The Electrofreezing of Supercooled Water Drops

Abstract

Studies have been made of the freezing of supercooled drops of radii ranging from 0.106 cm to 0.134 cm suspended from insulating supports in an environment whose temperature could be varied from 0°C to-45° C. It was found that the probability of a drop freezing within a fixed interval of time was appreciably greater over the temperature range 0 to-22°C if the surface of the drop was disrupted by either electrical or mechanical forces to produce a filament of liquid emanating from the localised area of rupture. For example, for a five-minute test-interval the fraction of drops studied that froze at temperatures-5°C,-10° C,-15°C and-20°C were 0.44, 0,62, 0.75 and 0.88 respectively if the drop wasl disrupted by means of an electric field, 0.25, 0.44, 0.50 and 0.58 respectively if the surface was penetrated by an insulating fibre or conducting wire at the same temperature as the drop, but only 0, 0.02, 0.07 and 0.18 respectively if the drop surface was not disrupted during this interval, but remained undisturbed, was situated in a strong electric field just below the disintegration threshold, or shaken violently on its support. These observations are totally inconsistent with the criterion for the occurrence of electrofreezing deduced by Pruppacher (1963 a, b), namely that the phenomenon is always associated with the movement of a triple-phase boundary. However, the observations are in agreement with the suggestion of Loeb (1963) that the most essential condition for the occurrence of electrofreezing is that a portion of the drop be drawn out into a thin filament; Loeb et al. (1938) had previously shown that such a filament may contain molecular aggregates which act as excellent freezing nuclei. These conclusions were reinforced by high-speed photographs demonstrating that freezing originated from the area of disruption. The influence upon the freezing probability of air-bubbles released into the liquid during the disruption process was shown to be secondary. A review of observations that have been made on natural clouds indicates that although no definitive evidence exists for the occurrence of electrofreezing, a considerable body of indirect information suggests strongly that electrofreezing was responsible for the observed existence of ice particles at extremely high temperatures within supercooled clouds. The characteristics of the observed frozen particles are consistent with those which would be expected from electrofreezing.