1 Introduction

Y-type hexagonal ferrites Ba2Me2Fe12O22, where Me represents a divalent ion such as cobalt, nickel and zinc, consist of S and T molecular unit having the unit cell formula 3(ST) [1]. Y ferrites were the first ferroxplana ferrites to be discovered, which have a preferred plane of magnetization perpendicular to the c-axis at room temperature. There is no five-coordinate site in the T-block, so the spin-orbit coupling contributes little to K. As a result, these ferrites have lager negative K values. Among these ferrites, Co2Y has the highest magnetic anisotropy in room temperature. Co2Y ferrite has the strong magnetic anisotropy field and high cut-off frequency, which is one order of magnitude higher than the traditional spinel [2–4]. Y-type hexagonal ferrites are qualified for various applications in modern information and communication processing devices, used as UHF soft magnetic material [5–10].

Widely used as high-frequency soft magnetic material, low initial permeability of Co2Y cannot meet the development of electronic devices any more. In order to apply in more extensive electronic areas, it’s necessary to use ion modification to improve properties of material. Cu2+ ion has similar radius as Fe2+ ion, and it is one of most effective ways to improve property of ferrites [11–15]. The melting point of CuO is 1026 °C, it melts during the sintering process and forms liquid phase, which could lower the sintering temperature and promote grain growth. Zhang et.al. [16] reported Cu substitution Z-type hexagonal barium ferrites, and the sintering temperature of ferrite has been lowered to 1050–1130 °C. It illustrated CuO additive could lower the sintering temperature about 150 °C. In ferrites, CuO substitution not only reduced the sintering temperature, but also adjusted the magnetic properties. In present work, we chose Cu2+ ions to substitute Co2+ ions of Co2Y ferrite (Ba2Co2−xCuxFe12O22, x = 0.0, 0.2, 0.4, 0.6, 0.8) at low temperature (~1100 °C). Phase formation, microstructure, magnetic properties and DC resistivity of samples were investigated. The magnetism of effect of Cu-substitution on structure and property was discussed.

2 Experiment procedure

Y-type hexagonal ferrites (Ba2Co2−xCuxFe12O22, x = 0.0, 0.2, 0.4, 0.6, 0.8) were prepared using solid-state reaction method. Analytic reagent grade raw materials BaCO3, CuO, CoO, and Fe2O3, were mixed in a ball mill for 6 h, using stainless-steel balls and deionized water as media. The mixed powders were dried and pre-sintered at 1050 °C for 3 h in air. Then the powders were ball-milled again for 12 h in deionized water. After drying, powders were granulated by adding 8 wt% of polyvinyl alcohol (PVA) as a binder and pressed into 2–3 mm thick plates. Then the samples were sintered at l100 °C for 4 h.

The phase compositions of the samples were determined using an X-ray diffractometer (XRD, DX-2700, Haoyuan Co.) with Cu Kα radiation. The microstructures of the samples were characterized using a scanning electron microscope (SEM, JEOL, JSM-6490). The bulk density was measured using an auto density tester (GF-300D, AND Co.) by Archimedes’ principle. Magnetization hysteresis loops were measured using a vibrating sample magnetometer (VSM, MODEL, BHL-525). Complex magnetic permeability was determined by using a HP-4991B RF impedance analyzer. DC resistivity of samples (Ag-Ag contact coated onto both sides) was measured by two probe method using a digital super megohmmeter (DSM-8104).

3 Results and discussion

3.1 Phase formation

Figure 1 showed the XRD patterns of the samples with different compositions compared with the standard XRD spectrum of Y-type phase. All the diffraction peaks were matched with standard patterns, which confirmed that all the samples were in single phase with hexagonal structure and no impurity phase was present. When x = 0.0, well defined Co2Y ferrite phase could be obtained. With Cu2+ ion adding, the peak intensity increased, which indicated that CuO addition could lower the sintering temperature and promoted the growth of crystalline grain. When the sintering temperature increased to 1100 °C, CuO began to wet and form liquid which spread across the surface of Co2Y ferrite [17].

Fig. 1
figure 1

XRD patterns of Ba2Co2−xCuxFe12O22 (x = 0.0, 0.2, 0.4, 0.6, 0.8) samples sintered at 1100 °C

Full size image

The melting CuO preferred to enter crystal lattice and occupy the lowest energy configuration. Hence, Cu-substitution not only modified properties of Co2Y ferrite, but also lowered the sintering temperature.

Lattice constants (a and c) of sintered samples were calculated using different diffraction peaks in the diffraction patterns as per the hexagonal crystal, and variations of lattice constants as a function of CuO content were shown in Fig. 2. There was a decrease in lattice constants a and c with CuO substitution from x = 0 to x = 0.8. First, the lattice constants (a = 0.587 nm, c = 4.356 nm) of pure Co2Y ferrites were close to the theoretical values. In consideration of ionic radii of cations, the ionic radius of Cu2+ ion is 0.73 Å and that of Co2+ ion is 0.745 Å. Cu2+ substituted Co2+ therefore brought a lattice distortion and led to a decrease of the lattice constants.

Fig. 2
figure 2

Effects of CuO substitution upon lattice parameters of samples

Full size image

3.2 Microstructure and densification

The cross-sectional SEM micrographs and bulk density of the sample were shown in Fig. 3. The microstructure of the samples was influenced by the variance of CuO contents. As shown in Fig. 3a, none substitute Co2Y ferrite (x = 0.0) presented plate-like grains, and had smaller average grain size and lower densification comparing with other samples. With CuO increasing, the grain size increased and densification enhanced, which led to increase of bulk density, as shown in Fig. 3f. CuO-substitute brought more compactly stacked and less pores between particles, which was ascribed to the appropriate segregation and volatilization of Cu [16]. When x = 0.8, part of the grain boundaries disappeared, and fewer pores left between particles and higher bulk density was enhanced.

Fig. 3
figure 3

SEM micrographs of the cross-sectional and bulk density of Ba2Co2−xCuxFe12O22 samples (a x = 0.0, b x = 0.2, c x = 0.4, d x = 0.6, e. x = 0.8, f bulk density)

Full size image

As shown in Fig. 3f, bulk density of the samples increased, correspond to SEM images. The density increased with CuO content greatly. The density of pure Co2Y sample was only 4.326 g/cm3. CuO addition improved the density dramatically and maximum density reached 5.015 g/cm3 when x = 0.8, which is 92.9% of the standard theoretical density (5.40 g/cm3) [1].

3.3 Magnetic properties

Magnetic hysteresis loops of the samples were shown in Fig. 4. Figure 5 showed values of saturation magnetization (M s) and coercivity (H c) of samples. All the samples exhibited typical hysteresis behavior under applied magnetic field at room temperature with complete saturation magnetization from ~H > 3000 Oe. This evidenced the magnetic nature of the samples. The theory saturation magnetization of pure Co2Y ferrite was 34 emu/g [1]. In present experiment, when x = 0.0, the saturation magnetization was 31.7 emu/g. The loss of M s was due to insufficient grain growth without CuO addition in 1100 °C. This result was consistent with previous analysis [17]. With CuO substitution, M s increased from 31.7 emu/g (x = 0.0) to 32.3 emu/g (x = 0.2), and then decreased to 27.7 emu/g (x = 0.8).

Fig. 4
figure 4

M-H hysteresis loops of the samples in room temperature

Full size image
Fig. 5
figure 5

Saturation magnetization and coercivity of Ba2Co2−xCuxFe12O22 (x = 0.2, 0.4, 0.6, 0.8) ferrite

Full size image

In ferrites, the saturation magnetization is mainly come from total net magnetic moment (m) and chemical composition. The value of m of Co2Y ferrite is 9.8μ B per unit formula. Co2+ ion has a marked preference for octahedral sites and carries a magnetic moment of 3.7μ B. Among the octahedral sites of Y structure, there are one spin down octahedral sublattice 6CVI (↓) and three spin up sublattices 3aVI(↓), 18 hVI(↓) and 3bVI(↓). 0.9 cobalt ions are located in the only spin down octahedral sublattice, whilst the remnant 1.1 are supposed to be distributed among the spin up sublattices [18]. Magnetization in a ferrite comes from the total net magnetic moment between tetrahedral and octahedral sublattices which align in an antiparallel arrangement with respect to each other due to super exchange [19].

When CuO substituted Co2Y ferrites, when x = 0.2, the increase of M s was due to high densification and few pores between grains. With increasing of CuO, the decrease of M s mainly came from the decrease of total magnetic moment. The magnetic moment of Cu2+ ions is 1.3μ B, which is smaller than that of Co2+ ions (3.7μ B). Cu2+ ions distributed in the spin up octahedral sub lattices 3aVI, 18 hVI, or 3bVI. The lower magnetic moment of Cu2+ ions in the vicinity of Co2+ ions diluted the strength of the super exchange interaction and reduced the total magnetic moment of ferrite. Thus, the saturation magnetization showed a downward trend with increase of CuO content after x = 0.2.

Coercivity (H c) is the magnetic field intensity required to reduce the magnetization of a magnetic sample to zero after magnetization saturation, and it is the principle parameter used to distinguish between soft and hard magnetic materials. Pure Co2Y-type barium ferrite is a soft magnetic material with an experimental maximum coercivity of about 377 Oe [1]. In the present study, a range of 172–136 Oe was obtained as shown in Fig. 5. When CuO substituted, coercivity dropped and the samples possessed better soft magnetic properties. Firstly, Cu2+ ion substitution for Co2+ ion weakened the magnetic planar anisotropy, which in turn reduced coercivity. Secondly, the different values of coercivity were related with the domain wall motion, which was affected by the grain size and could be enhanced by the increase of grain size. Lastly, lattice distortion and densification were another reason of coercivity change. Lattice distortion would decrease coercivity and high densification increased coercivity. Hence, the variation in coercivity may be due to the combined effects of ion composition, grain size, lattice distortion and densification.

The real (μ′) and imaginary (μ″) part of magnetic permeability versus frequency for the CuO substitution Co2−Y ferrite were determined from the magnetic measurement. Figure 6 indicated the frequency dependence of complex permeability with different CuO content. With CuO content increasing, the value of μ′ increased gradually at 0.1–1 GHz. Meanwhile, the μ″ increased slightly and kept a low value of under 0.25. Hence, the samples processed low magnetic loss (tanδ) (about 0.08) from 0.1 to 1 GHz.

Fig. 6
figure 6

Effect of CuO substitution for magnetic permeability of Co2Y ferrite, a measured real part of μ′ as function of field frequency, b measured imaginary part of permeability of μ″ as function of field frequency

Full size image

Magnetic permeability of ferrites can be expressed by

$$\mu \propto \frac{{M_{\text{s}}^{2} }}{{K_{1} + \lambda_{\text{s}} \sigma }}$$

where M s stands for saturation magnetization, K 1 stands formagneto-crystalline anisotropy, λ s stands for magnetostriction factor and σ stands for internal stress [8]. Hexagonal ferrites containing Co2+ ion inclines to exhibit planar anisotropy, such as Co2Y, which has the highest magnetic anisotropy constant among hexagonal ferrites, with a (K 1 + 2K 2) value of −2.6 × 105 Jm−3 [1]. The variation of saturation magnetization and anisotropy determined the variation permeability of the samples. Cu2+ ion substitution weakened the magnetic planar anisotropy, which in turn affected magnetic permeability. According to the discussion above, CuO substitution in Co2Y ferrite reduced saturation magnetization as well as the magnetic anisotropy. The increase of μ′ revealed that the declining of M s was shaper than that of K 1.

In ferrite, permeability is relate to microstructure [8]. In our experiment, appropriate amount of CuO led to more compactly stack and larger grains, which facilitated magnetization rotation. Meanwhile, less pores between boundaries made the domain wall motion easier. Besides, the liquid phase of CuO helped to form uniform microstructure that cut down the internal stress, which increased magnetic permeability. Thus, with CuO increasing from 0.0 to 0.8, real part of magnetic permeability increased gradually. For none substitution sample (x = 0.0), the immaturity of grain growth and low bulk density caused low magnetic permeability.

3.4 DC resistivity

The DC resistivity of the samples with different CuO substitution contents was shown in Fig. 7. The radius and thickness of the samples were measured then sliver was coated on both sides. Using two probe methods, resistance of the samples were measured by digital super megohmmeter (DSM-8140). The DC resistivity of the samples can be obtained by

Fig. 7
figure 7

The DC resistivity of the samples with different CuO substitution contents

Full size image
$$R = \rho \frac{s}{l}$$

The room temperature DC resistivity first increased from 3.65 × 105 Ω cm (x = 0.0) to 9.81 × 105 Ω·cm (x = 0.4) with CuO increasing, and then dropped to 3.14 × 104 Ω·cm (x = 0.8). The dominant conduction mechanism in ferrites is electron hopping between Fe2+ ion and Fe3+ ion at two interstitial sites. According to above discussion, Cu2+ ions preferentially occupy octahedral sites. The existence of Cu2+ ion would create internal stress and inhibit the electron hopping, and this led to the increase of DC resistivity [16]. When CuO content further increased, some grain boundaries disappeared and fewer pores left between particles, as shown seen from the SEM images. It has been reported by Yang Bai et al. [20] that deficiencies tend to gather at grain boundaries to form a more resistive surface. When some grain boundaries disappeared in the samples, it might lead to the decrease of resistivity. Besides, the increasing electrical resistivity was attributed to increasing number of pores by Mukhtar et al. [21]. Pores were thought to hinder the motion of charge carriers from their work. Hence, the decrease of pores could make electron hopping easier and cause the loss of resistivity. The decrease of grain boundaries and pores led to the decrease of resistivity.

4 Conclusions

In the present study, we have investigated the effect of CuO substitution upon the structural and magnetic properties of Co2Y barium ferrite. Single phase CuO substituted barium ferrites have been successfully synthesized at a low sintering temperature of 1100 °C. CuO substituted Y-type ferrites were suitable for high frequency applications because the magnetic permeability increased and remained steady from 100 to 700 MHz while the magnetic loss increased a little. This substitution not only improved microstructure but also brought changes to other magnetic properties. The saturation magnetization and magnetic anisotropy decreased as Cu2+ ion substituted Co2+ ion. DC resistivity of the samples reached maximum value 9.81 × 105 Ω cm at x = 0.4 and then decreased. It would be necessary to improve the DC resistivity in the subsequent research in order to cut down eddy current loss in high frequency.