Quantitative investigation of polymer composites’ destructions structures

The article discusses a technique for studying the structures of destruction of polymeric materials (crazes) using computer microscopy methods. Craze structure images were processed in two ways, one of which is based on image segmentation, and the second one is based on processing the image as a whole using the autocorrelation function method. The segmentation method made it possible, using threshold separation, to identify microvoids, which are typical for a craze structure, to determine their geometric characteristics, in particular, the area, to construct approximating ellipses, to determine the angle of inclination of the major semiaxis and the ratio of the lengths of the semiaxes. The display of these characteristics using histograms indicates the formation of a highly ordered structure in terms of angular distribution. Two characteristics - the ratio of the semiaxes of the approximating ellipse and the angle of its inclination are proposed as inplicators of internal deformations that have arisen during the formation of voids in the material. The implementation of the second method for studying images of a craze structure using the autocorrelation function confirmed the high orderliness of the structure of an individual craze. In addition, the results of the mapping of the autocorrelation function indicate the formation of a short-range order structure, the elements of which are individual crazes - a craze ensemble. The advantages of whole image methods for detecting weakly ordered structures in materials are shown.