Voltage-Driven Room-Temperature Resistance and Magnetization

Subscriber access provided by UNIV AUTONOMA DE COAHUILA UADEC

Functional Inorganic Materials and Devices

Voltage-Driven Room Temperature Resistance and Magnetization Switching in Ceramic TiO2/PAA Nanoporous Composite Films Xidi Sun, Xiaoming Mo, Lihu Liu, Huiyuan Sun, and Caofeng Pan ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b02593 • Publication Date (Web): 28 May 2019 Downloaded from http://pubs.acs.org on May 29, 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 21 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

Voltage-Driven Room Temperature Resistance and Magnetization Switching in Ceramic TiO2/PAA Nanoporous Composite Films Xidi Sun,†, ‡ Xiaoming Mo,† Lihu Liu,§ Huiyuan Sun,*,§ and Caofeng Pan,*, ‡ †Center on Nanoenergy Research, College of Physical Science and Technology, Guangxi University, Guangxi,

Nanning 530004, PR China ‡CAS Center for Excellence in Nanoscience, Beijing Key Laboratory of Micro-nano Energy and Sensor, Beijing

Institute of Nanoenergy and Nanosystems, Chinese Academy of Sciences, Beijing 100083, PR China §College of Physics Science & Information Engineering and Hebei Advanced Thin Film Laboratory, Hebei

Normal University, Shijiazhuang, Hebei 050024, China

ABSTRACT: Voltage control of room temperature ferromagnetism has remained a big challenge which will greatly influence multifunctional memory devices. In this paper, porous TiO2 thin films were deposited by DC-reactive magnetron sputtering onto ordered porous anodic alumina (PAA) substrates. Voltage-driving room temperature resistance and magnetization switching without external magnetic field are simultaneously found in an Ag/TiO2/PAA/Al (Ag/TP/Al) device. Further analysis indicates that the formation/rupture of oxygen vacancy defect based conductive filaments would be responsible for the changes of resistivity and magnetization. Our present results suggest that the TiO2/PAA (TP) nanoporous composite films material may therefore be used to achieve voltage control of magnetism and resistance switching in the future multifunctional memory devices. The Ag/TP/Al devices can also be used for new spintronics devices, neuromorphic operations, and alternative logic circuits and computing. KEYWORDS: storage materials, ferromagnetism, oxygen vacancy, nanoporous composite films, voltage control of magnetism

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

INTRODUCTION With the development of science and technology, ultra-high density and powerful memory devices are of particularly important. Therefore, multifunctional materials for achieving voltage control of room temperature ferromagnetism(RTFM) and resistance behavior are desirable in the development of multifunctional data storage and spintronics devices.1-7 Since H. Ohno et al discovered the electric manipulation of ferromagnetic properties in ferromagnetic semiconductor materials,8, 9 same behavior was also found in ABO3 perovskite structures10, 11 and metallic oxide semiconductor materials.12-14 The problem is that many magnetoelectric composites prepared on some special substrates have weak RTFM, which can not satisfied the practical needs of data storage and spintronics devices at all. Hence, search for multifunctional materials with obvious resistance switching and large room temperature ferromagnetism is one of the key problems. Recently, redox-based resistive switching memories (RSM) have attracted great attention in both scientific and industrial community for new data storage due to their advantages of high density, nonvolatile properties, fast operation speed, low power consumption, and simple synthesization.15-17 In addition, most of binary transition metal oxides have obvious resistive switching (RS) behavior, and which are compatible with the CMOS technology. Therefore, binary transition metal oxide system had been brought to the forefront by the researchers. In most literatures, the researchers believe that the physical origins of the RS phenomenon in these oxides is that conducting channels would be selectively formed under the applied external electric field at crystalline defects such as vacancies and grain boundaries.18-22 Ahmad et al21 and Huang et al22 demonstrated that the total oxygen vacancy density had obvious influence on the RS features of zinc oxide film separately, which indicated that the oxygen vacancies were in corresponding

2


Page 2 of 21

Page 3 of 21 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


positive charged state. Among the various binary oxides, TiO2 nano structures have been studied for decades due to their broad applications.23-26 It is noteworthy that the presence of oxygen vacancies in TiO2 nanotubes has a strong impact on its ferromagnetic properties.27 Then we assumed that whether it was possible to achieve the magnetization transform through the electric field control of RS in TiO2-based oxide materials to meet the burgeoning development of low-power-consumption multifunctional memory devices. With these motivations, the electric field control of non-volatile magnetization accompanied with RS characteristic in a complementary metal-oxide semiconductor compatible structure is greatly desirable. Based on the above discussion, TiO2/PAA (TP) nanoporous laminate films were prepared by magnetron sputtering method. The porous nanostructured TiO2 possess higher specific surface area than the bulk ones, which can provide a large number of oxygen vacancies in it and further performance improvements may be possible by controlling such defects. In our work, we have investigated the properties of the formed nanoscale oxide layers and analyzed the role of oxygen vacancies in room temperature resistance switching and magnetization switching characteristics without external magnetic field. EXPERIMENTAL SECTION To fabricate nanoporous PAA substrates, the commercial high-purity Al foils (99.999% purity, 0.5 mm thickness) were firstly annealed at 400℃ for 2 hours in argon atmosphere and then polished in homemade polishing solution of ethanol and perchloric acid (4:1 in volume) for 5 min. After that, the organics on the polished foils were removed by acetone and deionized water, and finally dried in air. In the anodization process, the Al foils and a platinum plate were used as the anode and cathode, respectively. The first anodization was carried out in 0.3 M oxalic acid

3



electrolytes under a constant DC voltage of 45 V at a temperature of 5 ℃ for 6 hours. Then, the alumina film produced was removed by chemical method in a mixture solution of phosphoric acid (6 wt%) and chromic acid (4 wt%) at 30 ℃ for 8 hours. Then, a second anodization was carried out under the same anodization conditions as in the first step. To get the desired pore channel length of the PAA thin films, we can change the oxidation time appropriately. After anodization, the highly ordered PAA substrates were obtained. Porous TiO2 thin films are deposited on the PAA substrates at 300℃ by using DC-reactive magnetron sputtering. To ensure the purity of the samples, a high purity titanium disc (99.99 %) was selected as sputtering target. On the other hand, the Ti target was pre-sputtered for 5 min is needed in order to eliminate surface impurities before depositing. In the sputtering process, high purity Ar and O2 (99.99 %) were using as working and reactive gases, respectively. The working pressure was kept constantly at 1.1 Pa by adjusting the flux of the argon at 30 sccm and oxygen at 3 sccm during sputtering. The experiment result show that the TiO2 films deposited for 60 minutes with the sputtering power of 60 W is appropriate in our work. At the final, the films were annealed in air and vacuum atmosphere at 400℃ for 3 hours, respectively. To study the RS properties of TP film, dot structural Ag top electrodes with a diameter of 200 μm were deposited with a metal shadow mask. In the subsequent measurement, the external electric field can be applied on the Ag top electrodes, the Al substrate serving as bottom electrode correspondingly. The morphology and constituent of the materials have been examined by using a field-emission scanning electron microscopy (SEM) with energy dispersive X-ray spectroscopy (EDX). The phase structure of TiO2 film was investigated by X-ray diffraction (XRD) with Cu Kα radiation.

4


Page 4 of 21

Page 5 of 21 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


The surface feature of the samples was obtained by using an atomic force microscope (AFM). The photoluminescence (PL) spectrum was achieved on a fluorescence spectrophotometer with the excitation wavelength of 340 nm. The magnetic properties of the samples had been tested on a superconducting quantum interference device (MPMSXL-5). In the measurements, all the samples were cut into a 2 mm×2 mm square due to the limitation of the sample holder and the magnetic field was applied perpendicular to the film plane. It should be noted that all the memory cells in this square area can be trigged to HRS or LRS at the same time before we measure their magnetism. Resistance switching experiments were performed in two-point, vertical geometry using a Keithley 2612A current source at room temperature, where the current is limited to 100 mA to guard the device from complete dielectric breakdown. It should be noted that all the tools used for handling the samples were non-ferromagnetic during sample preparation and magnetic measurement in order to eliminate any influence of alien magnetic impurities. RESULTS AND DISCUSSION Figure 1a shows the surface morphology and XRD pattern of the PAA thin film. It is obvious that the two-step anodization process yields well-defined hexagonal nanoporous arrays, and the PAA is in an amorphous state (inset of Figure 1a). We can see that the pores have diameters of about 45 nm, and the pore-to-pore length is approximately 125 nm. In our work, the nanoporous TP films are prepared by using DC magnetron sputtering instrument and PAA as template. Figure 1b and 1c display FE-SEM images of the surface and the cross-section of the as-sputtered TiO2 films on PAA substrates. We can see that the film is a highly ordered nanoporous array and the pore diameter of it is about 40 nm. It is clearly that the

5



Page 6 of 21

formation of TiO2 ordered pore arrays may attribute to the induction of the PAA substrate during the deposition process. But due to its growth pattern, the pore diameter of which decreased with the increasing of deposition time. And which will form a continuous film as the deposition time runs through the critical time. The thickness of TiO2 film layer is estimated to be about 140 nm from Figure 1(c). Figure 1(d) showed The EDX spectrum of the nanoporous TiO2 film. Only Al, Ti and O elements were detected, and there was not any form of transition metal can be found in the films. Obviously, the observed peak of Al and some of O should be ascribed to the PAA substrates. Figure 1(e) shows the X-ray diffraction pattern of the as-prepared TiO2 films deposited on PAA substrates, which show that the prepared sample TiO2 presenting the phase of tetragonal anatase (I41/amd), and there is no any impurity phases can be found within the detection limits.

(a)

(b)

(d)

(c)

(e)

Figure 1. (a) Surface SPM image and XRD pattern (Inset) of the PAA substrate. (b) Surface and (c) Cross-section SEM image of the nanoporous TP film. (d) EDX spectrum of the TP film. (e) XRD pattern for the porous TiO2 thin film.

6


Page 7 of 21 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


Previous works have proved that the physical properties of metal oxides are influenced by the oxygen vacancies.19, 27, 28 To examine the relationship between oxygen vacancy concentration and magnetism, the room-temperature magnetic moment versus magnetic field (M-H) curves with the magnetic field perpendicular to the TP film plane and PL spectra under different conditions were measured, as shown in Figure 2. As shown in Figure2(a), we draw the hysteresis loops using raw data in order to eliminate the correction error.7, 29 The well-defined hysteresis loops of the samples exhibit the RTFM nature of the TP films. In general, heat treating the oxide in reducing atmosphere (or vacuum) will increase the oxygen vacancy concentration. But contrarily, annealing in O2 gas condition decreases the oxygen vacancy concentration. Evidently, after annealing in vacuum, the magnetic properties of the TP film could be improved by compared to the one annealed in air condition.

Figure 2. (a) Hysteresis loops before and after annealing in air and vacuum condition at 400 ℃ for 3 hours. The bottom inset shows the magnified M-H curve. (b) Comparative analysis PL spectra for TP film annealed in air and vacuum at 400℃ for 3 hours.

As shown in figure 2(b), the room temperature PL spectra were obtained in the range of 310-590 nm with the excitation wavelength of 340 nm. In the PL spectra, there is a peak near 360 nm can be observed, which can be attributed to the intrinsic emission of TP. Furthermore, there is a broad emission band can be seen in the range of 400-520 nm, which may be attributed to the existence of oxygen vacancies.28, 30, 31 On the other hand, for some semiconductors, an increase in

7



defects can shrink the band gap, and then broaden the PL peak. On the contrary, the decrease of oxygen vacancy concentration will lead to the band tail decaying. And the results of which are to stop the band gap shrinkage and narrow the excitonic level distribution.31 Therefore, as the oxygen vacancy concentration decreases, the width of the spectrum becomes narrower. From figure 2(b), it can be seen that the intensity of the PL band weakens and the half-width also decreases (from 98 to 70 nm) in the TP film annealed in air compared to the TP film annealed in vacuum. Obviously, annealing in vacuum can increase oxygen vacancy concentration, whereas annealing in air can prevent the escape of oxygen and decrease the oxygen defects concentration existed on the surfaces of TP film. So, the change of PL spectrum intensity can be comprehended by the oxygen vacancy concentration change in different heat treatment procedures.31, 32 As discussed above, it is clear that the variation of PL spectrum (Figure 2(b)) is synchronous changes of magnetic moment (Figure 2(a)). Such a result suggests that the observed RTFM in the TP thin films is related to vacancies defects in the material. As a focal point, the possible magnetic exchange mechanism will be discussed in detail in the following parts of this paper. The resistive properties of the Ag/TP/Al structures were measured using a semiconductor characterization system (Keithley 2612A), and all the measurement were carried out on an aseismic platform. Figure 3(a) displays typical measured current-voltage (I-V) characteristics of the Ag/TP/Al structure with the application of a sweeping voltage from Ag to Al as 0 V→ 3 V→ 0 V→ -3 V→ 0 V, and the inset exhibits the schematic setup for I-V measurements. The current compliance was fixed at 10 mA to prevent permanent breakdown of the device. From Figure 3(a), there is no forming process can be found in the RS, which may attribute to the abundant oxygen vacancies existed in the pristine Ag/TP/Al nanoporous films.32-34 Under forward and reverse bias,

8


Page 8 of 21

Page 9 of 21 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


the switching hysteresis is asymmetric obviously. We can attribute this phenomenon to the ultra-high resistance of ~109 Ω of the sample, as described by Zoolfakar.33 The suddenly current increase when the voltage was swept from 0 V to +3 V may due to the forming process of conducting filaments, indicating that the cell switches from high resistance state (HRS) to low resistance state (LRS). However, it transformed to HRS again when the voltage sweeping from 0 V to −3 V. The filamentary model has been widely accepted to explain the switching behavior of TiO2-based resistance switching devices. The great number of oxygen vacancies in TP nanoporous composite films can form conducting filaments on the inner walls of the holes naturally. However, the formation/disruption of localized conducting filaments is preferred as the driving mechanism of resistance switching. For clarity, a simplified switching mechanism model is given below to explain the RS effect. Once a bias is applied, O2- ions will move towards the Ag top electrode more effectively. This movement of O2- ions is associated with formation of oxygen vacancies, and a conductive filament channels will be set up the leaving metal ions in TP film between the two electrodes. Consequently, with the set process (HRS to LRS) occurring, the incidence of oxygen vacancy capturing electron decreases. When applying a negative voltage, the O2- ions will move back from the Ag top electrode to the TP layer, and the consequent rupture of the conducting filament will occur. And the chance of oxygen vacancies capturing electrons improved greatly in the reset process (LRS to HRS) subsequently. On the other hand, once the conductive filament is formed (or disconnected), RS can still be maintained even if the voltage is dropped. So this resistance transformation is non-volatile. Figure 3(b) shows the I-V curve of the untreated Ag/TP/Al device over the course of a series of 10 repeated measurements, and a stable RS characteristic over the course of 10 switching cycles can be found. Where a higher voltage was

9



Page 10 of 21

applied to investigate the stability of the device and to ensure it can be used in a wide voltage range.

(a)

(b)

Figure 3. (a) I–V curves of the as-prepared Ag/TP/Al device by applying a sweeping voltage as 0 V→ 3V→ 0 V→ -3V→ 0 V. The arrows indicate the voltage sweeping directions. The inset in Figure 3(a)shows the schematic structure of the Ag/TP/Al device. (b) Endurance test of the bistable resistance in a Ag/TP/Al device over a course of 10 switching cycles.

Considering about the formation of oxygen vacancies, it is reasonable for us to hypothesize that an oxygen vacancy ( VO ) is formed and existed as positive center in TiO2 layer by removing an O2- ion. In other words, it is easy for oxygen vacancies to trap electrons. However, the trapped electrons will have net magnetic moment due to their localization, so we can sure that the VO (an oxygen vacancy occupied by one electron) will have a magnetic moment. The mobilities of electrons are different when they are in different resistive states for the sample, which lead to the different number of VO . Different resistive states, however, have different electron mobilities, therefore the number of VO in HRS and LRS is different. Based on the above analysis, we can conclude that the magnetism is different in HRS and LRS.7 Now we turn to the influnce of RS change on the magnetic properties of Ag/TP/Al devices, which is a particularly important question in this work. Figure 4 shows the M-H curves with the external magnetic field perpendicular to TP the film plane after triggering the memory devices to the HRS or LRS, respectively. Where we applied the magnetic field perpendicular to the films due to the magnetism of porous film material along the nano holes, i.e. perpendicular to the films, is bigger 10


Page 11 of 21 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


than that of parallel to the films.27, 28 Remarkably, the device at both LRS and HRS all shows clear M-H curves, which shows that RTFM still retains even the application of bias voltages. This result indicates that the effect of RS has signifacant influence on the magnetism of the Ag/TP/Al device. As suggested by the SQUID results, the MS for the device at HRS (about 200 emu/cm3) is larger

than that at LRS (about 110 emu/cm3). At the same time, the change of Ms of the sample keep same step with the one of RS, as shown in the inset of Figure 4. More interestingly, a magnetization switching was also found in the transformation process of HRS and LRS without obvious degradation. This is very crucial for the application of the information storage.

Figure 4. M-H loops in the initial state, HRS and LRS. The bottom inset is the switching of MS accompanying the RS effects.

Based on the experimental results described above, we believe that the oxygen vacancies play an essential role in the process of electric-field-induced resistance switching and magnetization change. In fact, up to now, the origin of ferromagnetism is still very confused. Although, there have been several different theoretical models on the origin of RTFM in dilute ferromagnetic semiconductor oxides. In 2000, Dietl proposed the first model to explain the origin of ferromagnetism in dilute magnetic semiconductor oxides,35 i.e. the Zener/Ruderman -Kittel-Kasuya-Yosida (RKKY) model, which believed that the dopant (transition metal cation) replace the positions on the lattice and generate polarized holes. The theoretical model considers

11



Page 12 of 21

that the magnetization of these samples is related to the doping concentration. However, the measured magnetization has been proved to be irrelevant on the consistence of dopants,36, 37 which indicates that the RKKY mechanism may not be the principal reason of ferromagnetism in dilute magnetic semiconductor oxides. Later, Coey developed several models to explain the magnetism of thin magnetic semiconductor films. For example, the bound magnetic polaron (BMP) model

38

and the charge-transfer ferromagnetic theoretical model,39 both of which emphasize the effect of defects in the origin of the RTFM in dilute magnetic semiconductor oxides. These models, however, cannot explain how ferromagnetism occurs in dopant-free dilute magnetic semiconductor oxides, because there is no additional carrier source for the absence of dopants. Yang et al reported the ferromagnetism in Al2O3 nanoparticles is mediated by the exchange coupling between single charged oxygen vacancy.40 This exchange mechanism assumes that single charged oxygen vacancy center will overlap each other once the density of them reaches the critical value for magnetic percolation, and which will bring about a long-range ferromagnetic ordering. But when the distance between oxygen vacancies is relatively far away, how the ferromagnetic order is generated is not explained by the research group. Very recently, a new defect-induced bound polaron model has been proposed in the undoped but with much defects dilute ferromagnetic semiconductor oxides.41 This extended model assumes that a narrow impurity band is formed just below the conduction band minimum because the overlap and hybridization of the defect states, i.e. the oxygen vacancies associated with different charge.41 So one view is that the ferromagnetic exchange interactions may arise between the VO and local 4d1 electron because their net spins possess the same direction.41 However, this model is insufficient to explain why some dilute magnetic semiconductor oxides without any unpaired d electrons also show

12


Page 13 of 21 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


ferromagnetism. As mentioned above, due to the oxygen vacancies are in a state of charge imbalance, which exists as electron trap easily to capture electrons.28 The neutral oxygen vacancy center can therefore be ascribed as an electron pair trapped in the hole left by the missing oxygen, thus the cancellation of paired electron spins occur because of the antiparallel electron spins. Moreover, the formation of neutral oxygen vacancy displays too high formation energies and may occur in negligible amounts. VO center containing only one single electron is similar to localized electron which has a net magnetic moment.7, 19 Moreover, VO center means strongly electron deficient with absence of both electrons, therefore, which made no contribution to the total magnetic moment at all. On the basis of direct exchange mechanism, the VO distributed at nearest neighbor sites will couple together directly for their spins arrange parallel.7, 19, 27, 28 In the case of the oxygen vacancies distributing far from each other, the conduction electrons will play the role of itinerant electrons in the material. The free electrons can easily be polarized by VO , and then the magnetic moment orientation of the polarized electrons will keep in line with that of VO . Thus the longer-range ferromagnetic exchange can occur by the intervention of polarized electrons for the VO , this is regarded as indirect exchange interactions process ( VO -e-- VO ). This insight may be used to interpret the RTFM of TP nanoporous films and also be applied to comprehend the observed distinct phenomenons of d0 ferromagnetism in low-dimensional nanostructures or thin films with high specific surface areas. Furthermore, the model we proposed indicates that magnetic ordering in might be obtained theoretically by introducing various anion vacancy defects in semiconductor oxides. Besides, the phenomenon that the saturated magnetization for the HRS are always larger than those for the LRS can be explained that many VO can be generated when

13



the oxygen ions driven by the electric field.7, 19 When applying a positive voltage to the Ag top electrode (see in Figure 3(a)), the device can be switched into the LRS, for which there is a greater electron mobility corresponding to a decreased probability that the VO vacancies capture electrons thereby reducing the number of VO vacancies. On the contrary, in the case of the application of negative bias, the device switches to HRS again. Compared with LRS, the probability of electron captured by VO vacancies increases due to the lower electron mobility, which means more VO present in the film. Now that a large number of VO vacancies can heighten the ferromagnetism, so the MS of nanoporous TP film in the LRS lower than that in the HRS is evident. Since the resistance switching is accompanied by the forming and breaking of conducting filaments under the modulating of external electric field, in the process of oxygen vacancies arranging orderly, the probability of trapping electrons of them will increase greatly, which results in the magnetism of HRS bigger than the pristine state. Based on the knowledge of solid state physics, it is easy to know that oxygen vacancy is easy to form on the sample surface, so the conductive filaments formed by oxygen vacancy are most likely to be formed on the inner surface of the pore. Therefore, compared with the pristine state, the probability of oxygen vacancy capturing electrons is higher no matter in the HRS or in the LRS. So the magnetization is higher in HRS and LRS than that in the as-prepared state, which is consistent with the expected result (see in Figure 4).2, 7, 19 CONCLUSIONS In summary, we focus on the resistance switching and electric field-induced magnetization modulation based on the Ag/TiO2/PAA/Al structure. The TiO2 film was synthesized by magnetron sputtering on PAA substrates. Large magnetization changes of the TP films emerge in response to

14


Page 14 of 21

Page 15 of 21 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


the resistance switching effect of the Ag/TP/Al heterostructure controlled by applied electric field. Our results provide convincing evidence for the influence oxygen vacancies on resistive behavior and magnetization modulation in undoped TiO2/PAA nanoporous films. Furthermore, the results suggest that the magnetic ordering might be obtained by regulating the kinds of anion vacancy defects in semiconductor oxides. The combination of the resistance switching and magnetism modulation caused by electric field at room temperature makes it possible to integrating both ferromagnetism and electrical properties into a simple Ag/TP/Al structure, which can be extended to produce other novel dilute magnetic semiconductors and is expected to be used in new memory and new spintronics devices. AUTHOR INFORMATION Corresponding Author *E-mail: [email protected] *E-mail: [email protected] ORCID Huiyuan Sun: 0000-0001-8462-7609 Caofeng Pan: 0000-0001-6327-9692 Notes The authors declare no competing financial interest. Acknowledgment The authors thank the support of the Natural Science Foundation of Hebei Province (no. A2016205237). REFERENCES [1] Cuellar, F. A.; Liu, Y. H.; Salafranca, J.; Nemes, N.; Iborra, E.; Sanchez-Santolino, G.; Varela, M.; Garcia Hernandez, M.; Freeland, J. W.; Zhernenkov, M.; Fitzsimmons, M. R.; Okamoto, S.; Pennycook, S. J.; Bibes, M.; Barthélémy, A.;

Velthuis, S.G.E.; Sefrioui, Z.; Leon1, C.;

15



Page 16 of 21

Santamaria, J. Reversible Electric-Field Control of Magnetization at Oxide Interfaces. Nat. Commun. 2014, 5, 4215. [2] Qi, L.; Pan, D.; Li, J.; Liu, L.; Sun, H. HfO2 / Porous Anodic Alumina Composite Films For Multifunctional Data Storage Media Materials under Electric Field Control. Nanotechnology 2017, 28, 115702. [3] Ohno, H. A Window on the Future of Spintronics. Nat. Mater. 2010, 9, 952-954. [4] Munjal, S.; Khare, N. Electroforming Free Controlled Bipolar Resistive Switching in Al/CoFe2O4/FTO Device. Appl. Phys. Lett. 2018, 112, 073502. [5] Zhao, M.; Zhu, Y.; Zhang, Y.; Zhang, T.; Qiu, D.; Lai, G.; Hu, C.; Wang, Q.; Zhang, F.; Li, M. Resistive

Switching

and

Related

Magnetization

Switching

in

Pt/BiFeO3/Nb:SrTiO3

Heterostructures. RSC Adv. 2017, 7, 23287-23292. [6] Munjal, S.; Khare, N. Multilevel Resistive and Magnetization Switching in Cu/CoFe2O4/Pt Device: Coexistence of Ionic and Metallic Conducting Filaments. Appl. Phys. Lett. 2018, 113, 243501. [7] Qi, L.; Liu, H.; Sun, H.; Liu, L.; Han, R. Electric Field Control of Magnetization in Cu2O/Porous Anodic Alumina Hybrid Structures at Room Temperature. Appl. Phys. Lett. 2016, 108, 142402. [8] Chiba, D.; Yamanouchi, M.; Matsukura, F.; Ohno, H. Electrical Manipulation of Magnetization Reversal in a Ferromagnetic Semiconductor. Science 2003, 301, 943-945. [9] Chiba, D.; Sawicki, M.; Nishitani, Y.; Nakatani, Y.; Matsukura, F.; Ohno, H. Magnetization Vector Manipulation by Electric Fields. Nature 2008, 455, 515-518. [10] Heron, J. T.; Trassin, M.; Ashraf, K.; Gajek, M.; He, Q.; Yang, S. Y.; Nikonov, D. E.; Chu, Y.

16


Page 17 of 21 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


H.; Salahuddin, S.; Ramesh, R. Electric-Field-Induced Magnetization Reversal in a Ferromagnet Multiferroic Heterostructure. Phys. Rev. Lett. 2011, 107, 217202. [11] Heron, J. T.; Bosse, J. L.; He, Q.; Gao, Y.; Trassin, M.; Ye, L.; Clarkson, J. D.; Wang, C.; Liu, J.; Salahuddin, S.; Ralph, D. C.; Schlom, D. G.; Iniguez1, J.; Huey, B. D.; Ramesh, R. Deterministic Switching of Ferromagnetism at Room Temperature Using an Electric Field. Nature 2014, 516, 370-373. [12] Chang, L.; Wang, C.; Tang, J.; Nie, T.; Jiang, W.; Chu, C.; Arafin, S.; He, L.; Afsal, M.; Chen, L. J.; Wang, K. L.; Electric-Field Control of Ferromagnetism in Mn-Doped ZnO Nanowires. Nano Lett. 2014, 14, 1823-1829. [13] Wang, Y.; Song, Y.; Tong, W.; Zhang, Y.; Qi, R.; Xiang, P.; Huang, R.; Zhong, N.;

Lin, H.;

Tang, X.; Peng, H.; Duan, C. Electric Field Control of Magnetism in Nickel with Coaxial Cylinder Structure at Room Temperature by Electric Double Layer Gating. J. Mater. Chem. C 2017, 5, 10609-10614. [14] Fix, T.; Choi, E.M.; Robinson, J. W. A.; Lee, S. B.; Chen, A.; Prasad, B.; Wang, H.; Blamire, M. G.; MacManus-Driscoll, J. L. Electric-Field Control of Ferromagnetism in a Nanocomposite via a ZnO Phase. Nano Lett. 2013, 13, 5886-5890. [15] Zaffora, A.; Cho, D. Y.; Lee, K. S.; Quarto, F. D.; Waser, R.; Santamaria, M.;

Valov, I.

Electrochemical Tantalum Oxide for Resistive Switching Memories. Adv. Mater. 2017, 29, 1703357. [16] Morishita, Y.; Hosono, T.; Ogawa, H. Fabrication of Resistive Switching Memory Structure Using Double-Sided-Anodized Porous Alumina. Solid State Electron. 2017, 131, 30-33. [17] Cho, D. Y. ; Luebben, M. ; Wiefels, S. ; Lee, K. S. ; Valov, I. Interfacial Metal-Oxide

17



Page 18 of 21

Interactions in Resistive Switching Memories. ACS Appl. Mater. Inter. 2017, 9, 19287-19295. [18] Lim, D. H.; Kim, G. Y.; Song, J. H.; Jeong, K. S.; Kim, D. C.; Nam, S. W.; Cho, M. H.; Lee, T. G. Electric Field Effect Dominated Bipolar Resistive Switching Through Interface Control in a Pt/TiO2/TiN Structure. RSC Adv. 2015, 5, 221-230. [19] Hou, X.; Sun, H.; Liu, L.; Jia, X.; Liu, H. Resistive Property and Ferromagnetism in ZnO/PAA Nanoporous Composite Films. J. Alloy. Compd. 2015, 640, 444-448. [20] Gao, S.; Zeng, F.; Li, F.; Wang, M.; Mao, H.; Wang, G.; Song, C.; Pan, F.

Forming-Free

and Self-Rectifying Resistive Switching of the Simple Pt/TaOx/n-Si Structure for Access Device-Free High-Density Memory Application. Nanoscale 2015, 7, 6031-6038. [21] Ahmad, S. Z.; Rosmalini, A. K.; Rozina, A. R.; Sivacarendran, B.; Liu, X. J.; Eugene, K.; Suresh, K. B.; Madhu, B. S.; Serge, Z.; Anthony, P. O.; Kourosh, K. Z. Engineering Electrodeposited ZnO Films and Their Memristive Switching Performance. Phys. Chem. Chem. Phys. 2013, 15, 10376-10384. [22] Huang, H. W.; Kang, C. F.; Lai, F. I.; He, J. H.; Lin, S. J.; Chueh, Y. L. Stability Scheme of ZnO-Thin Film Resistive Switching Memory: Influence of Defects by Controllable Oxygen Pressure Ratio. Nanoscale Res. Lett. 2013, 8, 483. [23] Hang, R.; Zhao, Y.; Bai, L.; Liu, Y.; Gao, A.; Zhang, X.; Huang, X.; Tang, B.;

Chu, P. K.

Fabrication of Irregular-Layer-Free and Diameter-Tunable Ni-Ti-O Nanopores by Anodization of NiTi Alloy. Electrochem. Commun. 2017, 76, 10-14. [24] Zheng, L. X.; Dong, Y. C.; Bian, H. D.; Lee, C.; Lu, J.; Li, Y. Y. Self-Ordered Nanotubular TiO2 Multilayers for High-Performance Photocatalysts and Supercapacitors. Electrochim. Acta 2016, 203, 257-264.

18


Page 19 of 21 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


[25] Yu, M.; Cui, H.; Ai, F.; Jiang, L.; Kong, J.; Zhu, X. Terminated Nanotubes: Evidence Against the Dissolution Equilibrium Theory. Electrochem. Commun. 2018, 86, 80-84. [26] Wu, H.; Li, D.; Zhu, X.; Yang, C.; Liu, D.; Chen, X.; Song, Y.; Lu, L. High- Performance and Renewable Supercapacitors Based on TiO2 Nanotube Array Electrodes Treated by an Electrochemical Doping Approach. Electrochim. Acta 2014, 116, 129-136. [27] Wu, T.; Sun, H.; Hou, X.; Liu, L.; Zhang, H.; Zhang, J. Significant Room-Temperature Ferromagnetism in Porous TiO2 Thin Films. Micropor. Mesopor. Mater. 2014, 190, 63-66. [28] Sun, H.; Zhang, H.; Hou, X.; Liu, L.; Wu, T.; Yang, S. Significant Room Temperature Ferromagnetism in PAA Thin Films. J. Mater. Chem. C 2013, 1, 3569-3572. [29] Gu, J.; Liu, L.; Qi, Y.; Xu, Q.; Zhang, H.; Sun, H. Magnetic Characterization of Diluted Magnetic Semiconductor Thin Films. J. Appl. Phys. 2011, 109, 023902. [30] Sood, S.; Kumar, S.; Umar, A.; Kaur, A.; Mehta, S. K.; Kansal, S. K. TiO2 Quantum Dots for the Photocatalytic Degradation of Indigo Carmine Dye. J. Alloy. Compd. 2015, 650, 193-198. [31] Zhang, H.; Xie, C.; Zhang, Y.; Liu, G.; Li, Z.; Liu, C.; Ma, X.; Zhang W. Effects of Thermal Treatment under Different Atmospheres on the Spectroscopic Properties of Nanocrystalline TiO2. J. Appl. Phys. 2008, 103, 103107. [32] Xu, Z.; Jin, K.; Gu, L.; Jin, Y.; Ge, C.; Wang, C.; Gu, H.; Lu, H.; Zhao, R.; Yang, G. Evidence for a Crucial Role Played by Oxygen Vacancies in LaMnO3 Resistive Switching Memories. Small 2012, 8, 1279-1284. [33] Zoolfakar, A. S.; Kadir, R. A.; Rani, R. A.; Balendhran, S.; Liu, X.; Kats, E.; Bhargava, S. K.; Bhaskaran, M.; Sriram, S.; Zhuiykov, S.; O’Mullane, A. P.; Kalantar-zadeh, K. Engineering Electrodeposited ZnO Films and Their Memristive Switching Performance. Phys. Chem. Chem.

19



Page 20 of 21

Phys. 2013, 15, 10376-10384. [34] Kwon, D. H.; Kim, K. M.; Jang, J. H.; Jeon, J. M.; Lee, M. H.; Kim, G. H.; Li, X.; Park, G.; Lee, B.; Han, S.; Kim, M.; Hwang, C. Atomic Structure of Conducting Nanofilaments in TiO2 Resistive Switching Memory. Nat. Nanotechnol. 2010, 5, 148-153. [35] Dietl, T.; Ohno, H.; Matsukura, F.; Cibert, J.; Ferrand, D. Zener Model Description of Ferromagnetism in Zinc-Blende Magnetic Semiconductors. Science 2000, 287, 1019-1022. [36] Hong, N. H.; Sakai, J.; Ruyter, A.; Brizé, V. Does Mn Doping Play any Key Role in Tailoring the Ferromagnetic Ordering of TiO2 Thin Films?. Appl. Phys. Lett. 2006, 89, 252504. [37] Xu, Q.; Schmidt, H.; Zhou, S.; Potzger, K.; Helm, M.; Hochmuth, H.; Lorenz, M.;

Setzer,

A.; Esquinazi, P.; Meinecke, C.; Grundmann, M. Room Temperature Ferromagnetism in ZnO Films due to Defects. Appl. Phys. Lett. 2008, 92, 082508. [38] Coey, J. M. D.; Venkatesan, M.; Fitzgerald, C. B. Donor Impurity Band Exchange in Dilute Ferromagnetic Oxides. Nat. Mater. 2005, 4, 173-179. [39] Coey, J. M. D.; Wongsaprom, K.; Alaria, J.; Venkatesan, M. Charge-Transfer Ferromagnetism in Oxide Nanoparticles. J. Phys. D: Appl. Phys. 2008, 41, 134012. [40] Yang, G.; Gao, D.; Zhang, J.; Shi, Z.; Xue, D. Evidence of Vacancy-Induced Room Temperature Ferromagnetism in Amorphous and Crystalline Al2O3 Nanoparticles. J. Phys. Chem. C 2011, 115, 16814-16818. [41] Rahman, M. A.; Rout, S.; Thomas, J. P.; McGillivray, D.; Leung, K. T. Defect-Rich Dopant-Free ZrO2 Nanostructures with Superior Dilute Ferromagnetic Semiconductor Properties. J. Am. Chem. Soc. 2016, 138, 11896-11906.

20


Page 21 of 21 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 captions: Figure 1. (a) Surface SPM image and XRD pattern (Inset) of the PAA substrate. (b) Surface and (c) Cross-section SEM image of the nanoporous TP film. (d) EDX spectrum of the TP film. (e) XRD pattern for the porous TiO2 thin film. Figure 2. (a) Hysteresis loops before and after annealing in air and vacuum condition at 400℃ for 3 hours. The bottom inset shows the magnified M-H curve. (b) Comparative analysis PL spectra for TP film annealed in air and vacuum at 400℃ for 3 hours. Figure 3. (a) I-V curves of the as-prepared Ag/TP/Al device by applying a sweeping voltage as 0 V→ 3V→ 0 V→ -3V→ 0 V. The arrows indicate the voltage sweeping directions. The inset in Figure 3(a)shows the schematic structure of the Ag/TP/Al device. (b) Endurance test of the bistable resistance in a Ag/TP/Al device over a course of 10 switching cycles. Figure 4. M-H loops in the initial state, HRS and LRS. The bottom inset is the switching of MS accompanying the RS effects.

21


Voltage-Driven Room-Temperature Resistance and Magnetization

Recommend Documents