Scale-bridging Microstructural Analysis – A Correlative Approach to Microstructure Quantification Combining Microscopic Images and EBSD Data

A comprehensive description of complex material structures may require characterization using different methods and observations across several scales. This work will present a correlative approach including light optical microscopy, scanning electron microscopy and electron backscatter diffraction, enabling microstructure quantification which combines microscopic images and electron backscatter diffraction data. The parameters obtained from electron backscatter diffraction such as misorientation parameters or grain and phase boundary data are an ideal source of information, complementing microscopic images. Two case studies performed on bainitic microstructures will be presented to demonstrate practical applications of this approach.

THEORETICAL-EXPERIMENTAL POSSIBILLITIES OF MICROSTRUCTURE QUANTIFICATION OF DISPERSION STRENGTHENED MATERIALS

The present paper is devoted to the possibilities to classify the spatial arrangement of the elements (features) of a stochastic process with geometrical objects. The fundamental quantities describing point processes were introduced. Experimental possibilities of structural objects determination, possibilities of evaluation of the size distribution of the secondary phases, testing of planar point structures (estimation of the process intensity, square method and characteristics of the second order) were estimated. Interparticle distances, namely mean interparticle distance, mean minimum distance, mean visibility and mean path of spherical contact were defined. Selected processes were described and demonstrated from simulated realizations on Al-Al4C3 dispersion strengthened material, prepared by a powder metallurgy method of reaction milling. Interparticle distances of Al4C3 particles were evaluated. Polygonal methods and quadrant counts method were used for characterization of the particle arrangement.

A Deep Learning Approach for Complex Microstructure Inference

Automated, reliable, and objective microstructure inference from micrographs is an essential milestone towards a comprehensive understanding of process-microstructure-property relations and tailored materials development. However, such inference, with the increasing complexity of microstructures, requires advanced segmentation methodologies. While deep learning (DL), in principle, offers new opportunities for this task, an intuition about the required data quality and quantity and an extensive methodological DL guideline for microstructure quantification and classification are still missing. This, along with a lack of open-access data sets and the seemingly intransparent decision-making process of DL models, hampers its breakthrough in this field. We address all aforementioned obstacles by a multidisciplinary DL approach, devoting equal attention to specimen preparation, contrasting, and imaging. To this end, we train distinct U-Net architectures with 30–50 micrographs of different imaging modalities and corresponding EBSD-informed annotations. On the challenging task of lath-bainite segmentation in complex-phase steel, we achieve accuracies of 90% rivaling expert segmentations. Further, we discuss the impact of image context, pre-training with domain-extrinsic data, and data augmentation. Network visualization techniques demonstrate plausible model decisions based on grain boundary morphology and triple points. As a result, we resolve preconceptions about required data amounts and interpretability to pave the way for DL's day-to-day application for microstructure quantification.

THEORETICAL-EXPERIMENTAL POSSIBILLITIES OF MICROSTRUCTURE QUANTIFICATION OF DISPERSION STRENGTHENED MATERIALS

<p>The present paper is devoted to the possibilities to classify the spatial arrangement of the elements (features) of a stochastic process with geometrical objects. The fundamental quantities describing point processes were introduced. Experimental possibilities of structural objects determination, possibilities of evaluation of the size distribution of the secondary phases, testing of planar point structures (estimation of the process intensity, square method and characteristics of the second order) were estimated. Interparticle distances, namely mean interparticle distance, mean minimum distance, mean visibility and mean path of spherical contact were defined. Selected processes were described and demonstrated from simulated realizations on Al-Al₄C₃ dispersion strengthened material, prepared by a powder metallurgy method of reaction milling. Interparticle distances of Al₄C₃ particles were evaluated. Polygonal methods and quadrant counts method were used for characterization of the particle arrangement.</p>