Assessment of Powder Solidification Structures in Tool Steels Using State-of-the-Art Microstructural Characterization Techniques

• Published in: Key engineering materials Vol. 968; pp. 81 - 86
• Main Authors: Benito, Santiago, Weber, Sebastian
• Format: Journal Article
• Language: English
• Published: Zurich Trans Tech Publications Ltd 06.12.2023
• Subjects: Atomizing; Carbide tools; Cellular structure; Dendritic powders; Elongated structure; Microstructure; Parameter identification; Powder metallurgy; Process variables; Solidification; State of the art; Stereology; Tool steels; Microstructural Characterization; Stereology; Image Analysis; Deep Learning
• ISSN: 1013-9826; 1662-9795; 1662-9795
• DOI: 10.4028/p-nea6Qp

The observation, description, and ultimate prediction of causal connections between processing and resulting macroscopic properties stand at the heart of Materials Science and Engineering. To that end, the microstructure is the subject of intense examination, as it is ultimately responsible for the observed emergent behavior. As a result, many of the scientific or technical questions that we strive to answer boil down to quantitatively studying the—sometimes subtle—effects of processing on the microstructure in terms of known or hypothesized thermodynamic and kinetic phenomena. This statement is naturally also true in the case of hot isostatically pressed powder metallurgy tool steels. In the 50 years since the process' popularization, many parameters have been identified as relevant to microstructure formation during consolidation. Among these process variables, the powder solidification structure distribution is probably the last to join the list. Dendritic solidification during the atomization of relatively massive particles produces slightly elongated carbides. On the other hand, cellular solidification in smaller powder particles is responsible for smaller and more angular carbides. Characterizing powder solidification structure as a function of particle size presents two main challenges: First, the assessment relies on examining cross-sections of the powder particles, which are most likely non-diametric. And, second, the manual identification exercise is tedious and highly subjective. In this work, we show how we achieve fast and reliable powder structure solidification distributions using deep learning combined with state-of-the-art stereology reconstruction techniques.