Journal of Physics D: Applied Physics
PAPER
Strain effects on conductivity and charge transport in La-doped BaTiO 3
thin films
To cite this article: Aihua Zhang et al 2020 J. Phys. D: Appl. Phys. 53 075305
View the article online for updates and enhancements.
This content was downloaded from IP address 210.28.139.88 on 07/05/2020 at 08:36
1. Introduction
Conductivity and charge transport in traditional insulating ferroelectric oxides have been cutting-edge research topics with rich physical connotations and wide-ranging prospects for applications in the field of ferroelectric materials [1–5].
Semiconductor and even metallic conduction in insulating fer- roelectric materials can be realized by various methods, such as ion doping [6, 7], oxygen reduction [8, 9], and so on. In the 1960s, Anderson and Blount developed a theory that intro- duced the idea of a polar metal [10], which suggested that two
seemingly incompatible states, such as metallic conduction and ferroelectricity, could exist simultaneously in a material.
Since then, significant theoretical [11–13] and experimental works [14–16] have been devoted to investigate the coexis- tence of a metallic phase and a ferroelectric or polar phase.
BaTiO3 (BTO) is a classic ferroelectric perovskite that can be used in various applications ranging from piezo-sensors and nonvolatile memories to resistive switching memories and photovoltaic solar cells [2, 17, 18]. Although stoichio- metric BTO is an insulator with a band gap of 3.2 eV [2], some results [3, 7–9, 19–21] indicate that BTO-based systems can
Journal of Physics D: Applied Physics
Strain effects on conductivity and charge transport in La-doped BaTiO 3 thin films
Aihua Zhang1, Qiang Li1, Dong Gao1, Min Guo1, Jiajun Feng2, Min Zeng1 , Zhen Fan1, Deyang Chen1, Xingsen Gao1, Guofu Zhou3,4, Xubing Lu1 and J-M Liu1,2
1 Guangdong Provincial Key Laboratory of Quantum Engineering and Quantum Materials and Institute for Advanced Materials, South China Academy of Advanced Optoelectronics, South China Normal University, Guangzhou 510006, People’s Republic of China
2 Laboratory of Solid State Microstructures and Innovation Center of Advanced Microstructures, Nanjing University, Nanjing 21009, People’s Republic of China
3 Guangdong Provincial Key Laboratory of Optical Information Materials, South China Academy of Advanced Optoelectronics, South China Normal University, Guangzhou 510006,
People’s Republic of China
4 National Center for International Research on Green Optoelectronics, South China Normal University, Guangzhou 510006, People’s Republic of China
E-mail: [email protected]
Received 10 October 2019, revised 12 November 2019 Accepted for publication 19 November 2019
Published 12 December 2019 Abstract
We investigate how epitaxial strain affects the conductivity and charge transport of Ba0.6La0.4TiO3 (BLTO) films deposited by pulsed laser deposition. Depositing BLTO films with thicknesses varying from 14.1 to 95.9 nm gives differing epitaxial strains, with a maximum strain up to ~6% for the thinnest film of 14.1 nm. Transport measurements demonstrate that film thickness (i.e. epitaxial strain) affects the conductivity, carrier mobility, and concentration. Measurements of resistivity versus temperature demonstrate that all the BLTO films undergo a distinct semiconductor-metal phase transition, and the transition temperature TSM also depends clearly on strain. For all the films, the charge transport follows the small-polaron hopping mechanism below TSM and the thermal phonon scattering mechanism above TSM. These results prove that strain in epitaxial films strongly affects the conductivity and charge transport of traditional insulating ferroelectric oxide films.
Keywords: La-doped BaTiO3 thin film, strain effect, charge transport, semiconductor-metal phase transition
(Some figures may appear in colour only in the online journal)
A Zhang et al
Printed in the UK 075305
JPAPBE
© 2019 IOP Publishing Ltd 53
J. Phys. D: Appl. Phys.
JPD
10.1088/1361-6463/ab590d
Paper
7
Journal of Physics D: Applied Physics IOP
2020
1361-6463
https://doi.org/10.1088/1361-6463/ab590d J. Phys. D: Appl. Phys. 53 (2020) 075305 (7pp)
A Zhang et al
2
exhibit semiconducting behavior, metal conductivity, or even metallic ferroelectricity, extending their range of application to novel optoelectronic devices. For decades, considerable effort has gone into the fabrication of conductive BTO oxides in both the bulk polycrystalline [22–25] and thin films [3, 7–9, 15, 20, 21, 26–28] by inserting ions of unequal valence into the Ba2+ or Ti4+ sites. In such a doped BTO system, some Ti4+ ions must be changed to Ti3+ ions to balance the elec- tric charge, and the resulting conductivity is due to the move- ment of 3d-orbital electrons among these Ti4+ and Ti3+ ions [19, 24]. For example, doping pentavalent (Nb5+, Ta5+) ions into Ti sites in BTO has been well studied to understand the insulating and/or semiconducting behavior [21]. Conversely, trivalent-ion (La3+, Y3+, Gd3+) doping into the Ba site has also been widely adopted [15, 22–25, 28]. La-doped BTO (Ba1−xLaxTiO3) is particularly intriguing. Takahashi et al prepared La-doped BTO epitaxial films with various doping contents and demonstrated the polar nature by band structure calculations in La-doped BTO with metallic conduction [15].
Strain in an epitaxial film, which is generated by the lattice mismatch between substrate and film, has been reported to significantly affect the structural, transport, and ferroelectric properties of materials [4, 29–36]. For example, Choi et al [4]
and Lyu et al [34] reported enhanced ferroelectricity in strained BTO thin films. Such strain engineering has been extensively used to induce the metal-insulator transition [29] and to pro- duce high-temperature perovskite ferromagnetic insulators [30]. In addition, epitaxial strain depends strongly on film thickness because strain relaxation occurs with increasing film thickness and the film relaxes beyond a critical thickness.
In the present work, BLTO thin films of varying thickness were prepared to obtain various strains and to investigate how strain affects the conductivity and charge-transport properties of BLTO thin films. The systematic results presented herein demonstrate how epitaxial strain critically affects the con- ductivity and charge transport of BLTO films, which should prove very helpful for clarifying the complex conduction and charge-transport mechanism in BLTO films and promote its application in various fields.
2. Experimental details
Ba0.6La0.4TiO3 (BLTO) films of varying thicknesses were grown epitaxially on (0 0 1) MgO single crystal substrates by using pulsed laser deposition with a KrF (λ = 248 nm) excimer laser (Coherent COMPexPro 205). For the deposi- tion, the substrate temperature was 650 °C, the laser energy density was 1.7 J·cm−2, the laser repetition rate was 2 Hz, and the pressure of the ambient oxygen was 3.0 × 10−4 Pa. After deposition, films were cooled down at the rate of 10 °C/min in a high vacuum. The film thickness was controlled by the number of laser pulses (i.e. 500, 1000, 1500, 2000, or 2500) and the film was characterized by using small-angle x-ray reflectivity.
The crystal structure of the films was determined by using x-ray diffraction (XRD, PANalytical X’Pert Pro diffractom- eter) with Cu Kα radiation (λ = 1.5406 Å). The surface
morphology of the BLTO films was investigated by using atomic force microscopy (AFM, Asylum Research Cypher).
The microstructure was determined by using cross-sectional high-resolution transmission electron microscopy (HRTEM, JEM2100F). The temperature dependence of the resistivity and the Hall Effect were determined by using a physical prop- erty measurement system (PPMS9, Quantum Design).
3. Results and discussion
Figure 1(a) shows x-ray diffraction patterns of BLTO films from 14.1 to 95.9 nm thick. All the films have (0 0 l) reflection planes, which correspond to the MgO (0 0 l) planes, indicating epitaxial growth of all the BLTO films. The epitaxial quality was further investigated by rocking curve measurements, from which the full width at half maximum value is very small (0.18°), implying very good epitaxial quality of the grown BLTO films. Note that the 14.1 nm-thick film has weak (0 0 1) and (0 0 2) peaks, and the (0 0 3) peak gradually appears with increasing film thickness. For clarity, the (0 0 1) and (0 0 2) dif- fraction peaks of BLTO films are presented on an expanded scale in figures 1(b) and (c), respectively. Note that both (0 0 2) and (0 0 1) reflection peaks are almost in the same 2θ value for the two thinner films (14.1 and 28.5 nm-thick films).
The reason is attributed to the high epitaxial strain. Further increasing the film thickness, they are shifted to a lower value of 2θ, which indicates an elongation of the out-of-plane lat- tice constant c due to the lattice relaxation [37]. Based on these reflection peaks and using Bragg’s law, the lattice constant c, defined as the out-of-plane orientation, increases from 4.089 Å for 14.1 nm-thick-film to 4.099 Å for 95.9 nm- thick-film. Figure 1(d) shows the small-angle x-ray reflec- tivity of BLTO films deposited on (0 0 1) MgO substrates.
The thickness of each film is calculated to be 14.1, 28.5, 52.5, 62.2, and 95.9 nm, which corresponds to the use of 500, 1000, 1500, 2000, and 2500 laser pulses, respectively, in the pulsed- laser deposition.
Figures 1(e) and (f) show two representative surface mor- phologies of the thinnest and thickest BLTO films, respec- tively. Although the film thickness varies from 14.1 to 95.9 nm, the film surfaces of all the films are very smooth with a root mean square roughness ranging from 142 to 270 pm.
In addition to the out-of-plane lattice parameters, the in- plane lattice parameters should also be determined to inves- tigate the epitaxial of BLTO/MgO heterostructures with different film thicknesses. Reciprocal space mapping (RSM) was done based on the (1 0 3) reflection, and the results are shown in figures 2(a)–(c) for the 14.1, 52.5, and 95.9 nm-thick films, respectively. These results reveal that all the BLTO films are of high epitaxial quality, and that the values of the epitaxial lattice parameters depend on the thickness.
On the basis of these RSM and XRD measurements, we confirmed all of the in-plane and out-of-plane lattice param- eters, which are summarized in figure 2(d). Upon increasing the film thickness from 14.1 to 95.9 nm, the lattice constant a of the BLTO films gradually decreases from 4.216 Å for the 14.1 nm-thick film to 3.997 Å for the 95.9 nm-thick film. The
J. Phys. D: Appl. Phys. 53 (2020) 075305
out-of-plane lattice parameter extracted by RSM is consistent with the results of XRD θ–2θ measurements (figure 1(a)).
Note that the diffraction intensity of the BLTO (1 0 3) peak of the thinnest film (figure 2(a)) is undefined due to the low intensity as a result of the low film thickness. The in-plane lat- tice constant a is calculated to be 4.216 Å based on the TEM result, which is discussed in detail below (figures 3(a)–(c)).
This result is the same as that for films deposited on MgO substrates (a = 4.216 Å [27]), which suggests that the film has the highest epitaxial strain.
The epitaxial strain ε in BLTO films was calculated by using the formula ε = (af − amin)/amin × 100%, where af is the lattice constant of BLTO films of various thicknesses, and amin is the minimum lattice constant of these films (i.e. for the completely relaxed BLTO film). The thinnest films are under tensile strain with ε ~ 6%, which decreases with increasing film thickness (see figure 2(e)). The strain ε approaches zero for the 52.5 nm-thick film, which indicates that the strain is completely relaxed at this thickness. The corresponding epi- taxial stress of BLTO films with different thicknesses is also evaluated in figure 2(e). The 14.1 nm-thick BLTO film has the largest stress of 19.0 GPa.
Cross-sectional TEM was also used to evaluate the thick- ness and epitaxial quality of the BLTO films, and the results are shown in figure 3 for the thinnest and thickest films (14.1 and 95.9 nm, see figures 3(a) and (d), respectively). These results are consistent with those of figure 1(d). Note that
the BLTO-MgO interfaces appear to be atomically sharp, as shown in figures 3(b) and (e) for the thinnest and thickest films, respectively. Moreover, figures 3(c) and (f) show the corre- sponding selected-area electron diffraction (SAED) patterns taken along the [0 1 0] MgO for the thinnest and the thickest film, respectively. The SAED spots from BLTO and MgO are sharp and identifiable, which indicates good single-crystal and epitaxial quality. Based on the TEM images and SAED results, we calculate the lattice parameters c and a for the two BLTO films to be c = 4.089 and 4.099 Å and a = 4.216 and 3.995 Å for the 14.1 and 95.9 nm-thick films, respectively, which is consistent with the RSM and XRD results.
To explore how strain affects the electrical transport proper- ties, we plot in figure 4(a) the electrical resistivity ρ as a func- tion of temperature T for the BLTO thin films with thickness ranging from 14.1 to 95.9 nm. The results show clearly that ρ depends strongly on thickness. Figure 4(b) plots ρ versus film thickness for temperatures of 80 K and 300 K. As the film thickness increases from 14.1 to 52.5 nm, ρ decreases for both low and high temperature, which is attributed mainly to strain relaxation. Upon further increasing the film thickness, the low-temperature resistivity increases slightly, whereas the high-temperature resistivity remains essentially constant.
Clearly, the mechanism responsible for the resistivity depends on temperature, as is discussed below.
The resistivity is low for relatively thick films (>52.5 nm, ρ ~0.1 Ω · cm), which is consistent with the fact that the lattice
Figure 1. (a) θ/2θ x-ray diffraction patterns of BLTO/MgO heterostructure with different film thicknesses. Expanded view of XRD peaks for (b) (0 0 1) and (c) (0 0 2) reflections. (d) X-ray reflectivity of BLTO films with thickness ranging from 14.1 to 95.9 nm. AFM image of (e) 14.1 nm-thick film and (f) 95.9 nm-thick film.
A Zhang et al
4
constants in these films approach that of bulk BLTO. As dem- onstrated by Fritsch et al in BLTO ceramics, a low resistivity ρ may be due to increased mobility of Ba and La ions [22]. In relatively thin films, an enhanced resistivity ρ may be due to the onset of tensile strain imposed by the substrate lattice. In a perovskite system, tensile strain is expected to increase the length and angle of the Ba–Ti or Ba–La bond and to influence
the 3d–2p orbital hybridization, which reduces hopping by charge carriers [21, 22, 32]. Many studies have reported that strain relaxation in epitaxial ferroelectric films may be helpful for modulating the electric transport properties of these systems [38, 39]. Yang et al reported that the compres- sive strain in an epitaxial ferroelectric film may induce an insulator–metal transition [39]. In contrast, a tensile strain
Figure 2. Reciprocal space mapping around (1 0 3) reflection from BLTO films (a) 14.1 nm thick, (b) 52.5 nm thick, and (c) 95.9 nm thick.
The dashed lines show the in- and out-of-plane peak positions for the film and substrate. (d) Lattice parameters a and c of BLTO films as a function of film thickness. (e) Strain and stress of BLTO films as a function of film thickness.
Figure 3. Cross-sectional HRTEM images and SAED patterns of epitaxial BLTO thin films deposited on MgO substrates. Panels (a)–(c) are for the 14.1 nm-thick film, and panels (d)–(f) are for 95.9 nm-thick film. The SAED measurement was made along the [0 1 0] MgO direction.
J. Phys. D: Appl. Phys. 53 (2020) 075305
enhances the insulating property. Furthermore, atomic-scale disorder or defects may also affect the electrical properties [40], although this phenomenon may be neglected in the pre- sent work because of the good epitaxial quality of the thin films (<100 nm).
Note that the ρ(T) curves reveal a clear semiconductor- metal transition. The transition temperature TSM and the strain state as a function of the film thickness are summarized in figure 4(c). With increasing film thickness, the TSM first decreases rapidly and then increases slightly. Similar rela- tionship is also found for the strain state. This implies that a direct relationship between TSM and strain state exists, and the dependence of TSM on the strain state is plotted in the inset of figure 4(c). It reveals that the larger is the strain, the higher is the TSM. In addition, the substrate-film lattice misfit induced strain also plays a critical role in determining the conductivity and charge transport behaviors. To differentiate between var- ious possible charge transport mechanisms in these BLTO thin films, we fit various models to the ρ(T) curves.
Figure 4(d) shows that, at high temperatures (T > TSM), ρ(T) is well fit by the thermal phonon scattering model, which predicts [21, 27]:
ρ∝T3/2,
(1)
Note that, at lower temperatures (T < TSM), the fit deviates from the thermal phonon scattering model, suggesting that this model is not valid at low temperatures. At low temper- atures, carriers hop smaller distances and are affected by mul- tiple activation energies, all of which is well described by the small-polaron hopping (SPH) model, which predicts [21, 41]:
ρ∝T3/2exp(WH/kBT),
(2) where WH is the thermal hopping activation energy, and kB
is the Boltzmann constant. Figure 4(e) shows that the low- temperature transport behavior fits well to the SPH model for all BLTO films studied.
Such a transition from thermally activated conduction to hopping conduction with decreasing temperature was previ- ously reported for Nb-doped BTO film [35, 37]. Jing et al attributed this type of transition to strongly localized electron states below TSM, which facilitates conduction in the SPH model [21]. The transition temperature TSM decreases with decreasing strain in the film, which implies that the charge- ordered state in films grows unstable with increasing strain.
Based on theory, Li et al predicted a decrease in TSM in BLTO thin films as tensile strain is relaxed [30]. In addition, we plot in figure 4(f) the activation energy WH extracted from the fits to the data as a function of thickness. The activation energy
Figure 4. (a) Resistivity ρ measured as a function of temperature T for BLTO films with various thicknesses. (b) Resistivity ρ at 80 K and 300 K as a function of BLTO film thickness. (c) Semiconductor-metal transition temperature TSM, and strain state ε as a function of BLTO film thickness, and the insert in (c) is the dependence of TSM on ε. Data for ρ as a function of temperature T fit (d) to thermal phonon scattering model for T > TSM and (e) to small-polaron hopping model for T < TSM. (f) Hopping energy given by fit in panel (e) to small- polaron hopping model plotted as a function of film thickness.
A Zhang et al
6
WH decays gradually with increasing thickness, implying that the energy band narrows as the film thickness increases, thereby allowing electrons to jump into the conduction band.
The slight increase in WH for the thicker films is attributed to film defects and/or to disorder among Ba and La atoms.
Figure 5(a) shows the Hall coefficient RH(T) measured as a function of temperature for BLTO films of various thick- nesses (14.1–95.9 nm). Note that RH(T) is negative, which indicates that electrons are the dominant carriers rather than holes, which is attributed to trivalent La3+ ion doping into the Ba-site. The interesting feature is that RH decreases gradu- ally with increasing temperature except for the 14.1 nm film, for which the opposite tendency is observed (below ~300 K), which proves that epitaxial strain strongly affects charge transport in these films. In addition, for BLTO films thicker than 52.5 nm, RH(T) is nearly independent of film thickness throughout the entire temperature range, which is attributed to the lack of strain in these thicker films (figure 2(e)).
Figure 5(b) shows the carrier density n determined from the Hall data as a function of temperature and for various film thicknesses. For the 14.1 nm-thick BLTO film, n(T) decreases with increasing temperature up to ~300 K, whereas n(T) increases with increasing temperature for the thicker films.
Note that n for all the films is on the order of 1021 cm−3, which is consistent with earlier reports [15].
Figure 5(c) shows the as-extracted carrier mobility µH as a function of T for all films studied herein. The car- rier mobility µH depends strongly on temperature for all the
films. It increases with increasing T below a critical temper- ature and then decreases with temperature above the critical temperature. The mobility of conduction-band electrons is typically thought to depend strongly on the scattering mech- anism. At low temperatures, scattering from ionized impu- rities dominates. Above the critical temperature, acoustic phonon scattering becomes dominant, and the mobility fol- lows µH∝T−3/2. Based on the results shown in figure 5(c), the high-temperature scattering mechanism in these films is most probably acoustic phonon scattering, whereas small- polaron hopping dominates in the low-temperature range.
These results are consistent with the resistivity results dis- cussed above.
Figures 5(d)–(f) show the Hall coefficient, carrier density, and carrier mobility at 300 K as a function of epitaxial strain.
It can be seen clearly that the thinner BLTO films with larger strain have a smaller carrier mobility, lower carrier density, and higher Hall coefficient, which implies that the epitaxial strain has a critical effect on carrier transport.
4. Conclusions
To summarize, we systematically investigated how epitaxial strain affects the microstructure and electrical transport nature of La-doped BTO films. The microstructural measurements, including AFM, HRTEM, and XRD, demonstrate that all the films studied were of good epitaxial quality, with atomic-level
Figure 5. (a) Hall coefficient RH, (b) carrier density n, and (c) Hall carrier mobility µH as functions of temperature and for various BLTO film thicknesses; (d) Hall coefficient RH, (e) carrier density n, and (f) Hall carrier mobility µH as functions of epitaxial strain. All the values shown in (d)–(f) were measured at 300 K.
J. Phys. D: Appl. Phys. 53 (2020) 075305
flat surfaces. Measurements of resistivity versus temperature demonstrated that the conductivity gradually increases with decreasing epitaxial strain. All the BLTO films undergo a semiconductor-metal transition, and the transition temper- ature first decreases and then increases with increasing film thickness. Measurements of the Hall Effect revealed that the majority carriers in these BLTO films are electrons, and their transport is by small-polaron hopping in the semiconductor phase and by thermal phonon scattering in the metallic phase.
In addition, the film thickness (i.e. epitaxial strain) clearly affects the carrier mobility and concentration. These find- ings provide deeper insights into the complex charge trans- port mechanism in conductive FE materials, which will boost potential applications in future information storage, sensors, and optoelectronic devices.
Acknowledgments
This work is supported by the National Natural Science Foundation of China (Contract No. 51431006, 51472093, 51872099). XB Lu acknowledges the support of the Proj- ect for Guangdong Province Universities and Colleges Pearl River Scholar Funded Scheme (2016). This work was also sup- ported by the Guangdong Innovative Research Team Program (No. 2013C102), the Guangdong Provincial Key Laboratory of Optical Information Materials and Technology (Grant No.
2017B030301007) and the 111 Project.
ORCID iDs
Min Zeng https://orcid.org/0000-0003-3594-7619 Xubing Lu https://orcid.org/0000-0002-3741-0500
References
[1] He X and Jin K-J 2016 Phys. Rev. B 94 224107
[2] Won C J, Park Y A, Lee K D, Ryu H Y and Hur N 2011 J. Appl. Phys. 109 084108
[3] Kolodiazhnyi T, Tachibana M, Kawaji H, Hwang J and Takayama-Muromachi E 2010 Phys. Rev. Lett.
104 147602
[4] Choi K J et al 2004 Science 306 1005–9 [5] Zhao X-W et al 2018 npj Quantum Mater. 3 52
[6] Huang C M, Lin C Y and Shieh J 2011 J. Phys. D: Appl. Phys.
44 345403
[7] Shao Y, Hughes R A, Dabkowski A, Radtke G, Gong W H, Preston J S and Botton G A 2008 Appl. Phys. Lett.
93 192114
[8] Jeong I K, Lee S H, Jeong S Y, Won C J, Hur N and Llobet A 2011 Phys. Rev. B 84 064125
[9] Page K, Kolodiazhnyi T, Proffen T, Cheetham A K and Seshadri R 2008 Phys. Rev. Lett. 101 205502
[10] Anderson P W and Blount E I 1965 Phys. Rev. Lett. 14 217–9 [11] Kim T H et al 2016 Nature 533 68–72
[12] Filippetti A, Fiorentini V, Ricci F, Delugas P and Íñiguez J 2016 Nat. Commun. 7 11211
[13] Ma C, Jin K-J, Ge C and Yang G-Z 2018 Phys. Rev. B 97 115103
[14] Shi Y et al 2013 Nat. Mater. 12 1024–7
[15] Takahashi K S, Matsubara Y, Bahramy M S, Ogawa N, Hashizume D, Tokura Y and Kawasaki M 2017 Sci. Rep.
7 4631
[16] Gu J-X, Jin K-J, Ma C, Zhang Q-H, Gu L, Ge C, Wang J- S, Wang C, Guo H-Z and Yang G-Z 2017 Phys. Rev. B 96 165206
[17] Si F, Tang B, Fang Z, Li H and Zhang S 2019 Ceram. Int.
45 17580–90
[18] Wen Z, Li C, Wu D, Li A and Ming N 2013 Nat. Mater.
12 617–21
[19] Raghavan S, Zhang J Y, Shoron O F and Stemmer S 2016 Phys. Rev. Lett. 117 037602
[20] Kolodiazhnyi T 2008 Phys. Rev. B 78 045107 [21] Jing X S et al 2017 Appl. Phys. Lett. 110 182903 [22] Fritsch V et al 2001 Phys. Rev. B 64 045113 [23] Mandal P 2005 Phys. Rev. B 71 165106
[24] Cai W, Fu C L, Lin Z B, Deng X L and Jiang W H 2011 Adv.
Mater. Res. 412 275–9
[25] Ben L B and Sinclair D C 2011 Appl. Phys. Lett. 98 092907 [26] Puli V S, Li P, Adireddy S and Chrisey D B 2015 J. Adv. Die.
05 1550027
[27] Liu L, Guo H, Lü H, Dai S, Cheng B and Chen Z 2005 J. Appl. Phys. 97 054102
[28] Ali A I, Park K, Ullah A, Huh R and Kim Y S 2014 Thin Solid Films 551 127–30
[29] Khare A et al 2017 Adv. Mater. 29 1606566
[30] Li Y L and Chen L Q 2006 Appl. Phys. Lett. 88 072905 [31] Yao D, Shi L, Zhou S, Liu H, Wang Y, Zhao J and Li Y 2016
J. Phys. D: Appl. Phys. 49 125301
[32] Zeng B, Long M, Zhang X, Dong Y, Li M, Yi Y and Duan H 2018 J. Phys. D: Appl. Phys. 51 235302
[33] Meng D et al 2018 Proc. Natl Acad. Sci. USA 115 2873–7 [34] Lyu J, Fina I, Solanas R, Fontcuberta J and Sánchez F 2018
Sci. Rep. 8 495
[35] Zheng X Y, Geng H, Yi Y P, Li Q K, Jiang Y Q, Wang D and Shuai Z G 2014 Adv. Funct. Mater. 24 5531–40
[36] Pang H, Zhang F, Zeng M, Gao X, Qin M, Lu X, Gao J, Dai J and Li Q 2016 npj Quantum Mater. 1 16015
[37] Ma C R, Liu M, Collins G, Liu J, Zhang Y M, Chen C L, He J, Jiang J C and Meletis E I 2012 Appl. Phys. Lett.
101 021602
[38] Fechner M, Ostanin S and Mertig I 2008 Phys. Rev. B 77 094112
[39] Yang L, Wang C A, Liu C, Qin M H, Lu X B, Gao X S, Zeng M and Liu J M 2016 Chin. Phys. B 25 077302 [40] Rotella H, Mazel Y, Brochen S, Valla A, Pautrat A, Licitra C,
Rochat N, Sabbione C, Rodriguez G and Nolot E 2017 J. Phys. D: Appl. Phys. 50 485106
[41] Kolodiazhnyi T and Wimbush S C 2006 Phys. Rev. Lett.
96 246404
