Observation of exchange bias below incommensurate antiferromagnetic (ICAFM) to canted A-type antiferromagnetic (cAAFM) transition in nanocrystalline orthorhombic EuMnO₃

Abstract

Detailed ac and dc magnetic properties of orthorhombic EuMnO₃ were studied using a nanocrystalline powder sample. Orthorhombic EuMnO₃ nanoparticles of 45 nm average size were synthesised using a hydrothermal method. Zero field cooled (ZFC) and field cooled (FC) magnetisation shows anomalies at ∼53, 44 and 24 K attributed to paramagnetic (PM) to an incommensurate antiferromagnetic phase (ICAFM), ICAFM to a canted A-type antiferromagnetic order (cAAFM) and spin reorientation transitions, respectively. No indication of a PM to ICAFM transition was observed in either the real or imaginary part of ac magnetisation. The real part of the ac magnetisation curve showed a peak at ∼24.7 K and anomalies at ∼46.8 and 21.8 K were observed in the imaginary part of ac magnetisation. Reduction in the width of the thermal hysteresis between the field cooling and heating magnetisation curve was observed with the nanocrystallites (∼1 K) compared with the bulk phase (∼3 K) EuMnO₃. Isothermal magnetisation showed a large value of the exchange bias (2799 Oe at 3 K) below the ICAFM to cAAFM transition temperature.

---

Introduction

The complex interlinks among lattice distortions, magnetism, dielectric and transport properties of RMnO₃ compounds makes them interesting materials to study.^1–3 Orthorhombic rare-earth manganites, RMnO₃ with R = Nd, Sm, Eu, Gd, Tb and Dy are interesting materials for study because changing the ionic radius of rare earth ions can tune the Mn magnetic structure, which in turn controls its magnetoelectric coupling.^4,5 The two extremes of magnetoelectric coupling in orthorhombic RMnO₃ are shown with EuMnO₃, which is nonmagnetoelectric, and GdMnO₃, which is magnetoelectric.⁶ In orthorhombic RMnO₃ multiferroics, ferroelectricity is attributed to a complex spiral spin order that breaks the inversion symmetry.^7,8 Therefore, it is important to investigate the temperature dependent magnetism of undoped RMnO₃ compounds in order to understand its relation to the dielectric and transport properties.

EuMnO₃ crystallises in an orthorhombic perovskite structure with Pbnm symmetry.⁹ At room temperature, it is paraelectric and PM. It undergoes a phase transition from PM to ICAFM at ∼50 K.⁵ It undergoes a further transition from ICAFM to cAAFM ordering at ∼44 K.¹⁰ Due to this cAAFM ordering, it shows ferromagnetic-like ordering below this temperature. At ∼24 K, it shows a spin reorientation transition.¹⁰ It shows a thermal hysteresis in a field cooled heating and cooling magnetisation curve.¹⁰ Detailed study of the thermal hysteresis has not been done so far. Studies of the thermal hysteresis and the transitions are needed in order to understand the origin of the ferromagnetism. Low temperature isothermal magnetisation has also not been studied so far in EuMnO₃, which could further give insight into these transitions.

Nanoparticles exhibit significant differences in their physical and chemical properties from their bulk counterparts due to their large surface to volume ratio.^11,12 Therefore, it is interesting to explore the corresponding effects like multiferroic properties in the nanoparticle form in order to integrate them into nano-electronics and other applications with potential benefits. Despite the great interest to study them in the nano-regime, there are no reports on the synthesis and magnetic properties of EuMnO₃ nanoparticles. In this paper, we are reporting for the first time the synthesis and magnetic properties of EuMnO₃ nanoparticles. Moreover, we are showing a detailed study on the thermal hysteresis and transitions. Isothermal magnetisation shows a large value of the exchange bias below the ICAFM to cAAFM transition temperature. The possible reason for the exchange bias is also studied.

Experimental section

EuMnO₃ nanoparticles were synthesised by a modified hydrothermal method.^13–16 In a typical reaction, stoichiometric amounts of europium(III) nitrate pentahydrate (Eu(NO₃)₃·5H₂O, Sigma-Aldrich, 99.9% purity) and manganese(II) nitrate x hydrate (Mn(NO₃)₂·xH₂O, Sigma-Aldrich, 99.9% purity) and an equal molar ratio of citric acid (C₆H₈O₇, metal/citric acid molar ratio = 1/1, Merck, 99.5% purity) were dissolved in deionized water. This solution was stirred for 6 h, followed by dropwise addition of ammonia solution (28 wt%) to neutralize the unreacted citric acid as well as to raise the pH value of the solution to near 9.2, resulting in the formation of a sol. The sol was stirred for 3 h and transferred into an 80 mL capacity autoclave with a Teflon liner. After the hydrothermal treatment at 180 °C for 48 h, the precipitate was in turn filtered, washed with deionized water and dried in a vacuum oven. Finally, the powder sample was calcined at 750 °C for 6 h and used for further study.

Characterization

An X-ray diffraction pattern (XRD) of EuMnO₃ was obtained by a PANalytical X'PERT PRO instrument using iron-filtered Cu K_α radiation (λ = 1.5406 Å) in the 2θ range of 10–80° with a step size of 0.02°. The morphology of the as-synthesized material was studied using an FEI (model Tecnai F30) high-resolution transmission electron microscope (HRTEM) equipped with a field emission source operating at 300 kV to image the EuMnO₃ nanocrystals on carbon-coated copper TEM grids. DC magnetisation versus temperature and magnetic field versus magnetization measurements were performed using a Physical Property Measurement System (PPMS) from Quantum Design Inc., San Diego, CA., equipped with a 9 T superconducting magnet and a vibrating sample magnetometer operating at 40 Hz. Temperature dependent magnetization (M)–magnetic field (H) loops were collected in a field sweep from −50 to +50 kOe at a rate of 75 Oe s⁻¹ by first demagnetizing the sample by heating it at 300 K in zero field before cooling it to the desired temperature. Magnetization versus temperature measurements in various field conditions in a broad temperature range from 3 to 300 K with cooling and heating rates of 2 K min⁻¹ were recorded. The ac magnetic measurements were performed using an AC Susceptibility & DC Magnetization Option (ACMS) from Quantum Design Inc., San Diego, CA. All the ac magnetic measurements were carried out with 5 Oe applied field.

Results and discussion

The X-ray diffraction pattern of the as-synthesised powder with the reference data from orthorhombic (JCPDS file no. 261126) phases of bulk EuMnO₃ is shown in Fig. 1. The XRD pattern matches nicely with the orthorhombic EuMnO₃ with Pbnm symmetry. A small variation in the relative peak intensity due to shape anisotropy was observed. The diffraction peaks are quite broad, indicating the formation of EuMnO₃ nanocrystallites. TEM images at different magnifications are shown in Fig. 2. The image in Fig. 2A suggests that the particles have a plate-like morphology. The average particle size is ∼45 nm. Fig. 2B shows the lattice fringes of the (202) plane of orthorhombic EuMnO₃ nanoparticles, which further confirm the phase and crystallinity of the sample.

Fig. 1 X-ray diffraction pattern of EuMnO₃ nanoparticles with the data from JCPDS card no. 261126 for the orthorhombic phase with Pbnm symmetry. Some peaks are not indexed for maintaining clarity of the figure. Inset shows the diffraction pattern from a single particle, indicating the formation of single crystalline orthorhombic EuMnO₃ crystals.

Fig. 2 (A) Transmission electron micrograph and (B) HRTEM image of orthorhombic EuMnO₃ nanocrystals.

Fig. 3 shows the temperature dependence of magnetisation in ZFC and FC modes at 500 Oe and 10 000 Oe applied magnetic field. At 500 Oe applied field, the ZFC and FC magnetisation curves show bifurcation at 53 K. The ZFC magnetisation curve shows two more anomalies as humps at 44.8 and 24.7 K. At 10 000 Oe applied field, the ZFC and FC magnetisation curves show bifurcation at 43 K. The ZFC curve shows a broad hump at 13 K. Above 100 K, Curie–Weiss fitting of the FC curve gave a Weiss temperature of −40 K and Curie constant, C = 4.5 K emu mol⁻¹. The Weiss temperature is greater than the reported value of −88 K for a bulk ceramic sample.¹⁰ Below bifurcation of the ZFC and FC magnetisation curves, the FC magnetisation curve shows a steep increase in magnetisation value. As the temperature increases from 3 K, FC magnetisation (500 Oe) shows a decreasing trend and ZFC magnetisation increases very slowly, showing a maximum at 24.7 K. Further heating causes a decrease in magnetisation and again produces a maximum at 44.8 K. The increase in magnetisation below ∼53 K in FC mode is an indication of ferromagnetic interaction. The bifurcation of the ZFC and FC magnetisation curves at 53 K is due to the PM to ICAFM transition. This is in good agreement with reported values for bulk samples.⁵ The maximum at ∼44.8 K in the ZFC magnetisation curve can be attributed to the ICAFM to cAAFM transition. The maximum at 24.7 K in the ZFC curve can be assigned to a spin reorientation transition.¹⁰ It can be noted that, for bulk ceramic samples, ZFC magnetisation shows a decreasing trend as the temperature increases from 3 K, but in our nanocrystallite particles, the ZFC magnetisation curve showed an increasing trend. This can be explained by the large value of the Weiss temperature, −40 K, which indicates reduced strength of the frustrated antiferromagnetism. At low temperature, cAAFM dominates and gives ferromagnetic-like behaviour. As the applied field is increased, these transitions are not seen in the ZFC magnetisation curve, which shows their weak interaction strength, so a higher field can cause them to vanish completely. At 10 000 Oe applied field, the ZFC and FC magnetisation curves show bifurcation at 43 K and ZFC shows a maximum at 13 K. The field dependence of these transitions were probed further and discussed below. The field cooling and heating curves (Fig. 4) at 500 Oe applied field show a thermal hysteresis between 19 and 56 K. The presence of the thermal hysteresis can be explained by competitive interaction between different spin orders. From 19–56 K, three phases of PM, ICAFM and cAAFM compete with each other during the heating and cooling cycle. So in the heating and cooling cycle, competitive interaction between the different types of spin ordering results in a thermal hysteresis. It is noteworthy that the width of the thermal hysteresis observed in the nanocrystallites (∼1 K) is lower than reported data for bulk samples (∼3 K).¹⁰ As the surface to volume ratio is higher in nanocrystallites, so surface disorder spins dominate over the bulk spin phenomenon. This is also evident from the steep increase of the FC magnetisation curve. To probe the field dependence of the transitions, we compared the ZFC magnetisation curves in Fig. 5. The spin reorientation transition at 24.7 K does not show field dependency up to 1000 Oe applied field, whereas the ICAFM to cAAFM transition varies from 48 K at 100 Oe to 45 K at 1000 Oe applied field. At 10 000 Oe applied field, the ZFC magnetisation curve shows a broad hump starting from 34 K with a pronounced peak ∼12 K.

Fig. 3 ZFC and FC magnetisation curves at 500 Oe and 10 000 Oe applied fields for orthorhombic EuMnO₃ nanoparticles.

Fig. 4 FC heating and cooling magnetisation curves at 500 Oe applied field for orthorhombic EuMnO₃ nanoparticles.

Fig. 5 ZFC magnetisation curves at various applied fields for orthorhombic EuMnO₃ nanoparticles showing the field dependency of transitions.

To probe further the nature of these transitions, ac magnetisation was performed in a wide temperature range at 5 Oe ac field at various frequencies. The real part of the ac magnetisation (Fig. 6A) curve showed a peak at ∼24.7 K, which is a spin reorientation transition observed in dc magnetisation. Except for this transition, none of the other transitions which were seen in the dc magnetisation data were observed in the real part of the ac magnetisation curve. The absence of a Néel transition (∼53 K) in the real part of the ac magnetisation was reported for many compounds of the same class in both the bulk and nano-regime.^17,18 No frequency dependency was observed for the spin reorientation transition. The imaginary part of the ac magnetisation (Fig. 6B) shows anomalies at ∼46.8 and 21.8 K. These can be associated to ICAFM to cAAFM and spin reorientation transitions, respectively, which are in good agreement with dc magnetisation data.

Fig. 6 Temperature dependence of (A) real and (B) imaginary part of ac magnetisation, measured in a 5 Oe field and at various frequencies as indicated in the figure for orthorhombic EuMnO₃ nanoparticles.

To probe further the nature of the transitions, isothermal field dependent magnetisation was performed around the transition regions. The M–H curves (Fig. 7) show non linearity below the Néel temperature and no saturation was observed in the whole temperature range as expected for AFM compounds. At 25 K, the coercivity value is 1025 Oe, which increases to 2991 Oe at 3 K. The increase in coercivity value at low temperature can be due to cAAFM or surface ferromagnetic-like uncompensated spins, which become more pronounced at low temperature. A large exchange bias (Fig. 7 inset) was observed below the ICAFM to cAAFM (∼43 K) transition. At 3 K, the value of the exchange bias was ∼2799 Oe and it decreased with an increase in temperature, becoming zero at 43 K. The presence of exchange bias in nanocrystallite particles is well explained in the literature, being due to the presence of a ferromagnetic-like surface layer. In the literature for bulk EuMnO₃, exchange bias is not reported. It is proposed in several previous studies that in smaller dimensions, the AFM systems may show uncompensated spins at the surface leading to net ferromagnetic-like behavior at low temperature. Moreover, due to the presence of indirect exchange interaction, the coupling is very weak and the shell-spins may not have a well-defined transition temperature. They are still part of the core lattice but due to the reduced coordination number of surface atoms, the surface spins remain disordered/canted leading to a ferromagnetic-like surface layer.^19–21 An exchange bias effect was also observed in antiferromagnetic nanoparticles, BiFe_0.8Mn_0.2O₃, La_0.2Ca_0.8MnO₃ due to uncompensated surface spins.^19,20 EuMnO₃ nanoparticles were cooled from 300 K to 3 K at the desired applied field, then an isothermal measurement was recorded. The applied field while cooling will align the surface uncompensated spins, which should increase the surface ferromagnetic-like character, thereby increasing the exchange bias. The M–H loops at 3 K under various field cooled conditions are shown in Fig. 8. It was found that at 30 000 Oe and 50 000 Oe field cooled conditions, there was an enhanced exchange bias (∼3400 Oe) at 3 K isothermal magnetisation. At 1000 Oe field cooled, which is less than coercivity at 3 K, the exchange bias decreased to 2260 Oe compared with 2799 Oe in the zero field cooled M–H. This could be due to the fact that, at fields less than the coercive field, the AFM spin could not align in the opposite direction thereby decreasing the exchange bias. Based on the above field cooled M–H loops at 3 K, we can model a nanoparticle as a core–shell structure having an inner antiferromagnetic core and ferromagnetic-like surface layer (Fig. 8 inset). The origin of the exchange bias below ICAFM to cAAFM (∼43 K) is due to the onset of antiferromagnetic nature in the core of the nanoparticles, which is weakly coupled with an uncompensated ferromagnetic-like surface layer. The temperature dependency of the exchange bias is due to an increase in the cAAFM ordering with decreasing temperature.

Fig. 7 M–H loops for orthorhombic EuMnO₃ nanoparticles at various temperatures. Inset shows temperature dependence of the exchange bias.

Fig. 8 M–H loops of orthorhombic EuMnO₃ nanoparticles at 3 K with different field cooled conditions. Inset shows schematic model of an individual orthorhombic EuMnO₃ nanoparticle.

In conclusion, we have synthesised single phase orthorhombic EuMnO₃ nanoparticles with 45 nm average particle size. The ZFC and FC magnetisation curves showed three anomalies at ∼53, 44.8 and 24.7 K, which were attributed to PM to ICAFM, ICAFM to cAAFM and spin reorientation transitions, respectively. Due to competitive interactions between different spin orderings, a thermal hysteresis (∼1 K) in between 19 and 56 K was observed in field cooled heating and cooling magnetisation. Based on the field cooled M–H loop measurement, the nanoparticle was modelled as a core–shell structure. The origin of the exchange bias below the ICAFM to cAAFM (∼43 K) transition is due to the weak coupling between the antiferromagnetic-core and the ferromagnetic-like surface layer of the nanoparticles.

Acknowledgements

PP acknowledges the centre for excellence in surface science at National Chemical Laboratory and network project Nano-SHE funded by the Council of Scientific and Industrial Research (CSIR), India and Department of Science & Technology (DST), India (DST/INT/ISR/P-8/2011). RD acknowledges CSIR, India for financial support.

References

1. M. N. Iliev, M. V. Abrashev, J. Laverdière, S. Jand, M. M. Gospodinov, Y. Q. Wang and Y. Y. Sun, Phys. Rev. B: Condens. Matter Mater. Phys., 2006, 73, 064302.
2. A. Munoz, J. A. Alonso, M. T. Casais, M. J. Martinez-Lope, J. L. Martínez and M. T. Fernández-Díaz, J. Phys.: Condens. Matter, 2002, 14, 3285.
3. B. Dabrowski, S. Kolesnik, A. Baszczuk, O. Chmaissem, T. Maxwell and J. Mais, J. Solid State Chem., 2005, 178, 629.
4. T. Kimura, S. Ishihara, H. Shintani, T. Arima, K. T. Takahashi, K. Ishizaka and Y. Tokura, Phys. Rev. B: Condens. Matter Mater. Phys., 2003, 68, 060403.
5. T. Goto, T. Kimura, G. Lawes, A. P. Ramirez and Y. Tokura, Phys. Rev. Lett., 2004, 92, 257201.
6. T. Arima, T. Goto, Y. Yamasaki, S. Miyasaka, K. Ishii, M. Tsubota, T. Inami, Y. Murakami and Y. Tokura, Phys. Rev. B: Condens. Matter Mater. Phys., 2005, 72, 100102.
7. T. Kimura, T. Goto, H. Shintani, K. Ishizaka, T. Arima and Y. Tokura, Nature, 2003, 426, 55.
8. S. W. Cheong and M. Mostovoy, Nat. Mater., 2007, 6, 13.
9. K. Noda, S. Nakamura, J. Nagayama and H. Kuwahara, J. Appl. Phys., 2005, 97, 10C103.
10. W. S. Ferreira, J. A. Moreira, A. Almeida, M. R. Chaves, J. P. Araújo, J. B. Oliveira, J. M. M. Da Silva, M. A. Sá, T. M. Mendonça, P. S. Carvalho, J. Kreisel, J. L. Ribeiro, L. G. Vieira, P. B. Tavares and S. Mendonça, Phys. Rev. B: Condens. Matter Mater. Phys., 2009, 79, 054303.
11. V. Bansal, P. Poddar, A. Ahmad and M. Sastry, J. Am. Chem. Soc., 2006, 128, 11958.
12. P. Poddar, T. Fried and G. Markovich, Phys. Rev. B: Condens. Matter Mater. Phys., 2002, 65, 172405.
13. R. Das, A. Jaiswal, S. Adyanthaya and P. Poddar, J. Phys. Chem. C, 2010, 114, 12104–12109.
14. R. Das, A. Jaiswal, S. Adyanthaya and P. Poddar, J. Appl. Phys., 2011, 109, 064309.
15. A. Jaiswal, R. Das, K. Vivekanand, T. Maity, P. M. Abraham, S. Adyanthaya and P. Poddar, J. Appl. Phys., 2010, 107, 013912.
16. A. Jaiswal, R. Das, K. Vivekanand, T. Maity, P. M. Abraham, S. Adyanthaya and P. Poddar, J. Phys. Chem. C, 2010, 114, 2108.
17. S. Harikrishnan, S. Rößler, C. M. N. Kumar, H. L. Bhat, U. K. Rößler, S. Wirth, F. Steglich and S. Elizabeth, J. Phys.: Condens. Matter, 2009, 21, 096002.
18. R. Das, A. Jaiswal and P. Poddar, J. Phys. D: Appl. Phys., 2013, 46, 045301.
19. P. K. Manna, S. M. Yusuf, R. Shukla and A. K. Tyagi, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 83, 184412.
20. V. Markovich, R. Puzniak, D. Mogilyansky, X. Wu, K. Suzuki, I. Fita, A. Wisniewski, S. Chen and G. Gorodetsky, J. Phys. Chem. C, 2011, 115, 1582.
21. H. Ahmadvand, H. Salamati, P. Kameli, A. Poddar, M. Acet and K. Zakeri, J. Phys. D: Appl. Phys., 2010, 43, 245002.

---