium ion is markedly positive, showing that the barium ion is, as expected, weakly solvated in pyridine. The enthalpies of transfer of the alkaline-earth-metal ions to pyridine are weakly exothermic, see Table 2. This indicates that the AtG" values are largely positive also for the other alkaline-earth- metal ions. The obtained entropy of transfer of the barium ion is very close to the values found for the copper(I1) and divalent d" metal The pyridine-solvated strontium and barium ions are octahedral in pyridine ~olution.~' The very negative entropies of transfer to pyridine are most probably because (1) a change in coordination number has taken place, the pyridine-solvated strontium and barium ions are ~ctahedral,~' and (2) pyridine molecules in a fairly large volume around the ion are oriented towards the ions, as dis- cussed el~ewhere.~,~ A large-angle X-ray scattering study on a pyridine solution of lead@) trifluoromethylsulfonate shows two large peaks at 5.5 and 10 A in the radial distribution function (RDF),70 while these peaks are much less pro- nounced in the RDFs of pyridine solutions of uncharged and 18971*72monovalent species. The observation that pyridine solutions of ions with a fairly high charge density have a higher bulk order than those of neutral complexes evidences that large negative entropies of transfer indicate increasing bulk order. N,N-Dimeth ylthioformamide The Gibbs energies of transfer to N,N-dimethyl-thioformamide are markedly exergonic and the enthalpies of transfer are very exothermic showing that the tin@) and lead(I1) ions are strongly solvated by the soft electron-pair donor solvent N,N-dime thy1 thioformamide.The entropies of transfer are small with a positive value for the tin@) ion and a negative value for the lead@) ion. The Gibbs energy of transfer of the silver ion,3 -98 kJ mol-', and of the tin(I1) and lead(@ ions, -70 and -94 kJ mol-', respectively, are of the same order of magnitude, while the enthalpies of transfer differ significantly, -140, -75 and -85 kJ mol-', respec-tively. This indicates that the silver-N,N-dimethyl-thioformamide solvate bond has a more covalent character than the corresponding tin@)- and lead(@-N,N-dimethyl-thioformamide bonds.The entropies of transfer of the mono- valent metal ions are negative, about -40 kJ mol-', while the divalent ions studied previously are close to zero, see Table 2 and ref. 3 and 4. The N,N-dimethylthioformamide solvated tin@) ion is most probably a flattened square- pyramid with the tin(@ ion at the top; the Sn-S bond dis- tance is 2.67 A,'' while the N,N-dimethylthioformamide solvated lead@) ion is most probably five-coordinated, a J. CHEM. SOC. FARADAY TRANS., 1994, VOL.90 smaller coordinated number than with the solvates of the other solvents in this study. Liquid Ammonia The large -AtG" and -AtH" values, though not accurate, show that the lead@) and alkaline-earth-metal ions are strongly solvated by liquid ammonia. This shows that liquid ammonia solvates hard and borderline electron-pair accep- tors very well. The entropies of transfer are slightly negative as has been found previously for the mercury(i1) Conclusions These results show that the entropies of transfer of the studied divalent metal ions are very similar to values obtained for other divalent metal ions, see ref. 4 and 5 and Table 2. This supports the view that the entropy of transfer term reflects the effect on the solvent and aqueous bulk at the introduction of an ion and that ions with similar charge density and size affect the solvent bulk to about the same extent.The enthalpies of transfer give a strong indication of the bond character of the ion-solvent bond, where an increasing degree of covalency in the interaction results in an increasingly exothermic enthalpy of transfer. The difference in solvation of an ion between two solvents is thus mainly dependent on the bonding character of the ion-solvent bonds and the way in which the ion affects the solvent bulk. This study further supports the assertion that the entropies of transfer indicate perturbation of the solvent bulk at the intro- duction of a metal ion and that this perturbation is mainly dependent on the charge density of the ion.This study also supports the previous observation that the soft electron-pair donor solvents pyridine and N,N-dimethylthioformamidedis-criminate significantly between hard and soft electron-pair acceptors at solvation, see Table 2, while hard electron-pair donor solvents solvate more equally hard and soft electron- pair acceptors, see Table 6 in ref. 3 and Table 3 in ref. 4. The Swedish Natural Research Council is acknowledged for their financial support of this project. References 1 M. Johnsson and 1. Persson, Znorg. Chim. Acta, 1987, 127, 15. 2 M. Johnsson and I. Persson, Znorg. Chim. Acta, 1987, 127, 25. 3 H. D. Inerowicz, W. Li and I. Persson, J. Chem, SOC., Faraday Trans., 1994,90,2223, and references therein.4 M. Chaudhry, K. C. Dash, E. Kamienska-Piotrowicz, Y. Kinjo and I. Persson, J. Chem. SOC.,Faraday Trans., 1994,90,2235. 5 M. Chaudhry and I. Persson, J. Chem. SOC., Faraday Trans., 1994,90,2243. 6 E. Grunwald, G. Baughman and C. Kohnstam, J. Am. Chem. SOC.,1960, 82, 5801. 7 E. M. Arnett and D. R. McKelvey, J. Am. Chem. SOC.,1966, 88, 2598. 8 B. G. Cox and A. J. Parker, J. Am. Chem. SOC., 1973,95402. 9 M. Johnsson and I. Persson, Znorg. Chim. Acta, 1987, 127,43. 10 S. Ahrland and S.-I. Ishiguro, Znorg. Chim. Acta, 1988,142, 277. 11 J. Sadlej, Spectrochim. Acta, Part A, 1979,35, 681. 12 K. Nilsson and I. Persson, Acta Chem. Scand., Ser A, 1987, 41, 139. 13 J. Selbin, W. E. Bull and L. H. Holmes Jr., J. Znorg.Nucl. Chem., 1961, 16, 219. 14 Y. Marcus, Pure Appl. Chem., 1983, 55, 977, and references therein. 15 Y. Marcus, Pure Appl. Chem., 1985, 57, 1103, and references therein. 16 I. Persson, M. Sandstrom and P. L. Goggin, Inorg. Chim. Acta, 1987,129, 183. 17 M. Sandstrom, I. Persson and P. Persson, Acta Chem. Scand., 1990,40,653. J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 2689 18 19 A. Iverfeldt and I. Persson, Znorg. Chim. Acta, 1986,111, 171. I. Persson, M. Sandstrom, C. T. Stllhandske, C. M. V. 47 S. Ahrland and N.-0. Bjork, Acta Chem Scand., Ser A, 1974,28, 823. 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 Stdlhandske and E. Kamienska-Piotrowicz, unpublished results. G. Gritzner, J. Chem. SOC.,Faruday Trans. 1, 1988,84, 1047, and references therein.G. Gritzner and S. Sperker, J. Solution Chem., 1990,19,543. G. Gritzner, 2. Phys. Chem. N. F., 1988,158,99. G. Gritzner and S. Sperker, J. Solution Chem., 1990, 17, 1133. F.Horzenberger and G. Gritzner, J. Chem. SOC., Faruday Trans., 1992,88,695. G. Gritzner, Inorg. Chim. Acta, 1977,24,5. R. M. Noyes, J. Am. Chem. SOC.,1962,84,513. D. R. Rosseinsky, Chem. Reo., 1965,65,467. B. G. Cox and A. J. Parker, J. Am. Chem. SOC., 1973,95,6879. G. Gritzner and G. Kraml, Znorg. Chem. Acta, 1989,156,227. F. Horzenberger and G. Gritzner, J. Chem. SOC.,Faraday Trans., 1992,88,3013. G. R. Hedwig, D. A. Owensby and A. J. Parker, J. Am. Chem. SOC., 1975,97, 3888. G. R. Hedwig and A. J. Parker, J. Am. Chem. SOC., 1974,%, 6589. K. Brauer and H. Strehlow, 2.Phys. Chem., N. F., 1958,17,336. 0.R. Brown and R. McIntyre, Electrochim. Acta, 1984,29,995. W. E. Elliott, J. R. Huff, R. W. Adler and W. L. Towle, Pro-ceedings of 20th Annual Power Sources Conference, 1966,p. 67. J. Farrer, R. Keller and M. M. Nicholson, High Energy Battery Systems, Final Report, No DA-28-043 AM (3-01394E), 1963, Rocketdyne, North American Aviation Inc. J. N. Butler, J. Electroanal. Chem., 1968, 17, 309. 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 V. Gutmann, K. Danksagmiiller and 0. Duschek, Z. Phys. Chem., N. F., 1974,92,199.T.Fujinaya and I. Sakamoto, J. Electroanal. Chem., 1976,73,21. G. Gritzner, J. Electroanal. Chem., 1983,144,259. G. Schwarzenbach and H. Flaschka, Die Komplexometrische Titration, F. Enke, Stuttgart, 1965.J. Gouy, J. Phys., 1895,4,320. H. E. Thompson, J. Phys. Chem., 1935,39,655. Handbook of Chemistry and Physics, CRC Press, Inc. Boca Raton, 73rd edn., 1992-1993,pp. 8-19. S.Sunner and I. Wadso, Sci. Tools, 1966, 13, 1. L.Kullberg, Acta Chem Scand., Ser. A, 1974,28,979. W. E.Dasent, Inorganic Energetics, Cambridge University Press, Cambridge, 2nd edn., 1982,p. 165. W. L. Jolly, Prog. Inorg. Chem., 1959,1,235. N. M. Senozan, J. Inorg. Nucl. Chem., 1973,35,727. 0.R. Brown and R. McIntyre, Electrochim. Acta, 1984, 29, 995, and references therein. R. Caminiti, G. Licheri, G. Piccalugo and G. Pinna, J. Appl. Cryst., 1979,12, 34. R. Caminiti, G. Licheri, G. Piccalugo and G. Pinna, Chem. Phys. Lett., 1979,61,45. M. G. B.Drew, P. P. K. Claire and G. R. Willey, J. Chem. SOC., Dalton Trans., 1988,215. I. Persson, J. E.Penner-Hahn and K. 0.Hodgson, Znorg. Chem., 1993,32,2497. K. Nilsson and I. Persson, Acta Chem. Scand., Ser. A, 1987, 41, 38 39 40 41 42 43 44 45 46 T. Mussini and A. Pagella, Chem. Eng. Data, 1971,16,49. E.Constantinescu, Rev. Roum. Chim., 1972,17,1819. T. Fujinaga and I. Sakamoto, J. Electroanal. Chem., 1976, 73, 235. V. Gutmann and G. Schober, 2. Anal. Chem., 1959,171,339. S. G. Bratsch and J. J. Lagowski, J. Solution Chem., 1987,16, 583. A. D.Covington and A. K. Covington, J. Chem. SOC.,Faraday Trans. I, 1975,71, 831. I. M. Kolthoff and J. F. Coetzee, J. Am. Chem. SOC., 1957, 79, 870. A. J. Bard and L. R. Faulkner, Electrochemical Methods, Wiley, New York, 1980,p. 160. S. Ahrland, K. Nilsson and B. Tagesson, Acta Chem. Scand., Ser A, 1983,37, 193. 66 67 68 69 70 71 72 139. T. Yamaguchi, G.Johansson, B. Holmberg, M. Maeda and H. Ohtaki, Acta Chem. Scand., Ser. A, 1984,38,437. M. Sandstrom, I. Persson, H. Yokoyama and M. Chaudhry, Znorg. Chem., submitted. G. Johansson and H. Ohtaki, Acta Chem. Scand., 1973,27,643. T. Yamaguchi, 0.Lindqvist, T. Claeson and J. B. Boyce, Chem. Phys. Lett., 1982, 93, 528. I. Persson, M. Chaudhry and M. Sandstrom, to be published. M. Sandstrom and I. Persson, J. Chem. SOC., Dalton Trans., 1985, 1597. F. HultCn and I. Persson, Acta Chem. Scand, Ser. A, 1987,41,87. Paper 3/06676B ; Received 8thNovember, 1993