December 15, 2010 10:7 WSPC/147-MPLB ...

December 15, 2010 10:7 WSPC/147-MPLB

S0217984911025626

Modern Physics Letters B, Vol. 25, No. 3 (2011) 211–222 c World Scientific Publishing Company

DOI: 10.1142/S0217984911025626

STRUCTURAL, ELECTRICAL AND MAGNETIC CHARACTERIZATION OF Ni Cu Zn SPINEL FERRITES

M. PENCHAL REDDY∗ , M. VENKATA RAMANA, N. RAMA MANOHAR REDDY, K. V. SIVA KUMAR and R. RAMA KRISHNA REDDY Department of Physics, Sri Krishnadevaraya University, Anantapur-515055, India ∗ [email protected] W. MADHURI School of Advanced Sciences, VIT University, Vellore 632014, India [email protected] K. SIVA KUMAR REDDY and P. SREEDHARA REDDY Department of Physics, Sri Venkateswara University, Tirupathi-517 502, India V. R. K. MURTHY Microwave Laboratory, Indian Institute of Science, Chennai 600036, India Received 19 April 2010 Revised 6 May 2010 Ni–Cu–Zn ferrite materials have been extensively used in multilayer chip inductors because of their remarkable properties at higher frequencies. In the present work, single phase Ni0.35 Cu0.05 Zn0.60 Fe1.98 O4−δ ferrite, has been prepared by microwave sintered (MS) method. In comparison with the conventional sintering method (CS), the sintering temperature and time for this MS method were significantly reduced to 30 min and 950◦ C from 5 h and 1250◦ C for the CS process. The frequency dependence of the dielectric properties such as dielectric constant (ε0 ), dielectric loss (tan δ) were studied. The temperature dependence of magnetic initial permeability (µi ) was studied. The saturation magnetization was also studied as a function of magnetic field. These microwave sintered ferrites results were compared with the properties of ferrites prepared by conventional sintering method in normal heating. Microwave sintering improves structural as well as electromagnetic parameters measured and thus makes the ferrite more suitable in microwave applications and electromagnetic devices. Keywords: Ferrites; microwave sintering; electrical properties; magnetic properties.

1. Introduction Spinel ferrites have very important structural, electrical and magnetic properties that are dependent on several factors such as method of preparation, sintering 211


212

S0217984911025626

M. P. Reddy et al.

temperature, sintering time, substitution of cations and microstructure, etc. MLCI is a passive surface mounting device (SMD) and widely-used component in various electronic circuits and they help greatly in the miniaturization of many latest electronic products including mobile phones, notebook computers and video cameras.1,2 MLCIs are produced by cofiring ferrite layers with an internal silver conductor. Ni–Cu–Zn ferrites have been the dominant ferrite materials for MLCI due to its better magnetic properties at high frequency and low sintering temperature.3 – 5 Since silver (internal conductor) has a melting point of 961◦ C, a much lower (< 961◦ C) sintering temperature of ferrite is preferred to suppress the diffusion of silver metal into the ferrite core. This metal diffusion into the ferrite body decreases the resistivity of core. The sintering temperature of the ferrite can be decreased by using microwave sintering (MS) method. Nowadays, microwave sintering method is used to prepare a Ni–Cu–Zn ferrites samples at a low sintering temperature (< 950◦C) without any sintering aids. The microwave sintering technique has unique advantages over conventional sintering (CS) technique. The major advantages of microwave sintering were found to include higher energy efficiency, higher post sintering density,6 – 9 and lower sintering temperatures10,11 compared with conventional sintering (CS). The essential difference in the microwave and conventional sintering process is in the heating mechanism. In conventional sintering, heat is generated by external heating elements and then diffuses into the test sample via radiation, conduction and convection producing high temperature gradients and internal stress.12 In microwave sintering, the heat is generated internally within the test sample, by rapid oscillation of dipoles at microwave frequencies,13 instead of diffusion from external sources. As a result of this internal and volumetric heating, it is possible to sinter the materials rapidly and uniformly. So the microwave sintering technique is expected to prepare the high-permeability Ni–Cu–Zn ferrites at relatively low sintering temperatures. The advantages of microwave sintering and experimental setup for ceramics have been reported in the earlier paper published from this laboratory.14 In this paper, we choose Ni–Cu–Zn ferrite with the composition of Ni0.35 Cu0.05 Zn0.60 Fe1.98 O4−δ , which had relatively small magnetocrystalline anisotropy, magnetostriction and high Curie temperature, to produce the high-permeability Ni–Cu–Zn ferrite. The low-sintering temperature Ni–Cu–Zn ferrite, Ni0.35 Cu0.05 Zn0.60 Fe1.98 O4−δ , was prepared using microwave sintering technique and the microstructure and electromagnetic properties were investigated and compared with the CS samples. 2. Experimental Details 2.1. Sample preparation Ni–Cu–Zn ferrites of the cubic spinel structure having the general formula Ni0.35 Cu0.05 Zn0.60 Fe1.98 O4−δ was prepared by employing the conventional double


S0217984911025626

Structural, Electrical and Magnetic Characterization of Ni–Cu–Zn Spinel Ferrites

213

sintering and microwave sintering methods using analytical grade NiO, CuO, ZnO and Fe2 O3 . The stoichiometric proportions of these constituent oxides were weighed, intimately mixed and the resulting powders were ball-milled using a planetary ball mill (Restch PM 200, Germany) in agate bowls with agate balls in aqueous medium for 20 h. The slurry was dried and loosely pressed into cakes using a hydraulic press. These cakes were pre-sintered at a temperature of 800 ◦ C for 4 h in closed alumina crucibles. The pre-sintered cakes removed from the furnace were crushed and ball-milled in an aqueous medium in agate bowls with agate balls for another 30 h to obtain fine particle size. These slurries after drying were sieved to obtain a uniform particle size. The green powder thus obtained was then pressed using a suitable die into two sets of toroids of dimensions 1.2 cm outer diameter (OD) and 0.8 cm inner diameter (ID) and 0.4 cm height with a hydraulic press at a pressure of 200 MPa using 2% PVA solution as a binder. The green pressed samples were divided into two batches (A and B): one was processed with conventional sintering process while another with the microwave sintering process. In CS process, the samples (batch A) were sintered in electrical programmable furnace at 1250 ◦C for 5 h and were cooled to room temperature at 80 ◦ C h−1 . The microwave sintering was carried out using a microwave furnace designed in these laboratories.15 Another set (batch B) of the green pressed samples were subjected to the microwave furnace. The microwave furnace used was a commercially available modified domestic (SHARP) microwave oven with a single magnetron operating at 2.45 GHz frequency, at a power output of 1.1 kW. The green pressed samples were finally sintered at 950◦ C for 30 min. 2.2. Characterization The phase structure for both sets of samples was characterized by X-ray diffraction (XRD) with CuKα radiation. The density of the materials was measured using Archimedes principle. The microstructure of the fracture surfaces was studied by CRL-ZESIS-EVO-MAI5 scanning electron microscopy (SEM). The compositions of the sintered samples were confirmed using an energy-dispersive X-ray spectrometry (EDS) facility attached to the scanning electron microscope. An HP3532-50 LCR Hitester impedance analyzer was used to measure the frequency dependence of dielectric properties from 100 Hz to 1 MHz, and also measure the temperature dependence of magnetic permeability from 30◦ C to 250◦ C. The magnetic characteristics were measured with VSM-Lakeshore 665 vibrating sample magnetometer (VSM). 3. Results and Discussion 3.1. Structural characterization Figure 1 shows the XRD patterns for the samples sintered by the microwave and conventional sintering techniques and shows a single-phase formation with spinel structure. No second phase was detected. The lattice constant ‘a’ for the MS sample


214

S0217984911025626

(400)

(511) (400)

(222)

(M S )

(422)

(220)

Intensity (a.u)

(311)

M. P. Reddy et al.

(C S ) 20

30

40

50

60

70

T h e ta (d e g )

Fig. 1. XRD patterns of Ni–Cu–Zn ferrite samples sintered by MS at 950◦ C for 30 min and CS at 1250◦ C for 5 h.

Table 1. Lattice constant, bulk density, average grain size and main electromagnetic properties of MS and CS samples.

Sample

‘a’ ˚) (A

Density (gm/cm3 )

Grain size (µm)

ρ (Ω cm)

ε0 at 1 MHz

tan δ at 1 MHz

µi at 1 MHz

Ms (emu/g)

MS CS

8.334 8.312

5.236 4.856

5.3 1.8

3.5 × 109 5.2 × 106

21.8 17.9

0.071 2.214

1690 1230

47.3 38.2

has been calculated and compared with that obtained from the MS sample and is depicted in Table 1. The value of ‘a’ for the MS samples is seen to be slightly lower than that obtained for the CS sample. The value of ‘a’ which is 8.334 ˚ A for the MS sample is closer to the lattice constant for the stoichiometric Ni-ferrite (8.331 ˚ A) as reported earlier.16 The closer value of ‘a’ for the MS sample, suggests that fewer oxygen ions have been lost during sintering by this technique. The density has been found to increase for the MS sample (Table 1) and this supports the idea that the microwave-sintering technique improves the density of ferrites due to enhanced reaction kinetics. In the present study, the composition prepared by the conventional sintering technique requires at least 5 long hours of sintering to reach the density of 94% of the theoretical density, whereas the microwave sintering technique gives 95% density which is almost similar, with only 30 minutes of soaking time. Hence, microwave sintering technique significantly enhanced densification in a much shorter time and it is more cost-effective.17 The average grain size for the MS sample as estimated from the SEM photographs have been compared with that of the CS sample in Table 1 and it


S0217984911025626


215

Fig. 2. SEM photographs of Ni–Cu–Zn ferrite materials processed by (a) MS process (950◦ C for 30 min) and (b) CS process (250◦ C for 5 h).

is found that the MS processed samples give much larger grains. Figure 2 shows the SEM photographs of the MS and CS sample. It is observed that the grains are much larger and more uniform for the MS sample. It can be seen that the grain size of MS sample is 5.3 µm, while that of CS samples is about 1.8 µm. Figure 2 also reveals that MS sample has a denser structure than CS samples. In order to confirm the chemical composition, EDS analysis was carried out for the microwave sintered Ni0.35 Cu0.05 Zn0.60 Fe1.98 O4−δ ferrite sample and is presented in Fig. 3. It may be mentioned, from the present study, that the compositional analysis obtained with this method could accurately quantify the Ni, Cu and Zn contents within 2–3% error, but it is unable to provide an accurate estimation of O content.18


216

S0217984911025626

M. P. Reddy et al.

Fig. 3.

EDS spectrum of microwave sintered Ni0.35 Cu0.05 Zn0.60 Fe1.98 O4−δ .

3.2. DC resistivity The DC electrical resistivity is an important property of low-temperature sintered (MS) ferrite for MLCI applications. The DC resistivity of the samples was measured using a two-probe method and the results are presented in Table 1. Table 1 shows that the DC resistivity for the MS sample is 103 times higher than that of the CS sample. This can be related with the smaller loss of nickel during sintering and hence fewer formation of ferrous ions which contributes mainly to the conductivity of the ferrites.

3.3. Frequency dependence of dielectric constant The dielectric properties such as dielectric constant (ε0 ), dielectric loss (tan δ) are important for multilayer chip inductors used in high frequency range. The frequency dependence of dielectric constant for Ni–Cu–Zn ferrite samples prepared by CS and MS process is shown in Fig. 4. The plots illustrate that the dielectric constant (ε0 ) decreases with increasing frequency reaching constant value at higher frequencies. The variation reveals the dispersion due to Maxwell–Wanger19,20 type interfacial polarization in agreement with Koop’s phenomenological theory.21 This is the normal behavior for ferrites.22 This is similar to the results reported earlier for Ni–Cu16 and Ni–Cu–Zn23 ferrites. In ferrites the mechanism of electric conduction and dielectric polarization are similar.24 The electron exchange between Fe2+ and Fe3+ ions gives local displacement of electrons in the direction of an applied electric field, which determine the polarization. Iwauchi and Ikeda25 has reported a strong correlation between the conduction mechanism and the dielectric behavior of the ferrites.


S0217984911025626


CS

217

MS

120

∋

dielectric Constant (ε )

100

80

60

40

20 2

10

3

10

4

10

5

10

6

10

Frequency (Hz)

Fig. 4. Frequency dependence of dielectric constant for Ni–Cu–Zn ferrite samples prepared by CS and MS techniques.

The polarization decreases substantially with increase in frequency and reaches constant ‘a’ value due to fact that, beyond certain frequency of external field the electron exchange between Fe2+ and Fe3+ ions cannot follow the alternating field. The higher values of dielectric constant observed at lower frequencies could be explained on the basis of Koops phenomenological theory. They are due to the predominance of species like Fe2+ ions, interfacial dislocation pile-ups, oxygen vacancies, and grain boundary defects.19,20 The decrease in dielectric constant with frequency due to the fact that any species contributing to polarizability is bound to show the lagging behind of the applied field at higher frequencies. The magnitude of the exchange controls the dielectric constant and this exchange depends on the concentration of the Fe3+ /Fe2+ ions pairs present on the B sites. Besides, the Fe2+ ion concentration in a ferrite material is a characteristic property dependent on many factors, such as method of preparation, sintering temperature, sintering time, atmosphere and grain structure, etc. It is known that some Fe2+ ions are formed from the evaporation of nickel ions during sintering. But microwave sintering may have gives less chance for evaporation of nickel and hence leading to the reduction of the Fe3+ to Fe2+ ions. These Fe2+ ions have the affinity for the B sites and therefore Fe3+ /Fe2+ concentration at the B sites is reduced. The electron exchange between the ions is reduced, leading to a decrease in the polarization and hence a decrease in the polarization and hence a decrease in dielectric constant expected for the MS sample. 3.4. Frequency dependence of dielectric loss tangent The frequency dependence of dielectric loss (tan δ) for Ni–Cu–Zn ferrite samples prepared by CS and MS process is shown in Fig. 5. It can be seen that the loss


218

S0217984911025626

M. P. Reddy et al. 6

CS

MS

Dielectric Loss (tan δ)

5

4

3

2

1

0 2

10

3

10

4

10

5

10

6

10

Frequency (Hz)

Fig. 5. Frequency dependence of dielectric loss for Ni–Cu–Zn ferrite samples prepared by CS and MS techniques.

tangent decreases continuously with increasing frequency. None of the samples exhibit the loss peak. Like dielectric constant, dielectric loss is also known to be inversely proportional to the frequency where the value is high at low frequency and low at higher frequency and becomes independent of frequency beyond certain limit.26 At higher frequencies, the dielectric loss (tan δ) is of the order of 10−2 . The low loss at higher frequencies identifies the potential of these ferrites for high-frequency applications.26 It can be seen from the figure that the MS samples (at higher frequency) has larger dielectric constant and smaller loss comparing to CS samples. Table 1 gives the room temperature values of dielectric constant and dielectric loss measured at 1 MHz. The dense structure of the MS sample, as confirmed by SEM result, indicates less air-holes which means low dielectric phase, thus lead to higher values of dielectric constant. Simultaneously, the MS low dielectric loss in MS samples results from the lower Fe2+ content in the materials.27 – 29 In the MS process, the shorter sintering time restrain the evaporation of O2− ions, and leading to the lower Fe2+ content in the material, so the dielectric loss decreased compared to the CS samples. Therefore, a better dielectric property of the Ni–Cu–Zn ferrite was obtained with MS process in comparison to CS process. 3.5. Temperature dependence of magnetic initial permeability For MLCI applications, the temperature dependence of magnetic initial permeability (µi ) is very important. The magnetic initial permeability (µi ) as a function of temperature (at constant frequency, 10 kHz) from room temperature


S0217984911025626


CS

4000

219

MS

3500

Initial Permeability (µι)

3000 2500 2000 1500 1000 500 0

30

60

90

120

150

180

210

240

o

Temperature ( C)

Fig. 6. Temperature dependence of magnetic initial permeability for Ni–Cu–Zn ferrite samples prepared by CS and MS techniques.

to Curie transition point was also studied. The temperature variation of magnetic initial permeability (µi ) for both samples is shown in Fig. 6. It is evident from figure that the magnetic initial permeability remains constant over a wide range for the microwave sintered sample. It can also be noticed from the figure that the microwave sintered ferrites show good thermal stability. This indicates that the shape of the permeability-temperature curves depends on the preparation conditions. It can be noted from Fig. 5 that as the temperature increases, the magnetic initial permeability (µi ) remains constant up to a certain temperature and increases to a peak value and then abruptly falls to a minimum value. The temperature at which this abrupt fall takes place is the magnetic Curie transition temperature (Tc ). At this temperature, the specimens transform from the ferrimagnetic phase to the paramagnetic phase. It suggests that MS sample has larger magnetic initial permeability than the CS samples. The magnetic initial permeability for the material is expected to strongly depend on the microstructure, as the magnetic initial permeability represents the mobility of magnetic domain wall in response to the small applied field.30 It is also seen from Fig. 2 that the MS samples result in large-grain samples leading to enhancement in permeability. 3.6. Magnetic hysteresis Figure 7 shows the magnetization versus external magnetic field (M–H curve) for the Ni–Cu–Zn ferrite materials prepared by MS and CS. For the CS materials, the saturation magnetization (Ms ) is 38.2 emu/g, while for MS samples the Ms is 47.3 emu/g. It is clear that the MS technique gives rise to larger grains as


220

S0217984911025626

M. P. Reddy et al.

60

CS

MS

Magnetization (emu/g)

40

20

0

-20

-40

-60 -6000

-4000

-2000

0

2000

4000

6000

Magnetic Field (Oe) Fig. 7.

M–H hysteresis loops for Ni–Cu–Zn ferrite samples prepared by CS and MS techniques.

compare with the CS technique. Larger grains indicate larger-sized domains and thus resulting in more magnetization in response to the external magnetic field. This statement is also asserted by the microstructure of MS sample showing completely crystallized bigger crystal grains (Fig. 2(a)). The larger Ms is also beneficial to minimize the devices size and decrease the loss. At the same time, the smaller Ms of the CS sample stems from the small crystal grain and this provides more pinning sites and grain boundaries.

4. Conclusion In this paper, we have prepared a suitable composition of Ni–Cu–Zn ferrite materials by the MS process and the microstructure and electromagnetic properties were investigated and compared to CS samples. Our results show that the MS material has excellent microstructure and good electromagnetic properties for application in MLCI. The particular composition of Ni–Cu–Zn (Ni0.35 Cu0.05 Zn0.60 Fe1.98 O4−δ ) was chosen on pragmatic considerations and was found to be superior in all respects when compared with the other samples reported earlier by using the same technique as well as different compositions of Ni–Cu–Zn prepared by the same technique. Moreover, using MS technique, the sintering temperature and soaking time can be decreased from conventional 1250◦C to 950◦ C and 5 h to 30 min, and the processing efficiency can also be improved significantly. Therefore, the MS process is an important processing technique to exploit these materials in MLCI technology.


S0217984911025626


221

Acknowledgments This work was supported by the financial assistance provided by the Defence Research and Development Organization (DRDO), New Delhi, India. The authors are thankful to the authorities of Sri Krishnadevaraya University, Anantapur, for providing facilities. References 1. X. Qi, J. Zhou, Z. Yue et al., Key Eng. Mater. 593 (2002) 224. 2. B. Li, Z. X. Yue, X. W. Qi, J. Zhou, Z. L. Gui and L. T. Li, Mater. Sci. Eng. B 99 (2003) 252. 3. T. Nakamura, J. Magn. Magn. Mater. 168 (1997) 285. 4. J. H. Jean, C. H. Lee and W. S. Kou, J. Am. Ceram. Soc. 82 (1999) 343. 5. K. O. Low and F. R. Sale, J. Magn. Magn. Mater. 246 (2002) 30. 6. V. Tsakaloudi, E. Papazoglou and V. T. Zaspalis, Mater. Sci. Eng. B 106 (2004) 289. 7. M. Sorescua, L. Diamandescua, R. Peelamedu, R. Roy and P. Yadoji, J. Magn. Magn. Mater. 279 (2004) 195. 8. N. A. Travitzky, A. Goldstein, O. Avsian and A. Singurindi, Mater. Sci. Eng. A 2 (2000) 225. 9. J. Majling, P. Znasik, J. Cheng, D. Agrawal and R. Roy, J. Mater. Res. 10 (1995) 2411. 10. A. Goldstein, N. Travitzky, A. Singurindi and M. Kravchik, J. Eur. Ceram. Soc. 19 (1999) 2067. 11. C. Siligardi, C. Leonelli, F. Bondioli, A. Corradi and G. C. Pellakani, J. Eur. Ceram. Soc. 20 (2000) 177. 12. A. C. F. M. Costa, E. Tortella, M. R. Morelli and R. H. G. A. Kiminami, J. Magn. Magn. Mater. 256 (2003) 174. 13. A. R. Phani and S. Santucci, J. Non-Cryst. Solids 352 (2006) 4093. 14. W. Madhuri, M. Penchal Reddy, N. Ramamanohar Reddy, K. V. Siva Kumar and V. R. K. Murty, J. Phys. D: Appl. Phys. 42 (2009) 165007. 15. M. Penchal Reddy, PhD thesis, Srikrishnadevaraya University, Anatapur, India, 2010. 16. D. R. Patil and B. K. Chougule, Mater. Chem. Phys. 117 (2009) 35. 17. S. R. Murthy, Bull. Mater. Sci. 26 (2003) 499. 18. J. A. Paulsen, C. C. H. Lo, J. E. Snyder, A. P. Ring, L. L. Jones and D. C. Jiles, IEEE Trans. Magn. 39 (2003) 3316. 19. K. C. Maxwell, Electricity and Magnetism (Oxford University Press, London, 1873), p. 328. 20. K. W. Wagner, Am. J. Phys. 40 (1913) 817. 21. C. G. Koops, Phys. Rev. 83 (1953) 121. 22. K. Katsmi, S. Mamoru, I. Tatrsuo and I. Katsuya, Bull. Chem. Soc. 48 (1975) 1764. 23. P. A. Jadhav, R. S. Devan, Y. D. Kolekar and B. K. Chougule, J. Phys. Chem. Solids 70 (2009) 396. 24. L. T. Robinkin and Z. I. Novikova, Ferrites (IZV Acad. Nauk USSR, Minsik, 1960), p. 146. 25. K. Iwauchi and Y. Ikeda, Phys. Status Solidi A 93 (1986) 309. 26. T. T. Ahmed, I. Z. Rahman and M. A. Rahman, J. Mater. Process. Technol. 52 (2004) 797. 27. Q. H. Yang, H. W. Zhang, Y. L. Liu and Q. Y. Wen, Mater. Lett. 62 (2008) 2647.


222

S0217984911025626

M. P. Reddy et al.

28. A. Verma and D. C. Dube, J. Am. Ceram. Soc. 88 (2005) 519. 29. M. L. S. Teo, L. B. Kong, Z. W. Li, G. Q. Lin and Y. B. Gan, J. Alloys Compd. 459 (2008) 567. 30. C. Y. Tsay, K. S. Liu and I. N. Lin, J. Eur. Ceram. Soc. 24 (2004) 1057.