Magnetic Properties of BaFeO3-BaTiO3 Bilayer/Multilayer Thin Films

Fozia Aziz1,a) , Ajay Vishwakarma2,b) Gireesh Soni2, c), Nitish Gupta3,d)

1Department of Physics, Government College Mandleshwar, India 2Department of Applied Physics and Optoelectronics, Shri G. S. Institute of Technology and Science, Indore, India

2Department of Applied Chemistry, Shri G. S. Institute of Technology and Science, Indore, Indiaa)Corresponding author: foziaaziz@gmail.com

b) ajayv032@gmail.comc)gireeshsoni@gmail.com

d)nitish.nidhi75@gmail.com

Abstract. BaFeO3 (BFO) and BaTiO3 (BTO) targets were used to deposit bi-layer (BL)/multilayer (ML) thin films on single crystal SrTiO3 (001) using pulsed laser deposition. X-Ray diffraction and Reciprocal Space Maps confirmed the formation of BL/ML. Magnetization studies revealed that the sandwiched BFO displayed weakest ferromagnetic interactions whereas the bilayer with BFO as top layer showed highest magnetization and growth of ferromagnetic phase. This study demonstrates that sandwiched interfaces restrict the degrees of freedom of BFO to grow ferromagnetic phase.

INTRODUCTION

The properties of perovskite oxides are dependent on the connectivity and distortion of BO 6 octahedra [1]. When grown over single crystals in perovskite heterostructure the octahedral connectivity must be maintained across the interface. This constraint enables the octahedral rotations or disortions to be transferred from substrate to film or from one layer to another (in case of hetrostructures). This feature makes perovskites particularly appealing for multilayer growth. Over past few years many unusual properties have been observed in perovskite multilayers and superlattices. For example, in LaAlO3/SrTiO3 multilayer, each of the constituting oxide is an insulator but interface gives rise to superconductivity [2,3]; antiferromagnetic LaMnO3 and SrMnO3 when combined as bilayer, show ferromagnetism with spin aligned in-plane at the interface [4,5]; multilayer of antiferromagnetic CaMnO 3 and paramagnetic CaRuO3 have ferromagnetic interfaces [6]; superlattices of ferroelectric, dielectric and paraelectric materials like BaTiO3/SrTiO3, KNbO3/KTaO3 and PbTiO3/SrTiO3 show enhanced ferroelectricity due to interface effect [7, 8]. In a trilayer of BaTiO3/SrTiO3/CaTiO3, interfaces lead to large polarization enhancement [9], BaTiO3/SrTiO3 superlattice has enhanced ferroelectricity [10] and BaFeO3/BaTiO3 superlattice shows coexisted ferromagnetic and ferroelectric state [11].

Paucity of single phase room-temperature multiferroic materials has led to the engineering of artificial multiferroic materials. An ideal way is to combine a ferroelectric and a magnetic material in form of multilayer or superlattice. BaFeO3 (BFO), being room temperature ferromagnet with correlated structure and magnetism, is a potential candidate for magnetoelectric materials. Recently, Fukatani et al [11] fabricated BFO and BaTiO3 (BTO) superlattices which showed ferromagnetism and ferroelectricity simultaneously. They found enhancement in magnetization of superlattice due to interface effects [11,9]. The magnetic properties of BFO at room temperature are so complex that they leave a huge scope of scientific exploration. BFO also has structural flexibility, that it forms in single-phase structure from hexagonal in bulk to cubic in thin film. The combination of BFO and BTO was studied by us from a different perspective as follows: when BFO is combined with BTO in form of multilayers then the structure and interfaces with BTO should influence the properties of BFO. The chances of influencing the magnetic properties are high because both the oxides have Ba ions at A-site and 3d transition metal ions at B-site.

Now any change in structure of BFO via interfacial effect, will in turn affect its magnetic properties. Motive of the present report was to synthesize multilayer and bilayers of BFO/BTO with different number of interfaces and to explore how the interfaces affect the magnetic properties of BFO.

EXPERIMENT

BaFeO3 (BFO) and BaTiO3 (BTO) pellets were used as targets for thin film synthesis. These targets were mounted on two different target holders. Multitarget rotation mechanism of PLD was employed for depositing the alternate layers of BFO and BTO. All the bilayers (BL) and multilayer (ML) structure were grown on SrTiO3 (001) (STO) single crystal substrates. The target to substrate distance was 4.2 cm. Substrate temperature was 700⁰C and an oxygen partial pressure of 2.8 mtorr was maintained during deposition. The samples were then cooled at the deposition pressure. Laser energy was set at 260 mJ and the laser repetition rate was 5 Hz. Two BL were prepared with equal proportion of BFO and BTO (1:1). In one BL, BFO was the top layer (BTO/BFOT) and in another, BFO was first deposited on STO then covered with BTO layer (BFO/BTOT). Therefore in this BL, BFO is sandwiched. Each layer had a thickness of 450 nm. Now keeping the ratio of the content of BFO and BTO constant (same as in bilayers), we deposited a ML with 36 alternate layers of BFO and BTO, each having a thickness of 12 nm (BFO/BTO)36. To compare the properties of these heterostructures with BFO we also deposited a BFO thin film of thickness 450nm on STO (001). The thickness was chosen such that the magnetometer can detect the signal from the weak magnetization of these layers and we can subtract the diamagnetic contribution of STO substrates, without noise in the net magnetization.

The out-of-plane structure of the films was studied by X-Ray diffraction (XRD) using Cu-Kα radiation. The Reciprocal Space Maps (RSMs), to study in-plane structure, were recorded using X-ray diffractometer with 5-axes cradle (Emperean, PanAlytical) (Cu-Kα). The magnetic properties were investigated using superconducting quantum interference device (SQUID) magnetometer (Quantum design).

RESULTS & DISCUSSION

Figure 1 shows the XRD pattern of BL and ML thin films. It can be seen that all the BL/ML are highly oriented without any impurity peaks. In each graph (Fig. 1 (a) & (b)) two XRD peaks from material clearly indicate the formation of two layers with discernible interfaces. In Fig. 1 c) XRD of ML also indicates the formation of two distinct material layers corresponding to BFO and BTO, however no discernible satellite peak is observed.

FIGURE 1. X-Ray Diffraction pattern of a) bilayer with BFO on top, b) bilayer with BFO sandwiched, c) Multilayer (BFO/BTO)36.

These BL/ML thin films have partially relaxed state of strain as apparent from RSM shown in Fig.2. RSM images again confirm the formation of bilayer (Fig. 2 (a) & (b)) with two in-plane peaks along with the peak of the substrate. The RSM of ML (BFO/BTO)36 Fig. 2 (c) clearly shows satellite peaks corresponding to the formation of multilayer structure.

If the XRDs and RSMs of both the bilayers are compared, they clearly show minor structural differences owing to different strain distributions. The different distribution of strain could be acquired by the samples because of different

bottom-layer/substrate.

FIGURE 2. Reciprocal space maps of a) bilayer with BFO on top, b) bilayer with BFO sandwiched, c) Multilayer (BFO/BTO)36.

Now we compare the magnetization of BFO film deposited on STO (001) with BL and ML of BFO/BTO. Field Cooled (FC) and Zero Field Cooled (ZFC) Magnetization versus Temperature curve were recorded for all the samples to observe the magnetic behavior of all the samples. In film and BL (BTO/BFOT), BFO acquires the position as top layer. The layer of BTO is diamagnetic and contributes negligibly in the magnetization. So whatever magnetic characteristics of bilayer is observed, originate from BFO only. Magnetization measurements showed that BTO/BFOT has a weak ferromagnetic state at room temperature (Fig. 3 (b)). An onset of short-range ferromagnetic phase can be observed for bilayer at ~ 260 K. Magnetization in BTO/BFOT has decreased as compared to BFO-S (Fig. 3(a)) as illustrated by the ZFC and FC magnetization curves. This could plausibly be due to difference in template on which the film grew (STO in case of BFO-S and BTO in case of BTO/BFO T). Difference in bottom layer will give rise to change in octahedral connectivity (changed B-O bond length and B-O-B bond angle) across the interface which then extends through the film volume.

FIGURE 3. Field cooled (FC) and Zero field cooled (ZFC) magnetization versus temperature curves of a) BFO thin film, b) bilayer with BFO on top, c) bilayer with BFO sandwiched at H = 500 Oe.

In the other bilayer BFO/BTOT, BFO was first deposited on STO then covered with BTO layer. Therefore in this bilayer, BFO is sandwiched. In BTO/BFOT, BFO is top-layer and had one free surface, but in, BFO is placed between STO (bottom) and BTO (top). When BFO was sandwiched between STO and BTO, magnetization showed a very different variation with temperature (Fig 3(c)) as compared to when it has a free surface on top (Fig. 3(b)). Separation between ZFC and FC magnetization curves is diminished for sandwiched BFO indicating diminished short-range ferromagnetic interactions. Also, a frustrated magnetic state arises in the low-temperature region. The magnetization isotherm at 300 K of BFO/BTOT shows a very weak ferromagnetic state as compared to BTO/BFOT

(Fig. 4) . Thus, the bilayers of BTO and BFO show dramatically different magnetic behavior in spite of the same proportions of BFO and BTO compounds. This difference of magnetic behavior can be plausibly explained as follows.

In BTO/BFOT bilayer, BFO encounters strain at one surface in contact with BTO, while the free surface of the bilayer is formed by BFO layer. Thickness of BFO layer in BTO/BFOT is ~ 450 nm and the chemically diffused region would extend a few nanometers beyond. In BFO/BTOT, BFO is sandwiched, and there are two interfaces of BFO: one with STO and another with BTO. Unlike BTO/BFOT, there is no free surface in BFO/BTOT, thus, latter has larger content of diffusion. In addition, the sandwiched layer would experience a larger distribution of strain in

FIGURE 4. Magnetization isotherm at 300 K of the two bilayers.

order to form the sandwiched structure between two different materials. In addition, there are lesser degrees of freedom to the sandwiched structure. Therefore, the sandwiched BFO behaves more like antiferromagnetic at low temperature and like a very weak ferromagnet at room temperature.

Now keeping the ratio of the content of BFO and BTO constant (same as in bilayers), we deposited a multilayer with 36 alternate layers of BFO and BTO, each having a thickness of 12 nm (BFO/BTO)36. With increase in number of interfaces, the magnetization enhances compared to that of BFO/BTOT, but the magnetization is slightly less than that of BTO/BFOT (Fig 5 (a)).

FIGURE 5. (a) ZFC-FC Magnetization versus Temperature curves, Magnetization isotherm at (b) 300 K and (c) 20K of multilayer (BFO/BTO)36.

The transition to frustrated magnetic state at low temperature diminished in multilayer. Thus, rather a long-range antiferromagnetic phase is found to arise dominatingly in case of (BTO/BFO)36 same as in BFO/BTOT bilayer. However, the magnetization measurements at 20 K (Fig 5 (b))also clearly illustrate the antiferromagnetic order with competing magnetic interactions, which also appears in large ZFC-FC separation at low-temperatures. This indicates highly competing short-range ferromagnetic phase embedded within antiferromagnetic background at low temperatures.

CONCLUSION

We have fabricated bilayer and multilayer structures of BFO and BTO. It was found that magnetization of BFO is suppressed when grown in sandwiched form. It shows sandwiched interfaces restricts the degrees of freedom of BFO to grow ferromagnetic phase.

ACKNOWLEDGEMENT

The authors are grateful to Dr Mukul Gupta, IUC, Indore for XRD studies and the Department of Physics, Banasthali Vidyapith University for providing various measurement facilities. Authors acknowledge the financial support by National Project Implementation Unit (NPU), New Delhi, India in form of TEQIP III CRS research project, CRS ID- 1-5759918216.

REFERENCES

1. J. M. Rondinelli, S. J. May, J. W. Freeland, MRS Bulletin 37, 261, 2012.2. K. Janicka, J. P. Velev, E. Y. Tsymbal, J. Appl. Phys. 103, 07B508, 2008.3. A. Ohtomo, H. Y. Hwang, Nature 427, 423, 2004. 4. N. Ogawa, Takuya Satoh, Yasushi Ogimoto, and Kenjiro Miyano , Phy. Rev. B 78, 212409, 2008.5. A. Bhattacharya, S. J. May, S. G. E. te Velthuis, M. Warusawithana, X. Zhai, Bin Jiang, J.-M. Zuo, M. R.

Fitzsimmons, S. D. Bader, and J. N. Eckstein, Phy. Rev. Lett. 100, 257203, 2008.6. C. He,  A. J. Grutter, M. Gu, N. D. Browning, Y. Takamura, B. J. Kirby, J. A. Borchers, J. W. Kim, M. R.

Fitzsimmons, X. Zhai, V. V. Mehta, F. J. Wong, and Y. Suzuki , Phys. Rev. Lett. 109, 197202, 2012.7. Oswaldo Diéguez, Karin M. Rabe, and David Vanderbilt, Phys. Rev. B 72,144101, 2005.8. M. Dawber , N. Stucki, C. Lichtensteiger, S. Gariglio, P. Ghosez , J.M. Triscone, Adv. Mater. 19, 4153, 2007.9. Ho Nyung Lee, Hans M. Christen, Matthew F. Chisholm, Christopher M. Rouleau & Douglas H. Lowndes ,

Nature 433, 395, 2005. 10. Ryosuke Fukatani, Hiroko Yokota, Shingo Ogura, Yoshiaki Uesu, Ausrine Bartasyte, Mamoru

Fukunaga and Yukio Noda, J. Appl. Phys. 51, 09LB01, 2012.11. H. Yokota, H. Tabata, G. B-Lee, R. Fukatani, S. Ogura, and Y. Uesu , , IEEE International Symposium on

Applications of Ferroelectrics (ISAF), 2010.12. M. K. Lee, T. K. Nath, C. B. Eom, M. C. Smoak, F. Tsui, Appl. Phys. Lett. 77, 3547, 2010.