Unexpected surface superparamagnetism in antiferromagnetic Cr₂O₃ nanoparticles

Abstract

We report an unexpected superparamagnetic behavior of antiferromagnetic Cr₂O₃ nanoparticles. The Cr₂O₃ particle cores retain their original antiferromagnetic phase, while the surfaces of the particles become superparamagnetic. The X-ray diffraction results confirm that the sample has a corundum structure without any other phases. Through X-ray photoelectron spectroscopy characterization, the particle surfaces present three different oxidation states: Cr³⁺ (antiferromagnetic), Cr⁴⁺ (ferromagnetic), and Cr⁶⁺ (nonferromagnetic). A bimagnetic particle model with Cr³⁺ cores and higher Cr oxidation surface states is used to explain the experimental results. In addition, we observe that spin-flop transitions occur in the antiferromagnetic cores below the Néel temperature (292 K). The spin-flop transition field is uncommon compared with other research, this novel behavior is attributed to the presence of superparamagnetism in the surfaces through the exchange field. These findings reveal the significance of surface states in mediating the magnetic properties in antiferromagnetic materials.

---

1 Introduction

Due to their important applications in exchange bias and spin-valve devices, antiferromagnetic (AF) oxides have been intensively investigated.^1–3 Special types of bimagnetic particles with core–shell structures receive more attention: particles where each CrO₂ particle is enclosed by a Cr₂O₃ layer, which is unexpected for ferromagnetic (FM)/AF core–shell structured particles, have been reported;⁴ the magnetic properties of an AF MnO core and a FM passivation shell (γ-Mn₂O₃ or Mn₃O₄) have been investigated;⁵ the core behaves as a regular antiferromagnet and the shell becomes a two-dimensional diluted antiferromagnet in Co₃O₄ nanowires;⁶ in NiO nanoparticles, the core retains its original AF order while the shell behaves as a spin glass system (Ni³⁺ ions).^7,8

Cr₂O₃, as a typical AF material, has received increased attention because of its abundant magnetic properties. Below the Néel temperature (T_N = 308 K)^4,9,10 in zero magnetic field, the Cr³⁺ spins align antiferromagnetically along the [111] easy axis.⁷ Because of the AF order of the spins, the crystal loses both space- and time-reversal symmetry below T_N, this symmetry reduction leads to that in sufficiently high magnetic fields along the z axis the spins flop into the basal plane maintaining their AF order, which is called spin-flop (SF). SF is a first-order transition that occurs at a critical field (H_SF; H_SF = 5.8 T below 90 K).⁹ Up to now, most groups had observed the SF transition experimentally at sufficiently high magnetic fields in a low temperature range;^10,11 among those, Foner et al. systematically investigated the H_SF of Cr₂O₃ and (Cr₂O₃)_0.9·(Al₂O₃)_0.1 from 4.2 K to about 0.95T_N,¹² which significantly broadened the temperature range where the SF transition happens compared with other experimental or theoretical conclusions.

In this paper, we report an unexpected superparamagnetic (SPM) behavior of AF Cr₂O₃ nanoparticles, in which the SPM signals could even exist above T_N. The magnetic properties of the bimagnetic nanoparticles with AF cores and SPM surface states are studied and the origin of the anomalous magnetic behavior is discussed in detail. More interestingly, the presence of SPM surface state influences the AF core, which means that the behavior of H_SF in the AF phase is unusual compared with other research.^9–12

2 Experimental section

2.1 Synthesis of the Cr₂O₃ nanoparticles

The chemical reagent used as the starting material was analytical grade and was used without any further treatment. Cr₂O₃ nanoparticles were synthesized by thermal decomposition of chromium chloride hexahydrate (CrCl₃·6H₂O). The starting material was sintered at 800 °C for 1 h in air to obtain the sample. The as-prepared Cr₂O₃ nanoparticles were annealed in an oxygen atmosphere (Cr₂O₃–O₂) at 500 °C for 1 h to allow study of the origin of the magnetism.

2.2 Characterization

X-ray diffraction (XRD; X’Pert PRO PHILIPS with CuK_α radiation, Almelo, Netherlands) was employed to study the structure of the sample. The morphology was characterized using scanning electron microscopy (SEM; Hitachi S-4800, Chiyoda-ku, Japan). The microstructure of the sample was investigated using transmission electron microscopy (TEM; Tecnai TMG2F30, FEI) and a high-resolution TEM (HRTEM) microscope equipped with an energy dispersive X-ray spectroscopy (EDX) spectrometer. Measurements of the magnetic properties were performed using a Quantum Design MPMS magnetometer based on a superconducting quantum interference device (SQUID; Quantum Design, Inc., San Diego, USA). X-ray photoelectron spectroscopy (XPS; VG Scientific ESCALAB-210 spectrometer, East Grinstead, UK) was utilized to determine the bonding characteristics and the composition of the nanoparticles.

3 Results and discussion

The XRD pattern of the Cr₂O₃ nanoparticles is shown in Fig. 1. The results indicate that the sample has a corundum structure [rhombohedral lattice system with space group of R 3̄c(167), JCPDS card no. 84-1616, [a = b = 4.97 Å, c = 13.67 Å]] without any other phases. The SEM images (Fig. 2) reveal that the sample is formed of aggregated microplates with a 0.2–0.4 μm thickness and 1–4 μm radius. In order to see the size of the crystallites aggregated into microplates, the sample was measured using TEM after a longtime ultrasonic treatment. As shown in Fig. 3(a), the sample dispersed into small particles after ultrasonic treatment. The particle diameters are distributed in the range of 60–80 nm. On the other hand, using the Scherrer formula for the full width at half-maximum of the main peaks in the XRD pattern, the average crystallite size is calculated to be 68.3 ± 3.9 nm which matches the particle size read from Fig. 3(a) well. From the HRTEM image shown in Fig. 3(b), a clear atom arrangement can be found inside of the particle, and the representative interplanar spacing is 0.25 nm which is equal to the (110) plane of Cr₂O₃. Fig. 3(c) depicts the corresponding EDX results for our sample, only the elements Cr, O, Cu, and C are present. It can be understood that Cu and C are from the carbon membranes which hold the sample during measurement. Besides, no impurity elements were detected.

Fig. 1 XRD pattern of the Cr₂O₃ nanoparticles, together with the theoretical pattern of Cr₂O₃ in a corundum structure [rhombohedral lattice system with space group of R3̄c(167), JCPDS card no. 84-1616].

Fig. 2 (a and b) SEM images of the Cr₂O₃ nanoparticles. The nanoparticles aggregate to form microplates.

Fig. 3 (a) TEM image of the Cr₂O₃ nanoparticles. (b) High resolution TEM image of the Cr₂O₃ nanoparticles, the interplanar spacing of (110) is 0.25 nm. Inset shows the region where the HRTEM image was obtained. (c) Corresponding EDX results for the Cr₂O₃ nanoparticles.

Fig. 4(a) shows the zero-field-cooled (ZFC) and field-cooled (FC) magnetization curves. It is worth noting that two temperature peaks were observed: the first broad one (T_p1) at around 91 K and the second one (T_p2) at about 292 K. When cooling down from 330 K, the ZFC and FC curves overlap closely until a peak at 292 K appears in both curves. Then they follow a similar trend, that is, the susceptibility increases slowly with the decreasing temperature. Below T_p1, the dividable curves reveal that the magnetization reversal is irreversible. We also obtained magnetization curves as a function of applied magnetic field (M–H) at different temperatures [Fig. 4(b)–(e)]. The curves after deducting the AF contribution are shown in Fig. S3.† It can be seen that all the M–H curves show an S-shaped signal in lower fields (<0.5 T), and the curves measured below T_p2 show a nonlinear behavior with the increasing of the magnetization in higher fields (>2 T). The latter curves are characteristic of SF transitions. In order to further study the SF transition, we have obtained the numerical derivative of the M–H curves, and H_SF is defined as the maximum in the derivative curves in the Fig. 4(f)–(i). The SF transition is observed from 20 K to 250 K and H_SF decreases firstly, then increases with the rise of the measurement temperature: the corresponding H_SF is 5.1, 4.3, and 6.1 × 10⁴ Oe at 20, 170, and 250 K, respectively. This phenomenon is different from those reported earlier where the H_SF increased monotonously with the increase of the critical temperature,^10,12 and that will be discussed in detail below.

Fig. 4 (a) ZFC–FC curves at the dc field of 100 Oe in the temperature range of 2–330 K. M–H curves at several temperatures [(b) 20 K, (c) 170 K, (d) 250 K, (e) 300 K]. The magnifications of the central part of the M–H curve are shown in the insets. The SF transition is indicated by the change of curvature in the M–H curves. Numerical derivatives of the M–H curves are shown for (f) 20 K, (g) 170 K, (h) 250 K and (i) 300 K.

The magnetization moments after deducting the AF signal at 300 K [Fig. S3(h)†] yield a clear S-shape dependence of the applied magnetic field. Remarkably, no coercivity was found, which might betoken an appearance of superparamagnetism. In fact, from the magnifications of the central part of the M–H curves obtained at 20 K, below T_p1, [the inset of Fig. 4(b)] and 170 K, between T_p1 and T_p2, [the inset of Fig. 4(c)], it can be clearly seen that coercivity exists at 20 K while it is absent at 170 K. This suggests that T_p1 might be the particle blocking temperature (T_B) of the SPM system. In order to further explore the origin of T_p1 and T_p2, ZFC magnetization curves were recorded under different magnetic fields (Fig. 5). The T_p1 (<150 K) shifts to lower temperature with the increasing magnetic field. The cusp temperature under different magnetic fields obeys the relationship H_p1² ∝ (1 − T_p1), and the curve of T_p vs. H_p [where T_p1 is defined as the maximum in Fig. 5(a)–(c)] is shown in Fig. 5(d), which is as expected for a SPM system.^13–15 These results confirm the above conclusions about superparamagnetism based on the M–H curves at different temperatures. For Fig. 5(e)–(g), T_p2 is defined as the inflection point of each curve, and there is a distinct difference between the plots of T_p2 and T_p1 vs. H_p. The plateau of T_p2 at ∼ 292 K [Fig. 5(h)] is in contrast to the sharp decrease of T_p1 with the decreasing H_p. In most AF systems, the field dependence of the critical phase boundary is very small within the range of the usually accessible experimental field values.¹⁶ Therefore, we can conclude that the T_p2 is the AF phase transition temperature (T_N). The T_N of our AF Cr₂O₃ particles is reduced in comparison with the bulk value (308 K),^4,9,10 which would be a result of the decrease in particle size.^10,16

Fig. 5 Field dependence of the ZFC curves at (a and e) 5 Oe; (b and f) 1000 Oe; and (c and g) 2000 Oe. (d and h) Plots of T_p1 or T_p2 vs. H_p (black circles); red lines show the fitted results.

It is common to observe hysteresis loops and divergence between ZFC/FC curves in low dimension AF materials. However, there is no recognized explanation for solving these problems until now. Different models are built for each AF system. The presence of an AF core and a ferromagnetic passivation shell was proposed during investigation of the magnetic properties of MnO nanoparticles.⁵ It was found that γ-Mn₂O₃ or Mn₃O₄ (depending on the particle size) would be present in the particle shell due to the surface passivation. Benitez et al. attributed the irreversible magnetization of AF Co₃O₄ nanowires to the wire shells which form two-dimensional diluted AF systems.⁶ This model also applies to AF BiFe_0.8Mn_0.2O₃ nanoparticles.¹⁷ The picture is more complicated for NiO systems. Claims such as the presence of Ni³⁺ ions within the AF NiO lattice,¹⁸ AF interaction with canted surface spins,³ or surface spin disorder induced spin-glass behavior rather than SPM¹⁹ were proposed to explain the anomalous magnetic properties in NiO nanoparticles. In our system, an unexpected SPM behavior and spin-flop (SF) transition have been observed. Specifically, the SPM signals could exist above the T_N of Cr₂O₃, which means the magnetic coupling of the SPM cannot be due to superexchange among Cr³⁺–O²⁻–Cr³⁺. What is the origin of superparamagnetism in AF Cr₂O₃ particles? How to generate net moments in this AF system? Many different mechanisms based on the origin of moments have been proposed for the oxide systems: (1) defects bear the brunt of the origin, especially O vacancies: CuO (O vacancies),²⁰ and CoO (O vacancies);²¹ (2) the mixed valences could cause ferromagnetism, such as for CeO₂ (the surface Ce³⁺/Ce⁴⁺ pairs),²² and VO₂ (the valence charge defects with unpaired electrons V⁵⁺ in VO₂ thin films).²³ In order to further identify the origin of the moments in our system, Cr₂O₃–O₂ samples were prepared. The M–H curve of Cr₂O₃–O₂ is shown in Fig. S5† (red line). The hysteresis loops of the as-prepared sample and the sample after annealed in oxygen (Cr₂O₃–O₂) are similar, which indicates that the observed SPM should not be directly related to O vacancies.

The chemical states of the compositional elements in the Cr₂O₃ particles were revealed using XPS. In Fig. 6, the survey spectrum, the indexed peaks only correspond to elements Cr, O, and C, where the binding energies are calibrated using the carbon C 1s peak (285.0 eV). The peak in the Cr 2p_3/2 spectrum (the inset of Fig. 6) is not totally symmetrical, and can be well fitted by three peaks with different binding energy components:²⁴ the dominant peak located at 577.2 eV is assigned to Cr³⁺ ions, and the other binding energy components can match Cr⁴⁺ ions (576.3 eV) and Cr⁶⁺ ions (579.0 eV). It is common to see the presence of a higher oxidation state at the surface of the transition metal oxides which have several stable oxidation states due to surface effects. For example, in epitaxial undoped VO₂ thin films which are grown by pulsed laser deposition, different valence states of the V ions were found at the surface by XPS characterization.²³ The major part of the surface V ions has a higher oxidation state (V⁵⁺) and makes up 47% of the V ions. Another example is AF MnO nanoparticles, the Mn ions at the particle surface are oxidized to a higher valence (Mn³⁺) due to the passivation effect and form a surface passivation shell of γ-Mn₂O₃ or Mn₃O₄.⁵ Similarly, the surface of the as-grown CoO nanoparticles appears oxidized up to Co₃O₄, and the surface effects here are described as a large amount of point defects which adsorb mainly oxygen species.²⁵ Valence states of +2, +3, +4, and +6 are familiar for Cr ions, and our samples were synthesized in the air. Therefore it is possible to obtain higher oxidation states of the Cr ions than “+3” at the particle surfaces. Reddy et al. investigated the magnetic coupling of Cr₂O_n (n = 1–6) and found that the moments at the Cr sites in Cr₂O₃ are antiferromagnetically coupled; the moments at the two Cr sites in CrO₂ are ferromagnetically coupled (T_C ≈ 396 K); while the Cr sites have negligible moments in CrO₃.²⁶ Thus, we further consider that the appearance of SPM in the Cr₂O₃ nanoparticles is owing to the presence of Cr⁴⁺ surface states rather than O vacancies.

Fig. 6 XPS survey spectrum. The inset shows a high resolution scan and fitted results of the Cr 2p_3/2 core level.

As discussed above, the anomalous behavior of the ZFC curve [Fig. 4(a)] could be explained by the formation of a Cr₂O₃ core with a CrO₂ surface state system: the moments (FM CrO₂ surface) of the particles are blocked below the T_B (∼91 K), and SPM occurs above T_B; the kinks at 292 K in the curve must be caused by the order/disorder transition of the AF Cr₂O₃ cores.¹³ Néel predicted that AF materials in fine particle form should exhibit some interesting magnetic properties including superparamagnetism and a weak ferromagnetism, which has been also observed in various AF particles as the particle size decreases. As the particle size decreases, a net magnetic moment is produced due to the nonexact compensation of the two magnetic sublattices, for example, imbalance in the number of “up and down” spins. A SPM susceptibility, due to uncompensated spins, can dominate over the AF contribution itself.^11,27,28 In the synthesized Cr₂O₃ nanoparticles (Fig. 7), we also observed similar results: the cores show regular AF orders, whereas the surface exhibits superparamagnetism from the symmetry breaking by surface Cr⁴⁺ cations (as well, Cr⁶⁺ cations have negligible moments). These results are similar to previous studies on Co₃O₄ nanostructures and MnO nanoparticles, where AF systems are usually governed by core–shell behavior.^5,16

Fig. 7 A schematic diagram (red arrows represent the spins). For the Cr₂O₃ nanoparticles, the core (inner circle) spins are almost antiparallel and in the surface there is a mixture of Cr⁶⁺ clusters with negligible moments and Cr⁴⁺ clusters which are so small that superparamagnetism appears.

The SF phenomenon as a kind of striking nonlinear effect in antiferromagnets is observed when H equals a critical field H_SF given by:^10,12

where H_E is the exchange field in the molecular-field approximation for two AF coupled sublattices, H_A is the anisotropy field, and χ_‖ and χ_⊥ are the parallel and perpendicular susceptibilities respectively. It is interesting that H_SF for our sample is much lower than in other reports,^10,12 which should be related to the appearance of the SPM phase. Bhowmik et al. provided a model for AF orders of spins in bulk and nanoscale samples:²⁷ (1) the long range AF interactions for the perfect bulk sample span up to many particles without any modulation (H_{E core} = H_{E shell}); (2) for nanoparticles, the core spins are almost antiparallel with slight distortion near the shell and deviation of the shell spins from the AF arrangement makes the H_E value become less negative in the shell region. Similarly, in our bimagnetic nanoparticles H_{E core} < H_{E shell} < 0 due to the appearance of surface moments. The smaller absolute value of H_E causes a lower H_SF compared with other Cr₂O₃ bulk systems.

Furthermore, through the molecular-field constant λ and the hypothesis that χ_⊥ ∼ 1/λ when H_E ≫ H_A, eqn (1) transforms into:^10–12

where K is the anisotropy constant which saturates in low temperature ranges;²⁹ the perpendicular susceptibility χ_⊥ is independent of temperature, and the parallel susceptibility χ_‖ decreases as the temperature decreases and vanishes at 0 K.^30,31 As can be seen, the SF transition provides information on the anisotropy, which in an AF system is generally difficult to obtain through other experiments.¹⁰ It is also worth remarking that in all Cr₂O₃ bulk and particle systems H_SF increases when the temperature rises monotonously.^10,12 This dependence is a characteristic of AF systems because the difference (χ_⊥ − χ_‖) decreases faster than the magnetic anisotropy constant K when the temperature increases.^10,30,31 However, it is puzzling that in our system the observed behavior of H_SF(T) is nonmonotonic, especially the increase of H_SF below T_B. At room temperature or any temperature above T_B, the SPM part has sufficient thermal energy to overcome the energy barrier and thus the magnetic spins are free to fluctuate between orientations owning to both applied magnetic field H and the AF core; below T_B, the spins are frozen, or blocked, along their anisotropy axis resulting in a disordered state common to the SPM one.³² In addition, we expect that our observations will become more pronounced with decreasing crystallite size. The unexpected SPM signal is attributed to the presence of Cr⁴⁺ at the particle surface. Just like the anomalous magnetic behavior which has been investigated for other low-dimensional AF systems, it can be expected that the one associated with the surface effect would further enhance with the decreasing scale of the system due to the higher surface to volume ratio.^{3,5,7,19,25,33} Similarly, the AF anisotropy will be weakened with the decreasing particle size, which leads to the magnetization reversal becoming easier under the applied field keeping the antiparallel arrangement of the spins. Then the SF transition might be observed in a lower field.

4 Conclusions

In summary, bimagnetic Cr₂O₃ nanoparticles with a corundum structure were directly synthesized by thermal decomposition. A SF transition in the AF core occurs below T_N (292 K), and the appearance of a SPM surface state is owing to the valence increase (Cr⁴⁺ state). In addition, the SPM surfaces influence the inner AF cores through the exchange field, which leads to the increase of H_SF below T_B. We hope our work can give an insight into the magnetic properties of Cr₂O₃ nanoparticles and provide a path to control magnetic ordering in AF materials.

Acknowledgements

This work is supported by the National Science Fund for Distinguished Young Scholars (grant nos 50925103 and 11034004), the Keygrant Project of Chinese Minisity of Education (grant no. 309027), and NSFC (grant no. 50902065).

References

1. B. Dieny, V. S. Speriosu, S. Metin, S. S. P. Parkin, B. A. Gurney and P. Baumgart, J. Appl. Phys., 1991, 69, 4774.
2. V. Skumryev, S. Stoyanov, Y. Zhang, G. Hadjipanayis, D. Givord and J. Nogues, Nature, 2003, 423, 850–853.
3. J. B. Yi, J. Ding, Y. P. Feng, G. W. Peng, G. M. Chow, Y. Kawazoe, B. H. Liu, J. H. Yin and S. Thongmee, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 224402.
4. R. K. Zheng, H. Liu, R. K. Zheng, H. Liu, Y. Wang and X. X. Zhang, Appl. Phys. Lett., 2004, 84, 702–704.
5. A. López-Ortega, D. Tobia, E. Winkler, I. V. Golosovsky, G. Salazar-Alvarez, S. Estradé, M. Estrader, J. Sort, M. A. González, S. Surinach, J. Arbiol, F. Peiró, R. D. Zysler, M. D. Baró and J. Nogués, J. Am. Chem. Soc., 2010, 132, 9398–9407.
6. M. J. Benitez, O. Petracic, E. L. Salabas, F. Radu, H. Tüysüz, F. Schüth and H. Zabel, Phys. Rev. Lett., 2008, 101, 097206.
7. S. Mandal, S. Banerjee and K. S. R. Menon, Phys. Rev. B: Condens. Matter Mater. Phys., 2009, 80, 214420.
8. M. Ghosh, K. Biswas, A. Sundaresan and C. N. R. Rao, J. Mater. Chem., 2006, 16, 106–111.
9. M. Fiebig, D. Fröhlich and H. J. Thiele, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, R12681–R12684.
10. D. Tobia, E. Winkler, R. D. Zysler, M. Granada and H. E. Troiani, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 78, 104412.
11. R. D. Zysler, D. Fiorani, A. M. Testa, L. Suber, E. Agostinelli and M. Godinho, Phys. Rev. B: Condens. Matter Mater. Phys., 2003, 68, 212408.
12. S. Foner and S. Hou, J. Appl. Phys., 1962, 33, 1289–1290.
13. F. C. Fonseca, G. F. Goya, R. F. Jardim, R. Muccillo, N. L. V. Carreno, E. Longo and E. R. Leite, Phys. Rev. B: Condens. Matter Mater. Phys., 2002, 66, 104406.
14. R. K. Zheng, H. Gu, B. Xu and X. X. Zhang, J. Phys.: Condens. Matter, 2006, 18, 5905–5910.
15. S. K. Sharma, J. M. Vargas, E. D. Biasi, F. Béron, M. Knobel, K. R. Pirota, C. T. Meneses, S. Kumar, C. G. Lee, P. G. Pagliuso and C. Rettori, Nanotechnology, 2010, 21, 035602.
16. M. J. Benitez, O. Petracic, H. Tüysüz, F. Schüth and H. Zabel, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 83, 134424.
17. P. K. Manna, S. M. Yusuf, R. Shukla and A. K. Tyagi, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 83, 184412.
18. I. S. Jacobs and C. P. Bean, in Magnetism, ed. G. T. Rado and H. Suhl, Academic Press, New York, 1963, vol. III, p. 294.
19. S. D. Tiwari and K. P. Rajeev, Phys. Rev. B: Condens. Matter Mater. Phys., 2005, 72, 104433.
20. D. Gao, G. Yang, J. Li, J. Zhang, J. Zhang and D. Xue, J. Phys. Chem. C, 2010, 114, 18347–18351.
21. G. Yang, D. Gao, Z. Shi, Z. Zhang, J. Zhang, J. Zhang and D. Xue, J. Phys. Chem. C, 2010, 114, 21989–21993.
22. M. Li, S. Ge, W. Qiao, L. Zhang, Y. Zuo and S. Yan, Appl. Phys. Lett., 2009, 94, 152511.
23. T.-H. Yang, S. Nori, H. Zhou and J. Narayan, Appl. Phys. Lett., 2009, 95, 102506.
24. G. Z. Xing, J. B. Yi, D. D. Wang, L. Liao, T. Yu, Z. X. Shen, C. H. A. Huan, T. C. Sum, J. Ding and T. Wu, Phys. Rev. B: Condens. Matter Mater. Phys., 2009, 79, 174406.
25. L. Soriano, M. Abbate, A. Fernández, A. R. González-Elipe, F. Sirotti and J. M. Sanz, J. Phys. Chem. B, 1999, 103, 6676–6679.
26. V. Reddy and S. N. Khanna, Phys. Rev. Lett., 1999, 83, 3170–3173.
27. R. N. Bhowmik, R. Nagarajan and R. Ranganathan, Phys. Rev. B: Condens. Matter Mater. Phys., 2004, 69, 054430.
28. S. A. Makhlouf, J. Magn. Magn. Mater., 2004, 272–276, 1530–1532.
29. J. O. Artman, J. C. Murphy and S. Foner, Phys. Rev., 1965, 138, 342–349.
30. T. Nagamiya, Prog. Theor. Phys., 1951, 6, 342–349.
31. S. Foner, Phys. Rev., 1963, 130, 183–197.
32. G. N. Rao, Y. D. Yao and J. W. Chen, IEEE Trans. Magn., 2005, 41, 3409–3411.
33. S. A. Makhlouf, H. Al-Allar and R. H. Kodama, Solid State Commun., 2008, 145, 1–4.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra04009d

---