Technical Information High performance NiZn Ferrite

High performance NiZn Ferrite
Y. Matsuo, M. Inagaki, T.Tomozawa and F. Nakao

Abstract--High-performance NiZn ferrite was developed for use in mini DC-DC converters and inductors. In the development process, attention was focused on a small amount of MnO2 contained in NiZn ferrite. The quantity of MnO2 was examined as to its effects on the crystal structure and magnetic properties of NiZn ferrite cores. It was confirmed that core loss can be reduced by optimizing the MnO2 content. This was attributed to the decrease of lattice strain in the crystal structure of the core by a change in MnO2 content, which causes a chain reaction of a coercive force decline and then a hysteresis loss decline. As a result, we were able to develop a topclass low-loss NiZn ferrite.

Index Terms-- Lattice strain, magnetic properties, MnO2 additive, NiZn ferrite.

I. Introduction

THE downsizing of various electronic equipment has been possible due to the improved performances of transformers, inductors, choke coils and other electromagnetic components. Yet further miniaturization of components is in demand by the debut of super lightweight equipment including mobile computers. As a promising method of downsizing, efforts are being made to package coil-containing electronic circuits into a module functioning like a component and easily mountable on electronic equipment. To realize such modularization, miniature DC-DC converters and inductors must be developed. Currently, high-frequency MnZn ferrite is used in most cases to develop mini DC-DC converters and inductors[1]. However, NiZn ferrite offers better miniaturization prospects because its high electric resistance exceeding 10⁷Ωm enables NiZn ferrite cores to be wound directly by coils[2]-[5]. In the present study, therefore, we examined the effects of MnO2 doping on the crystal structure and magnetic properties of NiZn ferrite cores.

II. SAMPLE PREPARATION AND MEASUREMENT

A. Experimental Procedure
Fig. 1 shows our experimental procedure in a flow chart. First, high-purity NiO, CuO, ZnO, Fe2O3 and MnO2 powders were weighed to five different formulations shown in Table I, and were wet-mixed in planetary mills. The mixtures, after drying, were respectively calcinated at 850oC in atmosphere. Then, the calcinated products were milled in planetary mills. Following the addition of PVA, the milled powders were granulated and were pressed into toroidal shapes, which were finally sintered at around 1090oC in atmosphere. The resultant toroidal cores had 25 mm outer diameter, 15 mm inner diameter, and 5 mm thickness. The core loss values of the toroids were measured by a B-H analyzer SY-8232 of Iwatsu made at operating temperatures from room temperature to 120oC. The Curie temperatures of the cores were determined from inductance fading temperature, using an impedance analyzer HP 4194A connected with an oven. Core density values were measured by the Archimedean method.

Fig. 1

B. Phase identification and lattice constant measurement
The X-ray diffraction (XRD) peaks of core samples were measured with X-ray powder diffractmeter RINT 2500 (Rigaku make, high resolution parallel beam optical system, Cu Kα). The measurement conditions were set at a 50 kV tube voltage, 250 mA tube current, and fixed time scan mode so that the maximum peak would exceed 10000 counts. As shown in Fig. 2, the high resolution parallel beam optical system differs from the Bragg-Brentano system in that the former allows the discarding of an absorption effect (error due to the intrusion and diffraction of X-rays at the inner depth of a core), horizontal divergence effect (error due to the absence on the core surface of the curvatures of concentrated circles), and eccentric effect (angle error due to the deviation of the core surface from the center of the goniometer). In the measurement of lattice constants, NIST640b Si powder was employed as an external standard sample to perform angle corrections. Further, zero point corrections were conducted on vertical divergence effects and optical adjustment defects by the external standard method.

Fig. 2

III. RESULTS AND DISCUSSION

As shown in Table 1, the mixing ratios of Fe2O3 (49.5 mol%), ZnO (31.0 mol%) and CuO (5.0 mol%) were kept constant, while the mixing ratio of NiO was adjusted to suit changes in the quantity of MnO2 additive. All the five core sample groups (A) to (E) were found to be single phase NiZn ferrite. Fig. 3 shows the XRD pattern of a sample. The lattice constant of each sample was determined from its XRD pattern. As higher angles provided greater sensitivity, diffraction planes (731), (751) and (844) at higher angles were selected for the calculation of a lattice constant.

Fig. 3

As a result, we confirmed a tendency of the lattice constant went up with the increase in MnO2 content as shown in Fig. 4. This was contrary to our initial prediction. Merely assuming the substitution of Ni2+ ions having a radius of 0.72Å with Mn4+ irons having a radius of 0.54Å, the lattice constant should decrease.
Fig. 4

Similarly, the lattice constant should decrease, even assuming the substitution of Ni2+ ions with other metallic ions (i.e., Fe3+:0.64Å, Zn2+:0.74Å, Cu2+:0.6-0.9Åin penta-coordination ionic radius). Since our measurements indicated an increase in the lattice constant, we may surmise that Mn4+ ions intruded into the interstitials instead of substituting for metallic ions at their ion sites, so that the inter-ion distances were widened. We plan to carry out additional studies on this hypothesis.
Relation between MnO2 content and core loss (50 kHz, 150 mT, 80oC) is presented in Fig. 5. Core loss declined with the increase in MnO2 content. To determine factors behind this phenomenon, relation between coercive force and MnO2 content was examined. As shown in Fig. 6, it was found that coercive force reached its minimum with an MnO2 content of 0.6 mol% for both groups, one measured at room temperature and the other at an 80oC core operating temperature.
Fig. 5

On the other hand, core density proved to reach its maximum with an MnO2 content of 0.6 mol% as shown Fig.7. From these results, we attributed the core loss decline to a decrease in crystal strain and stress under an increased MnO2 content. As shown in Equation(1)[6] below, initial permeability is known to decline with an increase in internal stress or in coercive force.

Initial permeability is believed to decline when domain wall motion is hindered by internal stress thereby causing coercive force to increase. In other words it is believed that, by increasing the lattice constant value, the strain or stress in the spinel NiZn ferrite is relaxed and result in a decline in coercive force as shown in Fig. 6[7]. This will in turn reduce hysteresis loss, thus, lead to a decrease in core loss. Fig. 8 shows the relation between Curie temperature and lattice constant. It was found that the Curie temperature of the sample cores changed by 20oC through the variation of their MnO2 content from 0 to 0.9 mol%[8]. It was also found that the Curie temperature took a sharp plunge as the lattice constant was increased to around 8.4095Å. This was attributed to the moderation of inter-ion exchange interactions by an increase in the lattice constant, which in turn reduces the resistance of magnetic ions to the crystal lattice vibrations induced by heat.

Fig. 7

In addition to the above observation, we attempted to optimize the core sintering conditions. As shown in Fig. 9, we succeeded in developing a low-loss NiZn ferrite, which provided a core loss of as low as 200 kW/m3 at an operating temperature of 80oC. The core loss of this NiZn ferrite was about 40% lower than that of conventional NiZn ferrite throughout the operating temperature range under the measurement conditions of 50 kHz and 150 mT.
Fig. 9

IV. CONCLUSIONS The amount of MnO2 additive contained in NiZn ferrite was studied as to its effects on the crystal structure and magnetic properties of NiZn ferrite cores, and the following conclusions were achieved. 1. By optimizing the MnO2 content, the lattice constant of the cores was increased and the core loss was decreased. 2. This was attributed to the relaxation of strains in the spinel crystal ion orientation as a result of an increase in the lattice constant. 3. Consequently a topclass low-loss NiZn ferrite was developed, yielding a core loss as low as 200 kW/m3 measured at 50 kHz, 150 mT and 80oC.

ACKNOWLEDGEMENT

We express our utmost gratitude to Mr. Fujinawa and Mr. Taniguchi of Rigaku Company for their valuable advice on achieving higher precision in X-ray powder diffraction and lattice constant measurement.

REFERENCES
[1] Y. Matsuo et al., "Decreasing core loss of Mn-Zn ferrite", J. Magn. Soc. Jpn.,Vol.20, No.2, 429 (1996).
[2] T. Nomura, A. Nakano, "New evolution of ferrite for multiplayer chip components", ICF-6 Tokyo, Japan 1198 (1992).
[3] H. Momoi, A. Nakano et al., "Nano-structure control of NiCuZn ferrites for multiplayer chip components", ICF-6 Tokyo, Japan 1202 (1992).
[4] T. Araki, H.Morinaga et al, "Low loss Ni-Zn-Cu ferrite for deflection yoke", ICF-6 Tokyo, Japan 1185 (1992).
[5] R. Lebourgeois, P. Perriat and M.Labeyrie, "High and low level frequency losses in NiZn and MnZn spinel ferrites", ICF-6 Tokyo, Japan 1159 (1992).
[6] Y. Matsuo, K. Ono and M. Ishikura, "Dependence of Mn-Zn ferrite properties on the Particle size of MoO3 additives", J. Magn. Soc. Jpn., Vol.23, No.4-2, 1413 (1999).
[7] K. Kondo, T. Chiba, E. Otsuki et al, "POWER LOSS ANALYSIS IN Ni-Zn FERRITES OF DIFFERENT CHEMICAL COMPOSITIONS", Digests of the ICF 8 Kyoto, Japan, 18PpI-28, 66 (2000).
[8] A.A. Sattar and A.M. Samy, "EFFECT Sm CONCENTRATION THE MAGNETIC PROPERTIES OF Cu-Zn FERRITE", Digests of the ICF 8 Kyoto, Japan, 19AaI-6, 86 (2000).

Close

[FDK Home Page]