TY - JOUR
AU1 - Kasahara, Takehiro
AU2 - Shindo, Daisuke
AU3 - Yoshikawa, Hideyuki
AU4 - Sato, Takafumi
AU5 - Kondo, Koichi
AB - Abstract Domain structures and magnetic flux distributions in Mn-Zn and Ni-Zn ferrites are investigated by in situ observations with Lorentz microscopy and electron holography. In situ Lorentz microscopic observation with the magnetic field applied reveals that the domain walls in Mn-Zn ferrite move easily across the grain boundary. On the other hand, each grain of Ni-Zn ferrite is magnetized by domain wall motion inside the grain. By taking a series of holograms with adjustment of the optical axis and astigmatism while the magnetic field is applied, we succeeded in observing the change in magnetic flux distribution quantitatively. Eventually, it is clarified that magnetization rotation does not take place in the magnetization process of Ni-Zn ferrite. The domain wall widths δ in Mn-Zn and Ni-Zn ferrites are evaluated to be 73 and 58 nm, respectively. Furthermore, through direct observation of the domain structure in Ni-Cu-Zn ferrite with Lorentz microscopy, it is found that the grains with size below 1.5 μm diameter are single domain. electron holography, domain wall motion, magnetic flux, ferrite, Lorentz microscopy, in situ observation Introduction With the reduction in size and weight of electronic appliances, transformer materials are expected to have lower power loss in high-frequency operations. In general, power loss is divided into hysteresis loss, eddy current loss and residual loss. Mn-Zn ferrites have low eddy current loss due to their higher resistivity than metals, and show good soft magnetic properties, so Mn-Zn ferrites are commonly used for transformers. On the other hand, Ni-Zn and Ni-Cu-Zn ferrites have such high-resistivity, >107 (Ω·m), that these ferrites can be expected as a material for higher frequency operations. Therefore, loss analyses of soft ferrites have been intensively carried out [1,2]. However, the details of such losses have not yet been understood. So, it is important to understand the motion of magnetic domain wall. Previously, we succeeded in observing the domain wall motions of Mn-Zn and Ni-Zn ferrites dynamically by Lorentz microscopy using the special specimen holder which can introduce the magnetic field to specimen position [3]. However, in order to understand details of magnetization process, it is necessary to observe the magnetization change by introducing the magnetic field. Recently, electron holography has caught the attention because this method can visualize the distribution of magnetic flux and electric potential. In this study, in situ observations of magnetic flux change in Mn-Zn and Ni-Zn ferrites with electron holography have been performed. It is noted that the domain wall widths of soft magnetic materials including Mn-Zn and Ni-Zn ferrites have not been previously reported. This is because the domain walls of soft magnetic materials tend to bend easily, and the thickness of specimens prepared by ion-milling is generally inhomogeneous so that it can be difficult to evaluate quantitatively their domain wall widths. In this paper, domain wall widths of Mn-Zn and Ni-Zn ferrites, with thin foils of homogeneous thickness prepared by focused-ion-beam (FIB) technique, have been measured quantitatively. Conversely, in order to reduce losses due to domain wall motion, it is efficient to prepare each grain consisting of single domain structure with no domain walls. There have been mainly two methods reported for evaluating single domain size. One is the use of neutron depolarization [4,5], and the other is the theoretical calculation method of the single domain size by using Brown's equation [4,6]. In this paper, single domain size in Ni-Cu-Zn ferrite has been evaluated by observing its domain structure with Lorenz microscopy. Methods Mn-Zn, Ni-Zn and Ni-Cu-Zn ferrites provided by NEC Tokin Corp. were prepared by powder metallurgy. Mean grain sizes of Mn-Zn and Ni-Zn ferrites are both 10 μm. Some of their magnetic and electric properties are presented in Table 1. It is noted that Mn-Zn ferrite has better soft magnetic properties than Ni-Zn ferrite, while the Ni-Zn ferrite has much higher electrical resistivity than Mn-Zn ferrite. In situ observations with Lorentz microscopy and electron holography were carried out using a JEM-3000F TEM with a field emission gun and a biprism. This microscope has a special polepiece designed for observing magnetic domains; i.e. the magnetic field at the specimen position can be reduced to 32 A m−1 (= 0.4 Oe) [7]. Thin foil specimens for TEM observation were prepared by a FIB technique (JEM-9310FIB), by which their size was controlled to be about 12 μm × 10 μm × 100 nm. The change in magnetic domains and lines of magnetic flux is studied with a magnetic holder that can produce a maximum magnetic field of about 16 kA m−1 (= 200 Oe) within the TEM [8]. However, it should be considered that only a part of applied magnetic field contributes to the magnetization process of thin foil specimen prepared by a FIB technique, because the thin foil specimen is surrounded by bulk areas ∼50 μm thickness. While the magnetic field is applied to specimen in TEM by using this specimen holder, domain wall movement has been observed by Lorentz microscopy while correcting the deflection of the electron beam due to the magnetic field. In addition, the change of magnetic flux has been also observed with electron holography in order to examine details of the magnetization process. In electron holography, a series of holograms were obtained by correcting the misalignment of optical axis and astigmatism caused by the applied magnetic field. Reconstructed phase images of the holograms were obtained from the digitized holograms with the Fourier transform operation. The phase shift φ(x, y) due to the magnetic field of the specimens is represented by cos φ(x, y) in the reconstructed phase images [9]. Table 1 Magnetic and electric properties of Mn-Zn, Ni-Zn and Ni-Cu-Zn ferrites Specimen  Bs (mT)  Hc (A m−1)  μi at 100 kHz  ρ (Ω·m)  Mn-Zn ferrite  520  13  2300  10  Ni-Zn ferrite  460  104  250  107  Ni-Cu-Zn ferrite  400  76  490  107  Specimen  Bs (mT)  Hc (A m−1)  μi at 100 kHz  ρ (Ω·m)  Mn-Zn ferrite  520  13  2300  10  Ni-Zn ferrite  460  104  250  107  Ni-Cu-Zn ferrite  400  76  490  107  View Large In order to estimate the domain wall width, the full width at half maximum (FWHM) of domain wall contrast has been measured by using Lorentz microscopy while changing the defocus value. Here, there are two contrasts of domain wall. One is the divergent wall image which appears to be dark, and the other is convergent wall image, which appears to be bright. The domain wall width of white contrast cannot be measured correctly at large defocus, because the domain wall contrast tends to have fine structure due to the interference effect. Therefore, two 180° domain walls were selected, and then FWHM were measured with dark contrast in the overfocused image and underfocused image. The intensity profiles were obtained in the narrowest areas of divergent wall images to avoid overestimating the wall width because of the inclination of the domain wall against the electron beam. FWHM were then plotted as a function of the defocus value, and the domain wall width was estimated from the intersection point of the extrapolated lines. Results and discussions Observations of domain structures in Mn-Zn and Ni-Zn ferrites Figure 1 shows bright-field images and Lorentz microscope images of Mn-Zn [(a), (b)] and Ni-Zn [(c), (d)] ferrites, respectively. The white dashed lines and black curved bands inside grains observed in Fig. 1a and c indicate the grain boundaries and the bend contours, respectively. In Lorentz microscope images of Fig. 1b and d, white lines and black bands marked with arrowheads correspond to the domain walls. Both Mn-Zn and Ni-Zn ferrites have the mean grain size of about 10 μm. Further, some pores (from 0.2 to 1.1 μm) produced during the sintering process can be seen in both specimens, which are located on the grain boundary or inside the grain. While most domain walls exist on the grain boundaries in Mn-Zn ferrite of Fig. 1b, the domain walls are located inside grains or pinned at some pores in Ni-Zn ferrite of Fig. 1d. Fig. 1 View largeDownload slide Bright-field images and Lorentz microscope images of Mn-Zn (a and b) and Ni-Zn (c and d) ferrites, respectively. The white dashed lines and black curved bands inside grains observed in (a and c) indicate the grain boundaries and the bend contours, respectively. Fig. 1 View largeDownload slide Bright-field images and Lorentz microscope images of Mn-Zn (a and b) and Ni-Zn (c and d) ferrites, respectively. The white dashed lines and black curved bands inside grains observed in (a and c) indicate the grain boundaries and the bend contours, respectively. The magnetic anisotropy constant of Ni-Zn ferrite is larger than that of Mn-Zn ferrite, so the axis of easy magnetization (the <111> directions) is not aligned in one direction. It is well known that magnetic poles can appear at the grain boundaries between grains if the directions in the axis of easy magnetization are different [10]. In Ni-Zn ferrite, the magnetic anisotropy constant is large, and thus the effect of magnetic poles may also be large. Eventually, closure domains are formed inside the grain in the Ni-Zn ferrite in order to disperse their magnetic poles, resulting in the reduction of magnetostatic enerrgy. On the other hand, in Mn-Zn ferrite, the magnetic anisotropy constant is so small, and thus the effect of magnetic pole may also be small. Eventually, few closure domains are formed at the grain boundaries in Mn-Zn ferrite. Figure 2 shows the relationship between the domain wall width and the defocus value. From the result, the domain wall width δ of Mn-Zn and Ni-Zn ferrites are estimated to be 73 and 58 nm, respectively. Here, δ is given as [11]   δ=πJS2Ka, (1) where J, S, K and a are the exchange constant, the spin, the anisotropy constant and the lattice constant, respectively. As shown in Table 2, it has been reported that K of Mn-Zn and Ni-Zn ferrites are −800 J m−3 and −4000 J m−3 at room temperature, respectively [12,13]. Thus, it is understood that the difference between domain wall widths of Mn-Zn and Ni-Zn ferrites is related to the difference of their anisotropy constants, i.e. the smaller anisotropy constant results in the wider domain wall width. It is noted that not only K but also other parameters (J, S and a) and the thin film effect should be taken into account to accurately evaluate the difference of domain wall widths between the two specimens. Fig. 2 View largeDownload slide Relationship between the domain wall width and the defocus value. Fig. 2 View largeDownload slide Relationship between the domain wall width and the defocus value. Table 2 Domain wall widths δ and the magnetic anisotropy constants K of Mn-Zn and Ni-Zn ferrites Specimen  δ (nm)  −K (J m−3)  Mn-Zn ferrite  72.8  K = 800[12]  Ni-Zn ferrite  57.6  K = 4000[12,13]  Specimen  δ (nm)  −K (J m−3)  Mn-Zn ferrite  72.8  K = 800[12]  Ni-Zn ferrite  57.6  K = 4000[12,13]  View Large In situ observation of domain wall motion and magnetic flux change in Mn-Zn and Ni-Zn ferrites Figure 3b–f are Lorentz microscope images of the Mn-Zn ferrite captured from videotape, showing domain wall motion with the increase in magnetic field. In Fig. 3a, the grain boundaries are schematically illustrated. The magnetic field is applied to the thin foil specimen parallel to the direction indicated by a big black arrow in Fig. 3c. It can be seen that most domain walls exist on the grain boundaries in Fig. 3b. With the increase in magnetic field, the domain walls move easily across the grain boundary and magnetic domains larger than the grain sizes are formed (Fig. 3c–e). Finally, the domain walls disappear and a large magnetic domain is formed in Fig. 3f, which is considered to be magnetized parallel to the direction of the magnetic field applied. Thus, it is reasonably considered that the easy domain wall movement without pinning in Mn-Zn ferrites results in high-permeability and low-hysteresis loss. Fig. 3 View largeDownload slide Lorentz microscope images of Mn-Zn ferrite captured from the videotape, directly showing the domain wall motion with the increase in magnetic field. In (a), the grain boundaries are schematically illustrated. The magnetic field is applied to the thin foil specimen parallel to the direction indicated by a large black arrow in (c). (b′–f′) are the schematic illustrations for (b–f). In the illustrations, red and blue lines indicate the white lines and black bands corresponding to the domain walls, respectively. Fig. 3 View largeDownload slide Lorentz microscope images of Mn-Zn ferrite captured from the videotape, directly showing the domain wall motion with the increase in magnetic field. In (a), the grain boundaries are schematically illustrated. The magnetic field is applied to the thin foil specimen parallel to the direction indicated by a large black arrow in (c). (b′–f′) are the schematic illustrations for (b–f). In the illustrations, red and blue lines indicate the white lines and black bands corresponding to the domain walls, respectively. Figure 4 shows Lorentz microscope images of Mn-Zn ferrite. While Fig. 4a is the demagnetized state, Fig. 4b shows the domain structure with the magnetic field of 2880 A m−1 applied. The sharp white lines and black bands correspond to the domain wall. With the increase in the magnetic field, the white line indicated by an arrowhead changes to the black band, and the domain wall appearing to be a white line arrowed at upper left moves to the lower position. In order to understand such a complicated demagnetization process, the magnetic flux change has been observed in detail by electron holography in the area surrounded by dashed lines, as described below. Fig. 4 View largeDownload slide Lorentz microscope images of Mn-Zn ferrite. (a) is the demagnetized state. (b) shows the domain structure with the magnetic field of 2880 A m−1 applied. Fig. 4 View largeDownload slide Lorentz microscope images of Mn-Zn ferrite. (a) is the demagnetized state. (b) shows the domain structure with the magnetic field of 2880 A m−1 applied. Figure 5a–i are the images selected from a series of reconstructed phase images while changing the applied magnetic field. Each time the magnetic field was changed, the misalignment of optical axis and astigmatism was corrected. White and black lines correspond to the lines of magnetic flux, and there exists the magnetic flux of h/e (= 4.1 × 10−15 Wb) between these lines. The boundaries at which the lines of magnetic flux change direction correspond to the domain wall. If the magnetic flux around the domain wall flows clockwise or counterclockwise, the domain walls are indicated by white and black dashed lines, respectively. The white arrows indicate the directions of lines of magnetic flux, and the direction of applied magnetic field is indicated by a big black arrow in Fig. 5b. The thick white line indicated as GB corresponds to the grain boundary. It is noted that some apparent bending of lines of magnetic flux is considered to result from the abrupt thickness change. From a demagnetized state to the state with 1600 A m−1 applied, there is no change in the distribution of magnetic flux (Fig. 5a and b). However, the distribution of magnetic flux changes from Fig. 5b to c with domain wall motion. In Fig. 5d, magnetic flux changes preferentially at the grain boundary, and a new domain wall, indicated by the black dashed line, appears at this boundary. The reason is considered to be related to the decrease in permeability at a grain boundary including precipitates, such as SiO2. On the other hand, it is observed that a new domain wall, indicated as white dashed line, is formed in Fig. 5f and this domain wall moves to the right in Fig. 5h with the increase in applied magnetic field. These changes of magnetic flux are considered to decrease the magnetostatic energy by forming the domain wall. In Fig. 5i with the applied magnetic field of 2880 A m−1, it is observed that the lines of magnetic flux have the same direction as the applied magnetic field in the right area of the image. On the contrary, there is antiparallel flux in the left area of the image. This is considered to be affected by thick area formed by the FIB milling process. Fig. 5 View largeDownload slide (a–i) Images selected from a series of reconstructed phase images with slight change of the applied magnetic field. White and black lines correspond to lines of magnetic flux. The white arrows indicate the directions of lines of magnetic flux, and the direction of applied magnetic field is indicated by a large black arrow in (b). White line indicated as GB corresponds to the grain boundary. Fig. 5 View largeDownload slide (a–i) Images selected from a series of reconstructed phase images with slight change of the applied magnetic field. White and black lines correspond to lines of magnetic flux. The white arrows indicate the directions of lines of magnetic flux, and the direction of applied magnetic field is indicated by a large black arrow in (b). White line indicated as GB corresponds to the grain boundary. Figures 6b–f show Lorentz microscope images of Ni-Zn ferrite captured from the videotape, directly showing domain wall motion with the increase in magnetic field. Figure 6a shows schematic illustration of the grain boundaries. The magnetic field is applied to the thin foil specimen parallel to the direction indicated by the large black arrow in Fig. 6c. Hereafter, we focus our attention to the motion of the domain walls indicated as DW (Fig. 6b). In Fig. 6b, these three domain walls indicated by arrows exist in each grain. With the increase in magnetic field, the domain walls arrowed move to the grain boundary as shown in Fig. 6c and d. In Fig. 6e, the domain wall which appears as a black band in the lower right of the image is pinned at the grain boundary, then the domain wall appearing as a white line moves to the pinning boundary and comes across it. From Fig. 6e to f, the contrast of the domain walls disappear at the grain boundary. Furthermore, from Fig. 6g to h, another domain wall appearing as a white line forms and moves from the left to the grain boundary, and it comes across the domain wall appearing as a black band being pinned at the grain boundary. Then, these two contrasts disappear as shown in Fig. 6g and h. Finally, the domain walls all disappear from inside the grains and grain boundary as shown in Fig. 6i. It is considered that the magnetizations are all along the same direction. Fig. 6 View largeDownload slide Lorentz microscope images of Ni-Zn ferrite captured from videotape, directly showing domain wall movement with the increase in magnetic field. (a) is the schematic illustration of the grain boundaries. The magnetic field applied to the thin foil specimen is indicated by a large black arrow in (c). (b′–i′) are the schematic illustrations for (b′–i′). In the illustrations, red and blue lines indicate the white lines and black bands corresponding to the domain walls, respectively. The area of the illustration corresponds to the lower right part outlined by dotted lines in (a). Fig. 6 View largeDownload slide Lorentz microscope images of Ni-Zn ferrite captured from videotape, directly showing domain wall movement with the increase in magnetic field. (a) is the schematic illustration of the grain boundaries. The magnetic field applied to the thin foil specimen is indicated by a large black arrow in (c). (b′–i′) are the schematic illustrations for (b′–i′). In the illustrations, red and blue lines indicate the white lines and black bands corresponding to the domain walls, respectively. The area of the illustration corresponds to the lower right part outlined by dotted lines in (a). Thus, it is seen that each grain of Ni-Zn ferrite is magnetized by domain wall movement inside the grain. On the other hand, the domain walls in Mn-Zn ferrite move easily across the grain boundary. So, it can be said that the magnetization process of Ni-Zn ferrite is quite different from that of Mn-Zn ferrite. We also observed the dynamic change in the lines of magnetic flux by electron holography in order to understand the detailed magnetization process. In Fig. 7, Lorentz microscope images of Ni-Zn ferrite are shown on the left, and the reconstructed phase images of the same area are shown on the right. The magnetic field is applied along the direction of the large black arrow in Fig. 7a and b. The white and black dashed lines in Fig. 7b correspond to the domain walls appearing as white lines and black bands, respectively. The thick white line indicated by GB corresponds to the grain boundary. In Fig. 7a, there are domain walls appearing as white lines in the grain, and the domain wall located on the grain boundary appears as a black band. The distribution of the lines of magnetic flux in Fig. 7b clearly indicates the closure domain inside the grain. Also, the change in direction of the lines of magnetic flux at the grain boundary is considered to result in the domain wall with dark contrast in the Lorentz microscope image. With the increase in magnetic field, the domain wall with bright contrast moves to the right and comes across the domain wall with dark contrast at the grain boundary, and then both contrasts disappear (Fig. 7c and e). It is noted that in the reconstructed phase images, the magnetic domain with magnetization antiparallel to the applied field becomes smaller, keeping the directions of the lines of magnetic flux unchanged in each domain. This indicates that magnetization rotation does not take place in the process of magnetization. Finally, the domain wall moves and disappears at the grain boundary, as shown in Fig. 7g and h. Fig. 7 View largeDownload slide (a, c, e and g) Lorentz microscope images (left) and reconstructed phase images (right) in the same region of the Ni-Zn ferrite obtained when the magnetic fields applied are 640, 1440, 2040, 3040 A m−1. The magnetic field is applied to the thin foil specimen parallel to the direction indicated by a large black arrow in (a) and (b). (b, d, f and h) The white and black dashed lines correspond to the bright and dark contrasts of the domain wall, respectively, and the white line indicated as GB corresponds to the grain boundary. Fig. 7 View largeDownload slide (a, c, e and g) Lorentz microscope images (left) and reconstructed phase images (right) in the same region of the Ni-Zn ferrite obtained when the magnetic fields applied are 640, 1440, 2040, 3040 A m−1. The magnetic field is applied to the thin foil specimen parallel to the direction indicated by a large black arrow in (a) and (b). (b, d, f and h) The white and black dashed lines correspond to the bright and dark contrasts of the domain wall, respectively, and the white line indicated as GB corresponds to the grain boundary. Evaluation of single-domain size in Ni-Cu-Zn ferrite In Ni-Cu-Zn ferrites prepared by changing sintering temperature, the relationship between the mean grain size and the losses are obtained as shown in Table 3. From Table 3, it is seen that the smaller mean grain size results in the smaller losses in high-frequency operations. It is considered that if the grain size becomes smaller, the number of domain walls inside the grain becomes smaller, resulting in decrease in the loss due to the dynamic motion of the domain walls in high-frequency operations. Here, the relationship between the number of magnetic domains inside the grain, and the grain diameter, has been examined in the specimen shown in Table 1 prepared by a FIB technique keeping the size of the specimens to 10 μm × 10 μm × 100 nm. Table 3 Mean grain size d and core loss Pcv of Ni-Cu-Zn ferrite prepared by changing the sintering temperature T No.  T (K)  d (nm)  Pcv (mW cc−1)  A  1253  2  750  B  1313  3.5  920  C  1353  28  2000  No.  T (K)  d (nm)  Pcv (mW cc−1)  A  1253  2  750  B  1313  3.5  920  C  1353  28  2000  View Large Figure 8a shows an example of bright field image of the Ni-Cu-Zn ferrite and Fig. 8b shows Lorentz micoscope image of the same area. The white dashed line corresponds to the grain boundary. As indicated by the white arrow in Fig. 8a, the longest diagonal line is regarded as the diameter D of its grain. Then, the underfocused and overfocused images of the same grain are observed, and the number of the magnetic domain existing in the grain is counted. If there are no domain walls, the grain is considered to be a single domain. In this study, 35 grains were examined. Figure 9 is the bar diagram showing the ratio of constituent domains per grain in Ni-Cu-Zn ferrite as a function of D. From this result, it is seen that the grains with sizes below 1.5 μm diameter have no domain walls and are regarded as single domains. Therefore, it is efficient to prepare grains with sizes below 1.5 μm diameter in order to achieve single domains for lower loss Ni-Cu-Zn ferrite in high-frequency operations. Fig. 8 View largeDownload slide Bright field image of the Ni-Cu-Zn ferrite (a) and Lorentz microscope image of the same area (b). The white dashed line corresponds to the grain boundary. As indicated by the white arrow in (a), the longest diagonal line is regarded as the diameter D of the grain. Fig. 8 View largeDownload slide Bright field image of the Ni-Cu-Zn ferrite (a) and Lorentz microscope image of the same area (b). The white dashed line corresponds to the grain boundary. As indicated by the white arrow in (a), the longest diagonal line is regarded as the diameter D of the grain. Fig. 9 View largeDownload slide Bar diagram showing the ratio of constituent domains per grain in Ni-Cu-Zn ferrite evaluated by Lorentz microscopy as a function of D. Fig. 9 View largeDownload slide Bar diagram showing the ratio of constituent domains per grain in Ni-Cu-Zn ferrite evaluated by Lorentz microscopy as a function of D. Concluding remarks In this paper, domain structure and domain wall motion in thin films of Mn-Zn and Ni-Zn ferrites are investigated by in situ observations with Lorentz microscopy and electron holography. The results are summarized as follows: Using Lorentz microscopy, the domain wall widths δ in Mn-Zn and Ni-Zn ferrites are evaluated to be 73 and 58 nm, respectively. Domain walls in Mn-Zn ferrite move easily across the grain boundary. On the other hand, each grain of Ni-Zn ferrite is magnetized by domain wall movement inside the grain. The magnetization processes of Mn-Zn and Ni-Zn ferrites are sharply different from each other. By taking a series of holograms with correcting the misalignment of optical axis and astigmatism while the magnetic field is applied, we succeeded in systematically observing the change in the distribution of magnetic flux. Eventually, it is found that the magnetization rotation does not take place in the magnetization process of Ni-Zn ferrite. Through direct observation of the domain structure in Ni-Cu-Zn ferrite with Lorentz microscopy, grains with sizes below 1.5 μm diameter are considered to be single domain. This work was partly supported by financial support from Grant-in-Aid for Scientific Research (A) from Japan Society for the Promotion of Science. References 1 Visser E G.  Analysis of the complex permeability of monocrystalline MnZnFeII ferrite,  J. Magn. Magn. Mater. ,  1984, vol.  42 (pg.  286- 290) Google Scholar CrossRef Search ADS   2 Kondo K,  Chiba T,  Yamada S,  Otsuki E.  Analysis of power loss in Ni-Zn ferrites,  J. Appl. Phys. ,  2000, vol.  87 (pg.  6229- 6231) Google Scholar CrossRef Search ADS   3 Kasahara T,  Park H S,  Shindo D,  Yoshikawa H,  Sato T,  Kondo K.  In situ observations of domain wall motion in Mn–Zn and Ni–Zn ferrites by Lorentz microscopy and electron holography,  J. Magn. Magn. Mater. ,  2006, vol.  305 (pg.  165- 170) Google Scholar CrossRef Search ADS   4 Zaag P J,  Valk P J,  Rekveldt M Th.  A domain size effect in the magnetic hysteresis of NiZn-ferrites,  Appl. Phys. Lett. ,  1996, vol.  69 (pg.  2927- 2929) Google Scholar CrossRef Search ADS   5 Aarts J,  Shiekah I A,  Zaag P J.  Domain structure in polycrystalline MnZn ferrite imaged by magnetic force microscopy,  J. Appl. Phys. ,  1999, vol.  85 (pg.  7302- 7309) Google Scholar CrossRef Search ADS   6 Brown W FJr.  The fundamental theorem of fine-ferromagnetic-particle theory,  J. Appl. Phys. ,  1968, vol.  39 (pg.  993- 994) Google Scholar CrossRef Search ADS   7 Shindo D,  Park Y-G,  Murakami Y,  Gao Y,  Kanekiyo H,  Hirosawa S.  Electron holography of Nd-Fe-B nanocomposite magnets,  Scr. Mater. ,  2003, vol.  48 (pg.  851- 856) Google Scholar CrossRef Search ADS   8 Inoue M,  Tomita T,  Naruse M,  Akase Z,  Murakami Y,  Shindo D.  Development of a magnetizing stage for in situ observations with electron holography and Lorentz microscopy,  J. Electron Microsc. ,  2005, vol.  54 (pg.  509- 513) Google Scholar CrossRef Search ADS   9 Shindo D,  Oikawa T. ,  Analytical Electron Microscopy for Materials Science ,  2002 Tokyo Springer-Verlag 10 Hubert A,  Schafer R. ,  Magnetic Domains ,  1998 New York Springer-Verlag 11 Chikazumi S. ,  Physics of Ferromagnetism ,  1997 New York Oxford University Press 12 Globus A,  Duplex P.  Effective anisotropy in polycrystalline materials. Separation of components,  J. Appl. Phys. ,  1968, vol.  39 (pg.  727- 729) Google Scholar CrossRef Search ADS   13 Groenou AB,  Schulkes JA,  Annis DA.  Magnetic anisotropy of some nickel zinc ferrite crystals,  J. Appl. Phys. ,  1967, vol.  38 (pg.  1133- 1134) Google Scholar CrossRef Search ADS   © The Author 2007. Published by Oxford University Press on behalf of Japanese Society of Microscopy. All rights reserved. For permissions, please email: journals.permissions@oxfordjournals.org
TI - In situ observations of domain structures and magnetic flux distributions in Mn-Zn and Ni-Zn ferrites by Lorentz microscopy and electron holography
JF - Journal of Electron Microscopy
DO - 10.1093/jmicro/dfl042
DA - 2007-01-17
UR - https://www.deepdyve.com/lp/oxford-university-press/in-situ-observations-of-domain-structures-and-magnetic-flux-GFa96SjAX6
SP - 7
EP - 16
VL - 56
IS - 1
DP - DeepDyve
ER -