Magnetization Process in Sm2Fe17N3 Fine Powders Studied Using Lorentz Microscopy and Electron Holography

Magnetization Process in Sm

Fe

N

Fine Powders Studied

Using Lorentz Microscopy and Electron Holography

Ki Hyun Kim

_{, Joong Jung Kim}

_{, Daisuke Shindo}

_{, Takashi Ishikawa}

_{and Kenji Ohmori}

1

Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan

2_{Ichikawa Research Laboratory, Sumitomo Metal Mining Co., Ltd., Ichikawa 272-8588, Japan}

The magnetization process for Sm2Fe17N3 fine powders was studied by Lorentz microscopy and electron holography. Inex situ observations under the external magnetic field produced by an electromagnet, the change of the domain wall position was observed by Lorentz microscopy, while the drastic change in the distribution of lines of magnetic flux outside the fine powders was clarified by electron holography. Throughin situobservations with a piezodriving holder with localized magnetic field produced by a sharp magnetic needle, the shift of domain walls was observed in real time by Lorentz microscopy. Finally, on the basis of Lorentz microscopy observations, the magnetization process of Sm2Fe17N3fine powders was briefly discussed taking into account a hysteresis loop. [doi:10.2320/matertrans.M2010081]

(Received March 2, 2010; Accepted April 28, 2010; Published June 16, 2010)

Keywords: samarium-iron-nitrogen fine powder, magnetization process, Lorentz microscopy, electron holography

1. Introduction

Sm2Fe17N3fine powders have been studied for use as high-performance permanent magnets due to their high saturation magnetization and strong uniaxial anisotropy.1–5)_In partic-ular, Sm2Fe17N3fine powders with average sizes of 2–3mm have been used as bonded magnets. However, fine powders suffer from poor oxidation resistance and thermal instability of coercivity. Furthermore, a high compaction pressure is required for the densification of these powders to obtain high-energy products. Thus, many efforts have been made to overcome these limitations. For example, several studies have attempted to obtain high coercivity in coarse powders through minor additions of other elements.6–8)

In general, the hardening mechanism of hard magnets can be explained in terms of the domain wall pinning or the nucleation mechanism. For Sm2Fe17N3 fine powders, it is well known that the magnetization reversal mechanism that determines its coercivity is controlled by nucleation.9)_Thus, in order to understand the magnetic properties of hard magnetic materials, it is necessary to investigate their magnetization process. However, direct observations of the magnetization process in these powders have not yet been carried out, since it is very difficult to apply a sufficiently strong magnetic field to induce magnetization in these high-coercivity powders inside a transmission electron microscope (TEM).

Recently the present authors and their colleagues have demonstrated the possibility of producing localized strong magnetic fields that utilize unique specimen holders equip-ped with a sharp needle.10) _{In this study, we utilized two} devices: an electromagnet and a piezodriving holder with a sharp magnetic needle made of sintered Nd2Fe14B (see Fig. 1). The former device is useful forex situexperiment, while the latter one is efficient for in situ experiment.

Observation of the magnetization process in Sm2Fe17N3fine powders was carried out by means of Lorentz microscopy and electron holography.

2. Experimental Method

Sm2Fe17N3fine powders were fabricated by nitrogenation of the Sm2Fe17 mother alloy obtained through a reduction and diffusion (RD) process.11) The coercivity Hc and maximum energy product ðBHÞmax of this powder were 800 kA/m and 292 kJ/m3_{, respectively. It should be noted} that in this study, we used degraded Sm2Fe17N3fine powders (Hc¼336kA/m) treated in a vacuum at 290C for 1 h in order to easily induce magnetization. For transmission electron microscopy, the powder was first mixed with epoxy and then coated on the Si substrate. Using the focused ion beam (FIB) method (JEM-9310FIB), we obtained thin foil specimens of Sm2Fe17N3fine powders.

Observations of the magnetization process were carried by Lorentz microscopy and electron holography using a JEM-3000F transmission electron microscope (TEM) system equipped with a field emission gun and biprism.12) The residual magnetic field around the specimen position in the TEM was reduced to around 32 A/m by switching off and then degaussing the objective lens.12) Lorentz microscopy was carried out using the Fresnel method.13) _{The exposure} time for obtaining electron holograms was set at 6 s. Reconstructed phase images of the electron holograms were obtained using a Fourier transform of the digitized holo-grams. Ex situ experiments were carried out using an electromagnet designed to apply a magnetic field of up to 2000 kA/m to a thin foil specimen placed in the TEM holder as shown in Fig. 1(a).10)_{In these experiments, the thin foil} specimen to be observed was removed from the TEM and subjected to a strong magnetic field. The remanent state of the specimen was then observed by inserting the specimen back into the TEM. In order to observe the magnetization processin situ, experiments using Lorentz microscopy were performed by moving a sharp magnetic needle made of sintered Nd2Fe14B toward the specimen and then

withdraw-*1_{Present address: Nanostructures Research Laboratory, Japan Fine}

Ceramics Center, Nagoya 456-8587, Japan

*2_{Present address: Samsung Electronics, Hwasung City 445-701, Korea} *3_{Corresponding author, E-mail: [email protected]}

ing it as indicated in Fig. 1(b).14) _{The Nd}

2Fe14B magnetic needle was able to apply a magnetic field of 560 kA/m within the TEM.

3. Results and Discussion

Figure 2 presents the typical electron holography results and shows the distribution of magnetic flux in Sm2Fe17N3 fine powders without the external magnetic flux applied. These data were obtained from the initial state of the powders. In the TEM image shown in Fig. 2(a), the Sm2Fe17N3 fine powders mixed with epoxy are observed on a Si substrate. Figures 2(b) and 2(c) show a hologram and a reconstructed phase image, respectively. In the reconstructed phase image, the direction and density of the white lines correspond to the direction and density of lines of magnetic flux projected along the electron beam, respec-tively. Figure 2(c) clearly shows that the lines of magnetic flux of Sm2Fe17N3 fine powders are aligned along the same direction; presumably, this aligned direction is the axis of easy magnetization of these powders.

As a study of the magnetization process in the Sm2Fe17N3 fine powders, we first carried out ex situ experiments by means of Lorentz microscopy in combination with electron holography. Figures 3 and 4 show the ex situ observation results obtained for the magnetization process in the Sm2Fe17N3fine powders (powders A and B). After applying

the magnetic field to the specimen using an electromagnet, both Lorentz microscopy (Fig. 3) and electron holography (Fig. 4) were carried out. The applied magnetic field is indicated in the top-right corner. Here, in the Lorentz microscope images, the domain wall contrasts with white lines indicated by red arrows in powders A and B are noted. It can be clearly seen that the domain wall contrast of white lines in powder A disappears under the applied magnetic field of 320–400 kA/m (magnetic flux density of 0.4–0.5 T) as shown in Figs. 3(d) and 3(e). In the corresponding holog-raphy results shown in Figs. 4(d) and 4(e), the drastic change in the direction of magnetic flux lines around powder A is also observed. In addition, with regard to powder B, when the applied magnetic field is between 480 and 640 kA/m (magnetic flux density between 0.6 and 0.8 T), a similar trend, i.e., the disappearance of domain wall contrast and the change of lines of magnetic flux around powder B are observed in the results obtained by Lorentz microscopy (Figs. 3(f) and 3(g)) and electron holography (Figs. 4(f) and 4(g)), respectively. The difference in the external magnetic field which shifts the domain walls in A and B particles is attributed to their different surface morphology. Eventually these results demonstrate that magnetization process in Sm2Fe17N3 fine powders can be analyzed through the ex

situ observation techniques by observing the domain wall contrast with Lorentz microscopy and also by visualizing the distribution of lines of magnetic flux outside the fine powders with electron holography.

In order to observe the formation and movement of the domain walls, in situ experiments are performed using a piezodriving holder with a sharp magnetic needle made of Nd2Fe14B. Figure 5 shows Lorentz microscope images (captured from videotape) that exhibit the magnetization process in Sm2Fe17N3fine powders. In this experiment, the magnetic needle is moved toward and then away from the specimen. In Fig. 5, the magnetic needle first approaches from the right side of this powder as indicated by a big red arrow. The time for each frame is indicated in the bottom-right corner. From these results, we found that the positions of the magnetic domain walls, as indicated by yellow arrows, varies with the approach and withdrawal of the magnetic needle, i.e., with the increase/decrease in the applied magnetic field strength. It should be noted that magnetic

(b)

(a)

Specimen holder

TEM specimen

Fig. 1 (a) Schematic illustration of part of a specimen holder equipped with a sharp magnetic needle made of sintered Nd2Fe14B. (b) Schematic illustration of the top view of an electromagnet designed to apply a magnetic field to a thin specimen in a TEM specimen holder.

(a)

(b)

(c)

1µm 1µm 1µm

Si substrate Epoxy

domains are mainly observed at the near-surface region of this powder. On the other hand, Fig. 5(d) shows almost no domain wall contrasts indicating the fully magnetized state of this powder.

Finally we carried out the domain structure analysis with ex situ experiments utilizing both an electromagnetic and a piezodriving holder with a sharp magnetic needle. Figure 6 shows Lorentz microscope images of another fairly thin foil specimen of Sm2Fe17N3 fine powders. It is noted that the contrast of the domain wall is observed more sharply than those in Figs. 3 and 5, where only white domain wall contrasts are observed. As shown in Fig. 6(a), the white lines and black bands correspond to domain walls at an initial state (see dotted lines in inset). Here, these domain walls clearly exhibit a rounded shape. After moving the magnetic needle toward the specimen—as indicated by the big red arrow—a

shift in the magnetic domain wall is noted in Fig. 6(b): the domain walls become parallel to the direction of the approaching magnetic needle, resulting in the increase in the size of domains indicated by red arrows. On the other hand, as shown in Fig. 6(c), a strong magnetic field of 2000 kA/m was applied to this specimen using an electro-magnet, as indicated by the big blue arrow. Under this condition, magnetic domain walls are not observed due to the application of the strong external magnetic field. Thus, the Sm2Fe17N3 fine powder is believed to become completely magnetized. On the other hand, as shown in Fig. 6(d), when a magnetic field was applied to the specimen using the magnetic needle—as indicated by the big red arrow—almost no changes in the magnetic domain wall are observed.

In order to qualitatively understand the change in the magnetic domain walls of the Sm2Fe17N3 fine powder as

revealed by the Lorentz microscope images in Fig. 6, we explain the phenomenon using a schematic illustration of the hysteresis loop shown in Fig. 7. In this figure, the typical hysteresis loop of a permanent magnetic material such as Sm2Fe17N3 fine powders is indicated by red dotted lines, and the expected hysteresis loop in the observed result is indicated by black solid lines. Furthermore, the changes observed in the magnetic domain walls are marked by ‘‘I’’, ‘‘II’’, and ‘‘III’’, which correspond to the sequences in Fig. 6(a)–(d). For region I ((a)!(b)), by applying a magnetic field using a magnetic needle, the magnetization curve undergoes a slightly increase. Some domain walls are believed to become aligned along the direction parallel to the magnetic needle. Next, by applying the magnetic field along the opposite direction using an electromagnet, the

magnetization curve changes to the opposite direction, as shown in region II ((b)!(c)). Thus, the direction of magnet-ization is believed to be the same throughout the powder and parallel to the direction of the applied external magnetic field. Eventually, the specimen was found to become saturated along the opposite direction.

For region III ((c)!(d)), although a magnetic field was applied to the specimen using the magnetic needle, the domain wall was not changed. It is considered that the applied magnetic field is not large enough to change the magnetic domain of the specimen because of the large distance from the specimen to the magnetic needle. Even-tually, it can be expected that when the magnetic field is removed, the magnetization curve reverts to around its previous condition.

4. Conclusions

The magnetization process for Sm2Fe17N3 fine powders, studied by Lorentz microscopy and electron holography, can be summarized as follows.

(1) Utilizing an electromagnet,ex situobservations on the magnetization process are carried out. While the change of the domain wall contrast is observed by Lorentz microscopy, the drastic change in the distribution of lines of magnetic flux outside the fine powders is clarified through electron holography.

(2) Throughin situobservations with a piezodriving holder with a sharp magnetic needle, the strong magnetic field can be produced near the specimen, and the shift of domain walls and the magnetization process are observed in real time by Lorentz microscopy.

(3) On the basis of Lorentz microscopy ex situ observa-tions, the magnetization process of Sm2Fe17N3fine powders is briefly discussed with a hysteresis loop.

4.57s

500nm

4.60s 5.00s

Magnetic needle

5.27s

Magnetic needle

11.77s 12.00s

(a)

(b)

(c)

(d)

(e)

(f)

Fig. 5 Lorentz microscope images showing the motion of the magnetic domain walls in the Sm2Fe17N3fine powders. The magnetic domain walls are indicated by yellow arrows. The time of each frame is indicated at the bottom-right corner of each figure. The direction of approach of magnetic needle is indicted by a red arrow.

1µm

(a) (b)

(c) (d)

Fig. 6 Lorentz microscope images showing the shift of the magnetic domain wall in Sm2Fe17N3fine powder. (a) Initial state. (b) State with the shift of domain wall. (c) Fully magnetized state after application of a magnetic field of 2000 kA/m using an electromagnet. (d) Magnetic domain wall image after applying the magnetic field using magnetic needle. The insets in (a)–(d) show schematic illustrations of the domain wall (black dotted lines) and magnetization directions (red and blue arrows). The big red and blue arrows indicate the direction of the applied external magnetic field.

I: (a) (b)

II: (b) (c) Magnetization

(B)

III: (c) (d)

Applied field (H) a

b

c

•

• •

•d

Acknowledgements

The authors are grateful to Y. Murakami and Z. Akase for their useful discussion and assistance. This study was partly supported financially by a Grant-in-Aid for Scientific Research (No. 19106002) from the Japan Society for the Promotion of Science, Global COE Program ‘‘Materials Integration (International Center of Education and Research), Tohoku University’’, MEXT, Japan and the special education and research expenses from Post-Silicon Materials and Devices Research Alliance.

REFERENCES

1) J. Ye, Y. Liu, M. Li, D. Song, S. Gao and M. Tu: Rare Metals25(2006) 568–571.

2) X. C. Kou: J. Alloy. Compd.281(1998) 41–45.

3) T. Iriyama, K. Kobayashi, N. Imakoa, T. Fukuda, H. Kato and Y. Nakagawa: IEEE Trans. Magn.28(1992) 2326–2331.

4) S. Suzuki, T. Miura and M. Kawasaki: IEEE Trans. Magn.29(1993) 2815–2817.

5) K. Majima, M. Ito, S. Katsuyama and H. Nagai: J. Appl. Phys.81 (1997) 4530–4532.

6) N. Imaoka, T. Iriyam, S. Itoh, A. Okamoto and T. Katsumata: J. Alloy. Compd.222(1995) 73–77.

7) K. Zuzek, P. J. McGuiness and S. Kobe: J. Alloy. Compd.289(1999) 265–269.

8) W. Kaszuwara, M. Leonowicz and J. A. Kozubowsk: Mater. Lett.42 (2000) 383–386.

9) X. C. Kou, W. J. Qiang, H. Kronmu¨ller and L. Schultz: J. Appl. Phys. 74(1993) 6791–6797.

10) D. Shindo, Y.-G. Park, Y. Gao and H. S. Park: J. Appl. Phys.95(2004) 6521–6526.

11) T. Iseki, T. Ishikawa, A. Kawamoto and K. Ohmori: J. Jpn. Soc. Powder Powder Metall. (in Japanese)50(2003) 633–637.

12) D. Shindo, Y.-G. Park, Y. Murakami, Y. Gao, H. Kanekiyo and S. Hirosawa: Scr. Mater.48(2003) 851–856.

13) D. Shindo and T. Oikawa: Analytical Electron Microscopy for Materials Science, (Springer, Tokyo, 2002) 111–125.