MICHAEL FALK. Can. J. Chem. 58, 1495 (1980). We have studied the infrared spectra of H 2 0 , D20, and HDO in hydrated Nafion membranes in the sodium form. The spectrum of isotopically isolated HDO exhibits two distinct absorption bands in the OH stretching region and similarly in the OD stretching region. These bands have been taken to correspond to two distinct environments of the OH groups in Nafion, identified as 0-H..O and 0-H...CF , . It appears that a substantial proportion of water molecules have an OH group (or occasionally both OH groups) exposed to the fluorocarbon environment. This implies that the hydrated ion clusters are either much smaller than previously estimated or, more likely, are non-spherical in shape with frequent local intrusions of the fluorocarbon phase. Water in Nafion is much less strongly hydrogen-bonded than water in aqueous salt solutions at the same temperature and may be expected to have distinct physical properties.

sl~ows nearly Gaussian contours, complete absence of shoulders, and a single lnaximun~ a t a frequency i~ltermediate betlveen those of HDO iri ice and in vapor. Illcrease of temperature causes a grnd~tal shift of each band in the direction of the f r e q~~e n c y in vapor. 'I'hese obscrvations fully support the "conti~li~urn" modcls of the structure of liql~id water and present strong cvidet~ce against the "rnizt~~re" rllotlels. It is concluded that hydrogeti bonds it1 liqliid \\.atcr have a broad, smooth, sing-le-pealced distributior~ of strengths. whicl~ gl-ad~ially shifts with temperatllre. 'l'he h\.drogen bonds a t one end of this distribution have strengths cornparable to thosc in ice, \vhere;~s those a t the other end are very wealc. These extremes do not correspond to disti~ict mo1ccul;ir species. Every description of licl~~id water ill terms of "bro1;en" a~l d "~~nbrolccn" bonds, or "monomers" and "clusters", is thcrcforc arbitrary. ISTI

The infrared spectra of CH3OH, CH3OD, CD3OH, and CD3OD in the five phases gas, liquid, vitreous solid, α-crystal, and (except perhaps for CD3OH and CD3OD for which the solid-solid transitions have not been studied) β-crystal have been recorded in the range 4000 to 300 cm−1. The Raman spectrum of liquid CD3OH has been recorded. A complete assignment of the internal modes is given, which differs somewhat from previous assignments for the CH3 bending and rocking vibrations. No significant difference in spectrum occurred between the α-crystal and β-crystal phases. Under the full symmetry of the β-phase determined by x-ray diffraction only one OH out-of-plane bending band should occur. Two bands are observed, and it is concluded that the carbon and oxygen atoms in one chain are not coplanar, as is required by the symmetry determined by x-ray diffraction [K. J. Tauer and W. N. Lipscomb, Acta Cryst. 5, 606 (1952)], but that the chains are puckered and the x-ray symmetry arises because the puckered chains are irregularly distributed, a structure that had been previously suggested by Tauer and Lipscomb tentatively on the basis of high apparent thermal amplitudes. Bands occur in the crystal spectra near 500 cm−1 and 340 cm−1 at −180°C. These are interpreted as lattice modes, probably the two infrared-active modes that involve translations of the molecules.

No abstract

No abstract