- Email: [email protected]
Effects of Li2CO3 addition on the microstructure and magnetic properties of low-temperature-fired NiCuZn ferrites
Ceramics International 42 (2016) 14609–14613
Contents lists available at ScienceDirect
Ceramics International journal homepage: www.elsevier.com/locate/ceramint
Effects of Li2CO3 addition on the microstructure and magnetic properties of low-temperature-fired NiCuZn ferrites Wengting Yang a, Xiaoli Tang a,n, Huaiwu Zhang a, Hua Su a,b,n a b
State Key Laboratory of Electronic Thin Films and Integrated Devices, University of Electronic Science and Technology of China, Chengdu 610054, China Institute of Electronic and Information Engineering, University of Electronic Science and Technology of China, Dongguan 518105, China
art ic l e i nf o
a b s t r a c t
Article history: Received 27 April 2016 Received in revised form 9 June 2016 Accepted 11 June 2016 Available online 13 June 2016
NiCuZn ferrites doped with 0.5 wt% Bi2O3 and different Li2CO3 contents (0–0.25 wt%) were sintered at 900 °C. The microstructure and magnetic properties of these materials were investigated. The addition of low-melting-point Li2CO3 led to large and uniform grains. However, excess Li2CO3 addition produced abnormal grains and many closed pores, thereby reducing density. Permeability initially increased and then decreased at the Li2CO3 content of 40.2 wt%. Maximum magnetic flux density (431.1 mT at room temperature, 339.6 mT at 100 °C) and minimum power loss were achieved at 0.2 wt% Li2CO3. These findings suggested the suitability of 0.2 wt% Li2CO3 for applications in low-temperature co-fired ceramic magnetic power components and modules. & 2016 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Keywords: LTCC Ferrite Magnetic flux density Power loss
1. Introduction Miniaturised, lightweight, integrated and multifunctional electronic devices are rapidly being developed. Portable devices are continuously fabricated with multifunctional and high-frequency applications [1]. Thus, the integration of low-temperature co-fired ceramic (LTCC) with inductors, capacitors and resistors, among others, into one monolithic chip component will offer enhanced performance and save space on the printed circuit board during assembly [2–5]. Low-temperature-fired NiCuZn ferrites have been widely used in manufacturing LTCC magnetic components, such as inductors and transformers, because of their superior magnetic properties. Ag has been chosen as the material for the internal conductor of multilayer chip magnetic components because of its low resistivity; thus, high-quality components are produced [6,7]. The interfacial reaction between Ag and ferrites can be inhibited by co-firing at a temperature lower than the melting point of Ag. Several methods such as ion exchange, fine powder and flux have been introduced to lower the sintering temperature of NiCuZn ferrites [8–15]. An increase in power density is important to further improve the integration of LTCC components and modules. However, few studies have investigated the power characteristics of low-temperature-fired NiCuZn ferrites to n
Corresponding authors. E-mail addresses: [email protected] (X. Tang), [email protected] (H. Su).
http://dx.doi.org/10.1016/j.ceramint.2016.06.078 0272-8842/& 2016 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
produce LTCC power magnetic elements and modules. In this study, we mainly analysed the influences of Li2CO3 addition on the microstructure and magnetic properties, especially power-related properties, such as magnetic flux density (Bs) and power loss ( Pcv ), of low-temperature-fired NiCuZn ferrites.
2. Experiment procedures NiCuZn ferrite samples (Co0.01Ni0.32Zn0.49Cu0.18Fe1.98O4) were prepared by a solid-state reaction method. Analytical grade Fe2O3, NiO, CuO, ZnO and Co2O3 were weighed following the required composition and then mixed for 4 h. The mixed oxide powders were calcined at 780 °C in air for 2 h. Different Li2CO3 contents (0, 0.05, 0.10, 0.15, 0.20 and 0.25 wt%) and 0.5 wt% Bi2O3 were added to the calcined powder, and the mixtures were further milled in ball mills. The resulting powders were granulated with 8% polyvinyl alcohol after they were dried. The powders were then pressed into toroidal shapes with the size of 18 mm (outside diameter) 8 mm (inside diameter) 4 mm (height), which were finally sintered at 900 °C with the heating rate of 2 °C/min and held for 3 h in air. To test permeability values of the samples, we first measured the outside diameter (R), inside diameter (r) and height (h) values of the sintered samples by vernier caliper. Then the samples were winded 20 turns wires respectively and measured by a precision LCR mete(Agilent 4284A) at the frequency of 10 kHz. Permeability
14610
W. Yang et al. / Ceramics International 42 (2016) 14609–14613
values could be calculated according to the following formula:
μ=
L × 104 2
2N ⋅h⋅ln ( R/r
)
in which the unit of R, r and h was mm, N was the wire turns, L (μH) was inductance values of the sample, which were directly measured by the LCR meter. Bs and Pcv were measured using a B-H analyser (IWATSU, SY-8232) at room temperature and 100 °C. The density of the sintered samples was calculated as the mass/volume ratio. The morphology of the samples was investigated using a scanning electron microscope (SEM, JSM-6490LV). X-ray diffractograms of the samples were recorded using an X-ray diffractometer with Cu Κα radiation.
3. Results and discussion Fig. 1 shows the typical SEM micrographs of the samples with different Li2CO3 contents. The sample without Li2CO3 presented a microstructure with some large grains and very small grains. The ratio of large grains gradually increased with the Li2CO3 content. The sample with 0.2 wt% Li2CO3 presented a microstructure with uniform and large grains, which were favourable to improve permeability and density. Subsequently, some large grains and many closed pores appeared, which might negatively affect the magnetic properties of the sample. The variation in microstructure was due to the ability of low-melting-point Li2CO3 ( 723 °C) to form a liquid phase during sintering, which further enhanced mass transfer and sintering by solid-state solubilisation and segregation [16–18]. However, an excess liquid phase for the sample with 0.25 wt% Li2CO3 might lead to the rapid growth of grains, resulting in
Fig. 1. Scanning electron micrographs of the samples with different Li2CO3 contents.
W. Yang et al. / Ceramics International 42 (2016) 14609–14613
abnormally large grains and closed pores. Thus, the addition of 0.2 wt% Li2CO3 was optimal to obtain large and uniform grains. Fig. 2 shows the X-ray diffractograms of the sintered NiCuZn samples doped with different Li2CO3 contents. All samples displayed the spinel ferrite phase, and no secondary phase was detected. This phenomenon indicated that Bi2O3 and Li2CO3 additives were completely soluted in the spinel structure within our doping range, or the content of the other phase was too small for XRD detection. Fig. 3 presents the initial permeabilities ( μi ) and sintered densities of the samples with different Li2CO3 contents. The permeability initially increased with the Li2CO3 content and then peaked with 0.20 wt% Li2CO3. Thereafter, it decreased with increasing Li2CO3 content. Density presented a similar trend but without any remarkable changes. These findings were consistent with the microstructural variations. The first increase in density was mainly due to the promotion of grain growth caused by Li2CO3 addition, and the final decrease in density was due to the appearance of very large grains and closed pores. The variation in permeability was influenced by both sintered density and microstructure. The initial increase in density and average grain size led to an obvious increase in permeability, whereas the final decrease in density and abnormal grains with closed pores led to decreased permeability. Thus, 0.2 wt% Li2CO3 was the optimal doping content to obtain high permeability and high sintered density. Fig. 4 shows the magnetic flux densities of NiCuZn ferrites with different Li2CO3 contents at room temperature and 100 °C. Bs initially increased and peaked with 0.20 wt% Li2CO3 (431.1 mT at
Fig. 2. X-ray diffractograms of the samples with different Li2CO3 contents.
Fig. 3. Variations in the permeabilities and densities of samples with different Li2CO3 contents.
14611
Fig. 4. Magnetic flux densities of the samples with different Li2CO3 contents.
room temperature, 339.6 mT at 100 °C). Subsequently, Bs slightly decreased with 0.25 wt% Li2CO3. This variation was also very similar to that of sintered density. Bs of ferrites was mainly determined by composition and sintered density. In this study, all samples had the same compositions. A slight difference in Li2CO3 addition slightly influenced the actual composition of the ferrites because some Li þ might have entered the crystal lattice. However, this influence could be ignored. Thus, Bs was mainly determined by sintered density. The sample with 0.20 wt% Li2CO3 obtained the highest magnetic flux density, which was highly favourable in power magnetic components. The Bs values at 100 °C were much lower than those at room temperature for all samples. This phenomenon was due to the fact that thermal agitation intensified with high temperature, which partially destroyed the reverse parallel arrangement of A and B molecular moments and decreased the sum of molecular moments [19]. Fig. 5 shows the frequency dependence of Pcv (under 50 mT) for the samples with different Li2CO3 contents. Pcv increased with frequency for all samples but initially decreased with the Li2CO3 content. The sample with 0.2 wt% Li2CO3 exhibited the lowest Pcv . Subsequently, Pcv increased when the Li2CO3 content reached 0.25 wt%. The trends at room temperature and 100 °C were the same. Pcv can be divided into hysteresis ( Pe ), eddy current loss ( Pe ) and residual loss ( Pr ) [20–24]. In this study, Pe was negligible because of the high resistivity of the NiCuZn ferrites. Pr was mainly determined by the Fe2 þ content in ferrites, and it was proportional to the square of the frequency. The frequency was tested within 20–120 kHz; thus, the ratio of Pr was small. Furthermore, a slight difference in Li2CO3 addition did not lead to a significant change in the Fe2 þ ions. Hence, the variation in Pcv was primarily caused by the variation in Ph . In this study, proper Li2CO3 addition could lead to more uniform and large grains with increased density. These characteristics facilitated domain wall movement and magnetic domain rotation, which resulted in decreased Ph and Pcv . However, a dual microstructure with both small and extremely large grains with closed pores appeared with increasing Li2CO3 doping. The presence of such grains hindered domain wall motion. Consequently, both Ph and Pcv increased. Meanwhile, the Pcv values were obviously reduced at 100 °C compared with that at room temperature, which indicated that the samples had negative temperature coefficients on Pcv. This characteristic was also beneficial for high power applications. Fig. 6 presents the relationship between the exciting flux density and Pcv of the samples at 100 kHz. Pcv increased obviously with exciting flux density for all samples. Under relatively low flux density (10 and 20 mT) conditions, Pcv first decreased, and then gradually increased with increasing Li2CO3 content. The sample with 0.15 wt% Li2CO3 obtained the lowest Pcv . However, when the
14612
W. Yang et al. / Ceramics International 42 (2016) 14609–14613
Fig. 5. Power losses of the samples with different Li2CO3 contents at (a) room temperature and (b) 100 °C.
exciting flux density exceeded 30 mT, the sample with 0.2 wt% Li2CO3 obtained the lowest Pcv . It is known that the predominant factor in Pcv is Ph . Big grain size led to lower Ph due to the decrease in the domain wall pinning factor grain boundary. However, the closed pores in big grain size also acted as domain wall pinning factor and increased Ph of samples. When exciting flux density was lower than 30 mT, the closed pores reduced the distance between pinned edges and easily block domain wall movement. So the microstructure with big grain size and closed pores led to a higher Pcv on the contrary. As a result, the sample with relatively high density (less open pores) and less closed pores presented the lowest Pcv. When the exciting flux density was larger (the critical flux density was around 30 mT), the closed pores were not easy to block domain wall movement and grain boundaries became the predominant domain wall pinning factor. Thus the sample with 0.2 wt% Li2CO3, which presented the biggest average gain size and lowest grain boundaries, obtained the lowest Pcv . All in all, the sample with 0.2 wt% Li2CO3 doping could obtain the highest permeability and Bs, and high performance on Pcv characteristics, especially under high exciting flux density conditions. It was very suitable for applications in LTCC power components and modules.
4. Conclusions We investigated the effects of Li2CO3 addition on the microstructure and magnetic properties of low-temperature-fired NiCuZn ferrites. Appropriate Li2CO3 doping promoted uniform and large grains, increased sintered density and Bs and decreased Pcv . However, a further increase in Li2CO3 doping led to contrasting effects. The sample with 0.20 wt% Li2CO3 exhibited optimal
Fig. 6. Relationship between the exciting flux density and power losses of the samples at 100 kHz and room temperature. (a) Bm ¼10 mT and 20 mT; (b) Bm ¼ 30 mT, 40 mT and 50 mT.
performance with relatively high permeability (172), high Bs (431.1 mT at room temperature, 339.6 mT at 100 °C) and low power loss. This material can be applied in LTCC power magnetic components and modules.
Acknowledgements This work was supported by the National Natural Science Foundation of China under Grant nos. 51372031, 61471096, National High-tech R&D Program of China under Grant no. 2015AA034102, Special Support Program of Guangdong Province under Grant no. 2014TX01C042 and Department of Science and Technology of Sichuan Province 2014GZ0015, 2015GZ0227.
References [1] L.Z. Li, Z.W. Lan, Z. Yu, K. Sun, Z.Y. Xu, H.N. Ji, Microstructure, electrical, and magnetic properties of ZrO2 added MnZn ferrites, IEEE Trans. Magn. 44 (2008) 2107–2112. [2] T. Rabe, H. Naghib-Zadeh, C. Glitzky, J. Topfer, Integration of Ni-Cu-Zn Ferrite in low temperature co-fired ceramics (LTCC) modules, Int. J. Appl. Ceram. Technol. 9 (2012) 18–28. [3] S.J. Feng, J. Li, S.G. Huang, X.S. Liu, Z.Y. Zhong, Magnetic hysteresis loss crossover in Ni0.4Zn0.6Fe1.95Ti0.05O4 ferrite, J. Alloy. Compd. 660 (2016) 398–401. [4] H. Su, X.L. Tang, H.W. Zhang, Y.L. Jing, Z.Y. Zhong, Low-temperature-fired NiCuZn ferrites with BBSZ glass, J. Magn. Magn. Mater. 323 (2011) 592–595. [5] K. Sun, Z.W. Lan, Z. Yu, L.Z. Li, J.M. Huang, X.N. Zhao, Effects of SnO2 addition on the microstructure and magnetic properties of NiZn ferrites, J. Magn. Magn. Mater. 320 (2008) 3352–3355. [6] H. Su, X. Tang, H. Zhang, Y. Jing, F. Bai, Low-loss magneto-dielectric materials: approaches and developments, J. Electron. Mater. 43 (2013) 299–307.
W. Yang et al. / Ceramics International 42 (2016) 14609–14613
[7] T. Nakamura, Low-temperature sintering of Ni-Zn-Cu ferrite and its permeability spectra, J. Magn. Magn. Mater. 168 (1997) 285–291. [8] W. Shen, H.K. Zhu, Y. Jin, H.Q. Zhou, Z.M. Xu, Sintering, microstructure and magnetic properties of low-temperature-fired NiCuZn ferrites doped with B2O3, Ceram. Int. 40 (2014) 9205–9209. [9] T. Krishnaveni, B.R. Kanth, V.S.R. Raju, S.R. Murthy, Fabrication of multilayer chip inductors using Ni-Cu-Zn ferrites, J. Alloy. Compd. 414 (2006) 282–286. [10] W. Shen, H.K. Zhu, H.Q. Zhou, Z.M. Xu, Y. Jin, Effects of borosilicate glass additions on microstructures and magnetic properties of low temperature cofired NiCuZn ferrites, J. Mater. Sci. – Mater. Electron. 24 (2013) 4713–4717. [11] H. Su, X.L. Tang, H.W. Zhang, Z.Y. Zhong, J. Shen, Sintering dense NiZn ferrite by two-step sintering process, J. Appl. Phys. 109 (2011) 3. [12] J.L. Ma, Z.F. Fu, P. Liu, B. Wang, Y. Li, Microwave dielectric properties of lowfired Li2TiO3-MgO ceramics for LTCC applications, Mater. Sci. Eng. B – Adv. Funct. Solid-State Mater. 204 (2016) 15–19. [13] A. Bhaskar, B.R. Kanth, S.R. Murthy, Electrical properties of Mn added MgCuZn ferrites prepared by microwave sintering method, J. Magn. Magn. Mater. 283 (2004) 109–116. [14] S.S. Kamble, V.S. Jagtap, P.C. Pingale, Synthesis of Mg0.48Cu0.12Zn0.40Fe2O4 ferrite and its aptness for multilayer chip component application, Ceram. Int. 39 (2013) 3597–3601. [15] E. Rezlescu, L. Sachelarie, P.D. Popa, N. Rezlescu, Effect of substitution of divalent ions on the electrical and magnetic properties of Ni-Zn-Me ferrites, IEEE Trans. Magn. 36 (2000) 3962–3967.
14613
[16] H.K. Zhu, Y. Jin, H. Zhu, W. Shen, Y.Q. Xu, H.Q. Zhou, Sintering, microstructures and magnetic properties of low temperature co-fired NiCuZn ferrites with B2O3 and WO3 additions, Mater. Res. Bull. 61 (2015) 32–35. [17] H. Su, H.W. Zhang, X.L. Tang, X.Y. Xiang, High-permeability and high-Curie temperature NiCuZn ferrite, J. Magn. Magn. Mater. 283 (2004) 157–163. [18] K. Sun, J.T. Wang, Y. Yang, Y. Li, Z. Yu, Z.W. Lan, X.N. Jiang, R.D. Guo, C.J. Wu, Influence of Ta2O5-Co2O23 co-doping on the magnetic property of NiMgCuZn ferrites, Physica B 476 (2015) 122–128. [19] H.N. Ji, Z.W. Lan, Z.Y. Xu, H.W. Zhang, J.X. Yu, M.Q. Li, Effects of second milling time on temperature dependence and improved steinmetz parameters of low loss MnZn power ferrites, IEEE Trans. Appl. Supercond. 24 (2014) 4. [20] H. Su, H.W. Zhang, X.L. Tang, Z.Y. Zhong, Y.L. Jing, Analysis of low-temperaturefired NiCuZn ferrites for power applications, Mater. Sci. Eng. B – Adv. Funct. Solid-State Mater. 162 (2009) 22–25. [21] S. Yamada, E. Otsuki, Analysis of eddy current loss in Mn–Zn ferrites for power supplies, J. Appl. Phys. 81 (1997) 4791–4793. [22] A. Znidarsic, M. Drofenik, Influence of oxygen partial pressure during sintering on the power loss of MnZn ferrites, IEEE Trans. Magn. 32 (1996) 1941–1945. [23] Hua Su, Huaiwu Zhang, Xiaoli Tang, Yulan Jing, Influence of microstructure on permeability dispersion and power loss of NiZn ferrite, J. Appl. Phys. 103 (2008) 093903. [24] X.L. Tang, H. Su, H.W. Zhang, Tailoring the microstructure of NiZn ferrite for power field applications, Chin. Phys. Lett. 29 (2012) 3.
Contents lists available at ScienceDirect
Ceramics International journal homepage: www.elsevier.com/locate/ceramint
Effects of Li2CO3 addition on the microstructure and magnetic properties of low-temperature-fired NiCuZn ferrites Wengting Yang a, Xiaoli Tang a,n, Huaiwu Zhang a, Hua Su a,b,n a b
State Key Laboratory of Electronic Thin Films and Integrated Devices, University of Electronic Science and Technology of China, Chengdu 610054, China Institute of Electronic and Information Engineering, University of Electronic Science and Technology of China, Dongguan 518105, China
art ic l e i nf o
a b s t r a c t
Article history: Received 27 April 2016 Received in revised form 9 June 2016 Accepted 11 June 2016 Available online 13 June 2016
NiCuZn ferrites doped with 0.5 wt% Bi2O3 and different Li2CO3 contents (0–0.25 wt%) were sintered at 900 °C. The microstructure and magnetic properties of these materials were investigated. The addition of low-melting-point Li2CO3 led to large and uniform grains. However, excess Li2CO3 addition produced abnormal grains and many closed pores, thereby reducing density. Permeability initially increased and then decreased at the Li2CO3 content of 40.2 wt%. Maximum magnetic flux density (431.1 mT at room temperature, 339.6 mT at 100 °C) and minimum power loss were achieved at 0.2 wt% Li2CO3. These findings suggested the suitability of 0.2 wt% Li2CO3 for applications in low-temperature co-fired ceramic magnetic power components and modules. & 2016 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Keywords: LTCC Ferrite Magnetic flux density Power loss
1. Introduction Miniaturised, lightweight, integrated and multifunctional electronic devices are rapidly being developed. Portable devices are continuously fabricated with multifunctional and high-frequency applications [1]. Thus, the integration of low-temperature co-fired ceramic (LTCC) with inductors, capacitors and resistors, among others, into one monolithic chip component will offer enhanced performance and save space on the printed circuit board during assembly [2–5]. Low-temperature-fired NiCuZn ferrites have been widely used in manufacturing LTCC magnetic components, such as inductors and transformers, because of their superior magnetic properties. Ag has been chosen as the material for the internal conductor of multilayer chip magnetic components because of its low resistivity; thus, high-quality components are produced [6,7]. The interfacial reaction between Ag and ferrites can be inhibited by co-firing at a temperature lower than the melting point of Ag. Several methods such as ion exchange, fine powder and flux have been introduced to lower the sintering temperature of NiCuZn ferrites [8–15]. An increase in power density is important to further improve the integration of LTCC components and modules. However, few studies have investigated the power characteristics of low-temperature-fired NiCuZn ferrites to n
Corresponding authors. E-mail addresses: [email protected] (X. Tang), [email protected] (H. Su).
http://dx.doi.org/10.1016/j.ceramint.2016.06.078 0272-8842/& 2016 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
produce LTCC power magnetic elements and modules. In this study, we mainly analysed the influences of Li2CO3 addition on the microstructure and magnetic properties, especially power-related properties, such as magnetic flux density (Bs) and power loss ( Pcv ), of low-temperature-fired NiCuZn ferrites.
2. Experiment procedures NiCuZn ferrite samples (Co0.01Ni0.32Zn0.49Cu0.18Fe1.98O4) were prepared by a solid-state reaction method. Analytical grade Fe2O3, NiO, CuO, ZnO and Co2O3 were weighed following the required composition and then mixed for 4 h. The mixed oxide powders were calcined at 780 °C in air for 2 h. Different Li2CO3 contents (0, 0.05, 0.10, 0.15, 0.20 and 0.25 wt%) and 0.5 wt% Bi2O3 were added to the calcined powder, and the mixtures were further milled in ball mills. The resulting powders were granulated with 8% polyvinyl alcohol after they were dried. The powders were then pressed into toroidal shapes with the size of 18 mm (outside diameter) 8 mm (inside diameter) 4 mm (height), which were finally sintered at 900 °C with the heating rate of 2 °C/min and held for 3 h in air. To test permeability values of the samples, we first measured the outside diameter (R), inside diameter (r) and height (h) values of the sintered samples by vernier caliper. Then the samples were winded 20 turns wires respectively and measured by a precision LCR mete(Agilent 4284A) at the frequency of 10 kHz. Permeability
14610
W. Yang et al. / Ceramics International 42 (2016) 14609–14613
values could be calculated according to the following formula:
μ=
L × 104 2
2N ⋅h⋅ln ( R/r
)
in which the unit of R, r and h was mm, N was the wire turns, L (μH) was inductance values of the sample, which were directly measured by the LCR meter. Bs and Pcv were measured using a B-H analyser (IWATSU, SY-8232) at room temperature and 100 °C. The density of the sintered samples was calculated as the mass/volume ratio. The morphology of the samples was investigated using a scanning electron microscope (SEM, JSM-6490LV). X-ray diffractograms of the samples were recorded using an X-ray diffractometer with Cu Κα radiation.
3. Results and discussion Fig. 1 shows the typical SEM micrographs of the samples with different Li2CO3 contents. The sample without Li2CO3 presented a microstructure with some large grains and very small grains. The ratio of large grains gradually increased with the Li2CO3 content. The sample with 0.2 wt% Li2CO3 presented a microstructure with uniform and large grains, which were favourable to improve permeability and density. Subsequently, some large grains and many closed pores appeared, which might negatively affect the magnetic properties of the sample. The variation in microstructure was due to the ability of low-melting-point Li2CO3 ( 723 °C) to form a liquid phase during sintering, which further enhanced mass transfer and sintering by solid-state solubilisation and segregation [16–18]. However, an excess liquid phase for the sample with 0.25 wt% Li2CO3 might lead to the rapid growth of grains, resulting in
Fig. 1. Scanning electron micrographs of the samples with different Li2CO3 contents.
W. Yang et al. / Ceramics International 42 (2016) 14609–14613
abnormally large grains and closed pores. Thus, the addition of 0.2 wt% Li2CO3 was optimal to obtain large and uniform grains. Fig. 2 shows the X-ray diffractograms of the sintered NiCuZn samples doped with different Li2CO3 contents. All samples displayed the spinel ferrite phase, and no secondary phase was detected. This phenomenon indicated that Bi2O3 and Li2CO3 additives were completely soluted in the spinel structure within our doping range, or the content of the other phase was too small for XRD detection. Fig. 3 presents the initial permeabilities ( μi ) and sintered densities of the samples with different Li2CO3 contents. The permeability initially increased with the Li2CO3 content and then peaked with 0.20 wt% Li2CO3. Thereafter, it decreased with increasing Li2CO3 content. Density presented a similar trend but without any remarkable changes. These findings were consistent with the microstructural variations. The first increase in density was mainly due to the promotion of grain growth caused by Li2CO3 addition, and the final decrease in density was due to the appearance of very large grains and closed pores. The variation in permeability was influenced by both sintered density and microstructure. The initial increase in density and average grain size led to an obvious increase in permeability, whereas the final decrease in density and abnormal grains with closed pores led to decreased permeability. Thus, 0.2 wt% Li2CO3 was the optimal doping content to obtain high permeability and high sintered density. Fig. 4 shows the magnetic flux densities of NiCuZn ferrites with different Li2CO3 contents at room temperature and 100 °C. Bs initially increased and peaked with 0.20 wt% Li2CO3 (431.1 mT at
Fig. 2. X-ray diffractograms of the samples with different Li2CO3 contents.
Fig. 3. Variations in the permeabilities and densities of samples with different Li2CO3 contents.
14611
Fig. 4. Magnetic flux densities of the samples with different Li2CO3 contents.
room temperature, 339.6 mT at 100 °C). Subsequently, Bs slightly decreased with 0.25 wt% Li2CO3. This variation was also very similar to that of sintered density. Bs of ferrites was mainly determined by composition and sintered density. In this study, all samples had the same compositions. A slight difference in Li2CO3 addition slightly influenced the actual composition of the ferrites because some Li þ might have entered the crystal lattice. However, this influence could be ignored. Thus, Bs was mainly determined by sintered density. The sample with 0.20 wt% Li2CO3 obtained the highest magnetic flux density, which was highly favourable in power magnetic components. The Bs values at 100 °C were much lower than those at room temperature for all samples. This phenomenon was due to the fact that thermal agitation intensified with high temperature, which partially destroyed the reverse parallel arrangement of A and B molecular moments and decreased the sum of molecular moments [19]. Fig. 5 shows the frequency dependence of Pcv (under 50 mT) for the samples with different Li2CO3 contents. Pcv increased with frequency for all samples but initially decreased with the Li2CO3 content. The sample with 0.2 wt% Li2CO3 exhibited the lowest Pcv . Subsequently, Pcv increased when the Li2CO3 content reached 0.25 wt%. The trends at room temperature and 100 °C were the same. Pcv can be divided into hysteresis ( Pe ), eddy current loss ( Pe ) and residual loss ( Pr ) [20–24]. In this study, Pe was negligible because of the high resistivity of the NiCuZn ferrites. Pr was mainly determined by the Fe2 þ content in ferrites, and it was proportional to the square of the frequency. The frequency was tested within 20–120 kHz; thus, the ratio of Pr was small. Furthermore, a slight difference in Li2CO3 addition did not lead to a significant change in the Fe2 þ ions. Hence, the variation in Pcv was primarily caused by the variation in Ph . In this study, proper Li2CO3 addition could lead to more uniform and large grains with increased density. These characteristics facilitated domain wall movement and magnetic domain rotation, which resulted in decreased Ph and Pcv . However, a dual microstructure with both small and extremely large grains with closed pores appeared with increasing Li2CO3 doping. The presence of such grains hindered domain wall motion. Consequently, both Ph and Pcv increased. Meanwhile, the Pcv values were obviously reduced at 100 °C compared with that at room temperature, which indicated that the samples had negative temperature coefficients on Pcv. This characteristic was also beneficial for high power applications. Fig. 6 presents the relationship between the exciting flux density and Pcv of the samples at 100 kHz. Pcv increased obviously with exciting flux density for all samples. Under relatively low flux density (10 and 20 mT) conditions, Pcv first decreased, and then gradually increased with increasing Li2CO3 content. The sample with 0.15 wt% Li2CO3 obtained the lowest Pcv . However, when the
14612
W. Yang et al. / Ceramics International 42 (2016) 14609–14613
Fig. 5. Power losses of the samples with different Li2CO3 contents at (a) room temperature and (b) 100 °C.
exciting flux density exceeded 30 mT, the sample with 0.2 wt% Li2CO3 obtained the lowest Pcv . It is known that the predominant factor in Pcv is Ph . Big grain size led to lower Ph due to the decrease in the domain wall pinning factor grain boundary. However, the closed pores in big grain size also acted as domain wall pinning factor and increased Ph of samples. When exciting flux density was lower than 30 mT, the closed pores reduced the distance between pinned edges and easily block domain wall movement. So the microstructure with big grain size and closed pores led to a higher Pcv on the contrary. As a result, the sample with relatively high density (less open pores) and less closed pores presented the lowest Pcv. When the exciting flux density was larger (the critical flux density was around 30 mT), the closed pores were not easy to block domain wall movement and grain boundaries became the predominant domain wall pinning factor. Thus the sample with 0.2 wt% Li2CO3, which presented the biggest average gain size and lowest grain boundaries, obtained the lowest Pcv . All in all, the sample with 0.2 wt% Li2CO3 doping could obtain the highest permeability and Bs, and high performance on Pcv characteristics, especially under high exciting flux density conditions. It was very suitable for applications in LTCC power components and modules.
4. Conclusions We investigated the effects of Li2CO3 addition on the microstructure and magnetic properties of low-temperature-fired NiCuZn ferrites. Appropriate Li2CO3 doping promoted uniform and large grains, increased sintered density and Bs and decreased Pcv . However, a further increase in Li2CO3 doping led to contrasting effects. The sample with 0.20 wt% Li2CO3 exhibited optimal
Fig. 6. Relationship between the exciting flux density and power losses of the samples at 100 kHz and room temperature. (a) Bm ¼10 mT and 20 mT; (b) Bm ¼ 30 mT, 40 mT and 50 mT.
performance with relatively high permeability (172), high Bs (431.1 mT at room temperature, 339.6 mT at 100 °C) and low power loss. This material can be applied in LTCC power magnetic components and modules.
Acknowledgements This work was supported by the National Natural Science Foundation of China under Grant nos. 51372031, 61471096, National High-tech R&D Program of China under Grant no. 2015AA034102, Special Support Program of Guangdong Province under Grant no. 2014TX01C042 and Department of Science and Technology of Sichuan Province 2014GZ0015, 2015GZ0227.
References [1] L.Z. Li, Z.W. Lan, Z. Yu, K. Sun, Z.Y. Xu, H.N. Ji, Microstructure, electrical, and magnetic properties of ZrO2 added MnZn ferrites, IEEE Trans. Magn. 44 (2008) 2107–2112. [2] T. Rabe, H. Naghib-Zadeh, C. Glitzky, J. Topfer, Integration of Ni-Cu-Zn Ferrite in low temperature co-fired ceramics (LTCC) modules, Int. J. Appl. Ceram. Technol. 9 (2012) 18–28. [3] S.J. Feng, J. Li, S.G. Huang, X.S. Liu, Z.Y. Zhong, Magnetic hysteresis loss crossover in Ni0.4Zn0.6Fe1.95Ti0.05O4 ferrite, J. Alloy. Compd. 660 (2016) 398–401. [4] H. Su, X.L. Tang, H.W. Zhang, Y.L. Jing, Z.Y. Zhong, Low-temperature-fired NiCuZn ferrites with BBSZ glass, J. Magn. Magn. Mater. 323 (2011) 592–595. [5] K. Sun, Z.W. Lan, Z. Yu, L.Z. Li, J.M. Huang, X.N. Zhao, Effects of SnO2 addition on the microstructure and magnetic properties of NiZn ferrites, J. Magn. Magn. Mater. 320 (2008) 3352–3355. [6] H. Su, X. Tang, H. Zhang, Y. Jing, F. Bai, Low-loss magneto-dielectric materials: approaches and developments, J. Electron. Mater. 43 (2013) 299–307.
W. Yang et al. / Ceramics International 42 (2016) 14609–14613
[7] T. Nakamura, Low-temperature sintering of Ni-Zn-Cu ferrite and its permeability spectra, J. Magn. Magn. Mater. 168 (1997) 285–291. [8] W. Shen, H.K. Zhu, Y. Jin, H.Q. Zhou, Z.M. Xu, Sintering, microstructure and magnetic properties of low-temperature-fired NiCuZn ferrites doped with B2O3, Ceram. Int. 40 (2014) 9205–9209. [9] T. Krishnaveni, B.R. Kanth, V.S.R. Raju, S.R. Murthy, Fabrication of multilayer chip inductors using Ni-Cu-Zn ferrites, J. Alloy. Compd. 414 (2006) 282–286. [10] W. Shen, H.K. Zhu, H.Q. Zhou, Z.M. Xu, Y. Jin, Effects of borosilicate glass additions on microstructures and magnetic properties of low temperature cofired NiCuZn ferrites, J. Mater. Sci. – Mater. Electron. 24 (2013) 4713–4717. [11] H. Su, X.L. Tang, H.W. Zhang, Z.Y. Zhong, J. Shen, Sintering dense NiZn ferrite by two-step sintering process, J. Appl. Phys. 109 (2011) 3. [12] J.L. Ma, Z.F. Fu, P. Liu, B. Wang, Y. Li, Microwave dielectric properties of lowfired Li2TiO3-MgO ceramics for LTCC applications, Mater. Sci. Eng. B – Adv. Funct. Solid-State Mater. 204 (2016) 15–19. [13] A. Bhaskar, B.R. Kanth, S.R. Murthy, Electrical properties of Mn added MgCuZn ferrites prepared by microwave sintering method, J. Magn. Magn. Mater. 283 (2004) 109–116. [14] S.S. Kamble, V.S. Jagtap, P.C. Pingale, Synthesis of Mg0.48Cu0.12Zn0.40Fe2O4 ferrite and its aptness for multilayer chip component application, Ceram. Int. 39 (2013) 3597–3601. [15] E. Rezlescu, L. Sachelarie, P.D. Popa, N. Rezlescu, Effect of substitution of divalent ions on the electrical and magnetic properties of Ni-Zn-Me ferrites, IEEE Trans. Magn. 36 (2000) 3962–3967.
14613
[16] H.K. Zhu, Y. Jin, H. Zhu, W. Shen, Y.Q. Xu, H.Q. Zhou, Sintering, microstructures and magnetic properties of low temperature co-fired NiCuZn ferrites with B2O3 and WO3 additions, Mater. Res. Bull. 61 (2015) 32–35. [17] H. Su, H.W. Zhang, X.L. Tang, X.Y. Xiang, High-permeability and high-Curie temperature NiCuZn ferrite, J. Magn. Magn. Mater. 283 (2004) 157–163. [18] K. Sun, J.T. Wang, Y. Yang, Y. Li, Z. Yu, Z.W. Lan, X.N. Jiang, R.D. Guo, C.J. Wu, Influence of Ta2O5-Co2O23 co-doping on the magnetic property of NiMgCuZn ferrites, Physica B 476 (2015) 122–128. [19] H.N. Ji, Z.W. Lan, Z.Y. Xu, H.W. Zhang, J.X. Yu, M.Q. Li, Effects of second milling time on temperature dependence and improved steinmetz parameters of low loss MnZn power ferrites, IEEE Trans. Appl. Supercond. 24 (2014) 4. [20] H. Su, H.W. Zhang, X.L. Tang, Z.Y. Zhong, Y.L. Jing, Analysis of low-temperaturefired NiCuZn ferrites for power applications, Mater. Sci. Eng. B – Adv. Funct. Solid-State Mater. 162 (2009) 22–25. [21] S. Yamada, E. Otsuki, Analysis of eddy current loss in Mn–Zn ferrites for power supplies, J. Appl. Phys. 81 (1997) 4791–4793. [22] A. Znidarsic, M. Drofenik, Influence of oxygen partial pressure during sintering on the power loss of MnZn ferrites, IEEE Trans. Magn. 32 (1996) 1941–1945. [23] Hua Su, Huaiwu Zhang, Xiaoli Tang, Yulan Jing, Influence of microstructure on permeability dispersion and power loss of NiZn ferrite, J. Appl. Phys. 103 (2008) 093903. [24] X.L. Tang, H. Su, H.W. Zhang, Tailoring the microstructure of NiZn ferrite for power field applications, Chin. Phys. Lett. 29 (2012) 3.