Examination of the theory of heterogeneous nucleation of ice has led to a number of observations which indicate several inconsistencies between this theory and experimental observations. Using activity curves for AgI-bearing smokes proposed by Fletcher, we have calculated the hypothetical efficiency curves from log-normal distributions of such smokes. We have also generated ΔT vs radius (Ra) data from fitted curves to experimental activity data to test the values of contact and other parameters in the theory. As a result of the above analysis, we observe that:
Using a KI-AgI activity curve and log-normal smoke distributions, the resulting efficiency curve shows a sharp-kneed trend completely unlike the smooth sweeping trends of actual curves observed by experiments.
Some of the data involving means of distributions of smoke sizes are considerably different from those estimated from electron micrographs.
When inserted into the theoretical expressions the generated ΔT and Ra data result in values of σ12, m and δ (pit area factor) and/or trends of f(m,xi) and xi which are incompatible with the theoretical treatment and reasonable values for these physical constants.
Recognizing that a semi-empirical theory of this type has its place in the development of a scientific discipline, it is nevertheless apparent that analyses of this type based upon scanty experimental values for the physical parameters contained therein will continue to present an inaccurate picture of nucleation processes unless a greater effort is given to the determination of such parameters as contact angle, interfacial energies, and actual surface structure of nucleant materials.