Microwave absorption properties of Sr2FeMoO6 nanoparticles

Physica B 406 (2011) 2168–2171

Contents lists available at ScienceDirect

Physica B journal homepage: www.elsevier.com/locate/physb

Microwave absorption properties of Sr2FeMoO6 nanoparticles L. Xi n, X.N. Shi, Z. Wang, Y.L. Zuo, J.H. Du Key Laboratory for Magnetism and Magnetic Materials of Ministry of Education, Lanzhou University, Lanzhou 730000, PR China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 26 November 2010 Received in revised form 11 March 2011 Accepted 11 March 2011 Available online 22 March 2011

The microwave absorption properties of nanosized double perovskite Sr2FeMoO6 and epoxy resin composites were investigated in the frequency range of 2–18 GHz using the coaxial method. The Sr2FeMoO6 composites with an optimal 20 wt% epoxy resin showed a strong electromagnetic attenuation of  49.3 dB at 8.58 GHz with a matching thickness of 2.15 mm. Moreover the optimum absorption frequency at which the reflection loss is less than  20 dB, which corresponds to 99% reflection loss of the incident microwave, is from 5.7 to 13.2 GHz with the matching thickness ranging from 3.0 to 1.5 mm. The excellent microwave-absorption properties are a consequence of a proper electromagnetic match due to the existence of the insulating matrix of anti-site defects and anti-phase domains, which not only contribute to the dielectric loss but also to the reduced eddy current loss. & 2011 Elsevier B.V. All rights reserved.

Keywords: Double perovskites Microwave absorption properties Magnetic properties

1. Introduction Recently with the rapid development of information technology, serious electromagnetic inference pollution has triggered great interest in finding effective electromagnetic wave absorption materials with properties of wide and strong absorption frequency range, low density and high resistivity. Up to now several systems were investigated, such as nanoparticles of 3d-transition metals [1] and improved microwave absorption properties of their core/shell structures [2] due to the enhanced dielectric loss and the reduced eddy current loss by forming insulating shells. However, the process to obtain the core/shell structure is usually complicated. For application, it is essential to develop a material with good microwave absorption performance by a simple method. Double perovskite Sr2FeMoO6 is a kind of spintronics material due to its half-metallic properties [3], which has large potential application in spintronic devices. Samples of Sr2FeMoO6 are usually prepared by the solid-state reaction or sol–gel method. However, it is prone to form anti-site defects (ASD), anti-phase domains (APD) and intrinsic disorder under certain conditions. The insulating matrix of the anti-phase defect and intrinsic disorder inside the Sr2FeMoO6 nanoparticles can reduce the eddy current loss and make it possible to be used as a microwave absorption material. In this paper, Sr2FeMoO6 nanoparticles were fabricated by the sol–gel method at different sintering temperatures, which are the key factors to get the ideal half-metallic properties of S2FeMoO6 [4].

n

Corresponding author. E-mail address: [email protected] (L. Xi).

0921-4526/$ - see front matter & 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.physb.2011.03.024

The microwave absorption properties of 80 wt% Sr2FeMoO6 nanoparticles and 20 wt% epoxy resin composites in the range of 2–18 GHz were determined by the coaxial method and a remarkable performance of electromagnetic wave absorption properties was reported.

2. Experiment details Sr2FeMoO6 nanoparticles were prepared by the sol–gel method. Stoichiometric amounts of analytical grade (NH4)6Mo7 O24  4H2O, Sr(NO3)2 and Fe(NO3)3  9H2O were used as starting materials. These compounds were dissolved in deionized water. Citric acid and glycol were added to the solutions with constant churning using a glass rod. The gels were dried at 110 1C for 2 days and then decomposed at 600 1C for 4 h in air in order to remove the elements of C, H and N. Later, the precursors were ground in an agate mortar and pelletized for further annealing at 900 1C for 6 h in air. The pelletized sample was annealed at different temperatures of 900, 950 and 1000 1C, for 6 h in air. Finally, the samples were annealed at the above-mentioned corresponding temperatures for 2.5 h in a H2/Ar (3.7%/96.3%) reduction flow and named as S900, S950 and S1000, respectively. Phase analysis was carried out using powder X-ray diffraction (XRD) technique with Cu Ka radiation. Magnetic properties were measured using a Vibrating Sample Magnetometer at room temperature. The Sr2FeMoO6 sample was evenly mixed with epoxy at a mass ratio of 80% by ultrasound dispersion at 42 1C for 30 min. Toroidal Sr2FeMoO6/epoxy samples with inner diameter of 3.04 mm, outer diameter of 7 mm and thickness of 1–2 mm were prepared to fit well with the coaxial sample holder for microwave measurements. The complex relative permeability

L. Xi et al. / Physica B 406 (2011) 2168–2171

(mr ¼ m0 jm00 ) and permittivity (er ¼ e0  je00 ) of the composites were measured by the coaxial method on an Agilent E8363B vector network analyzer within the frequency range of 2–18 GHz.

3. Results and discussion

(044)

(116)()332)

(022)

2000

Fig. 2. One can see that although Ms increases significantly from 4.0 to 11.0 emu/g with the sintering temperature, it is quite small compared to the reported one at room temperature [4]. The low Ms indicates the existence of disorder in the sample since the crystal structure and the cell parameters are the same as those of the ideal Sr2FeMoO6. Effects of thermal disturbance, size effect, anti-site defects (ASD), anti-phase domains (APD) and intrinsic disorder are believed to be the most probable sources reducing the magnetic moment in double perovskite [5]. ASD results from the exchange of Fe and Mo moments among the B0 and B sites of a double perovskite structure. APD originates from two coherent crystallites facing each other with different octahedral B0 and B sites but is occupied by similar Fe or Mo atoms forming the planes of either strong antiferromagnetic Fe–O–Fe or weak antiferromagnetic Mo–O–Mo bonds [8–10]. On the other hand, the intrinsic disorder in the lattice structure increases with the sintering temperature. This is indirectly proved by the presence of a magnetic spin glass phase by the ac susceptibility measurement [5]. Besides, for nanoparticles, the broken exchange bonds and the translational symmetry breaking of the lattice at the surface will induce extra disordered spins. All of the disorders lead to decrease of the saturation magnetization and formation of insulator matrix [11,12], which may be used to improve the dielectric loss and reduce the eddy current loss. Fig. 3 shows the relative complex permeability and permittivity spectra of Sr2FeMoO6/epoxy composites. As shown in Fig. 3(a), for all three samples, values of m0 and m00 change slightly in the range of 2–18 GHz. However, the values of m00 decrease in the low frequency range and then vary with the increase of frequency for each sample. The variations occurred at different frequencies, which have been marked by black arrows and the changing frequency moves to low value as the sintering temperature increases. It is generally agreed that the permeability spectrum is mainly determined by the natural resonance in the magnetic microwave absorption material [13]. According to the natural resonance equation [14] 2pfr ¼ gHa, where g is the gyromagnetic ratio and Ha ¼49K19/3m0Ms, increasing Ms will result in decreasing Ha with the sintering temperature increasing as shown in Fig. 2. Moreover, the lattice defects, interior stress as well as magnetic exchange coupling, etc. resulting from the non-equilibrium Sr2FeMoO6 particles can also induce the extra effective anisotropy field [15–17]. Thus, the effective anisotropy field decreases with temperature increase. As a result, it is reasonable to consider that the natural resonance frequencies for the three samples shift to 12

observed calculated background difference phase

S950 D = 48nm

10 S900 S950 S1000

8 6

1000

4 M (emu/g)

intensity (arb. unit)

1000

(024)(132)

S1000 D = 49nm

2000

(004)(220)

(112)(020)

3000

(224)(040)

Fig. 1 shows the XRD patterns of the three samples. One can see that the diffraction peak positions of different samples are the same. All peaks in the diffraction patterns are consistent with those of the pure Sr2FeMoO6 phase [5] with tetragonal crystal structure. The Rietveld refinement method was used to analyse the occupation probability of Fe and Mo ions, since the anti-site defects were usually observed for double perovskites. Typical experimental, calculated XRD patterns and their difference of S950 are shown in Fig. 1(b). One can see that the calculated and the experimental XRD patterns are consistent with each other. The fitted reliable parameter Rwp values are 5.9%, 2.6% and 3.4% for S900, S950 and S1000 samples, respectively, indicating the reliable refinement results. The fitted anti-site defects decrease from 0.375 to 0.207 as the annealing temperature increases from 900 to 1000 1C, respectively. Thus, one can see a decreasing trend of the anti-site defects with the increase of annealing temperature, which is consistent with literature [6]. The average size of grains was calculated according to Scheller’s formula D ¼kl/B cos y, where k is a constant (0.89), l is the wavelength of the X-ray, B is the width at the halfmaximum of the peak (FWHM) and y is the diffraction angle of the strongest diffraction peak. The Sr2FeMoO6 grain size is estimated to be about 42, 48 and 49 nm for S900, S950 and S1000, respectively. Thus, an increasing trend of grain size with the sintering temperature is shown. The theoretical saturation magnetization of Sr2FeMoO6 is around 4mB/f.u, which equals 52.88 emu/g, and the ferromagnetism results from a long range ferrimagnetic coupling with Fe–O–Mo orderly arrangement as shown by Monte Carlo simulation [7]. The hysteresis loops of S900, S950 and S1000 are displayed in

2169

0

S900 D = 42nm

1000

2 0 -2 -4 -6 -8 -10

0 10

20

30

40

50

60

70

80

90

2θ (degrees) Fig. 1. XRD pattern of S900, S950 and S1000 and the Rietveld refinement of powder XRD data for S950.

-12 -15000

-10000

-5000

0 H (Oe)

5000

10000

Fig. 2. Magnetic hysteresis loop of S900, S950 and S1000.

15000

2170

L. Xi et al. / Physica B 406 (2011) 2168–2171

1.6

0

complex permeability

-10 1.2

-20 μ', μ', μ',

0.8

μ" S900 μ" S950 μ" S1000

-40

0.4

S900

-50 0 ε', ε', ε',

30 25

Reflection loss (dB)

0.0

Complex permittivity

1.5mm 2.15mm 2.5mm 3.0mm

-30

ε" S900 ε" S950 ε" S1000

20 15 10

-10 -20 -30

1.5mm 2.0mm 2.5mm 3.0mm

S950

0

5

-5

0 2

4

6

8

10

12

14

16

18

-10 1.5mm 2.0mm 2.5mm 3.0mm

f (GHz)

-15 Fig. 3. Frequency dependence of the relative complex permeability and permittivity of SFMO/epoxy composites. The black arrow shows the points of the resonance frequency.

S1000 -20 2

low frequency band as temperature increases due to the remarkable decrease of the anisotropy field of Ha. This is significant when using them as electromagnetic wave absorption materials in the microwave range [2]. Fig. 3(b) shows the real part e0 and imaginary part e00 of the relative complex permittivity spectra of S900, S950 and S1000. One can see that the value of e00 of S900 is smaller than that of S1000 in the whole frequency range. It can be explained as follows: according to the free-electron theory [18], e00 ¼1/2prf, where r is the resistivity. Thus, the higher resistivity indicates the lower value of e00 of a sample. In our sample the ASD and APD form an insulating matrix, which increases the resistivity of the sample. Thus, as mentioned above, S900 has the lowest saturation magnetization, indicating the largest defect in this sample, which results in the highest resistivity of S900 and the lowest permittivity. Fig. 4 shows the frequency dependence of the reflection loss (RL) of Sr2FeMoO6/epoxy composites, which is calculated from the relative permeability and permittivity at given frequency and absorber thickness based on the transmission line theory [19]: Zin ¼ Z0 ðmr =er Þ1=2 tanh½jð2pfd=cÞ=ðmr er Þ1=2 

ð1Þ

RL ¼ 20log9ðZin Z0 Þ=ðZin þ Z0 Þ9

ð2Þ

where f is the frequency of incident electromagnetic wave, d is the absorber thickness, c is the velocity of light, Z0 is the impedance of free space and Zin is the input impedance of absorber. From Fig. 4, one can see that the microwave absorption frequency at which the reflection loss is less than  10 dB, which corresponds to 90% reflection loss of incident microwave, ranges from 5.3 to 14.0 GHz for absorber thickness of 3.0–1.5 mm for sample S950. The absorption frequency of reflection loss less than 20 dB ranges from 5.7 to13.2 GHz for the absorption thickness of 3.0–1.5 mm for sample S950. It is noteworthy that the width of the frequency interval is distinctly wider than that for most existent absorbers [20]. Furthermore, the absorption peak of the

4

6

8

10 f (GHz)

12

14

16

18

Fig. 4. Frequency dependence of the microwave reflection loss of the SFMO/epoxy composites.

maximum reflection loss is located at 8.58 GHz with matching thickness of 2.15 mm, and reaches  49.3 dB for sample S900. Thus, Sr2FeMoO6/epoxy composites can be considered as an effective microwave absorbent for practical applications.

4. Conclusion In conclusion, Sr2FeMoO6/epoxy composites exhibit strong microwave absorption with RL less than 10 dB in the frequency range of more than 8 GHz for absorber thicknesses of 1.5–3.0 mm. The optimal reflection loss of  49.3 dB is obtained at 8.58 GHz and absorber thickness of 2.15 mm. The excellent microwave absorption properties mainly result from the proper EM matching achieved by ASD, APD and the intrinsic disorder in the lattice structure. Thus, double perovskite Sr2FeMoO6 is also promising as a new type of microwave absorption material with a relative simple preparation procedure.

Acknowledgment This work is supported by National Natural Science Foundation of China (no. 50701021) and the Fundamental Research Funds for the Central Universities (no. lzujbky-2009-53). References [1] Z. Han, D. Li, M. Tong, X. Wei, R. Skomski, W. Liu, Z.D. Zhang, D.J. Sellmyer, J. Appl. Phys. 107 (2010) 09A929. [2] X.F. Zhang, X.L. Dong, H. Huang, B. Lv, J.P. Lei, C.J. Choi, J. Phys. D 40 (2007) 5383.

L. Xi et al. / Physica B 406 (2011) 2168–2171

[3] Claudia Felser, Gerhard H. Fecher, Benjamin Balke, Angew. Chem. Int. Ed. 46 (2007) 668. [4] B. Jurca, J. Berthon, N. Dragoe, P. Berthet, J. Alloys Compd. 474 (2009) 416. [5] R.N. AsokPoddar, I. Bhowmik, Panneer Muthuselvam, Nilanjan Das, J. Appl. Phys. 106 (2009) 073908. [6] L. Balcells, J. Navarro, M. Bibes, A. Roig, B. Martı´nez, J. Fontcuberta, Appl. Phys. Lett. 78 (2001) 781. [7] A.S. Ogale, S.B. Ogale, R. Ramesh, T. Venkatesan, Appl. Phys. Lett. 75 (1999) 537. [8] D. Topwal, D.D. Sarma, H. Kato, Y. Tokura, M. Avignon, Phys. Rev. B 73 (2006) 094419. [9] J.M. Greneche, M. Venkatesan, R. Suryanarayanan, J.M.D. Coey, Phys. Rev. B 63 (2001) 174403. [10] T. Asaka, X.Z. Yu, Y. Kaneko, T. Nagai, K. Kimoto, K. Ishizuka, Y. Tokura, Y. Matsui, Phys. Rev. B 75 (2007) 184440.

2171

[11] T. Zhu, B.G. Shen, J.R. Sun, H.W. Zhao, W.S. Zhan, Appl. Phys. Lett. 78 (2001) 3863. [12] Ll. Balcells, J. Fontcuberta, B. Martınez, X. Obradors, Phys. Rev. B 58 (1998) R14697. [13] J. Ma, J.G. Li, X. Ni, X.D. Zhang, J.J. Huang, Appl. Phys. Lett. 95 (2009) 102505. [14] C. Kittel, Phys. Rev. 73 (1948) 155. [15] D.A. Dimitrov, G.M. Wysin, Phys. Rev. B 51 (1995) 11947. [16] V.P. Shilov, J.C. Bacri, F. Gazeau, F. Gendron, R. Perzynski, Y.L. Raikher, J. Appl. Phys. 85 (1999) 6642. [17] F. Bødker, S. Mørup, S. Linderoth, Phys. Rev. Lett. 72 (1994) 282. [18] S. Ramo, J.R. Whinnery, T.V. Duzer, Fields and Waves in Communication Electronics, Wiley, New York, 1984, p. 181. [19] Y. Naito, K. Suetake, IEEE Trans. Microwave Theory Tech. 19 (1971) 65. [20] Y.L. Cheng, J.M. Dai, X.B. Zhu, D.J. Wu, Z.R. Yang, Y.P. Sun, Nanoscale Res. Lett. 4 (2009) 1153.