6 Concentrated Electrolyte Solutions and Fused Salts Raman and Infrared Spectroscopy of By RONALD E. HESTER Department of Chemistry University of York York YO4 5AL THE TASK of reviewing progress made in any branch of chemical spectroscopy is now greatly simplified by the availability of the Chemical Society Specialist Periodical Reports on ‘Spectroscopic Properties of Inorganic and Organo-metallic Compounds,’ written by Professor Greenwood and his colleagues. In addition this review benefits from the thorough coverage of ‘Electrolyte Solutions’ made by Pethybridge and Prue’ in the Annual Reports for 1968. Together these publications have removed the need for a comprehensive report and for any extensive discussion of pre-1969 developments so that this reporter can present his personal view of recent progress and its significance.Advances in instrumentation and technique particularly concerned with laser Raman spectroscopy and with interferometric far4.r. spectroscopy have largely been responsible for the most significant developments in this area. The most detailed studies of interionic association and ion-solvation processes have been made with metal nitrates in water and in a number of non-aqueous solvents of high dielectric constant but many other salts have been subjected to vibrational spectroscopic investigation and polyelectrolyte systems also have yielded interest-ing results. It is perhaps appropriate though to begin with the traditional solvent, liquid water. Water.-The interpretation of the shape and temperature dependence of vibra-tional bands in the stretching region of liquid water has become a thoroughly controversial issue.Little disagreement remains over the form of the spectra themselves but it is still far from clear whether they are better explained by a continuum theory or by a two-state structural theory. In late 1968 Schiffer and Hornig3 presented a new and interesting interpretation of the i.r. spectrum of the liquid. In making comparisons with spectra from simple salt hydrates these ’ N. N. Greenwood J. W. Akitt W. Errington T. C. Gibb and B. P. Straughan ‘Spec-troscopic Properties of Inorganic and Organometallic Compounds,’ The Chemical Society vols. 1 and 2 1967 and 1968. A. D. Pethybridge and J. E. Prue ‘Electrolyte Solutions,’ Annual Reports ( A ) 1968,65, 129.J . Schiffer and D. F. Hornig J . Chem. Phys. 1968 49,4150 80 R. E. Hester authors noted an anomalously high intensity for the low-frequency component (ca. 3250 cm- ') of the broad band in the stretching region of the liquid water spectrum this band being assigned to a Fermi resonance-enhanced 2v2 (deforma-tion mode v2 x 1640 cm-' for liquid water at 20 "C) because of spectral similar-ities with hydrates. In addition they showed that the width at half-height of the uncoupled 0-H stretching band of liquid water (from dilute HDO in D20) is anomalously broad being ca. 250 cm- compared with ca. 6 cm- ' for hydrates. These anomalies were explained in terms of a Maxwellian distribution of col-lisional interactions between water molecules demonstrating that (a) most molecules in the liquid are highly distorted by collisional perturbations and (b) there is a broad distribution of distortion among the molecules in the liquid.These concepts led Schiffer and Hornig to propose a continuum distribution of v1 and v 3 modes of variously distorted molecules. Their theory provides a satisfactory interpretation not only of the water band shape and width but also of the variations induced by temperature changes. In direct conflict with this continuum interpretation of water is the explanation proposed by Walrafen4 for the vibrational band shapes. This author's most recent experimental contribution was the discovery of a pronounced asymmetry in the Raman band due to 0 - H stretching of HDO in 1 mole % solution in D20. A high frequency shoulder near 3630 cm- ' was reported separated from the contour maximum at ca.3430 cm- ' by an inflexion near 3600 cm- '. Resolu-tion of the band into two Gaussian components at ca. 3455 and 3628 cm- was achieved and Walrafen presented this as further evidence in favour of a model of water structure involving broken (or distorted) and unbroken (linear or near-linear) hydrogen bonds the higher frequency component being assigned to non-hydrogen-bonded water molecules and the lower to hydrogen-bonded water molecules. The various aspects of the two-state model versus the continuum model conflict have been reviewed by Schiffer,' who argued that even Walrafen's data on the variations with temperature of the Raman 0-H stretching band were best interpreted by the continuum theory.Walrafen6 answered this by seeming to demonstrate that there are several aspects of the vibrational spectrum of liquid water in clear contradiction to the continuum model. The assumption of a two-state model however appears to ignore differences between water mole-cules which are hydrogen-bonded to one two three or four neighbours and it leans heavily on Gaussian band contour curve-fitting procedures an approach which has been severely criticised by Perram.' Among the more significant of the many other attempts made recently to throw further light on this subject is that of Ford and Falk,* who made an i.r. absorption study of hydrogen bonding in water and ice. Their results were correlated with the intermolecular potential energies of both H 2 0 and DzO, G.E. Walrafen J . Chem. Phys. 1969 50 560. J. Schiffer J . Chem. Phys. 1969 50 566. G. Walrafen J . Chem. Phys. 1969 50 567. T. A. Ford and M. Falk Canad. J . Chem. 1968,46 3579. ' J . W. Perram J . Chem. Phys. 1968 49 4245 Spectroscopy of Concentrated Electrolyte Solutions and Fused Salts 81 both showing smooth and very broad distributions which appear to conform to a continuum model for the liquid structure. Another examination of the Raman bands from ca. 2 mole % HDO in liquid water (HzO and DzO) has been made by Wall.9 His analysis of the band shapes in terms of the intermolecular correla-tion function developed by Gordon" led him to conclude that the individual water molecules are only approximately independent scattering centres even the internal motions being tied to the structure of the liquid as a whole.Lattice-like motions of the liquid are very much involved in Wall's vibrational band-shape analysis. Stimulated Raman scattering experiments' ' have provided further evidence for hydrogen bonding in water and other i.r. studies have been reported concerned with the low-frequency librational vibrations,12 with the high-frequency near4.r. absorption characteristics " 9 l4 and with the weak 2130 cm- band a~signment.'~ Further work on the i.r. absorption of different forms of ice has also been reported.16 As mentioned earlier Raman and i.r. studies of crystalline hydrates have been used to provide information on the state of co-ordinated water particularly for comparison with liquid water.Hydrated divalent metal ions in combination with halide ions or oxyanions have been most studied.' 7-2 ' Several years ago Derjaguin and his co-workers at the Karpov Institute of Physical Chemistry in Moscow reported the synthesis of a completely new form of liquid water.22 It is only during the past year that this discovery has begun to stir up great interest in the West but already the properties of this new material have been excitedly described in the popular press and by radio news-reporters, and a writer to Nature23 has described it as possibly 'the most dangerous material on earth' and cautioned us to treat it as the 'most deadly virus'! 'Anomalous water' or 'polywater' as it also has been somewhat prematurely named has been prepared by the condensation of normal water vapour into freshly drawn fine quartz capillary tubes contained in a chamber held at a slightly higher tempera-ture than the liquid water source.24 The material is highly viscous with a density T.T. Wall J . Chem. Phys. 1969 51 113. l o R. G. Gordon J . Chem. Phys. 1965,43 1307. I 1 V. Parkash M. K. Dheer and T. S. Jaseja Phys. Letters A 1969 29 221. l 2 Yu. V. Gurikov L. V. Moiseeva and A. 1. Sidorova Dokfady Akad. Nauk S.S.S.R., l 3 G. B. Woolsey Diss. Abs. 1968 28B 4972. l 4 W. A. P. Luck and W. Ditter Z . Naturforsch. 1969 24b 482. l 5 A. I. Sidorova Opt. i Spektroskopiya 1969,26 1055. l 6 J. E. Bertie H. J. Labbe and E. Whalley J . Chem. Phys. 1968 49 2141; 1969 50, l 7 R. A. Fifer and J. Schiffer J . Chem. Phys. 1969 50 21. l 9 R. E. Hester K.Krishnan and C. W. J. Scaife J. Chem. Phys. 1968 49 1100. 2 o J. Guillermet and A. Novak J . Chem. Phys. 1969 66 68. 2 1 A. V. Karyakin and G. A. Muradova Zhur. j i z . Khim. 1968,42,2735. 2 2 B. V. Derjaguin M. V. Talaev and N . N . Fedyakin Dokfady Akad. Nauk S.S.S.R., 3965 165 597 (Phys. Chem. Sect.); translation in Proc. Acad. Sci. (U.S.S.R.) Phys. Chem. 1965,165 807. 1968 182 1044. 4501. V. Seidl 0. Knop and M. Falk Canad. J . Chem. 1969 47 1361. 2:' F. J. Donahoe Nature 1969,224 198. 2 4 L. J. Bellamy A. R. Osborn E. R. Lippincott and A. R. Bandy Chem. and Ind. 1969, 686 82 R. E. Hester of ca. 1.4gcm-3 and the ability to remain liquid over the temperature range - 70 "C (where a glass is formed) to something in excess of 500 "C. It has a refrac-tive index of 1.48 and a binding energy in the range -100 kcal mole- '.The several groups working independently on anomalous water are convinced that the extraordinary physical properties are not due to the presence of dissolved but rather to the polymeric nature of the material. The Raman spectrum2' shows a high intensity band at 630 cm- which has been compared with bands given by strongly hydrogen-bonded systems such as the HF2- ion, and which leads to an estimated 0 . . * 0 distance of cu. 2.3 A and a hydrogen-bond energy of cu. 30 kcal mol- '. Based on this unusually strong hydrogen bond, polymeric structures have been suggested for the liquid containing negatively charged hexagonal layers of molecules the layers being held together by protons or hydronium ions or containing highly branched chain polymers with few normal OH groups.Much of the structural theory appears to be mere speculation at this stage however and a good deal more experimental work needs to be done. The suggestions that this is the secret of the planet Venus's missing water,23 and that water in biological systems might be of this anomalous ~ariety,~' similarly await experimental tests. The fact that anomalous water has been recovered unchanged after mixing with normal water reassures us slightly that we are not in immediate danger from total conversion of the metastable phase, and it has been pointed out2* that although the conditions for formation of anomalous water (quartz surface and greater than 95% humidity) are wide-spread in nature the Earth's waters have withstood the test for billions of years.Wt Soluhns-It is well known that the structure of normal liquid water is greatly modified by the presence of dissolved electrolytes though while the structure of the pure solvent itself remains undetermined it must be anticipated that information on the detailed nature of the changes induced by electrolytes will remain imprecise. The thorough investigations of the Hornig and of Wa1rafen3* have been followed recently by Kecki and c o - ~ o r k e r s ~ ~ in studm of the effect of electrolytes on the i.r. bands of water. Nitrates. An isolated and unperturbed NO3- ion would have D3, symmetry and give rise to only four vibrational frequencies corresponding to the species Al' (R) + A2" (i-r.) + 2E' (R i.r.) where R and i.r.indicate Raman and i.r. activity respectively. Lowering of the NO3 - ion symmetry by loss of the equivalence of 2 5 E. Willis G. R. Rennie C. Smart and B. A. Pethica Nature 1969 222 159. 26 E. R. Lippincott R. R. Stromberg W. H. Grant and G. L. Cessac Science 1969 164, 2 7 E. R. Lippincott reported at the International Conference on Raman Spectroscopy, 2 8 3. D. Bernal P. Barnes I. A. Cherry and J. L. Finney Nature 1969 224 393; D. H. 29 3- W. Schultz and D. F. Hornig J. Phys. Chem. 1961,65 2136; T. T. Wall and D. F. 30 G. E. Walrafen J . Chem. Phys. 1962 36 1035; 1966,44 1546. 1482. Ottawa August 1969. Everett J. M. Hayna and P. J. McElroy ibid. 394. Hornig J. Chem. Phys. 1967 47 784. Z Kecki P. Dryjanski and E. Kozlowska Roczniki Chem. 1968,42,1749; P.Dryjanski and Z . Kecki ibid. 1969 43 1053 Spectroscopy of Concentrated Electrolyte Solutions and Fused Salts 83 the three 0-atoms is expected to lead to loss of degeneracy from the E’ modes.32 It is somewhat mysterious then to find that aqueous alkali-metal nitrates reveal that v3E’ appears as a doublet split by ca. 56 cm- for all dilute solutions studied to date by both Raman and i.r. spectroscopy whereas v4E’ remains an unper-turbed singlet.33 This anomalous splitting has been ascribed to NO3- ion per-turbation by water molecules and it persists in ion-pairs which are believed to be of the solvent-separated type but gives way to the anticipated splitting of both E‘ bands when contact ion pairs are formed.34 On this basis it appears that calcium cadmium mercury(u) copper(II) indium(m) cerium(Iv) and bismuth nitrates form inner-sphere metal-nitrate complexes (contact ion-pairs) whereas the alkali-metal ions and the zinc(I1) ion form only outer-sphere complexes (solvent-separated ion-pairs).’Although experimentally sound the theoretical basis for this distinction between the two types of perturbed NO3- spectrum is obscure. An alternative to the symmetry point group treatment which might have some value in this context is a quasi-lattice interpretation in which correla-tion field coupling effects between neighbouring anions in solution might play a significant role. Cooney and Hall have found this type of analysis useful in accounting for the nitrate spectrum of the crystalline monohydrate Hg(N03)2, H20 which they regard as an extension of the aqueous solutions of the com-pound.35 The perturbation of the dilute solution nitrate spectrum which has been ascribed to hydration has been verified by comparison with i.r.studies of chloroform and methanol solutions of tetraphenylarsonium nitrate.36 These produced evidence for hydrogen-bonded interactions between NO3 - and the solvent molecules suggesting that in water also the interaction takes the form of specific hydrogen-bond formation. Stepwise complex formation between bismuth(u1) and nitrate ions followed by Raman intensity changes has been reported by Oertel and Plane.37 Their spectra were consistent with C2” nitrate symmetry and the polarisation of the highest frequency Raman band (at ca. 1500 cm- ’) inferred bidentate co-ordina-tion.No evidence for polynuclear species incorporating bridging nitrate groups was found but co-ordination of up to four nitrates per bismuth was established, with water molecules also bound to the metal as shown by a polarised band at ca. 370 cm- ’. This was attributed to the symmetric Bi-OH2 stretching mode. Similar Raman intensity studies of aqueous solutions of a variety of metal nitrates by Russian workers3* have shown a decrease in the vlAl’ band integrated intensity which correlates with the splitting of the v3E’ band these effects again being ascribed to cation-nitrate ion interactions. 32 R. E. Hester and W. E. L. Grossman Znorg. Chem. 1966 5 1308; H. Brintzinger and 3 3 D. E. Irish and A. R. Davis Canad. J. Chem. 1968,46,943. 34 D.E. Irish A. R. Davis and R. A. Plane J. Chem. Phys. 1969,50 2262. ” R. P. J. Cooney and J. R. Hall Austral. J . Chem. 1969 22 337. 36 A. R. Davis J. W. Macklin and R. A. Plane J . Chem. Phys. 1969 50 1478. 37 R. P. Oertel and R. A. Plane Znorg. Chem. 1968 7 1192; R. P. Oertel Diss. Abs., ’* L. V. Volod’ko and X. Le TharS Zhur. priklad. Spektroskopii 1968 9 644. R. E. Hester ibid. 1966 5 980. 1969,29B 4083 84 R. E. Hester Irish and Plane and their students have shown that solvent interaction with the nitrate ion taken together with metal ion complexation can cause the bands derived from v3E' to have complex multiplet structures in spectra of concentrated metal nitrate solutions. The significance of the quantitative splitting data ob-tained from a recent study of acetone solutions of a wide range of metal nitrates,39 (most of them being used as hydrates) must therefore be considered dubious.A much more thorough investigation of acetonitrile solutions of the anhydrous nitrates of Zn Cd and Hg" has produced further spectroscopic evidence for strong perturbation of nitrate ions by solvated cations.40 The presence of the major solvated species [Zn(CH3CN),] (N03)2 has been established from Raman intensities measured on zinc nitrate solutions and a band at 248 cm- assigned to an Hg-0NO2 stretching mode in the mercury complex. The acetonitrile solvent spectrum perturbations were found to be consistent with co-ordination of CH3CN to the metal ions through the nitrogen atom though the CH3CN v2 band assignment appears to have been in error in this work.41 Halides.Solutions of halide complexes formed by several non-transition elements have been studied by Raman and i.r. methods. Aqueous solutions containing various ratios of chloride to Bi"' ions were reported to contain the species BiCI4-, BiClS2- and BiCl,,- and in addition lower halide complexes with three two, and possibly one C1- per Bi"' were identified by their Raman spectra.42 However, in acetone solutions containing I - and Bi"' only the Bi14- and Bi163- species appear to have any stability.43 The hexahalide ions were identified as having octahedral geometry but the tetrahalides both have lower symmetry than tetrahedral. The Bi14- ion appears to have C2" symmetry suggesting a trigonal bipyramidal structure with one equatorial position occupied by an electron lone-pair and it appears that the Bi"' 'inert pair' is similarly stereochemically active in the BiC14- species.Related work on AsCl solutions in H20 D20 CH30H, C2H50H CH3CHOHCH3 and (CH3CH2)20 has shown the species formed to be much more sensitive to solvent effects.44 A pH-dependent equilibrium between the species H20AsC13 and HOAsC1,- has been postulated but no evidence for loss of C1- or formation of As"' species with fewer than three chlorides was found up to the pH at which As406 was precipitated. The AsC1,- complex evidently forms only on addition of C1- to aqueous AsCl,. This tetrachloro-arsenite has been characterised by Raman spectra of ether extracts of AsC1,-A useful review of metal-halogen stretching vibrations in co-ordination com-plexes of gallium indium and thallium has been prepared by cart^,^ though ~ ~ 1 .4 5 3 9 G. Norwitz and D. E. Chasin J . Inorg. Nuclear Chem. 1969 31 2267. 4 0 C. C. Addison D. W. Amos and D. Sutton J . Chem. SOC. ( A ) 1968 2285. 4 1 J. C. Evans J . Chem. SOC. ( A ) 1969 1849. 4 2 R. P. Oertel and R. A. Plane Inorg. Chem. 1967 6 1960. 43 R. A. Spragg H. Stammreich and Y. Yawano J . Mol. Structure 1969 3 305. 44 T. M. Loehr and R. A. Plane Inorg. Chem. 1969 8 73. 4 5 J. E. D. Davies and D. A. Long J . Chem. SOC. ( A ) 1968 1757 1761. 46 A. J. Carty Co-ordination Chem. Rev. 1969 4 29 Spectroscopy of Concentrated Electrolyte Solutions and Fused Salts 85 little attention has been paid to solvent effects. Hanson and Plane47 have used Raman spectroscopy to establish the presence of the species InBr2+ InBr,', and InBr3 in aqueous solutions of indium bromide and of the following chloride complexes in indium chloride solutions InC12 + InC12 + and InCl - .The spec-tra here showed strong evidence for the participation of water in the inner co-ordination sphere of the metal atom. Similar mixed aquo-fluoro-complexes of Be" have been characterised by an aqueous solution study in which Raman spectroscopy has played a part.48 Far4.r. absorption spectra have been reported for the anionic tetrachloro-complexes of Al Ga"' and In"' in benzene solution, as well as for Zn Mn" Co'I and Ni" tetrachlorometallates.49 Some evidence for anionsation interactions in these solutions was found when cations having acidic protons were used and it was assumecbtbat hydrogen bonding was involved.Raman studies of mixed halide complexes of cadmium in water and organic solvent^,^' and of mixed trihalogenomercurate(D) complex anions in alcoholic solutions5 have been reported. Some solid-state spectroscopic work which has a bearing on earlier solution studies is Beattie and co-workers' recent examina-t i ~ n ~ ~ of single crystal Raman spectra from systems containing CuC142- and ZnClh2- ions and from CuC1,,2H20. Raman spectra of some interhalogen complex ions have been reported by Shamir.53 Vibrational spectra of a large number of transition-metal-halide complex ions have been reviewed by James and Nolan,54 who also have applied the spectroscopic data to the determination of the nature of metal-ligand bonding in complexes.Creighton's' recent Raman spectroscopic study of the hexa-chlorotitanate(1v) ion in non-aqueous solution confirms the earlier e~idence'~ indicating that aqueous hydrochloric acid solutions of titanium(1v) chloride do not contain appreciable quantities of TiC16 - anions. Creighton found his Raman spectra of WCl and WC162 - to be consistent with their having octahedral symmetry in solution in a variety of solvents. WF6 has also been shown to have octahedral symmetry in solution,57 in spite of claims that it forms 1 1 charge transfer complexes with molecules of the solvents used. Interesting far-i.r. work on the assignment of bridging metal-halogen stretching vibrations in several co-ordination compounds in the solid state has been r e p ~ r t e d ~ * ? ~ ~ this work 4 7 M.P. Hanson and R. A. Plane Inorg. Chem. 1969,8 746. 4 8 R. E. Mesmer and C. F. Baes jun. Inorg. Chem. 1969 8 618. 4 9 M. L. Good C.-C. Chang D. W. Wertz and J. R. Durig Spectrochim. Acta 1969, 5 0 N. Yellin Israel J . Chem. 1969 7 43. 5 1 J. R. Saraf R. C. Aggarwal and J. Prasad J . Inorg. Nuclear Chem. 1969 31 2123. 5 2 I. R. Beattie T. R. Gilson and G. A. Ozin J . Chem. SOC. ( A ) 1969 534. J. Shamir Israel J . Chem. 1969 7 495. 5 4 D. W. James and M. J. Nolan Progr. Znorg. Chem. 1968 9 195. 5 5 J . A. Creighton Chem. Comm. 1969 163. 5 6 J. E. D. Davies and D. A. Long J . Chem. SOC. (A) 1968 2560. 5 7 H. J. Clase A. M. Noble and J. M. Winfield Spectrochim. Acta 1969 25A 293. 5 8 C. Postmus J. R. Ferraro A. Quattrochi K.Shobatake and K. Nakamoto Inorg. 5 9 I . E. Grey and P. W. Smith Austral. J . Chem. 1969 22 1627. 25A 1303. Chem. 1969 8 1851 86 R. E. Hester again having a direct bearing on assignments made for polynuclear species present in solution. Other Salts. A recent demonstration of the power of the vibrational spectroscopic method for establishing the presence of contact ion-pairs in solution is provided by Edgell and Pauuwe’s6’ work on Na2Cr2(CO)lo in THF and DMSO. This showed the presence of a [Cr2(CO)10] 2 - anion having D4d symmetry as expected for an anion environment in which one axial CO is contact-ion-paired with a Na’ ion and the other CO groups have solvent molecules as nearest neighbours. Effects of this type can easily lead to erroneous interpretation of spectra from ionic species in solvents of moderate to low dielectric constant if the possibility of ion-pair formation is neglected.The nature of the species present in aqueous solutions of vanadates(v) per-oxovanadates(v) molybdates and tungstates(v1) over a range of pH values and with metal concentrations at ca. IM has been investigated by Raman and i.r. spectroscopy.6 The spectra are very complex and the analysis was necessarily highly empirical in parts there being little hope of rigorous theoretical analysis when species such as variously protonated [Vlo02,]6- [V20 l(H20)2]4-, and [Mo7024]4- contribute. However this work illustrates how useful spectros-copic data can be for such complicated systems when used together with other physical evidence. Evidence for polycondensation interactions in concentrated aqueous solutions of per-rhenic acid also has been obtained from a combined Raman i.r.and ‘H n.m.r. study,62 and new work on aqueous and crystalline K2Cr207 has been reported which is in serious disagreement with earlier Griffith64 has examined vibrational spectra from a number of monomeric oxy-complexes of the form [MO,X6-,IY- with n = 2,3,4 and 6; M = Os Re Mo, and W ; X = C1 Br OH CN NCS etc. and also the spectra of dinuclear species involving one or two bridging oxygen atoms. Again the spectra are exceedingly complex but much useful structural information has been derived from their analysis. A classical study of solution formation of the four species As(OH)~, AsO(OH)~- AsO~(OH)~- and A s O ~ ~ - has been reported by Loehr and Plane,65 applying the Job method of continuous variations to their Raman spectra from aqueous solutions of As”’ containing a wide range of OH - concentrations.Blatz and Waldstein66 recently reported what they called the first vibrational spectral evidence of ionic association in aqueous solution. This extravagant claim was in the context of their interesting study of low-frequency Raman spectra from aqueous solutions of sodium and ammonium formate and acetate, wherein they also determined bands believed to characterise complexes formed between the anions and water. Other recent spectroscopic work on oxyanion 6 o W. F. Edgell and N. Pauuwe Chem. Comm. 1969,284. 6 1 W. P. Griffith and P. J. B. Lesniak J . Chem. SOC. ( A ) 1969 1066. 6 2 K.Ulbright R. Radaglia and H. Kriegsmann 2. anorg. Chem. 1968 356 22. 6 3 M. S. Mathur C. A. Frenzel and E. B. Bradley J . Mol. Structure 1968 2 429. 64 W. P. Griffith J . Chem. SOC. (A) 1969 211. 6 5 T. M. Loehr and R. A. Plane Znorg. Chem. 1968,7 1708. 6 6 L. A. Blatz and P. Waldstein J . Phys. Chem. 1968,72 2614 Spectroscopy of Concentrated Electrolyte Solutions and Fused Salts 87 species in solution has been concerned with silicates,67 phosphates,68 sul-p h a t e ~ ~ ~ ~ ’ and trihalogenomethane s~lphonates.~ Comparisons of vibrational spectra from Hg(SCN)2 in the solid state and in DMSO solution have shown the solution species to be co-ordinated by DMSO molecules though it appears to retain a linear SHgS str~cture.~’ Similarly, marked frequency shifts in v(Hg-S) accompanying dissolution of CH,HgSCN in methanol have been observed and related to structural changes.Thiocyanate solution spectra also have been examined for a correlation between the integrated absorption of the C-N i.r. band and the type of co-ordination (M-SCN or M-NCS).73 The intensities were found to be very sensitive to solvent changes, so that the criteria established are useful only if the solvent is kept constant. Raman spectra of aqueous edta and its complexes with a wide range of metal ions have been studied and characteristic metal-nitrogen bands assigned.74 Taking the appearance of a Raman band as indicating covalent character in the chemical bond undergoing vibrations it seems that the metal-oxygen bonds in edta complexes are mainly electrostatic though significant covalent character is to be associated with metal-nitrogen bonds in most cases.Metal ion complexa-tion by o-phenanthroline in aqueous solution also has been studied by Raman spectro~copy,~~ and metal-nitrogen stretching frequencies characterised. For solution i.r. work on some trisacetylacetonato-metal(Ir1) complexes chloroform has been used as the solvent.76 The magnitudes of M-0 vibration absorption intensities are correlated with the covalent character of the M - O bonds for the metals Al Cr Fe and Co. A final example of specific solvation studies is the far-i.r. characterization of vibrational modes due to alkali-metal ions and ammonium ions in solvent cages.77 The lithium ion was shown to be solvated by four mole-cules of 1-methyl-2-pyrrolidinone in a mixed solvent system dioxan with the pyrrolidinone but the absence of Raman bands assignable to metal-ligand vibrations indicated that the bond with solvent is essentially ionic.Fused Salts.-Nitrates. Metal nitrates have been investigated more thoroughly as melts than have most other salts. This is due mainly to the fact that they are low-melting have long stable liquid ranges and provide a polyatomic ion the vibrational modes of which can be used as an indicator of interionic interactions. Evidence is accumulating which suggests that the alkali-metal nitrate melts might have substantial lattice-like structure necessitating a factor-group analysis of their vibrational modes but that the quasi-lattice interactions between 67 K. Ichikawa and I.Iwasaki J . Chem. SOC. Japan 1968,89 1217. 6 8 S. Pinchas and D. Sadeh J. Inorg. Nuclear Chem. 1968,30 1785. 6 9 L. V. Volod’ko and L. T. Khoakh Zhur. priklad. Spektroskopii 1969 10 779. 7 0 R. S. Katiyar and N. Krishnamurthy Indian J . Pure Appl. Phys. 1969 7 95. 7 ’ M. G. Miles G. Doyle R. P. Cooney and R. S . Tobias Spectrochim. Acta 1969, 7 2 R. P. J. Cooney and J. R. Hall Austral. J . Chem. 1969 22 21 17. 7 3 R. Larsson and A. Miezis Acta Chem. Scand. 1969 23 37. 7 4 K. Krishnan and R. A. Plane J . Amer. Chem. SOC. 1968,90 3 195. 7 5 K. Krishnan and R. A. Plane Spectrochim. Acta 1969,25A 831. 7 6 R. Larsson and 0. Eskilsson Acta Chem. Scand. 1969 23 1765. 7 ’ J. L. Wuepper and A. I. Popov J. Amer. Chem. SOC. 1969,91,4352. 25A 1515 88 R. E. Hester components of fused nitrates containing multiply-charged metal ions are weak compared with more specific pair-wise interactions which can be characterised as complex ion formation.Devlin and his c o - ~ o r k e r s ~ ~ have produced the most extensive set of data supporting the liquid quasi-lattice model. These have largely been based on i.r. attenuated total reflection studies of alkali-metal nitrates, showing multiplet structure in bands due to the internal modes of the nitrate ion, which have been interpretable in terms of orthorhombic or cubic lattice structures. Most recently frequency shifts produced by isotopic isolatiQn of the I4No3-ion in fused Li' 5N03 have been interpreted as the result of decoupling of I4NO3-ions in a perturbed liquid lattice structure.Support for these proposals is found in a number of studies of low frequency Raman and far4.r. bands produced by fused alkali ' though some disagreement exists between the various contributors as to which is the most appropriate type of lattice structure. In the region below 700 cm- there are no vibrational bands due to internal modes of NO3- and bands which have been observed below 200 cm- have been attributed to phonon modes in the liquid quasi-lattice. The observation of progressive shifts in the frequencies and widths of bands in the lattice region of crystalline alkali-metal nitrates up to and through the melting point has suggested specific assign-ments of these to phonons associated with librational motions of NO3- groups.8' Their depolarised (Raman) character is consistent with this though the recent report of essentially identical bands at ca.120 cm- ' from fused AgN03 and from concentrated solutions of AgN03 in acetonitrile82 clouds the issue considerably. It is difficult to conceive of identical quasi-lattice structures for a fused salt and its solution in any solvent and there is other evidence to show that in acetonitrile the Ag+ ion is strongly ~ o l v a t e d . ~ ~ An attempt has been made84 to relate the shape of the 830 cm- i.r. band which arises from out-of-plane deformation vibrations of nitrate ions in alkali-metal nitrates to intermolecular potentials through the theory developed by Gordon.85 Interestingly failure to explain the experimental result has been attributed to limitations in the experimental determination of the bands rather than in the theoretical approach.The subject of band shapes and their dependence on mole-cular motion and intermolecular potentials is currently receiving a lot of attention in a much wider context than that of electrolyte solutions and fused salts,86 and ' 1 3 J. P. Devlin P. C. Li and G. Pollard J . Chem. Phxs. November 1969; P. C. Li and J. P. Devlin ibid. 1968 49 1441 ; K. Williamson P. C. Li and J. P. Devlin ibid., 1968,48 3891. G. H. Wegdam R. Bonn and J. van der Elsken Chem. Phys. Letters 1968 2 182. G. J. Janz and K . Balasubrahmanyam in 'Proceedings of the International Conference on Raman Spectroscopy,' Ottawa 1969. 7 9 D. W. James and W. H. Leong J . Chem. Phys. 1969 51 640. 13' J. H. R. Clarke Chem. Phys. Letters 1969 4 39.83 C. B. Baddiel M. J. Tait and G. J. Janz J . Phys. Chem. 1965 69 3634. 8 4 R. Bonn G H. Wegdam and J. van der Elsken J . Chem. Phys. 1969,50 1901. * 5 R. G. Gordon J. Chem. Phys. 1963,38 1724; 1963,39,2788; 1964,41,2819. 8 6 R. G. Gordon W. Klemperer and J. I. Steinfield Ann. Rev. Phys. Chem. 1968 19, 215; M. J. Jacob J. Leclerc and J. Vincent-Geisse J . Chim. phys. 1969,66,970; W. R. Hess and J. Brandmuller Z . Physik 1969,224 144; G. E. Ewing Accounts Chem. Res., 1969 2 168; M. Davies G. W. F. Pardoe J. Chamberlain and H. A. Gebbie Chem. Phys. Letters 1968 2 411; B. Bulkin Helv. Chim. Acta 1969 52 1348 Spectroscopy of Concentrated Electrolyte Solutions and Fused Salts 89 doubtless the fruits of all this activity will eventually provide further insight into the nature of the low frequency ‘external’ modes as well as the ‘internal’ modes of nitrates.Raman band width data and their dependence on temperature for a series of fused monovalent metal nitrates have been used to calculate the time for orientational relaxation of NO3- in the melts.87 The relation of this time to the potential barrier for reorientation of NO3- and the activation energy for viscous flow has been examined. Raman and i.r. spectra from a large number of mixed alkali-metal nitrate-multiply-charged metal nitrate systems in the molten state have been studied in recent years.87 These have in all cases shown strong evidence for specific pair-wise metal-nitrate interactions and a complex-ion model has been found most suited to their interpretation.Many of these mixtures have been found to be glass-forming systems which provide the possibility of low-temperature fused salt s t ~ d i e s . ~ ~ ~ ’ Mixed Ca(N03)2-KN03 and Ca(N0&NaN03 systems have been studied recently by Vallier and Ricodeau,” who have found some temperature-dependent frequency shifts. Halides. Maroni and Cairns’ have reviewed the whole field of Raman studies of fused salts with particular attention to halide-containing systems. Their own work on the system MgCl,-KCl showed evidence for a magnesiumxhloride complex equilibrium involving several structurally different species. MgC14’ -, a less well-defined MgC1,’”- ’)- and (MgCl,) polymeric species were proposed. A similar analysis of Raman spectra for the MgBr2-KBr system was reported, and spectra from the Hg12-LiI-KI system were shown to be consistent with formation of planar Hg13-.Raman spectroscopy has been used to establish the presence in mixed InC13-MCl systems (with M being Li K or Cs) of the tetrachloroindate(Ir1) complex ion InC14-.” The spectrum of this complex over the temperature range 4 W 7 0 0 ” ~ is very closely similar to that given by the same species in ether solution at normal room temperature showing a surprising lack of environmental influence on the internal modes of vibration. At high chloride indium ratios formation of InClS2 - and InC163 - species appears likely though environmental perturba-tions (judged by band width effects and sensitivity to the nature of the alkali-metal ion present) here are much more severe.Dissolution of indium metal in molten InC13 results in the formation of lower chlorides of indium containing the metal in a lower oxidation state than In”‘. The nature of species formed in these melts has been investigated by Raman spectro~copy.’~ InCl has the 8 8 R. E. Hester and K. Krishnan J . Chem. Phys. 1967 46 3405; 1967 47 1747; R. E. 8 9 R. E. Hester and K. Krishnan J. Chem. SOC. ( A ) 1968 1955. N. A. Ponyatenko and I. V. Radchenko Opt. i Spektroskopiya 1969,26,645. Hester and C. J. W. Scaife ibid. 1967 47 5253. J. Vallier and J. Ricodeau Compt. rend. 1968 267 B 890; 1969,268 B 622; J. Vallier, J . Chim. phys. 1968 65 1762. 9 1 V. A. Maroni and E. J. Cairns in ‘Molten Salts; Characterisation and Analysis,’ ed. Mamantov Marcel Dekker Ltd. Maidenhead 1969.92 J. H. R. Clarke and R. E. Hester J . Chem. Phys. 1969 50 3106. 9 3 J. H. R. Clarke and R. E. Hester Inorg. Chem. 1969 8 1 1 13 ; Chem. Comm. 1968, 1072 90 R. E. Hester constitution In+ InC1,- in the liquid and In,Cl appears to be In' IIIC~,~- but the tendency of InC1,3- to lose C1- and form the thermodynamically more stable InC14- ion coupled with an apparent disproportionation of In' in the melts, complicates this system considerably. Support for these conclusions has come from independent Raman studies of both solid and molten indium chloride^.^^,^^ Raman spectra of the molten salts MnC12 TlCl and SnBr have been reported recently,96 as have studies of the mercuric halides HgI, HgIBr and HgBr2 in the molten state.97 Other Salts. A comparison of Raman and i.r.spectra from crystalline and fused anhydrous lithium perchlorate has been made.98 Although the melt spectra are simpler than those from the solid state a complete lack of coincidences between Raman and i.r. frequencies has been reported and interpreted in terms of co-operative vibrational modes of a pseudo-lattice having cubic symmetry. Fused alkali-metal sulphates however show signs of asymmetric cation-anion interac-t i o n ~ . ~ ~ The appearance of the formally forbidden v2E mode at ca. 450 cm- in i.r. spectra indicates some weak asymmetric perturbation of the tetrahedral SO4 - ion. More pronounced perturbations result from the addition of divalent metal ions to a K2SO4 melt with lifting of mode degeneracies indicating the presence of S042- ions with symmetry lowered to C3".Vibrational spectra from fused alkali-metal thiocyanates and from their solutions of a large number of divalent metal salts have been analysed in terms of ion-pair or complex-ion formation.'00 The spectra enable S- and N-co-ordination of metal ions with the SCN- ion to be distinguished. Fused metal carbonates have proved more difficult to study due to their highly corrosive nature and thermal instability but some spectroscopic work has been completed ' and reports by Maroni and Cairns and by Hester and Krishnan will be published in 1970. The nature and strength of metal-carbonate interactions in the alkali-metal fused salts appears to be very similar to those already reported for the corresponding nitrates. Polyelectrolytes in Solution.-Although this area of spectroscopic activity barely qualifies for a place in this particular review it is worthy of note that this is an area of rapidly increasing endeavour.Current interest centres on systems of biological significance and much of the work in progress involves Raman spectra of aqueous solutions. Lord and co-workers"' have studied Raman spectra of the enzyme molecules lysozyme ribonuclease and a-chymotrypsin in aqueous solution where changes in the solution conditions (pH temperature ionic 94 J. T. Kenney and F. X. Powell J . Phys. Chem. 1968,72 3094. 9 5 F. J. J. Brinkmann and H. Gerding Rec. Trav. chim. 1969 88 275. 96 J. T. Kenney and F. X. Powell U.S.G.R.D.R. 1969,69(5) 56 AD680 039. 9 7 J. H. R. Clarke and C. Solomons J . Chem.Phys. 1968,48 528. 9 8 W. H. Leong and D. W. James Austral. J . Chem. 1969 22,499. 99 R. E. Hester and K. Krishnan J . Chem. Phys. 1968 49 4356. l o o R. E. Hester and K. Krishnan J . Chem. Phys. 1968 48 825. I o 1 R. C. Lord and G. J. Thomas jun. Spectrochim. Acta 1967 23A 2551; Biochim. Biophys. Acta 1967,142 1 ; Developments Appl. Spectroscopy 1968,6 179; R . C . Lord and N. Yu to be published Spectroscopy of Concentrated Electrolyte Solutions and Fused Salts 91 strength etc.) can induce conformational changes. Components of enzyme molecules such as the amino-acids and small peptide units and of some common-ly occurring purine and pyrimidine derivatives of RNA and DNA have also been investigated spectroscopically. Information on the structure and conformation of these macromolecules in solution appears to be available from such studies.Other Raman studies in this area have been reported by Okamura'" and by Hiranolo3. Smith and co-workers'04 have obtained Raman spectra of poly-L-proline in the solid state and in aqueous solution and have concluded that only very minor conformational changes accompany dissolution. Rimai and co-worker~"~ have used their Raman spectra of adenosine mono- di- and tri-phosphates (AMP ADP and ATP) in aqueous solution over the pH range 0-5-1 3.5 to distinguish analytically between these three molecules and have shown that the degree of ionisation of the terminal phosphate group could be estimated from bands at ca. 960 and 1100 cm- '. Zundello6 has applied i.r. spectroscopy to the problem of determining the role played by water molecules in governing conformation and the nature of interionic interactions in polyelectrolytes basing his studies on organic ion-exchangers which he considers are simple model substances for systems of biopolymers. The most recent contribution dealing directly with biomacromolecules is Tobin'~''~ study of the Raman spectra of two forms of DNA as solids and as aqueous gels. K. Okamura Seibutsu Butsuri 1966 6 183. l o 3 K. Hirano Bull. Chem. SOC. Japan 1968,41 731. lo' M. Smith Biopolymers 1969 8 175. I o 5 L. Rimai T. Cole J. L. Parsons J. T. Hickmott jun. and E. B. Carew Biophysics, lo' G. Zundel Angew. Chem. Internat. Edn. 1969 8 499. ' 0 7 M. C. Tobin Spectrochim. Acta 1969 25A 1855. 1969 9 320