Journal of Magnetism and Magnetic Materials 416 (2016) 53–60
Contents lists available at ScienceDirect
Journal of Magnetism and Magnetic Materials journal homepage: www.elsevier.com/locate/jmmm
Electromagnetic properties of Co flaky particles prepared via ballmilling method Chao Liu a, Jian-Tang Jiang a, Yong Yuan b, Yuan-Xun Gong c, Liang Zhen a,d,n a
School of Materials Science and Engineering, Harbin Institute of Technology, Harbin 150001, China Precision Machinery Research Institute of Shanghai Space Flight Academy, Shanghai 201600, China c Aerospace Research Institute of Special Material and Processing Technology, Beijing 100074, China d MOE Key Laboratory of Micro-System and Micro-Structures Manufacturing, Harbin Institute of Technology, Harbin 150080, China b
art ic l e i nf o
a b s t r a c t
Article history: Received 22 December 2015 Received in revised form 14 April 2016 Accepted 15 April 2016 Available online 19 April 2016
Flaky cobalt particles with different aspect ratio were produced with ball-milling method. The phase structure and morphology of the particles were identified by XRD analysis and SEM observation. The static magnetic and electromagnetic properties of the particles were measured and effects of shape, microstructure and filling fraction were investigated. Phase transition from fcc lattice to hcp lattice occur due to the drive of ball-milling is responsible for the largely increased coercivity. Particles with high aspect ratio are found to possess high permittivity and permeability, compelling the frequency of absorption peak to shift to low frequency. Coatings using cobalt particles milled for 20 h as fillers present a RL peak of 33 dB at 8 GHz at the thickness of 2.5 mm together with a broad effective absorbing (RL below 10 dB) bandwidth covering 6–10 GHz. & 2016 Elsevier B.V. All rights reserved.
Keywords: Flaky cobalt Ball-milling Shape Electromagnetic properties
1. Introduction Electromagnetic wave absorbing (EMA) materials that serve at elevated temperature have become more and more vital in defense systems. Carbon fibers and SiC fibers have been explored for application as high temperature EMA materials for years [1,2]. These series of materials present high dielectric loss efficiency but do not possess ferromagnetic properties and thus exhibit unsound electromagnetic matching characteristics, which then inhibits the improvement of EMA performances. Additionally, application of C fibers and SiC fibers is usually combined into complicated fabrication process of components and the flexibility is thus limited. Particles of Fe and related alloys are excellent EMA materials [3–5] as they possess high permeability and tailorable permittivity when used at mild temperature. Their permeability however decreases quickly when exposed to temperature near to the Curie point (770 °C), which thus limits the application at high temperatures. Cobalt, possessing much higher Curie temperature (1130 °C) and satisfactory saturation magnetization (Ms, 165 emu/g), exhibits the potency to present high permeability at higher temperature. Considering the fact that permittivity is usually much higher than permeability in metallic particles, it's then significant to obtain n Corresponding author at: School of Materials Science and Engineering, Harbin Institute of Technology, Harbin 150001, China. E-mail address:
[email protected] (L. Zhen).
http://dx.doi.org/10.1016/j.jmmm.2016.04.037 0304-8853/& 2016 Elsevier B.V. All rights reserved.
high permeability for obtaining good impedance matching and for obtaining excellent EMA performances. Besides, application of ferromagnetic particles is more flexible than C and SiC fibers since they can be filled into matrix to fabricate coatings rather than being built-into rigid components. High and properly tailored electromagnetic properties are crucial for obtaining excellent EMA performances, as previous research suggested [6,7]. Among various factors that affect electromagnetic properties, particle shape is found the most vital one [8– 11]. Recent work has indicated that, flaky particles usually possess much higher anisotropy to compare with spherical ones due to the lower geometric symmetry. Meanwhile, the thin thickness of flaky particles may also sever the purpose of suppressing the eddy current effect that occurs in metallic particles. Both these merits are helpful for obtaining high permeability and have inspired a lot of research in the past a few years. Gong [8] synthesized CoFe alloy flakes (NF) and spheres (NP) of nanometer scale and observed that CoFe NFs presented higher permeability than CoFe NPs. Also, coatings using CoFe NFs presented a higher EMA efficiency (RLmax of 57.8 dB, at about 2.3 GHz) to compare with that observed in case of CoFe NP ( 16.6 dB, at 18 GHz). Ma [12] prepared Co flakes several micrometer in diameter and around 80 nm in thickness through hydrothermal reduction and observed excellent EMA performance. Li [13] prepared Co nano-flakes via liquid phase reduction method and observed m′ as high as 2 at 2 GHz in a specimens with quite low filling ratio (13 vol%). Flaky cobalt particles of micrometer or nanometer scale are
54
C. Liu et al. / Journal of Magnetism and Magnetic Materials 416 (2016) 53–60
usually prepared through solution chemistry methods with very small yield in previous research, which limits the practical application of this series of particles as EMA fillers. Besides, fine particles prepared through this method are chemically active and likely to oxide, which also restricts applications exposed to open weather. Ball milling is a convenient method to produce flaky particles in large quantity. It has been successfully applied in preparing flaky Fe or Fe based alloy particles [14–17] and high performance EMA fillers were where from fabricated. The application of ball milling in Co particles however has not been reported and the evolution in particle shape and EM properties was not clear. The current study was thus inspired to investigate the evolution in morphology, microstructure and EM properties of Co particles. Flaky cobalt particles of different micro sizes were prepared in large scale via a ball-milling process in the current research. Influences of particles' shape and microstructure on electromagnetic properties were systematically investigated.
2. Materials and experimental details Commercially available gas-atomized cobalt particles (mesh -400, 99.9 wt% in purity) were used as the starting material. The as-received particles, certain anhydrous ethanol and some GCr15 steel balls were put into jars of 250 ml and then the ball-milling process was carried out with a planetary ball-mill system at 500 rpm for up to 20 h. For all processes, the weights of ball, particle and ethanol were kept at 350 g, 70 g and 45 g, respectively. Milled particles were gathered after the milling process completed, washed for 3 times with anhydrous ethanol and then dried for 12 hours at 60 °C in an oven. Phase composition of raw and as-milled cobalt particles was examined on a X-ray diffractometer (XRD, PANalytical X'Pert PRO, CuKα). The morphology of cobalt particles was observed with a scanning electron microscope (SEM, Philips FEI Siron). Static magnetic properties were measured with a vibrating sample magnetometer (VSM, Lake Shore 7404) at room temperature. Electromagnetic properties of specimens with milled cobalt as fillers and paraffin as matrix were measured on a vector network analyzer (VNA, Agilent N5230A) in the frequency range of 2– 18 GHz. The volume fraction of fillers in VNA specimens was set at 10 vol%, 23 vol% and 30 vol% respectively to examine the effect of filling rate. The fabricated VNA specimens were coaxial toroidal, 7 mm in outer diameter, 3 mm in inner diameter and 3–3.5 mm in thickness. Microwave absorption performance can be evaluated by the transmission line theory on the basis of the following related formulas [7,18]: 1/2
Zin = Z 0(μr /εr )
1/2
tanh[j(2πfd/c )(μr εr )
]
RL = 20log (Zin − Z 0)/(Zin + Z 0)
(1)
(2)
3. Results and discussion
Fig. 1. X-ray diffraction patterns of cobalt particles milled for different time.
that β-Co is transferred to α-Co during the ball-milling. On the other hand, characteristic peaks of α-Co become broader and some peaks disappear as the ball-milling proceeds, as shown in the figure. Similar results were observed in our previous work [9] and other researches [19,20]. This variation is attributed to the increased defects, the decreased grain size and the amorphous phase's forming in the surface layer of the particles due to the intense impact. Grain size of different samples is calculated according to Scherer formula and the calculated results are listed in Table 1. It is found that grain size decease before 8 h and then keep unchanged during the ball milling. Meanwhile, the fact that average grain sizes calculated by (101) peak and (002) peak are different is reasonable on the basis that ball milling will result in non-uniform deformation of grains. Figs. 2 and 3 shows the morphology of raw and milled particles. The raw particles are spheres with 5–30 μm in diameter, as shown in Fig. 2(a). After milled for 4 h, particles are flattened into flakes with 3–4 μm in thickness. After milled for 8 h, flakes with diameter up to 40 μm and thickness below 1 μm are observed, contributing to a very high aspect ratio (diameter/thickness, AR). As the ball-milling further proceeds, flaky particles are broken into small scraps while the thickness remains at about 1 μm. The largest particle diameter observed at 12 h, 16 h and 20 h is around 25 μm, 15 μm and 10 μm, respectively. Through the whole process, the AR firstly increases and then decreases. 3.2. Static magnetic properties Fig. 4 shows the magnetic hysteresis loops of raw cobalt particles and particles milled for different time. The measured static magnetic properties are listed on Table 2. Saturation magnetization (Ms) fluctuates slightly all along the milling proceeds but remains at the level similar to that of raw particles. Since Ms of β-Co (165 emu/g) is quite near to that of α-Co (162 emu/g) [21], Table 1 Grain size calculated by different x-ray diffraction peaks according to Scherer formula.
3.1. Phase structure and morphology Fig. 1 shows the x-ray diffraction patterns of cobalt particles as received or ball-milled for different time. The XRD pattern of raw cobalt particles presents two sets of diffraction peaks, one of which can be indexed to the hcp lattice of α-Co, another to fcc lattice of β-Co. Peaks corresponding to β-Co disappear while those corresponding to α-Co enhanced in the pattern, after ball milled for 4 h, as shown in Fig. 1. This variation in XRD pattern indicates
Samples (milling time)
Grain size (nm) calculated Grain size (nm) calculated by (002) peak by (101) peak
Raw (0 h) 4h 8h 12 h 16 h 20 h
36 19 17 21 18 19
21 13 12 10 11 10
C. Liu et al. / Journal of Magnetism and Magnetic Materials 416 (2016) 53–60
55
Fig. 2. Morphology of raw cobalt particles (magnified 500 times), (a), and particles ball milled for (b), 4 h, (c), 8 h, (d), 12 h, (e), 16 h, and (f) 20 h.
Fig. 3. Morphology of raw cobalt particles (magnified 2000 times), (a), and particles ball milled for (b), 4 h, (c), 8 h, (d), 12 h, (e), 16 h, and (f) 20 h.
transition from β-Co to α-Co does not influence the Ms of the particle apparently. The variation of Ms is believed related to the increased density of defects in particles. On the other hand, coercivity (Hc) of milled particles is observed around twice higher than that of raw Co particles (73 Oe). Sun et al. [22] observed that coercivity of cobalt nano-crystals of mixed phases was much lower than that of pure hcp phase. Sato et al. [23] noted that Hc of Co powders kept increasing as the content of hcp phase increased.
The increase of coercivity observed in the current research is also related to the increase content of hcp Co. Hc is positively correlated to the magneto crystalline anisotropy constants (K) which is much higher in hcp cobalt than in fcc cobalt. Due to the higher coercivity, the remanent magnetization (Mr) increases from 2.1 emu/g of raw particles to 8.1 emu/g of miller particles for 12 h and then drops slightly as ball-milling proceeds.
56
C. Liu et al. / Journal of Magnetism and Magnetic Materials 416 (2016) 53–60
3.3. Electromagnetic properties
Fig. 4. Magnetic hysteresis (M–H) loops of cobalt particles milled for different time. Table 2 Magnetic properties of cobalt powders milled for different time. No.
Ms (emu/g)
Hc (Oe)
Mr (emu/g)
Raw 4h 8h 12 h 16 h 20 h
149 158 154 154 159 164
73 150 149 148 142 126
2.1 5.8 7.1 8.1 7.3 7.8
The effective permittivity and effective permeability of composite specimens containing cobalt particles milled for different time as EMA fillers were measured in 2–18 GHz range and representative results are illustrated in Figs. 5 and 6. As is shown in Fig. 5(a) and (b), the real part (ε′) as well as the imaginary part (ε″) of permittivity first increases and then decreases over the measured frequency range as the ball-milling proceeds. Specifically, highest permittivity was observed when particles milled for 8 h are used as fillers. Interfacial polarization between conductive metal particles and the insulating matrix is the main mechanism that domains the dielectric properties of composite medias [14,24,25]. On the other hand, the forming of local conducting network (LCN) can also be contributive for obtaining high permittivity according to the percolation theory [26,27]. Co particles of flake shape present larger specific surface area to compare with spherical particles and thus contribute to higher overall interfacial polarization. Additionally, flaky particles are more likely to form LCN in the wax matrix and thus further enhance the permittivity. Another factor that influences the interfacial polarization is the conductivity of metal fillers. As ball-milling proceeds, conductivity of cobalt particles are believed decrease due to increased defects, just as we observed in precious study [9]. Decreased conductivity is helpful to hinder the transfer of electro charge and thus lead to the decreased permittivity in the later ball-milling process. Obvious dielectric relaxation is observed, especially when Co particles milled for 8 or 12 h are used as fillers, as shown in Figs. 5 (a) and 4(b). This relaxation arises from the interfacial polarization.
Fig. 5. The frequency dependence of electromagnetic properties of specimens using cobalt particles milled for different time as fillers, (a) the real part of permittivity, (b) the imaginary part of permittivity, (c) the real part of permeability, (d) the imaginary part of permeability, the volume fraction for all specimens is 23 vol%.
C. Liu et al. / Journal of Magnetism and Magnetic Materials 416 (2016) 53–60
57
Fig. 6. The frequency dependence of electromagnetic parameters of samples with cobalt particles milled for 12 h as fillers, the volume fraction of fillers is 10 vol%, 23 vol% and 30 vol%, respectively, (a) the real part of permittivity, (b) the imaginary part of permittivity, (c) the real part of permeability and (d) the imaginary part of permeability.
Similar behavior was previously predicted by Youngs [28,29] and Bowler [24]. It was pointed out that the relaxation frequency (fr) was positively related to the conductivity of metal particles and often located in the optical or infrared band. It's possible to decrease fr to microwave band by decreasing the conductivity of metal particles or replacing the metal particles with metal granular film. In the present work, flaky Co particles possess much lower conductivity comparing to bulk Co and thus contribute to the dielectric relaxation in microwave band. As is shown in Fig. 5(c) and (d), the real part (μ′) and the imaginary part (μ″) of permeability present tendency similar to that of permittivity as ball-milling proceeds and peak permeability is observed when particles milled for 12 h are used as fillers. Besides, μ″ is observed decrease as the frequency increases from 2 to 18 GHz, indicating a resonance occur at frequency below 2 GHz. This resonance is observed shift to higher frequency and then drop down to lower frequency in Fig. 5(d). The natural resonance is believed dominating the complex permeability within the microwave band, as previously reported [15,30]. Basically, the natural resonance frequency fr for a particles can be expressed as follows [31]:
fr =
γ [Hk+4πMs(D⊥ − De )][Hk+4πMs(Dh − De )] 2π
(3)
where Hk and γ are the magneto crystalline anisotropy field and gyromagnetic ratio, D⊥ is the perpendicular demagnetization factor, Dh and De are the parallel demagnetization factors along with the hard and easy magnetization axis, respectively. According to Osborn' theoretical work on demagnetizing factors of the general ellipsoid, there are approximate relationships as follows [32]:
D⊥ = Dh = De = 1/3 for spheres
(4)
D⊥ ≈ 1, Dh = De ≈ 0 for thin films
(5)
Thus, combined Eqs. (3)–(5), the following relationships are obtained,
fr =
γ Hk for spheres 2π
(6)
fr =
γ Hk[Hk + 4πMsD⊥] for thin films 2π
(7)
According to Eqs. (6) and (7), Hk and D⊥ greatly influence the resonance frequency. For a ferromagnetic crystal with an easy axis [100] in cubic system or [0001] in hexagonal system, Hk can be expressed as follows:
Hk =
K1 μ 0 Ms
(8)
where K1 and μ0 are the magneto crystalline anisotropy constant and static initial susceptibility. Once milled, sphere particles are flattened into flake ones, which result in a great increase of D⊥ . Besides, The transformation from fcc to hcp phase leads to increased K1, which then contributes to the increase in Hk according to Eq. (8). Eventually, both changes lead to increased fr and the permeability in the band is thus greatly improved. Slight decrease is observed when the ball milling is prolonged to more longer than 12 h, as shown in the figure. This is due to that D⊥ decreased while
58
C. Liu et al. / Journal of Magnetism and Magnetic Materials 416 (2016) 53–60
aspect ratio of particles decreased since D⊥ is positively related to aspect ratio [32], which has been discussed in our previous work about FeSi flaky particles [9]. Additionally, eddy current effect is believed partially suppressed when thin flaky particles are used as fillers. The μ′ of raw cobalt particles is observed drops below 1 when frequency is higher than 2.5 GHz, indicating the influence from the eddy current effect. Similar phenomenon is also observed in 10.5–18 GHz band when particles milled for 4 hours are used. After milled for longer time, μ′ below 1 is no longer observed. The decreased thickness is helpful to suppress the eddy current effects. However, μ′ of particles milled for 12 h decrease sharply and is lower than that of particles milled for 8 or 16 h at high frequency band, suggesting the effect from an extra factor. The locally distributed eddy current related to LCNs may lead to the decrease of permeability. Co flakes with high aspect ratio are likely to overlap for forming LCNs in the wax matrix. The effect of filling fraction was also investigated in the current work. As shown in Fig. 6(a) and (b), both ε′ and ε″ keep increasing as the filler's fraction increases and meanwhile enhancement of dielectric relaxation appears. The surface area and total number of LCN increase as the filling fraction of Co fillers increase, which is helpful to enhance interfacial polarization. This result further convinces the LCN's influence and interfacial polarization in previous explanation about Fig. 5(a) and (b). Both μ′ and μ″ in the whole frequency range keeps increasing except that μ′ of 30 vol% filled specimens is lower than 1 when the frequency is higher than 8 GHz, as is shown in Fig. 6(c) and (d).This exception is reasonable according to the LCN model discussed previously. Increased
number of LCN strongly enhanced the eddy current effects, thus leads to a dramatically decrease in μ′ in the high frequency band. 3.4. EMA performance The frequency dependency of reflection loss (RL) of coatings containing cobalt particles as fillers is illustrated on Fig. 7. As is shown, the frequency of RL peak ( fm ) is observed shift first to lower and then back to higher frequency as the milling time increases, in coatings of various thickness. Specifically, the fm are about 9, 4.5, 4.4, 6.3, 8 GHz for particles milled for 4, 8, 12, 16, 20 h at a given thickness of 2.5 mm, respectively. This indicates that flaky particles with larger aspect ratio are likely to present RL peak at lower frequency, which is of technology importance for application in L–S band. Similar result are found about controlling of fm in our previous research on FeSi particles [9]. The tendency of fm along with milling time at a given thickness can be explained by quarter-wavelength cancellation theory. The cancellation is described as follows. When an electromagnetic wave is normally incident to a coating with two interfaces A (outside) and B (inside), it will be reflected back at both interfaces. If the coating thickness is equal to odd times of quarter-wavelength, two reflected wave will cancel at interfaces A as they are out of phase by odd times of π. Therefore, there is a formula to describe this relationship [33]:
dm =
nc (n=1, 3, 5…) 4fm Abs( μr εr )
(9)
Fig. 7. The frequency dependence of the reflection loss of samples with cobalt milled for different time as fillers at different coating thickness, the volume fraction of fillers is 23 vol%, (a) 1.0mm, (b) 1.5 mm, (c) 2.00 mm, and (d) 2.5 m.
C. Liu et al. / Journal of Magnetism and Magnetic Materials 416 (2016) 53–60
59
suppress eddy current effects while large diameter will enhance this effect at high frequency band due to local connected network. The relationship between RL peak frequency and the coating thickness was successfully illustrated by quarter-wavelength cancellation. It also indicates that increased permittivity and permeability is helpful to compel RL peak shifting to lower frequency or reduce the coating thickness. In addition, coatings with cobalt particles milled for 20 h present a RL of 33 dB at 8 GHz at the thickness of 2.5 mm. Its effective absorption bandwidth (below 10 dB) almost covers 6–10 GHz. Moreover, coatings with cobalt particles milled for 12 h present a RL of 9 dB at 8.4 GHz at the thin thickness of 1.5 mm.
Acknowledgment This work was financially supported by the Natural Science Foundation of China (Grant no. 51201048, and 51503192). J.T. Jiang thanks the Ph.D. Programs Foundation of Ministry of Education of China (Grant no. 20112302120021) and the SAST Foundation for the support.
References
Fig. 8. The frequency dependency of the reflection loss of samples with cobalt particles milled for 12 h at different thickness, the volume fraction of fillers is 23 vol%; and the relationship between RL peak frequency (fm) and matching thickness, dot point data is measured and solid line data is calculated based on Eq. (9).
where Abs( μr εr ) is absolute value of μr εr . This formula has been successfully utilized in previous researches [33,34]. In current work, as flake particles with higher aspect ratio possess higher product of μr εr , it presents a lower fm at a given thickness. From Fig. 8, it not only suggests a relationship between fm and dm but also states that variation of fm along thickness is less sensitive to variation of dm at lower fm . This is meaningful when fabricating coatings in practice since a precise control of coating thickness is very difficult. Compared to coatings with raw particles or particles milled for 8 h or 12 h as fillers, coatings with particles milled for 4 h, 16 h or 20 h possess much lower RL at thickness of 1.5 mm, 2.0 mm and 2.5 mm. Specifically, coating using particles milled for 20 h as fillers presents a RL of 33 dB at around 8 GHz together with a wide effective absorbing (RL below 10 dB) bandwidth covering 6– 10 GHz at thickness of 2.5 mm. This lowest RL is almost twice that of coatings using particles milled for 12 h as fillers. This enhanced EMA performance is due to improved impedance matching.
4. Summary In this work, flaky cobalt particles of different size were prepared in large quantity through a ball-milling process. Particles with large diameter and thin thickness present high permittivity and high permeability due to enhanced interfacial polarization and intensified natural resonance. Besides, phase transition for cobalt from fcc phase to hcp phase are believed contribute to enhancing of the natural resonance. Thin thickness is helpful to
[1] M.S. Cao, W.L. Song, Z.L. Hou, B. Wen, J. Yuan, The effects of temperature and frequency on the dielectric properties, electromagnetic interference shielding and microwave-absorption of short carbon fiber/silica composites, Carbon 48 (2010) 788–796. [2] J. Yuan, H.J. Yang, Z.L. Hou, W.L. Song, H. Xu, Y.Q. Kang, H.B. Jin, X.Y. Fang, M. S. Cao, Ni-decorated SiC powders: enhanced high-temperature dielectric properties and microwave absorption performance, Powder Technol. 237 (2013) 309–313. [3] M. Han, W. Tang, W. Chen, H. Zhou, L. Deng, Effect of shape of Fe particles on their electromagnetic properties within 1–18 GHz range, J. Appl. Phys. 107 (2010) 09A958. [4] X. Li, X. Guo, T. Liu, X. Zheng, J. Bai, Shape-controlled synthesis of Fe nanostructures and their enhanced microwave absorption properties at L-band, Mater. Res. Bull. 59 (2014) 137–141. [5] X. Zheng, J. Feng, Y. Zong, H. Miao, X. Hu, J. Bai, X. Li, Hydrophobic graphene nanosheets decorated by monodispersed superparamagnetic Fe3O4 nanocrystals as synergistic electromagnetic wave absorbers, J. Mater. Chem. C 3 (2015) 4452–4463. [6] S.S. Kim, S.T. Kim, Y.C. Yoon, K.S. Lee, Magnetic, dielectric, and microwave absorbing properties of iron particles dispersed in rubber matrix in gigahertz frequencies, J. Appl. Phys. 97 (2005) 10F905. [7] Y. Naito, K. Suetake, Application of ferrite to electromagnetic wave absorber and its characteristics, Microw. Theory Tech., IEEE Trans. 19 (1971) 65–72. [8] Y. Gong, L. Zhen, J. Jiang, C. Xu, W. Shao, Synthesis and microwave electromagnetic properties of CoFe alloy nanoflakes prepared with hydrogen-thermal reduction method, J. Appl. Phys. 106 (2009) 064302. [9] C. Liu, Y. Yuan, J.T. Jiang, Y.X. Gong, L. Zhen, Microwave absorption properties of FeSi flaky particles prepared via a ball-milling process, J. Magn. Magn. Mater. 395 (2015) 152–158. [10] J. Liu, Y. Feng, T. Qiu, Synthesis, characterization, and microwave absorption properties of Fe–40 wt% Ni alloy prepared by mechanical alloying and annealing, J. Magn. Magn. Mater. 323 (2011) 3071–3076. [11] W. Yang, L. Qiao, J. Wei, Z. Zhang, T. Wang, F. Li, Microwave permeability of flake-shaped FeCuNbSiB particle composite with rotational orientation, J. Appl. Phys. 107 (2010) 033913. [12] F. Ma, Y. Qin, Y.Z. Li, Enhanced microwave performance of cobalt nanoflakes with strong shape anisotropy, Appl. Phys. Lett. 96 (2010) 202507. [13] J. Li, J. Huang, Y. Qin, F. Ma, Magnetic and microwave properties of cobalt nanoplatelets, Mater. Sci. Eng.: B 138 (2007) 199–204. [14] Y. Feng, T. Qiu, Enhancement of electromagnetic and microwave absorbing properties of gas atomized Fe-50 wt% Ni alloy by shape modification, J. Magn. Magn. Mater. 324 (2012) 2528–2533. [15] M. Han, D. Liang, J. Xie, L. Deng, Effect of attrition time on the microwave permeability of magnetic Fe–Si–Al flakes, J. Appl. Phys. 111 (2012) 07A317. [16] R. Yang, W. Liang, Microwave absorbing characteristics of flake-shaped FeNiMo/epoxy composites, J. Appl. Phys. 113 (2013) 17A315. [17] R.B. Yang, W.F. Liang, S.T. Choi, C.K. Lin, The effects of size and shape of iron particles on the microwave absorbing properties of composite absorbers, Magn., IEEE Trans. 49 (2013) 4180–4183. [18] S. Kim, S. Jo, K. Gueon, K. Choi, J. Kim, K. Churn, Complex permeability and permittivity and microwave absorption of ferrite-rubber composite at X-band frequencies, Magn., IEEE Trans. 27 (1991) 5462–5464. [19] Y. Duan, S. Gu, Z. Zhang, M. Wen, Characterization of structures and novel
60
[20]
[21] [22] [23] [24]
[25]
[26]
[27]
C. Liu et al. / Journal of Magnetism and Magnetic Materials 416 (2016) 53–60
magnetic response of Fe 87.5 Si 7 Al 5.5 alloy processed by ball milling, J. Alloy. Compd. 542 (2012) 90–96. X. Wang, R. Gong, P. Li, L. Liu, W. Cheng, Effects of aspect ratio and particle size on the microwave properties of Fe–Cr–Si–Al alloy flakes, Mater. Sci. Eng.: A 466 (2007) 178–182. M. Nishikawa, E. Kita, T. Erata, A. Tasaki, Enhanced magnetization in Co/MgO multilayer thin films, J. Magn. Magn. Mater. 126 (1993) 303–306. S. Sun, C. Murray, Synthesis of monodisperse cobalt nanocrystals and their assembly into magnetic superlattices, J. Appl. Phys. 85 (1999) 4325–4330. H. Sato, O. Kitakami, T. Sakurai, Y. Shimada, Y. Otani, K. Fukamichi, Structure and magnetism of hcp-Co fine particles, J. Appl. Phys. 81 (1997) 1858–1862. N. Bowler, Designing dielectric loss at microwave frequencies using multilayered filler particles in a composite, Dielectr. Electr. Insul., IEEE Trans. 13 (2006) 703–711. L. Zhen, Y. Gong, J. Jiang, C. Xu, W. Shao, P. Liu, J. Tang, Synthesis of CoFe/Al2O3 composite nanoparticles as the impedance matching layer of wideband multilayer absorber, J. Appl. Phys. 109 (2011) 07A332. J. Clerc, G. Giraud, J. Laugier, J. Luck, The electrical conductivity of binary disordered systems, percolation clusters, fractals and related models, Adv. Phys. 39 (1990) 191–309. A. Lagarkov, A. Sarychev, Electromagnetic properties of composites containing
elongated conducting inclusions, Phys. Rev. B 53 (1996) 6318. [28] I. Youngs, N. Bowler, K. Lymer, S. Hussain, Dielectric relaxation in metal-coated particles: the dramatic role of nano-scale coatings, J. Phys. D: Appl. Phys. 38 (2005) 188. [29] I. Youngs, N. Bowler, O. Ugurlu, Dielectric relaxation in composites containing electrically isolated particles with thin semi-continuous metal coatings, J. Phys. D: Appl. Phys. 39 (2006) 1312. [30] P. Zhou, J. Xie, Y. Liu, L. Deng, Composition dependence of microstructure, magnetic and microwave properties in ball-milled FeSiB nanocrystalline flakes, J. Magn. Magn. Mater. 320 (2008) 3390–3393. [31] R. Walser, W. Win, P. Valanju, Shape-optimized ferromagnetic particles with maximum theoretical microwave susceptibility, Magn., IEEE Trans. 34 (1998) 1390–1392. [32] J. Osborn, Demagnetizing factors of the general ellipsoid, Phys. Rev. 67 (1945) 351. [33] Z. Li, G. Lin, L. Kong, Microwave reflection characteristics of Co Z barium ferrite composites with various volume concentration, Magn., IEEE Trans. 44 (2008) 2255–2261. [34] T. Wang, R. Han, G. Tan, J. Wei, L. Qiao, F. Li, Reflection loss mechanism of single layer absorber for flake-shaped carbonyl-iron particle composite, J. Appl. Phys. 112 (2012) 104903.