Transfer of Protons through “Pure” Ice Ih Single Crystals. II. Molecular Models for Polarization and Conduction

Abstract

Neither the polarization of pure ice single crystals can be understood on the basis of rotating H₂O molecules nor the conduction through such crystals by assuming migrating H₃O⁺ and OH⁻ ions. Bjerrum's concepts of L‐, D‐defect and ionic‐defect formation are a useful starting point, but the subsequent development of double‐well models leads into serious difficulties. A new interpretation of the intrinsic polarization of ice $I_h$ visualizes $\mathrm{L,\; D}$ pair formation as a one‐step process: a proton shifted by near‐infrared phonon excitation to an empty corner of its H₂O tetrahedron. Subsequent intramolecular proton transfers, slightly field directed, inscribe a dipole moment into the disordered proton system of the ice crystal; the defects die off by recombination. Thus the polarization builds up from a statistically triggered pulse spectrum; its correlation period is the waiting time, until the same H₂O molecule forms again a defect pair. This waiting period, the relaxation time of the Debye spectrum, connects the low‐frequency response of the ice to events in the near infrared. A simple quantitative formulation is given. Next the role of the ionic defects is investigated. These defects cannot follow the field in the ideal crystal: If they move downstream, the antipolarization created by this intermolecular proton transfer pumps the corresponding charge up‐stream. Thus only the $\mathrm{L,\; D}$ defects carry charge through the volume of an ideal ice crystal. Interconversion of $\mathrm{L,\; D}$ defects and ionic defects can occur at the electrodes and may create low‐dielectric‐constant layers in front of them. In addition to these intrinsic effects, extrinsic phenomena are expected and observed. Lattice defects, dislocations, and doping can cause complexities akin to “color centers” in alkali halides. “Catalytic defect mills” may arise when extrinsic effects couple into the proton‐ordering mechanisms of the ice crystal. Part III resumes the experimental account.