Sol–gel synthesis and characterization of single-phase CoLa_xFe_2−xO₄ ferrite nanoparticles dispersed in a SiO₂ matrix

Abstract

Single-phase CoLa_xFe_2−xO₄ (x = 0, 0.05, 0.10, 0.15, 0.20) nanoparticles dispersed in a SiO₂ (30 wt%) matrix were synthesized by a sol–gel method. The structure, magnetic properties and cation distribution of the nanocomposites were studied at room temperature. The X-ray diffraction results revealed that the lattice constant of CoLa_xFe_2−xO₄ increased initially and then decreased with increasing La³⁺ concentration x, and the ferrite average grain size was in the range of 26–39 nm, which was close to the particle size observed from transmission electron microscopy images. Field emission scanning electron microscopy analysis indicated the ferrite nanoparticles were almost spherically shaped and homogeneous. Mössbauer spectroscopy measurements suggested all the samples were completely magnetically ordered, and the distribution of cations between tetrahedral and octahedral sites changed with increasing x. The saturation magnetization and coercivity of the nanocomposites determined from vibrating sample magnetometry showed the maximum values of 43.66 emu g⁻¹ at x = 0.05 and 1685.0 Oe at x = 0.10, respectively. The improved magnetic properties will make the nanocomposites useful for applications in high density magnetic recording media.

---

1. Introduction

The spinel ferrites are technologically an important class of magnetic oxides due to their versatile magnetic and electrical properties.^1–5 Among the spinel ferrites, CoFe₂O₄ has been regarded as one of the competitive candidates for high density magnetic recording media because of its moderate saturation magnetization, high coercivity, mechanical hardness and chemical stability.^6–9 Magnetic properties of ferrites can be suitably tailored by varying the composition of the cations.^10–13 Due to rare earth elements (RE) presenting large ionic radii, their substitution into the spinel structure may determine a change in cell symmetry and thus generate internal stress. As a consequence, not only the structural properties are changed, but also the magnetic properties of ferrites.^14,15 Ashiq et al. prepared CoHo_xFe_2−xO₄ ferrites by the co-precipitation technique and found the coercivity increased with increasing Ho³⁺ concentration.¹⁶ Chin et al. synthesized CoM_xFe_2−xO₄ (where M = Gd and Pr and x = 0, 0.1 and 0.2) powders by a citrate precursor technique, and concluded that the substitution of RE in the cobalt ferrite materials annealed in the temperature range of 800–900 °C might improve magnetic properties.¹⁷ Moreover, the effects of Nd³⁺, Eu³⁺and Gd³⁺ substitution at a cobalt site on the magnetic and structural properties of CoFe₂O₄ synthesized by co-precipitation were studied by Shokrollahi et al.¹⁸

The aim of present investigation is to enhance the coercivity of CoFe₂O₄ to make it useful for applications in high density magnetic recording media. For this purpose the rare earth ion La³⁺ is substituted instead of iron in cobalt ferrites. Difference with Ho³⁺, Gd³⁺, Pr³⁺, Nd³⁺, Eu³⁺ and Gd³⁺, La³⁺ is non-magnetic rare earth cation as it has no 4f electrons.^19,20 Moreover, to make the magnetic particles have reduced size and uniform distribution, we put the ferrite nanoparticles disperse in silica matrix to synthesize CoLa_xFe_2−xO₄/SiO₂ nanocomposites.^21,22 Sol–gel process offers some advantages in making silica composite materials containing highly dispersed magnetic nanoparticles, but it normally provides multi-phased nanoparticles in the nano-composites.^23,24 The keys to get single phase ferrite nanoparticles dispersed in SiO₂ matrix depend on the content of silica and annealing temperature. When the mass ratio of ferrite and SiO₂ was 7 : 3 and the annealing temperature was in the range of 800–1200 °C, the single-phase ferrite nanoparticles dispersed in SiO₂ matrix were synthesized in our previous works.^25–27

In this paper, we prepared CoLa_xFe_2−xO₄/SiO₂ nanocomposites with 30 wt% of SiO₂ using sol–gel method and the effects of La³⁺ concentration on the structural, magnetic properties and cation distribution of the nanocomposites are investigated by using X-ray diffraction (XRD), field emission scanning electron microscopy (FE-SEM), transmission electron microscopy (TEM), infrared spectrum (IR), vibrating sample magnetometer (VSM), and Mössbauer spectroscopy (MS) at room temperature.

2. Experimental section

2.1. Synthesis

CoLa_xFe_2−xO₄/SiO₂ (x = 0, 0.05, 0.10, 0.15, 0.20) nanocomposites that the mass ratio of ferrite and SiO₂ being 7 : 3 were prepared by sol–gel method. A solution was obtained from the mixture of analytical grade Fe(NO₃)₃·9H₂O, Co(NO₃)₂·6H₂O, La(NO₃)₂·6H₂O, tetraethylorthosilicate (TEOS), deionized water, nitrate, ethylene glycol and methoxyethanol. The solution was continuously stirred for 5 h and placed at room temperature for 24 h to form a gel. The gel was evaporated on a water bath at 60 °C for 12 h, and then was dried at 100 °C in drying oven for 24 h to form the xerogel. At last, the xerogels were annealed at 1050 °C for 2 h under air and cooled slowly in the furnace. The chemical reaction is shown in Fig. 1.

Fig. 1 Chemical reaction of CoFe_2−xLa_xO₄/SiO₂ synthesized by sol–gel method.

2.2. Characterization

The structures of CoLa_xFe_2−xO₄/SiO₂ nanocomposite were characterized by XRD (Rigaku D/max-2500/PC). The microstructural and compositional analyses were performed using FE-SEM-EDAX (JEOL JSE-7800F, Oxford SDD EDS). The morphology and size of the nanoparticles were measured by TEM (JEM-2100HR). IR spectra were recorded as KBr pellets in the range of 4000–400 cm⁻¹ using a Perkin-Elmer Frontier FT-IR. The magnetic properties of the samples were measured using VSM (Lake Shore 7407) with a maximum applied field of 20 kOe. The ⁵⁷Fe Mössbauer spectra were collected on a FAST Comtec Mössbauer systems at room temperature, using a ⁵⁷Co(Pd) source and a constant acceleration mode. All isomer shifts were given relative to that of α-Fe at room temperature. The spectra were fitted with Lorentzian lines via the least squares method.

3. Results and discussion

Fig. 2 shows the XRD patterns of CoLa_xFe_2−xO₄/SiO₂ nanocomposites. For all the samples the peaks were corresponded to cubic spinel structure. No any trace of impurity peaks were observed, which confirmed La³⁺ ions had been incorporated into the spinel lattice. Otherwise, there were no characteristic peaks from SiO₂ matrix, which suggested SiO₂ remained amorphous nature in the nanocomposites, and its content was not enough to show a hump around 23°.²⁸

Fig. 2 XRD patterns of CoFe_2−xLa_xO₄/SiO₂ nanocomposite.

Fig. 3 gives the variations of the lattice constant (a) and average crystallite size (d) of CoLa_xFe_2−xO₄ in CoLa_xFe_2−xO₄/SiO₂ as functions of La³⁺ concentration x. It could be seen that a and d increased with increasing x, and reached the maximum values of 0.8413 nm and 39 nm when x was 0.10, respectively. For the samples with x ≤ 0.10, the increase in a is attributed to the replacement of the smaller ionic crystal radius of Fe³⁺ (0.064 nm) by the larger La³⁺ (0.106 nm).²⁹ With increasing of x, more and more La³⁺ will reside at the grain boundaries.³⁰ Table 1 gives the EDAX data analysis of atomic percentage at the grain and grain boundaries for the samples with x = 0.10 and 0.15. The concentration of La element in the grain boundaries was slightly smaller than that in the grains for x = 0.10. However, results was just the opposite for x = 0.15. In other words, the La³⁺ ions in the present ferrite system tended to concentrate in the grain boundaries when x was above 0.15, and they can not only hinder the grain growth, and may exert a pressure on the grains which leads the lattice constant to decrease for the samples with x ≥ 0.15. The high percentage of Si and O elements was due to the measurement completed through putting the samples on oxidized Si (100) substrates.

Fig. 3 Variations of the lattice constant a and average grain sizes d of CoFe_2−xLa_xO₄ in CoFe_2−xLa_xO₄/SiO₂ as functions of La³⁺ concentration x.

Table 1 The EDAX data analysis of atomic percentage at the grain and grain boundaries for the sample x = 0.10 and 0.15

x		Si	O	Fe	Co	La
0.10	Grain	31.00	56.67	3.09	1.74	0.11
	Grain boundary	32.20	53.09	2.44	1.22	0.08
0.15	Grain	29.30	51.96	3.68	2.07	0.12
	Grain boundary	29.72	51.04	2.97	1.73	0.26

---

FE-SEM images of the samples are shown in Fig. 4(a)–(e). Sizes of the ferrite particles were almost uniform and they were spherical in shape. The EDAX patterns of the samples are shown in Fig. 4(f)–(j). The data analyses of the mass percentage are listed in Table 2. It showed the samples did not contain any impurity elements, and they corresponded to the composition of CoLa_xFe_2−xO₄/SiO₂ (x = 0, 0.05, 0.10, 0.15, 0.20) nanocomposite with 30 wt% of SiO₂. The present of Pt elements was due to the spraying gold platinum on the surface of the samples to make them conducting.

Fig. 4 (a)–(e) The FE-SEM images of samples x = 0, 0.05, 0.1, 0.15, 0.20, (f)–(j) The typical EDAX patterns of the samples x = 0, 0.05, 0.1, 0.15, 0.20, respectively.

Table 2 The EDAX data analysis of the mass percentage for CoLa_xFe_2−xO₄/SiO₂ nanocomposites

x	Fe (%)	Co (%)	La (%)	Si (%)	O (%)	C (%)	Pt (%)
0	30.96	15.31	0	11.84	31.20	3.14	7.55
0.05	28.32	14.97	2.61	12.27	34.448	2.86	4.49
0.10	27.54	15.08	3.85	12.51	33.99	3.33	3.70
0.15	22.45	14.09	5.09	11.75	28.89	5.18	12.56
0.2	24.57	14.18	6.35	12.43	31.32	3.46	7.69

---

Fig. 5(a)–(c) show TEM images of samples with x = 0, 0.10 and 0.20. Most of the nanoparticles were spherical in shape and homogeneously dispersed in the SiO₂, there was not obvious agglomerate. The silica matrix in the nanocomposite to serve as spatial nucleation sites for CoLa_xFe_2−xO₄, not only to confine the coarsening of nanoparticles but also to minimize the degree of crystallite aggregation.³¹ The lattice fringe in HRTEM image (Fig. 5(d)) for the sample with x = 0.10 showed an inter-planar spacing of approximately 0.255 nm corresponding to (311) crystal plane of the spinel phase. Average particle sizes calculated from the TEM images for the five samples were 27, 40, 42, 35 and 29 nm, respectively. Values were almost comparable with the average grain size obtained from XRD. The histogram of particle size distribution drawn for x = 0.10 are shown in Fig. 5(e).

Fig. 5 (a)–(c) The TEM images of the samples x = 0, 0.1, 0.2, (d) the HREM image and (e) the particle size histogram of the sample x = 0.1.

Fig. 6 shows the IR spectra of CoFe_2−xLa_xO₄/SiO₂ nanocomposite in the range 400–4000 cm⁻¹. The band (ν₁) at about 593 cm⁻¹ was attributing to the metal–oxygen stretching vibration of unit cell of the spinel in the tetrahedral A sites.³² There was a very slight change in the frequency of ν₁, which confirmed the cations distribution in A sites were variation with increasing x. The bands at 467, 804 and 1082 were corresponding to the silica matrix. There was no signature of Si–O–Fe bond vibration at about 950 cm⁻¹ which was reflected the coupling between silica matrix and iron.³³ The pronounced bands of ferrite and silica matrix indicated completed formation of single-phase of CoLa_xFe_2−xO₄ ferrite nanoparticles dispersed in SiO₂ matrix. The two bands at about 3437 and 1632 cm⁻¹ were ascribed to the stretching modes of O–H group and H–O–H bending vibration of the free and absorbed water.

Fig. 6 IR spectra of CoFe_2−xLa_xO₄/SiO₂ nanocomposite.

The Mössbauer spectra of CoLa_xFe_2−xO₄/SiO₂ nanocomposites are shown in Fig. 7. The fitted Mössbauer parameters and cation distribution are listed in Table 3. All spectra for the nanocomposites were fitted with two Zeeman sextets, which indicated that the samples were completely magnetic order. In Table 3, the isomer shift (IS) value varied in the range of 0.243–0.400 mm s⁻¹, consistent with that of the high spin Fe³⁺ state. The component A and B corresponded to the Fe³⁺ on the A sites and the octahedron B sites, respectively. The quadrupole splitting (QS) value was increasing with the increase of x, which was due to the doped La³⁺ ions leading to structural distortion and promoting the unsymmetric structure of the nanoparticles.³⁴

Fig. 7 Mössbauer spectra of CoFe_2−xLa_xO₄/SiO₂ nanocomposite.

Table 3 Mössbauer parameters and cation distribution for CoLa_xFe_2−xO₄/SiO₂ nanocomposites^a

x		IS (mm s^−1)	QS (mm s^−1)	Hf (kOe)	S (%)	Cation distribution
a Note: IS, QS, Hf, and S represent the isomer shift, the quadrupole splitting, the hyperfine field, and the relative absorption area, respectively.
0	A	0.262	0.018	487.8	37.1	(Co0.258Fe0.742)A[Co0.742Fe1.258]O4
	B	0.399	0.022	494.8	62.9
0.05	A	0.261	0.019	488.1	41.2	(Co0.197Fe0.803)A[Co0.803Fe1.147La0.050]O4
	B	0.400	0.028	495.0	58.8
0.10	A	0.263	0.022	487.4	41.4	(Co0.213Fe0.787)A[Co0.787Fe1.113La0.100]O4
	B	0.398	0.034	494.1	58.6
0.15	A	0.243	0.024	485.1	39.9	(Co0.262Fe0.738)A[Co0.738Fe1.112La0.150]O4
	B	0.384	0.036	491.4	60.1
0.20	A	0.260	0.024	484.9	37.9	(Co0.318Fe0.682)A[Co0.682Fe1.118La0.200]O4
	B	0.370	0.038	490.5	62.1

---

The relative absorption area of the two Zeeman sextets reflects the content of Fe³⁺ ions on the A and B sites, and the cation distribution in CoLa_xFe_2−xO₄ is written as (Co_αFe_1−α)_A[Co_1−αFe_1+α−xLa_x]_BO₄. Thus, the absorption area ratio (S) of A to B sites, S_A/S_B, based on the above distribution are written as

where f_A and f_B represents the recoil-free fractions for A and B sites Fe³⁺, respectively. In the present work, we assumed that f_A was equal to f_B.³⁵ Obtained cation distribution is listed in the 7th column of Table 3. It could be seen that the CoFe₂O₄ nanoparticles was partially inverted spinel structure. Co²⁺ ions were forced to migrate from A sites to B sites, and then got back to A sites with increasing x. The migration of Co²⁺ ions would influence directly the coercivity (H_c) of the nanocomposites. In addition, as the net magnetic moments in the ferrite materials depend on the number of magnetic ions occupying the A sites and B sites, the reduction of magnetization is clear for the substitution of magnetic Fe³⁺ by non-magnetic La³⁺ on the B sites. So the weighted average of the hyperfine field (H_f) decreased from 492.2 kOe for x = 0 to 488.3 kOe for x = 0.20.

The hysteresis loops of CoLa_xFe_2−xO₄/SiO₂ nanocomposites are shown in Fig. 8(a). At high field, magnetization increased almost linearly with the external field and showed a lack of saturation at a field as high as 20 kOe. The values of the saturation magnetization (M_S) were obtained by extrapolation to infinite field in an M vs. 1/H² plot (see Fig. 8(b)).³⁶

Fig. 8 (a) Hysteresis loops of CoFe_2−xLa_xO₄/SiO₂ nanocomposites, (b) variations of M with 1/H².

Fig. 9 gives the variation of M_S of CoLa_xFe_2−xO₄/SiO₂ as a function of x. M_S increased firstly and then decreased with increasing x, showing a maximum value of 43.66 emu g⁻¹ at x = 0.05. The decrease in magnetization from the change of cation distribution (in Table 3) is obvious. However, the variations in the size and surface effects of the nano-materials may complicate the above arguments.³⁷ The fact was evident as we observed that the larger saturation magnetization values for the samples with x = 0.05 and 0.10 had larger grain sizes compared with x = 0. In addition, the values of M_S in present are comparable with the results of Kumar et al.³⁸

Fig. 9 Variations of the saturation magnetization M_S and coercivity H_c of CoFe_2−xLa_xO₄/SiO₂ as functions of La³⁺ concentration x.

The variation of H_c of CoLa_xFe_2−xO₄/SiO₂ as a function of x also been showed in Fig. 9. The value of H_c increased firstly and then decreases, showing a maximum value of 1685 Oe at x = 0.10. This behavior can be explained by Brown's relation³⁹ given by

where K₁ is the magnetic crystalline anisotropy constant and μ₀ is the permeability of the free space. The improved coercivity for the samples with x = 0.05 and 0.10 is attributed to the increase of K₁ which is related to the high concentration of Co²⁺ at the B sites (in Table 3). As for the reason why the increase in H_c for x = 0.05 being not obvious is that there is also a significantly increase in M_S. The values of H_c in present are larger than those of Kim et al.,⁴⁰ Kar et al.⁴¹ and Kumar et al.,⁴² which is due to the interfacial diffusion of CoLa_xFe_2−xO₄ layer and SiO₂.

4. Conclusions

In conclusion, single-phase CoLa_xFe_2−xO₄ nanoparticles dispersed in SiO₂ (30 wt%) matrix were synthesized successfully by sol–gel method. When x increased from 0 to 0.20, the lattice constant and the average grain size of CoLa_xFe_2−xO₄ in the nanocomposite showed the maximum values at x = 0.10, respectively. It was due to more and more La³⁺ ions resided at the grain boundaries when x was above 0.15. Mössbauer spectroscopy measurements indicated that all the samples were completely magnetic order, and Co²⁺ ions were forced to migrate from A sites to B sites, and then got back to A sites with increasing La³⁺ content. The saturation magnetization and coercivity of the nanocomposites showed the maximum values of 43.66 emu g⁻¹ at x = 0.05 and 1685.0 Oe at x = 0.10, respectively. The change in the saturation magnetization was attributed to the cation distribution, size and surface effects of the nanoparticles. The improved coercivity depended on the variations of the magnetic crystalline anisotropy and the saturation magnetization with increasing La³⁺ content. The moderate saturation magnetization and high coercivity will make CoLa_xFe_2−xO₄/SiO₂ nanocomposites useful for applications in high density magnetic recording media.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Nos. 21371071 and 11504132).

References

1. Y. S. Yu, A. Mendoza-Garcia, B. Ning and S. H. Sun, Adv. Mater., 2013, 25, 3090.
2. A. T. Nelson, J. T. White, D. A. Andersson, J. A. Aguiar, K. J. McClellan, D. D. Byler, M. P. Short and C. R. Stanek, J. Am. Ceram. Soc., 2014, 97, 1559.
3. J. Jacob and M. A. Kuadar, J. Appl. Phys., 2010, 107, 114310.
4. J. Zhang, J. Fu, G. Tan, F. Li, C. Luo, J. Zhao, E. Xie, D. Xue, H. Zhang, N. J. Mellors and Y. Peng, Nanoscale, 2012, 4, 2754.
5. Y. B. Yong and K. Seung-Yeop, Dalton Trans., 2011, 40, 9989.
6. J. Blasco, G. Subías, J. García, C. Popescu and V. Cuartero, J. Solid State Chem., 2015, 221, 173.
7. T. Gaudisson, M. Artus, U. Acevedo, F. Herbst, S. Nowak, R. Valenzuela and S. Ammar, J. Magn. Magn. Mater., 2014, 370, 87.
8. S. R. Naik and A. V. Salker, J. Mater. Chem., 2012, 22, 2740.
9. M. Mustaqima, P. Yoo, W. Huang, W. L. Bo and C. Liu, Nanoscale Res. Lett., 2015, 10, 1.
10. D. S. Nikam, S. V. Jadhav, V. M. Khot, R. A. Bohara, C. K. Hong, S. S. Mali and S. H. Pawar, RSC Adv., 2015, 5, 2338.
11. G. Mustafa, M. U. Islam, W. Zhang, Y. Jamil, A. W. Anwar, M. Hussain and M. Ahmad, J. Alloys Compd., 2015, 618, 428.
12. L. Wang, M. Lu, Y. Liu, J. Li, M. Liu and H. B. Li, Ceram. Int., 2015, 41, 4176.
13. C. Murugesan and G. Chandrasekaran, RSC Adv., 2015, 5, 73714.
14. S. E. Shirsath, R. H. Kadam, S. M. Patange, M. L. Mane, A. Ghasemi and A. Morisako, Appl. Phys. Lett., 2012, 100, 042407.
15. H. S. Mund, J. Sahariya, R. J. Choudhary, D. M. Phase, A. Dashora, M. Itou, Y. Sakural and B. L. Ahuja, Appl. Phys. Lett., 2013, 102, 232403.
16. I. Ali, M. U. Islam, M. Ishaque, H. M. Khan, M. N. Ashiq and M. U. Rana, J. Magn. Magn. Mater., 2012, 324, 3773.
17. R. N. Panda, J. C. Shih and T. S. Chin, J. Magn. Magn. Mater., 2003, 257, 79.
18. S. Amiri and H. Shokrollahi, J. Magn. Magn. Mater., 2013, 345, 18.
19. M. A. Ahmed, N. Okasha and M. Gabal, Mater. Chem. Phys., 2004, 83, 107.
20. L. Ben Tahar, M. Artus, S. Ammar, L. S. Smiri, F. Herbst, M.-J. Vaulay, V. Richard, J.-M. Grenèche, F. Villain and F. Fiévet, J. Magn. Magn. Mater., 2008, 320, 3242.
21. N. Rajic, M. Ceh, R. Gabrovsek and V. Kaucic, J. Am. Ceram. Soc., 2002, 86, 1719.
22. K. Wang, Y. Sun, Y. Li and X. He, J. Alloys Compd., 2010, 494, 236.
23. K. H. Wu, T. H. Ting, M. C. Li and W. D. Ho, J. Magn. Magn. Mater., 2006, 298, 25.
24. K. H. Wu, W. C. Huang, C. C. Yang and J. S. Hsu, Mater. Res. Bull., 2005, 40, 239.
25. J. M. Li, L. Wang, Z. H. Bai, J. Li, J. B. Lu and H. B. Li, J. Magn. Magn. Mater., 2014, 354, 190.
26. L. Wang, J. Li, M. Liu, Y. M. Zhang, J. B. Lu and H. B. Li, J. Magn. Magn. Mater., 2012, 324, 4200.
27. L. Wang, J. Li, M. Liu, H. Dong, J. Hua, S. C. Xu and H. B. Li, J. Supercond. Novel Magn., 2015, 28, 191.
28. R. Liu, X. Q. Shen, C. T. Jiang, F. Z. Song and H. X. Li, J. Alloys Compd., 2012, 511, 163–168.
29. X. H. Ren and G. L. Xu, J. Magn. Magn. Mater., 2014, 354, 44.
30. M. M. Eltabey, W. R. Agami and H. T. Mohsen, J. Adv. Res., 2014, 5, 601.
31. K. Praveena, K. Sadhana and S. R. Murthy, J. Magn. Magn. Mater., 2011, 323, 2122.
32. R. D. Waldron, Phys. Rev., 1955, 99, 1727.
33. K. Nadeem, T. Traussnig, I. Letofsky-Papst, H. Krenn, U. Brossmann and R. Würschum, J. Alloys Compd., 2010, 493, 385.
34. S. M. Masoudpanah, S. A. S. Ebrahimi, M. Derakhshani and S. M. Mirkazemi, J. Magn. Magn. Mater., 2014, 370, 122.
35. L. Wang, J. B. Lu, J. Li, J. Hua, M. Liu, Y. M. Zhang and H. B. Li, Phys. B, 2013, 421, 8.
36. L. Kumar and M. Kar, J. Magn. Magn. Mater., 2011, 323, 2042.
37. A. T. Ngo, P. Bonville and M. P. Pileni, J. Appl. Phys., 2001, 89, 3370.
38. P. Kumar, S. K. Sharma, M. Knobel and M. Singh, J. Alloys Compd., 2010, 508, 115.
39. J. M. D. Coey, in Rare Earth Permenant Magnetixm, John Wiley and Sons, New York, 1996, p. 220.
40. W. C. Kima, S. J. Kima, J. C. Surb and C. S. Kim, J. Magn. Magn. Mater., 2002, 242–245, 197.
41. L. Kumar and M. Kar, Ceram. Int., 2012, 38, 4771.
42. P. Kumar, S. K. Sharma, M. Knobel and M. Singh, J. Alloys Compd., 2010, 508, 115.

---