La–Mg-Substituted M-Type Sr-Hexaferrites for Enhanced Hard Magnetic Properties

Abstract

This study investigates the structural and magnetic properties of La–Mg co-substituted M-type strontium hexaferrites with the compositions Sr_1−xLa_xFe_12−xMg_xO₁₉ (x = 0, 0.05, 0.1, 0.15), synthesized via a conventional solid-state reaction route. X-ray diffraction confirmed the formation of single-phase M-type hexaferrites up to x = 0.15. The Sr₀.₉La₀.₁Fe₁₁.₉Mg₀.₁O₁₉ composition (x = 0.1) composition exhibited the best magnetic performance, achieving a saturation magnetization, M_S = 71.9 emu/g, remanent magnetization, 4πM_r = 2280 G, and coercivity, H_C = 3549 Oe. These improvements were attributed to the synergistic effect of La³⁺ and Mg²⁺ substitution, which enhanced magnetocrystalline anisotropy and refined grain morphology. Further enhancement was obtained by adding sintering additives (0.8 wt% SiO₂, 0.5 wt% CaO, and 0.7 wt% SrCO₃), resulting in an H_C of 4489 Oe and a 4πM_r of 2285 G. The results demonstrate that La–Mg co-substitution, combined with optimized sintering conditions, provides a cobalt-free strategy for developing high-performance, environmentally benign M-type ferrite magnets for advanced motor and electronic applications.

1. Introduction

M-type hexaferrites, with the general formula AFe₁₂O₁₉ (A = Ba, Sr, or Pb), are among the most important classes of hard magnetic materials due to their high magnetocrystalline anisotropy, large coercivity (H_C), high Curie temperature, and excellent chemical stability [1,2,3]. These properties make them highly attractive for permanent magnet applications such as motors, loudspeakers, microwave devices, and magnetic recording media. Sr or Ba hexaferrites have been widely studied and utilized because of their superior hard magnetic properties and ease of synthesis.

The SrFe₁₂O₁₉ (SrM) compound possesses a crystal structure composed of alternating spinel (S, Fe₆O₈) and hexagonal (R, SrFe₆O₁₁) blocks, arranged in the sequence RSR* S*, where the asterisk denotes a 180° rotation around the hexagonal c-axis [4]. Within one unit cell, twenty-four Fe³⁺ ions occupy five distinct crystallographic sites in the oxygen lattice: one tetrahedral (4f₁), three octahedral (12k, 2a, and 4f₂), and one trigonal bipyramidal (2b) site. The magnetic structure originates from superexchange interactions between Fe³⁺ and O²⁻ ions, resulting in parallel spin alignment along the c-axis for Fe³⁺ ions at the 12k, 2a, and 2b sites, and antiparallel alignment for those at the 4f₁ and 4f₂ sites. Consequently, the net magnetic moment per formula unit of SrM (half of the unit cell) is 20 μ_B at 0 K. However, this theoretical value essentially defines the upper limit of the saturation magnetization (M_S) of SrM. Although SrM exhibits relatively high M_S and strong magnetocrystalline anisotropy, which make it suitable for permanent magnet applications, its intrinsic magnetic properties remain inadequate for advanced applications, particularly in automotive traction motors, high-efficiency BLDC (Blushless DC) motors, and energy-harvesting or power-generation systems.

To address these limitations, extensive studies have focused on modifying the intrinsic magnetic behavior of M-type hexaferrites through the substitution of rare-earth (RE) ions at either the A-site (Ba, Sr) or Fe-site positions [5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20]. Early systematic investigations by Grossinger et al. revealed that RE ions such as La, Ce, and Nd can substantially enhance magnetocrystalline anisotropy and H_C by inducing lattice distortion within the hexagonal magnetoplumbite framework [5]. More recently, Lobo Guerrero et al. confirmed through a comprehensive review that RE substitution (La, Ce, Gd, Sm) tends to increase H_C and thermal stability while simultaneously improving grain homogeneity in both BaM and SrM systems [6]. Furthermore, complementary experiments by Todkar et al. demonstrated that Ce–Dy co-doping in BaM nanoparticles yields remarkably high H_C suitable for microwave absorber applications, illustrating the technological significance of RE incorporation for permanent magnet design [7].

Additionally, several mechanistic studies have clarified how RE and transition-metal co-substitution modify magnetic anisotropy and exchange coupling. Ueda et al. reported that La–Co substitution in SrM enhances the anisotropy field through spin–orbit coupling between Co²⁺ and Fe²⁺ ions [8], while Cao et al. highlighted that partial Ce³⁺ ion substitution in SrM stabilizes the M-type phase but can also lead to secondary Ce-rich phases at higher concentrations [9]. Similarly, Huang et al. revealed that Gd–Zn co-doping in Ba–Sr ferrites improves M_S and suppresses abnormal grain growth during sol–gel synthesis [10]. Furthermore, Islam et al. demonstrated that Cu–Gd co-substitution in SrM increases both H_C and magnetization simultaneously due to synergistic modifications of Fe–O superexchange pathways [11]. Despite these advances, most RE substitution studies have primarily focused on single-site doping or RE–transition-metal co-doping, whereas systematic investigations combining rare-earth and alkaline-earth-modifying cations such as Mg²⁺ remain limited. Therefore, exploring RE–Mg co-substitution offers a promising approach to simultaneously enhance magnetocrystalline anisotropy and refine the microstructure, paving the way for environmentally benign, Co-free hard ferrite magnets optimized for high-performance applications. In addition, Mg²⁺ plays a crucial role in controlling grain growth and refining the microstructure, which is expected to improve the hard magnetic properties further.

In this context, the present study investigates the synthesis and characterization of La–Mg-substituted SrM prepared via a conventional ceramic route. Systematic substitutions at both the Sr and Fe sites were carried out to optimize the balance between remanence and H_C. Furthermore, various sintering additives, including SiO₂, CaO, CaCO₃, Al₂O₃, Cr₂O₃, SrCO₃, and Bi₂O₃, were examined to improve densification and microstructural uniformity. The correlations among substitution composition, processing parameters, microstructure, and the resulting hard magnetic properties are discussed in detail, providing valuable insights for the design of advanced hexaferrite permanent magnets suitable for high-performance applications.

2. Materials and Methods

M-type hexaferrites with the chemical compositions Sr_1−xLa_xFe_12−xMg_xO₁₉ (x = 0, 0.05, 0.1, 0.15) were synthesized via the conventional solid-state reaction method. The precursor and sintering additive powders—Fe₂O₃ (99.5%, industrial use), SiO₂ (99.0%, industrial use), SrCO₃, La₂O₃, MgO, CaO, CaCO₃, Al₂O₃, Cr₂O₃, CuO (all 99.9% purity, Kojundo Chemical Lab. Co., Ltd., Sakado, Japan), and Bi₂O₃ (99.9% purity, Aldrich, St. Louis, MO, USA) —were weighed according to the desired cation ratios. Each powder mixture (50.00 g) was placed in a polypropylene bottle with yttria-stabilized ZrO₂ balls and deionized (DI) water and ball-milled at a rotation speed of 100 rpm for 24 h. Then, the mixed powder was dried in an oven, placed in an alumina crucible, and calcined at a temperature of 1100 °C in air for 4 h. The calcined powders were ground, sieved, ball-milled again for 20 h, and dried. The ball-milled powders were pressed in a disk-shaped mold with a diameter of 30.0 mm at a pressure of approximately 15 MPa to produce green compact pellets. The pelletized samples were sintered in air at a temperature of 1230 °C for 2 h.

The sample density was calculated based on the weight and geometric dimensions of the disk-shaped samples. X-ray diffraction (XRD; D2 Phaser, Bruker, Billerica, MA, USA) was conducted for phase identification using a Cu-Kα radiation source (λ = 0.15406 nm) over a 2θ range of 20–70°, with a step size of 0.0197° and a total acquisition time of 18 min per sample. Field-emission scanning electron microscopy (FE-SEM; JSM-7610F, JEOL, Akishima, Japan) was used for phase identification. The average grain size was determined from the SEM micrographs using the line-intercept method. The magnetic hysteresis curves were measured at room temperature using a vibrating-sample magnetometer (VSM; Lakeshore 7410, Westerville, OH, USA) and a B–H loop tracer (BH-5501, Denshijiki Industry, Tokyo, Japan). For the VSM measurements, rectangular sintered samples (~2.0 mm × ~1.5 mm × ~0.8 mm) were tested under magnetic fields ranging from −23 to 23 kOe. The B–H loop tracer measurements were conducted on disk-shaped samples (diameter ~24 mm, thickness ~6 mm) by applying a maximum field of 20 kOe and then decreasing it to −20 kOe to obtain the demagnetization curves.

3. Results and Discussion

Figure 1 shows the powder XRD patterns of the Sr_1−xLa_xFe_12−xMg_xO₁₉ (x = 0, 0.05, 0.1, 0.15) samples sintered at 1230 °C. The diffraction peaks for the M-type hexaferrites were indexed based on hexagonal magnetoplumbite crystal structures with space group P6₃/mmc. All the diffraction peaks are well matched with the standard data from the International Center for Diffraction Data (ICDD, PDF no. 00-33-1340), and no secondary phases were detected, confirming the formation of a single-phase M-type hexaferrite structure in all samples. This result indicates that La and Mg ions were successfully incorporated into the Sr and Fe sites, respectively. However, no significant XRD peak shift was observed with varying x values; therefore, the variation in lattice parameters was not determined.

Figure 2a–d show the SEM micrographs of Sr_1−xLa_xFe_12−xMg_xO₁₉ (x = 0, 0.05, 0.1, 0.15) samples, illustrating the microstructural evolution with varying La–Mg substitution levels. All samples were sintered at 1230 °C with the addition of 0.5 wt% SiO₂ and 1 wt% CaCO₃ as sintering aids. The grain size distributions of each sample are presented as histograms in Figure 3a–d. In the case of the unsubstituted SrFe₁₂O₁₉ sample (Figure 2a), the average grain size was 1.45 μm. With La–Mg substitutions, the average grain size slightly increased to about 1.5 μm at x = 0.05, while it was refined to around 1.0 μm at higher substitution levels of x = 0.1 and 0.15.

Figure 4a shows the M–H curves of Sr_1−xLa_xFe_12−xMg_xO₁₉ (x = 0, 0.05, 0.1, 0.15) samples. The specimens used for the VSM measurements were rectangular sintered bodies (~2.0 mm × ~1.5 mm × ~0.8 mm), and the magnetization (M) values represent the magnetic moment measured in emu normalized by the sample mass. In the VSM measurements, attention was focused on the intrinsic property of each cation-substituted sample—specifically, the variation in the M_S per unit mass. The compositional dependence of M_S and Hc values with increasing substitution level x is shown in Figure 4b, and the numerical values are summarized in

Table 1. Sample density, saturation magnetization (M_S) and coercivity (H_C) measured by VSM and remanent magnetization (4πM_r) and coecivity (H_C) measured by B-H loop tracer of the Sr_1−xLaFe_12−xMg_xO₁₉ (x = 0, 0.05, 0.1, 0.15) powders after sintering at 1230 °C.

Composition	x	Density (g/cm3)	VSM		B-H Loop Tracer
			MS (emu/g)	HC (Oe)	4πMr (G)	HC (Oe)
Sr1−xLaFe12−xMgxO19	0	4.78	69.71 ± 0.28	2708	2261	3263
	0.05	4.72	70.98 ± 0.28	3185	2282	3289
	0.1	4.77	71.86 ± 0.29	3347	2280	3549
	0.15	4.73	68.20 ± 0.27	3249	2271	3487

.

Both M_S and H_C increased as x increased from 0 to 0.1, but decreased simultaneously at x = 0.15. As confirmed in Figure 1, all four samples consist of a single M-type phase; therefore, the change in M_S can be attributed to variations in the net spin moment of Fe³⁺ ions contributing to M_S. Since La³⁺ substitutes for Sr²⁺ and Mg²⁺ substitutes for Fe³⁺ simultaneously, the valence state of the remaining Fe³⁺ ions is not expected to change. Under this assumption, the enhanced M_S observed at x = 0.1 indicates that the nonmagnetic Mg²⁺ ions preferentially occupy the down-spin Fe³⁺ sites in the M-type hexaferrite structure.

Such an increase in M_S due to selective cation substitution has also been reported for minor La–Zn and Mn–Zn substitutions in M-type hexaferrites. In the case of La–Zn substitution, the highest M_S was obtained at x = 0.3 [13], whereas for Mn–Zn substitution, the maximum M_S occurred at x = 0.1 [14]. However, when the substitution level of nonmagnetic ions exceeds a certain limit, M_S inevitably decreases, as also observed for the La–Mg system, where M_S begins to decrease beyond x = 0.15.

Generally, when M_S increases, the H_C tends to decrease because of the weakening of the magnetic anisotropy. However, in the La–Mg substituted samples, both M_S and H_C increased simultaneously from x = 0.05 to x = 0.1, which is a noteworthy result. Since H_C is also influenced by extrinsic factors such as the average grain size and its distribution, these effects should be considered together. The H_C is known to follow the relationship:

H_C = D_cri/D^1.08

(1)

where D_cri is the critical grain size for a single domain [15]. The fact that the x = 0.05 sample exhibited a higher H_C value despite having a larger average grain size than the unsubstituted sample suggests that the La–Mg substitution enhanced the intrinsic hard magnetic properties, such as the magnetocrystalline anisotropy constant (K_u), of the M-type hexaferrite. Because H_C is proportional to the anisotropy field (H_ani), which is related to M_S by H_ani = 2K_u/M_S, a simultaneous increase without decreasing M_S implies that the K_u increased with La–Mg substitution, resulting in enhanced H_ani and consequently higher H_C.

Figure 5a presents the demagnetization curves of sintered Sr_1−xLa_xFe_12−xMg_xO₁₉ (x = 0, 0.05, 0.1, 0.15) samples, measured using a B–H loop tracer. While the M–H curves in Figure 4 were obtained by a VSM using small specimens (~2.0 mm × ~1.5 mm × ~0.8 mm) under vibration, the demagnetization curves in Figure 5 were acquired from disk-shaped samples with a much larger volume (diameter ~24 mm, thickness ~6 mm). In this measurement, a varying magnetic field was applied to the specimen, and the magnetic flux density change was detected through a J-coil. Thus, the results in Figure 5 more closely represent the actual magnetic performance of isotropic sintered magnets. Figure 5b show the variations in remanent flux density (4πM_r) and H_C as a function of the substitution level x. The trend of 4πM_r with increasing x is consistent with that of the VSM-derived results in Figure 4b. However, while M_S obtained from the VSM represents an intrinsic magnetic property (magnetization per unit mass), 4πM_r is influenced by extrinsic factors such as sample density and crystallographic alignment.

Practically, the 4πM_r value (in units of Gauss) obtained from the B–H loop tracer can be related to the M_S (in emu/g) measured by the VSM through the following equation [20]:

4πM_r = 4π∙M_S∙ρ∙S∙(1 − f_i)

(2)

Here, ρ represents the density of the sintered sample (in g/cm³), and S denotes the grain orientation factor, which takes a value of 1 when all grains are perfectly aligned along the c-axis and 0.5 when they are completely randomly oriented. f_i is the volumetric fraction of nonmagnetic impurities. Since the present samples are randomly oriented isotropic magnets, S value is expected to be close to 0.5 [2]. For the La–Mg substituted samples, in which no secondary phase was detected, the f_i value can be regarded as zero (f_i ≈ 0).

The dependence of H_C on x also follows a similar trend to that shown in Figure 4b,c, although the absolute values differ slightly. Because the B–H loop tracer data are obtained from specimens with a volume roughly 1000 times larger than those used in the VSM, they are considered more representative of the macroscopic magnetic behavior and compositional dependence. The apparent sintered density of the disk samples, along with the 4πM_r and H_C values obtained from the B–H loop tracer measurements, are also presented in Table 1. Considering all results, the Sr₀.₉La₀.₁Fe₁₁.₉Mg₀.₁O₁₉ sample (x = 0.1) exhibited the most superior hard magnetic properties, with the highest M_S of 71.9 emu/g, a remanent flux density of 2280 G, and a H_C of 3549 Oe.

Finally, to further optimize the hard magnetic properties, additional B–H loop tracer measurements were performed on Sr₀.₉La₀.₁Fe₁₁.₉Mg₀.₁O₁₉ samples fabricated using various sintering additives, as shown in Figure 6a. Among the sintering additives, SiO₂ and Al₂O₃ are known to effectively suppress grain growth, while CaCO₃, CaO, SrCO₃, Bi₂O₃, and CuO promote densification and grain coarsening [21,22,23,24]. Figure 6a presents the demagnetization curves for samples prepared with six different combinations of sintering aids. Curve A represents the sample sintered with the original additives, 0.5 wt% SiO₂ and 1.0 wt% CaCO₃. When 0.5 wt% SiO₂ and 0.5 wt% SrCO₃ were additionally introduced, as shown in curve B, both 4πM_r and H_C improved simultaneously. In curve C, 4πM_r slightly decreased compared with B, while H_C increased significantly (when 1 wt% of CaCO₃ was replaced by 0.5 wt% of CaO, and the SiO₂ and SrCO₃ contents were adjusted to 0.8 wt% and 0.7 wt%, respectively). For the samples containing Al₂O₃ (curves D and E), high H_C values were obtained; however, this was accompanied by a significant reduction in 4πM_r. Among them, the sample sintered with 0.8 wt% SiO₂, 0.3 wt% SrCO₃, 0.3 wt% Al₂O₃, and 0.5 wt% CaO exhibited the highest H_C = 4811 Oe (curve E). On the other hand, when Bi₂O₃ or CuO was used instead of SrCO₃, the sintered density increased most significantly, but H_C decreased considerably.

Figure 6b summarizes these results by plotting 4πM_r versus H_C. Since these two parameters generally exhibit a trade-off relationship, samples located toward the upper-right region of the graph are considered to have superior hard magnetic performance. Accordingly, the sample corresponding to curve C, prepared with 0.8 wt% SiO₂, 0.5 wt% CaO, and 0.7 wt% SrCO₃, demonstrated the best isotropic magnetic properties, achieving H_C = 4489 Oe and 4πM_r = 2285 G. Furthermore, if this composition were subjected to magnetic field alignment before sintering—thereby increasing the grain orientation factor S in Equation (2) to approximately 0.95—and further improving the sintered density, it is expected that an anisotropic magnet with 4πM_r ≈ 4500 G could be fabricated, albeit with a slight decrease in H_C. The values of 4πM_r and H_C are also presented in

Table 2. Sintering density, remanent magnetization (4πM_r), and coercivity (H_C) of M-type hexaferrite samples according to sintering additives.

Composition	Sintering Additives (wt%)							Density (g/cm3)	4πMr (G)	HC (Oe)
	SiO2	CaCO3	SrCO3	Al2O3	CaO	Bi2O3	CuO
Sr0.9La0.1 Fe11.9Mg0.1O19	0.5	1	-	-	-	-	-	4.77	2280	3549
	1	1	0.5	-	-	-	-	4.73	2316	3697
	0.8	-	0.7	-	0.5	-	-	4.64	2285	4489
	0.8	-	0.5	0.2	0.5	-	-	4.60	2186	4512
	0.8	-	0.3	0.3	0.5	-	-	4.57	2073	4811
	0.6	0.7	-	-	-	0.3	0.3	4.90	2264	3285

. These results demonstrate a remarkable enhancement in magnetic properties achieved without the use of Co, which is typically considered an essential element for high-performance ferrite fabrication, indicating the great potential of this approach.

4. Summary

La–Mg co-substituted M-type Sr-hexaferrites, Sr_1−xLa_xFe_12−xMg_xO₁₉ (x = 0, 0.05, 0.1, 0.15) were successfully synthesized using a conventional solid-state reaction method, and their structural and magnetic properties were systematically investigated. XRD analysis confirmed that La–Mg substitution yielded single-phase M-type hexaferrites up to x = 0.15. Among all compositions, the Sr₀.₉La₀.₁Fe₁₁.₉Mg₀.₁O₁₉ sample (x = 0.1) exhibited the optimal hard magnetic properties, achieving a high M_S of 71.9 emu/g, a remanent flux density (4πM_r) of 2280 G, and a H_C of 3549 Oe. The observed enhancement in both M_S and H_C can be attributed to the synergistic effect of La³⁺ and Mg²⁺ substitution, which increased M_S and K_U while refining grain morphology. Additional improvements were achieved by employing sintering additives, which played a crucial role in controlling densification and grain growth. In particular, the combination of 0.8 wt% SiO₂, 0.5 wt% CaO, and 0.7 wt% SrCO₃ resulted in the best hard magnetic properties, realizing an H_C of 4489 Oe and a 4πM_r of 2285 G. These findings highlight that the co-substitution of rare-earth and Mg ions, combined with precise control of sintering additives and microstructure, can effectively enhance hard magnetic performance without the use of cobalt. This approach provides a promising and environmentally friendly route for developing next-generation high-performance ferrite magnets for advanced motor, energy, and electronic applications.