- Email: [email protected]
Flowerlike iron oxide nanostructures and their application in microwave absorption
Journal of Alloys and Compounds 631 (2015) 183–191
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom
Flowerlike iron oxide nanostructures and their application in microwave absorption Chenyang Guo, Fangyuan Xia ⇑, Zhen Wang, Li Zhang, Li Xi, Yalu Zuo ⇑ 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
Article history: Received 22 August 2014 Received in revised form 14 January 2015 Accepted 15 January 2015 Available online 22 January 2015 Keywords: Self-assembled flowerlike iron oxides Microwave absorption Electromagnetic impedance match Natural resonance
a b s t r a c t Self-assembled flowerlike a-Fe2O3, Fe3O4 and c-Fe2O3 were fabricated by a simple calcination procedure. The structure characterization shows that the flowerlike morphology and the size of the nanostructures are perfectly maintained in the conversion of precursor to a-Fe2O3, Fe3O4, and c-Fe2O3. The complex permittivity and permeability results indicate that the dielectric and magnetic loss of Fe3O4 flower are both higher than those of c-Fe2O3 flower. In addition, Fe3O4 flower shows a good electromagnetic impedance match and its microwave absorption mainly originates from magnetic loss rather than dielectric loss. An optimal reflection loss of 46.0 dB is found at 3.4 GHz for flowerlike Fe3O4, which indicates that the sample can be used as a highly efficient microwave absorber. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Microwave absorbing materials based on the magnetic materials with low reflection and high absorption properties are receiving extensive attention because of their promising application in electronic devices used in commerce, industry, and military affairs [1]. In recent years, carbonyl irons were widely investigated as one of typical magnetic microwave absorbing materials [2–4]. However, the low density and absorption properties of the commercial carbonyl iron restrict its use in the applications. Hollow iron particle with low density is also a new material to improve the microwave absorption properties [5,6]. Besides, iron-based materials such as c-Fe2O3, and Fe3O4 are recognized as promising materials for microwave applications because of their relative high magnetization and antioxidation properties in air. Various iron oxides or iron composites with different morphologies such as rods, wires, tubes, flakes, urchins and dendrites have been successfully reported with their application as electromagnetic wave absorber [7–12]. However, the relationship among the structure, morphology and magnetic properties of iron-based nanomaterials has not been fully understood. As we known, large surface areas and high magnetization will improve the microwave absorption properties. Herein, we report the synthesis of novel three-dimensional (3D) flowerlike iron oxide nanostructures with large surface areas by an ethylene glycol (EG)-mediated self-assembly process. The assynthesized iron oxide precursor was transformed into iron oxide. ⇑ Corresponding author. Tel.: +86 0931 8914036; fax: +86 0931 8914160. E-mail address: [email protected] (Y. Zuo). http://dx.doi.org/10.1016/j.jallcom.2015.01.087 0925-8388/Ó 2015 Elsevier B.V. All rights reserved.
Moreover, the phase of the final product can be easily controlled to be a-Fe2O3, c-Fe2O3, or Fe3O4, three of the most common iron oxides, by altering the calcination conditions. The microwave absorption properties of a-Fe2O3, c-Fe2O3, and Fe3O4 flowerlike nanostructures with almost the same large surface areas are presented in this study. 2. Experimental The thermal decomposition of metal alkoxide is a simple route to achieving a tailored metal oxide. In a typical procedure, 2.4 g of ferric chloride (FeCl36H2O), 5.4 g of urea, and 14.4 g of tetrabutylammonium bromide were dissolved in 360 mL of ethylene glycol. The red solution was refluxed at 195 °C for 30 min. After cooling, the as-synthesized iron oxide precursor was collected as a green precipitate after centrifugation and ethanol-washing cycles. The morphology of the precursor was studied by scanning electron microscopy (SEM, Hitachi S-4800). Fig. 1a shows the SEM image of a typical sample composed of many uniform, flowerlike nanostructures approximately 3 lm in diameter. The detailed morphology of the flowerlike nanostructures is shown in Fig. 1b. The entire structure was found to be built from several dozen nanopetals with smooth surfaces. These nanopetals were 1.5 lm wide and connected to each other through the center to form 3D flowerlike structures. A series of other measurements was also performed to investigate the as-obtained iron oxide precursor. The X-ray diffraction (XRD, Philips X’Pert, with Cu Ka radiation) pattern (Fig. 1c) shows the emergence of diffraction peaks similar to those of other metal oxide precursors reported in the literature [13], especially the strong peak located in the low-angle region (2h = 11°), although the exact crystal structure of the iron oxide precursor is not yet to be determined. The effect of calcinations on the crystallization and morphology of the iron oxide precursor was investigated by thermogravimetric analysis–differential thermal analysis (TG–DTA, Perkin Elmer SII) with Ar protection. Fig. 1d shows the TG–DTA curves of the as-synthesized iron oxide precursor. A weight loss was observed at about 250–400 °C in the TG curve. The weight loss can be attributed to the combustion of the resultant organics. One sharp exothermic peak was found at 320 °C, which was accompanied by the aforementioned weight loss. No any other
184
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
Fig. 1. (a) SEM image of the as-synthesized iron oxide precursor. (b) SEM image of a single flowerlike iron oxide precursor. (c) XRD and (d) TG–DTA curves of the assynthesized iron oxide precursor.
Fig. 2. XRD patterns of the samples: (a) c-Fe2O3, (b) Fe3O4, and (c) a-Fe2O3 flowers.
weight loss in TG or any peak in DTA was observed over 450 °C, confirming that all organic components were burned out at 450 °C. Hence, the as-synthesized iron oxide precursor was transformed into a-Fe2O3 (or Fe3O4) at 450 °C in air (or under N2 protection) for 3 h. Then, phase transformation was achieved from Fe3O4 to c-Fe2O3 by an oxidation process. The microstructure of the sample was characterized by high-resolution transmission electron microscopy (HRTEM; JEOL, JEM-2010; at 200 kV) and selected area electron diffraction (SAED) analyses. X-ray photoelectron spectroscopy (XPS; VG
Fig. 3. Fe 2p core-level XPS spectra of a-Fe2O3, Fe3O4, and c-Fe2O3 flowers.
ESCALAB MK II) was used to test the chemical valence of Fe ions. The magnetic properties of the samples were measured with a vibrating sample magnetometer (VSM; LakeShore 7304 model). A toroidal sample (inner diameter, 3.04 mm; outer diameter, 7 mm; and thickness, 3 mm) was prepared to fit well the coaxial sample holder for microwave measurements. The complex relative permeability (lr = l0 jl00 ) and permittivity (er = e0 je00 ) of the composite samples were measured by the coaxial method on an Agilent E8363B vector network analyzer within the range of 0.1–18 GHz.
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
185
Fig. 4. (a) Low-magnification and (b) high-magnification SEM images of a-Fe2O3 flower. (c) Low-magnification and (d) high-magnification TEM images of a-Fe2O3 flower. (e) HRTEM image and (f) SAED patterns of a-Fe2O3 flower.
3. Results and discussion The diffraction peaks of the as-obtained red product well agreed with those for a-Fe2O3 powder (JCPDS card 80-2377) (Fig. 2c), when the precursor was calcined at 450 °C in air for 3 h. When the precursor was calcined at 450 °C under N2 protection for 3 h, the positions of all diffraction peaks of the corresponding product (Fig. 2b), well matched those of Fe3O4 (JCPDS card 87-0245). The formation of Fe3O4 can be attributed to the reductive ability of the organic species (glycolates) in the precursor. Furthermore, when as-obtained Fe3O4 was heated at 250 °C in air for 3 h, the color changed from black to red-brown. The XRD pattern of this product (Fig. 2a) was well agreed with the standard XRD pattern
of c-Fe2O3 (JCPDS card 39-1346). The three peaks between 50 and 70 °C from the red-brown sample slightly shifted to higher angles [14]. These results showed that Fe3O4 was transformed into c-Fe2O3. Herein, a simple calcination procedure was used to obtain all three common iron oxides from the same iron oxide precursor. The patterns observed in Fig. 2b and c are almost the same, matching both the patterns of Fe3O4 and c-Fe2O3. Therefore, XPS measurements (Fig. 3) were performed to unambiguously assign the crystal phase because XPS is very sensitive to Fe2+ and Fe3+ cations. The XPS spectra have been charge corrected to the adventitious C1s with binding energy of 284.6 eV. It can be seen that the peaks generally shift to a high binding energy and broaden for Fe3O4 because of the appearance of Fe2+(2p3/2) and Fe2+(2p1/2). In
186
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
Fig. 5. (a) Low-magnification and (b) high-magnification SEM images of Fe3O4 flower. (c) Low-magnification and (d) high-magnification TEM images of Fe3O4 flower. (e) HRTEM image and (f) SAED patterns of Fe3O4 flower.
this study, the levels of Fe 2p3/2 are 711.1 and 710.7 eV, Fe 2p1/2 are 724.7 and 724.3 eV for c-Fe2O3 and a-Fe2O3, respectively, as well as Fe 2p3/2 and Fe 2p1/2 are 711.2 and 724.8 eV for Fe3O4. A satellite peak at 719.6 and 719.2 eV is observed in c-Fe2O3 and a-Fe2O3 [15]. In addition, no satellite for Fe3O4 is identified, excluding the presence of c-Fe2O3 or a-Fe2O3 in the Fe3O4 samples, which well agrees with the above mentioned literature. The XPS patterns well agree with the XRD data and reveal that complete phase transformation can be achieved by a simple calcination procedure. The SEM images of the as-synthesized samples are shown in panel (a) of Figs. 4–6. Compared with the conversion of the precursor, the flowerlike morphology and size of the samples are perfectly maintained in the conversion of a-Fe2O3 to Fe3O4 and
c-Fe2O3. Panel (b) of Figs. 4–6 shows the high-magnification images of a single flower, and all flowers exhibit a similar structure, which is the same as the precursor nanoflowers. Further structural characterization of a-Fe2O3, Fe3O4, and c-Fe2O3 flowers was performed using HRTEM, as shown in panels c and d of Figs. 4–6. The results show the same morphology and size as those in the SEM images. A representative HRTEM image taken from a-Fe2O3 flower is shown in Fig. 4e. The lattice fringes are clearly visible with a spacing of 0.252 nm, corresponding to the spacing of the (1 1 0) planes of a-Fe2O3. The lattice fringes of the (2 2 0) planes of Fe3O4 with a d-spacing of 0.297 nm and the (3 1 1) planes of c-Fe2O3 with a d-spacing of 0.253 nm are also clearly observed in the insets of Figs. 5e and 6e, respectively. Furthermore, the SAED patterns were
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
187
Fig. 6. (a) Low-magnification and (b) high-magnification SEM images of c-Fe2O3 flower. (c) Low-magnification and (d) high-magnification TEM images of c-Fe2O3 flower. (e) HRTEM image and (f) SAED patterns of c-Fe2O3 flower.
measured at the edge of the nanopetals of all the flowerlike samples. Fig. 4f shows that edge of the nanopetals of a-Fe2O3 flower is single crystalline. The crystallographic orientation (1 1 0) is also observed. However, for a whole a-Fe2O3 flower, it should be polycrystalline. In addition, SAED patterns of Fe3O4 and c-Fe2O3 (Figs. 5f and 6f) exhibit diffraction rings, indicating the polycrystalline nature of the nanoflowers, and the positions of all diffraction rings of the corresponding calcined products well match those in the XRD patterns. Comparing to a-Fe2O3, the polycrystalline nature may be due to the N2 protection and further oxidation process in the calcination procedure for Fe3O4 and c-Fe2O3, respectively. Iron-based materials often have intriguing magnetic properties due to the variation of their crystal structure, shape anisotropy, and crystallinity. Especially, the value of Ms for ferromagnetic
nanostructures is associated with grain size, crystallinity, and noncollinear magnetic structure. Therefore, the magnetic properties of a-Fe2O3, Fe3O4, and c-Fe2O3 flowers were examined by VSM. Fig. 7 shows the magnetic hysteresis loops (M–H loops) of the flowerlike samples measured at room temperature. The antiferromagnetic a-Fe2O3 flowers exhibit quite small saturation magnetization (Ms) around 1 emu/g at 10 kOe, which may be caused by the uncompensated surface spin. The Ms of Fe3O4 and c-Fe2O3 nanoflowers are about 80 and 55 emu/g at 1500 Oe, which are respectively lower than 92 and 76 emu/g for bulk Fe3O4 and c-Fe2O3 materials because the spin disorder on the surface would reduce the total magnetic moment. This is commonly observed for the nanostructure materials, for example, the Ms of Fe3O4 and c-Fe2O3 dendritic microstructures are around 75 and 42 emu/g, respectively [16].
188
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
to the transmission line theory [17], the normalized input impedance Zin of a microwave absorber is given by
rffiffiffiffiffi Z in ¼
lr 2p pffiffiffiffiffiffiffiffiffi lr er fd tan h j er c
where lr and er are the relative permeability and permittivity of the composite medium, c is the velocity of electromagnetic (EM) waves in free space, f is the frequency of the microwave, and d is the thickness of the absorber. The reflection loss is related to Zin as RL pffiffiffiffiffiffiffiffiffiffiffiffiffi 0 where Z 0 ¼ l0 =e0 is the characteristic ðdBÞ ¼ 20 log ZZinin Z þZ 0 ,
Fig. 7. Magnetic hysteresis loops of (a) a-Fe2O3, (b) Fe3O4, and (c) c-Fe2O3 flowers measured at 300 K.
Fig. 8. Microwave reflection loss of epoxy resin with 80 wt.% (a) c-Fe2O3 and (b) Fe3O4 flowers at different thicknesses.
The magnetic loss and microwave absorption property of
a-Fe2O3 flower are weak because of its small saturation magnetization. Thus, the microwave absorption property was investigated only for c-Fe2O3 and Fe3O4 flowers. To reveal the microwave absorption properties, the reflection loss (RL) values of epoxy resin with 80 wt.% c-Fe2O3 and Fe3O4 flowers were calculated. According
impedance of free space. The thickness of the sample is one of the crucial parameters that affects the intensity and position of the frequency at the RL minimum. Therefore, we prepared c-Fe2O3 and Fe3O4 samples at different thicknesses in order to eliminate the influence resulting from the thickness of the samples. The results are shown in Fig. 8. It was found that the two samples exhibit excellent microwave absorption properties. For Fe3O4 flower, there is a sharp and strong peak at 3.4 GHz with the minimum RL of 46.0 dB when the thickness of the Fe3O4 sample is 5.0 mm, and a broad peak (15.0 dB, 13.0 GHz) is observed with the thickness of 2.0 mm (Fig. 8b). It is slightly higher than the result of Fe3O4 urchins, which have the optimal reflection loss of 43.2 dB at high frequency of 16.8 GHz [18]. Compared to Fe3O4 nanospheres and sponges, the author pointed out that the excellent microwave absorption performance was ascribed to their unique morphologies. Such morphologies resulted in reinforced EM parameters and multiresonant behavior. However, the result of the Fe3O4 flower is obviously better than other reports: Fe3O4 nanospheres with sizes of about 300 nm (30.3 dB, 5.5 mm), and Fe3O4 nanoparticles with the sizes of about 50–100 nm (35.0 dB, 4.4 mm) [19,20]. In addition, for c-Fe2O3 flower, the minimum RL reaches 15 dB with the thickness of 3 mm (Fig. 8a). Thus, the microwave absorption property of Fe3O4 flower is better than c-Fe2O3 flower. To investigate the possible mechanism of microwave absorption of the above two samples, we independently investigated the complex relative permittivity (e0 and e00 ) and permeability (l0 and l00 ) of the samples. Generally, complex permittivity and permeability represent the dielectric and dynamic magnetic properties of materials. The real parts of complex permittivity and permeability (e0 and l0 ) symbolize the storage and transformation capability of electromagnetic (EM) energy. The imaginary parts (e00 and l00 ) represent the loss of EM energy. The dielectric and magnetic loss tangent (tan dE = e00 /e0 and tan dM = l00 /l0 ) represent the tangent of the angle between the induced electric and magnetic field in the microwave absorber and the electric and magnetic field of the EM wave. Excellent microwave-absorbing materials are well known to have a good match between the magnetic loss and dielectric loss. The frequency dependence of e0 and e00 of c-Fe2O3 and Fe3O4 flowers is shown in Fig. 9a and b, respectively. It can be seen that both e0 and e00 have fluctuating behavior, which can be attributed to the displacement current lag at the interface between epoxy resin and flowerlike samples. This behavior was also observed in Fe3O4/ Bi nanocomposites and ZnO-coated Fe nanocapsules [21,22]. The e0 of Fe3O4 flower is higher than that of c-Fe2O3 flower. However, the e00 values of Fe3O4 and c-Fe2O3 flowers are similar, indicating that the dielectric loss of Fe3O4 and c-Fe2O3 flowers is similar. Otherwise, based on the free electron theory, e00 = 1/2pe0qf (where q is the resistivity), e00 has a lower value (0–2.2) within the entire range. Thus, low values of e00 indicate high electric resistivity. Fig. 9c shows relative complex permeability of c-Fe2O3 and Fe3O4 flowers. The real part (l0 ) of Fe3O4 flower exhibits a sharp decrease from 3.4 to 0.7 within the 0.1–18.0 GHz range. However, l0 of c-Fe2O3 flower gradually decreases from 1.5 to 0.9. Regarding
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
189
Fig. 9. Complex permittivity and permeability of epoxy resin with 80 wt.% c-Fe2O3 and Fe3O4 flowers as a function of frequency.
Fig. 10. Tan dE and tan dM of c-Fe2O3 and Fe3O4 flowers, respectively.
to the l00 –f spectra of both c-Fe2O3 and Fe3O4 flowers in Fig. 9d, three peaks are observed at about 3.0, 11.0, and 15.0 GHz, respectively. The first resonance peak at around 3.0 GHz can be attributed to natural resonance. The resonance frequency strongly depends on the saturation magnetization, crystalline anisotropy, and magnetic particle geometry [23]. The resonance peaks at around 11.0 and 15.0 GHz may be ascribed to exchange resonance. To our knowledge, multiresonance can be observed when the size of the magnetic particles is reduced. Among the modes that deal with multiresonance, the most accepted one is the exchange resonance mode placed forward by Aharoni [24]. Surface anisotropy and exchange energy caused by the exchange effect between grains have been proven to be important for nanocrystalline particles [25]. Generally speaking, the multiresonance phenomena were contributed from the natural resonance and exchange resonance [26,27]. In Fig. 9b and d, the values of e00 and l00 for Fe3O4 flower are both higher than those of c-Fe2O3 flower. This finding indicates that the dielectric and magnetic loss of Fe3O4 flower are both higher than those of c-Fe2O3 flower. For c-Fe2O3 flower in Fig. 10a, tan dE increases and exhibits a peak at 3.0 GHz, whereas tan dM shows a peak at about 13.0 GHz. The magnetic loss and dielectric loss of c-Fe2O3 flower are reportedly a mismatch. Meanwhile, for Fe3O4 flower in Fig. 10b, the trend of the tan dM curve is similar to that of tan dE, which has a few peaks within the 0.1–18.0 GHz range. This result means that the sample has a good match between the magnetic loss and dielectric loss. Thus, Fe3O4 flower has great potential for being a highly efficient microwave absorber. Furthermore, tan dM is larger than tan dE, indicating that the microwave absorption of Fe3O4 flower mainly originates from magnetic loss rather than dielectric loss. To further understand the magnetic loss, permeability spectra of c-Fe2O3 and Fe3O4 were fitted as the linear overlap of three resonance peaks P1, P2 and P3. Then, the P1, P2, P3 could be separated from each other by fitting of the Gilbert modification of Landau–Lifshitz equation [22] as
190
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
Fig. 11. The fitting curves of real part (a) and imaginary part (b) of permeability for c-Fe2O3, and real part (c) and imaginary part (d) of permeability for Fe3O4.
Table 1 Fitting parameters for permeability spectra of c-Fe2O3 (black) and Fe3O4 (blue). The f(exp) and f(fit) are the experimental and fitted resonance frequency, respectively.
l0 ¼ B þ
3 X Ii i¼1
l00 ¼
3 X Ii i¼1
c-Fe2O3/Fe3O4
f (exp) (GHz)
f (fit) (GHz)
Ii
ai
Natural resonance (P1) Exchange resonance (P2) Exchange resonance (P3)
3.0/3.2 10.3/10.9 14.5/15.4
2.9/2.9 10.3/10.7 14.4/15.5
0.460/1.61 0.014/0.07 0.008/0.02
0.75/1.88 0.11/0.20 0.10/0.07
2
½1 ðf =f i Þ ð1 a2i Þ 2 2
2
½1 ðf =f i Þ ð1 þ ai Þ þ 4a
2 2 i ðf =f i Þ
ð1Þ
2
ðf =f i Þai ½1 þ ðf =f i Þ ð1 þ a2i Þ 2
2 2
½1 ðf =f i Þ ð1 þ ai Þ þ 4a2i ðf =f i Þ
2
ð2Þ
where f is the frequency, fi is the resonance frequency, ai is the damping coefficient, and Ii is the intensity of a peak. First, the triple resonance of the l00 curve was fitted with the variation of fi, ai and Ii. Then, the l0 curve was calculated using the obtained fitting parameters. Fig. 11a–d shows the fitted real part and imaginary part of the permeability spectra of c-Fe2O3 and Fe3O4. All the experimental and fitting parameters and results are listed in Table 1. In Fig. 11a–d, the fitting curves of real part (l0 ) and imaginary part (l00 ) of the permeability spectra show a good agreement with the experimental curves both for c-Fe2O3 and Fe3O4. Moreover, the fitted values of resonance frequencies f(fit) of P1, P2 and P3 correspond to the experimental result in Table 1. As above discussed, the first resonance peaks (P1) for c-Fe2O3 and Fe3O4 respectively can be ascribed to the natural resonance. The resonance peaks (P2 and P3) are ascribed to exchange resonance. In Table 1, Ii is
the intensity of resonance peak. Both for c-Fe2O3 and Fe3O4, the value of I1 is much bigger than that of I2 or I3, which indicate the natural resonance peak (P1) is much stronger than the two exchange resonance peaks. Thus, the magnetic loss may mainly result from the natural resonance.
4. Conclusions Self-assembled flowerlike a-Fe2O3, Fe3O4 and c-Fe2O3 were fabricated by a simple calcination procedure. The complex permittivity and permeability results indicate that the dielectric loss and magnetic loss of Fe3O4 flower are both higher than those of c-Fe2O3 flower. Compared with c-Fe2O3, the microwave absorption property of Fe3O4 is enhanced as evidenced by its higher saturation magnetization, crystalline anisotropy and better EM impedance match. An optimal RL of 46.0 dB was found at 3.4 GHz for Fe3O4 flower, which indicates that the sample can be used as a highly efficient microwave absorber. The further analysis shows that microwave absorption of Fe3O4 flower mostly originates from magnetic loss rather than dielectric loss. Moreover, the magnetic loss mainly results from the natural resonance.
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
Acknowledgment This work is supported by National Natural Science Foundation of China (Nos: 51101079 and 51171076). References [1] F. Qin, C. Brosseau, J. Appl. Phys. 111 (2012) 061301–061324. [2] R. Han, L. Qiao, T. Wang, F.-S. Li, J. Alloys Comp. 509 (2011) 2734–2737. [3] D. Yuping, W. Guangli, G. Shuchao, L. Shuqing, M. Guojia, Appl. Surf. Sci. 258 (2012) 5746–5752. [4] Y.Y. Zhou, A.C. Zhou, Y.C. Qing, F. Luo, D.M. Zhu, J. Magn. Magn. Mater. 74 (2015) 345–349. [5] D. Kim, J. Park, K. An, N.K. Yang, J.G. Park, T. Hyeon, J. Am. Chem. Soc. 129 (2007) 5812–5813. [6] C.L. Yin, Y.B. Cao, J.M. Fan, L.Y. Bai, F. Ding, F.L. Yuan, Appl. Surf. Sci. 270 (2013) 432–438. [7] Z.C. Wang, W. Yang, J.J. Wei, F.B. Meng, X.B. Liu, Mater. Lett. 123 (2014) 6–9. [8] X.Q. Shen, F.Z. Song, X.C. Yang, Z. Wang, M.X. Jing, Y.D. Wang, J. Alloys Comp. 621 (2015) 146–153. [9] C.J. Jia, L.D. Sun, Z.G. Yan, L.P. You, F. Luo, X.D. Han, Y.C. Pang, Z. Zhang, C.H. Yan, Angew. Chem. Int. Ed. 44 (2005) 4328–4333. [10] G.X. Tong, W.H. Wu, J.G. Guan, H.S. Qian, J.H. Yuan, W. Li, J. Alloys Comp. 509 (2011) 4320–4326. [11] H. Wu, L.D. Wang, H.J. Wu, Appl. Surf. Sci. 290 (2014) 388–397.
191
[12] Y.W. Zhu, T. Yu, C.H. Sow, Y.J. Liu, A.T.S. Wee, X.J. Xu, C.T. Lim, J.T.L. Thong, Appl. Phys. Lett. 87 (2005) 023103-1–023103-3. [13] X.C. Jiang, Y.L. Wang, T. Herricks, Y.N. Xia, J. Mater. Chem. 14 (2004) 695–703. [14] H. Deng, X.L. Li, Q. Peng, X. Wang, J.P. Chen, Y.D. Li, Angew. Chem. Int. Ed. 44 (2005) 2782–2785. [15] C.J. Jia, L.D. Sun, F. Luo, X.D. Han, L.J. Heyderman, Z.G. Yan, C.H. Yan, K. Zheng, J. Am. Chem. Soc. 130 (2008) 16968–16977. [16] G.B. Sun, B.X. Dong, M.H. Cao, B.Q. Wei, C.W. Hu, Chem. Mater. 23 (2011) 1587–1593. [17] Z.J. Fan, G.H. Luo, Z.F. Zhang, L. Zhou, F. Wei, Mater. Sci. Eng. B 132 (2006) 85– 89. [18] G.X. Tong, W.H. Wu, R. Qiao, J.H. Yuan, J.G. Guan, H.S. Qian, J. Mater. Res. 26 (2011) 1639–1645. [19] K. Jia, R. Zhao, J. Zhong, X. Liu, J. Magn. Magn. Mater. 332 (2010) 2167–2171. [20] C. Qiang, J. Xu, Z. Zhang, L. Tian, S. Xiao, Y. Liu, P. Xu, J. Alloys Comp. 506 (2010) 93–97. [21] J.H. Wang, H. Wang, J.J. Jiang, W.J. Gong, D. Li, Q. Zhang, X.G. Zhao, S. Ma, Z.D. Zhang, Cryst. Growth Des. 12 (2012) 3499–3504. [22] X.G. Liu, D.Y. Geng, H. Meng, P.J. Shang, Z.D. Zhang, Appl. Phys. Lett. 92 (2008) 173117-1–173117-3. [23] C. Kittel, Phys. Rev. 73 (1948) 155–161. [24] A. Aharoni, J. Appl. Phys. 69 (1991) 7762–7764. [25] W.H. Zhong, C.Q. Sun, S. Li, H.L. Bai, E.Y. Jiang, Acta Mater. 53 (2005) 3207– 3214. [26] Q.C. Liu, Z.F. Zi, D.J. Wu, Y.P. Sun, J.M. Dai, J. Mater. Sci. 47 (2012) 1033–1037. [27] N.N. Song, H.T. Yang, H.L. Liu, X. Ren, H.F. Ding, X.Q. Zhang, Z.H. Cheng, Sci. Rep. 3 (2013) 3161-1–3161-5.
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom
Flowerlike iron oxide nanostructures and their application in microwave absorption Chenyang Guo, Fangyuan Xia ⇑, Zhen Wang, Li Zhang, Li Xi, Yalu Zuo ⇑ 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
Article history: Received 22 August 2014 Received in revised form 14 January 2015 Accepted 15 January 2015 Available online 22 January 2015 Keywords: Self-assembled flowerlike iron oxides Microwave absorption Electromagnetic impedance match Natural resonance
a b s t r a c t Self-assembled flowerlike a-Fe2O3, Fe3O4 and c-Fe2O3 were fabricated by a simple calcination procedure. The structure characterization shows that the flowerlike morphology and the size of the nanostructures are perfectly maintained in the conversion of precursor to a-Fe2O3, Fe3O4, and c-Fe2O3. The complex permittivity and permeability results indicate that the dielectric and magnetic loss of Fe3O4 flower are both higher than those of c-Fe2O3 flower. In addition, Fe3O4 flower shows a good electromagnetic impedance match and its microwave absorption mainly originates from magnetic loss rather than dielectric loss. An optimal reflection loss of 46.0 dB is found at 3.4 GHz for flowerlike Fe3O4, which indicates that the sample can be used as a highly efficient microwave absorber. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Microwave absorbing materials based on the magnetic materials with low reflection and high absorption properties are receiving extensive attention because of their promising application in electronic devices used in commerce, industry, and military affairs [1]. In recent years, carbonyl irons were widely investigated as one of typical magnetic microwave absorbing materials [2–4]. However, the low density and absorption properties of the commercial carbonyl iron restrict its use in the applications. Hollow iron particle with low density is also a new material to improve the microwave absorption properties [5,6]. Besides, iron-based materials such as c-Fe2O3, and Fe3O4 are recognized as promising materials for microwave applications because of their relative high magnetization and antioxidation properties in air. Various iron oxides or iron composites with different morphologies such as rods, wires, tubes, flakes, urchins and dendrites have been successfully reported with their application as electromagnetic wave absorber [7–12]. However, the relationship among the structure, morphology and magnetic properties of iron-based nanomaterials has not been fully understood. As we known, large surface areas and high magnetization will improve the microwave absorption properties. Herein, we report the synthesis of novel three-dimensional (3D) flowerlike iron oxide nanostructures with large surface areas by an ethylene glycol (EG)-mediated self-assembly process. The assynthesized iron oxide precursor was transformed into iron oxide. ⇑ Corresponding author. Tel.: +86 0931 8914036; fax: +86 0931 8914160. E-mail address: [email protected] (Y. Zuo). http://dx.doi.org/10.1016/j.jallcom.2015.01.087 0925-8388/Ó 2015 Elsevier B.V. All rights reserved.
Moreover, the phase of the final product can be easily controlled to be a-Fe2O3, c-Fe2O3, or Fe3O4, three of the most common iron oxides, by altering the calcination conditions. The microwave absorption properties of a-Fe2O3, c-Fe2O3, and Fe3O4 flowerlike nanostructures with almost the same large surface areas are presented in this study. 2. Experimental The thermal decomposition of metal alkoxide is a simple route to achieving a tailored metal oxide. In a typical procedure, 2.4 g of ferric chloride (FeCl36H2O), 5.4 g of urea, and 14.4 g of tetrabutylammonium bromide were dissolved in 360 mL of ethylene glycol. The red solution was refluxed at 195 °C for 30 min. After cooling, the as-synthesized iron oxide precursor was collected as a green precipitate after centrifugation and ethanol-washing cycles. The morphology of the precursor was studied by scanning electron microscopy (SEM, Hitachi S-4800). Fig. 1a shows the SEM image of a typical sample composed of many uniform, flowerlike nanostructures approximately 3 lm in diameter. The detailed morphology of the flowerlike nanostructures is shown in Fig. 1b. The entire structure was found to be built from several dozen nanopetals with smooth surfaces. These nanopetals were 1.5 lm wide and connected to each other through the center to form 3D flowerlike structures. A series of other measurements was also performed to investigate the as-obtained iron oxide precursor. The X-ray diffraction (XRD, Philips X’Pert, with Cu Ka radiation) pattern (Fig. 1c) shows the emergence of diffraction peaks similar to those of other metal oxide precursors reported in the literature [13], especially the strong peak located in the low-angle region (2h = 11°), although the exact crystal structure of the iron oxide precursor is not yet to be determined. The effect of calcinations on the crystallization and morphology of the iron oxide precursor was investigated by thermogravimetric analysis–differential thermal analysis (TG–DTA, Perkin Elmer SII) with Ar protection. Fig. 1d shows the TG–DTA curves of the as-synthesized iron oxide precursor. A weight loss was observed at about 250–400 °C in the TG curve. The weight loss can be attributed to the combustion of the resultant organics. One sharp exothermic peak was found at 320 °C, which was accompanied by the aforementioned weight loss. No any other
184
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
Fig. 1. (a) SEM image of the as-synthesized iron oxide precursor. (b) SEM image of a single flowerlike iron oxide precursor. (c) XRD and (d) TG–DTA curves of the assynthesized iron oxide precursor.
Fig. 2. XRD patterns of the samples: (a) c-Fe2O3, (b) Fe3O4, and (c) a-Fe2O3 flowers.
weight loss in TG or any peak in DTA was observed over 450 °C, confirming that all organic components were burned out at 450 °C. Hence, the as-synthesized iron oxide precursor was transformed into a-Fe2O3 (or Fe3O4) at 450 °C in air (or under N2 protection) for 3 h. Then, phase transformation was achieved from Fe3O4 to c-Fe2O3 by an oxidation process. The microstructure of the sample was characterized by high-resolution transmission electron microscopy (HRTEM; JEOL, JEM-2010; at 200 kV) and selected area electron diffraction (SAED) analyses. X-ray photoelectron spectroscopy (XPS; VG
Fig. 3. Fe 2p core-level XPS spectra of a-Fe2O3, Fe3O4, and c-Fe2O3 flowers.
ESCALAB MK II) was used to test the chemical valence of Fe ions. The magnetic properties of the samples were measured with a vibrating sample magnetometer (VSM; LakeShore 7304 model). A toroidal sample (inner diameter, 3.04 mm; outer diameter, 7 mm; and thickness, 3 mm) was prepared to fit well the coaxial sample holder for microwave measurements. The complex relative permeability (lr = l0 jl00 ) and permittivity (er = e0 je00 ) of the composite samples were measured by the coaxial method on an Agilent E8363B vector network analyzer within the range of 0.1–18 GHz.
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
185
Fig. 4. (a) Low-magnification and (b) high-magnification SEM images of a-Fe2O3 flower. (c) Low-magnification and (d) high-magnification TEM images of a-Fe2O3 flower. (e) HRTEM image and (f) SAED patterns of a-Fe2O3 flower.
3. Results and discussion The diffraction peaks of the as-obtained red product well agreed with those for a-Fe2O3 powder (JCPDS card 80-2377) (Fig. 2c), when the precursor was calcined at 450 °C in air for 3 h. When the precursor was calcined at 450 °C under N2 protection for 3 h, the positions of all diffraction peaks of the corresponding product (Fig. 2b), well matched those of Fe3O4 (JCPDS card 87-0245). The formation of Fe3O4 can be attributed to the reductive ability of the organic species (glycolates) in the precursor. Furthermore, when as-obtained Fe3O4 was heated at 250 °C in air for 3 h, the color changed from black to red-brown. The XRD pattern of this product (Fig. 2a) was well agreed with the standard XRD pattern
of c-Fe2O3 (JCPDS card 39-1346). The three peaks between 50 and 70 °C from the red-brown sample slightly shifted to higher angles [14]. These results showed that Fe3O4 was transformed into c-Fe2O3. Herein, a simple calcination procedure was used to obtain all three common iron oxides from the same iron oxide precursor. The patterns observed in Fig. 2b and c are almost the same, matching both the patterns of Fe3O4 and c-Fe2O3. Therefore, XPS measurements (Fig. 3) were performed to unambiguously assign the crystal phase because XPS is very sensitive to Fe2+ and Fe3+ cations. The XPS spectra have been charge corrected to the adventitious C1s with binding energy of 284.6 eV. It can be seen that the peaks generally shift to a high binding energy and broaden for Fe3O4 because of the appearance of Fe2+(2p3/2) and Fe2+(2p1/2). In
186
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
Fig. 5. (a) Low-magnification and (b) high-magnification SEM images of Fe3O4 flower. (c) Low-magnification and (d) high-magnification TEM images of Fe3O4 flower. (e) HRTEM image and (f) SAED patterns of Fe3O4 flower.
this study, the levels of Fe 2p3/2 are 711.1 and 710.7 eV, Fe 2p1/2 are 724.7 and 724.3 eV for c-Fe2O3 and a-Fe2O3, respectively, as well as Fe 2p3/2 and Fe 2p1/2 are 711.2 and 724.8 eV for Fe3O4. A satellite peak at 719.6 and 719.2 eV is observed in c-Fe2O3 and a-Fe2O3 [15]. In addition, no satellite for Fe3O4 is identified, excluding the presence of c-Fe2O3 or a-Fe2O3 in the Fe3O4 samples, which well agrees with the above mentioned literature. The XPS patterns well agree with the XRD data and reveal that complete phase transformation can be achieved by a simple calcination procedure. The SEM images of the as-synthesized samples are shown in panel (a) of Figs. 4–6. Compared with the conversion of the precursor, the flowerlike morphology and size of the samples are perfectly maintained in the conversion of a-Fe2O3 to Fe3O4 and
c-Fe2O3. Panel (b) of Figs. 4–6 shows the high-magnification images of a single flower, and all flowers exhibit a similar structure, which is the same as the precursor nanoflowers. Further structural characterization of a-Fe2O3, Fe3O4, and c-Fe2O3 flowers was performed using HRTEM, as shown in panels c and d of Figs. 4–6. The results show the same morphology and size as those in the SEM images. A representative HRTEM image taken from a-Fe2O3 flower is shown in Fig. 4e. The lattice fringes are clearly visible with a spacing of 0.252 nm, corresponding to the spacing of the (1 1 0) planes of a-Fe2O3. The lattice fringes of the (2 2 0) planes of Fe3O4 with a d-spacing of 0.297 nm and the (3 1 1) planes of c-Fe2O3 with a d-spacing of 0.253 nm are also clearly observed in the insets of Figs. 5e and 6e, respectively. Furthermore, the SAED patterns were
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
187
Fig. 6. (a) Low-magnification and (b) high-magnification SEM images of c-Fe2O3 flower. (c) Low-magnification and (d) high-magnification TEM images of c-Fe2O3 flower. (e) HRTEM image and (f) SAED patterns of c-Fe2O3 flower.
measured at the edge of the nanopetals of all the flowerlike samples. Fig. 4f shows that edge of the nanopetals of a-Fe2O3 flower is single crystalline. The crystallographic orientation (1 1 0) is also observed. However, for a whole a-Fe2O3 flower, it should be polycrystalline. In addition, SAED patterns of Fe3O4 and c-Fe2O3 (Figs. 5f and 6f) exhibit diffraction rings, indicating the polycrystalline nature of the nanoflowers, and the positions of all diffraction rings of the corresponding calcined products well match those in the XRD patterns. Comparing to a-Fe2O3, the polycrystalline nature may be due to the N2 protection and further oxidation process in the calcination procedure for Fe3O4 and c-Fe2O3, respectively. Iron-based materials often have intriguing magnetic properties due to the variation of their crystal structure, shape anisotropy, and crystallinity. Especially, the value of Ms for ferromagnetic
nanostructures is associated with grain size, crystallinity, and noncollinear magnetic structure. Therefore, the magnetic properties of a-Fe2O3, Fe3O4, and c-Fe2O3 flowers were examined by VSM. Fig. 7 shows the magnetic hysteresis loops (M–H loops) of the flowerlike samples measured at room temperature. The antiferromagnetic a-Fe2O3 flowers exhibit quite small saturation magnetization (Ms) around 1 emu/g at 10 kOe, which may be caused by the uncompensated surface spin. The Ms of Fe3O4 and c-Fe2O3 nanoflowers are about 80 and 55 emu/g at 1500 Oe, which are respectively lower than 92 and 76 emu/g for bulk Fe3O4 and c-Fe2O3 materials because the spin disorder on the surface would reduce the total magnetic moment. This is commonly observed for the nanostructure materials, for example, the Ms of Fe3O4 and c-Fe2O3 dendritic microstructures are around 75 and 42 emu/g, respectively [16].
188
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
to the transmission line theory [17], the normalized input impedance Zin of a microwave absorber is given by
rffiffiffiffiffi Z in ¼
lr 2p pffiffiffiffiffiffiffiffiffi lr er fd tan h j er c
where lr and er are the relative permeability and permittivity of the composite medium, c is the velocity of electromagnetic (EM) waves in free space, f is the frequency of the microwave, and d is the thickness of the absorber. The reflection loss is related to Zin as RL pffiffiffiffiffiffiffiffiffiffiffiffiffi 0 where Z 0 ¼ l0 =e0 is the characteristic ðdBÞ ¼ 20 log ZZinin Z þZ 0 ,
Fig. 7. Magnetic hysteresis loops of (a) a-Fe2O3, (b) Fe3O4, and (c) c-Fe2O3 flowers measured at 300 K.
Fig. 8. Microwave reflection loss of epoxy resin with 80 wt.% (a) c-Fe2O3 and (b) Fe3O4 flowers at different thicknesses.
The magnetic loss and microwave absorption property of
a-Fe2O3 flower are weak because of its small saturation magnetization. Thus, the microwave absorption property was investigated only for c-Fe2O3 and Fe3O4 flowers. To reveal the microwave absorption properties, the reflection loss (RL) values of epoxy resin with 80 wt.% c-Fe2O3 and Fe3O4 flowers were calculated. According
impedance of free space. The thickness of the sample is one of the crucial parameters that affects the intensity and position of the frequency at the RL minimum. Therefore, we prepared c-Fe2O3 and Fe3O4 samples at different thicknesses in order to eliminate the influence resulting from the thickness of the samples. The results are shown in Fig. 8. It was found that the two samples exhibit excellent microwave absorption properties. For Fe3O4 flower, there is a sharp and strong peak at 3.4 GHz with the minimum RL of 46.0 dB when the thickness of the Fe3O4 sample is 5.0 mm, and a broad peak (15.0 dB, 13.0 GHz) is observed with the thickness of 2.0 mm (Fig. 8b). It is slightly higher than the result of Fe3O4 urchins, which have the optimal reflection loss of 43.2 dB at high frequency of 16.8 GHz [18]. Compared to Fe3O4 nanospheres and sponges, the author pointed out that the excellent microwave absorption performance was ascribed to their unique morphologies. Such morphologies resulted in reinforced EM parameters and multiresonant behavior. However, the result of the Fe3O4 flower is obviously better than other reports: Fe3O4 nanospheres with sizes of about 300 nm (30.3 dB, 5.5 mm), and Fe3O4 nanoparticles with the sizes of about 50–100 nm (35.0 dB, 4.4 mm) [19,20]. In addition, for c-Fe2O3 flower, the minimum RL reaches 15 dB with the thickness of 3 mm (Fig. 8a). Thus, the microwave absorption property of Fe3O4 flower is better than c-Fe2O3 flower. To investigate the possible mechanism of microwave absorption of the above two samples, we independently investigated the complex relative permittivity (e0 and e00 ) and permeability (l0 and l00 ) of the samples. Generally, complex permittivity and permeability represent the dielectric and dynamic magnetic properties of materials. The real parts of complex permittivity and permeability (e0 and l0 ) symbolize the storage and transformation capability of electromagnetic (EM) energy. The imaginary parts (e00 and l00 ) represent the loss of EM energy. The dielectric and magnetic loss tangent (tan dE = e00 /e0 and tan dM = l00 /l0 ) represent the tangent of the angle between the induced electric and magnetic field in the microwave absorber and the electric and magnetic field of the EM wave. Excellent microwave-absorbing materials are well known to have a good match between the magnetic loss and dielectric loss. The frequency dependence of e0 and e00 of c-Fe2O3 and Fe3O4 flowers is shown in Fig. 9a and b, respectively. It can be seen that both e0 and e00 have fluctuating behavior, which can be attributed to the displacement current lag at the interface between epoxy resin and flowerlike samples. This behavior was also observed in Fe3O4/ Bi nanocomposites and ZnO-coated Fe nanocapsules [21,22]. The e0 of Fe3O4 flower is higher than that of c-Fe2O3 flower. However, the e00 values of Fe3O4 and c-Fe2O3 flowers are similar, indicating that the dielectric loss of Fe3O4 and c-Fe2O3 flowers is similar. Otherwise, based on the free electron theory, e00 = 1/2pe0qf (where q is the resistivity), e00 has a lower value (0–2.2) within the entire range. Thus, low values of e00 indicate high electric resistivity. Fig. 9c shows relative complex permeability of c-Fe2O3 and Fe3O4 flowers. The real part (l0 ) of Fe3O4 flower exhibits a sharp decrease from 3.4 to 0.7 within the 0.1–18.0 GHz range. However, l0 of c-Fe2O3 flower gradually decreases from 1.5 to 0.9. Regarding
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
189
Fig. 9. Complex permittivity and permeability of epoxy resin with 80 wt.% c-Fe2O3 and Fe3O4 flowers as a function of frequency.
Fig. 10. Tan dE and tan dM of c-Fe2O3 and Fe3O4 flowers, respectively.
to the l00 –f spectra of both c-Fe2O3 and Fe3O4 flowers in Fig. 9d, three peaks are observed at about 3.0, 11.0, and 15.0 GHz, respectively. The first resonance peak at around 3.0 GHz can be attributed to natural resonance. The resonance frequency strongly depends on the saturation magnetization, crystalline anisotropy, and magnetic particle geometry [23]. The resonance peaks at around 11.0 and 15.0 GHz may be ascribed to exchange resonance. To our knowledge, multiresonance can be observed when the size of the magnetic particles is reduced. Among the modes that deal with multiresonance, the most accepted one is the exchange resonance mode placed forward by Aharoni [24]. Surface anisotropy and exchange energy caused by the exchange effect between grains have been proven to be important for nanocrystalline particles [25]. Generally speaking, the multiresonance phenomena were contributed from the natural resonance and exchange resonance [26,27]. In Fig. 9b and d, the values of e00 and l00 for Fe3O4 flower are both higher than those of c-Fe2O3 flower. This finding indicates that the dielectric and magnetic loss of Fe3O4 flower are both higher than those of c-Fe2O3 flower. For c-Fe2O3 flower in Fig. 10a, tan dE increases and exhibits a peak at 3.0 GHz, whereas tan dM shows a peak at about 13.0 GHz. The magnetic loss and dielectric loss of c-Fe2O3 flower are reportedly a mismatch. Meanwhile, for Fe3O4 flower in Fig. 10b, the trend of the tan dM curve is similar to that of tan dE, which has a few peaks within the 0.1–18.0 GHz range. This result means that the sample has a good match between the magnetic loss and dielectric loss. Thus, Fe3O4 flower has great potential for being a highly efficient microwave absorber. Furthermore, tan dM is larger than tan dE, indicating that the microwave absorption of Fe3O4 flower mainly originates from magnetic loss rather than dielectric loss. To further understand the magnetic loss, permeability spectra of c-Fe2O3 and Fe3O4 were fitted as the linear overlap of three resonance peaks P1, P2 and P3. Then, the P1, P2, P3 could be separated from each other by fitting of the Gilbert modification of Landau–Lifshitz equation [22] as
190
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
Fig. 11. The fitting curves of real part (a) and imaginary part (b) of permeability for c-Fe2O3, and real part (c) and imaginary part (d) of permeability for Fe3O4.
Table 1 Fitting parameters for permeability spectra of c-Fe2O3 (black) and Fe3O4 (blue). The f(exp) and f(fit) are the experimental and fitted resonance frequency, respectively.
l0 ¼ B þ
3 X Ii i¼1
l00 ¼
3 X Ii i¼1
c-Fe2O3/Fe3O4
f (exp) (GHz)
f (fit) (GHz)
Ii
ai
Natural resonance (P1) Exchange resonance (P2) Exchange resonance (P3)
3.0/3.2 10.3/10.9 14.5/15.4
2.9/2.9 10.3/10.7 14.4/15.5
0.460/1.61 0.014/0.07 0.008/0.02
0.75/1.88 0.11/0.20 0.10/0.07
2
½1 ðf =f i Þ ð1 a2i Þ 2 2
2
½1 ðf =f i Þ ð1 þ ai Þ þ 4a
2 2 i ðf =f i Þ
ð1Þ
2
ðf =f i Þai ½1 þ ðf =f i Þ ð1 þ a2i Þ 2
2 2
½1 ðf =f i Þ ð1 þ ai Þ þ 4a2i ðf =f i Þ
2
ð2Þ
where f is the frequency, fi is the resonance frequency, ai is the damping coefficient, and Ii is the intensity of a peak. First, the triple resonance of the l00 curve was fitted with the variation of fi, ai and Ii. Then, the l0 curve was calculated using the obtained fitting parameters. Fig. 11a–d shows the fitted real part and imaginary part of the permeability spectra of c-Fe2O3 and Fe3O4. All the experimental and fitting parameters and results are listed in Table 1. In Fig. 11a–d, the fitting curves of real part (l0 ) and imaginary part (l00 ) of the permeability spectra show a good agreement with the experimental curves both for c-Fe2O3 and Fe3O4. Moreover, the fitted values of resonance frequencies f(fit) of P1, P2 and P3 correspond to the experimental result in Table 1. As above discussed, the first resonance peaks (P1) for c-Fe2O3 and Fe3O4 respectively can be ascribed to the natural resonance. The resonance peaks (P2 and P3) are ascribed to exchange resonance. In Table 1, Ii is
the intensity of resonance peak. Both for c-Fe2O3 and Fe3O4, the value of I1 is much bigger than that of I2 or I3, which indicate the natural resonance peak (P1) is much stronger than the two exchange resonance peaks. Thus, the magnetic loss may mainly result from the natural resonance.
4. Conclusions Self-assembled flowerlike a-Fe2O3, Fe3O4 and c-Fe2O3 were fabricated by a simple calcination procedure. The complex permittivity and permeability results indicate that the dielectric loss and magnetic loss of Fe3O4 flower are both higher than those of c-Fe2O3 flower. Compared with c-Fe2O3, the microwave absorption property of Fe3O4 is enhanced as evidenced by its higher saturation magnetization, crystalline anisotropy and better EM impedance match. An optimal RL of 46.0 dB was found at 3.4 GHz for Fe3O4 flower, which indicates that the sample can be used as a highly efficient microwave absorber. The further analysis shows that microwave absorption of Fe3O4 flower mostly originates from magnetic loss rather than dielectric loss. Moreover, the magnetic loss mainly results from the natural resonance.
C. Guo et al. / Journal of Alloys and Compounds 631 (2015) 183–191
Acknowledgment This work is supported by National Natural Science Foundation of China (Nos: 51101079 and 51171076). References [1] F. Qin, C. Brosseau, J. Appl. Phys. 111 (2012) 061301–061324. [2] R. Han, L. Qiao, T. Wang, F.-S. Li, J. Alloys Comp. 509 (2011) 2734–2737. [3] D. Yuping, W. Guangli, G. Shuchao, L. Shuqing, M. Guojia, Appl. Surf. Sci. 258 (2012) 5746–5752. [4] Y.Y. Zhou, A.C. Zhou, Y.C. Qing, F. Luo, D.M. Zhu, J. Magn. Magn. Mater. 74 (2015) 345–349. [5] D. Kim, J. Park, K. An, N.K. Yang, J.G. Park, T. Hyeon, J. Am. Chem. Soc. 129 (2007) 5812–5813. [6] C.L. Yin, Y.B. Cao, J.M. Fan, L.Y. Bai, F. Ding, F.L. Yuan, Appl. Surf. Sci. 270 (2013) 432–438. [7] Z.C. Wang, W. Yang, J.J. Wei, F.B. Meng, X.B. Liu, Mater. Lett. 123 (2014) 6–9. [8] X.Q. Shen, F.Z. Song, X.C. Yang, Z. Wang, M.X. Jing, Y.D. Wang, J. Alloys Comp. 621 (2015) 146–153. [9] C.J. Jia, L.D. Sun, Z.G. Yan, L.P. You, F. Luo, X.D. Han, Y.C. Pang, Z. Zhang, C.H. Yan, Angew. Chem. Int. Ed. 44 (2005) 4328–4333. [10] G.X. Tong, W.H. Wu, J.G. Guan, H.S. Qian, J.H. Yuan, W. Li, J. Alloys Comp. 509 (2011) 4320–4326. [11] H. Wu, L.D. Wang, H.J. Wu, Appl. Surf. Sci. 290 (2014) 388–397.
191
[12] Y.W. Zhu, T. Yu, C.H. Sow, Y.J. Liu, A.T.S. Wee, X.J. Xu, C.T. Lim, J.T.L. Thong, Appl. Phys. Lett. 87 (2005) 023103-1–023103-3. [13] X.C. Jiang, Y.L. Wang, T. Herricks, Y.N. Xia, J. Mater. Chem. 14 (2004) 695–703. [14] H. Deng, X.L. Li, Q. Peng, X. Wang, J.P. Chen, Y.D. Li, Angew. Chem. Int. Ed. 44 (2005) 2782–2785. [15] C.J. Jia, L.D. Sun, F. Luo, X.D. Han, L.J. Heyderman, Z.G. Yan, C.H. Yan, K. Zheng, J. Am. Chem. Soc. 130 (2008) 16968–16977. [16] G.B. Sun, B.X. Dong, M.H. Cao, B.Q. Wei, C.W. Hu, Chem. Mater. 23 (2011) 1587–1593. [17] Z.J. Fan, G.H. Luo, Z.F. Zhang, L. Zhou, F. Wei, Mater. Sci. Eng. B 132 (2006) 85– 89. [18] G.X. Tong, W.H. Wu, R. Qiao, J.H. Yuan, J.G. Guan, H.S. Qian, J. Mater. Res. 26 (2011) 1639–1645. [19] K. Jia, R. Zhao, J. Zhong, X. Liu, J. Magn. Magn. Mater. 332 (2010) 2167–2171. [20] C. Qiang, J. Xu, Z. Zhang, L. Tian, S. Xiao, Y. Liu, P. Xu, J. Alloys Comp. 506 (2010) 93–97. [21] J.H. Wang, H. Wang, J.J. Jiang, W.J. Gong, D. Li, Q. Zhang, X.G. Zhao, S. Ma, Z.D. Zhang, Cryst. Growth Des. 12 (2012) 3499–3504. [22] X.G. Liu, D.Y. Geng, H. Meng, P.J. Shang, Z.D. Zhang, Appl. Phys. Lett. 92 (2008) 173117-1–173117-3. [23] C. Kittel, Phys. Rev. 73 (1948) 155–161. [24] A. Aharoni, J. Appl. Phys. 69 (1991) 7762–7764. [25] W.H. Zhong, C.Q. Sun, S. Li, H.L. Bai, E.Y. Jiang, Acta Mater. 53 (2005) 3207– 3214. [26] Q.C. Liu, Z.F. Zi, D.J. Wu, Y.P. Sun, J.M. Dai, J. Mater. Sci. 47 (2012) 1033–1037. [27] N.N. Song, H.T. Yang, H.L. Liu, X. Ren, H.F. Ding, X.Q. Zhang, Z.H. Cheng, Sci. Rep. 3 (2013) 3161-1–3161-5.