Statistical Characterization of Morphological Features of Layer-by-Layer Polymer Films by Image Analysis

This article presents an image analysis-based approach to the statistical characterization of mor- phological features in layer-by-layer (LBL) polymer e lms, which depend on the experimental condi- tions of fabrication. The method is based on the recently introduced concept of exact dilations. This concept allows the precise determination of all possible parallel contours around any general object and, as a consequence, the respective fully accurate Voronoi partition dee ned by the objects. With the use of a fast wavefront propagation scheme, such diagrams can be obtained at high speed. The method has been applied to statistically characterize features such as grain size and distribution on LBL polymer e lms. To do this, the polymer surface is tracked from its highest z values downward to its bottom, and each new connected component is detected as a Voronoi seed. Through deter- mination of equivalent radii of the Voronoi cells, the probability density function characterizing the distribution of the domain radius can then be obtained.