Effect of Co₂Y additive on power loss of Ni_0.8Zn_0.2Fe₂O₄ ferrites

Abstract

Ba₂Co₂Fe₁₂O₂₂ (Co₂Y, 0 wt%, 2 wt%)-added Ni_0.8Zn_0.2Fe₂O₄ polycrystalline ferrites are prepared by solid-state reaction method, and the magnetic properties of ferrites are investigated. Magnetic measurements show that the magnetic loss is obviously decreased with Co₂Y addition. Loss separation reveals that both hysteresis loss and eddy current loss reduce significantly after Co₂Y addition. The mechanism of magnetic loss decreasing due to Co₂Y additive is discussed, and the decrease of magnetic hysteresis loss and eddy current loss caused by Co²⁺ cation doping was regarded as the main reason for the observed power loss decrease in ferrites and insulating grain boundary composed by barium compound which plays a role in decreasing eddy current loss. The Q value of Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ ferrite was much higher than that of Ni_0.8Zn_0.2Fe₂O₄ sample which can be used as inductors for high-frequency applications.

1 Introduction

Ni–Zn ferrites are important soft magnetic materials with numerous applications in electronics and communication technologies. High electrical resistivity, high cutoff frequency, broadband and low magnetic losses make Ni–Zn ferrite an obvious choice as a magnetic core material for high-frequency applications. In comparison, M-type ferrites with hexagonal structure are suitable as higher-frequency soft magnet ferrites [1,2,3,4,5,6], but they have a relatively large hysteresis loss, so Ni–Zn ferrites are suitable for application in the 1–200 MHz band. Different applications need different characteristics of ferrites, which can be achieved by doping of suitable cations in materials [5,6,7,8,9,10,11,12,13,14,15,16]. For example, manganese dioxide doped in nickel–zinc ferrites can decrease the magnetic loss in materials [17,18,19]. The nickel–zinc ferrites applied as inductance cores in radio frequency band-pass filters need high-temperature stability of permeability and low magnetic loss, and the temperature stability of common nickel zinc ferrites cannot meet this requirement. It was reported that Co₂Y added to nickel–zinc ferrites can obviously improve the temperature stability of permeability [20,21,22,23,24] and improve DC-bias superposition characteristics [25]. However, only a few works have been done on the effect of Co₂Y addition on magnetic power loss in Ni–Zn ferrites [25, 26], so the influence of Co₂Y additive on magnetic loss in Ni–Zn ferrites is investigated in this paper.

2 Experiment

The ferrite samples are prepared by solid-state reaction method. Reagent-grade Ni₂O₃, ZnO, and Fe₂O₃ are used in the required proportion in the chemical formula Ni_0.8Zn_0.2Fe₂O₄, the mixtures are milled thoroughly in agate mortar to ensure the raw materials are mixed evenly and then calcined at 1050 °C for 3 h in air. The resulting mixtures are reground to reduce them into micron-sized particles. The obtained powder is divided into two parts, and 2% weight ratio Co₂Y powder is added in one part. Every part is mixed thoroughly and then pressed in a toroidal shape in 200 MPa pressure using 5% PVA solution as the binder. The toroidal green samples are sintered for 3 h at 1220 °C in air followed by natural cooling to room temperature. The crystal structure of obtained ferrite samples is investigated by X-ray diffractometer (XRD, Mac Science MPX18AHF, Cu Kα radiation), and the power losses of ferrite are measured by BH analyzer (Iwatsu SY-8258) and the measurement error is controlled at 1%.

3 Results and discussion

As shown in Fig. 1a, b, the XRD patterns of Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ ferrite powders match well with the standard XRD data for spinel phase; no other phases can be found in the patterns. Initial permeability measurement shows that Co₂Y addition caused decrease in permeability (from 28 to 20). Power loss data at different magnetic flux densities and frequencies are measured for ferrite samples. The magnetic flux density (B_m) dependence of power loss of Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ ferrites at 200 kHz is shown in Fig. 2. It is found that the power loss of Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ ferrite is about half of Ni_0.8Zn_0.2Fe₂O₄ sample, which indicated that moderate addition of Co₂Y in Ni–Zn ferrites could dramatically decrease the power loss of sample [26].

Fig. 1

Powder XRD patterns of Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ ferrites

Fig. 2

Variation of power loss versus magnetic flux density of Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ ferrites at 200 kHz

The measured magnetic loss dependence on the frequency of two samples at B_m = 6 mT is shown in Fig. 3 and it is found that the measured power loss increases with increase of frequency. Compared with Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄, the power loss of Ni_0.8Zn_0.2Fe₂O₄ sample is about twice of the former, which also shows that the power loss of Ni_0.8Zn_0.2Fe₂O₄ obviously decreased with Co₂Y additive.

Fig. 3

Variation of power loss versus frequency of Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ferrites at B_m = 6 mT

The measured power loss includes hysteresis loss, eddy current loss and residual loss. It is regarded that the power loss can be separated using the empirical Legg formula at low frequency and low magnetic flux density range [27]. To investigate the Co₂Y additive influence on hysteresis loss, the relative loss factor (tanδ/µ′) versus magnetic flux density at 500 kHz is measured for Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ and Ni_0.8Zn_0.2Fe₂O₄ samples. The measured data are shown in Fig. 4. Theoretically, tanδ/µ' vs B_m obeys a simple linearity, as predicted by the Legg formula:

$$\frac{\tan\delta }{\mu ^{\prime}}=\frac{1}{2\pi }(a{B}_{\mathrm{m}}+ef+c)$$

(1)

where f is frequency, a, e and c are hysteresis loss coefficient, eddy current loss coefficient and residual loss coefficient, respectively. The Legg formula tells that the slope represents hysteresis loss coefficient. Compared to the experimental results in Fig. 4, it can be found that the linear slope of Ni_0.8Zn_0.2Fe₂O₄ ferrites is bigger than that of Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ sample. The hysteresis loss coefficient obtained from the linear fitting of measured data in Fig. 4 is 3.4 × 10^–5/mT and 2.3 × 10^–5/mT for Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ samples, respectively, which indicates that the hysteresis loss of Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ is about two-thirds of Ni_0.8Zn_0.2Fe₂O₄ under the same test condition.

Fig. 4

Variation of relative loss factor versus magnetic flux density of Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ samples at 500 kHz

The hysteresis loss performance of ferrite can be observed by measuring its hysteresis loop. The amount of energy lost in the material in one cycle of the applied field is proportionate to the area inside the hysteresis loop. Figure 5 gives the room temperature magnetic hysteresis loop for Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ samples measured at 10 mT and 1 MHz. It can be found in Fig. 5 that the Co₂Y addition obviously reduced the area of hysteresis loop, in spite of the decrease in permeability. The hysteresis loops in Fig. 5 also indicate that hysteresis loss decreases with Co₂Y addition in Ni–Zn ferrites.

Fig. 5

Room temperature magnetic loops of Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ ferrites measured at 1 MHz and 10 mT

The influence of Co₂Y addition in Ni_0.8Zn_0.2Fe₂O₄ ferrite on the eddy current loss can also be investigated using the Legg formula. The measured ${{\tan \delta } \mathord{\left/ {\vphantom {{\tan \delta } {\mu^{^{\prime}} }}} \right. \kern-\nulldelimiterspace} {\mu^{^{\prime}} }}$ as a function of frequency for Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ samples at B_m = 6 mT is shown in Fig. 6. It was noted that the measured ${{\tan \delta } \mathord{\left/ {\vphantom {{\tan \delta } {\mu^{^{\prime}} }}} \right. \kern-\nulldelimiterspace} {\mu^{^{\prime}} }}$ linearly increased with increasing f, which was consistent with Formula (1). The eddy current loss coefficient obtained from the linear fitting of measured data in Fig. 6 is 6.9 × 10⁻¹¹ s and 2.3 × 10⁻¹¹ s for Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ samples, respectively, which indicates that the eddy current loss of Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ is about one-third of Ni_0.8Zn_0.2Fe₂O₄ under the same measurement conditions. Experimental data indicate that Co₂Y added in Ni_0.8Zn_0.2Fe₂O₄ ferrite can obviously reduce the eddy current loss. Considering that the eddy current loss would play a dominant role at high frequency, Co₂Y-added Ni–Zn ferrites are more suitable for high-frequency applications [25].

Fig. 6

Variation of relative magnetic loss factor ${{\tan \delta } \mathord{\left/ {\vphantom {{\tan \delta } {\mu^{^{\prime}} }}} \right. \kern-\nulldelimiterspace} {\mu^{^{\prime}} }}$ versus frequency of Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ samples at 6 mT

The above experimental results show that the addition of Co₂Y reduced the power loss of ferrite to about half of the Ni_0.8Zn_0.2Fe₂O₄ sample at the same frequency and magnetic flux density. Considering that the permeability of Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ is less than that of Ni_0.8Zn_0.2Fe₂O₄, the field energy stored in the Co₂Y-added Ni–Zn is more than that in Ni–Zn ferrite with the same magnetic flux density. The reduction of power loss due to the addition of Co₂Y is underestimated, in case of the same magnetic flux density. The Q factor of two samples is measured to compare the ratio of stored field energy to power loss in one cycle, and the variation of Q factor with magnetic flux density and frequency is shown in Figs. 7 and 8. It could be found that the Q value of Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ ferrite is about three times higher than that of Ni_0.8Zn_0.2Fe₂O₄ sample in Figs. 7 and 8, which shows that the difference in Q value between the two samples is more obvious than that of power loss.

Fig. 7

Variation of relative magnetic quality factor (Q) versus magnetic flux density of Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ samples at 6 mT

Fig. 8

Variation of relative magnetic quality factor (Q) versus frequency of Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ samples at 200 kHz

The reason for power loss reduction caused by Co₂Y addition in Ni–Zn ferrites can be analyzed as follows. Co₂Y cannot form a single phase, which reveals that Y-type ferrite is unstable at a high temperature [28]. It is considered that Co₂Y added to Ni–Zn ferrites decomposes at a high temperature, and Co²⁺ ion enters the spinel lattice. The addition of Co²⁺ in spinel improves the performance, as validated by the reduction of power loss, however at the expense of μ_i [29]. It is regarded that Co²⁺ doping is the main reason for hysteresis loss reduction in Ni–Zn ferrites. Ba²⁺ ion cannot enter lattice due to its big radius, so it concentrates on the grain boundary to form a high-resistance layer, which leads to a reduction in eddy current loss. On the other hand, the presence of Co²⁺ stabilizes Fe³⁺, as its third ionization energy is higher than that of Fe²⁺, which also reduces eddy current loss [29].

4 Conclusions

In summary, Ni_0.8Zn_0.2Fe₂O₄- and Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ ferrite samples are prepared by solid-state reaction method and their magnetic properties are investigated. Magnetic measurements show that the magnetic loss obviously decreased after Co₂Y addition. Magnetic loss separation indicates that the hysteresis loss coefficient and eddy current loss coefficient both decreased due to the addition of Co₂Y to Ni–Zn ferrites. The Q value of Co₂Y-added Ni_0.8Zn_0.2Fe₂O₄ ferrite is about three times higher than that of Ni_0.8Zn_0.2Fe₂O₄ sample. The reduction of hysteresis loss caused by Co²⁺ cation doping in Ni–Zn ferrite and decrease in eddy current loss are considered to be the formation of high-resistance insulating grain boundary containing the barium compound. The obvious reduction of eddy current loss reveals that Co₂Y-added Ni–Zn ferrites are more suitable for high-frequency applications.