Characterization of Micro-Structures in Materials

Licentiate thesis, 2015

On developing the next generation sustainable soft materials, it is often crucial to understand and control their properties and function. Whereas a characterization of three-dimensional data is desirable in corresponding studies, often two-dimensional data are less time consuming to obtain. Consequently, characterizing the three-dimensional micro-structure of a material from two-dimensional data would enable efficient screening of its properties. In this work, the challenge of characterizing two different materials from two-dimensional images obtained by scanning (transmission) electron microscopy is overcome by using tools from image analysis and spatial statistics. The two different materials are a colloidal nanoparticle gel and a polymer blended film. The characterization of the micro-structures in the materials is conducted in two main steps. First, the microscopy images are processed in order to identify the objects of interest. Second, the structures are characterized according to the objects of interest. In particular, the spatial arrangement of nanoparticles is evaluated by summary functions from spatial statistics. One such function based on the size of a cluster of particles has been developed in this project. For the pore space analysis of a polymer blended film, tools from image analysis are applied to measure and compare the shapes of pores in a statistical analysis. The results obtained in this work, may be useful for reconstructions of the micro-structure in materials in three dimensions. Such reconstructions can be used to analyze materials, which may not yet have even been synthesized, in simulation studies without experiments involving valuable resources. In that way, new sustainable high quality products may be developed.