Integration of BaTiO3/CoFe2O4 multiferroic heterostructure on GaN semiconductor

Guanjie Liac, Xiaomin Li*ab, Qiuxiang Zhu*a, Junliang Zhaod and Xiangdong Gaoab
aState Key Laboratory of High Performance Ceramics and Superfine Microstructure, Shanghai Institute of Ceramics, Chinese Academy of Science, No. 1295 Dingxi Road, Shanghai, 200050, PR China. E-mail:;
bCenter of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Science, No. 19A Yuquan Road, Beijing, 100049, PR China
cUniversity of Chinese Academy of Science, No. 19A Yuquan Road, Beijing, 100049, PR China
dNanjing NanoArc New Materials Technology Co., Ltd., No. 37 Jiangjun Avenue, Nanjing, 211106, PR China

Received 16th June 2019 , Accepted 15th August 2019

First published on 19th August 2019

Epitaxial integration of BaTiO3 (BTO)/CoFe2O4 (CFO) multiferroic heterostructure directly on GaN semiconductor was demonstrated using pulsed laser deposition. The domain matching epitaxy mechanism was revealed to be (111)[1[1 with combining macron]0] BTO//(111)[1[1 with combining macron]0] CFO//(0002)[11[2 with combining macron]0] GaN. Spinel CFO thin films with a layer-by-layer growth mode on GaN not only served as the ferrimagnetic functional layer, but also as a buffer layer, inducing an epitaxial growth of perovskite BTO ferroelectric thin films on wurtzite GaN by greatly reducing lattice mismatch at the BTO/GaN interface. The designed BTO/CFO/GaN heterostructure displayed high crystallinity, dense microstructure and good interfacial state. More importantly, good ferroelectric properties for the BTO layer with a remanent polarization of 5.5 μC cm−2 and magnetic properties for the CFO layer with a saturation magnetization of 169 emu cm−3 at room temperature were also demonstrated. Thus, the epitaxial integration of high performance BTO/CFO multiferroic heterostructure with GaN could add more functional degrees of freedom for designing advanced microelectronic devices on a GaN semiconductor platform.

1. Introduction

Monolithic integration of functionalized oxides on complementary metal oxide semiconductor (CMOS) platforms offers promising routes to improve device performances, such as a high-k oxide gate and ferroelectric negative capacitance gate,1,2 or to add new functionalities on a chip, including integrated photonics, hybrid spintronics and ferroelectric field effect transistors.3–5 Compared to traditional Si, Ge and GaAs semiconductors, the third-generation GaN semiconductor, with wider band gap and higher electron mobility and breakdown voltage, would provide a great platform for oxide microelectronic devices.6,7 Presently, various functional oxides with ferroelectric, piezoelectric or magnetic functionalities have been integrated on GaN. Nevertheless, functional thin films grown on GaN are currently focused on single magnetic oxides, such as EuO and Fe3O4,8,9 or single ferroelectric oxides, including BiFeO3, PbZr0.8Ti0.2O3 and LiNbO3.10–12

In the research surge on multiferroic materials with coupled ferroelectric and magnetic properties, multiferroic oxides integrated on semiconductors could offer more functional degrees of freedom for exploring integrated oxide devices, especially for multiferroic heterostructures with strong simultaneous ferroelectric and magnetic properties.13 Therefore, the epitaxial integration of high performance multiferroic heterostructures on GaN semiconductors displays great potential for exploring multifunctional GaN-based integrated oxide devices. Meanwhile, on account of large lattice mismatch and different crystal structures between functional oxides and semiconductors,14 complex buffer layers are normally introduced to enable epitaxial integration, such as BaTiO3/CoFe2O4 multiferroic heterostructure on Si with LaNiO3/CeO2/YSZ buffer layer and BiFeO3/La0.7Sr0.3MO3 multiferroic heterostructure on GaAs with SrTiO3 buffer layer.15,16 However, buffer layers weaken the coupling effect at the multiferroic/semiconductor interface. Thus, there is tremendous interest in the epitaxial construction of multiferroic heterostructures on GaN without a buffer layer by artificial lattice design, which could simplify the device structure and enhance the performance of GaN-based integrated devices.

CoFe2O4 (CFO) is a typical spinel ferrimagnetic insulator with high Curie temperature, large coercive field and strong magnetostriction for wide application in magnetic tunnel junctions and magnetic data storage.17 In addition, as the first discovered high performance perovskite ferroelectric, BaTiO3 (BTO) is suitable for application in integrated ferroelectric devices and has been integrated on various semiconductor substrates.18,19 More interestingly, obvious magnetoelectric coupling effect has been detected in BTO/CFO multiferroic heterostructures grown on SrTiO3 substrates.20,21 Thus, the monolithic integration of BTO/CFO multiferroic heterostructure on GaN might further expand the functionality of GaN-based integrated oxide devices.

In this research, the epitaxial growth of the designed (111) BTO/CFO multiferroic heterostructure on (0002) GaN semiconductor without a buffer layer was realized using pulsed laser deposition (PLD). The spinel CFO ferrimagnetic functional thin film directly grown on GaN could serve as the buffer layer to induce the epitaxial integration of the perovskite BTO ferroelectric thin film with a wurtzite GaN semiconductor. Thus, a BTO/CFO/GaN multiferroic-semiconductor heterostructure with high crystallinity, dense microstructure and good interfacial state was constructed, which displayed good ferroelectric and magnetic properties.

2. Experimental

In this study, BTO/CFO multiferroic heterostructures were constructed on commercial n-type (0002) GaN-on-sapphire substrates using the PLD technique with a KrF excimer laser (248 nm). The heavily Si-doped n-type GaN substrates could serve as bottom electrodes with electron concentrations over 5 × 1018 cm−3. First, CFO thin films were deposited on GaN substrates using CFO ceramic targets and the corresponding growth conditions were set to a background oxygen pressure of 5 Pa, temperature of 600 °C, growth rate of 5 Å min−1 and laser energy density of 3 J cm−2. After the deposition of CFO thin films, BTO thin films were fabricated on CFO/GaN under background oxygen pressure of 2 Pa, growth temperature of 680 °C, growth rate of 2 nm min−1 and laser energy density of 3 J cm−2 with BaTiO3 ceramic targets.

During the whole growth process, the growth modes of BTO and CFO thin films were studied by in situ reflection high-energy electron diffraction (RHEED) with electron energy at 20 keV. A high-resolution X-ray diffractometer (HRXRD) (Bruker D8 Discover) was employed to investigate the epitaxial quality and the relationship of BTO/CFO/GaN heterostructures. Surface and sectional morphologies of the heterostructures were observed by atomic force microscopy (AFM) (Ntegra NT-MDT) and field emission scanning electron microscopy (FESEM) (Hitachi SU-8220), respectively. Using a high-resolution transmission electron microscope (HRTEM) (Tecnai G2 F20 S-Twin), the epitaxial relationship and interfacial state of BTO/CFO/GaN heterostructures were further analyzed. Piezoelectric force microscopy (PFM) measurements were conducted on BTO using a PFM system (Ntegra NT-MDT). Circular Pt top electrodes with a diameter of 150 μm and spacing of 1 mm were deposited on the BTO surface through a shadow mask by PLD for ferroelectric tests. The ferroelectric properties of BTO/CFO/GaN heterostructures were studied using a double beam laser interferometer (aixDBLI) with a top–top electrode test mode and the corresponding magnetic properties were studied by a superconducting quantum interference device (SQUID) (MPMS3, Quantum Design).

3. Results and discussion

RHEED patterns of GaN substrates and CFO and BTO thin films were detected along the (0002)[11[2 with combining macron]0] GaN and (0002)[10[1 with combining macron]0] GaN azimuths, respectively. Fig. 1c and d show bright streaky RHEED patterns of CFO thin films along the above two azimuths, indicating the high epitaxial quality of spinel CFO on wurtzite GaN with a layer-by-layer mode.22 From the RHEED patterns shown in Fig. 1e and f, it could be inferred that the perovskite BTO thin films were epitaxially grown on CFO-buffered GaN with a Stranski–Krastanov growth mode. Thus, BTO/CFO/GaN epitaxial heterostructure is constructed and the RHEED patterns were resolved as marked in the patterns.14,17 By resolving the above RHEED patterns, the epitaxial relationship of BTO/CFO/GaN heterostructures could be inferred to be (111)[1[1 with combining macron]0] BTO//(111)[1[1 with combining macron]0] CFO//(0002)[11[2 with combining macron]0] GaN and (111)[11[2 with combining macron]] BTO//(111)[11[2 with combining macron]] CFO//(0002)[10[1 with combining macron]0] GaN along the two observed azimuths. In addition, six repeated RHEED patterns were detected for 360° rotation around the normal of CFO and BTO thin films as GaN substrates, revealing the six-fold symmetry of (111) CFO and (111) BTO thin films on GaN. Therefore, on account of the three-fold symmetry for (111) CFO and (111) BTO bulk crystals, intrinsic twin domain structures in epitaxial CFO and BTO thin films are formed.
image file: c9ce00932a-f1.tif
Fig. 1 RHEED patterns for BTO/CFO/GaN heterostructure taken along (0002)[11[2 with combining macron]0] GaN azimuths of (a) wurtzite GaN, (c) spinel CFO and (e) perovskite BTO; (0002)[10[1 with combining macron]0] GaN azimuths of (b) wurtzite GaN, (d) spinel CFO and (f) perovskite BTO.

The crystalline structure and epitaxial relationship of the designed BTO/CFO/GaN heterostructures were further analyzed by HRXRD. As shown in the HRXRD θ–2θ scan pattern in Fig. 2a, only (111)-orientation diffraction peaks could be observed for thick CFO thin films on (0002) GaN substrates, which reveals an out-of-plane (111) CFO//(0002) GaN epitaxial relationship. Fig. 2b demonstrates the rocking curve of the (222) CFO diffraction peak; a corresponding full-width half-maximum (FWHM) value of merely 0.10° was calculated, which is much lower than the CFO grown on Si semiconductors and even SrTiO3 single crystal substrates.23,24 Thus, (111) CFO thin films on GaN possess ultrahigh crystallinity and epitaxy quality. BTO thin films directly deposited on GaN possess polycrystalline structures and weak ferroelectric properties for large lattice mismatch at the BTO/GaN interface.25,26 However, as shown in Fig. 2c, only (111)-orientation diffraction peaks could be observed for perovskite BTO on GaN with the insertion of spinel CFO, indicating that CFO could serve as a buffer layer to induce the epitaxial integration of BTO on GaN. In addition, the rocking curve taken of (111) BTO with FWHM value of 0.81° shown in Fig. 2d demonstrates the good crystallinity of integrated BTO thin films.

image file: c9ce00932a-f2.tif
Fig. 2 (a) XRD θ–2θ scan pattern of CFO on (0002) GaN substrate. (b) Rocking curve for the (222) CFO diffraction peak. (c) XRD θ–2θ scan pattern of BTO/CFO/GaN heterostructure. (d) Rocking curve for the (111) BTO diffraction peak. (e) XRD Φ scan patterns detected on (1[1 with combining macron]0) BTO and (10[1 with combining macron]1) GaN crystal planes. (f) Schematic diagram of the epitaxial relationship of BTO/CFO/GaN heterostructure and corresponding lattice mismatch at interface.

HRXRD Φ scans were conducted on (1[1 with combining macron]0) BTO and (10[1 with combining macron]1) GaN crystal planes to further reveal the epitaxial relationship. As shown in Fig. 2e, six (1[1 with combining macron]0) BTO diffraction peaks 60° apart from each other could be observed. Thus, a twin domain in (111) BTO thin films with six-fold symmetry on GaN is confirmed, which is consistent with the RHEED results. Twin domain structures could cause massive intrinsic grain boundaries for domain walls, which could contribute to the release of the crystal lattice strain but would not be beneficial for the ferroelectric properties of epitaxial (111) BTO thin films. By analysing the relative positions of (1[1 with combining macron]0) BTO (2θ = 32.50°, χ = 35.26°) and (10[1 with combining macron]1) GaN (2θ = 36.85°, χ = 61.95°) crystal planes, an in-plane [1[1 with combining macron]0] BTO//[11[2 with combining macron]0] GaN orientation relationship could be proved,14 which is consistent with the RHEED conclusions.

The epitaxial relationship of the designed BTO/CFO/GaN heterostructures is fully resolved by RHEED and HRXRD results. As shown in Fig. 2f, the corresponding epitaxy model is proposed to be (111)[1[1 with combining macron]0] BTO//(111)[1[1 with combining macron]0] CFO//(0002)[11[2 with combining macron]0] GaN with an in-plane 180° rotation twin domain for BTO and CFO. Based on the proposed model, the lattice mismatch at CFO/GaN interface is +6.9%, calculated by (4aGaNimage file: c9ce00932a-t1.tif)/4aGaN. The lattice plane spacing of (111) CFO is obtained as 2.40 Å from HRXRD results using Bragg's diffraction equation, indicating a tensile strain in (111) CFO on (0002) GaN compared with the (111) CFO bulk crystal (2.42 Å), which is consistent with a positive lattice mismatch. However, the relatively large lattice mismatch value seems large for the epitaxial growth of CFO on GaN with a layer-by-layer mode. (111) CFO could be epitaxially grown on mica and Al2O3 with the layer-by-layer mode for the low surface energy of (111) spinel crystal plane.22,27 Thus, the low surface energy of (111) CFO crystal plane and relatively small lattice mismatch could jointly induce the epitaxy of high quality CFO thin films on GaN substrates. In addition, on account of the lattice constants of BTO (a = 4.01 Å) and CFO (a = 8.39 Å),23,28 the possible domain matching epitaxy mechanism could be 2aBTO = aCFO. Thus, the lattice mismatch at the BTO/CFO interface is +4.4%, calculated by image file: c9ce00932a-t2.tif. Note that the lattice mismatch decreased from +10.8% to +4.4% with the insertion of the CFO spinel buffer layer, thus inducing the epitaxial growth of perovskite BTO on wurtzite GaN. In addition, the lattice spacing of about 2.33 Å in (111) BTO calculated by HRXRD results is close to that of the bulk crystal, indicating the relaxed state of BTO thin film for large thicknesses and massive twin grain boundary defects. From the sectional HRSEM image of the BTO/CFO/GaN multilayer shown in Fig. 3a, CFO and BTO thin films grown on GaN display dense microstructures with the thicknesses of about 20 nm and 205 nm, respectively. Note that the dense microstructure of BTO could enhance the ferroelectric properties for a relatively low leakage current. AFM characterization was further conducted to evaluate the growth quality of the designed heterostructure. The AFM image in Fig. 3b demonstrates the high quality step-shaped surface of epitaxial (111) CFO thin films on GaN. In addition, an ultralow root-mean-square (RMS) roughness of merely 0.282 nm and height variations within 1 nm along the line profile could be observed, which is consistent with the layer-by-layer growth mode of CFO revealed by RHEED. Note that the ultralow FWHM and RMS values of CFO reveal the ultrahigh epitaxial growth quality of (111) CFO thin films integrated on GaN, which could be compared with the CFO grown on lattice mismatched MgO substrates.24 Therefore, the high quality CFO layer offers a great template for the epitaxial growth of BTO on GaN and other ferroelectric materials, such as PZT and PMN-PT, with similar lattice constants. In addition, the AFM image of the BTO surface shown in Fig. 3c displays an RMS value of 0.740 nm, also revealing a good surface and microstructure of BTO thin films on GaN.

image file: c9ce00932a-f3.tif
Fig. 3 (a) Cross-sectional HRSEM image of the BTO/CFO/GaN heterostructure. AFM images of (b) CFO surface and (c) BTO surface with height line profile.

Sectional HRTEM was conducted on the designed BTO/CFO/GaN heterostructures. As seen in the interface structure in Fig. 4a, the clear lattice fringes observed in the CFO and BTO layers directly confirm the whole epitaxy structure of BTO/CFO/GaN heterostructures and the thickness of the high quality CFO layer is determined to be 15 nm. Fig. 4b and c present clear HRTEM images of the BTO/CFO and CFO/GaN interfaces, respectively, after processing by fast Fourier transform (FFT). The interfacial state, especially for CFO/GaN oxide-semiconductor interfaces, is pivotal for integrated oxide devices. No obvious interfacial oxide layer or chemical diffusion reaction could be observed at the interface, revealing the good interfacial state of the BTO/CFO/GaN heterostructure. Average lattice spacings in BTO and CFO thin films are calculated to be 4.02 Å and 2.23 Å using the Digital Micrograph software, corresponding to the (001) BTO and (222) CFO crystal planes, respectively. Thus, the proposed (111) BTO//(111) CFO//(0002) GaN epitaxial relationship could be proved. In addition, the lattice spacing of 2.23 Å in (111) CFO thin film is slightly smaller than 2.42 Å in the corresponding bulk crystal; the above lattice distortion indicates strong tensile strain in CFO, which originated from the relatively large lattice mismatch of +6.9% at the CFO/GaN interface. Thus, the epitaxial BTO/CFO multiferroic heterostructure with high crystallinity and good microstructure and interface state was constructed on GaN semiconductor, demonstrating potential applications in integrated oxide microelectronics.

image file: c9ce00932a-f4.tif
Fig. 4 HRTEM images of (a) BTO/CFO/GaN heterostructure, (b) BTO/CFO interface and (c) CFO/GaN interface after processing by FFT.

The ferroelectric properties of BTO thin films in BTO/CFO/GaN heterostructures were systemically studied. On account of the ferroelectric–insulator–semiconductor structure of BTO/CFO/GaN, the ferroelectric domain of BTO should be pinned by the depletion layer in the GaN semiconductor under the ferroelectric (FE) test process and asymmetric PE loops should be detected. Thus, the ferroelectric properties of BTO thin films were evaluated by the top–top electrode testing mode, as shown in the insert of Fig. 5a.29 From the polarization–voltage (PV) loops displayed in Fig. 5a, the ferroelectric properties of BTO thin films on CFO-buffered GaN were investigated under different external voltages with 100 Hz at room temperature. The polarization state of BTO increases with enhanced applied external test voltage, revealing a good ferroelectric property of BTO thin films. Fig. 5b shows PV loops of BTO near saturated state and the corresponding ferroelectric switching IV curve. The remnant ferroelectric polarization (Pr) value of BTO on GaN is determined to be 5.5 μC cm−2. In addition, switching peaks near coercive voltage observed in the IV curve further prove the good ferroelectric switching property of BTO thin films. Moreover, as the PFM phase and magnitude images demonstrate in Fig. S1 (ESI), clear phase contrast could be observed after +10 V and −10 V polarization treatment, which directly proves the ferroelectric domain switching property of BTO in the BTO/CFO/GaN heterostructure. The Pr value is relatively lower than that of BTO thin films on STO substrates with SrRuO3 bottom electrodes and MgO substrates with La0.5Sr0.5CoO3 bottom electrodes.30,31 The ferroelectric domain should be pinned by the space charge depletion layer in the GaN semiconductor and weakened by the large depolarization field of incomplete screening charges at the BTO/CFO ferroelectric/insulator interface.28 Therefore, ferroelectric properties would be normally weaker in a metal–ferroelectric–insulator–semiconductor than in a metal–ferroelectric–metal–insulator–semiconductor structure, including Pr value and ferroelectric domain switching capability, which is the main challenge in integrated ferroelectric devices for non-volatile memory and logical applications. For example, epitaxial BTO thin films integrated on GaAs with BTO–MgO–GaAs and BTO–STO–GaAs structures displayed Pr values lower than 2.5 μC cm−2.32,33 In addition, the massive intrinsic grain boundary defects for the six-fold symmetry in (111) BTO thin films could also restrain the ferroelectric properties.

image file: c9ce00932a-f5.tif
Fig. 5 (a) PV hysteresis loops of BTO ferroelectric thin films tested under different voltages; (insert) electrical test schematic diagram for BTO/CFO/GaN heterostructure. (b) PV hysteresis loop and corresponding IV switching curve and (c) LV and (d) CV curves of BTO/CFO/GaN heterostructures.

Fig. 5c shows the leakage–voltage (LV) property of BTO/CFO/GaN heterostructure at room temperature. The leakage current is lower than 3.5 μA cm−2 with an applied voltage of 10 V, displaying low leakage current in the heterostructure, which is vital for integrated ferroelectric device applications. The low leakage current could originate from the low intrinsic leakage property of BTO, the thick CFO insulator layer, and the formation of a wide depletion region in the GaN semiconductor. Fig. 5d displays the capacitance–voltage (CV) property of BTO/CFO/GaN heterostructure at 1 kHz. The observed butterfly-shaped CV curve reveals the ferroelectric properties of BTO thin films. Note that a wide capacitance variation from 97 nF to 59 nF of the heterostructure could originate from the thickness modulation of the depletion layer in GaN by loaded external voltage during the test process and the partial dielectric tunability of BTO. Thus, the epitaxial integration of BTO ferroelectric thin films with good performance on GaN semiconductor was achieved by inserting a ferrimagnetic CFO buffer layer.

The magnetic property of the designed BTO/CFO/GaN heterostructure was further investigated. In-plane and out-of-plane magnetization–magnetic field (MH) hysteresis loops of BTO/CFO/GaN heterostructure at 300 K are shown in Fig. 6b. After the subtraction of the diamagnetic contribution from GaN substrates (Fig. 6c), Fig. 6a demonstrates the in-plane and out-of-plane MH hysteresis loops of CFO thin film in the designed heterostructure. The in-plane and out-of-plane saturation magnetizations (Ms) are determined to be about 169 emu cm−3 and 156 emu cm−3, respectively. Meanwhile, the in-plane and out-of-plane coercive fields (Hc) are determined to be 2.7 kOe and 1.6 kOe, respectively. The above Ms and Hc values are comparable to the epitaxial (111) CFO thin films deposited on Si and mica substrates,17,22 indicating good magnetic properties of the ferrimagnetic CFO functional layer. Note that the magnetic anisotropy property in CFO thin film is sensitive to the strain state.17,24 Thus, the observed magnetic anisotropy of CFO thin film in BTO/CFO/GaN heterostructure could originate from the strong tensile strain of the CFO layer confirmed by HRTEM results.

image file: c9ce00932a-f6.tif
Fig. 6 (a) In-plane and out-of-plane MH hysteresis loops of CFO ferrimagnetic thin film after the subtraction of diamagnetic contribution from GaN substrate. Insert (b) MH hysteresis loops of BTO/CFO/GaN heterostructure; insert (c) MH hysteresis loop of GaN substrate.

Therefore, the monolithic integration of BTO/CFO multiferroic heterostructure with good ferroelectric and magnetic properties directly on GaN semiconductor is realized. Utilizing the ferroelectric function of BTO, the BTO/CFO heterostructure could serve as a ferroelectric polarization gate on AlGaN/GaN for normally-off high electron mobility transistors and nonvolatile ferroelectric memory devices.34 Meanwhile, BTO/CFO/GaN heterostructure shows potential application in GaN-based spintronic devices for its magnetic layer of CFO. More interestingly, good magnetoelectric coupling effect might exist in the BTO/CFO multiferroic heterostructure, which could pave the way for exploring advanced magnetoelectric multiferroic devices, such as magnetoelectric memory and magnetic sensors, on the GaN semiconductor platform.35

4. Conclusions

In conclusion, we have demonstrated the epitaxial construction of high quality BTO/CFO/GaN multiferroic-semiconductor heterostructure by the PLD technique. A domain epitaxy matching model of (111)[1[1 with combining macron]0] BTO//(111)[1[1 with combining macron]0] CFO//(0002)[11[2 with combining macron]0] GaN is revealed by RHEED, HRXRD and HRTEM results. Spinel CFO on GaN could simultaneously serve as ferrimagnetic function layer and a buffer layer that induces the epitaxial growth of perovskite BTO on wurtzite GaN by reducing the lattice mismatch from +10.8% to +4.4% at the BTO/GaN interface. The BTO/CFO/GaN heterostructure with high crystallinity, dense microstructure and good interfacial state also demonstrates good room temperature ferroelectric properties with Pr of 5.5 μC cm−2 and magnetic properties with Ms of 169 emu cm−3. Thus, the designed BTO/CFO/GaN multiferroic-semiconductor heterostructure could broaden the way to explore advanced multiferroic microelectronic devices on a GaN platform.

Conflicts of interest

There are no conflicts of interest to declare.


This work was supported by the National Natural Science Foundation of China (No. 51572280 and 51602329), the National Key R&D Program of China (No. 2016YFA0201103).


  1. R. A. Mckee, F. J. Walker and M. F. Chisholm, Science, 2001, 293, 468–471 CrossRef CAS PubMed.
  2. M. Hoffmann, M. Pešić, K. Chatterjee, A. I. Khan, S. Salahuddin, S. Slesazeck, U. Schroeder and T. Mikolajick, Adv. Funct. Mater., 2016, 26, 8643–8649 CrossRef CAS.
  3. S. Abel, T. Stoferle, C. Marchiori, C. Rossel, M. D. Rossell, R. Erni, D. Caimi, M. Sousa, A. Chelnokov, B. J. Offrein and J. Fompeyrine, Nat. Commun., 2013, 4, 1671 CrossRef PubMed.
  4. L. L. Lev, D. V. Averyanov, A. M. Tokmachev, F. Bisti, V. A. Rogalev, V. N. Strocov and V. G. Storchak, J. Mater. Chem. C, 2017, 5, 192–200 RSC.
  5. S. R. Bakaul, C. R. Serrao, M. Lee, C. W. Yeung, A. Sarker, S. L. Hsu, A. K. Yadav, L. Dedon, L. You, A. I. Khan, J. D. Clarkson, C. Hu, R. Ramesh and S. Salahuddin, Nat. Commun., 2016, 7, 10547 CrossRef CAS PubMed.
  6. S. J. Pearton, F. Ren, A. P. Zhang and K. P. Lee, Mater. Sci. Eng., R, 2000, 30, 55–212 CrossRef.
  7. F. Roccaforte, P. Fiorenza, G. Greco, R. Lo Nigro, F. Giannazzo, F. Iucolano and M. Saggio, Microelectron. Eng., 2018, 187–188, 66–77 CrossRef CAS.
  8. A. Schmehl, V. Vaithyanathan, A. Herrnberger, S. Thiel, C. Richter, M. Liberati, T. Heeg, M. Rockerath, L. F. Kourkoutis, S. Muhlbauer, P. Boni, D. A. Muller, Y. Barash, J. Schubert, Y. Idzerda, J. Mannhart and D. G. Schlom, Nat. Mater., 2007, 6, 882–887 CrossRef CAS PubMed.
  9. P. K. J. Wong, W. Zhang, X. G. Cui, Y. B. Xu, J. Wu, Z. K. Tao, X. Li, Z. L. Xie, R. Zhang and G. van der Laan, Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 81, 035419 CrossRef.
  10. W. Tian, V. Vaithyanathan, D. G. Schlom, Q. Zhan, S. Y. Yang, Y. H. Chu and R. Ramesh, Appl. Phys. Lett., 2007, 90, 172908 CrossRef.
  11. L. Li, Z. Liao, N. Gauquelin, M. D. Nguyen, R. J. E. Hueting, D. J. Gravesteijn, I. Lobato, E. P. Houwman, S. Lazar, J. Verbeeck, G. Koster and G. Rijnders, Adv. Mater. Interfaces, 2018, 5, 1700921 CrossRef.
  12. C. T. Lee, C. L. Yang, C. Y. Tseng, J. H. Chang and R. H. Horng, IEEE Trans. Electron Devices, 2015, 62, 2481–2487 CAS.
  13. Y. Kim, C. Pham and J. P. Chang, J. Phys. D: Appl. Phys., 2015, 48, 063001 CrossRef.
  14. G. Li, X. Li, Y. Chen, S. Jia and X. Xu, Appl. Surf. Sci., 2019, 465, 1055–1060 CrossRef CAS.
  15. M. Scigaj, N. Dix, J. Gazquez, M. Varela, I. Fina, N. Domingo, G. Herranz, V. Skumryev, J. Fontcuberta and F. Sanchez, Sci. Rep., 2016, 6, 31870 CrossRef CAS PubMed.
  16. M. S. Rahman, S. Ghose, L. Hong, P. Dhungana, A. Fahami, J. R. Gatabi, J. S. Rojas-Ramirez, A. Zakhidov, R. F. Klie, R. K. Pandey and R. Droopad, J. Mater. Chem. C, 2016, 4, 10386–10394 RSC.
  17. Y. K. Wakabayashi, Y. Nonaka, Y. Takeda, S. Sakamoto, K. Ikeda, Z. Chi, G. Shibata, A. Tanaka, Y. Saitoh, H. Yamagami, M. Tanaka, A. Fujimori and R. Nakane, Phys. Rev. B, 2017, 96, 104410 CrossRef.
  18. Y. Huang, B. Lu, Y. Zou, D. Li, Y. Yao, T. Tao, B. Liang and S. Lu, J. Inorg. Chem., 2018, 33, 767–772 Search PubMed.
  19. P. Ponath, K. Fredrickson, A. B. Posadas, Y. Ren, X. Wu, R. K. Vasudevan, M. B. Okatan, S. Jesse, T. Aoki, M. R. McCartney, D. J. Smith, S. V. Kalinin, K. Lai and A. A. Demkov, Nat. Commun., 2015, 6, 6067 CrossRef CAS PubMed.
  20. Y. Zhang, C. Deng, J. Ma, Y. Lin and C. W. Nan, Appl. Phys. Lett., 2008, 92, 062911 CrossRef.
  21. N. Jedrecy, T. Aghavnian, J. B. Moussy, H. Magnan, D. Stanescu, X. Portier, M. A. Arrio, C. Mocuta, A. Vlad, R. Belkhou, P. Ohresser and A. Barbier, ACS Appl. Mater. Interfaces, 2018, 10, 28003–28014 CrossRef CAS PubMed.
  22. H. J. Liu, C. K. Wang, D. Su, T. Amrillah, Y. H. Hsieh, K. H. Wu, Y. C. Chen, J. Y. Juang, L. M. Eng, S. U. Jen and Y. H. Chu, ACS Appl. Mater. Interfaces, 2017, 9, 7297–7304 CrossRef CAS PubMed.
  23. R. Bachelet, P. de Coux, B. Warot-Fonrose, V. Skumryev, G. Niu, B. Vilquin, G. Saint-Girons and F. Sánchez, CrystEngComm, 2014, 16, 10741–10745 RSC.
  24. T. Dhakal, D. Mukherjee, R. Hyde, P. Mukherjee, M. H. Phan, H. Srikanth and S. Witanachchi, J. Appl. Phys., 2010, 107, 053914 CrossRef.
  25. B. K. Pandey, T. N. Bhat, B. Roul, K. K. Nanda and S. B. Krupanidhi, J. Phys. D: Appl. Phys., 2018, 51, 045104 CrossRef.
  26. T. N. Bhat, B. K. Pandey and S. B. Krupanidhi, J. Phys. D: Appl. Phys., 2017, 50, 275101 CrossRef.
  27. A. V. Ramos, T. S. Santos, G. X. Miao, M. J. Guittet, J. B. Moussy and J. S. Moodera, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 78, 180402 CrossRef.
  28. C. Dubourdieu, J. Bruley, T. M. Arruda, A. Posadas, J. Jordan-Sweet, M. M. Frank, E. Cartier, D. J. Frank, S. V. Kalinin, A. A. Demkov and V. Narayanan, Nat. Nanotechnol., 2013, 8, 748–754 CrossRef CAS PubMed.
  29. Y. Bai, Z. J. Wang, J. Z. Cui and Z. D. Zhang, J. Phys. D: Appl. Phys., 2018, 51, 205303 CrossRef.
  30. A. P. Chen, F. Khatkhatay, W. Zhang, C. Jacob, L. Jiao and H. Wang, J. Appl. Phys., 2013, 114, 124101 CrossRef.
  31. Y. Gagou, J. Belhadi, B. Asbani, M. El Marssi, J. L. Dellis, Y. I. Yuzyuk, I. P. Raevski and J. F. Scott, Mater. Des., 2017, 122, 157–163 CrossRef CAS.
  32. T. E. Murphy, D. Chen and J. D. Phillips, Appl. Phys. Lett., 2004, 85, 3208 CrossRef CAS.
  33. W. Huang, Z. P. Wu and J. H. Hao, Appl. Phys. Lett., 2009, 94, 032905 CrossRef.
  34. L. Z. Hao, J. Zhu, Y. J. Liu, X. W. Liao, S. L. Wang, J. J. Zhou, C. Kong, H. Z. Zeng, Y. Zhang, W. L. Zhang and Y. R. Li, Thin Solid Films, 2012, 520, 6313–6317 CrossRef CAS.
  35. R. Thomas, J. F. Scott, D. N. Bose and R. S. Katiyar, J. Phys.: Condens. Matter, 2010, 22, 423201 CrossRef PubMed.


Electronic supplementary information (ESI) available. See DOI: 10.1039/c9ce00932a

This journal is © The Royal Society of Chemistry 2019