Electric field control of phase transition and tunable resistive switching

Subscriber access provided by EKU Libraries

Surfaces, Interfaces, and Applications

Electric field control of phase transition and tunable resistive switching in SrFeO2.5 Muhammad Shahrukh Saleem, Bin Cui, Cheng Song, Yiming Sun, Youdi Gu, Ruiqi Zhang, Muhammad Umer Fayaz, Xiaofeng Zhou, Peter Werner, Stuart S. P. Parkin, and Feng Pan ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b18251 • Publication Date (Web): 21 Jan 2019 Downloaded from http://pubs.acs.org on January 23, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Electric field control of phase transition and tunable resistive switching in SrFeO2.5

Muhammad Shahrukh Saleem,1 Bin Cui,2 Cheng Song,1,* Yiming Sun,1 Youdi Gu,1 Ruiqi Zhang,1 Muhammad Umer Fayaz,1 Xiaofeng Zhou,1 Peter Werner,2 Stuart S.P. Parkin,2 and Feng Pan1,* 1Key

Laboratory of Advanced Materials (MOE), School of Materials Science and

Engineering, Tsinghua University, Beijing 100084, China 2Max

Planck Institute for Microstructure Physics, 06120 Halle, Germany

E-mail: [email protected]; [email protected]

Abstract SrFeOx (SFOx) compound exhibits ionic conduction and oxygen-related phase transformation, having potential applications in solid-oxide fuel cells, smart windows, and memristive devices. The phase transformation in SFOx typically requires thermal annealing process under various pressure conditions, hindering their practical applications. Here we have achieved a reversible phase transition from brownmillerite (BM) to perovskite (PV) in SrFeO2.5 (SFO2.5) film through ionic liquid (IL) gating. The real-time phase transformation is imaged using in-situ high resolution transmission electron microscopy. The magnetic transition in SFO2.5 is identified by fabricating an assisted La0.7Sr0.3MnO3 (LSMO) bottom layer. The IL gating converted PV phase of SrFeO3-δ (SFO3-δ) layer shows a ferromagnetic-like behavior but applies a huge pinning effect on LSMO magnetic moments, which consequently leads to a prominent exchange bias phenomenon, suggesting an uncompensated helical magnetic structure of SFO3-δ. Whereas the suppression of both magnetic and 1

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

exchange coupling signals for BM phased SFO2.5 layer elucidates its fully compensated G-type antiferromagnetic nature. We also demonstrated that the phase transition by IL gating is an effective pathway to tune the resistive switching (RS) parameters, such as set, reset and high/low-resistance ratio in SFO2.5-based resistive random access memory devices.

Keywords: Electric field control of phase transition, tunable resistive switching, ionic liquid gating, SrFeO3, SrFeO2.5

2


Page 2 of 34

Page 3 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction Complex oxides possess widely tunable oxygen stoichiometry that can host a variety of phases1-3. For example, SrFeO2.5 (SFO2.5) undergoes oxygen vacancy (VO) tunable brownmillerite (BM) to perovskite (PV) structure transition along with a dramatic change in its intrinsic properties4-6. Also, the unoccupied lattices in the crystal and motion of oxygen vacancies make the SFO2.5 fundamentally important for ionic conduction, which is the key for a number of energy storage and environmental monitoring devices, such as fuel cells, smart windows and sensor technologies7-11. Therefore, the understanding and controlling of the oxygen content is significantly important to the device performances. Notably, the tremendous efforts have been made to control the oxygen-related phase transformation in SrFeO3-δ (SFO3-δ) since several decades ago, where the phase transformation typically requires the elevated temperature and specific pressure conditions11-14. However, the compatibility issues make this method impractical for many technological applications. Alternatively, in recent years, the ionic liquid (IL) gating method has been extensively employed to modulate the VO concentration in oxide materials at room temperatur15-17. It provides a unique opportunity for electric field controlled ions transport in a reversible manner that is relatively compatible with device and battery integration. In this study, we demonstrate the IL gating method to modulate the phase transformation in SFO2.5 which has not been reported yet. To gain more insight into the magnetic properties of SFO2.5, we deposited an additional La0.7Sr0.3MnO3 (LSMO) layer before the growth of SFO2.5. The robust magnetic sensitivity of manganite makes the LSMO-SFO2.5 an appropriate system for the study of magnetic phase transition in SFO2.5 during phase modulation18-19. Interestingly, in this study, with the help of exchange bias, the tuneable G-type AFM to helical magnetic transition is 3



identified. Ionic conductor is an important candidate of memristive devices, whose working is based on the resistive switching (RS) phenomena by controlling the cations or anions motions20-21. In respect of memristive devices, the SrFeOx (SFOx) is an attractive material due to its two resistance states: insulating BM phase and conducting PV phase4,11,12. Such a phase transformation property could introduce a new degree of freedom in tuning the RS behavior in memristive devices. In last few years, several methods have been employed to tune the RS behavior such as thickness variation, doping, and changing the crystal orientation22-24. However, these kinds of tuning parameters must be introduced during the fabrication of devices, and their effects are difficult to be altered once the devices are put into use. Although some other kinds of operating stimulants like light or magnetic field have also been tested to tune the RS parameters after fabrication of device25,26, the materials must be sensitive enough to fulfill the desired requirements. In fact, the underlying mechanism in oxide-based RS devices is attributed to the transport of ions (i.e. O2−) and VO distribution20,21. Therefore, with modifying the oxygen concentration inside the oxide layer of the device, it is highly expected that the RS parameters could be tuned more elegantly even after the fabrication of the devices. In the experiment below, we combine the IL gating method with SFOx based resistive random access memory (RRAM) cell. Then on the basis of O2− modulation inside the device by IL gating, we highlighted an effective way to control the operating parameters such as set, reset, and RS window. In previous works, we focus on the mechanisms of IL gating on SrCoOx: e.g. the anisotropic oxygen migration27 and the competition of phase transitions in SrCoOx and LSMO28. Based on these understandings, here we investigate the phase transition in SFOx system as well as the use of IL gating to tune the memory storage 4


Page 4 of 34

Page 5 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


devices with a stronger application background.

Experiment The LSMO-SFO2.5 bilayers were grown on (001) SrTiO3 (STO) substrate from stoichiometric La0.7Sr0.3MnO3 and SrFeO2.91 targets by using pulsed laser deposition (PLD) technique. Firstly, epitaxial 7.8 nm LSMO film was grown at 650 oC under 100 mTorr oxygen partial pressure. Then 30 nm layer SFO2.5 was deposited on top of the LSMO layer at the temperature of 750 oC under 20 mTorr oxygen partial pressure. The precise control of epitaxial growth was monitored by an in-situ reflection high-energy electron diffraction (RHEED) system (Supporting Information Figure S1). After the deposition, 2-hours post-growth annealing was carried out under 0.1 mTorr low oxygen pressure to get the good crystalline BM phase, and then sample was cooled down to room temperature with a rate of 10 oC per minute at the same oxygen pressure. Such growth and annealing parameters were also applied for single layer SFO2.5 on STO and Nb-doped (001) SrTiO3 (NSTO) substrates. Single layer SFO3-δ was achieved by a growth under 200 mTorr oxygen partial pressure, and cooled down to room temperature under 500 mTorr oxygen pressure without post-annealing. We use a target of intermediate oxygen content SrFeO2.91, which is suitable for the growth of both SFO2.5 and SFO3-δ films. The IL N, N-diethyl-N-(2-methoxyethyl)-N-ethyl ammonium bis-(trifluoromethyl sulfonyl)-imide (DEME-TFSI) was used as an electrolyte to gate the 5 × 2.5 mm2 area of LSMO-SFO2.5 heterostructure. For IL gating, the surface of the sample was covered by IL and then a bias voltage was applied between the LSMO-SFO2.5 heterostructure and an external aluminium top electrode. After gating, the IL was removed by washing the sample with acetone and then alcohol. Both gated and 5



pristine samples were used to characterize the x-ray diffraction (XRD), scanning transmission electron microscopy (STEM), x-ray absorption spectroscopy (XAS), superconducting quantum interface device (SQUID) magnetometer. For RRAM device, the BM phased SFO2.5 film was fabricated on conductive 0.7% Nb-doped (001) SrTiO3 substrate and then 70 nm platinum (Pt) pads with an area of 177 µm2 was deposited on the sample surface by magnetron sputtering unit.

Results and discussion The XRD θ-2θ scan of the pristine LSMO-SFO2.5 sample is given in Figure 1a. It clearly exhibits (080) plane diffraction at 45.7 degrees (see the enlarged right panel), in addition of half order (020), (040), (060) and (0100) plane diffractions. The half order peaks identify the doubling of the unit cell normal to the film surface, which is the b-axis of SFO2.5 crystal due to the alternate stacking of oxygen octahedral and tetrahedral layers29. The identified peaks are the characteristic signal of the BM phase11,29. The diffraction signal from LSMO was not identified, because it merges together with the signal of SrTiO3 substrate due to their close lattice parameters (aLSMO = 3.87 Å, aSTO = 3.90 Å). Then IL was dropped on the whole surface of bilayer sample and a negative 2 V d.c voltage was applied between aluminium gate electrode and LSMO-SFO2.5 film at room temperature as shown in Figure 1b. After applying a gate voltage (Vg) of –2 V for 45 minutes to the LSMO-SFO2.5 sample the ex-situ XRD was performed. It should be noted that the non-volatile feature of gating effect assures the feasibility of the ex-situ measurements17,30. The θ-2θ graph of sample gated by Vg = –2 V does not show any half order peaks and only one peak at 47.5 degrees was identified (see the curve labelled by –Vg). This newly emerged peak is the characteristic diffraction for 6


Page 6 of 34

Page 7 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


SFO3-δ perovskite structure that confirms the successful conversion of BM to PV phase11,31. After achieving the PV phase, we applied inverse d.c. bias to the same sample through IL gating. A positive Vg of 0.7 V was applied for 20 minutes, it reverts back the BM phase by extracting the oxygen ions from highly oxygenated SFO3-δ film (Figure 1c). In the reverse process, the half order peaks and (080) diffraction peak appear again at the same positions as they were in pristine BM phase (+Vg curve in Figure 1a). Hence, the well-defined and high-quality diffraction pattern confirms the feasibility of the reversible phase transition in SFOx via IL gating. Moreover, no any noticeable damage at the surface of the film was found, indicating the reliability of the IL gating experiment (see the Supporting Information Figure S2). In fact, fast topotactic phase transformations were found in SFOx by annealing the sample in different oxygen environments for ~10 minutes. However, relative high temperatures of 300–700 °C are still needed to realize the completed and homogeneous phase transition in the thin films11,31. Taking the fast decayed oxygen diffusion coefficient with the decrease of temperature into consideration (0.5 and 2 orders per 100 °C for some BM and PV phase oxides, respectively32), the gating time of 45 minutes at room temperature in our work is reasonable. Furthermore, Fe4+ in SFO3-δ is not a stable valence state and requires strong oxidizing conditions14, which might also induce a longer gating time in our case. For direct visualization of crystalline structure, a high-angle annular dark field scanning transmission electron microscopy (HAADF-STEM) was measured for gated samples. The cross-sectional images of the gated samples are given in Figure 2. In the negative gated sample, a fully oxygenated uniform perovskite structure was observed at SFO3-δ region (Figure 2a), while in the positive gated sample the alternate stacking of oxygen deficient and fully oxygenated layers was observed (Figure 2b). These 7



alternate layers manifested by the appearance of dark and bright stripes, implying the staking of tetrahedral and octahedral sites layer. The oxygen migration could be realized by O2– jumping to neighbouring vacancy. The ordered vacancy channels in the FeO4 tetrahedral of BM phase compound provide the easiest diffusion pathway with lowest energy barrier32-34, which guarantee the lateral oxygen ion diffusion in our thin films. Nevertheless, the continuous phase transition throughout the whole thin film also calls for oxygen migration perpendicular to the FeO4 and FeO6 layers through vacancy jumping. Although the interstitial and interstitialcy mechanisms are also generally accepted in many systems32, we believe that the vacancy mechanism of diffusion plays a prominent role in our case. Furthermore, we performed the in-situ transmission electron microscopy (TEM) characterization to demonstrate the dynamic process of structural transition with time during IL gating (Figure 2c). The experiment detail of the in-situ transmission electron microscopy setup is given in Ref. 27. A series of images have been shown with respect to 15 minutes time elapsed, in which gradual phase transition from BM to PV can be directly visualized (Figure 2c). Once a negative Vg of –2 V was applied to the LSMO-SFO2.5 bilayer sample, the injected oxygen ions start to fill the oxygen vacant tetrahedral sites in the SFO2.5 layer and resultantly dark and bright strips contrast of alternate layers start to disappear gradually with time. Finally, a complete homogeneous highly oxygenated PV phase was achieved after 45 minutes. It is noteworthy that the phase transition from SFO2.5 to SFO3-δ happens firstly at the interface of SFO/LSMO. In the in-situ TEM experiment, ionic liquid is close to but not on the imaged area27. Thus the phase transition is mainly caused by the lateral diffusion of oxygen ions in the thin film. Compared with the bulk of thin film, the interface between SFO2.5 and LSMO should have more defects or vacancies35, which 8


Page 8 of 34

Page 9 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


could supply a faster oxygen transport lane and result in the prior phase transition near the SFO/LSMO interface. Meanwhile, such a lateral oxygen migration could also be favored by ordered oxygen vacancy channels in SFO2.5 parallel to interface. This phase modulation can also be identified by initial and final stage super-lattice reflections in the FFT (fast Fourier transform) patterns, given in Figure 2d and 2e. The circled reflections at initial 0-minute stage was removed after in-situ 45-minutes gating, indicating the disappearance of BM phased supper-lattice and formation of the PV phase. It is important to know that the dissolved oxygen (or H2O) inside the ionic liquid cannot be completely removed even in a high vacuum environment, which can supply enough oxygen ion for the phase transition during the gating process15,27. IL gating enables not only the structural change but also the electronic structure variation in the SFO2.5 layer. During a cycle of positive and negative gating, the elemental-resolved details of iron valence state and oxygen stoichiometry were investigated by XAS. The oxygen K-edge spectra of LSMO-SFO2.5 samples of different gating states are given in Figure 3a. They were collected in total electron yield (TEY) mode. Since the probing depth of TEY is several nanometres (6–8 nm) that ensure the signal entirely come from the SFO2.5 top layer36. The pre-edge of oxygen K-edge spectra consist of two peaks at a and b positions which belong to the O 2p−Fe 3d band hybridization37,38. In terms of oxygen concentration, an obvious difference was found between two phases. The pristine state only shows one peak at position b and complete absence of the signal at the position a. It indicates the loss of spectral feature associated with O 2p−Fe 3d hybridization37-38. On the other hand, the −Vg state of sample exhibit two peaks at a and b positions, indicating the increase of oxygen content in SFO3-δ as compared to the pristine state. Subsequently, after applying +Vg to the same sample, the single peak at a disappears again, signifying the 9



Page 10 of 34

gap opening between O 2p−Fe 3d band hybridization. The feature of spectral loss favours the existence of VO and insulating behaviour of positive gated sample. Indeed, the change of oxygen stoichiometry is also related to the valence states of Fe cations. Therefore, oxygen stoichiometry was also determined from Fe L-edge spectra given in Figure 3b. Spin-orbit interaction split the Fe spectra into 2p3/2 and 2p1/2 multiples39. In comparison, the peak positions of the 2p3/2 and 2p1/2 shoulders slightly shift toward the higher energy side in the −Vg state as compared to that of the +Vg state. Approximately 0.35 eV energy shift was observed, suggesting an increase of Fe valance after negative gating to the sample. Moreover, both the −Vg and +Vg states show considerable change at 2p1/2 region. The shape of the Fe multiples for the −Vg state (i.e. 2p3/2 and 2p1/2) has the consistency with previous reported SFO3 spectra39-40.

The

SFO3

compound

should

mainly

possesses

3d4

(Fe4+)

configuration14,41. But a considerable amount of 3d5 (Fe3+) configuration was also reported because SFO3 is in the negative charge-transfer regime with holes in the O 2p levels39. It might be the reason why only a 0.35 eV shift in the XAS of Fe is observed in our SFO2.5-SFO3 phase change, and also the reason of broad pre-edge peak due to strong covalence in O K-edge spectra. Nevertheless, SFO2.5 in the +Vg state shows two peaks with equal intensity at 2p1/2 region, which are similar to the previously reported spectra of LaFeO3 with 3d5 (Fe3+) electronic configuration40,42. It confirms the valence of Fe in the +Vg state is close to +3. The XAS results of +Vg and –Vg gated SFO-based heterostructure samples shows similar peak position and shape to those of single layer SFO2.5 and SFO3-δ thin films prepared by different growth parameters (Supporting Information Figure S3). The correlation between electronic and magnetic properties motivates us to elucidate the magnetic behaviors in SFOx with different oxygen vacancy 10


Page 11 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


concentrations. In-plane hysteresis loops (M-H) along the [100] direction for pristine and gated LSMO-SFO2.5 samples were measured by SQUID magnetometer at 10 K (Figure 4a). A 5 kOe cooling field was applied before getting the M-H curves. It can be easily observed that the hysteresis loop for the −Vg state prominently shifts toward the negative field direction with an exchange bias field (HEB) of 150 Oe. While the hysteresis loops for both pristine and +Vg state only show very tiny HEB of 15 and 28 Oe, respectively, together with less saturation magnetization (Ms) comparative to −Vg state. The exchange bias in SFO3–x/LSMO should not be caused by the interfacial strain because of the negligible interfacial lattice mismatch of 0.5% (aSFO3 = 3.85 Å and aLSMO = 3.87 Å). To understand the coupling behaviour in bilayer system, the magnetic hysteresis loops of single layer SFO2.5 and SFO3-δ were also measured. The thicknesses and sizes of the single layer samples are the same as that of bilayer sample. As shown in the Figure 4b, the single layer SFO3-δ sample shows a ferromagnetic (FM)-like loop with a Ms value of 0.65×10–5 emu. In contrast, no significant magnetic signal was observed in SFO2.5 single layer sample. It is well understood that the exchange bias phenomena is highly related to the pinning of FM spin caused by the uncompensated AFM spins at the interface. Despite the fact of magnetic signal in SFO3-δ single layer, the SFO3-δ at low temperature possess helical magnetic order which is actually tilted AFM structure4,43-45. The helical magnetic structure is an uncompensated AFM structure in which Fe4+-Fe4+ spins are not completely antiparallel but at some angle (Supporting Information Figure S4). Therefore SFO3-δ itself could exhibit weak but ferromagnetic-like signal in magnetic measurements as we observed in the single layer. Remarkably, the uncompensated tilted spin of SFO3-δ at the interface is expected to induce the exchange bias phenomena in LSMO-SFO3-δ system, along with the increase of 11



saturation magnetization due to the combined magnetic signal from SFO3-δ and LSMO. The increase of magnetic signal in the −Vg states is comparable with the Ms value of single layer SFO3-δ sample. On the other hand, it is frequently given in literature that the combination of fully compensated G-type AFM with FM compound does not pursue the spin pinning effects at the interface and resultantly exchange biased phenomenon cannot be originated46,47. Our results evidently support that: the SFO2.5 in pristine and +Vg gating states are indeed G-type AFM compounds and negligible exchange bias fields are found. Note that the coercive values of samples in pristine, +Vg and −Vg states are 159, 186 and 189 Oe, respectively, which are close to each other but much bigger than that usually observed in single LSMO layer (e.g. –2V) in Figure 5b, c and f are corresponding to a backward rectification that might be attributed to the nanobattery effect52. Compared with other stages, the nanobattery effect is more obvious in 20 minutes gating stage. Apparently this effect changes the shape and window of the I-V curve. Consequently, the values of the window and HRS become larger than those of 10 minute gating state. In fact, the 20 minute stage devices attain 16


Page 16 of 34

Page 17 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


a low resistance state just at 1.95 V as compared to the 3.6 V for 10 minute gating stage. It represents the high oxygen concentration and easy formation of conducting filament as compared to the 10 minute gating stage. Previous works have shown excellent RS properties (low switching voltages and high uniformity) for SFO2.524,29. Different from the previous ones, this study shows that the IL gating here directly influences the leakage conduction and resultant resistive hysteresis of the SFO2.5 functional layer. Our results demonstrate that IL gating is an effective way to tuning the resistive switching (RS) parameters, such as set, reset and window (high/low-resistance ratio) in RRAM even after the fabrication of the device.

Conclusion We have demonstrated the phase transition in SFO2.5 layer by using IL gating in the presence of LSMO assistant layer that was used as an electrode. During the phase modulation, the helical magnetic order in SFO3-δ and G-type AFM magnetic structure in SFO2.5 produces prominent exchange bias phenomenon in adjacent FM LSMO layer which might be significant for the development of multifunctional materials and exchange coupling devices in spintronics. Furthermore, we highlight, the combination of IL method with RRAM device provides an effective way to tune the set, reset operating-voltages and the ratio of HRS/LRS by changing the mobile ions concentration.

Acknowledgements: C.S. acknowledges the support of Young Chang Jiang Scholars Program and Beijing Advanced Innovation Center for Future Chip (ICFC). B.C. thanks the Alexander von Humboldt Foundation for their support. This work was supported by the National Key 17



R&D Program of China (Grant No. 2017YFB0405604) and the National Natural Science Foundation of China (Grant Nos. 51571128 and 51671110). B.C., P.W., and S.S.P.P. acknowledge partial funding from the EU H2020 program “Phase Change Switch”.

Supporting Information The growth dynamics of samples, atomic force microscope images, XAS spectra of Oxygen K-edge and iron L-edge, Magnetic structure of SrFeO3, Temperature dependent magnetization measurement,

References (1) Tsujimoto, Y.; Tassel, C.; Hayashi, N.; Watanabe, T.; Kageyama, H.; Yoshimura, K.; Takano, M.; Ceretti, M.; Ritter, C.; Paulus, W. Infinite-layer Iron Oxide With a Square-Planar Coordination. Nature 2007, 450, 1062-1065. (2) Gu, Y., D.; Song, C.; Zhang, H.; Wang, Z,; Cui, B., Li, F., Peng, J., J.; Saleem, M., S.; Wang, G.; Zhong, X.; Wang, F.; Ma, S.; Sun, J.; Liu, W.; Pan, F.; Zhang, Z. Controllable Oxygen Vacancies, Orbital Occupancy and Magnetic Ordering in SrCoO3-δ Flms. J. Magn. Magn. Mater. 2018, 454, 228-236. (3) Wu, H.-C.; Lin, Y. S. Effects of Oxygen Vacancy Order–Disorder Phase Transition on Air Separation by Perovskite Sorbents. Ind. Eng. Chem. Res. 2017, 56, 6057-6064. (4) Lebon, A.; Adler, P.; Bernhard, C.; Boris, A. V.; Pimenov, A. V.; Maljuk, A.; Lin, C. T.; Ulrich, C.; Keimer, B. Magnetism, Charge Order, and Giant Magnetoresistance in SrFeO3-δ Single Crystals. Phys. Rev. Lett. 2004, 92, 037202. (5) Hodges, J. P.; Short, S.; Jorgensen, J. D.; Xiong, X.; Dabrowski, B.; Mini, S. M.; 18


Page 18 of 34

Page 19 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Kimball, C. W. Evolution of Oxygen-Vacancy Ordered Crystal Structures in the Perovskite Series SrnFenO3n−1 (n=2, 4, 8, and ∞), and the Relationship to Electronic and Magnetic Properties. J. Solid State Chem. 2000, 151, 190-209. (6) Galakhov, V. R.; Kurmaev, E. Z.; Kuepper, K.; Neumann, M.; McLeod, J. A.; Moewes, A.; Leonidov, I. A.; Kozhevnikov, V. L. Valence Band Structure and X-ray Spectra of Oxygen-Deficient Ferrites SrFeOx. J. Phys. Chem. C 2010, 114, 5154-5159. (7) Cheng, J.; Navrotsky, A. Thermochemistry of La1-xSrxFeO3 -δ Solid Solutions (0.0 ≤ x ≤ 1.0, 0.0 ≤ δ ≤ 0.5). Chem. Mater. 2005, 17, 2197-2207 (8) Wang, Y.; Chen, J.; Wu, X. Preparation and gas-sensing properties of perovskite-type SrFeO3 oxide. Mater. Lett. 2001, 49, 361-364 (9) Su, C.; Wang, W.; Liu, M.; Tadé, M. O.; Shao, Z. Progress and Prospects in Symmetrical Solid Oxide Fuel Cells with Two Identical Electrodes. Adv. Energy Mater. 2015, 5, 1500188. (10)Bakken, E.; Norby, T.; Stølen, S. Redox Energetics of Perovskite-Related Oxides. J. Mater. Chem. 2002, 12, 317-323. (11)Khare, A.; Shin, D.; Yoo, T. S.; Kim, M.; Kang, T. D.; Lee, J.; Roh, S.; Jung, I. H.; Hwang, J.; Kim, S. W.; Noh, T. W.; Ohta, H.; Choi, W. S. Topotactic Metal-Insulator Transition in Epitaxial SrFeOx Thin Films. Adv. Mater. 2017, 29, 1606566. (12)Yamada, H.; Kawasaki, M.; Tokura, Y. Epitaxial Growth and Valence Control of Strained Perovskite SrFeO3 Films. Appl. Phys. Lett. 2002, 80, 622-624. (13)Beurmann, P. S.; Thangadurai, V.; Weppner, W. Phase Transitions in the SrSnO3–SrFeO3 Solid Solutions: X-ray Diffraction and Mössbauer Studies. J. Solid State Chem. 2003, 174, 392-402. 19



(14)MacChesney, J. B.; Sherwood, R. C.; Potter, J. F. Electric and Magnetic Properties of the Strontium Ferrates. J. Chem. Phys. 1965, 43, 1907-1913. (15)Lu, N.; Zhang, P.; Zhang, Q.; Qiao, R.; He, Q.; Li, H. B.; Wang, Y.; Guo, J.; Zhang, D.; Duan, Z.; Li, Z.; Wang, M.; Yang, S.; Yan, M.; Arenholz, E.; Zhou, S.; Yang, W.; Gu, L.; Nan, C. W.; Wu, J.; Tokura, Y.; Yu, P. Electric-field Control of Tri-State Phase Transformation With a Selective Dual-Ion Switch. Nature 2017, 546, 124-128. (16)Li, M.; Han, W.; Jiang, X.; Jeong, J.; Samant, M. G.; Parkin, S. S. Suppression of Ionic Liquid Gate-Induced Metallization of SrTiO3(001) by Oxygen. Nano Lett. 2013, 13, 4675-4678. (17)Cui, B.; Song, C.; Mao, H.; Yan, Y.; Li, F.; Gao, S.; Peng, J.; Zeng, F.; Pan, F. Manipulation of Electric Field Effect by Orbital Switch. Adv. Funct. Mater. 2016, 26, 753-759. (18)Li, F.; Song, C.; Wang, Y. Y.; Cui, B.; Mao, H. J.; Peng, J. J.; Li, S. N.; Wang, G. Y.; Pan, F. Tilt Engineering of Exchange Coupling at G-type SrMnO3/(La,Sr)MnO3 Interfaces. Sci. Rep. 2015, 5, 16187. (19)Wu, S. M.; Cybart, S. A.; Yi, D.; Parker, J. M.; Ramesh, R.; Dynes, R. C. Full Electric Control of Exchange Bias. Phys. Rev. Lett. 2013, 110, 067202. (20)Yang, J. J.; Pickett, M. D.; Li, X.; Ohlberg, D. A.; Stewart, D. R.; Williams, R. S. Memristive Switching Mechanism For Metal/Oxide/Metal Nanodevices. Nat. Nanotechnol. 2008, 3, 429-433. (21)Skinner, J. S.; Kilner, A. J. Oxygen Ion Conductors Mater. Today 2003, 6, 30-37. (22)Gao, S.; Zeng, F.; Wang, M.; Wang, G.; Song, C.; Pan, F. Tuning the Switching Behavior of Binary Oxide-Based Resistive Memory Devices by Inserting an Ultra-Thin Chemically Active Metal Nanolayer: A Case Study on the Ta2O5-Ta 20


Page 20 of 34

Page 21 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


System. Phys. Chem. Chem. Phys. 2015, 17, 12849-12856. (23)Kim, S.; Choi, S. H.; Lee, J.; Lu D. W. Tuning Resistive Switching Characteristics of Tantalum Oxide Memristors Through Si Doping. ACS Nano 2014, 8, 10262-10269 (24)Acharya, S. K.; Jo, J.; Raveendra, N. V.; Dash, U.; Kim, M.; Baik, H.; Lee, S.; Park, B. H.; Lee, J. S.; Chae, S. C.; Hwang, C. S.; Jung, C. U. Brownmillerite Thin Films as Fast Ion Conductors For Ultimate-Performance Resistance Switching Memory. Nanoscale 2017, 9, 10502-10510. (25)Sun, Y.; Tai, M.; Song, C.; Wang, Z.; Yin, J.; Li, F.; Wu, H.; Zeng, F.; Lin, H.; Pan, F. Competition Between Metallic and Vacancy Defect Conductive Filaments in a CH3NH3PbI3-Based Memory Device. J. Phys. Chem. C 2018, 122, 6431-6436. (26)Sun, B.; Liu, Y.; Zhao, W.; Chen, P. Magnetic-Field and White-Light Controlled Resistive Switching Behaviors in Ag/[BiFeO3/γ-Fe2O3]/FTO device. RSC Advances 2015, 5, 13513-13518. (27)Cui, B.; Werner, P.; Ma, T.; Zhong, X.; Wang, Z.; Taylor, J. M.; Zhuang, Y.; Parkin, S. S. P. Direct Imaging of Structural Changes Induced by Ionic Liquid Gating Leading to Engineered Three-Dimensional Meso-Structures. Nat. Commun. 2018, 9, 3055. (28)Cui, B.; Song, C.; Li, F.; Zhong, X. Y.; Wang, Z. C.; Werner, P.; Gu, Y. D.; Wu, H. Q.; Saleem, M. S.; Parkin, S. S. P.; Pan, F. Electric-Field Control of Oxygen Vacancies and Magnetic Phase Transition in a Cobaltite/Manganite Bilayer. Phys. Rev. Appl. 2017, 8, 044007. (29)Acharya, S. K.; Nallagatla, R. V.; Togibasa, O.; Lee, B. W.; Liu, C.; Jung, C. U.; Park, B. H.; Park, J. Y.; Cho, Y.; Kim, D. W.; Jo, J.; Kwon, D. H.; Kim, M.; Hwang, C. S.; Chae, S. C. Epitaxial Brownmillerite Oxide Thin Films for Reliable Switching 21



Memory. ACS Appl. Mater. Interfaces 2016, 8, 7902-7911. (30)Cui, B.; Song, C.; Gehring, G. A.; Li, F.; Wang, G.; Chen, C.; Peng, J.; Mao, H.; Zeng, F.; Pan, F. Electrical Manipulation of Orbital Occupancy and Magnetic Anisotropy in Manganites. Adv. Funct. Mater. 2015, 25, 864-870. (31)Roh, S.; Lee, S.; Lee, M.; Seo, Y.-S.; Khare, A.; Yoo, T.; Woo, S.; Choi, W. S.; Hwang, J.; Glamazda, A.; Choi, K. Y. Oxygen Vacancy Induced Structural Evolution of SrFeO3−x Epitaxial Thin Film From Brownmillerite to Perovskite. Phys. Rev. B 2018, 97, 075104. (32)Chroneos, A.; Yildiz, B.; Tarancón, A.; Parfitt, D.; Kilner, J. A. Oxygen Diffusion in Solid Oxide Fuel Cell Cathode and Electrolyte Materials: Mechanistic Insights From Atomistic Simulations. Energy & Environ. Sci. 2011, 4, 2774-2789. (33)Paulus, W.; Schober, H.; Eibl, S.; Johnson, M.; Berthier, T.; Hernandez, O.; Ceretti, M.; Plazanet, M.; Conder, K.; Lamberti, C. Lattice Dynamics To Trigger Low Temperature Oxygen Mobility in Solid Oxide Ion Conductors. J. Am. Chem. Soc. 2008, 130, 16080-16085 (34)Mitra, C.; Meyer, T.; Lee, H. N.; Reboredo, F. A. Oxygen Diffusion Pathways in Brownmillerite SrCoO2.5: Influence of Structure and Chemical Potential. J. Chem. Phys. 2014, 141, 084710. (35)Navickas, E.; Chen, Y.; Lu, Q.; Wallisch, W.; Huber, T. M.; Bernardi, J.; Stoger-Pollach, M.; Friedbacher, G.; Hutter, H.; Yildiz, B.; Fleig, J. Dislocations Accelerate Oxygen Ion Diffusion in La0.8Sr0.2MnO3 Epitaxial Thin Films. ACS Nano 2017, 11 , 11475-11487. (36)Saleem, M. S.; Song, C.; Peng, J. J.; Cui, B.; Li, F.; Gu, Y. D.; Pan, F. Metal-Insulator-Metal Transition in NdNiO3 Films Capped by CoFe2O4. Appl. Phys. Lett. 2017, 110, 072406. 22


Page 22 of 34

Page 23 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(37)Galakhov, V. R.; Kurmaev, E. Z.; Kuepper, K.; Neumann, M.; McLeod, J. A.; Moewes, A.; Leonidov, I. A.; Kozhevnikov, V. L. Valence Band Structure and X-ray Spectra of Oxygen-Deficient Ferrites SrFeOx. J. Phys. Chem. C 2010, 114, 5154-5159 (38)Karvonen, L.; Valkeapää, M.; Liu, R.-S.; Chen, J.-M.; Yamauchi, H.; Karppinen, M. O-K and Co-L XANES Study on Oxygen Intercalation in Perovskite SrCoO3-δ. Chem. Mater. 2010, 22, 70-76. (39)Abbate, M.; Zampieri, G.; Okamoto, J.; Fujimori, A.; Kawasaki, S.; Takano, M. X-ray Absorption of the Negative Charge-Transfer Material SrFe1−xCoxO3. Phys. Rev. B 2002, 65, 165120. (40) Abbate, M.; de Groot, F. M. F.; Fuggle, J. C.; Fujimori, A.; Strebel, O.; Lopez, F.; Domke, M.; Kaindl, G.; Sawatzky, G. A.; Takano, M.; Takeda, Y.; Eisaki, H.; Uchida, S. Controlled-valence properties of La1−xSrxFeO3 and La1−xSrxMnO3 studied by soft-x-ray absorption spectroscopy. Phys. Rev. B 1992, 46, 4511-4519. (41)Takeda, T.; Yamaguchi, Y.; Watanabe, H. Magnetic Structure of SrFeO3. J. Phys. Soc. Jpn. 1972, 33, 967-969 (42)Lüning, J.; Nolting, F.; Scholl, A.; Ohldag, H.; Seo, J. W.; Fompeyrine, J.; Locquet, J. P.; Stöhr, J. Determination of the Antiferromagnetic Spin Axis in Epitaxial LaFeO3 Films by X-Ray Magnetic Linear Dichroism Spectroscopy. Phys. Rev. B 2003, 67, 214433. (43)Zhao, Y. M.; Zhou, P. F. Metal–Insulator Transition in Helical SrFeO3−δ Antiferromagnet. J. Magn. Magn. Mater. 2004, 281, 214-220. (44)Bocquet, A. E.; Fujimori, A.; Mizokawa, T.; Saitoh, T.; Namatame, H.; Suga, S.; Kimizuka, N.; Takeda, Y.; Takano, M. Electronic Structure of SrFe4+O3 and Related Fe Perovskite Oxides. Phys. Rev. B 1992, 45, 1561-1570. (45)Reehuis, M.; Ulrich, C.; Maljuk, A.; Niedermayer, C.; Ouladdiaf, B.; Hoser, A.; 23



Hofmann, T.; Keimer, B. Neutron Diffraction Study of Spin and Charge Ordering in SrFeO3−δ. Phys. Rev. B 2012, 85, 184109. (46)Nowak, U.; Usadel, K. D.; Keller, J.; Miltényi, P.; Beschoten, B.; Güntherodt, G. Domain State Model for Exchange Bias. I. Theory. Phys. Rev. B 2002, 66, 014430. (47)Dong, S.; Yamauchi, K.; Yunoki, S.; Yu, R.; Liang, S.; Moreo, A.; Liu, J. M.; Picozzi, S.; Dagotto, E. Exchange Bias Driven by the Dzyaloshinskii-Moriya Interaction and Ferroelectric Polarization at G-type Antiferromagnetic Perovskite Interfaces. Phys. Rev. Lett. 2009, 103, 127201. (48) Huijben, M.; Martin, L. W.; Chu, Y. H.; Holcomb, M. B.; Yu, P.; Rijnders, G.; Blank, D. H. A.; Ramesh, R. Critical Thickness and Orbital Ordering in Ultrathin La0.7Sr0.3MnO3 Films. Phys. Rev. B 2008, 78, 094413. (49) Zhao, Y. M.; Mahendiran, R.; Nguyen, N.; Raveau, B.; Yao, R. H. SrFeO2.95: A Helical Antiferromagnet with Large Magnetoresistance. Phys. Rev. B 2001, 64, 024414. (50)Takeda, T.; Yamaguchi, Y.; Watanabe, H.; Tomiyoshi, S.; Yamamoto, H. Crystal and Magnetic Structure of Sr2Fe2O5. J. Phys. Soc. Jpn. 1969, 26, 1320 (51)Schmidt, M.; Hofmann, M.; Campbell, S. J. Magnetic Structure of Strontium Ferrite Sr4Fe4O11, J. Phys.: Condens. Matter 2003, 15, 8691–8701. (52) Midya, R.; Wang, Z.; Zhang, J.; Savel'ev, S. E.; Li, C.; Rao, M.; Jang, M. H.; Joshi, S.; Jiang, H.; Lin, P.; Norris, K.; Ge, N.; Wu, Q.; Barnell, M.; Li, Z.; Xin, H. L.; Williams, R. S.; Xia, Q.; Yang, J. J. Anatomy of Ag/Hafnia-Based Selectors with 1010 Nonlinearity. Adv. Mater. 2017, 29, 1604457

24


Page 24 of 34

Page 25 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 1. Gating time and corresponding RS parameters. The ratio of HRS/LRS was taken at –2 V from I-V curves. –Vg

Resistance

Set VSET Reset

(min.)

(Ω)

(+V)

HRS/LRS

VRESET (–V)

0

1 GΩ

unstable unstable

10

190 MΩ

3.6

6.5

104

20

110 MΩ

1.95

9.7

105

30

9 MΩ

1.92

9.7

103

40

800 kΩ

1.7

9.7

101

25


unstable


Figure Captions Figure 1. (a) XRD θ-2θ patterns of pristine, negative and positive gated LSMO-SFO2.5 samples. The right side panel shows the enlarged view of BM-(080) and PV-(020) plane peak positions. (b) and (c) are the schematic cross-sectional view of LSMO-SFO2.5 sample under negative and positive gate voltages respectively. The applied potential induces the redistribution of TFSI− (red spheres with negative sign) and DEME+ (blue spheres with positive sign) ions at the surface of the SFO2.5 layer and gate electrode respectively. Small red color spheres indicate the O2− ions. With a negative gate bias (–Vg), the driving potential of electric double layer (EDL) can insert the O2− ions into the SFO2.5 layer, while with positive gate biasing (+Vg), the O2− ions can extract from the SFO2.5 layer as indicated by arrows.

Figure 2. (a) and (b) HAADF-STEM images of LSMO-SFO2.5 bilayer after negative and positive IL gating respectively. (c) The series of in-situ TEM images with respect to time under the negative biased –2 V gate voltage. Each image was taken after 15 minutes. The scale-bars indicate the 2 nm length. (d) and (e) are the FFT images taken at the position of SFO2.5 when the gating times were 0 and 45 minutes respectively.

Figure 3. (a) Oxygen K-edge (b) Iron L-edge XAS curves of pristine, positive and negative gated LSMO-SFO2.5 sample. The shaded region of oxygen K-edge spectra denote the Fe-O hybridization, and in iron L2,3-edge spectra the shaded area denotes the shoulders of Fe-2p3/2 and Fe-2p1/2 region.

Figure 4. (a) In-plane hysteresis loops of pristine, negative and positive gated 26


Page 26 of 34

Page 27 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


LSMO-SFO2.5 sample at 10 K. (b) The in-plane hysteresis loops of single layer SFO2.5 and SFO3-δ samples at 10 K. All of the samples with area of 5×2.5 mm2 were used for magnetic measurements.

Figure 5. (a-f) I-V switching curves for 50 cycles collected from Pt/SFO2.5/NSTO RRAM devices. In Figure (a), the arrows indicate the sequence of voltage sweeping and inset indicates the schematic model of devices configuration. The IL gating was applied on the device to inject the oxygen ions into the SFO2.5 film. The time duration of gating is labelled together with I-V curves. (g) Retention data of HRS and LRS state collected after the 20 minutes –Vg application.

Figure 6. (a) The pristine SFO2.5 layer consisting of tetrahedral (triangles) and octahedral (squares) layers. The vertical octahedral row indicates the SFO3 conduction filament routs under the LRS state of Pt/SFO2.5/NSTO devices. (b) The schematic diagram of thick Y shape SFO3 conduction filament inside the SFO2.5 matrix after applying the IL gating to the device. (c) All the tetrahedral site converted into octahedral and device consisting of a nearly SFO3 layer after applying the long-time gating.

27



Figure 1 147x100mm (299 x 299 DPI)


Page 28 of 34

Page 29 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2 128x123mm (299 x 299 DPI)



Figure 3 133x75mm (299 x 299 DPI)


Page 30 of 34

Page 31 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4 81x135mm (299 x 299 DPI)



Figure 5 161x271mm (299 x 299 DPI)


Page 32 of 34

Page 33 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 6 125x45mm (299 x 299 DPI)



ToC figure 239x208mm (96 x 96 DPI)


Page 34 of 34

Electric field control of phase transition and tunable resistive switching

Recommend Documents