- Email: [email protected]
Facile synthesis of novel cobalt particles by reduction method and their microwave absorption properties
Powder Technology 264 (2014) 128–132
Contents lists available at ScienceDirect
Powder Technology journal homepage: www.elsevier.com/locate/powtec
Facile synthesis of novel cobalt particles by reduction method and their microwave absorption properties Shulai Wen, Ying Liu ⁎, Xiuchen Zhao, Jingwei Cheng, Hong Li School of Materials Science and Engineering, Beijing Institute of Technology, Beijing 100081, People's Republic of China
a r t i c l e
i n f o
Article history: Received 9 January 2014 Received in revised form 12 May 2014 Accepted 17 May 2014 Available online 24 May 2014 Keywords: Cobalt particles Permeability Permittivity Reflection loss
a b s t r a c t Three kinds of cobalt particles with different morphologies and crystal structures were synthesized via reducing cobaltous sulfate (CoSO4·7H2O) by hydrazine hydrate (N2H4·H2O) under ultrasonic wave, in which seignette salt (C4H4KNaO6·4H2O) or C6O7H5Na3·2H2O acted as complexing agent, C16H33(CH3)3NBr acted as surfactant agent and sodium hydroxide (NaOH) acted as pH regulator. Less saturation magnetization and more coercivity were obtained compared to single hcp-cobalt and bulk cobalt, respectively. The electromagnetic properties of cobalt particles dispersed in paraffin (70 wt.%) were measured in the microwave frequency range of 1–18 GHz. The three kinds of cobalt particles all have multi-nonlinear dielectric resonances due to atomic and electronic polarization. The cobalt particles composed by nanosheets exhibit higher real part of permeability in the range of 10–18 GHz compared to spherical cobalt particles. The reflection loss values of cobalt particles were calculated according to the transmission-line theory, and the minimal reflection loss of − 19.06 dB at 17.42 GHz was observed corresponding to a thickness of 5 mm. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Magnetic materials have been widely investigated to eliminate the serious electromagnetic interference problems recently. For the ferromagnetic metal particles, large values of permeability can be obtained in GHz range due to the Snoek's limit, because their saturation magnetization is higher than ferrites, such as FeCo nanoparticles [1], hexagonal Fe microflakes [2], Fe nanotubes [3], and hierarchical dendrite-like Fe [4]. As a typical magnetic metal, the electromagnetic and microwave properties of cobalt particles have been investigated in recent years. Wang et al. [5] reported that at thickness of 2 mm, flower-like cobalt with sharp petals possessed the maximum reflection loss of −13.6 dB. Tong et al. [6] found that flower-like cobalt superstructures showed excellent microwave absorption performances, with a minimum reflection loss of −40 dB, corresponding to a matching thickness of 2.5 mm. It is well-known that the electromagnetic microwave absorption properties of the cobalt particles depend on their morphologies and crystal structure, so present research is focused on their synthesis of cobalt particles with specific morphologies and crystal structure. Cobalt particles possess three crystal structures (FCC, HCP and BCC) [7] and various morphologies, such as hollow cobalt mesospheres [8], ring-shaped cobalt nanomaterials [9], cobalt tubes [10,11], cobalt rods [12,13], and cobalt wires [14–17]. However, few investigations have been focus on dependence of electromagnetic and microwave properties on the crystal structures and morphologies of cobalt particles. It is valuable to ⁎ Corresponding author. E-mail addresses: [email protected], [email protected] (Y. Liu).
http://dx.doi.org/10.1016/j.powtec.2014.05.030 0032-5910/© 2014 Elsevier B.V. All rights reserved.
obtain excellent electromagnetic microwave absorption via adjusting the crystal structures and morphologies of cobalt particles. In the present paper, three kinds of cobalt particles, with nanosheets, porous surface and smooth surface, were successfully synthesized by facile reduction method via reducing cobaltous sulfate (CoSO4·7H2O) by hydrazine hydrate (N2H4·H2O) under ultrasonic wave, in which seignette salt (C4H4KNaO6·4H2O) or C6O7H5Na3·2H2O acted as complexing agent, C16H33(CH3)3NBr acted as surfactant agent and sodium hydroxide (NaOH) acted as pH regulator. We investigated the electromagnetic and microwave properties of cobalt particles dispersed in paraffin (70 wt.%) in detail. The three kinds of cobalt particles all have multi-nonlinear dielectric resonances due to atomic and electronic polarization. The cobalt particles composed by nanosheets exhibit higher real part of permeability in the microwave frequency range of 10–18 GHz compared to spherical cobalt particles. The high yields, simple instrument and mild conditions would make the method good prospect in future large-scale applications. 2. Experimental 2.1. Preparation of cobalt particles Cobaltous sulfate (CoSO4·7H2O), seignette salt (C4O6H4KNa·4H2O), sodium citrate (C6O7H5Na3·2H2O), hydrazine hydrate (N2H4·H2O), cetyl trimethyl ammonium bromide (C16H33(CH3)3NBr) and sodium hydroxide (NaOH) were purchased from Beijing Tongguangxincheng Co. Ltd. All chemicals used in this work were of analytical grade and were used as received without further purification. The synthesis of
S. Wen et al. / Powder Technology 264 (2014) 128–132
129
cobalt particles was carried out in a 3-necked flask equipped with an ultrasound reactor. The three kinds of cobalt particles, S1, S2 and S3, were processed as follows: For S1, 2.8 g CoSO4·7H2O, 1 g C16H33(CH3)3NBr (C16TAB) and 12.6 g C6O7H5Na3·2H2O were dissolved in 150 mL deionized water under vigorous stirring for 40 min at 40 °C, followed by 4 g NaOH. The reaction temperature, the power and the ultrasonic frequency were fixed at 90 °C, 1400 W and 40 kHz, respectively. After that, 0.5 mL N2H4·H2O (80 wt.%) was quickly added to the above solution, then 10 min later, the solution was cooled to room temperature. The dark gray precipitates were separated, washed with deionized water and absolute ethanol several times, and dried under vacuum at 40 °C for 24 h to collect the cobalt particles. For S2, 1.4 g CoSO4·7H2O, 1 g C16H33(CH3)3NBr (C16TAB) and 8.4 g C4O6H4KNa·4H2O were dissolved in 150 mL deionized water under vigorous stirring for 40 min at 40 °C, followed by 14 g NaOH. Then the reaction temperature, the power and the ultrasonic frequency were fixed at 65 °C, 1400 W and 40 kHz, respectively. After that, 0.5 mL N2H4·H2O (80 wt.%) was quickly added to the above solution, and 10 min later, the solution was cooled to room temperature. The dark gray precipitates were separated, washed with deionized water and absolute ethanol several times, and dried under vacuum at 40 °C for 24 h to collect the cobalt particles. For S3, 2.8 g CoSO4·7H2O, 4 g C16H33(CH3)3NBr (C16TAB) and 12.6 g C4O6H4KNa·4H2O were dissolved in 150 mL deionized water under vigorous stirring for 40 min at 40 °C, followed by 14 g NaOH. Then the reaction temperature, the power and the ultrasonic frequency were fixed at 90 °C, 1400 W and 40 kHz, respectively. After that, 1.3 mL N2H4·H2O (80 wt.%) was quickly added to the above solution, and 10 min later, the solution was cooled to room temperature. The dark gray precipitates were separated, washed with deionized water and absolute ethanol several times, and dried under vacuum at 40 °C for 24 h to collect the cobalt particles. 2.2. Instruments and measurements The X-ray diffraction (XRD) patterns were recorded on a Bruker D8 Advance diffractometer in Mo Kα radiation (λ = 0.7093 Å) operated at 50 kV and 30 mA. The scanning electron microscopy (SEM) images were obtained using a QUANTA600 scanning electron microscope operated at 25 kV. The magnetization measurement was performed by a DMS vibrating sample magnetometer (VSM) at room temperature. The relative complex permittivity (εr = ε′ − jε″) and relative complex permeability (μr = μ′ − jμ″) were determined using the T/R coaxial line method in the range of 2–18 GHz with an O-ring shaped sample (i.d. = 3 mm and o.d. = 7 mm, thickness = 2 mm) using an HP8722ESS vector network analyzer. All measurements were performed at room temperature. 3. Results and discussion The crystal structure and phase purity of the as-synthesized samples were determined by XRD. Fig. 1 shows the typical X-ray diffraction patterns of S1, S2 and S3. For S1, the characteristic peaks at 2θ = 18.856°, 20.193°, 21.402°, 32.911°, 35.957° and 38.865° can be well indexed to the (100), (002), (101), (110), (103) and (112) planes of hexagonalphase cobalt (space group: P63/mmc (194); JCPDS card: 05-0727, a = 2.503 Å, c = 4.0621 Å), so S1 is hcp-cobalt structure. The peaks of S2 and S3 at 2θ = 18.856°, 21.402° and 35.957° match very well with the (100), (101) and (103) planes of hcp-cobalt (space group: P63/mmc (194); JCPDS card: 05-0727, a = 2.503 Å, c = 4.0621 Å), respectively, and the peaks at 2θ = 19.957°, 32.878°, and 38.759° can be well
Fig. 1. The XRD patterns of the as-prepared samples.
indexed to the (111), (220) and (311) planes of fcc-cobalt (space group: Fm3m (225); JCPDS card: 15-0806, a = 3.545 Å), respectively. Due to several fcc-cobalt peaks overlapping with hcp-cobalt peaks, the structures of samples are distinguished by the relative intensity. The above result shows that S1 and S2 are mixture of fcc-cobalt and hcpcobalt, respectively. Fig. 2 shows the SEM images of cobalt particles. Cobalt particles with a diameter of about 5 μm were obtained (S1). The cobalt particles were assembled by nanosheets with a thickness of about 200 nm. S2 particles are spherical cobalt particles with porous surface, and the diameter is about 6 μm. S3 particles are spherical cobalt particles with smooth surface, and the size is about 7 μm. The magnetic property of the cobalt particles is manifested in the M–H loop acquired by VSM measurement, as shown in Fig. 3. The saturation magnetization (Ms) values of S1, S2 and S3 are 90.559, 117.50 and 123.00 emu/g, respectively. The coercivity values of S1, S2 and S3 are 176.67, 190.89 and 156.80 Oe, respectively. Less saturation magnetization was achieved compared to the hcp-Co single crystal (162 emu/g) [18]. This may be attributed to the effective Bohr magneton number reduced due to the pinning effect of the ions or atom absorbed on the surface of flow-like cobalt particles, such as [C4H4O6]2 − and oxygen atom. Moreover, as-obtained cobalt particles exhibit larger coercivity compared to bulk cobalt (10 Oe) [19,20], and there maybe three reasons: firstly, according to the fanning mode [21,22], Hc is inversely proportional to R3, so Hc of cobalt particles is larger because cobalt particles are in micron scale; secondly, hcp-cobalt possesses larger magnetocrystalline anisotropy energy, which would enhance coercivity of cobalt particles; thirdly, it is well known that coercivity is related to magnetic domain and magnetic moment, and the interface between hcp-cobalt and fcc-cobalt can improve the difficulty of movement of magnetic domain and rotation of magnetic moment. The coercivity of S1, S2 and S3 are larger than that of bulk cobalt crystal, respectively. The frequency dependency of the complex permittivity (εr = ε′ − jε″) and complex permeability (μr = μ′ − jμ″) for cobalt particles–paraffin composites are shown in Fig. 4. For S3, the real part of permittivity is larger than that for S1 and smaller than that for S2 (Fig. 4a). The phenomenon can be explained on the basis of space charge polarization model of Wagner [23] and Maxwell [24] and is in agreement with the Koop's phenomenological theory. According to the Koop's model, complex permittivity behaviors are attributed to the interfacial space-charge polarization arising from the heterogeneous structure of the samples. Cobalt particles dispersed in paraffin can act as charge centers, and the interfacial space-charge polarization would appear due to charges' motion hindered in these interfaces differently when charges in cobalt are made to move by the external electric field. For S1, there is only one kind of interface: cobalt particle and paraffin. For S2 and S3, there are two kinds of interfaces: cobalt particle and paraffin and fcc-cobalt and hcp-cobalt, composed by two crystal structures, respectively. Moreover, S2 has more surface than
130
S. Wen et al. / Powder Technology 264 (2014) 128–132
Fig. 2. The SEM images of the cobalt particles. (a) S1, (b) S2, and (c) S3.
S3 as shown in Fig. 2. The curve of the real part of permittivity is S2 N S3 N S1. The imaginary parts of S1, S2 and S3 exhibit multiple non-linear dielectric resonances (Fig. 4b). A similar phenomenon has been observed in Fe-phthalocyanine oligomer/Fe3O4 [25]. The permittivity originates from orientation polarization, atomic polarization and electronic polarization. Normally, the resonance caused by vacancy or pores usually dominates in the low-frequency regions. The resonances in middle and high frequency are attributed to atomic and electronic polarization. Multiple non-linear resonance peaks of complex permittivity can be interpreted as the results of atomic and electronic polarization [26,27]. For S1, S2 and S3 as shown in Fig. 4(c), due to both eddy current loss and ferromagnetic resonance, the real part of permeability deceases with frequency, exhibiting excellent frequency dispersion. Compared to S1, higher real parts of permeability for S1 and S2 were observed in the frequency range of 10–18 GHz. In general, the real part of permeability decreases with the frequency, due to both the current loss and ferromagnetic resonance. At the beginning due to the negatively effect on magnetic permeability caused by having larger grain boundaries, the cobalt particles assembled by nanosheets exhibit smaller values than spherical cobalt particles. With frequency increasing, the spherical cobalt particles have larger current loss compared to the cobalt particles assembled by nanosheets. So the spherical cobalt particles possess a smaller value of μ′ than that of cobalt particles assembled by nanosheets in the range of 10–18 GHz. A wide resonance peak is only observed for S3 whilst two overlapped resonances are observed for both S1 and S2 (Fig. 4d). A similar broad magnetic-resonance peak has been observed for Co/TiO2 [28], (Fe, Co)/Al2O3 [29] and Ni/TiO2 [30]. For two overlapped resonances, the first one may be the natural resonance of hcp-
cobalt particles (about 6.5 GHz) [31], and the second one may be in relation to the exchange resonance of cobalt particles [32]. The resonance broadened may be interpreted as a consequence of size and morphology of the cobalt particles. On one hand, the micrometer-sized particles, owing to their large size with respect to a magnetic wall, are made up of several magnetic domains, and the large resonance band may be interpreted as a consequence of their magnetic polydomain configuration. Indeed, it is well known that, for partially magnetized samples, the internal demagnetizing fields resulting from the distribution of magnetic domains contribute to the resonance broadening. On the other hand, the frequency band broadening is also related to the morphology of the particles because of the effect of the demagnetization fields which are related to the shapes of cobalt particles [33,34]. The similar phenomenon has been reported for nickel hollow sphere [35]. As a typical magnetic material, the loss of cobalt particles mainly originates from the magnetic loss. In general, the microwave magnetic loss of magnetic materials is mostly associated with magnetic hysteresis, domain wall resonance, natural resonance [36], and exchange resonance [37]. In the current study, the permeability of the as-obtained cobalt particles could be ascribed to natural resonance, and exchange resonance, rather than magnetic hysteresis and domain wall resonance. This is because magnetic hysteresis stemming from irreversible magnetization occurs only in a highly applied field, whereas resonance due to domain wall movement derived from multi-domain materials occurs only in the low frequency (b2 GHz). However, resonance resulting from spin rotational component occurs at high-frequency range [38]. −1 2 ¼ 23 πμ 0 d σ To confirm the above results, the equation μ ″ ðμ 0 Þ−2 f [39,40], where μ0 is the vacuum permeability, was used to investigate the eddy current loss contribution to the imaginary part of permeability μ″ relative to the thickness d and the electric conductivity σ of the composites. According to the equation, if the magnetic loss only stems from the eddy current loss, then the values of μ″(μ′)−2f−1 should be constant when the frequency is changed. The value of μ″(μ′)−2f−1 of the cobalt particles was shown in Fig. 5. The values changing with frequency in the range of 1–18 GHz were observed, suggesting that the magnetic loss of S1, S2 and S3 chiefly caused not only by eddy current loss but also by natural resonance and exchange resonance, respectively. Interestingly, two peaks for S1 and S2 were observed, and the mechanism was further investigated. Dependence of the reflection loss of cobalt particles–paraffin composites on the frequency in the range of 1–18 GHz is shown in Fig. 5. It is calculated according to the transmission-line theory as follows [41]: Z in ¼
Fig. 3. The hysteresis loop of the cobalt particles (at room temperature).
rffiffiffiffiffi pffiffiffiffiffiffiffiffiffi 2πfd μ r εr μr tanh j c εr
Z −1 RL ¼ 20 log in Z in þ 1
ð1Þ
ð2Þ
S. Wen et al. / Powder Technology 264 (2014) 128–132
131
Fig. 4. Electromagnetic parameter of cobalt particles dispersed in paraffin in the range of 1–18 GHz. (a) Real part of permittivity; (b) imaginary part of permittivity; (c) real part of permeability; (d) imaginary part of permeability; (e) dielectric loss; (f) magnetic loss.
where f is the frequency of the electromagnetic wave, d the thickness of the absorber, c the velocity of light, Z0 the impedance of free space, and Zin the input impedance of absorber. According to Eqs. (1) and (2), the simulations of the reflection loss for S1, S2 and S3 were shown in Fig. 6, respectively.
As shown in Fig. 6a, the minimal reflection loss of S1 is − 7.25 dB at 18 GHz corresponding to 5.5 mm, and reflection loss values exceeding − 10 dB are not obtained. The minimal reflection loss of − 12.57 dB for S2 is observed at 14.94 GHz corresponding to the thickness of 5 mm, and the reflection loss values exceeding − 10 dB are obtained in the range from 13.99 to 15.84 GHz (Fig. 6b). An optimal reflection loss value of S3 is − 19.06 dB at 17.42 GHz. For the thickness of 5 mm the reflection loss values exceeding − 10 dB are observed in the range from 16.15 to 18 GHz (Fig. 6c). From all abovementioned results, less reflection loss for S1 is achieved compared to S2 and S3, which resulted from lower dielectric and magnetic loss. For S2 and S3, it is suggested that larger dielectric or magnetic loss results in larger reflection loss. The minimal reflection loss of S2 and S3 not corresponding to the magnetic loss implies an efficient complementation between the permittivity and permeability. The reflection loss of the cobalt particles dispersed in paraffin is dielectric and magnetic dual mechanism, not magnetic single mechanism. 4. Conclusions
Fig. 5. Frequency dependence of μ″(μ′)−2f−1.
In summary, we successfully synthesized three kinds of cobalt particles with different morphologies and crystal structures by reduction method in liquid phase. We investigated the electromagnetic and
132
S. Wen et al. / Powder Technology 264 (2014) 128–132
Fig. 6. Frequency dependence of reflection loss of cobalt particles dispersed in paraffin for layer of different thickness. (a) S1; (b) S2; (c) S3.
microwave absorption properties of cobalt particles dispersed in paraffin (70 wt.%) in the microwave frequency range of 1–18 GHz. The three kinds of cobalt particles all have multi-nonlinear dielectric resonances due to both atomic polarization and electronic polarization. The cobalt particles assembled by nanosheets exhibit higher real part of permeability in the frequency range of 10–18 GHz. The reflection losses of cobalt particles were calculated according to the transmission-line theory, and the minimal reflection loss of − 19.06 dB at 17.42 GHz was observed corresponding to a thickness of 5 mm. References [1] W.S. Seo, J.H. Lee, X.M. Sun, Y. Suzuki, D. Mann, Z. Liu, M. Terashima, P.C. Yang, M.V. Mcconnell, D.G. Nishimura, H.J. Dai, Nat. Mater. 5 (2006) 971–976. [2] L.S. Fu, J.T. Jiang, C.Y. Xu, L. Zhen, CrystEngComm 14 (2012) 6827–6832. [3] G.X. Tong, M. Hong, J.G. Guan, J.H. Yuan, W.H. Wu, H.S. Qian, Micro Nano Lett. 6 (2011) 722–724. [4] G.B. Sun, B.X. Dong, M.H. Cao, B.Q. Wei, C.W. Hu, Chem. Mater. 23 (2011) 1587–1593. [5] C. Wang, X.J. Han, X.L. Zhang, S.R. Hu, T. Zhang, J.Y. Wang, Y.C. Du, X.H. Wang, P. Xu, J. Phys. Chem. C 114 (2010) 14826–14830. [6] G.X. Tong, J.H. Yuan, W.H. Wu, Q. Hu, H.S. Qian, L.C. Li, J.P. Shen, CrystEngComm 14 (2012) 2071–2079. [7] Victor F. Puntes, Kannan Krishnan, A. Paul Alivisatos, Top. Catal. 19 (2002) 145–148. [8] G. Lin, L. Fang, X.G. Wen, L. He, W.Z. Zheng, C.P. Chen, Q.P. Zhong, Adv. Funct. Mater. 17 (2007) 425–430. [9] L. Guo, F. Liang, N. Wang, D.S. Kong, S.M. Wang, L. He, C.P. Chen, X.M. Meng, Z.Y. Wu, Chem. Mater. 20 (2008) 5163–5168. [10] G. Ji, H. Su, S. Tang, Y. Du, B. Xu, Chem. Lett. 34 (2005) 86. [11] T.A. Crowley, K.J. Ziegler, D.M. Lyons, D. Erts, H. Olin, M.A. Morris, J.D. Holmes, Chem. Mater. 15 (2003) 3519. [12] F. Xu, Y. Bando, D. Gelberg, M. Hasegawa, M. Mitome, Acta Mater. 52 (2004) 601. [13] M. Aslam, R. Bhobe, N. Alem, S. Donthu, V.P. Dravid, J. Appl. Phys. 98 (2005) 074311. [14] D. Qin, M.H. Liu, Chem. Phys. Lett. 350 (2001) 51.
[15] A. Bantu, J. Rivas, G. Zaragoza, M.A. López-Quintela, M.C. Blanco, J. Non-Cryst. Solids 287 (2001) 5. [16] J. Xu, X. Huang, G. Xie, Y. Fang, D. Liu, Mater. Lett. 50 (2005) 981. [17] L. Vila, P. Vincent, L.D. Pra, G. Pirio, E. Minoux, L. Gangloff, S. Demoustier-Champagne, N. Sarazin, E. Ferain, R. Legras, L. Piraux, P. Legagneux, Nano Lett. 4 (2004) 521. [18] Y.L. Li, J.Z. Zhao, X.D. Su, J. Colloid Interface Sci. 336 (2009) 41–45. [19] S. Chikazumi, Physics of Magnetism, JohnWiley & Sons, NewYork, 1964. [20] Y.J. Zhang, Q. Yao, Y. Zhang, T.Y. Cui, Cryst. Growth Des. 8 (2008) 3206–3212. [21] I.S. Jacobs, C.P. Bean, Phys. Rev. 100 (1955) 1060–1067. [22] W. Zhou, K. Zheng, L. He, R.M. Wang, L. Guo, C.P. Chen, X.D. Han, Z. Zhang, Nano Lett. 4 (2008) 1147–1152. [23] K.W. Wagner, Ann. Phys. (Leipzig) 40 (1913) 817. [24] J.C. Maxwell, Electricity and magnetism, vol 1, Oxford University Press, Oxford, 1929. Section 328. [25] F.B. Meng, R. Zhao, Y.Q. Zhan, Y.J. Lei, J.C. Zhong, X.B. Liu, Appl. Surf. Sci. 257 (2011) 5000–5006. [26] Z.B. Li, B. Shen, Y.D. Deng, L. Liu, W.B. Hu, Appl. Surf. Sci. 255 (2009) 4542–4546. [27] Z.B. Li, Y.D. Deng, B. Shen, W.B. Hu, J. Phys. D. Appl. Phys. 42 (2009) 145002–145006. [28] X. Ni, J. Ma, J.G. Li, D.M. Jiao, J.J. Huang, X.D. Zhang, J. Alloys Compd. 468 (2009) 368. [29] X.G. Liu, D.Y. Geng, H. Meng, B. Li, Q. Zhang, D.J. Kang, Z.D. Zhang, J. Phys. D. Appl. Phys. 41 (2008) 175001. [30] N.K. Sun, B.S. Du, F. Liu, P.Z. Si, M.X. Zhao, X.Y. Zhang, G.M. Shi, J. Alloys Compd. 577 (2013) 533–537. [31] G. Viau, F. Ravel, O. Acher, F. Fiévet-Vincent, F. Fiévet, J. Appl. Phys. 76 (1994) 6570. [32] Philippe Toneguzzo, Olivier Acher, IEEE Trans. Magn. 35 (1999) 3469–3471. [33] D. Polder, J. Smit, Rev. Mod. 89 (1953) 25. [34] Ph. Toneguzzo, G. Viau, O. Acher, J. Mater. Sci. 35 (2000) 3767–3784. [35] Y.D. Deng, L. Zhao, B. Shen, L. Liu, W.B. Hu, J. Appl. Phys. 100 (2006) 014304. [36] M. Jamet, W. Wernsdorfer, C. Thirion, D. Mailly, V. Dupuis, P. Méinon, A. Pérez, Phys. Rev. Lett. 86 (2001) 4676. [37] G. Bertotti, J. Appl. Phys. 57 (1985) 2110–2117. [38] Z.B. Li, Y.D. Deng, B. Shen, W.B. Hu, Mater. Sci. Eng. B 164 (2009) 112–115. [39] A. Aharoni, J. Appl. Phys. 39 (1991) 7762–7764. [40] F.S. Wen, H.B. Yi, L. Qiao, H. Zheng, D. Zhou, F.S. Li, Appl. Phys. Lett. 92 (2008) 042507. [41] M.Z. Wu, T.D. Zhang, S. Hui, T.D. Xiao, S.H. Ge, W.A. Hines, J.L. Budnick, G.W. Taylor, Appl. Phys. Lett. 80 (2002) 4404–4406.
Contents lists available at ScienceDirect
Powder Technology journal homepage: www.elsevier.com/locate/powtec
Facile synthesis of novel cobalt particles by reduction method and their microwave absorption properties Shulai Wen, Ying Liu ⁎, Xiuchen Zhao, Jingwei Cheng, Hong Li School of Materials Science and Engineering, Beijing Institute of Technology, Beijing 100081, People's Republic of China
a r t i c l e
i n f o
Article history: Received 9 January 2014 Received in revised form 12 May 2014 Accepted 17 May 2014 Available online 24 May 2014 Keywords: Cobalt particles Permeability Permittivity Reflection loss
a b s t r a c t Three kinds of cobalt particles with different morphologies and crystal structures were synthesized via reducing cobaltous sulfate (CoSO4·7H2O) by hydrazine hydrate (N2H4·H2O) under ultrasonic wave, in which seignette salt (C4H4KNaO6·4H2O) or C6O7H5Na3·2H2O acted as complexing agent, C16H33(CH3)3NBr acted as surfactant agent and sodium hydroxide (NaOH) acted as pH regulator. Less saturation magnetization and more coercivity were obtained compared to single hcp-cobalt and bulk cobalt, respectively. The electromagnetic properties of cobalt particles dispersed in paraffin (70 wt.%) were measured in the microwave frequency range of 1–18 GHz. The three kinds of cobalt particles all have multi-nonlinear dielectric resonances due to atomic and electronic polarization. The cobalt particles composed by nanosheets exhibit higher real part of permeability in the range of 10–18 GHz compared to spherical cobalt particles. The reflection loss values of cobalt particles were calculated according to the transmission-line theory, and the minimal reflection loss of − 19.06 dB at 17.42 GHz was observed corresponding to a thickness of 5 mm. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Magnetic materials have been widely investigated to eliminate the serious electromagnetic interference problems recently. For the ferromagnetic metal particles, large values of permeability can be obtained in GHz range due to the Snoek's limit, because their saturation magnetization is higher than ferrites, such as FeCo nanoparticles [1], hexagonal Fe microflakes [2], Fe nanotubes [3], and hierarchical dendrite-like Fe [4]. As a typical magnetic metal, the electromagnetic and microwave properties of cobalt particles have been investigated in recent years. Wang et al. [5] reported that at thickness of 2 mm, flower-like cobalt with sharp petals possessed the maximum reflection loss of −13.6 dB. Tong et al. [6] found that flower-like cobalt superstructures showed excellent microwave absorption performances, with a minimum reflection loss of −40 dB, corresponding to a matching thickness of 2.5 mm. It is well-known that the electromagnetic microwave absorption properties of the cobalt particles depend on their morphologies and crystal structure, so present research is focused on their synthesis of cobalt particles with specific morphologies and crystal structure. Cobalt particles possess three crystal structures (FCC, HCP and BCC) [7] and various morphologies, such as hollow cobalt mesospheres [8], ring-shaped cobalt nanomaterials [9], cobalt tubes [10,11], cobalt rods [12,13], and cobalt wires [14–17]. However, few investigations have been focus on dependence of electromagnetic and microwave properties on the crystal structures and morphologies of cobalt particles. It is valuable to ⁎ Corresponding author. E-mail addresses: [email protected], [email protected] (Y. Liu).
http://dx.doi.org/10.1016/j.powtec.2014.05.030 0032-5910/© 2014 Elsevier B.V. All rights reserved.
obtain excellent electromagnetic microwave absorption via adjusting the crystal structures and morphologies of cobalt particles. In the present paper, three kinds of cobalt particles, with nanosheets, porous surface and smooth surface, were successfully synthesized by facile reduction method via reducing cobaltous sulfate (CoSO4·7H2O) by hydrazine hydrate (N2H4·H2O) under ultrasonic wave, in which seignette salt (C4H4KNaO6·4H2O) or C6O7H5Na3·2H2O acted as complexing agent, C16H33(CH3)3NBr acted as surfactant agent and sodium hydroxide (NaOH) acted as pH regulator. We investigated the electromagnetic and microwave properties of cobalt particles dispersed in paraffin (70 wt.%) in detail. The three kinds of cobalt particles all have multi-nonlinear dielectric resonances due to atomic and electronic polarization. The cobalt particles composed by nanosheets exhibit higher real part of permeability in the microwave frequency range of 10–18 GHz compared to spherical cobalt particles. The high yields, simple instrument and mild conditions would make the method good prospect in future large-scale applications. 2. Experimental 2.1. Preparation of cobalt particles Cobaltous sulfate (CoSO4·7H2O), seignette salt (C4O6H4KNa·4H2O), sodium citrate (C6O7H5Na3·2H2O), hydrazine hydrate (N2H4·H2O), cetyl trimethyl ammonium bromide (C16H33(CH3)3NBr) and sodium hydroxide (NaOH) were purchased from Beijing Tongguangxincheng Co. Ltd. All chemicals used in this work were of analytical grade and were used as received without further purification. The synthesis of
S. Wen et al. / Powder Technology 264 (2014) 128–132
129
cobalt particles was carried out in a 3-necked flask equipped with an ultrasound reactor. The three kinds of cobalt particles, S1, S2 and S3, were processed as follows: For S1, 2.8 g CoSO4·7H2O, 1 g C16H33(CH3)3NBr (C16TAB) and 12.6 g C6O7H5Na3·2H2O were dissolved in 150 mL deionized water under vigorous stirring for 40 min at 40 °C, followed by 4 g NaOH. The reaction temperature, the power and the ultrasonic frequency were fixed at 90 °C, 1400 W and 40 kHz, respectively. After that, 0.5 mL N2H4·H2O (80 wt.%) was quickly added to the above solution, then 10 min later, the solution was cooled to room temperature. The dark gray precipitates were separated, washed with deionized water and absolute ethanol several times, and dried under vacuum at 40 °C for 24 h to collect the cobalt particles. For S2, 1.4 g CoSO4·7H2O, 1 g C16H33(CH3)3NBr (C16TAB) and 8.4 g C4O6H4KNa·4H2O were dissolved in 150 mL deionized water under vigorous stirring for 40 min at 40 °C, followed by 14 g NaOH. Then the reaction temperature, the power and the ultrasonic frequency were fixed at 65 °C, 1400 W and 40 kHz, respectively. After that, 0.5 mL N2H4·H2O (80 wt.%) was quickly added to the above solution, and 10 min later, the solution was cooled to room temperature. The dark gray precipitates were separated, washed with deionized water and absolute ethanol several times, and dried under vacuum at 40 °C for 24 h to collect the cobalt particles. For S3, 2.8 g CoSO4·7H2O, 4 g C16H33(CH3)3NBr (C16TAB) and 12.6 g C4O6H4KNa·4H2O were dissolved in 150 mL deionized water under vigorous stirring for 40 min at 40 °C, followed by 14 g NaOH. Then the reaction temperature, the power and the ultrasonic frequency were fixed at 90 °C, 1400 W and 40 kHz, respectively. After that, 1.3 mL N2H4·H2O (80 wt.%) was quickly added to the above solution, and 10 min later, the solution was cooled to room temperature. The dark gray precipitates were separated, washed with deionized water and absolute ethanol several times, and dried under vacuum at 40 °C for 24 h to collect the cobalt particles. 2.2. Instruments and measurements The X-ray diffraction (XRD) patterns were recorded on a Bruker D8 Advance diffractometer in Mo Kα radiation (λ = 0.7093 Å) operated at 50 kV and 30 mA. The scanning electron microscopy (SEM) images were obtained using a QUANTA600 scanning electron microscope operated at 25 kV. The magnetization measurement was performed by a DMS vibrating sample magnetometer (VSM) at room temperature. The relative complex permittivity (εr = ε′ − jε″) and relative complex permeability (μr = μ′ − jμ″) were determined using the T/R coaxial line method in the range of 2–18 GHz with an O-ring shaped sample (i.d. = 3 mm and o.d. = 7 mm, thickness = 2 mm) using an HP8722ESS vector network analyzer. All measurements were performed at room temperature. 3. Results and discussion The crystal structure and phase purity of the as-synthesized samples were determined by XRD. Fig. 1 shows the typical X-ray diffraction patterns of S1, S2 and S3. For S1, the characteristic peaks at 2θ = 18.856°, 20.193°, 21.402°, 32.911°, 35.957° and 38.865° can be well indexed to the (100), (002), (101), (110), (103) and (112) planes of hexagonalphase cobalt (space group: P63/mmc (194); JCPDS card: 05-0727, a = 2.503 Å, c = 4.0621 Å), so S1 is hcp-cobalt structure. The peaks of S2 and S3 at 2θ = 18.856°, 21.402° and 35.957° match very well with the (100), (101) and (103) planes of hcp-cobalt (space group: P63/mmc (194); JCPDS card: 05-0727, a = 2.503 Å, c = 4.0621 Å), respectively, and the peaks at 2θ = 19.957°, 32.878°, and 38.759° can be well
Fig. 1. The XRD patterns of the as-prepared samples.
indexed to the (111), (220) and (311) planes of fcc-cobalt (space group: Fm3m (225); JCPDS card: 15-0806, a = 3.545 Å), respectively. Due to several fcc-cobalt peaks overlapping with hcp-cobalt peaks, the structures of samples are distinguished by the relative intensity. The above result shows that S1 and S2 are mixture of fcc-cobalt and hcpcobalt, respectively. Fig. 2 shows the SEM images of cobalt particles. Cobalt particles with a diameter of about 5 μm were obtained (S1). The cobalt particles were assembled by nanosheets with a thickness of about 200 nm. S2 particles are spherical cobalt particles with porous surface, and the diameter is about 6 μm. S3 particles are spherical cobalt particles with smooth surface, and the size is about 7 μm. The magnetic property of the cobalt particles is manifested in the M–H loop acquired by VSM measurement, as shown in Fig. 3. The saturation magnetization (Ms) values of S1, S2 and S3 are 90.559, 117.50 and 123.00 emu/g, respectively. The coercivity values of S1, S2 and S3 are 176.67, 190.89 and 156.80 Oe, respectively. Less saturation magnetization was achieved compared to the hcp-Co single crystal (162 emu/g) [18]. This may be attributed to the effective Bohr magneton number reduced due to the pinning effect of the ions or atom absorbed on the surface of flow-like cobalt particles, such as [C4H4O6]2 − and oxygen atom. Moreover, as-obtained cobalt particles exhibit larger coercivity compared to bulk cobalt (10 Oe) [19,20], and there maybe three reasons: firstly, according to the fanning mode [21,22], Hc is inversely proportional to R3, so Hc of cobalt particles is larger because cobalt particles are in micron scale; secondly, hcp-cobalt possesses larger magnetocrystalline anisotropy energy, which would enhance coercivity of cobalt particles; thirdly, it is well known that coercivity is related to magnetic domain and magnetic moment, and the interface between hcp-cobalt and fcc-cobalt can improve the difficulty of movement of magnetic domain and rotation of magnetic moment. The coercivity of S1, S2 and S3 are larger than that of bulk cobalt crystal, respectively. The frequency dependency of the complex permittivity (εr = ε′ − jε″) and complex permeability (μr = μ′ − jμ″) for cobalt particles–paraffin composites are shown in Fig. 4. For S3, the real part of permittivity is larger than that for S1 and smaller than that for S2 (Fig. 4a). The phenomenon can be explained on the basis of space charge polarization model of Wagner [23] and Maxwell [24] and is in agreement with the Koop's phenomenological theory. According to the Koop's model, complex permittivity behaviors are attributed to the interfacial space-charge polarization arising from the heterogeneous structure of the samples. Cobalt particles dispersed in paraffin can act as charge centers, and the interfacial space-charge polarization would appear due to charges' motion hindered in these interfaces differently when charges in cobalt are made to move by the external electric field. For S1, there is only one kind of interface: cobalt particle and paraffin. For S2 and S3, there are two kinds of interfaces: cobalt particle and paraffin and fcc-cobalt and hcp-cobalt, composed by two crystal structures, respectively. Moreover, S2 has more surface than
130
S. Wen et al. / Powder Technology 264 (2014) 128–132
Fig. 2. The SEM images of the cobalt particles. (a) S1, (b) S2, and (c) S3.
S3 as shown in Fig. 2. The curve of the real part of permittivity is S2 N S3 N S1. The imaginary parts of S1, S2 and S3 exhibit multiple non-linear dielectric resonances (Fig. 4b). A similar phenomenon has been observed in Fe-phthalocyanine oligomer/Fe3O4 [25]. The permittivity originates from orientation polarization, atomic polarization and electronic polarization. Normally, the resonance caused by vacancy or pores usually dominates in the low-frequency regions. The resonances in middle and high frequency are attributed to atomic and electronic polarization. Multiple non-linear resonance peaks of complex permittivity can be interpreted as the results of atomic and electronic polarization [26,27]. For S1, S2 and S3 as shown in Fig. 4(c), due to both eddy current loss and ferromagnetic resonance, the real part of permeability deceases with frequency, exhibiting excellent frequency dispersion. Compared to S1, higher real parts of permeability for S1 and S2 were observed in the frequency range of 10–18 GHz. In general, the real part of permeability decreases with the frequency, due to both the current loss and ferromagnetic resonance. At the beginning due to the negatively effect on magnetic permeability caused by having larger grain boundaries, the cobalt particles assembled by nanosheets exhibit smaller values than spherical cobalt particles. With frequency increasing, the spherical cobalt particles have larger current loss compared to the cobalt particles assembled by nanosheets. So the spherical cobalt particles possess a smaller value of μ′ than that of cobalt particles assembled by nanosheets in the range of 10–18 GHz. A wide resonance peak is only observed for S3 whilst two overlapped resonances are observed for both S1 and S2 (Fig. 4d). A similar broad magnetic-resonance peak has been observed for Co/TiO2 [28], (Fe, Co)/Al2O3 [29] and Ni/TiO2 [30]. For two overlapped resonances, the first one may be the natural resonance of hcp-
cobalt particles (about 6.5 GHz) [31], and the second one may be in relation to the exchange resonance of cobalt particles [32]. The resonance broadened may be interpreted as a consequence of size and morphology of the cobalt particles. On one hand, the micrometer-sized particles, owing to their large size with respect to a magnetic wall, are made up of several magnetic domains, and the large resonance band may be interpreted as a consequence of their magnetic polydomain configuration. Indeed, it is well known that, for partially magnetized samples, the internal demagnetizing fields resulting from the distribution of magnetic domains contribute to the resonance broadening. On the other hand, the frequency band broadening is also related to the morphology of the particles because of the effect of the demagnetization fields which are related to the shapes of cobalt particles [33,34]. The similar phenomenon has been reported for nickel hollow sphere [35]. As a typical magnetic material, the loss of cobalt particles mainly originates from the magnetic loss. In general, the microwave magnetic loss of magnetic materials is mostly associated with magnetic hysteresis, domain wall resonance, natural resonance [36], and exchange resonance [37]. In the current study, the permeability of the as-obtained cobalt particles could be ascribed to natural resonance, and exchange resonance, rather than magnetic hysteresis and domain wall resonance. This is because magnetic hysteresis stemming from irreversible magnetization occurs only in a highly applied field, whereas resonance due to domain wall movement derived from multi-domain materials occurs only in the low frequency (b2 GHz). However, resonance resulting from spin rotational component occurs at high-frequency range [38]. −1 2 ¼ 23 πμ 0 d σ To confirm the above results, the equation μ ″ ðμ 0 Þ−2 f [39,40], where μ0 is the vacuum permeability, was used to investigate the eddy current loss contribution to the imaginary part of permeability μ″ relative to the thickness d and the electric conductivity σ of the composites. According to the equation, if the magnetic loss only stems from the eddy current loss, then the values of μ″(μ′)−2f−1 should be constant when the frequency is changed. The value of μ″(μ′)−2f−1 of the cobalt particles was shown in Fig. 5. The values changing with frequency in the range of 1–18 GHz were observed, suggesting that the magnetic loss of S1, S2 and S3 chiefly caused not only by eddy current loss but also by natural resonance and exchange resonance, respectively. Interestingly, two peaks for S1 and S2 were observed, and the mechanism was further investigated. Dependence of the reflection loss of cobalt particles–paraffin composites on the frequency in the range of 1–18 GHz is shown in Fig. 5. It is calculated according to the transmission-line theory as follows [41]: Z in ¼
Fig. 3. The hysteresis loop of the cobalt particles (at room temperature).
rffiffiffiffiffi pffiffiffiffiffiffiffiffiffi 2πfd μ r εr μr tanh j c εr
Z −1 RL ¼ 20 log in Z in þ 1
ð1Þ
ð2Þ
S. Wen et al. / Powder Technology 264 (2014) 128–132
131
Fig. 4. Electromagnetic parameter of cobalt particles dispersed in paraffin in the range of 1–18 GHz. (a) Real part of permittivity; (b) imaginary part of permittivity; (c) real part of permeability; (d) imaginary part of permeability; (e) dielectric loss; (f) magnetic loss.
where f is the frequency of the electromagnetic wave, d the thickness of the absorber, c the velocity of light, Z0 the impedance of free space, and Zin the input impedance of absorber. According to Eqs. (1) and (2), the simulations of the reflection loss for S1, S2 and S3 were shown in Fig. 6, respectively.
As shown in Fig. 6a, the minimal reflection loss of S1 is − 7.25 dB at 18 GHz corresponding to 5.5 mm, and reflection loss values exceeding − 10 dB are not obtained. The minimal reflection loss of − 12.57 dB for S2 is observed at 14.94 GHz corresponding to the thickness of 5 mm, and the reflection loss values exceeding − 10 dB are obtained in the range from 13.99 to 15.84 GHz (Fig. 6b). An optimal reflection loss value of S3 is − 19.06 dB at 17.42 GHz. For the thickness of 5 mm the reflection loss values exceeding − 10 dB are observed in the range from 16.15 to 18 GHz (Fig. 6c). From all abovementioned results, less reflection loss for S1 is achieved compared to S2 and S3, which resulted from lower dielectric and magnetic loss. For S2 and S3, it is suggested that larger dielectric or magnetic loss results in larger reflection loss. The minimal reflection loss of S2 and S3 not corresponding to the magnetic loss implies an efficient complementation between the permittivity and permeability. The reflection loss of the cobalt particles dispersed in paraffin is dielectric and magnetic dual mechanism, not magnetic single mechanism. 4. Conclusions
Fig. 5. Frequency dependence of μ″(μ′)−2f−1.
In summary, we successfully synthesized three kinds of cobalt particles with different morphologies and crystal structures by reduction method in liquid phase. We investigated the electromagnetic and
132
S. Wen et al. / Powder Technology 264 (2014) 128–132
Fig. 6. Frequency dependence of reflection loss of cobalt particles dispersed in paraffin for layer of different thickness. (a) S1; (b) S2; (c) S3.
microwave absorption properties of cobalt particles dispersed in paraffin (70 wt.%) in the microwave frequency range of 1–18 GHz. The three kinds of cobalt particles all have multi-nonlinear dielectric resonances due to both atomic polarization and electronic polarization. The cobalt particles assembled by nanosheets exhibit higher real part of permeability in the frequency range of 10–18 GHz. The reflection losses of cobalt particles were calculated according to the transmission-line theory, and the minimal reflection loss of − 19.06 dB at 17.42 GHz was observed corresponding to a thickness of 5 mm. References [1] W.S. Seo, J.H. Lee, X.M. Sun, Y. Suzuki, D. Mann, Z. Liu, M. Terashima, P.C. Yang, M.V. Mcconnell, D.G. Nishimura, H.J. Dai, Nat. Mater. 5 (2006) 971–976. [2] L.S. Fu, J.T. Jiang, C.Y. Xu, L. Zhen, CrystEngComm 14 (2012) 6827–6832. [3] G.X. Tong, M. Hong, J.G. Guan, J.H. Yuan, W.H. Wu, H.S. Qian, Micro Nano Lett. 6 (2011) 722–724. [4] G.B. Sun, B.X. Dong, M.H. Cao, B.Q. Wei, C.W. Hu, Chem. Mater. 23 (2011) 1587–1593. [5] C. Wang, X.J. Han, X.L. Zhang, S.R. Hu, T. Zhang, J.Y. Wang, Y.C. Du, X.H. Wang, P. Xu, J. Phys. Chem. C 114 (2010) 14826–14830. [6] G.X. Tong, J.H. Yuan, W.H. Wu, Q. Hu, H.S. Qian, L.C. Li, J.P. Shen, CrystEngComm 14 (2012) 2071–2079. [7] Victor F. Puntes, Kannan Krishnan, A. Paul Alivisatos, Top. Catal. 19 (2002) 145–148. [8] G. Lin, L. Fang, X.G. Wen, L. He, W.Z. Zheng, C.P. Chen, Q.P. Zhong, Adv. Funct. Mater. 17 (2007) 425–430. [9] L. Guo, F. Liang, N. Wang, D.S. Kong, S.M. Wang, L. He, C.P. Chen, X.M. Meng, Z.Y. Wu, Chem. Mater. 20 (2008) 5163–5168. [10] G. Ji, H. Su, S. Tang, Y. Du, B. Xu, Chem. Lett. 34 (2005) 86. [11] T.A. Crowley, K.J. Ziegler, D.M. Lyons, D. Erts, H. Olin, M.A. Morris, J.D. Holmes, Chem. Mater. 15 (2003) 3519. [12] F. Xu, Y. Bando, D. Gelberg, M. Hasegawa, M. Mitome, Acta Mater. 52 (2004) 601. [13] M. Aslam, R. Bhobe, N. Alem, S. Donthu, V.P. Dravid, J. Appl. Phys. 98 (2005) 074311. [14] D. Qin, M.H. Liu, Chem. Phys. Lett. 350 (2001) 51.
[15] A. Bantu, J. Rivas, G. Zaragoza, M.A. López-Quintela, M.C. Blanco, J. Non-Cryst. Solids 287 (2001) 5. [16] J. Xu, X. Huang, G. Xie, Y. Fang, D. Liu, Mater. Lett. 50 (2005) 981. [17] L. Vila, P. Vincent, L.D. Pra, G. Pirio, E. Minoux, L. Gangloff, S. Demoustier-Champagne, N. Sarazin, E. Ferain, R. Legras, L. Piraux, P. Legagneux, Nano Lett. 4 (2004) 521. [18] Y.L. Li, J.Z. Zhao, X.D. Su, J. Colloid Interface Sci. 336 (2009) 41–45. [19] S. Chikazumi, Physics of Magnetism, JohnWiley & Sons, NewYork, 1964. [20] Y.J. Zhang, Q. Yao, Y. Zhang, T.Y. Cui, Cryst. Growth Des. 8 (2008) 3206–3212. [21] I.S. Jacobs, C.P. Bean, Phys. Rev. 100 (1955) 1060–1067. [22] W. Zhou, K. Zheng, L. He, R.M. Wang, L. Guo, C.P. Chen, X.D. Han, Z. Zhang, Nano Lett. 4 (2008) 1147–1152. [23] K.W. Wagner, Ann. Phys. (Leipzig) 40 (1913) 817. [24] J.C. Maxwell, Electricity and magnetism, vol 1, Oxford University Press, Oxford, 1929. Section 328. [25] F.B. Meng, R. Zhao, Y.Q. Zhan, Y.J. Lei, J.C. Zhong, X.B. Liu, Appl. Surf. Sci. 257 (2011) 5000–5006. [26] Z.B. Li, B. Shen, Y.D. Deng, L. Liu, W.B. Hu, Appl. Surf. Sci. 255 (2009) 4542–4546. [27] Z.B. Li, Y.D. Deng, B. Shen, W.B. Hu, J. Phys. D. Appl. Phys. 42 (2009) 145002–145006. [28] X. Ni, J. Ma, J.G. Li, D.M. Jiao, J.J. Huang, X.D. Zhang, J. Alloys Compd. 468 (2009) 368. [29] X.G. Liu, D.Y. Geng, H. Meng, B. Li, Q. Zhang, D.J. Kang, Z.D. Zhang, J. Phys. D. Appl. Phys. 41 (2008) 175001. [30] N.K. Sun, B.S. Du, F. Liu, P.Z. Si, M.X. Zhao, X.Y. Zhang, G.M. Shi, J. Alloys Compd. 577 (2013) 533–537. [31] G. Viau, F. Ravel, O. Acher, F. Fiévet-Vincent, F. Fiévet, J. Appl. Phys. 76 (1994) 6570. [32] Philippe Toneguzzo, Olivier Acher, IEEE Trans. Magn. 35 (1999) 3469–3471. [33] D. Polder, J. Smit, Rev. Mod. 89 (1953) 25. [34] Ph. Toneguzzo, G. Viau, O. Acher, J. Mater. Sci. 35 (2000) 3767–3784. [35] Y.D. Deng, L. Zhao, B. Shen, L. Liu, W.B. Hu, J. Appl. Phys. 100 (2006) 014304. [36] M. Jamet, W. Wernsdorfer, C. Thirion, D. Mailly, V. Dupuis, P. Méinon, A. Pérez, Phys. Rev. Lett. 86 (2001) 4676. [37] G. Bertotti, J. Appl. Phys. 57 (1985) 2110–2117. [38] Z.B. Li, Y.D. Deng, B. Shen, W.B. Hu, Mater. Sci. Eng. B 164 (2009) 112–115. [39] A. Aharoni, J. Appl. Phys. 39 (1991) 7762–7764. [40] F.S. Wen, H.B. Yi, L. Qiao, H. Zheng, D. Zhou, F.S. Li, Appl. Phys. Lett. 92 (2008) 042507. [41] M.Z. Wu, T.D. Zhang, S. Hui, T.D. Xiao, S.H. Ge, W.A. Hines, J.L. Budnick, G.W. Taylor, Appl. Phys. Lett. 80 (2002) 4404–4406.