- Email: [email protected]
Sol–gel auto-combustion synthesis of SiO2-doped NiZn ferrite by using various fuels
ARTICLE IN PRESS
Journal of Magnetism and Magnetic Materials 298 (2006) 25–32 www.elsevier.com/locate/jmmm
Sol–gel auto-combustion synthesis of SiO2-doped NiZn ferrite by using various fuels K.H. Wua,, T.H. Tinga, M.C. Lia, W.D. Hob a
Department of Applied Chemistry, Chung Cheng Institute of Technology, NDU, No. 190, Sanyuan 1st Street, Dashi Jen, Tahsi, Taoyuan 335, Taiwan b Chemical Systems Research Division, Chung Shan Institute of Science and Technology, Taoyuan, Taiwan Received 22 December 2004; received in revised form 14 February 2005 Available online 24 March 2005
Abstract A nitrate–chelate–silica gel was prepared from metallic nitrates, citric acid and tetraethoxysilane (TEOS) by sol–gel process with different complexing agents such as glycine, hydrazine and citric acid, and it was further used to synthesize Ni0.5Zn0.5Fe2O4/20 wt% SiO2 nanocomposites by auto-combustion. The effect of varying complexing agent on the structural and magnetic properties of the composites was studied by FTIR, 29Si CP/MAS NMR, XRD, TEM, EPR and SQUID measurements. The complexing agent in the starting solution influenced the magnetic interaction between NiZn ferrite and silica, and then determined on the particle size. Further, the complexing agent type had a direct effect on the EPR parameters (DHPP, g-factor and T2) and SQUID parameters (Ms, Mr and Hc) of the as-synthesized powder. r 2005 Elsevier B.V. All rights reserved. Keywords: Complexing agent; Fuel; Ferrite; Silica; EPR
1. Introduction NiZn ferrites are one of the most versatile magnetic materials for general use, which have many applications in both low and high frequency devices and play a useful role in many technological applications such as microwave devices, power transformers in electronics, rod Corresponding author. Tel.: +886 33891716324; fax: +886 33808906. E-mail address: [email protected] (K.-H. Wu).
antennas, read/write heads for high speed digital tape, etc. because of their high resistivity, low dielectric losses, mechanical hardness, high Curie temperature and chemical stability [1–4]. The amorphous matrixes have been shown to play an important role in retarding the motion of the particles as well as the grain growth during the formation of nanocrystals. In addition, the electromagnetic properties of composite are affected not only by the compositions, additives and annealing conditions but also by the raw materials [5,6].
0304-8853/$ - see front matter r 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jmmm.2005.03.008
ARTICLE IN PRESS 26
K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32
Many synthetic approaches have been employed to prepare magnetic nanocrystals [7–10]. The sol–gel auto-combustion technique is a novel way with a unique combination of the chemical sol–gel process and the combustion process. The synthesis has been used to create different ceramic systems [11,12]. The success of the process is due to an intimate blending among the constituents using a suitable fuel or complexing agent (e.g. citric acid, urea, glycine, etc.) in an aqueous media and an exothermic redox reaction between the fuel and an oxidizer (i.e. nitrates) [13]. The powder characteristics like crystallite size, surface area, extent, and nature of agglomeration are primarily governed by enthalpy or flame temperature generated during combustion, which itself is dependent on nature of the fuel and fuel-to-oxidant ratio [14]. Chelating ligands, which contain carboxylate groups or aliphatic amine groups, are essential in the water-soluble complex precursor synthesis route. Citric acid (containing carboxylate groups), glycine (containing carboxylate and aliphatic amine groups) and hydrazine (containing aliphatic amine groups) were often used before in the synthesis of metallic oxides. Such types of complexing agent can effectively complex metal ions of varying ionic sizes, which helps in preventing their selective precipitation to maintain compositional homogeneity among the constituents. On the other hand, these can also serve as a fuel in the combustion reaction, being oxidized by nitrate ions. In our previous papers [15–17], SiO2-doped NiZn ferrite nanocomposites were prepared using sol–gel auto-combustion method. Many initial synthesis conditions such as silica content, calcinations temperature, solution pH and fuel-to-oxidant ratio have been varied in order to determine the optimal conditions for synthesizing the material. The present paper describes the synthesis of SiO2-doped NiZn ferrite nanocomposites by hydrolysis of tetraethoxysilane (TEOS) onto NiZn ferrite particle and the effect of varying the organic fuel on the chemical and electromagnetic properties. The spectroscopic characterization, crystallite sizes, EPR parameters (DHPP, g-factor and T2) and SQUID parameters (Ms, Mr and Hc) of composites are studied using FTIR, 29Si CP/MAS NMR, XRD, EPR and SQUID.
2. Experimental 2.1. Preparation of NiZn ferrite/SiO2 composite Analytical grade nickel nitrate, zinc nitrate, iron nitrate, fuel and TEOS were used as raw materials to prepare Ni0.5Zn0.5Fe2O4/20 wt% SiO2 nanocomposite. The initial molar ratio was Ni:Zn:Fe ¼ 1:1:4. First, 2.0 g Ni(NO3)2 6H2O, 2.05 g Zn(NO3)2 6H2O, and 11.12 g Fe(NO3)3 9H2O were dissolved in 20 ml of ethanol, then 20 wt% of TEOS and H2O in a molar ratio of 1:4 and the fuel, such as citric acid, glycine and hydrazine were added into the solution. The molar ratio of nitrates to fuel was 1:1. A small amount of ammonia was added to the solution to adjust the pH value to about 5. The entire mixture was thoroughly stirred for 6 h at 70 1C. Then, the mixed solution was poured into a teflon dish and heated 24 h at 60 1C and 3 h at 100 1C under a vacuum to obtain a dried gel. When ignited at any point, the dried gel burnt in a self-propagating combustion manner until all the gels were burnt out completely to form a loose powder. 2.2. Characterization of NiZn ferrite/SiO2 composite The phase identification of the as-burnt powder was performed using X-ray diffraction (XRD; SIEMENS D5000) with Cu Ka radiation. Average grain sizes (D) were determined from the XRD peaks using Scherrer’s formula as well as by a PHILIPS CM-200 transmission electron microscopy (TEM). Infrared spectra (IR) of the as-burnt powder were recorded on a Bomem DA 3.002 spectrophotometer from 400 to 4000 cm–1 by the KBr pellet method. The solid-state 29Si-NMR spectra of the gel precursor were determined using a Bruker MSL-400 with the cross-polarization combined with magic angle spinning (CP/MAS). The 29Si CP/MAS NMR provides a unique way to follow the structure of silica network and the magnetic interaction between iron (III) and silica. The electron paramagnetic resonance (EPR) spectra of the composites were recorded on a Bruker EMX-10 spectrometer operating at X-band ðn ¼ 9:6 GHzÞ with 100 kHz field modulations. DPPH
ARTICLE IN PRESS K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32
ðg ¼ 2:0036Þ was used as a field marker. The EPR g-value of the unknown sample can be calculated from g ¼ gs ðDH=H o Þgs ; where gs is the g-value of the reference sample, DH is the separation of the centers of the two spectra, Ho is the strength of the applied external field. The EPR DHPP was calculated from peak-to-peak linewidth. The EPR spectra were recorded at variable temperatures (200–400 K) using variable temperature controller. Magnetization measurements were performed in fields of up to 5 T using a Quantum Design SQUID magnetometer (Model MPMS5).
3. Results and discussion 3.1. Phase analysis and particle size of powder Fig. 1 shows the X-ray diffraction patterns of microcrystalline NiZn ferrite/SiO2 powders. All the as-burnt powders are a single phase NiZn ferrite with a spinel structure. This result indicates that the NiZn ferrites are directly formed after auto-combustion of gels. The direct transformation of crystalline ferrite from amorphous gel during combustion is surely due to the greater heat generated from the exothermic reaction of nitrates and fuel [18]. Crystallite sizes of the composite are calculated from the X-ray peak
Intensity (a.u.)
(311)
(a)
(b)
(c)
20
30
40 50 2 degrees
60
70
Fig. 1. XRD spectra of NiZn ferrite/SiO2 synthesized with (a) glycine, (b) citric acid and (c) hydrazine.
27
broadening of the (3 1 1) diffraction peak using Scherrer’s formula [12] D ¼ 0:9l=b cos y,
(1)
where D is the crystallite size in nanometres, l the radiation wavelength (0.154056 nm for Cu Ka), b the bandwidths at half-height and y the diffraction peak angle. The calculated crystallite sizes are 34, 22 and 18 nm, respectively, for crystalline synthesized with glycine, hydrazine and citric acid. It was evident that the crystallite size of the ferrite phase depending on the fuel, which was fairly consistent with the particle size determined by transmission electron microscopy (Fig. 2). TEM micrographs show that the composite consisted of a mixture of crystalline NiZn ferrite and amorphous silica at the nanometer level and homogeneous distribution. Due to the rapid explosive pyrolysis and the high fame temperature [19], the glycine method have larger particle size and good crystallinity, thus the signal to noise of the XRD is significantly higher than other. Furthermore, citric acid is a more effective complexing agent than hydrazine and glycine in producing fine ferrite powder with sol–gel auto-combustion method. 3.2. Infrared and
29
si cp/mas nmr spectra
Fig. 3 shows the IR spectra of the NiZn ferrite/ SiO2 with different complexing agent. The silica network is characterized by the absorptions at 1090, 805 and 462 cm–1, corresponding to the Si–O–Si anti-symmetric stretch, symmetric stretch and bending mode. The absorptions in the range 3700–3200 and 1628 cm–1 assigned to the stretching vibrations of the silanol groups (Si–OH). The bands at about 1384 and 570 cm–1 are attributed to stretching vibration of the anti-symmetric NO 3 and tetrahedral complexes of ferrite. The ferritestretching band is shifted to higher wave number as comparison with in NiZn ferrite (560 cm–1) [20], which can be ascribed to the formed interaction between the NiZn ferrite and silica through Si–O–Fe bonds [7]. On the other hand, the disappearance or decrease at 1384 cm–1 of the asburnt powder revealed that the NO 3 ions take part in the reaction during combustion. Therefore, the
ARTICLE IN PRESS K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32
570
(a)
4000
3500
3000 2500 2000 1500 Wavenumber (cm-1)
1000
570 462
1090
805
1384
(c)
574
(b) 1628
Transmittance (arb. unit)
28
500
Fig. 3. IR spectra of NiZn ferrite/SiO2 synthesized with (a) citric acid, (b) glycine and (c) hydrazine.
Fig. 2. TEM photograph of NiZn ferrite/SiO2 synthesized with (a) glycine, (b) hydrazine and (c) citric acid.
combustion can be considered as a thermally induced anionic redox reaction of the gel wherein the fuel acts as a reductant and the NO–3 ion acts as an oxidant [21]. 29 Si-NMR measurements confirm FTIR data showing the formation of magnetic interaction between the NiZn ferrite and silica. However, we got additional information on the morphologies, especially on the distribution of the transition metal in the amorphous system. The 29Si CP/MAS NMR spectra are collected for the dried gel and as burnt powder as shown in Fig. 4. The signals of Q3 (100 ppm) and Q4 (110 ppm) structural units in the dried gels (Figs. 4a–c) are partially overlapped and the intensities are weakened in the order: citric acid sample4hydrazine sample4 glycine sample, indicating that the glycine has the strongest complex ability. The results may be due to the Ni2+, Zn2+ and Fe3+ ions dispersion homogeneously in silica gel and in close proximity to the paramagnetic center. On the other hand, there are no signals for the as-burnt powder in Fig. 4d. The disappearance of silica signals in NiZn ferrite/SiO2 composite is due to dipolar interactions between 29Si nuclei and the paramagnetic Fe(III) cation, which provide efficient NMR relaxation sinks. This fact further reveals that a magnetic ferrite is formed from the as-burnt powder.
ARTICLE IN PRESS K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32
3.3. Effect of fuel on the EPR and SQUID spectra
Q4 Side band Q3
Side band
(a)
(b)
(c) (d) -200
-150
-100 ppm
-50
0
Fig. 4. 29Si NMR spectra of the dried gel obtained from different complexing agent (a) citric acid, (b) glycine, (c) hydrazine and (d) NiZn ferrite/SiO2 composite.
dH/dP
(a)
The EPR spectra of NiZn ferrite/SiO2 nanocomposites have been recorded at room temperature and the dependence of fuel is shown in Fig. 5. The EPR parameters (DHPP, g-factor and T2) obtained from Fig. 5 are given in Table 1. It can be observed that the EPR spectra show a single broad signal, indicating that the isolated Fe3+, Ni2+ and Zn2+ ions do not exist. Moreover, the EPR parameters (i.e. DHPP and g-factor) of composites decrease in the order: glycine sample4hydrazine sample4citric acid sample (Table 1). The results are due to the decreased particle size of NiZn ferrite for citric acid-assisted methods. This could make the dipole–dipole interactions in NiZn ferrite particles decrease, and then decrease of DHPP and g-factor [22]. The spin–spin relaxation process is the energy difference (DE) transferred to neighboring electrons and the relaxation time (T2) can be determined from the peak-to-peak linewidth (DHPP) according to gbDH 1=2 1 ; ¼ T2 _
(b)
(c)
0
1000
2000
29
3000 4000 Field (G)
5000
6000
Fig. 5. EPR spectra of NiZn ferrite/SiO2 synthesized with (a) glycine, (b) citric acid and (c) hydrazine.
DH 1=2 ¼
pffiffiffi 3DH PP
(2)
(in s1), where b is the Bohr magneton (9.274 1021 erg G1), DH1/2 the linewidth (in G) at half-height of the absorption peak, _ a constant (1.054 1027 erg s) [23]. In our case the T2 decreases in the order: citric acid sample4hydrazine sample4glycine sample. It is interesting to note that magnetic ions in the NiZn ferrite/SiO2 composites with different complexing agent can be found in different environments. The result may be due to the motion of ferrites is restricted by
Table 1 EPR and SQUID characteristics of NiZn ferrite/20 wt% SiO2 composites synthesized with different complexing agent at room temperature Sample
D (nm)
DHPP (G)
g-factor
T2 ( 1011 s)
Ms (emu/g)
Mr (emu/g)
Hc (G)
Glycine Hydrazine Citric acid
34 22 18
1740 1311 1174
2.31 2.30 2.27
3.11 4.12 5.39
10 6 12
0.80 0.46 2.33
33 29 69
D is the crystallite size, DHPP the peak-to-peak linewidth, T2 the spin–spin relaxation time, Ms the saturation magnetization, Mr the remnant magnetization, Hc the coercivity.
ARTICLE IN PRESS K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32
30
agglomeration at glycine sample. Thus, the T2 increases with decreasing particle size. The magnetic properties are the most important properties for ferrites depending on the processing conditions, microstructure, chemical composition and the amount and type of the additives [24,25]. Fig. 6 shows the magnetic hysteresis loops of NiZn ferrite/SiO2 nanocomposites with different fuel at room temperature. The hysteresis loops were measured to determine parameters such as the saturation magnetization (Ms), remnant magnetization (Mr) and coercivity (Hc), as shown in Table 1. It is evident from the Fig. 6 that the Ms, Mr and Hc decrease in the order: citric acid sample4glycine sample4hydrazine sample. High value of the Ms and low value of the Hc is required for soft magnetic applications, assuring high magnetic permeability and minimal energy losses [26].
Magnetization (emu/g)
15 10
citric acid glycine hydrazine
5 0 -5 -10 -15 -3000
-2000
-1000
(a)
0 1000 Field (G)
2000
3000
Therefore, we conclude that glycine sample processes produced more excellent magnetic properties. 3.4. Temperature dependence EPR spectra Fig. 7 shows the variation in the peak-to-peak linewidth (DHPP) with temperature. The values of DHPP continuously decrease with increase in temperature, indicating that the energy difference (DE) transferred to neighboring electrons ease with temperature. This is due to the fact that in a randomly oriented dispersed ferromagnet the absorption linewidth turns out to be a nonmonotonic function of temperature. At low temperature, the linewidth is large due to the scatter in direction of anisotropic field of particles. As the temperature increases, the tendency to make magnetic moment isotropic causes linewidth to decrease [27]. The energy between two adjacent degenerate spin energy levels DE has the same behavior of the linewidth. The reduction of linewidth may cause a reduction in the separate energy DE. The value of DE is given by the relation DE ¼ hn ¼ gbH 0 [23]. Thus, the g-value is defined as the constant of proportionality between the frequency and the field at which resonance occurs, and is proportional to the magnetic moment of the molecule being studied. Fig. 8 shows the variation in the g-factor with temperature of the NiZn ferrite/SiO2 composites. It was observed that g2200
4
glycine hydrazine citric acid
2000 1800
2
∆HPP (G)
Magnetization (emu/g)
6
citric acid glycine hydrazine
0 -2
1600 1400 1200
-4 1000
-6 -100 -80 -60 -40 -20 0 20 Field (G) (b)
40
60
80 100
Fig. 6. The magnetic hysteresis loops of NiZn ferrite/SiO2 composites synthesized with different complexing agent (a), low-field range (b).
800 180 200 220 240 260 280 300 320 340 360 380 400 420
Temperature (K) Fig. 7. The linewidth (DHPP) as a function of temperature of NiZn ferrite/SiO2 composites synthesized with different complexing agent.
ARTICLE IN PRESS K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32 2.60 glycine hydrazine citric acid
2.55 2.50
g factor
2.45 2.40 2.35 2.30 2.25 2.20
31
resonance parameters, g-factor and DHPP. Strong dipole interactions give a large DHPP and g-factor; further, strong superexchange interactions produce a small DHPP and g-factor [29]. The increased temperature should increase the motion of electrons, causing stronger superexchange interactions among the cations through oxygen ions and a decrease in DHPP and g-factor. Therefore, the T2 value increases with increasing temperature.
2.15 2.10 180 200 220 240 260 280 300 320 340 360 380 400 420
4. Conclusions
Temperature (K) Fig. 8. The g-factor as a function of temperature of NiZn ferrite/SiO2 composites synthesized with different complexing agent.
5.5 5.0
glycine hydrazine citric acid
T2 (10-11s)
4.5 4.0 3.5 3.0 2.5 2.0 1.5 180 200 220 240 260 280 300 320 340 360 380 400 420
Temperature (K) Fig. 9. The spin–spin relaxation time (T2) as a function of temperature of NiZn ferrite/SiO2 composites synthesized with different complexing agent.
value decreases with increasing temperature. The weakening of magnetic moment is responsible for the observed reduction in the g-factor [28]. The spin–spin relaxation T2 can be easily measured from the peak-to-peak EPR linewidth DHPP and g-factor according to the Eq. (2). The temperature dependence of T2 values in the NiZn ferrite/SiO2 composites with different fuel are shown in Fig. 9. Magnetic dipole interactions among particles and superexchange interactions between the magnetic ions though oxygen ions are two predominant factors that determine the EPR
Ni0.5Zn0.5Fe2O4/SiO2 nanocomposites have been prepared by a gel combustion technique without further calcinations. This route was based on the combustion of dried precursor mass due to an exothermic redox reaction between nitrate ions and fuel. The experiment shows that fuel type dramatically influenced the phase formation and electromagnetic properties of the final products. The following conclusions can be drawn from this study: 1. The particle size for preparing ferrite powders decreased in the order: glycine method4hydrazine method4citric acid method, thus citric acid was an effective chelating agent in producing fine ferrite powder. 2. From the FTIR and 29Si-NMR results, the magnetic interaction between NiZn ferrite and silica through Si–O–Fe bonds was formed and decreased in the order: glycine sample4hydrazine sample4citric acid sample. 3. From the elecrtomagnetic properties, we can see that glycine was an effective fuel in our combustion system. It was observed that glycine possessed the best magnetic properties with high value of Ms and low value of Hc.
Acknowledgements The authors thank the National Science Council of the Republic of China (Grant NSC 93-2113-M014-001). Authors wish to express their gratitude to Miss J.C. Chen of NSC Instrument Center for EPR analysis.
ARTICLE IN PRESS 32
K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32
References [1] J.B. da Silva, N.D.S. Mohallem, J. Magn. Magn. Mater. 226–230 (2001) 1393. [2] A.M. Abden, J. Magn. Magn. Mater. 185 (1998) 199. [3] K.I. Arshak, A. Ajina, D. Egan, Microelectron. J. 32 (2001) 113. [4] X.D. Tong, B. Xue, Y. Sun, Biotech. Prog. 17 (2001) 134. [5] Y. Choi, H.S. Shim, J.S. Lee, J. Alloy. Compd. 326 (2001) 56. [6] C. Caizer, Mater. Sci. Eng. B 100 (2003) 63. [7] G.S. Li, L.P. Li, R.L. Smith Jr., H. Inomata, J. Mol. Struct. 560 (2001) 87. [8] V.M. Bujoreamu, E. Segal, Solid State Sci. 3 (2001) 407. [9] J. Plocek, A. Hutlova, D. Niznansky, J. Bursik, J.L. Rehspringer, Z. Micka, J. Non-Cryst. Solids 315 (2003) 70. [10] C. Caizer, M. Popovici, C. Savii, Acta Mater. 51 (2003) 3607. [11] Z. Yue, J. Zhou, L. Li, H. Zhang, Z. Gui, J. Magn. Magn. Mater. 208 (2000) 55. [12] Z. Yue, L. Li, J. Zhou, H. Zhang, Z. Gui, Mater. Sci. Eng. B 64 (1999) 68. [13] R.K. Selvan, C.O. Augustin, L.J. Berchmans, R. Saraswathi, Mater. Res. Bull. 38 (2003) 41. [14] J. Huang, H. Zhuang, W.L. Li, Mater. Res. Bull. 38 (2003) 149. [15] K.H. Wu, Y.C. Chang, G.P. Wang, J. Magn. Magn. Mater. 269 (2004) 150.
[16] K.H. Wu, H.Y. Li, C.C. Yang, H.B. Chen, Appl. Surf. Sci. 228 (2004) 285. [17] K.H. Wu, W.C. Huang, J. Solid State Chem. 177 (2004) 3052. [18] R.D. Purohit, B.P. Sharma, K.T. Pillia, A.K. Tyagi, Mater. Res. Bull. 36 (2001) 2711. [19] L.A. Chick, L.R. Pederson, G.D. Maupin, J.L. Bates, L.E. Thomas, G.J. Exarhos, Mater. Lett. 10 (1990) 6. [20] N.S. Gajbhiye, U. Bhattacharya, V.S. Darshane, Thermochim. Acta 264 (1995) 219. [21] G. Ting-Kuo Fey, R.F. Shiu, V. Subramanian, C.L. Chen, Solid State Ion. 148 (2002) 291. [22] X. Li, G. Lu, S. Li, J. Alloys Compd. 235 (1996) 150. [23] J.F. Raber, Experimental Methods in Polymer Chemistry—Physical Principal and Applications, Wiley, New York, 1980, p. 332. [24] A. Verma, A.K. Saxena, D.C. Dube, J. Magn. Magn. Mater. 263 (2003) 228. [25] X.H. Wang, L.T. Li, Z.X. Yue, S.Y. Su, J. Magn. Magn. Mater. 271 (2004) 301. [26] B. Skolyszewska, W. Tokaz, K. Przybylski, Z. Kakol, Physica C 387 (2003) 290. [27] R. Massart, D. Zins, F. Gendron, M. Rivoire, R.V. Mehta, R.V. Upadhyay, P.S. Goyal, V.K. Aswal, J. Magn. Magn. Mater. 201 (1999) 73. [28] P. Kinnari, R.V. Upadhyay, R.V. Mehta, J. Magn. Magn. Mater. 252 (2002) 35. [29] L. Li, G. Li, R.L. Smith, H. Inomata, Chem. Mater. 12 (2000) 3705.
Journal of Magnetism and Magnetic Materials 298 (2006) 25–32 www.elsevier.com/locate/jmmm
Sol–gel auto-combustion synthesis of SiO2-doped NiZn ferrite by using various fuels K.H. Wua,, T.H. Tinga, M.C. Lia, W.D. Hob a
Department of Applied Chemistry, Chung Cheng Institute of Technology, NDU, No. 190, Sanyuan 1st Street, Dashi Jen, Tahsi, Taoyuan 335, Taiwan b Chemical Systems Research Division, Chung Shan Institute of Science and Technology, Taoyuan, Taiwan Received 22 December 2004; received in revised form 14 February 2005 Available online 24 March 2005
Abstract A nitrate–chelate–silica gel was prepared from metallic nitrates, citric acid and tetraethoxysilane (TEOS) by sol–gel process with different complexing agents such as glycine, hydrazine and citric acid, and it was further used to synthesize Ni0.5Zn0.5Fe2O4/20 wt% SiO2 nanocomposites by auto-combustion. The effect of varying complexing agent on the structural and magnetic properties of the composites was studied by FTIR, 29Si CP/MAS NMR, XRD, TEM, EPR and SQUID measurements. The complexing agent in the starting solution influenced the magnetic interaction between NiZn ferrite and silica, and then determined on the particle size. Further, the complexing agent type had a direct effect on the EPR parameters (DHPP, g-factor and T2) and SQUID parameters (Ms, Mr and Hc) of the as-synthesized powder. r 2005 Elsevier B.V. All rights reserved. Keywords: Complexing agent; Fuel; Ferrite; Silica; EPR
1. Introduction NiZn ferrites are one of the most versatile magnetic materials for general use, which have many applications in both low and high frequency devices and play a useful role in many technological applications such as microwave devices, power transformers in electronics, rod Corresponding author. Tel.: +886 33891716324; fax: +886 33808906. E-mail address: [email protected] (K.-H. Wu).
antennas, read/write heads for high speed digital tape, etc. because of their high resistivity, low dielectric losses, mechanical hardness, high Curie temperature and chemical stability [1–4]. The amorphous matrixes have been shown to play an important role in retarding the motion of the particles as well as the grain growth during the formation of nanocrystals. In addition, the electromagnetic properties of composite are affected not only by the compositions, additives and annealing conditions but also by the raw materials [5,6].
0304-8853/$ - see front matter r 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jmmm.2005.03.008
ARTICLE IN PRESS 26
K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32
Many synthetic approaches have been employed to prepare magnetic nanocrystals [7–10]. The sol–gel auto-combustion technique is a novel way with a unique combination of the chemical sol–gel process and the combustion process. The synthesis has been used to create different ceramic systems [11,12]. The success of the process is due to an intimate blending among the constituents using a suitable fuel or complexing agent (e.g. citric acid, urea, glycine, etc.) in an aqueous media and an exothermic redox reaction between the fuel and an oxidizer (i.e. nitrates) [13]. The powder characteristics like crystallite size, surface area, extent, and nature of agglomeration are primarily governed by enthalpy or flame temperature generated during combustion, which itself is dependent on nature of the fuel and fuel-to-oxidant ratio [14]. Chelating ligands, which contain carboxylate groups or aliphatic amine groups, are essential in the water-soluble complex precursor synthesis route. Citric acid (containing carboxylate groups), glycine (containing carboxylate and aliphatic amine groups) and hydrazine (containing aliphatic amine groups) were often used before in the synthesis of metallic oxides. Such types of complexing agent can effectively complex metal ions of varying ionic sizes, which helps in preventing their selective precipitation to maintain compositional homogeneity among the constituents. On the other hand, these can also serve as a fuel in the combustion reaction, being oxidized by nitrate ions. In our previous papers [15–17], SiO2-doped NiZn ferrite nanocomposites were prepared using sol–gel auto-combustion method. Many initial synthesis conditions such as silica content, calcinations temperature, solution pH and fuel-to-oxidant ratio have been varied in order to determine the optimal conditions for synthesizing the material. The present paper describes the synthesis of SiO2-doped NiZn ferrite nanocomposites by hydrolysis of tetraethoxysilane (TEOS) onto NiZn ferrite particle and the effect of varying the organic fuel on the chemical and electromagnetic properties. The spectroscopic characterization, crystallite sizes, EPR parameters (DHPP, g-factor and T2) and SQUID parameters (Ms, Mr and Hc) of composites are studied using FTIR, 29Si CP/MAS NMR, XRD, EPR and SQUID.
2. Experimental 2.1. Preparation of NiZn ferrite/SiO2 composite Analytical grade nickel nitrate, zinc nitrate, iron nitrate, fuel and TEOS were used as raw materials to prepare Ni0.5Zn0.5Fe2O4/20 wt% SiO2 nanocomposite. The initial molar ratio was Ni:Zn:Fe ¼ 1:1:4. First, 2.0 g Ni(NO3)2 6H2O, 2.05 g Zn(NO3)2 6H2O, and 11.12 g Fe(NO3)3 9H2O were dissolved in 20 ml of ethanol, then 20 wt% of TEOS and H2O in a molar ratio of 1:4 and the fuel, such as citric acid, glycine and hydrazine were added into the solution. The molar ratio of nitrates to fuel was 1:1. A small amount of ammonia was added to the solution to adjust the pH value to about 5. The entire mixture was thoroughly stirred for 6 h at 70 1C. Then, the mixed solution was poured into a teflon dish and heated 24 h at 60 1C and 3 h at 100 1C under a vacuum to obtain a dried gel. When ignited at any point, the dried gel burnt in a self-propagating combustion manner until all the gels were burnt out completely to form a loose powder. 2.2. Characterization of NiZn ferrite/SiO2 composite The phase identification of the as-burnt powder was performed using X-ray diffraction (XRD; SIEMENS D5000) with Cu Ka radiation. Average grain sizes (D) were determined from the XRD peaks using Scherrer’s formula as well as by a PHILIPS CM-200 transmission electron microscopy (TEM). Infrared spectra (IR) of the as-burnt powder were recorded on a Bomem DA 3.002 spectrophotometer from 400 to 4000 cm–1 by the KBr pellet method. The solid-state 29Si-NMR spectra of the gel precursor were determined using a Bruker MSL-400 with the cross-polarization combined with magic angle spinning (CP/MAS). The 29Si CP/MAS NMR provides a unique way to follow the structure of silica network and the magnetic interaction between iron (III) and silica. The electron paramagnetic resonance (EPR) spectra of the composites were recorded on a Bruker EMX-10 spectrometer operating at X-band ðn ¼ 9:6 GHzÞ with 100 kHz field modulations. DPPH
ARTICLE IN PRESS K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32
ðg ¼ 2:0036Þ was used as a field marker. The EPR g-value of the unknown sample can be calculated from g ¼ gs ðDH=H o Þgs ; where gs is the g-value of the reference sample, DH is the separation of the centers of the two spectra, Ho is the strength of the applied external field. The EPR DHPP was calculated from peak-to-peak linewidth. The EPR spectra were recorded at variable temperatures (200–400 K) using variable temperature controller. Magnetization measurements were performed in fields of up to 5 T using a Quantum Design SQUID magnetometer (Model MPMS5).
3. Results and discussion 3.1. Phase analysis and particle size of powder Fig. 1 shows the X-ray diffraction patterns of microcrystalline NiZn ferrite/SiO2 powders. All the as-burnt powders are a single phase NiZn ferrite with a spinel structure. This result indicates that the NiZn ferrites are directly formed after auto-combustion of gels. The direct transformation of crystalline ferrite from amorphous gel during combustion is surely due to the greater heat generated from the exothermic reaction of nitrates and fuel [18]. Crystallite sizes of the composite are calculated from the X-ray peak
Intensity (a.u.)
(311)
(a)
(b)
(c)
20
30
40 50 2 degrees
60
70
Fig. 1. XRD spectra of NiZn ferrite/SiO2 synthesized with (a) glycine, (b) citric acid and (c) hydrazine.
27
broadening of the (3 1 1) diffraction peak using Scherrer’s formula [12] D ¼ 0:9l=b cos y,
(1)
where D is the crystallite size in nanometres, l the radiation wavelength (0.154056 nm for Cu Ka), b the bandwidths at half-height and y the diffraction peak angle. The calculated crystallite sizes are 34, 22 and 18 nm, respectively, for crystalline synthesized with glycine, hydrazine and citric acid. It was evident that the crystallite size of the ferrite phase depending on the fuel, which was fairly consistent with the particle size determined by transmission electron microscopy (Fig. 2). TEM micrographs show that the composite consisted of a mixture of crystalline NiZn ferrite and amorphous silica at the nanometer level and homogeneous distribution. Due to the rapid explosive pyrolysis and the high fame temperature [19], the glycine method have larger particle size and good crystallinity, thus the signal to noise of the XRD is significantly higher than other. Furthermore, citric acid is a more effective complexing agent than hydrazine and glycine in producing fine ferrite powder with sol–gel auto-combustion method. 3.2. Infrared and
29
si cp/mas nmr spectra
Fig. 3 shows the IR spectra of the NiZn ferrite/ SiO2 with different complexing agent. The silica network is characterized by the absorptions at 1090, 805 and 462 cm–1, corresponding to the Si–O–Si anti-symmetric stretch, symmetric stretch and bending mode. The absorptions in the range 3700–3200 and 1628 cm–1 assigned to the stretching vibrations of the silanol groups (Si–OH). The bands at about 1384 and 570 cm–1 are attributed to stretching vibration of the anti-symmetric NO 3 and tetrahedral complexes of ferrite. The ferritestretching band is shifted to higher wave number as comparison with in NiZn ferrite (560 cm–1) [20], which can be ascribed to the formed interaction between the NiZn ferrite and silica through Si–O–Fe bonds [7]. On the other hand, the disappearance or decrease at 1384 cm–1 of the asburnt powder revealed that the NO 3 ions take part in the reaction during combustion. Therefore, the
ARTICLE IN PRESS K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32
570
(a)
4000
3500
3000 2500 2000 1500 Wavenumber (cm-1)
1000
570 462
1090
805
1384
(c)
574
(b) 1628
Transmittance (arb. unit)
28
500
Fig. 3. IR spectra of NiZn ferrite/SiO2 synthesized with (a) citric acid, (b) glycine and (c) hydrazine.
Fig. 2. TEM photograph of NiZn ferrite/SiO2 synthesized with (a) glycine, (b) hydrazine and (c) citric acid.
combustion can be considered as a thermally induced anionic redox reaction of the gel wherein the fuel acts as a reductant and the NO–3 ion acts as an oxidant [21]. 29 Si-NMR measurements confirm FTIR data showing the formation of magnetic interaction between the NiZn ferrite and silica. However, we got additional information on the morphologies, especially on the distribution of the transition metal in the amorphous system. The 29Si CP/MAS NMR spectra are collected for the dried gel and as burnt powder as shown in Fig. 4. The signals of Q3 (100 ppm) and Q4 (110 ppm) structural units in the dried gels (Figs. 4a–c) are partially overlapped and the intensities are weakened in the order: citric acid sample4hydrazine sample4 glycine sample, indicating that the glycine has the strongest complex ability. The results may be due to the Ni2+, Zn2+ and Fe3+ ions dispersion homogeneously in silica gel and in close proximity to the paramagnetic center. On the other hand, there are no signals for the as-burnt powder in Fig. 4d. The disappearance of silica signals in NiZn ferrite/SiO2 composite is due to dipolar interactions between 29Si nuclei and the paramagnetic Fe(III) cation, which provide efficient NMR relaxation sinks. This fact further reveals that a magnetic ferrite is formed from the as-burnt powder.
ARTICLE IN PRESS K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32
3.3. Effect of fuel on the EPR and SQUID spectra
Q4 Side band Q3
Side band
(a)
(b)
(c) (d) -200
-150
-100 ppm
-50
0
Fig. 4. 29Si NMR spectra of the dried gel obtained from different complexing agent (a) citric acid, (b) glycine, (c) hydrazine and (d) NiZn ferrite/SiO2 composite.
dH/dP
(a)
The EPR spectra of NiZn ferrite/SiO2 nanocomposites have been recorded at room temperature and the dependence of fuel is shown in Fig. 5. The EPR parameters (DHPP, g-factor and T2) obtained from Fig. 5 are given in Table 1. It can be observed that the EPR spectra show a single broad signal, indicating that the isolated Fe3+, Ni2+ and Zn2+ ions do not exist. Moreover, the EPR parameters (i.e. DHPP and g-factor) of composites decrease in the order: glycine sample4hydrazine sample4citric acid sample (Table 1). The results are due to the decreased particle size of NiZn ferrite for citric acid-assisted methods. This could make the dipole–dipole interactions in NiZn ferrite particles decrease, and then decrease of DHPP and g-factor [22]. The spin–spin relaxation process is the energy difference (DE) transferred to neighboring electrons and the relaxation time (T2) can be determined from the peak-to-peak linewidth (DHPP) according to gbDH 1=2 1 ; ¼ T2 _
(b)
(c)
0
1000
2000
29
3000 4000 Field (G)
5000
6000
Fig. 5. EPR spectra of NiZn ferrite/SiO2 synthesized with (a) glycine, (b) citric acid and (c) hydrazine.
DH 1=2 ¼
pffiffiffi 3DH PP
(2)
(in s1), where b is the Bohr magneton (9.274 1021 erg G1), DH1/2 the linewidth (in G) at half-height of the absorption peak, _ a constant (1.054 1027 erg s) [23]. In our case the T2 decreases in the order: citric acid sample4hydrazine sample4glycine sample. It is interesting to note that magnetic ions in the NiZn ferrite/SiO2 composites with different complexing agent can be found in different environments. The result may be due to the motion of ferrites is restricted by
Table 1 EPR and SQUID characteristics of NiZn ferrite/20 wt% SiO2 composites synthesized with different complexing agent at room temperature Sample
D (nm)
DHPP (G)
g-factor
T2 ( 1011 s)
Ms (emu/g)
Mr (emu/g)
Hc (G)
Glycine Hydrazine Citric acid
34 22 18
1740 1311 1174
2.31 2.30 2.27
3.11 4.12 5.39
10 6 12
0.80 0.46 2.33
33 29 69
D is the crystallite size, DHPP the peak-to-peak linewidth, T2 the spin–spin relaxation time, Ms the saturation magnetization, Mr the remnant magnetization, Hc the coercivity.
ARTICLE IN PRESS K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32
30
agglomeration at glycine sample. Thus, the T2 increases with decreasing particle size. The magnetic properties are the most important properties for ferrites depending on the processing conditions, microstructure, chemical composition and the amount and type of the additives [24,25]. Fig. 6 shows the magnetic hysteresis loops of NiZn ferrite/SiO2 nanocomposites with different fuel at room temperature. The hysteresis loops were measured to determine parameters such as the saturation magnetization (Ms), remnant magnetization (Mr) and coercivity (Hc), as shown in Table 1. It is evident from the Fig. 6 that the Ms, Mr and Hc decrease in the order: citric acid sample4glycine sample4hydrazine sample. High value of the Ms and low value of the Hc is required for soft magnetic applications, assuring high magnetic permeability and minimal energy losses [26].
Magnetization (emu/g)
15 10
citric acid glycine hydrazine
5 0 -5 -10 -15 -3000
-2000
-1000
(a)
0 1000 Field (G)
2000
3000
Therefore, we conclude that glycine sample processes produced more excellent magnetic properties. 3.4. Temperature dependence EPR spectra Fig. 7 shows the variation in the peak-to-peak linewidth (DHPP) with temperature. The values of DHPP continuously decrease with increase in temperature, indicating that the energy difference (DE) transferred to neighboring electrons ease with temperature. This is due to the fact that in a randomly oriented dispersed ferromagnet the absorption linewidth turns out to be a nonmonotonic function of temperature. At low temperature, the linewidth is large due to the scatter in direction of anisotropic field of particles. As the temperature increases, the tendency to make magnetic moment isotropic causes linewidth to decrease [27]. The energy between two adjacent degenerate spin energy levels DE has the same behavior of the linewidth. The reduction of linewidth may cause a reduction in the separate energy DE. The value of DE is given by the relation DE ¼ hn ¼ gbH 0 [23]. Thus, the g-value is defined as the constant of proportionality between the frequency and the field at which resonance occurs, and is proportional to the magnetic moment of the molecule being studied. Fig. 8 shows the variation in the g-factor with temperature of the NiZn ferrite/SiO2 composites. It was observed that g2200
4
glycine hydrazine citric acid
2000 1800
2
∆HPP (G)
Magnetization (emu/g)
6
citric acid glycine hydrazine
0 -2
1600 1400 1200
-4 1000
-6 -100 -80 -60 -40 -20 0 20 Field (G) (b)
40
60
80 100
Fig. 6. The magnetic hysteresis loops of NiZn ferrite/SiO2 composites synthesized with different complexing agent (a), low-field range (b).
800 180 200 220 240 260 280 300 320 340 360 380 400 420
Temperature (K) Fig. 7. The linewidth (DHPP) as a function of temperature of NiZn ferrite/SiO2 composites synthesized with different complexing agent.
ARTICLE IN PRESS K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32 2.60 glycine hydrazine citric acid
2.55 2.50
g factor
2.45 2.40 2.35 2.30 2.25 2.20
31
resonance parameters, g-factor and DHPP. Strong dipole interactions give a large DHPP and g-factor; further, strong superexchange interactions produce a small DHPP and g-factor [29]. The increased temperature should increase the motion of electrons, causing stronger superexchange interactions among the cations through oxygen ions and a decrease in DHPP and g-factor. Therefore, the T2 value increases with increasing temperature.
2.15 2.10 180 200 220 240 260 280 300 320 340 360 380 400 420
4. Conclusions
Temperature (K) Fig. 8. The g-factor as a function of temperature of NiZn ferrite/SiO2 composites synthesized with different complexing agent.
5.5 5.0
glycine hydrazine citric acid
T2 (10-11s)
4.5 4.0 3.5 3.0 2.5 2.0 1.5 180 200 220 240 260 280 300 320 340 360 380 400 420
Temperature (K) Fig. 9. The spin–spin relaxation time (T2) as a function of temperature of NiZn ferrite/SiO2 composites synthesized with different complexing agent.
value decreases with increasing temperature. The weakening of magnetic moment is responsible for the observed reduction in the g-factor [28]. The spin–spin relaxation T2 can be easily measured from the peak-to-peak EPR linewidth DHPP and g-factor according to the Eq. (2). The temperature dependence of T2 values in the NiZn ferrite/SiO2 composites with different fuel are shown in Fig. 9. Magnetic dipole interactions among particles and superexchange interactions between the magnetic ions though oxygen ions are two predominant factors that determine the EPR
Ni0.5Zn0.5Fe2O4/SiO2 nanocomposites have been prepared by a gel combustion technique without further calcinations. This route was based on the combustion of dried precursor mass due to an exothermic redox reaction between nitrate ions and fuel. The experiment shows that fuel type dramatically influenced the phase formation and electromagnetic properties of the final products. The following conclusions can be drawn from this study: 1. The particle size for preparing ferrite powders decreased in the order: glycine method4hydrazine method4citric acid method, thus citric acid was an effective chelating agent in producing fine ferrite powder. 2. From the FTIR and 29Si-NMR results, the magnetic interaction between NiZn ferrite and silica through Si–O–Fe bonds was formed and decreased in the order: glycine sample4hydrazine sample4citric acid sample. 3. From the elecrtomagnetic properties, we can see that glycine was an effective fuel in our combustion system. It was observed that glycine possessed the best magnetic properties with high value of Ms and low value of Hc.
Acknowledgements The authors thank the National Science Council of the Republic of China (Grant NSC 93-2113-M014-001). Authors wish to express their gratitude to Miss J.C. Chen of NSC Instrument Center for EPR analysis.
ARTICLE IN PRESS 32
K.-H. Wu et al. / Journal of Magnetism and Magnetic Materials 298 (2006) 25–32
References [1] J.B. da Silva, N.D.S. Mohallem, J. Magn. Magn. Mater. 226–230 (2001) 1393. [2] A.M. Abden, J. Magn. Magn. Mater. 185 (1998) 199. [3] K.I. Arshak, A. Ajina, D. Egan, Microelectron. J. 32 (2001) 113. [4] X.D. Tong, B. Xue, Y. Sun, Biotech. Prog. 17 (2001) 134. [5] Y. Choi, H.S. Shim, J.S. Lee, J. Alloy. Compd. 326 (2001) 56. [6] C. Caizer, Mater. Sci. Eng. B 100 (2003) 63. [7] G.S. Li, L.P. Li, R.L. Smith Jr., H. Inomata, J. Mol. Struct. 560 (2001) 87. [8] V.M. Bujoreamu, E. Segal, Solid State Sci. 3 (2001) 407. [9] J. Plocek, A. Hutlova, D. Niznansky, J. Bursik, J.L. Rehspringer, Z. Micka, J. Non-Cryst. Solids 315 (2003) 70. [10] C. Caizer, M. Popovici, C. Savii, Acta Mater. 51 (2003) 3607. [11] Z. Yue, J. Zhou, L. Li, H. Zhang, Z. Gui, J. Magn. Magn. Mater. 208 (2000) 55. [12] Z. Yue, L. Li, J. Zhou, H. Zhang, Z. Gui, Mater. Sci. Eng. B 64 (1999) 68. [13] R.K. Selvan, C.O. Augustin, L.J. Berchmans, R. Saraswathi, Mater. Res. Bull. 38 (2003) 41. [14] J. Huang, H. Zhuang, W.L. Li, Mater. Res. Bull. 38 (2003) 149. [15] K.H. Wu, Y.C. Chang, G.P. Wang, J. Magn. Magn. Mater. 269 (2004) 150.
[16] K.H. Wu, H.Y. Li, C.C. Yang, H.B. Chen, Appl. Surf. Sci. 228 (2004) 285. [17] K.H. Wu, W.C. Huang, J. Solid State Chem. 177 (2004) 3052. [18] R.D. Purohit, B.P. Sharma, K.T. Pillia, A.K. Tyagi, Mater. Res. Bull. 36 (2001) 2711. [19] L.A. Chick, L.R. Pederson, G.D. Maupin, J.L. Bates, L.E. Thomas, G.J. Exarhos, Mater. Lett. 10 (1990) 6. [20] N.S. Gajbhiye, U. Bhattacharya, V.S. Darshane, Thermochim. Acta 264 (1995) 219. [21] G. Ting-Kuo Fey, R.F. Shiu, V. Subramanian, C.L. Chen, Solid State Ion. 148 (2002) 291. [22] X. Li, G. Lu, S. Li, J. Alloys Compd. 235 (1996) 150. [23] J.F. Raber, Experimental Methods in Polymer Chemistry—Physical Principal and Applications, Wiley, New York, 1980, p. 332. [24] A. Verma, A.K. Saxena, D.C. Dube, J. Magn. Magn. Mater. 263 (2003) 228. [25] X.H. Wang, L.T. Li, Z.X. Yue, S.Y. Su, J. Magn. Magn. Mater. 271 (2004) 301. [26] B. Skolyszewska, W. Tokaz, K. Przybylski, Z. Kakol, Physica C 387 (2003) 290. [27] R. Massart, D. Zins, F. Gendron, M. Rivoire, R.V. Mehta, R.V. Upadhyay, P.S. Goyal, V.K. Aswal, J. Magn. Magn. Mater. 201 (1999) 73. [28] P. Kinnari, R.V. Upadhyay, R.V. Mehta, J. Magn. Magn. Mater. 252 (2002) 35. [29] L. Li, G. Li, R.L. Smith, H. Inomata, Chem. Mater. 12 (2000) 3705.