Presentation Information
[P1-16]Magnetization reversal of core-shell structured grain of GBDP Nd-Fe-B sintered magnet
Minghe Liu1, Shuai Cao1, Xinlong Yang1, Wenhe Liu1, *Weixing Xia1, Renjie Chen1, Aru Yan1, Wei Li1 (1. Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences (China))
Keywords:
GBDP Nd-Fe-B sintered magnet,Magnetization reversal,Magnetic domain,In-situ observation,Lorentz TEM
The GBDP sintered Nd-Fe-B magnet exhibits improved magnetic properties without excessive use of expensive heavy rare earths such as Dy and Tb. The Dy- or Tb-rich shell, which possesses a high anisotropy field, along with an improved grain boundary, has been attributed to the increased coercivity and minimal degradation of remanent magnetization [1]. However, the mechanism behind coercivity enhancement remains not fully understood. Various experimental techniques, including Kerr microscopy, MFM, and Lorentz TEM, have been employed to observe the magnetization reversal process [2,3]. Most observations conducted under varying magnetic fields and temperatures focus on domain evolution on the surface perpendicular to the easy axis. However, the demagnetization field significantly alters the domain structure, making the observed results not directly correlated with the magnet's magnetic properties. In this study, utilizing a self-developed strong in-plane magnetic field holder in Lorentz TEM [4], we observed core-shell structured grains along the plane of the easy axis, obtaining direct experimental insights into the magnetization reversal process of the magnet.
GBDP Nd-Fe-B was prepared by diffusing DyH3 into an N55 commercial Nd-Fe-B magnet. This process increased the coercivity from 14.2 kOe to 21.8 kOe. A TEM sample was cut using FIB at a depth of 40 μm from the diffusion surface. Fig. 1(a) presents the SEM image of the TEM sample, where grain ⑥, located at the center, exhibits a full core-shell structure. It is surrounded by some RE-rich phases and six uncomplete grains, the remaining portions after FIB etching. Fig. 1(b) shows the thermally demagnetized state, revealing a multi-domain structure with domain walls spanning across all grains except for grain ③. The domain walls deflect within the grains along their easy axes. As the applied field increased slightly, the domain walls shifted and merged rapidly, bringing the sample to a saturated state—an essential characteristic of nucleation-type magnetic domains. When an opposing field was applied, as indicated by the long red arrows in Fig. 1, magnetization reversal occurred grain by grain. Fig. 1(c)-(f) are four frames from a continuous video, before which grains ①–③ had already reversed their magnetization. The magnetization reversal is evident as an instantaneous flip in diffraction contrast, clearly observable in the video. Another notable feature is the contrast change at the grain boundaries, as exemplified in Fig. 1(c) and 1(d) when magnetization reversal occurred in grain ④. The grain boundary, marked by white arrows in Fig. 1(c), became “black” in Fig. 1(d), which is attributed to changes in Lorentz forces between magnetization of grains ④ and ⑥, as well as ④ and ⑦. The directions of the magnetization are indicated by yellow and red arrows. As the field increased to 1.21 T and 1.29 T, magnetization reversal was observed in grains ⑤ and ⑥, as shown in Fig. 1(e) and 1(f), respectively. It is worth noting that the contrast of grain boundaries indicated by white arrows in Fig. 1(f), becomes almost identical in Fig. 1(c), which is the result of magnetization reversals in all three grains ④-⑥.
In this experiment, we observed that magnetization reversals occurred first in the surrounding grains, followed by the central grain. This is in anticipation as the defected regions of the surrounding grains have lower anisotropy fields. The exception of grain ⑦ is likely due to its thick Dy-rich shell layer. More importantly, we found that diffraction contrast jumped only seven times, and no domain walls like those in Fig. 1(b) were observed during the process. This indicates that the magnetization reversal of each grain happens simultaneously, with no distinct sequence between the core and shell regions. Although micromagnetic simulations suggested that demagnetization could nucleate in either the core or shell first, this phenomenon was not observed within the frame rate of 30 fps (33.3 ms per image) in the video, which suggests strong exchange coupling between the core and shell within individual grains. The one-by-one reversal of grains demonstrates good magnetic decoupling by the grain boundaries. Magnetization reversal at high temperatures will also be reported at the conference.
References
[1] Sepehri-Amin, T. Ohkubo, K. Hono, Acta Materialia 61 (2013) 1982.
[2] Tim Helbig, Konrad Loewe, Simon Sawatzki, Min Yi, Bai-Xiang Xu, Oliver Gutfleisch, Acta Materialia 127 (2017) 498.
[3] Yuqing Li, Ming Yue, Weixing Xia, et al., Journal of Alloys and Compounds 906 (2022) 164414.
[4] Tian Bai, Weixing Xia, Renjie Chen, Aru Yan, Wei Li, et al., Ultramicroscopy 260 (2024) 113950.
GBDP Nd-Fe-B was prepared by diffusing DyH3 into an N55 commercial Nd-Fe-B magnet. This process increased the coercivity from 14.2 kOe to 21.8 kOe. A TEM sample was cut using FIB at a depth of 40 μm from the diffusion surface. Fig. 1(a) presents the SEM image of the TEM sample, where grain ⑥, located at the center, exhibits a full core-shell structure. It is surrounded by some RE-rich phases and six uncomplete grains, the remaining portions after FIB etching. Fig. 1(b) shows the thermally demagnetized state, revealing a multi-domain structure with domain walls spanning across all grains except for grain ③. The domain walls deflect within the grains along their easy axes. As the applied field increased slightly, the domain walls shifted and merged rapidly, bringing the sample to a saturated state—an essential characteristic of nucleation-type magnetic domains. When an opposing field was applied, as indicated by the long red arrows in Fig. 1, magnetization reversal occurred grain by grain. Fig. 1(c)-(f) are four frames from a continuous video, before which grains ①–③ had already reversed their magnetization. The magnetization reversal is evident as an instantaneous flip in diffraction contrast, clearly observable in the video. Another notable feature is the contrast change at the grain boundaries, as exemplified in Fig. 1(c) and 1(d) when magnetization reversal occurred in grain ④. The grain boundary, marked by white arrows in Fig. 1(c), became “black” in Fig. 1(d), which is attributed to changes in Lorentz forces between magnetization of grains ④ and ⑥, as well as ④ and ⑦. The directions of the magnetization are indicated by yellow and red arrows. As the field increased to 1.21 T and 1.29 T, magnetization reversal was observed in grains ⑤ and ⑥, as shown in Fig. 1(e) and 1(f), respectively. It is worth noting that the contrast of grain boundaries indicated by white arrows in Fig. 1(f), becomes almost identical in Fig. 1(c), which is the result of magnetization reversals in all three grains ④-⑥.
In this experiment, we observed that magnetization reversals occurred first in the surrounding grains, followed by the central grain. This is in anticipation as the defected regions of the surrounding grains have lower anisotropy fields. The exception of grain ⑦ is likely due to its thick Dy-rich shell layer. More importantly, we found that diffraction contrast jumped only seven times, and no domain walls like those in Fig. 1(b) were observed during the process. This indicates that the magnetization reversal of each grain happens simultaneously, with no distinct sequence between the core and shell regions. Although micromagnetic simulations suggested that demagnetization could nucleate in either the core or shell first, this phenomenon was not observed within the frame rate of 30 fps (33.3 ms per image) in the video, which suggests strong exchange coupling between the core and shell within individual grains. The one-by-one reversal of grains demonstrates good magnetic decoupling by the grain boundaries. Magnetization reversal at high temperatures will also be reported at the conference.
References
[1] Sepehri-Amin, T. Ohkubo, K. Hono, Acta Materialia 61 (2013) 1982.
[2] Tim Helbig, Konrad Loewe, Simon Sawatzki, Min Yi, Bai-Xiang Xu, Oliver Gutfleisch, Acta Materialia 127 (2017) 498.
[3] Yuqing Li, Ming Yue, Weixing Xia, et al., Journal of Alloys and Compounds 906 (2022) 164414.
[4] Tian Bai, Weixing Xia, Renjie Chen, Aru Yan, Wei Li, et al., Ultramicroscopy 260 (2024) 113950.
