Quantitative Characterization of the Spatial Distribution of Corrosion Pits Based on Nearest Neighbor Analysis

Abstract

Nearest neighbor analysis (NNA)-based procedures are proposed for the quantitative characterization of the spatial distribution of corrosion pits in metals. After the exposure of a carbon steel to a 3.5% NaCl solution mist, the results derived from observation of corrosion pit initiation and growth were used to justify the applicability of this approach. The pits initially comprised clusters that were superimposed on a randomly distributed background set. The clustered pits subsequently coalesced, evolving into a more random pit arrangement. Furthermore, it was revealed that in the early stages, the spatial pit distribution can be predicted via inspection of surface inclusions prior to the corrosion process.