Quantitative three-dimensional imaging of chemical short-range order via machine learning enhanced atom probe tomography

• Published in: Nature communications 2023-11, Vol.14 (1), p.7410-7410, Article 7410
• Main Authors: Li, Yue, Wei, Ye, Wang, Zhangwei, Liu, Xiaochun, Colnaghi, Timoteo, Han, Liuliu, Rao, Ziyuan, Zhou, Xuyang, Huber, Liam, Dsouza, Raynol, Gong, Yilun, Neugebauer, Jörg, Marek, Andreas, Rampp, Markus, Bauer, Stefan, Li, Hongxiang, Baker, Ian, Stephenson, Leigh T., Gault, Baptiste
• Format: Article
• Language: English
• Subjects: 147/143; 639/301/1023/1026; 639/301/930/12; Alloys; Electrical resistivity; Ferrous alloys; Heat treatment; Heat treatments; Humanities and Social Sciences; Iron; Learning algorithms; Machine learning; Microscopy; multidisciplinary; Nanohardness; Science; Science (multidisciplinary); Short range order; Three dimensional imaging; Tomography
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-023-43314-y

Chemical short-range order (CSRO) refers to atoms of specific elements self-organising within a disordered crystalline matrix to form particular atomic neighbourhoods. CSRO is typically characterized indirectly, using volume-averaged or through projection microscopy techniques that fail to capture the three-dimensional atomistic architectures. Here, we present a machine-learning enhanced approach to break the inherent resolution limits of atom probe tomography enabling three-dimensional imaging of multiple CSROs. We showcase our approach by addressing a long-standing question encountered in body-centred-cubic Fe-Al alloys that see anomalous property changes upon heat treatment. We use it to evidence non-statistical B 2 -CSRO instead of the generally-expected D0 3 -CSRO. We introduce quantitative correlations among annealing temperature, CSRO, and nano-hardness and electrical resistivity. Our approach is further validated on modified D0 3 -CSRO detected in Fe-Ga. The proposed strategy can be generally employed to investigate short/medium/long-range ordering phenomena in different materials and help design future high-performance materials. Quantifying chemical short-range order (CSRO) remains a formidable for volume-averaged or 2D microscopy methods. Here the authors introduce a machine-learning approach that breaks the resolution limitations of atom probe tomography to reveal the 3D atomistic architecture of CSRO in Fe-based alloys.