Correlation of Proposed Structure for Superwater with Observed Spectra

The Raman and infrared bands for superwater are correlated with the previous proposal that superwater consists of stable ice-II-type water clusters, whereas normal water consists of stable ice-I-type water clusters. The strong Raman band for superwater at 620 ${\mathrm{cm}}^{-1\mathrm{}}$ and its shoulder at 640 ${\mathrm{cm}}^{-1\mathrm{}}$ correspond to the strong infrared band for ice II at 642 ${\mathrm{cm}}^{-1\mathrm{}}$ and its shoulder at 660 ${\mathrm{cm}}^{-1\mathrm{}}$. The slight shift can be attributed to temperature variations (from -173 to 25°C) of the infrared band. Examples are given to show that stretching frequencies (2500-4000 ${\mathrm{cm}}^{-1\mathrm{}}$) of hydrates or aqueous salt solutions can be shifted to bending frequencies (around 1400 to about 1600 ${\mathrm{cm}}^{-1\mathrm{}}$) because of imposed stresses on hydrogen bonds. These stresses also involve a change in the librational bands, such as a shift from 153 cm to 410 ${\mathrm{cm}}^{-1\mathrm{}}$ for concentrated HCl solutions. Consequently, the reduction of bands between 2500 and 4000 ${\mathrm{cm}}^{-1\mathrm{}}$ and the appearance of both low-frequency bands at 1400 and 1595 ${\mathrm{cm}}^{-1\mathrm{}}$ and strong librational bands at 620 and 640 ${\mathrm{cm}}^{-1\mathrm{}}$, in going from normal water to superwater, can be explained on the basis of imposed stresses on hydrogen bonds in superwater. That is, the hydrogen bonds in ice-II structures of superwater will have greater stresses than those in ice-I structures of normal water. Hence, for superwater, but not for normal water, the bending-frequency bands at 1400 and 1595 ${\mathrm{cm}}^{-1\mathrm{}}$ become more important dissipators (or absorbers) of photons than the stretching-frequency bands between 2500 and 4000 ${\mathrm{cm}}^{-1\mathrm{}}$. Results on concentrated salt solutions show that one bending-frequency band can be shifted to a different one when greater stress is imposed on the hydrogen bonds. In contrast, the hindered rotational bands (600 ${\mathrm{cm}}^{-1\mathrm{}}$ or so) do not shift. Consequently, the 642-${\mathrm{cm}}^{-1\mathrm{}}$ band of ice II can be correlated with superwater, but the bending-frequency bands of ice II at 1690, 1066, and 960 ${\mathrm{cm}}^{-1\mathrm{}}$ cannot. The 1690-${\mathrm{cm}}^{-1\mathrm{}}$ band of ice II most likely corresponds to the 1595-${\mathrm{cm}}^{-1\mathrm{}}$ band of superwater, and the 1066- and 960-${\mathrm{cm}}^{-1\mathrm{}}$ bands of ice II to the doublet at 1400 ${\mathrm{cm}}^{-1\mathrm{}}$. Experimental data on superwater show that both the ionic model of Lippincott and co-workers and the tetramer model of Bolander and co-workers are erroneous. The very small spread of relaxation times in normal water and the spontaneous settling of superwater from normal water at low temperatures support the proposed cluster-aggregate model for both normal and superwater. Stable clusters would produce cooperative effects in relaxation times and an inability of ice-I clusters (normal water) to interact with ice-II clusters (superwater).