Relationship between TiB2 agglomerate size and grain refinement effectiveness of Al-Ti-B grain refiner

• Published in: Journal of Japan Institute of Light Metals Vol. 71; no. 9; pp. 409 - 414
• Main Authors: Minagawa, Akihiro, Furuta, Shogo, Omura, Naoki, Murakami, Yuichiro, Tsunekawa, Masanori
• Format: Journal Article
• Language: Japanese; English
• Published: Tokyo The Japan Institute of Light Metals 15.09.2021; Japan Science and Technology Agency
• Subjects: Agglomerates; Agglomeration; Aluminum; Casting defects; Computed tomography; Crystal defects; DC casting; Direct chill casting; Grain boundaries; Grain refinement; grain refiner; Grain size; grain size prediction model; Grain structure; Lager; Mathematical models; Prediction models; Predictions; Size distribution; TiB2 agglomerate; Titanium diboride
• ISSN: 0451-5994; 1880-8018
• DOI: 10.2464/jilm.71.409

In order to reduce the risk of the casting defects, fine grains are required in the aluminum DC casting process. It is commonplace to use the grain refiner as a means of obtaining the desired grain structure. Typical grain refiners are Al-Ti-B system. In this study, TiB2 agglomerates in Al-Ti-B grain refiner were observed by multiscale X-ray computed tomography (CT), which has Micro-CT and Nano-CT. The grain size of cast sample with grain refiner was calculated by grain size prediction model with TiB2 agglomerate size distribution. The predicted grain size and measured grain size showed good agreement in case that assuming free growth will initiate on larger TiB2 agglomerate prior. Additionally, the TiB2 agglomerates at the center or boundary of α-Al grains were observed by multiscale X-ray CT. The TiB2 agglomerates at the center of grain were larger than these at the grain boundary. Therefore, the assumption of free growth will initiate on lager TiB2 agglomerates was estimated to be valid. From these results, it was estimated that the grain refinement effectiveness of grain refiner was varied by the TiB2 agglomerate size distribution, and it can be calculated by grain size prediction model with TiB2 agglomerate size distribution.