- Email: [email protected]
In-situ Lorentz microscopy of Fe85Si2B8P4Cu1 nanocrystalline soft magnetic alloys
Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Q1 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 Q2 62 63 64 65 66
Contents lists available at ScienceDirect
Journal of Magnetism and Magnetic Materials journal homepage: www.elsevier.com/locate/jmmm
In-situ Lorentz microscopy of Fe85Si2B8P4Cu1 nanocrystalline soft magnetic alloys Zentaro Akase a,b,n, Shinji Aizawa b, Daisuke Shindo a,b, Parmnand Sharma c, Akihiro Makino c,d a
Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan Center for Emergent Matter Science, RIKEN, Wako, Saitama 351-0198, Japan c Research and Development Center for Ultra High Efficiency Nano-crystalline Soft Magnetic Materials, Institute for Materials Research, Tohoku University, Sendai 980-8577, Japan d Institute for Materials Research, Tohoku University, Sendai 980-8577, Japan b
art ic l e i nf o
a b s t r a c t
Article history: Received 30 April 2014 Received in revised form 30 June 2014
Microstructure dependence of magnetic properties of soft magnetic Fe–Si–B–P–Cu nanocrystalline alloys were studied by using in-situ Lorentz microscopy in a transmission electron microscope equipped with a magnetizing system. In particular, we investigated in detail motion of magnetic domain walls in heattreated Fe85Si12B6P4Cu1 amorphous ribbons. Smooth motion of domain walls was observed for the optimally heat-treated (at 430 1C) nano-crystalline alloy. Pinning of domain walls was observed for higher-temperature-heat-treated (470 1C) ribbons. Both ribbons showed a nanocrystalline structure containing α-Fe crystallites of about 15 nm in size. Electron diffraction patterns indicated that the higher-temperature-heat-treated samples contained boride precipitates, which is considered to cause less smooth domain wall motion. & 2014 Elsevier B.V. All rights reserved.
Keywords: Nanocrystalline soft magnetic alloy Magnetic domain-wall motion Lorentz microscopy
1. Introduction Soft magnetic materials are used as magnetic cores in a variety of applications such as motors, transformers, inductors, sensors, actuators, power circuits, and electronic communication devices [1,2]. Electrical energy loss, which manifests as heat, is inherent in materials used in these applications. Metallic alloys in amorphous and/or nano-crystalline states exhibit low magnetic core loss (W) [1,2]. In addition to low W, soft magnetic materials are required to have a high saturation magnetic flux density (Bs) for device miniaturization. Unfortunately, there are very few amorphous/ nanocrystalline materials that have at the same time W lower than that of grain oriented steel and Bs close to that of the grain oriented steel. Recently, Fe-rich Fe–Si–B–P–Cu nanocrystalline alloys have been reported to have these necessary characteristics [3–5]. Therefore, these alloys have large application potential in saving electrical energy in various devices. The Fe–Si–B–P–Cu alloys have a uniform nanocrystalline structure composed of high density α-Fe grains with the size less than 20 nm surrounded by remaining amorphous matrix. A uniform
n Corresponding author at: Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan. E-mail address: [email protected] (Z. Akase).
nanocrystalline structure is obtained after an optimum heattreatment of an as-quenched amorphous alloy. Sharma et al. [5] reported that heat-treatment at temperatures lower than the optimum temperature results in a significantly larger volume of remaining amorphous phase, whereas higher temperature heattreatment leads to crystallization of iron-boride phases. Magnetic properties of nano-crystalline alloys strongly depend on the type of phases (crystalline and amorphous), distribution, and their volume fraction. High density α-Fe nanocrystals are required for strong magnetic exchange coupling, which is necessary to average out the magnetocrystalline anisotropy of α-Fe [1]. A larger separation between α-Fe grains i.e. relatively more volume fraction of amorphous phase, leads to indirect exchange coupling among αFe grains (through the amorphous matrix), and affect the magnetic properties. Similarly appearance of a minor amount of iron-boride phase can significantly increase the coercivity (Hc) [6]. In soft magnetic materials coercivity is mainly governed by the motion of domain walls, and their propagation is affected by the presence of inhomogeneities, which can be amorphous/iron boride phases. Therefore, it is important to study motion of domain walls in the newly developed Fe–Si–B–P–Cu alloys heat-treated at different temperatures. The Fe85Si2B8P4Cu1 alloys were selected because of good magnetic properties (Hc 6 A/m and Bs 1.85 T) and an ability to produce as-quenched X-ray amorphous ribbons [5]. Motion of magnetic domain walls was studied by using in-situ Lorentz
http://dx.doi.org/10.1016/j.jmmm.2014.08.101 0304-8853/& 2014 Elsevier B.V. All rights reserved.
Please cite this article as: Z. Akase, et al., Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j. jmmm.2014.08.101i
67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96
Z. Akase et al. / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 Q4
microscopy in a transmission electron microscope (TEM) equipped with a magnetizing system [7–9] in order to clarify the dependence of domain walls motion on microstructure.
2. Materials and methods The Fe85Si2B8P4Cu1 ribbons (width 4–6 mm and thickness 20 μm) were made by high frequency induction melting (in vacuum) and melt-spinning (in air) techniques. Ribbons were heat-treated in an infra-red furnace under Ar-gas flow condition. Heating rate to achieve the required heat-treatment temperature was 400 1C/min. A magnetic flux density (B) versus magnetic field (H) loop analyzer was used to measure the magnetic properties. Three samples were selected based on their heat-treatment temperature and magnetic properties: Sample 1 is heat-treated at 330 1C, which is lower than the optimum heat-treatment temperature. Sample 2 is heat-treated under the optimum conditions, i.e., at 430 1C. Sample 3 is heat-treated at a higher temperature ( 470 1C) than the optimum temperature. The coercivities (Hc) of these samples were as follows: Sample 1: 21.2 A/m; Sample 2: 6.29 A/m; Sample 3: 259 A/m. The Hc of the as-quenched amorphous ribbon was 17 A/m. The TEM samples were prepared using a JIB-4500Multibeam focused ion beam (FIB) system. They were crosssectioned in the direction of the thickness of the ribbons with dimensions of 10 μm wide, 2–5 μm long and 0.1 μm thick. These rectangular samples were attached to a collodion film of TEM meshgrids by using a micro-manipulator with a glass probe. For conventional TEM observations and Lorentz microscope observations, we used HF-3300S (Hitachi High-Technologies). For in-situ Lorentz microscopy observations with a static or alternating external magnetic fields, we used JEM-3000F (JEOL) equipped with a magnetizing specimen holder, which can apply a magnetic field horizontal to the sample [7–9].
3. Results and discussion 3.1. Microstructures Fig. 1 shows bright field images (a–c) and selected area electron diffraction patterns (d–f) for Fe85Si2B8P4Cu1 ribbons: Sample 1
(330 1C heat-treated), Sample 2 (430 1C heat-treated), and Sample 3 (470 1C heat-treated). Sample 1 has randomly oriented α-Fe grains of size 20–40 nm sparsely distributed in the remaining amorphous matrix (Fig. 1a and d). The α-Fe grain size decreases significantly at higher-temperature heat-treatment. Dense and uniform distributions of α-Fe grains with the size of 15 nm are noticeable in the TEM images of Sample 2 and Sample 3 (Fig. 1b and c). Such a large difference in grain sizes is due to changes in the mechanism of crystallization at different heat-treatment temperatures. Grain nucleation density, which is dependent on heat-treatment temperature, governs the grain size. The diffraction patterns of all the heat-treated ribbons (Figs. 1d–f) shows Debye–Scherrer rings corresponding to α-Fe nano-crystallites. Additional diffraction spots can be noticed for Sample 3 (Fig. 1f, marked with red circles). These spots can be indexed as 112, 202 and 310 reflections from Fe2B crystals. Low Hc ( 6 A/m) for Sample 2 (430 1C heat-treated) is due to suppression of magneto-crystalline anisotropy (K1) of α-Fe grains through magnetic exchange interactions [6]. The necessary requirement for this to happen is that the grain size should be smaller than the ferromagnetic exchange length, which is about 20–40 nm for Fe-based alloys. The grain size of α-Fe for Sample 3 (470 1C heat-treated) is approximately 15 nm, which is similar to that of Sample 2 (430 1C heat-treated). Even though the grain size is similar to each other, Hc is very different (6.29 A/m for Sample 2; 259 A/m for Sample 3). This may be due to the presence of Fe2B grains. Huge magnetocrystalline anisotropy of Fe2B (430 kJ/m3, more than 10 times as large as that of α-Fe) may be responsible for this. The presence of high anisotropy grains in a nanocrystalline alloy can influence motion of domain walls, which then can be pinned by Fe2B grains, and can lead to an increase in Hc.
3.2. Magnetic domain structure and motion of magnetic domain walls The structural characterization confirmed that Sample 1 (heattreated at 330 1C), Sample 2 (heat-treated at 430 1C) and Sample 3 (heat-treated at 470 1C) are very different from each other in terms of microstructure, which is also reflected in their coercivity values. Fig. 2 shows magnetic domain walls in these ribbon samples imaged by Lorentz microscopy, where magnetic domain walls
Fig. 1. Bright field images (a–c) and selected area diffraction patterns (d–f) obtained from heat-treated Fe85Si12B6P4Cu1 amorphous-ribbons. The heat-treatment temperatures are 330 1C, 430 1C, and 470 1C. (For interpretation of the references to color in this figure the reader is referred to the web version of this article.)
Please cite this article as: Z. Akase, et al., Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j. jmmm.2014.08.101i
67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132
Z. Akase et al. / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
3
Fig. 2. Lorentz micrographs of heat-treated Fe85Si12B6P4Cu1 samples. The heat-treatment temperatures are 330 1C, 430 1C, and 470 1C.
Fig. 3. Snapshots of Lorentz micrographs of the heat-treated Fe85Si12B6P4Cu1 samples in an AC magnetic field H at 0.5 Hz, (a) H¼ 2.5 kA/m and (b) H¼ 2.0 kA/m from video recording for Sample 2 (430 1C heat-treated) and Sample 3 (470 1C heattreated). The direction of the applied magnetic field is indicated by the arrow at the top right.
appear as dark and bright line contrasts. All the samples exhibited a closure domain structures along the long axis of the sample. The domain structure depends on the size and the shape of the TEM samples, while the smoothness of the domain walls and their motion in each TEM sample depends on magnetic properties, as discussed in detail in the following. Note that the shape of domain walls in Sample 3 (470 1C heat-treated) is not as straight as for that of other samples. Kinks in domain walls can originate from pinning. Without pinning, domain walls will move smoothly with the increase in external magnetic field. To study motion of domain walls in the heat-treated samples, an alternating magnetic field was applied. Although Sample 2 (heat-treated at 430 1C) and Sample 3 (heat-treated at 470 1C) have a similar grain size of α-Fe, there is a large difference in coercivity Hc (6.29 A/m for Sample 2; 259 A/m for Sample 3). Therefore we investigated the difference in relation to the domain structure in more detail. Motion of domain wall was recorded from pictures on a small fluorescent screen using a conventional video camera. Fig. 3a and b shows the Lorentz micrographs of Sample 2 (430 1C heat-treated) and Sample 3 (470 1C heat-treated). These micrographs were obtained from snapshots of the recorded video. The original video files are included as a supplementary materials in this paper. Video 1 and Video 2 correspond to Fig. 3a and b, respectively. The frequency and amplitude of alternating magnetic field were 0.5 Hz and 2.5 kA/m for Fig. 3a (Sample 2), and 0.5 Hz and 2.0 kA/m for Fig. 3b (Sample 3). The direction of applied
Fig. 4. Lorentz micrographs of the specimen heat-treated at 470 1C under a static external magnetic field H of (a) 3.8 kA/m, (b) 4.3 kA/m, (c) 4.6 kA/m, and (d) 4.8 kA/m. The direction of the external magnetic field is indicated by the arrow at the top right. The right column shows magnified images of the corresponding rectangular regions on the left column. The positions of the magnetic domain wall and its positions at lower magnetic fields are indicated by dotted lines.
magnetic field is indicated by the arrow shown at the top of Fig. 3. Straight and smooth motion of magnetic domain walls were observed in Sample 2, while irregular motion of domain walls were observed in Sample 3. The irregular motion may result from the presence of Fe2B grains, because they have higher magnetocrystalline anisotropy than that of α-Fe.
Please cite this article as: Z. Akase, et al., Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j. jmmm.2014.08.101i
67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132
Z. Akase et al. / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎
4
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
Supplementary material related to this article can be found online at http://dx.doi.org/10.1016/j.jmmm.2014.08.101. In order to gain better understanding of domain wall motion in Sample 3, we observed domain wall behaviors by increasing static magnetic field H. The static magnetic field was increased continuously until the domain structure changed. A charge-coupled device (CCD) camera was used to obtain images of domain walls. Fig. 4a–d shows the Lorentz micrographs under a static external magnetic fields of 3.8 kA/m (Fig. 4a), 4.3 kA/m (Fig. 4b), 4.6 kA/m (Fig. 4c), and 4.7 kA/m (Fig. 4d). The direction of the magnetic field is indicated by the arrow shown at the top right of Fig. 4. Data on the left column show low-magnification images, whereas magnified images of the rectangular regions marked in the left column are shown on the right column. The position of the magnetic domain wall is marked by dotted lines. To understand the effects of magnetic field on the motion of domain walls, domain wall positions at smaller magnetic fields are also shown in the images. From Fig. 4, we find that the motion of domain walls is not smooth, in particular it is pinned in the area marked with a white circle in Fig. 4c. We conjecture that pinning of domain walls appeared at this region, where the magnetic flux density fluctuated due to the presence of Fe2B precipitates. 4. Conclusions Microstructures were observed on heat-treated Fe85Si2B8P4Cu1 amorphous ribbon samples by using transmission electron microscope. Results show that (1) Sample 2 (heat-treated at 430 1C) and Sample 3 (heat-treated at 470 1C) have a nano-crystalline structure with a grain size of 15 nm, (2) Sample 1 (heat-treated at 330 1C) has a larger grain size (20–40 nm), and (3) Sample 3 (heat-treated at 470 1C) has precipitates of Fe2B in addition to α-Fe grains. Domain walls were imaged in-situ by Lorentz microscopy by varying the external magnetic field H. Domain walls move smoothly in an as-quenched ribbon sample as well as in Sample 1 (330 1C heat-treated) and Sample 2 (430 1C heat-treated), while domain walls move less smoothly in Sample 3 (470 1C heattreated). Domain walls are pinned at highly anisotropic Fe2B
precipitates. We conjecture that the differences in coercivity between Sample 2 (430 1C heat-treated) and Sample 3 (470 1C heat-treated) is due to less smooth motion and/or domain wall pinning at Fe2B precipitates.
Acknowledgement We would like to thank Dr. Y. A. Ono for critical reading of the manuscript and for comments and suggestions. This work was supported by the “Tohoku Innovative Materials Technology Initiatives for Reconstruction (TIMT)” funded by the Ministry of Education, Culture, Sports, Science and Technology (MEXT) and Reconstruction Agency, Japan. This research was partly supported by the grant from the Japan Society for the Promotion of Science (JSPS) through the ‘“Funding Program for World-Leading innovative R&D on Science and Technology (FIRST program)” initiated by the Council for Science and Technology Policy (CSTP). References [1] G. Herzer, Modern soft magnets: amorphous and nanocrystalline materials, Acta Mater. 61 (2013) 718–734. [2] R. Hasegawa, Applications of amorphous magnetic alloys in electronic devices,, J. Non-cryst. Solids 287 (2001) 405–412. [3] A. Makino, Nanocrystalline soft magnetic Fe–Si–B–P–Cu alloys with high B of 1.8–1.9 T contributable to energy saving, IEEE Trans. Magn. 48 (2012) 1331–1335. [4] Z.Q. Zhang, P. Sharma, A. Makino, Role of Si in high Bs and low core-loss Fe85.2B10 XP4Cu0.8SiX nano-crystalline alloys, J. Appl. Phys. 112 (2012) 103902. [5] P. Sharma, X. Zhang, Y. Zhang, A. Makino, Influence of microstructure on soft magnetic properties of low coreloss and high Bs Fe85Si2B8P4Cu1 nanocrystalline alloy, J. Appl. Phys. 115 (2014) 17A340. [6] G. Herzer, Anisotropies in soft magnetic nanocrystalline alloys, J. Magn. Magn. Mater. 294 (2005) 99–106. [7] Z. Akase, D. Shindo, In situ Lorentz microscopy in an alternating magnetic field, J. Electron Microsc. 59 (2010) 207–213. [8] Z. Akase, D. Shindo, M. Inoue, A. Taniyama, Lorentz microscopic observations of electrical steel sheets under an alternating current magnetic field, Mater. Trans. 48 (2007) 2626–2630. [9] M. Inoue, T. Tomita, M. Naruse, Z. Akase, Y. Murakami, D. Shindo, Development of a magnetizing stage for in situ observations with electron holography and Lorentz microscopy, J. Electron Microsc. 54 (2005) 509–513.
Please cite this article as: Z. Akase, et al., Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j. jmmm.2014.08.101i
41 42 43 44 45 46 47 48 49 50 51 Q3 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Q1 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 Q2 62 63 64 65 66
Contents lists available at ScienceDirect
Journal of Magnetism and Magnetic Materials journal homepage: www.elsevier.com/locate/jmmm
In-situ Lorentz microscopy of Fe85Si2B8P4Cu1 nanocrystalline soft magnetic alloys Zentaro Akase a,b,n, Shinji Aizawa b, Daisuke Shindo a,b, Parmnand Sharma c, Akihiro Makino c,d a
Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan Center for Emergent Matter Science, RIKEN, Wako, Saitama 351-0198, Japan c Research and Development Center for Ultra High Efficiency Nano-crystalline Soft Magnetic Materials, Institute for Materials Research, Tohoku University, Sendai 980-8577, Japan d Institute for Materials Research, Tohoku University, Sendai 980-8577, Japan b
art ic l e i nf o
a b s t r a c t
Article history: Received 30 April 2014 Received in revised form 30 June 2014
Microstructure dependence of magnetic properties of soft magnetic Fe–Si–B–P–Cu nanocrystalline alloys were studied by using in-situ Lorentz microscopy in a transmission electron microscope equipped with a magnetizing system. In particular, we investigated in detail motion of magnetic domain walls in heattreated Fe85Si12B6P4Cu1 amorphous ribbons. Smooth motion of domain walls was observed for the optimally heat-treated (at 430 1C) nano-crystalline alloy. Pinning of domain walls was observed for higher-temperature-heat-treated (470 1C) ribbons. Both ribbons showed a nanocrystalline structure containing α-Fe crystallites of about 15 nm in size. Electron diffraction patterns indicated that the higher-temperature-heat-treated samples contained boride precipitates, which is considered to cause less smooth domain wall motion. & 2014 Elsevier B.V. All rights reserved.
Keywords: Nanocrystalline soft magnetic alloy Magnetic domain-wall motion Lorentz microscopy
1. Introduction Soft magnetic materials are used as magnetic cores in a variety of applications such as motors, transformers, inductors, sensors, actuators, power circuits, and electronic communication devices [1,2]. Electrical energy loss, which manifests as heat, is inherent in materials used in these applications. Metallic alloys in amorphous and/or nano-crystalline states exhibit low magnetic core loss (W) [1,2]. In addition to low W, soft magnetic materials are required to have a high saturation magnetic flux density (Bs) for device miniaturization. Unfortunately, there are very few amorphous/ nanocrystalline materials that have at the same time W lower than that of grain oriented steel and Bs close to that of the grain oriented steel. Recently, Fe-rich Fe–Si–B–P–Cu nanocrystalline alloys have been reported to have these necessary characteristics [3–5]. Therefore, these alloys have large application potential in saving electrical energy in various devices. The Fe–Si–B–P–Cu alloys have a uniform nanocrystalline structure composed of high density α-Fe grains with the size less than 20 nm surrounded by remaining amorphous matrix. A uniform
n Corresponding author at: Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan. E-mail address: [email protected] (Z. Akase).
nanocrystalline structure is obtained after an optimum heattreatment of an as-quenched amorphous alloy. Sharma et al. [5] reported that heat-treatment at temperatures lower than the optimum temperature results in a significantly larger volume of remaining amorphous phase, whereas higher temperature heattreatment leads to crystallization of iron-boride phases. Magnetic properties of nano-crystalline alloys strongly depend on the type of phases (crystalline and amorphous), distribution, and their volume fraction. High density α-Fe nanocrystals are required for strong magnetic exchange coupling, which is necessary to average out the magnetocrystalline anisotropy of α-Fe [1]. A larger separation between α-Fe grains i.e. relatively more volume fraction of amorphous phase, leads to indirect exchange coupling among αFe grains (through the amorphous matrix), and affect the magnetic properties. Similarly appearance of a minor amount of iron-boride phase can significantly increase the coercivity (Hc) [6]. In soft magnetic materials coercivity is mainly governed by the motion of domain walls, and their propagation is affected by the presence of inhomogeneities, which can be amorphous/iron boride phases. Therefore, it is important to study motion of domain walls in the newly developed Fe–Si–B–P–Cu alloys heat-treated at different temperatures. The Fe85Si2B8P4Cu1 alloys were selected because of good magnetic properties (Hc 6 A/m and Bs 1.85 T) and an ability to produce as-quenched X-ray amorphous ribbons [5]. Motion of magnetic domain walls was studied by using in-situ Lorentz
http://dx.doi.org/10.1016/j.jmmm.2014.08.101 0304-8853/& 2014 Elsevier B.V. All rights reserved.
Please cite this article as: Z. Akase, et al., Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j. jmmm.2014.08.101i
67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96
Z. Akase et al. / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 Q4
microscopy in a transmission electron microscope (TEM) equipped with a magnetizing system [7–9] in order to clarify the dependence of domain walls motion on microstructure.
2. Materials and methods The Fe85Si2B8P4Cu1 ribbons (width 4–6 mm and thickness 20 μm) were made by high frequency induction melting (in vacuum) and melt-spinning (in air) techniques. Ribbons were heat-treated in an infra-red furnace under Ar-gas flow condition. Heating rate to achieve the required heat-treatment temperature was 400 1C/min. A magnetic flux density (B) versus magnetic field (H) loop analyzer was used to measure the magnetic properties. Three samples were selected based on their heat-treatment temperature and magnetic properties: Sample 1 is heat-treated at 330 1C, which is lower than the optimum heat-treatment temperature. Sample 2 is heat-treated under the optimum conditions, i.e., at 430 1C. Sample 3 is heat-treated at a higher temperature ( 470 1C) than the optimum temperature. The coercivities (Hc) of these samples were as follows: Sample 1: 21.2 A/m; Sample 2: 6.29 A/m; Sample 3: 259 A/m. The Hc of the as-quenched amorphous ribbon was 17 A/m. The TEM samples were prepared using a JIB-4500Multibeam focused ion beam (FIB) system. They were crosssectioned in the direction of the thickness of the ribbons with dimensions of 10 μm wide, 2–5 μm long and 0.1 μm thick. These rectangular samples were attached to a collodion film of TEM meshgrids by using a micro-manipulator with a glass probe. For conventional TEM observations and Lorentz microscope observations, we used HF-3300S (Hitachi High-Technologies). For in-situ Lorentz microscopy observations with a static or alternating external magnetic fields, we used JEM-3000F (JEOL) equipped with a magnetizing specimen holder, which can apply a magnetic field horizontal to the sample [7–9].
3. Results and discussion 3.1. Microstructures Fig. 1 shows bright field images (a–c) and selected area electron diffraction patterns (d–f) for Fe85Si2B8P4Cu1 ribbons: Sample 1
(330 1C heat-treated), Sample 2 (430 1C heat-treated), and Sample 3 (470 1C heat-treated). Sample 1 has randomly oriented α-Fe grains of size 20–40 nm sparsely distributed in the remaining amorphous matrix (Fig. 1a and d). The α-Fe grain size decreases significantly at higher-temperature heat-treatment. Dense and uniform distributions of α-Fe grains with the size of 15 nm are noticeable in the TEM images of Sample 2 and Sample 3 (Fig. 1b and c). Such a large difference in grain sizes is due to changes in the mechanism of crystallization at different heat-treatment temperatures. Grain nucleation density, which is dependent on heat-treatment temperature, governs the grain size. The diffraction patterns of all the heat-treated ribbons (Figs. 1d–f) shows Debye–Scherrer rings corresponding to α-Fe nano-crystallites. Additional diffraction spots can be noticed for Sample 3 (Fig. 1f, marked with red circles). These spots can be indexed as 112, 202 and 310 reflections from Fe2B crystals. Low Hc ( 6 A/m) for Sample 2 (430 1C heat-treated) is due to suppression of magneto-crystalline anisotropy (K1) of α-Fe grains through magnetic exchange interactions [6]. The necessary requirement for this to happen is that the grain size should be smaller than the ferromagnetic exchange length, which is about 20–40 nm for Fe-based alloys. The grain size of α-Fe for Sample 3 (470 1C heat-treated) is approximately 15 nm, which is similar to that of Sample 2 (430 1C heat-treated). Even though the grain size is similar to each other, Hc is very different (6.29 A/m for Sample 2; 259 A/m for Sample 3). This may be due to the presence of Fe2B grains. Huge magnetocrystalline anisotropy of Fe2B (430 kJ/m3, more than 10 times as large as that of α-Fe) may be responsible for this. The presence of high anisotropy grains in a nanocrystalline alloy can influence motion of domain walls, which then can be pinned by Fe2B grains, and can lead to an increase in Hc.
3.2. Magnetic domain structure and motion of magnetic domain walls The structural characterization confirmed that Sample 1 (heattreated at 330 1C), Sample 2 (heat-treated at 430 1C) and Sample 3 (heat-treated at 470 1C) are very different from each other in terms of microstructure, which is also reflected in their coercivity values. Fig. 2 shows magnetic domain walls in these ribbon samples imaged by Lorentz microscopy, where magnetic domain walls
Fig. 1. Bright field images (a–c) and selected area diffraction patterns (d–f) obtained from heat-treated Fe85Si12B6P4Cu1 amorphous-ribbons. The heat-treatment temperatures are 330 1C, 430 1C, and 470 1C. (For interpretation of the references to color in this figure the reader is referred to the web version of this article.)
Please cite this article as: Z. Akase, et al., Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j. jmmm.2014.08.101i
67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132
Z. Akase et al. / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
3
Fig. 2. Lorentz micrographs of heat-treated Fe85Si12B6P4Cu1 samples. The heat-treatment temperatures are 330 1C, 430 1C, and 470 1C.
Fig. 3. Snapshots of Lorentz micrographs of the heat-treated Fe85Si12B6P4Cu1 samples in an AC magnetic field H at 0.5 Hz, (a) H¼ 2.5 kA/m and (b) H¼ 2.0 kA/m from video recording for Sample 2 (430 1C heat-treated) and Sample 3 (470 1C heattreated). The direction of the applied magnetic field is indicated by the arrow at the top right.
appear as dark and bright line contrasts. All the samples exhibited a closure domain structures along the long axis of the sample. The domain structure depends on the size and the shape of the TEM samples, while the smoothness of the domain walls and their motion in each TEM sample depends on magnetic properties, as discussed in detail in the following. Note that the shape of domain walls in Sample 3 (470 1C heat-treated) is not as straight as for that of other samples. Kinks in domain walls can originate from pinning. Without pinning, domain walls will move smoothly with the increase in external magnetic field. To study motion of domain walls in the heat-treated samples, an alternating magnetic field was applied. Although Sample 2 (heat-treated at 430 1C) and Sample 3 (heat-treated at 470 1C) have a similar grain size of α-Fe, there is a large difference in coercivity Hc (6.29 A/m for Sample 2; 259 A/m for Sample 3). Therefore we investigated the difference in relation to the domain structure in more detail. Motion of domain wall was recorded from pictures on a small fluorescent screen using a conventional video camera. Fig. 3a and b shows the Lorentz micrographs of Sample 2 (430 1C heat-treated) and Sample 3 (470 1C heat-treated). These micrographs were obtained from snapshots of the recorded video. The original video files are included as a supplementary materials in this paper. Video 1 and Video 2 correspond to Fig. 3a and b, respectively. The frequency and amplitude of alternating magnetic field were 0.5 Hz and 2.5 kA/m for Fig. 3a (Sample 2), and 0.5 Hz and 2.0 kA/m for Fig. 3b (Sample 3). The direction of applied
Fig. 4. Lorentz micrographs of the specimen heat-treated at 470 1C under a static external magnetic field H of (a) 3.8 kA/m, (b) 4.3 kA/m, (c) 4.6 kA/m, and (d) 4.8 kA/m. The direction of the external magnetic field is indicated by the arrow at the top right. The right column shows magnified images of the corresponding rectangular regions on the left column. The positions of the magnetic domain wall and its positions at lower magnetic fields are indicated by dotted lines.
magnetic field is indicated by the arrow shown at the top of Fig. 3. Straight and smooth motion of magnetic domain walls were observed in Sample 2, while irregular motion of domain walls were observed in Sample 3. The irregular motion may result from the presence of Fe2B grains, because they have higher magnetocrystalline anisotropy than that of α-Fe.
Please cite this article as: Z. Akase, et al., Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j. jmmm.2014.08.101i
67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132
Z. Akase et al. / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎
4
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
Supplementary material related to this article can be found online at http://dx.doi.org/10.1016/j.jmmm.2014.08.101. In order to gain better understanding of domain wall motion in Sample 3, we observed domain wall behaviors by increasing static magnetic field H. The static magnetic field was increased continuously until the domain structure changed. A charge-coupled device (CCD) camera was used to obtain images of domain walls. Fig. 4a–d shows the Lorentz micrographs under a static external magnetic fields of 3.8 kA/m (Fig. 4a), 4.3 kA/m (Fig. 4b), 4.6 kA/m (Fig. 4c), and 4.7 kA/m (Fig. 4d). The direction of the magnetic field is indicated by the arrow shown at the top right of Fig. 4. Data on the left column show low-magnification images, whereas magnified images of the rectangular regions marked in the left column are shown on the right column. The position of the magnetic domain wall is marked by dotted lines. To understand the effects of magnetic field on the motion of domain walls, domain wall positions at smaller magnetic fields are also shown in the images. From Fig. 4, we find that the motion of domain walls is not smooth, in particular it is pinned in the area marked with a white circle in Fig. 4c. We conjecture that pinning of domain walls appeared at this region, where the magnetic flux density fluctuated due to the presence of Fe2B precipitates. 4. Conclusions Microstructures were observed on heat-treated Fe85Si2B8P4Cu1 amorphous ribbon samples by using transmission electron microscope. Results show that (1) Sample 2 (heat-treated at 430 1C) and Sample 3 (heat-treated at 470 1C) have a nano-crystalline structure with a grain size of 15 nm, (2) Sample 1 (heat-treated at 330 1C) has a larger grain size (20–40 nm), and (3) Sample 3 (heat-treated at 470 1C) has precipitates of Fe2B in addition to α-Fe grains. Domain walls were imaged in-situ by Lorentz microscopy by varying the external magnetic field H. Domain walls move smoothly in an as-quenched ribbon sample as well as in Sample 1 (330 1C heat-treated) and Sample 2 (430 1C heat-treated), while domain walls move less smoothly in Sample 3 (470 1C heattreated). Domain walls are pinned at highly anisotropic Fe2B
precipitates. We conjecture that the differences in coercivity between Sample 2 (430 1C heat-treated) and Sample 3 (470 1C heat-treated) is due to less smooth motion and/or domain wall pinning at Fe2B precipitates.
Acknowledgement We would like to thank Dr. Y. A. Ono for critical reading of the manuscript and for comments and suggestions. This work was supported by the “Tohoku Innovative Materials Technology Initiatives for Reconstruction (TIMT)” funded by the Ministry of Education, Culture, Sports, Science and Technology (MEXT) and Reconstruction Agency, Japan. This research was partly supported by the grant from the Japan Society for the Promotion of Science (JSPS) through the ‘“Funding Program for World-Leading innovative R&D on Science and Technology (FIRST program)” initiated by the Council for Science and Technology Policy (CSTP). References [1] G. Herzer, Modern soft magnets: amorphous and nanocrystalline materials, Acta Mater. 61 (2013) 718–734. [2] R. Hasegawa, Applications of amorphous magnetic alloys in electronic devices,, J. Non-cryst. Solids 287 (2001) 405–412. [3] A. Makino, Nanocrystalline soft magnetic Fe–Si–B–P–Cu alloys with high B of 1.8–1.9 T contributable to energy saving, IEEE Trans. Magn. 48 (2012) 1331–1335. [4] Z.Q. Zhang, P. Sharma, A. Makino, Role of Si in high Bs and low core-loss Fe85.2B10 XP4Cu0.8SiX nano-crystalline alloys, J. Appl. Phys. 112 (2012) 103902. [5] P. Sharma, X. Zhang, Y. Zhang, A. Makino, Influence of microstructure on soft magnetic properties of low coreloss and high Bs Fe85Si2B8P4Cu1 nanocrystalline alloy, J. Appl. Phys. 115 (2014) 17A340. [6] G. Herzer, Anisotropies in soft magnetic nanocrystalline alloys, J. Magn. Magn. Mater. 294 (2005) 99–106. [7] Z. Akase, D. Shindo, In situ Lorentz microscopy in an alternating magnetic field, J. Electron Microsc. 59 (2010) 207–213. [8] Z. Akase, D. Shindo, M. Inoue, A. Taniyama, Lorentz microscopic observations of electrical steel sheets under an alternating current magnetic field, Mater. Trans. 48 (2007) 2626–2630. [9] M. Inoue, T. Tomita, M. Naruse, Z. Akase, Y. Murakami, D. Shindo, Development of a magnetizing stage for in situ observations with electron holography and Lorentz microscopy, J. Electron Microsc. 54 (2005) 509–513.
Please cite this article as: Z. Akase, et al., Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j. jmmm.2014.08.101i
41 42 43 44 45 46 47 48 49 50 51 Q3 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79