A laboratory study of charge transfer accompanying the collision of ice crystals with a simulated hailstone

In a series of experiments, conducted inside a cold‐room, clouds of small ice crystals of mean diameter 10μm were drawn at velocities between 1.5 and 18ms −1 past simulated hailstones. Two of these ice targets, side by side but under different conditions, were subjected to crystals from the same clouds simultaneously, so that the effects of various parameters on charge transfer could be isolated. The concentrations, sizes and shapes of the ice crystals were determined. A target heated radiatively with respect to the crystals became charged more negatively than an unheated target. A colder target became charged more positively. However, the magnitude of the charging did not depend on the temperature difference between target and crystals, provided this was greater than about 1 K. Attempts to find evidence of discharge‐limited charging produced negative results. A strong velocity dependence of charging currents was found, but subsidiary experiments on the collection of crystals by the target suggested that this was largely due to the variation in the probability of separation of the crystals, which appeared to be close to zero at the lowest velocities employed. The effects of doping the targets with HF and NH 3 , and of surface roughness, were investigated and were found to be small. These results were found to be in reasonable qualitative and quantitative agreement with the predictions of the charging mechanisms proposed by Auf der Maur and Buser, involving differences in the work functions of ice surfaces of different nature. It was estimated that the charge, q ', transferred in a rebounding collision between an ice crystal and a heated ice target is about 900 aC. It is shown that this is roughly compatible with the general pattern of results of other workers (with one exception) if the charging increases with the square of the diameter, d , of the small particles. There are good theoretical reasons for such a relationship, which may be described by the equation q '(aC)=5 d 2 , for 10μm⩽d⩽200 μm. Charge transfer of this magnitude is estimated to be just sufficient to account for the electrification of thunderclouds.