Quantitative assessment of the interfacial roughness in multi-layered materials using image analysis: Application to oxidation in ceramic-based materials

A method is developed to characterize interfacial roughness of layered materials. A series of contiguous images is acquired with a Scanning Electron Microscope on polished cross-sections. Using image analysis algorithms based on mathematical morphology operations, interface profiles are extracted and mathematically processed to calculate standard surface roughness parameters. In addition, morphological criteria describing the interface tortuosity are determined, such as the “normalized rumpling index” (based on the profile developed length) and the “barbule density” that quantifies the protrusions of the layer (using the skeletonization function). The validity and accuracy of the method is evaluated by comparison with measurements performed by the conventional surface metrology techniques. This method is used to study internal oxidation of two ceramic-based materials (Si3N4–TiN composite/thermal barrier coatings) upon thermal aging. In both cases, a link is established between the roughening of the oxide layer interface and a decrease in the mechanical strength of the multi-layered material.