The M-type hexaferrite, Strontium hexaferrite, is an excellent candidate for technological applications because of its high uniaxial magneto-crystalline anisotropy, large magnetization, high permeability, low conductive losses, excellent chemical stability, corrosion resistance and excellent high-frequency response [1] [2] [3]. Strontium hexaferrite has been widely used as materials for industrial applications, such as in microwave devices, small motors, electromagnetic wave absorber, and ferroxdures [4] [5]. Besides this, the popularity of strontium hexaferrite is also due to its economic success, which is its low price per unit available magnetic energy and its wide availability. So far, efforts have been made to further improve electrical, dielectric and magnetic properties of strontium hexaferrite by means of doping, heat treatment, ion substitution, and processing conditions [6] [7] [8] [9] [10].

In M-type hexaferrite, the structure is comprised of 64 ions per hexagonal unit cell on 11 distinct basis sites. The 24 Fe³⁺ iron atoms of a unit cell occupy five different interstitial sites: three octahedral sites (2a, 12k, and 4f₂), one tetrahedral site (4f₁) and one trigonal bi-pyramidal site (2b). The coupling of these sites by superexchange interaction via O− 2 gives rise to a ferrimagnetic structure. The sites 12k, 2a, and 2b are sites with spin up while sites 4f₂ and 4f₁ are sites with the downs-spin [10]. This provides an ample opportunity to tune the magnetic properties of M-type hexaferrite compound by carefully engineering site occupancy in favor of increasing net magnetization of the compound. In view of this, attempts are made with either partial substitution of Sr²⁺ or Fe³⁺ sites. For example, substitution with non-magnetic ions such as Al³⁺ [11] [12], Zn³⁺ [13], Ga³⁺ [14] and Cd³⁺ [15] [16] and magnetic ions such as Co²⁺ [17] and Cr³⁺ [18] at Fe sites, or with the partial substitution of Sr²⁺ site by RE³⁺ such as La³⁺ [19] [20], Nd³⁺ [21], Sm³⁺ [22] [23] and Pr³⁺ [24] ions, and substitution of Sr²⁺/Fe³⁺ together with Pr-Zn [25], La-Cu [26], and La-Zn [27]. The majority of these studies have attributed changes in the magnetic behavior of doped SrM either to the site-occupancy, which perturbs exchange interaction between Fe³⁺-O²⁻Fe³⁺, anisotropy changes occurring at 2b sites due to perturbation in electric field gradient or extrinsic features such as particle size.

Doping of non-magnetic Al³⁺ in SrFe_12−xAl_xO₁₉ has been reported to bring a considerable enhancement in the coercivity along with a reduction in saturation magnetization [11] [28] [29]. Thus with the Al³⁺ substitution, in spite of a large increase in the coercivity, the compound attains abysmal magnetization value. A strategy could be designed to keep both coercivity and magnetization value high. In the present work, an attempt is made to co-doped Co²⁺-Al³⁺ in the SrFe₁₂O₁₉ compound. Magnetic ion Co²⁺ (~3.7 μ_B) is chosen to maintain high magnetization value, while Al³⁺ doping is expected to maintain high coercivity value in SrFe₁₂O₁₉. Furthermore, Lechevallier et al. reported that the only light rare-earth enters the structure, which a solubility that is related to the shape of the charge distribution of the 4f electrons to its surroundings in the crystal structure. Thus, only those rare-earth ions have easy accommodation at the Sr²⁺ site, whose surroundings favor an oblate electronic distribution. This makes Pr and Nd (with oblate 4f charge distribution, negative Stevens constant) ideal candidate to occupy the Sr²⁺ site [25]. Furthermore, Co²⁺ ions were reported to modify the surroundings of the Sr²⁺ site, improving the introduction of rare-earth ios with oblate electronic distributions, in fact, the presence of Co²⁺ increases the solubility of Pr³⁺ [25]. Also, RE³⁺-Co²⁺-Al³⁺, co-substitution naturally maintains the charge neutrality condition in the compound. With the substitution of RE³⁺ for the Sr²⁺, the extra positive charge is compensated by the substitution of Co²⁺ at the Fe³⁺ site. In the absence of Co²⁺, one of the Fe³⁺ has to convert to Fe²⁺, which has a deleterious effect on the magnetization. With this strategy in mind, the present study investigates, the influence of the 1) rare-earth substitution, Pr³⁺, and La³⁺, at the Sr²⁺ site and 2) Co²⁺-Al³⁺ substitution at Fe³⁺ sites on the overall magnetic properties of the Sr_0.7RE_0.3Fe_12−2xCo_xAl_xO₁₉ compound.

The series of Sr_0.7RE_0.3Fe_12−2xCo_xAl_xO₁₉ (RE: La³⁺ and Pr³⁺, x = 0, 0.2, 0.4, 0.6 and 0.8) compounds were prepared by auto-combustion method [11]. Nitrate salts of Fe(NO₃)₃∙9H₂O (Sigma Aldrich, 99%), Sr(NO₃)₂ (Sigma Aldrich, 99%), Al(NO₃)₃∙9H₂O (Sigma Aldrich, 99%), RE(NO₃)₃∙9H₂O (Sigma Aldrich, 99%) and C₆H₈O₇ (Sigma Aldrich, 99%) were selected for the synthesis Sr_0.7RE_0.3Fe_12−2x Co_xAl_xO₁₉ compounds. The stoichiometric amount of nitrates salts and citric acid were mixed in 25 ml of deionized water under constant stirring on a hot plate at 80˚C for about 30 minutes. Aqueous ammonia was added dropwise in the homogeneous mixture until the pH value of the solution reached 6.5. The intermediate precipitate was heated at 100˚C on a hot plate under stirring until the viscous brown gel is formed. The temperature is suddenly increased to 300˚C to turn it into magnetic powder. Powders were calcined in air at 1000˚C for twelve hours. The precursor used in the preparation of 1 g of sample is listed in Table 1.

The crystal structure and phase purity of the obtained samples were identified by analyzing x-ray diffraction (XRD) pattern collected via Bruker D8 Advance X-ray diffractometer using Cu K_α radiation source (λ ~ 0.154056 nm). Vibrating sample magnetometer (VSM) was used to collect hysteresis loops of the samples at room temperature in the maximum field of 1.2 T. Curie temperature, Tc, were measured using Thermogravimetric analyzer, TGA, (DuPont 910) equipped with a permanent magnet. Room temperature ⁵⁷Fe Mossbauer spectroscopy was

employed to derive hyperfine parameters. The Mossbauer spectrometer (SEE Co. Minneapolis, MN USA) was calibrated against α-Fe foil. The Mossbauer spectra were analyzed using WMOSS software (SEE Co. Minneapolis, MN USA).

The room temperature XRD pattern of Sr_0.7RE_0.3Fe_12−2xCo_xAl_xO₁₉ (RE: La³⁺ and Pr³⁺, x = 0.0, 0.2, 0.4, 0.6, and 0.8) is shown in Figure 1. XRD patterns of the substituted compounds indicated the presence of a magnetoplumbite structure (ICCD 080-1198) corresponding to the hexagonal P63/mmc symmetry phase group without any secondary phase. The lattice parameters of the compound were calculated from their inter-planar spacing d_hkl corresponding to the major peaks (107), (114), and (008) using equation [30]:

d ( h k l ) = ( 4 ( h 2 + h k + k 2 ) 3 a 2 + l 2 c 2 ) − 1 2 (1)

where h, k, and l are Miller indices. The calculated lattice parameters a and c of the hexaferrites are shown in Table 2. The plot for a and c as a function of doping content is shown in Figure 2. A linear decrease in the lattice parameters a and c with the doping content was observed. The linear decrease in the lattice parameter may occur due to the substitution of Al³⁺ (ionic radii, r ~ 0.51 Å) and

Co²⁺ (r ~ 0.72 Å) for Fe³⁺ (r ~ 0.64 Å) and smaller La³⁺ (r ~ 1.172 Å) and Pr³⁺ (r ~ 1.13 Å) for the Sr²⁺ (r ~ 1.32 Å) ions [31]. According to Wagner [32], an examination of c/a parameter ratio may be used to quantify the structure type, as the M-type (magnetoplumbite) structure can be assumed if the ratio is observed to be in the range 3.917 and 3.963. As per Table 2, the c/a ratios of as-prepared samples are in the range of 3.908 to 3.919, assuring that the as-prepared samples have maintained the M-type structure.

The crystallite size of as-synthesized particles was calculated using Scherrer’s equation [33]:

D ( h k l ) = k λ / ( β cos ϑ ) (2)

where k denotes the Scherrer constant (k = 0.9), λ is the wavelength of x-ray source (λ = 0.154056 nm), β is the full-width-half-maximum of a diffraction peak and θ is the diffraction angle. As listed in Table 2, the crystallite size of as-synthesized samples is in the range of 54 to 77 nm. In the comparison of the pure SrFe₁₂O₁₉ sample with a crystallite size of 80.7 nm, the doped compounds show reduced crystallite size. Grain refinement is usually observed in rare-earth and doped oxide compounds [34]. This grain refinement is reported to result from the 1) increased microstrain and defect density with substitution content and 2) diffusion of substituent element to the grain boundaries, the migration that restrains the grain growth by lowering down the grain growth mobility. If the retarding force generated is more than the driving force for the grain growth due to dopants, the movement of the grain boundary is impeded [35].

The room temperature hysteresis loops, M vs. H, of Sr_0.7RE_0.3Fe_12−2xCo_xAl_xO₁₉ are shown in Figure 3. The magnetic parameters viz. saturation magnetization, Ms, remanence, Mr, and coercivity, Hc were extracted from the hysteresis loops and are listed in Table 3. The M vs. H curves show hard ferrites behaviors with high coercivity. The magnetic properties have been changed significantly upon RE³⁺ and Co²⁺-Al³⁺ substitution in Sr_0.7RE_0.3Fe_12−2xCo-Al_xO₁₉ as compared to undoped ferrite SrFe₁₂O₁₉. The Ms value of Pr³⁺-SrM is observed to have a maximum value of 86 emu/g for x = 0.2 but decreases with Co²⁺-Al³⁺ content and achieves a minimum value of 65 emu/g for x = 0.8. However, in the case of La³⁺-SrM, the Ms increases with x and attains a maximum value of 87 emu/g at x = 0.8. As compared to pure SrFe₁₂O₁₉, La³⁺-SrM displays a noticeable 20% enhancement while Pr³⁺-SrM shows a 10% reduction in the Ms value. The variation of Mr and Ms as a function of x content is shown in Figure 4. The value of Mr follows a trend similar to that of Ms. As explained in the Mossbauer section, for series of compounds shows the marked preference of Co²⁺-Al³⁺ for the 2a site, a site with the spin-up moment. Despite the substitution of Co²⁺-Al³⁺ at the 2a site with the up-spin moment, the magnetization of La³⁺-SrM displayed a marked increase in the magnetization value with the substitution. A possible explanation for this observation is that the substitution of non-magnetic La³⁺ at the Sr²⁺ 1) could enhance hyperfine fields at 12k site due to strengthening in the Fe³⁺-O²⁻Fe³⁺ superexchange interaction giving higher net magnetization. This increase in the hyperfine field of 12k site is evident from the Mossabuer analysis, and 2) may increase the spin canting at the near-neighbor 4f2 site (Sr²⁺-4f2 ~ 0.366 nm), which results in a positive contribution to the overall magnetic moment. However, the possibility of Co²⁺ occupying a 4f2 site cannot be ignored, which increases the net positive moment per unit cell of the compound [36] [37]

[39]. On the other hand, the substitution of magnetic Pr³⁺ at the Sr²⁺ site may not allow spin-canting at the 4f2 site, and with the replacement of Fe³⁺ at the 2a site, the overall magnetic moment decreases in the Pr³⁺-SrM compounds.

The variation of coercivity, Hc, of La³⁺-SrM and Pr³⁺-SrM compounds as a function of x content is shown in Figure 5. Consistently La³⁺-SrM samples display higher Hc values than Pr³⁺-SrM compounds for all x values. The Hc value of La³⁺-SrM displays a marked ~50% increase at x = 0.2 when compared to that of pure SrFe₁₂O (Hc ~ 3918 Oe). On the other hand, Hc value decreases linearly with increasing the x content for the Pr³⁺-SrM compound. Being an extrinsic

property, the coercivity depends on factors such as crystal structure, morphological features, elemental composition, and defects. In general, Hc can be expressed as,

H c = ( α ( 2 K 1 / ( μ o M s ) ) − N ( M r + M s ) / μ o (3)

where α is the microstructure factor which has a reciprocal dependence on the grain size, N is the demagnetization factor, 2K1/(μ_oM_s) is the Ha is the magnetocrystalline anisotropy field [39]. In the M-type hexaferrite, magnetocrystalline anisotropy mainly originates from Fe³⁺ ions at the 2b and 4f1 site [40] [41] ]. It has been reported that Al³⁺ ions have a preference for 4f2, 2a, and 12k sites, [42] [43], and Co²⁺ ions mainly prefer 4f2 (mainly) and 2a sites [37] [38] [44]. In accordance with the Mossbauer analysis, discussed later, it is observed that in both series of samples studied herein, the Co³⁺-Al³⁺ ions have a preference for the 2a site. Thus the intrinsic coercivity as per Equation (3), should increase for both series of samples due to the increase in the magnetocrystalline anisotropy, as the site 2b remains unaffected with the said substitution. However, from Figure 5, it is observed that the La³⁺-SrM sample shows a maximum (5829 Oe) in coercivity value at x = 0.2, followed by a decrease with increasing x content in the sample. On the other hand, the Hc value of the Pr³⁺-SrM sample shows a monotonous decrease with the x content in the sample. Also, the La³⁺-SrM sample displays consistently higher coercive field value as compared to Pr³⁺-SrM. In a comparative study on Sr(La/Pr)Fe₁₂O₁₉ samples it was reported that the Pr³⁺ doped Sr(Pr)Fe₁₂O₁₉ displayed higher coercivity than La³⁺ doped Sr(Pr)Fe₁₂O₁₉ sample because of the higher magnetocrystalline anisotropy of the former. However, the contrary results in the present study can be explained by taking microstructure factor, α, in account in Equation (3) above. The La³⁺-SrM samples have relatively smaller crystallite size as compared to Pr³⁺-SrM; as a result, the microstructure factor, α, will have a higher value than that of Pr³⁺-SrM sample. This will make the first term in the Equation (3) above greater than for La³⁺-SrM as compared to that for Pr³⁺-SrM. The observed decrease in coercivity with the substitution could be attributed to the fact that the second term in the Equation (3) decreases much rapidly as compared to the αHa term.

The Curie temperature, Tc, of Sr_0.7RE_0.3Fe_12−2xCo_xAl_xO₁₉ are shown in Figure 6. The maximum Tc value for La³⁺-SrM and Pr³⁺-SrM was observed to be 505 and 525 K, respectively at x = 0.40. After attaining maximum value, the Tc decreases gradually for x > 0.40. The overall decrease in Tc after attaining maximum can be explained on the basis of three combined effects [45] [46], 1) substitution of Co²⁺-Al³⁺ ions for Fe³⁺ ions reduce the Fe³⁺ ions and hence leads to a reduction of Fe³⁺-O²⁻-Fe³⁺ number of superexchange interaction and strength and 2) lattice contraction with the substitution alters the bond length and angle of Fe³⁺-O²⁻-Fe³⁺ from its optimum interaction strength value. However, in the beginning, Tc drops down for x = 0.20, where presumably Fe³⁺ might be substituted more with non-magnetic Al³⁺ and above combined effect become prevalent. However, the increasing magnetic Co²⁺ ion content helps regain the strength of the superexchange interaction and hence increases the Tc value for x = 0.40 [47].

In order to investigate the site occupancies of transition metal ions and resulting hyperfine parameters in the samples, Mossbauer spectra were collected at room temperature. Figure 7 shows the fitted Mossbauer spectra as a function of content x in the as-synthesized samples. All spectra consist of five Lorentzian sextets originating from 12k, 4f1, 4f2, 2a, and 2b crystallographic sites of Fe³⁺ ions. The spectra were fitted with the constrain that all the linewidths of the absorption lines were the same. The extracted Mossbauer parameters are listed in Table 4(a) and Table 4(b). The observed sequences of the magnetic hyperfine magnetic fields, HF (4f2 > 2a > 4f1 > 12k > 2b) and isomer shift, δ (the isomer shifts follow the sequence of 4f2 > 12k > 2a > 2b > 4f1) for both samples are in agreement with the reported results [47].

It is well known that the magnitude of the hyperfine magnetic field at the Fe³⁺ site depends on the distribution of neighboring magnetic cations. The hyperfine field, HF, the parameter of the most intense 12k line effectively shows no variation

with x in Pr³⁺-SrM, while HF increases with x in the La³⁺-SrM sample. The increase in HF value of La³⁺-SrM is in agreement with the observed increase in the saturation magnetization, Ms, with content x. Other lines 4f1, 2a, and 4f2 show a monotonous decrease in HF with x.

The isomer shift plot as a function of x is shown in Figure 8. The isomer shift, δ, for both series of samples for all sites visibly remains invariant except for the 2b site. Also, the 2b site’s isomer shift value clearly reflects the influence of the

rare-earth ion in the compound. In the presence of rare-earth ion, the isomer shift value of 2b drops significantly as compared to that of a pure SrFe₁₂O₁₉ compound. Isomer shift is a measure of 3d electron density at the nucleus of the Fe atom. The closest sites of Sr²⁺ sites are 12k, 4f2, and 2b. The 2b sites are located in the same plane as Sr²⁺ ions (at a distance of 0.340 nm), while the 4f2 and 12k sites are located in the adjacent planes at nearly the same distance from Sr²⁺ viz. Sr²⁺-12k ~ 0.365 nm and Sr²⁺-4f2 ~ 0.366) nm [48] [49]. Pr [Xe] 4f³ 6s² and La[Xe] 5d¹ 6s² with more electrons as compared to Sr (Kr[5s²⁺]) has a negative effect on the isomer shift, that is increased substitution of Pr³⁺ or La³⁺ increases the net s-electron charge density at the 2b site, which in turn decreases the isomer shift value at the 2b site.

The quadrupole shift, QS, values of 2b site show a similar ~10% drop in both Pr³⁺ and La³⁺ substituted samples. The quadrupole values for all sites remain invariant with x for both series of samples. It is known that QS is a measure of the asymmetry of a crystallographic site. Since the 2b sites are located in the same plane as Sr²⁺, the substitution of the smaller Pr³⁺ and La³⁺ for Sr²⁺ ions induces a deformation, which affects the electronic site symmetry of the nearby 2b site. Due to the substitution of smaller La³⁺ and Pr³⁺ ion for Sr²⁺ ion, the lattice constant c contracts. Consequently, large Fe-O distances parallel to the c-axis decreases. As a result, the oxygen bipyramid of the 2b-site becomes more symmetric, which is in agreement with the observed reduction in the quadrupole shift value with content x. A similar decrease in 2b quadrupole splitting originating from the presence of La³⁺ substitution has been reported in Sr_1−xLa_xFe_12−yCo_yO₁₉ [39].

A close examination of the fitted spectra shows a decreasing trend of 2a absorption intensities with increasing x. Without Co²⁺-Al³⁺ co-doping, the intensity of 2a site lines in both Sr_0.7La_0.3Fe₁₂O₁₉ and Sr_0.7Pr_0.3Fe₁₂O₁₉ are nearly the same as that of undoped SrFe₁₂O₁₉. However, with the substitution of Co²⁺-Al³, the 2a absorption area decreased linearly and finally disappeared at x = 0.6 and 1.0 for La³⁺-SrM and Pr³⁺-SrM sample, respectively. This suggests preferential occupation of Co²⁺-Al³ for the 2a site. It has already been reported that Al³⁺ ion prefers 12k and 2a site [50] while Co²⁺ was reported to enter 4f2 site in Sr_1−xLa_xFe_12−xCo_xO₁₉ [51]. There is only one 2a site in every one formula unit of M-type hexaferrite. Therefore, the fact that the 2a absorption line of Sr_0.7La_0.3Fe_12−2xCo_xAl_xO₁₉ disappeared completely at x = 0.6 means that all the Co²⁺-Al³ dopants almost exclusively go into 2a site. In the case when x > 0.5, the excessive Co²⁺-Al³ dopant should occupy the crystallographic sites other than the 2a site. On the other hand, in the case of Sr_0.7Pr_0.3Fe_12−2xCo_xAl_xO₁₉, 2a absorption intensity has decreased linearly and has completely diminished at x = 1.0, which indicates that all the 2a sites become full when x = 1.0. This suggests that only half of the Co²⁺-Al³⁺ dopant occupies the 2a site compared to the case of the La³⁺-SrM sample. Then, the other half of the 2a site should be occupied by Fe³⁺ ions through the range of x examined. In the case of Pr³⁺ series samples, a slight decline in the absorption area for the most intense 12k site is observed, while in the La³⁺ containing samples, 12k areal almost remains constant. Thus it seems that the excessive Co²⁺-Al³⁺ dopant occupies the 2a site and 12k site in Sr_0.7Pr_0.3Fe_12−2xCo_xAl_xO₁₉. The above observation clearly indicates that the tendency of 12k site occupation by the Co²⁺-Al³⁺ dopant is affected by the species of rare-earth ions at the Sr²⁺ site. Where Co²⁺-Al³⁺ impurity mostly prefers 2a site and its preference for 12k site is enhanced in the presence of Pr³⁺ rather than La³⁺, once the 2a site is completely filled. This observation is interesting considering the fact that there are no 2a sites in the vicinity of Sr²⁺, and hence the substitution at Sr²⁺ with rare-earth ions is not expected to influence the site occupancy at the 2a site. Furthermore, the physical reason for the preferential site occupancy of Co²⁺-Al³⁺ in the presence of La³⁺ or Pr³⁺ is not clear at this stage. Mossbauer spectroscopy cannot discriminate against the site occupancies of both Co²⁺ and Al³⁺ ions separately. But one scenario is that Co²⁺ ions might not prefer the 2a site in the presence of Pr³⁺ ion, which needs further investigation.

The structural, magnetic, and Mossbauer spectrum study of the influence of rare-earth La³⁺ and Pr³⁺ on Sr_0.7RE_0.3Fe_12−2xCo_xAl_xO₁₉, (RE: La and Pr) hexaferrite compounds were investigated. Overall, dopant ion brings a unit cell contraction, which affects the Curie temperature and saturation magnetization of both series of samples. La³⁺-SrM samples show marked improvement in the magnetization value, while Pr³⁺-SrM samples show deterioration in Ms value with the Co²⁺-Al³⁺ substitution. This is suggestive of the fact that magnetic Pr³⁺, unlike non-magnetic La³⁺ (no 4f electrons), substitution for Sr²⁺ could be detrimental to the overall magnetization behavior of the SrFe₁₂O₁₉ compound. The Co²⁺-Al³⁺ substitution brings in grain refinement, more for the La³⁺-SrM than Pr³⁺-SrM, which has a considerable impact on the coercivity values of the compound. The Mossbauer spectral analysis revealed 2a as a preferred substitution for Co²⁺-Al³⁺ ions for both series of samples. Due to lattice contraction, the 2b site displayed a considerable reduction in its quadrupole shift value. The improved hyperfine field for the 12k site in the La³⁺-SrM series is attributed to the increase in its magnetization value with the substitution of Co²⁺-Al³⁺ ions for Fe³⁺. For the study, it is concluded that the substitution of rare-earth ions with 4f charge distribution for Sr²⁺ could be detrimental to the overall magnetic performance of the M-type hexaferrite materials.

The authors declare no conflicts of interest regarding the publication of this paper.

Ghimire, M.L., Kunwar, D.L., Dahal, J.N., Neupane, D., Yoon, S. and Mishra, S.R. (2020) Co-Doped Rare-Earth (La, Pr) and Co-Al Substituted M-Type Strontium Hexaferrite: Structural, Magnetic, and Mossbauer Spectroscopy Study. Materials Sciences and Applications, 11, 474-493. https://doi.org/10.4236/msa.2020.117033