Type of presentation: Poster

MS-12-P-1511 Determination of magnetic flux density of grain boundary phase in Nd-Fe-B permanent magnets

Murakami Y.1,2, Tanigaki T.2, Sasaki T.3, Takeno Y.1, Park H. S.2, Matsuda T.4, Ohkubo T.3, Hono K.3, Shindo D.1,2
1Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai, Japan, 2Center for Emergent Matter Science (CEMS), RIKEN, Wako, Japan, 3National Institute for Materials Science, Tsukuba, Japan, 4Japan Science and Technology Agency, Kawaguchi, Japan
murakami@tagen.tohoku.ac.jp

Nanostructure optimization is crucial for enhancing the coercivity of Nd-Fe-B permanent magnets, which allow a significant degree of miniaturization of electric parts because of the large energy product. Regarding the materials science/engineering of Nd-Fe-B magnets, an essential problem is to understand the magnetism of the ultrathin grain boundary (GB) phase which envelopes the individual crystal grains of Nd2Fe14B [1]. However, revealing the magnetism in the GB region (~3 nm in width) remains challenging. Here, we used electron holography to determine the magnetic flux density in the GB phase using a sintered magnet subjected to optimal annealing.

As schematically shown in Fig. 1(a), which represents the cross section of a thin-foil specimen, the phase shift of electrons was measured in the line connecting the points R and S. The thin GB phase was tilted away from the direction of incident electrons: i.e., the trace of GB phase (WGB) was approximately 110 nm in this experimental setup. The magnetic flux density in the GB phase (BGB) can be determined from Δφ, which represents the deviation of the observed phase shift (in the GB area) from the extrapolated curve assuming the absence of the GB phase [Fig. 1(b)].

Figure 1(c) shows a transmission electron microscope image of the thin-foil specimen, which contained five Nd2Fe14B grains, A-E. Using the reconstructed phase image such as shown in Fig. 1(d), we accurately measured the phase shift in the R-S line which crossed the GB (shown in yellow) between Nd2Fe14B grains A and B. The observations are plotted with light-blue dots in Fig. 2(a). Split-illumination electron holography [2] achieved the sufficient precision for measuring the phase shift of ±0.08 rad. To obtain the phase information due to the GB phase, curve fitting was carried out for area A (containing only grain A); refer to the red curve in Fig. 2(a). The deviation Δφ was plotted as a function of position along the R-S line, as shown in Fig. 2(b). The value of Δφ continued to decrease over the GB area, reaching a minimum of −0.34 rad at the border of this area. Following the simulations of Δφ, the value of BGB that explains well the observation (i.e., −0.34 rad, at the border of GB area) is ~1.0 T [Fig. 2(c)]. The result explicitly indicates that the GB phase is ferromagnetic, contrary to the traditional understanding. Our observation implies significant exchange coupling between Nd2Fe14B grains, which explains the avalanche-propagation of magnetization reversal observed in sintered magnets.

References:

[1] H. Sepehri-Amin et al., Acta Mater. 60 (2012) 819.

[2] T. Tanigaki et al., Appl. Phys. Lett. 101 (2012) 043101.


This study was supported by grants “Funding Program for World-Leading Innovative R&D on Science and Technology (FIRST Program)” and “JST-CREST”.

Fig. 1: (a) Cross-sectional view of thin-foil specimen containing Nd2Fe14B grains (A and B), and a thin GB phase. (b) Phase shift (schematic representation) observed in the R-S line. (c) TEM image of the thin-foil specimen. (d) Reconstructed phase image of the rectangle area shown in (c). The red allows indicate the direction of magnetic flux.

Fig. 2: (a) Phase shift observed along the R-S line shown in Fig. 1(c). (b) Difference between the observations and the fitting curve, Δφ, which determined the phase shift due to the GB phase to be −0.34 rad at the border of area GB. (c) Comparison between observations and calculations (simulations) of Δφ.