Interface State-Induced Negative Differential ... - ACS Publications

Subscriber access provided by Kaohsiung Medical University

Surfaces, Interfaces, and Applications

Interface-state Induced Negative Differential Resistance Observed in Hybrid Perovskite Resistive Switching Memory Hanlu Ma, Wei Wang, Hai Yang Xu, Zhongqiang Wang, Ye Tao, Peng Chen, Weizhen Liu, Xintong Zhang, Jiangang Ma, and Yichun Liu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b07850 • Publication Date (Web): 06 Jun 2018 Downloaded from http://pubs.acs.org on June 6, 2018

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

Interface-state Induced Negative Differential Resistance Observed in Hybrid Perovskite Resistive Switching Memory

Hanlu Ma†, Wei Wang†, Haiyang Xu*, Zhongqiang Wang*, Ye Tao, Peng Chen, Weizhen Liu, Xintong Zhang, Jiangang Ma and Yichun Liu

Key Laboratory for UV Light-Emitting Materials and Technology (Northeast Normal University), Ministry of Education, 5268 Renmin Street, Changchun, China

KEYWORDS: hybrid perovskite, bipolar resistive switching, negative differential resistance, interface-state, iodine vacancies.

ACS Paragon Plus Environment

1

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

Page 2 of 34

ABSTRACT

Hybrid organic−inorganic perovskite, well known as light-absorbing materials in solar cells, have recently attracted considerable interest for applications in resistive switching (RS) memory. A better understanding of the role of interface-state in hybrid perovskite materials on RS behavior is essential for the development of practical devices. Here, we study the influence of interface-state on the RS behavior of an Au/CH3NH3PbI3/FTO memory device using a simple air-exposure method. We observe a transition of RS hysteresis behavior with exposure time. Initially no hysteresis is apparent but air exposure induces bipolar RS and a negative differential resistance (NDR) phenomenon. The reductions of I/Pb atomic ratio and work function on film surface are examined using XPS spectra and Kelvin Probe technique, verifying the produce of donor-type interface-states (e.g. iodine vacancies) during CH3NH3PbI3 film degradation. Studies on complex impedance spectroscopy confirm the responsibility of interface-states in NDR behavior. Eventually, the trapping/de-trapping of electrons in bulk defects and at interface-states accounts for the bipolar RS behavior accompanied with NDR effect.


2

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


1. INTRODUCTION Resistive random access memory (RRAM) devices have emerged as a next generation memory technology owing to attractive properties, including high storage density, fast switching speed, and low power consumption. Resistive switching (RS) characteristics have been observed in various materials, such as metal oxides1-7, chalcogenides8, inorganic perovskite9, and organic materials10. The RS behavior can bring advantages in low power operation as well as multilevel memory11-12. The switching mechanism is generally related to cation/anion (e.g., Ag+, Cu+, and O2−) migration in the switching layer. Recently, hybrid organic−inorganic perovskites (e.g. CH3NH3PbBr3, CH3NH3PbI3, CH3NH3PbI3-xClx) have demonstrated remarkable RS behavior based on the migration of CH3NH3+ (MA) and halogen ions13-15, and trapping defects within the films16. These devices have attracted growing attention due to their good memory performance. As well known, hybrid perovskites also have excellent optical absorption and a tunable band gap, making them suitable for a variety of optoelectronic applications including light-emitting devices17, photodetectors18, and especially solar cells19. Thus, hybrid perovskite based RRAM devices might allow some unique multi-functionality through the integration of electrical and light-sensitive properties, such as photo-read synaptic functions20 and electric-SET/light-RESET operations21. the realization of electric-SET/light-RESET operations Zhu and Lu developed the optogenetics-inspired artificial synapse based on organic−inorganic halide perovskite due to its high light sensitivity 22. Furthermore, the good mechanical properties of hybrid perovskite show potential applications in flexible RRAM devices by taking advantage of their low temperature processing conditions. Many efforts have also been made to understand the switching mechanism of hybrid perovskites. An analytical compact model has been proposed in our previous work to describe the RS behavior of organic–inorganic perovskite-based RRAM

23

.


3


Page 4 of 34

Huang et al. confirmed the migration of ions (e.g., CH3NH3+, I−) under an electric field in CH3NH3PbI3 (MAPbI3), which can induce p-type/n-type doping effects and modulate the material’s electrical properties24. Yoo et al. proposed that charge trapping in the bulk material plays an important role in RS operation15. However, the mechanism of RS requires further investigation in these hybrid perovskites because other factors can also affect the RS behaviors. For example, interface-state can influence RS behavior by trapping electrons or adjusting the Schottky barrier of the devices25-28. Interface-state are also a general concern in hybrid perovskites and their presence can greatly affect the electronic transport properties of the devices24. However, there have been few reports available on the influence of interface-state on RS behavior in hybrid perovskite materials. In this work, we focused on the effect of interface-state on RS behavior in two-terminal Au/CH3NH3PbI3/FTO devices. Herein, we used a simple air-exposure method to produce and adjust the presence of interface-state of the CH3NH3PbI3film, where atmospheric moisture induced chemical decomposition of the film. Interestingly, the devices showed a transition from no hysteresis to bipolar RS, and finally a negative differential resistance (NDR) phenomenon with increasing air-exposure time. Furthermore, we systematically investigated the evolution of interface-sate during film degradation and its corresponding role in RS behavior using various experimental techniques, such as x-ray photoelectron spectroscopy (XPS), Kevin probe method and complex impedance spectroscopy. 2. RESULTS AND DISCUSSION: Figure 1a shows the structure of the memory cell studied in this work, where a MAPbI3 layer was sandwiched between an Au top electrode and FTO substrate. The top-view SEM image


4

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


in Figure 1b the granular MAPbI3 film fully covered the FTO substrate, in which the grain size varied from 300 to 500 nm. The granular film was typical in organic−inorganic perovskite, and similar results has been reported in previous literatures16,29 .Cross-sectional SEM and AFM images (Figure 1b and c) show the MAPbI3 layer had a thickness of about 1 µm and a roughness of 45 nm. The thickness was higher than that of a typical metal oxide based RRAM30, but was necessary to ensure a continuous thin film of the MAPbI3 material, as previously reported16. The tetragonal perovskite structure of the as-prepared films was confirmed by XRD analysis, as shown in Figure 1d. As previously reported, hybrid perovskite materials suffer from chemical decomposition in the presence of moisture, which induces defects on the surface of their films31-36. Herein, to generate and adjust the presence of interface-state in our MAPbI3 films, we simply exposed the MAPbI3 film to ambient air with relative humidity (RH) of 48±5% at room temperature under dark condition. Figure 2 illustrates the evolution of the I-V curve of the Au/MAPbI3/FTO device with air-exposure time, in which the bias was applied in a bipolar dual-sweep mode. We defined the positive voltage as the current flowing from the Au to FTO electrodes in the measurements. A change of the RS behavior can be clearly observed under the influence of air-exposure. For the pristine device, there was no noticeable hysteresis behavior even when a relatively large bias (±5 V) was applied, as shown in Figure 2a. However, a self-limited bipolar RS behavior appeared after 3 days and 5 days of air-exposure (Figure 2b-c), with the devices showing reversible switching between a high resistance state (HRS) and low resistance state (LRS). The device set / reset at voltages of +1 V / −1 V in the positive / negative sweeps, respectively. These devices can present 100 repeatable switching cycles without degradation (See Figure S1 in the Supporting Information). Notably, after further exposure time, of up to 7 days, another asymmetric


5


Page 6 of 34

hysteresis phenomenon accompanying the bipolar RS was found in our devices. As indicated by arrows in Figure 2d, instead of the direct switching from the HRS to LRS during the positive voltage sweep (Figure 2b-c), the current reached a relatively high value around 1 V and then decreased as the applied forward bias was increased to 3 V. The phenomenon of the current decreasing with increasing voltage is known as a NDR effect37-39 and has been widely studied for applications in low-power40,41 memories and logic circuits42-45. In fact, the NDR behavior was also observed in organic–inorganic perovskite in the report of Hasegawa et al46. However, the switching mechanism of NDR effect was not experimentally studied in their work. According to previous studies, NDR behavior is usually related to charge trapping at interface-state of metaloxide RRAM38. This effect likely also explained the NDR behavior of our MAPbI3 device. The appearance of NDR behavior was consistent with our experimental design that allowed interfacestate to develop through an air-exposure method. The resistance state of NDR eventually reached a LRS after the reverse sweep from 3 V to 0 V, where the current increased when the bias was below 1 V. A reset process could also be achieved during the negative sweep. Thus, successful implementation of set and reset processes indicated that this RS characteristic accompanied with NDR effect could be used for memory applications. Furthermore, the NDR effect dominated the RS behavior and remained stable even after longer air exposure times. In addition, it should be noted that forming process is invariably required to active the switching operations in BRS and NDR effect. This forming gives rise to the great reduction of HRS in following RS cycles, as illustrated in Figure 2b-d. The transition from no hysteresis to NDR effect can be also obtained within 1 hour by increasing the humidity to 90% (See Figure S2 in the Supporting Information). The RS characteristics of data retention, switching speed and the evolution of HRS/LRS with cycles were evaluated in Figure 3. Figure 3a shows that both the HRS and LRS state can be


6

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


retained without degradation for over 104 s, indicating the good retention of the memory cell. The resistance states were read at50 mV. As shown in Figure 3b, the evolution of HRS/LRS with switching cycles were conducted by recording 100 consecutive RS cycles. There was no obvious degradation of HRS and LRS with switching cycles except for the acceptable fluctuation, confirming its reproducible read/write characteristic. Furthermore, the fast-speed switching was performed using a pulsed setup (Figure 3c). As illustrated in Figure 3d and 3e, we can monitor the transient resistance change of the device by checking the output voltage of a load resistor (200 Ω). Read pulses [-0.4 V/500 ns] were input before and after the set/reset pulses. Figure 3d and 3e show the enlarged figure of set /reset processes. Note that the switching time should be where the difference of input and output signal happens. As indicated by dash lines in Figure 3d and 3e, the set/reset switching times are around 60 ns and 100 ns respectively, indicating the feasibility of the fast switching memory device. The evolution of RS behavior with increasing air-exposure time can be summarized that the absence of hysteresis was replaced by bipolar RS before finally showing a NDR effect. The transitions of RS behavior may be related to the presence of interface-state induced by the chemical instability of the MAPbI3 layer under air-exposure. Potential degradation processes may occur on the film surface as follows47: H 2O

CH3 NH3 PbI 3 (s) ↔ CH3 NH3 I (aq) + PbI2 (s) CH3 NH3 I (aq) ↔ CH3 NH2 (aq) + HI(aq)

(1) (2)

4HI(aq) + O2 ↔ 2I 2 (s) + 2H 2O

(3)

The MAPbI3 film can easily decompose to CH3NH3I solution and PbI2 under our exposure conditions (48±5% RH) according to reaction (1). The CH3NH3I solution can then further


7


Page 8 of 34

decompose to CH3NH2 and HI by reaction (2). A redox reaction may eventually occur between the HI solution and oxygen in reaction (3). To determine whether the degradation of the MAPbI3 film actually occurred during our air-exposure experiment, we further studied the XPS spectra of film surface to verify the decomposition process with air-exposure time, as shown in Figure 4. For the pristine MAPbI3 film the N1s, Pb 4f and I 3d core levels were located at 401.80, 137.31, and 618.28 eV and were consistent with the core levels in perovskite, respectively48. With increasing air-exposure time, one more peak at 399.5 eV appeared for N 1s, as shown in Figure 3a. This peak suggested the presence of CH3NH249, which was agreed with reaction (2). In addition, the binding energy of Pb 4f and I 3d shifted from 137.52 eV to 137.40 eV and from 618.46 eV to 618.25 eV with exposure up to 7 days, respectively. The 0.2 eV down-shift of Pb 4f and I 3d presented the similar result with that of Guo et al., indicating the weakened the bonding strength between Pb and I48. Several literatures also reported the generation of PbI2 and Pb0, which may be too weak to be detected in our experiment

50,51

. Instead, the X-ray diffraction

patterns of MAPbI3 film can confirm the existence of PbI2 after 7days exposure (See Figure S3 in the Supporting Information). The changes in N 1s, Pb 4f and I 3d were clear enough to confirm the degradation of the MAPbI3 film, which lead to a reduction of the iodine content at the film surface and generate iodine vacancies52,53. Iodine vacancies are a type of donor-like defect46, which can induce n-doping effects and affect the carrier transport of MAPbI3, even causing a transition from p-type to n-type material52. The degradation rate is quite different for the interfaces near air side and FTO side during exposure. According to Shao et al.’s report

54

,

the generation of defects could be ignorable in air exposure, which will not be considered in the current work.


8

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


Besides the analysis of core levels in perovskite, we further examined the elemental composition on the surface, as shown in Figure 5. Initially, the atomic ratio of I/Pb (3.3) was close to the stoichiometry of MAPbI3, indicating no degradation occurred in the pristine films. The I/Pb ratio decreased to 2.2 after 7 days of air exposure, which indicated a large amount of iodine vacancies were generated during the degradation process. This effect was also further confirmed by monitoring the evolution of the work function on the surface with air-exposure time, as shown in Figure 5. The work function decreased with time from its initial 5.3 to 4.5 eV after 7 days of air exposure. This result verified the up-shift of the surface Fermi level induced by the generation of iodine-vacancy donor defects. The conduction-band and valence-band energy level of MAPbI3 film have been reported as 3.9 and 5.4 eV, respectively35. Thus, the surface Fermi level of the pristine film was 0.1 eV above the valence-band maximum, indicating p-type conduction. However, after air exposure for 7 days, the surface Fermi level moved upward by 0.8 eV, that is, to 0.15 eV above the middle of band gap, resulting in a transition of MAPbI3 surface from p-type to weak n-type conduction. Thus, iodine vacancies can be regarded as a major donor-type interface-state generated during degradation under air-exposure, which affects the electrical conduction of the film surface. How the interface-state changed the electrical transport properties of our Au/MAPbI3/FTO device? We studied rectifying I-V curves of the device over a relatively small voltage range (from −0.8 to 0.8 V) to determine it. Notably, the experiment should be carried out in a fresh device without undergoing forming process since which can largely change the film electrical transport properties (Figure 1). Figure 6a and b show the I-V curve and the band alignment of the pristine device. We observed that both forward and reverse currents were inhibited during the voltage sweep because of the Schottky-like barriers at the Au/MAPbI3 and MAPbI3/FTO


9


Page 10 of 34

interfaces. Thus, electrical transport of the pristine device can be described by an equivalent circuit, as shown in the inset of Figure 6a. This circuit consisted of two back-to-back Schottkydiodes and one bulk resistor that represented the MAPbI3 layer (Rb). The effects of air-exposure of the bottom MAPbI3/FTO interface were negligible compared with those relating to degradation on top surface54. Therefore, the Schottky barrier of MAPbI3/FTO remained constant over the whole experiment. Conversely, the Schottky barrier of the top Au/MAPbI3 interface declined because of the decrease of the surface work function with air-exposure (Figure 5). This effect was verified by the fact that the device under the forward bias is was gradually turn on with exposure time, as shown in Figure 6c-e. The corresponding equivalent circuit diagrams are also provided in the insets of Figure 6c-e. After exposing the device for 3–5 days, the work function of MAPbI3 surface decreased from 5.2 to 5.0 eV (Figure 4), which reduced the barrier of Au/MAPbI3 and finally resulted in the formation of an ohmic contact (Figure 6c-d and f-g). After further air-exposure for 7 days, the Fermi level shifted up into the middle of the bandgap (Ei) leading to the transition from p-type to weak n-type conduction behavior (Figure 5), eventually resulting in inversion of the Schottky barrier at the Au/MAPbI3interface. This inversion was confirmed from the results presented in Figure 6e and h, which show that the negative current was further suppressed by this additional barrier. The changes in electrical transport of Au/MAPbI3/FTO device, which can be well understood in terms of the shift of MAPbI3 surface Fermi level, in turn, further confirmed the formation of surface donor defects induced by airexposure. The RS transition in Figure 2 is a key indicator for studying the effects of interfacestate. In fact, the transition from no hysteresis to bipolar RS can be explained by the decrease of the Schottky barrier at the Au/MAPbI3 interface, as shown in Figure 6. The absence of hysteresis behavior may be explained by the large bias drop at the Schottky barrier of Au/MAPbI3


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


compared with that on the MAPbI3 layer, as indicated in Figure 6a. When the Schottky barrier decreased after exposure for 3 days, the bias drop on the MAPbI3 layer became larger and eventually resulted in bipolar RS. Next we studied the transition from bipolar RS to the NDR effect in greater detail. The further question based on above discussion is what is the relationship between interface-state and NDR effect. Before elucidating this issue, the switching mechanism of NDR should be investigated in the devices exposed to air for 7 days. Figure 7a shows the dependence of the HRS and LRS on the size of electrode area. It was clear that both HRS and LRS scaled with electrode area, suggesting typical homogeneous-type RS in the present devices55, which is quite different from a filamentary-type RS mechanism proposed in previous reports of hybrid perovskite based RRAM29,56. This homogeneous-type RS behavior is likely related to the degradation induced interface-state. We further verified this deduction by analyzing the complex impedance spectroscopy of HRS and LRS in Figure 7b and c. The spectra showed two semicircles for the HRS and a single semicircle for the LRS. The inset of Figure 7b shows the equivalent circuits for the HRS, which included three contributions to the impedance responses. The circuit consisted of one series resistor Rc and two parallel resistor / capacitor combinations with respective equivalent values of 100 kΩ / 0.25 nF (Rb/Cb) and 3 kΩ / 11.5 nF (Rs/Cs). The parallel RsCs represents a response at lower frequencies with a low equivalent resistance and high capacitance, implying that it derived from the Au/MAPbI3 interface owing to its lower thickness57. It should be emphasized that such RsCs was not observed in the device after 5 days exposure when only the MAPbI3/FTO junction should exist. This means the effect of MAPbI3/FTO interface is negligible in bipolar RS and NDR process. The forming process may be related to this phenomenon and will be discussed in later section. In contrast, RbCb represents


11


Page 12 of 34

the response of the film interior in the higher frequency region. Rc stands for the ohmic resistance, e.g. the contact resistance of the measurement57. Importantly, the larger equivalent resistance of the MAPbI3 film indicated that the bias mostly dropped across the bulk layer, rather than at the Au/MAPbI3 interface, at the start of a voltage sweep. Unlike the HRS, the equivalent circuit for the LRS was composed of a series resistor Rc and parallel one resistor / capacitor combination with equivalent values of 1.4 kΩ / 0.19 nF. Because the parallel component in the LRS had a similar capacitance value (0.25 nF and 0.19 nF) and high response frequency to RbCb of the HRS, we also attributed this component to the response of the film interior. The existence of a single semicircle in the LRS also indicated that the components showed semiconducting rather than metallic conduction behavior. We summarize the differences before and after the appearance of NDR accompanying RS behavior: (i) the disappearance of RsCs in the LRS indicated that the interface-state played a great role in the NDR behavior, which is consistent with the film degradation at its interface (Figures 4-6); (ii) Rb decreased from 100 to 1.4 kΩ when the device transitioned from the HRS to the LRS, suggesting that the switching also involved the interior of the MAPbI3 layer as well as the interface-state. Now, let’s further examine the roles of bulk film and interface-state in the NDR process. We replotted the I-V curves for HRS and LRS, during a positive sweep, on a log-log scale, as shown in Figure 8a. Compared with the typical Ohmic conduction seen for the LRS, the HRS showed a complicated conduction behavior. When the positive voltage was less than Vset (1 V), the conduction of the HRS could be divided into three regions with different fitting slopes (1, 1.9, and 5.9). This result is in good agreement with the trap-controlled space charge limited conduction (SCLC) mechanism58. Similar behavior was also observed in bipolar RS of devices after 3-5 days exposure (See Figure S1 in the Supporting Information). When the positive bias


12

Page 13 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


was larger than Vset, the current decreased with increasing voltages between 1 to 3 V and presented a typical NDR effect. This SCLC conduction was attributed to the trapping / detrapping of electrons in bulk defects (e.g. iodine vacancy)16, and thus accounts for the change of Rb during the HRS-to-LRS transition, as discussed in Figures 7b-c. The NDR behavior with a negative slope (~ −1) was likely related to interfacial defects rather than bulk defects, as has been reported in the literature38. It should be clarified that forming process may also strongly impact the bulk defects according to the fact that obvious HRS reduction was obtained after it (Figure 2). As previously reported13,24, the migration of halide ion likely occurs under high electric field in perovskite, resulting in a great deal of halide vacancies and associated move up of Femi level in film. This theory may be also applicable in the forming process of our devices, which generates more iodine vacancies and raises Femi level along the conductive path. It can be one of the reasons for HRS reduction in following RS operations (Figure 2) and elimination of the effect of MAPbI3/FTO interface (Figure 7b). Another possible reason can be attributed to the uncomplete release of trapped electrons during reset process from the full-filled LRS state with ohmic conduction. To better understand how interface-state affected the NDR behavior, we studied the relationship between the amplitude of the applied positive bias and the LRS state, as shown in Figure 7b. The negative sweep remained constant from 0 to −3 V, and no bipolar RS or NDR occurred when the applied positive bias was zero or smaller than 1.5 V. In other words, a sufficiently large forward bias (≥1.5 V) was necessary to induce the RS and NDR phenomena. Importantly, the switching window and reset current became larger as the applied positive voltage was increased from 1.5 V to 3 V in the NDR region. This result implies that although the bulk resistance switching happens during the SCLC process, the final switching state is controlled by the applied voltage in the NDR region. In addition, we also observed that the HRS


13


Page 14 of 34

decreases with increasing the positive applied bias in Figure 8b. It can be explained that the value of positive bias determines the maximum trapped level and larger bias leads to more remaining electrons in some traps after release process in reset. Considering all the above experimental results, the origin of NDR accompanying RS phenomenon can be explained by a proposed model, as illustrated in Figure 9. (I) Initially, the HRS contained many bulk defects in the film and donor-like interfacial defects (e.g., iodine vacancies) in the forbidden band39. (II) When a small positive voltage was applied to the Au electrode (1 V) led to the Fermi level at the Au/MAPbI3 interface moving up considerably and electrons become trapped by interface-states below the Fermi level. As a result, the current decreased with increasing voltage. This explains the NDR phenomenon achieved during the set process. (IV) Once the positive bias was swept back to a value smaller than 1 V (3 V→ 0 V), the Fermi level dropped but interface-state above the Fermi level remained filled. (V) Thus, the current sharply increased owing to the disappearance of the interfacial limitations and the resistance changed from the HRS to the LRS, as shown in the I-V curve of Figure 9. (VI) During the reset process, the Fermi level was lowered as a negative bias was applied to Au. Trapped electrons at interfacestate as well as those in bulk defects detrapped under the effect of the bias and high Joule


14

Page 15 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


heating59. Thus, the LRS-to-HRS transition occurred in the reset process. The reset current and switching window should depend on the height of levels trapped by interface-state in the set process, as observed in Figure 8b. Another interesting advantage of hybrid organic−inorganic perovskite is their low processing temperatures60, which motivated us to assess the feasibility of flexible RRAMs based on MAPbI3 materials. We fabricated flexible devices on ITO/PET substrates and studied their stability and endurance by mechanical bending tests. As shown in Figure 10a, the 1.5-cm-long flat device was bent to lengths of 1.0, 1.2, and 1.4 cm for in-situ electrical measurements; the inset is an image of bending the device. The device performance was evaluated by recording I–V curves for 30 RS cycles. The statistical data in Figure 10a reveal that the HRS and LRS remain∼3×104Ω and ∼500Ω respectively, ensuring error-free ratio of ∼60 nearly the same as those of the unbent device. Figure 10b shows a continuous bending test carried out 800 times at a high frequency of 10bends/s; reliable switching still can occur with increasing bending cycles. Figure 10c and d show the I-V curves and retention characteristics of the memory device before and after bending. In addition, 100 switching cycles were also performed to verify the reliability of the memory devices after bending test. It can be observed that there is no degradation for the NDR behavior was induced by the mechanical bending. The HRS and LRS can be retained for over 104 s before and after bending test, indicating good information retention. These results suggest that the device operation is stable towards mechanical stress and strain for flexible memory applications. 3. CONCLUSIONS:


15


Page 16 of 34

In summary, the interface-state effects on the RS behavior of hybrid organic−inorganic perovskite were demonstrated. An interesting NDR phenomenon presented with degradation of MAPbI3 under air-exposure. More importantly, we verified that the occurrence of NDR behavior can be attributed to degradation induced interface-state, such as iodine vacancies. The trapping / detrapping of electrons in bulk defects and interface-state accounts for the NDR and bipolar RS behavior. The current work not only presents a unique NDR behavior, but also clarifies the role of interface-state in hybrid perovskite RS memories. Interface-state and chemical instability are critical issues for hybrid perovskite materials. Our study of the interface-state effects on RS behavior will help guide the development of more accurate switching mechanisms for memory applications based on hybrid organic−inorganic perovskites. Furthermore, the hybrid perovskite device also exhibits potential for applications in flexible memory. 4. EXPERIMENTAL SECTION: 4.1 CH3NH3PbI3 Thin Film Synthesis and Device Fabrication: First, FTO substrates were washed sequentially by ultrasonic treatment in acetone and ethanol. The MAPbI3 layer was then fabricated by a two-step deposition method: (i) PbI2 thin films were first grown by a sol-gel method in a glove box (filled with N2), maintaining the substrate temperature at 90 °C during preparation; (ii) After deposition of the PbI2 layer, samples were immersed into a mixed solution of methylammonium iodide (MAI) (10 mg mL-1) and 2-propanol for 120 s in air, and were then annealed at 100 °C for 30 min to obtain the MAPbI3 layer. Finally, the Au top-electrodes (TEs) were thermally evaporated and patterned into circular pads with a diameter of 0.5 mm using a shadow mask.


16

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


4.2 Characterization: The perovskite film was characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), atomic force microscopy (AFM) and XPS. The work function of the MAPbI3surface was determined by a KP020 Kelvin Probe system. The electrical properties of the devices were characterized by a Keithley 2636A semiconductor analyzer in a voltage-sweep mode under ambient and dark conditions. In addition, simultaneous impedance characterization was performed using a CHI 600D electrochemical workstation over a frequency range from 0.5 kHz to 100 kHz under an amplitude of 50 mV. ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: XXX I-V curves and HRS/LRS resistance after different exposure days; I-V characteristic under the condition of different humidity; XRD patterns; I-V curves of log-log scale after different exposure days. AUTHOR INFORMATION Corresponding Author *E-mail: [email protected] (Haiyang Xu); [email protected] (Zhongqiang Wang). †The authors Hanlu Ma and Wei Wang contributed equally to this work.

ORCID: Haiyang Xu: 0000-0001-6034-8736 Zhongqiang Wang: 0000-0002-1133-6058 Notes


17


Page 18 of 34

The authors declare no competing financial interest. ACKNOWLEDGMENT This work was supported by the NSFC for Excellent Young Scholars (No. 51422201), the NSFC Program (Nos. 51701037, 51732003, 61774031, and 61574031), the “111” Project (No. B13013), the Fund from Jilin Province (Nos. 20160101324JC, and 20180520186JH), the Fundamental Research Funds for the Central Universities (No. 2412018ZD004). REFERENCES (1) Waser, R.; Dittmann, R.; Staikov, G.; Szot, K. Redox-Based Resistive Switching Memories– Nanoionic Mechanisms, Prospects, and Challenges. Adv. Mater. 2009, 21, 2632-2663. (2) Yang, F. P. Y.; Liu, Q.; Liu, M.; Zeng, F. Fully Room-Temperature-Fabricated Nonvolatile Resistive Memory for Ultrafast and High-Density Memory Application. Nano Lett. 2009, 9, 1636-1643. (3) Zhu, X.; Su, W.; Liu, Y.; Hu, B.; Pan, L.; Lu, W.; Zhang, J.; Li, R.-W. Observation of Conductance Quantization in Oxide-Based Resistive Switching Memory. Adv. Mater. 2012, 24, 3941-3946. (4) Wang, Z.; Xu. H.; Zhang, L.; Li, X.; Ma, J.; Zhang, X.; Liu, Y. Performance Improvement of Resistive Switching Memory Achieved by Enhancing Local-Electric-Field Near Electromigrated Ag-Nanoclusters. Nanoscale 2013, 5, 4490-4494. (5) Zhang, L.; Xu, H.; Wang, Z.; Yu, H.; Zhao, X.; Ma, J.; Liu, Y. Oxygen-Concentration Effect on P-Type CuAlOx Resistive Switching Behaviors and the Nature of Conducting Filaments. Appl. Phys. Lett. 2014, 104, 093512. (6) Liu, S.; Lu, N.; Zhao, X.; Xu, H.; Banerjee, W.; Lv, H.; Long, S.; Li, Q.; Liu, Q.; Liu, M. Eliminating Negative-SET Behavior by Suppressing Nanofilament Overgrowth in Cation-Based Memory. Adv. Mater. 2016, 28, 10623-10629. (7) Shin, K.-Y.; Kim, Y.; Antolinez, F. V.; Ha, J. S.; Lee, S.-S.; Park, J. H. Controllable Formation of Nanofilaments in Resistive Memories via Tip-Enhanced Electric Fields. Adv. Electron. Mater. 2016, 2, 1600233. (8) Ambrogio, S.; Balatti, S.; Choi, S.; Ielmini, D. Impact of the Mechanical Stress on Switching Characteristics of Electrochemical Resistive Memory. Adv. Mater. 2014, 26, 3885-3892. (9) Asanuma, S.; Akoh, H.; Yamada, H.; Sawa, A. Relationship between Resistive Switching Characteristics and Band Diagrams of Ti/Pr1-xCaxMnO3 Junctions. Phys. Rev. B 2009, 80, 235113.


18

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


(10) Lee, B.-H.; Bae, H.; Seong, H.; Lee, D.-I.; Park, H.; Choi, Y. J.; Im, S.-G.; Kim, S. O.; Choi, Y.-K. Direct Observation of A Carbon Filament in Water-Resistant Organic Memory. ACS Nano 2015, 9, 7306-7313. (11) Chakrabarti, S.; Maikap, S.; Samanta, S. Scalable Cross-Point Resistive Switching Memory and Mechanism through an Understanding of H2O2/Glucose Sensing Using an IrOx/Al2O3/W Structure. Phys. Chem. Chem. Phys. 2017, 19,25938-25948. (12) Chakrabarti, S.; Ginnaram, S.; Jana, S. Negative Voltage Modulated Multi-Level Resistive Switching by Using a Cr/BaTiO x /TiN Structure and Quantum Conductance through Evidence of H 2O2 Sensing Mechanism. Sci Rep. 2017, 7,4735. (13) Zhou, F.; Liu, Y.; Shen, X.; Wang, M.; Yuan, F.; Chai, Y. Low-Voltage, Optoelectronic CH3NH3PbI3-xClx Memory with Integrated Sensing and Logic Operations. Adv. Funct. Mater. 2018, 1800080. (14) Yang, K.; Li, F.; Veeramalai, C. P. A Facile Synthesis of CH3NH3PbBr3 Perovskite Quantum Dots and Their Application in Flexible Nonvolatile Memory. Appl. Phys. Lett. 2017, 110,14097. (15) Cheng, X. F.; Hou, X.; Zhou, J. Pseudohalide-Induced 2D (CH3NH3)2 PbI2 (SCN)2 Perovskite for Ternary Resistive Memory with High Performance. Small. 2018,1703667. (16) Yoo, E. J.; Lyu, M.; Yun, J. H.; Kan,g C. J.; Choi, Y. J.; Wang, L. Resistive Switching Behavior in Organic–Inorganic Hybrid CH3NH3PbI3-xClx Perovskite for Resistive Random Access Memory Devices. Adv. Mater. 2015, 27, 6170-6175. (17) Tan, Z.-K.; Moghaddam, R. S.; Lai, M. L.; Docampo, P.; Higler, R.; Deschler, F.; Price, M.; Sadhanala, A.; Pazos, M. L.; Credgington, D.; Hanusch, F.; Bein, T.; Snaith, H. J.; Friend, R. H. Bright Light-Emitting Diodes Based on Organometal Halide Perovskite. Nat. Nanotechnol. 2014, 9, 687-692. (18) Dou, L.; Yang, Y. M.; You, J.; Hong, Z.; Chang, W.-H.; Li, G.; Yang, Y. SolutionProcessed Hybrid Perovskite Photodetectors with High Detectivity. Nat. Commun. 2014, 5, 5404. (19) Bi, D.; Yi, C.; Luo, J.; Décoppet, J.-D.; Zhang, F.; Zakeeruddin, S. M.; Li, X.; Hagfeldt, A.; Grätzel, M. Polymer-Templated Nucleation and Crystal Growth of Perovskite Films for Solar Cells with Efficiency Greater Than 21%. Nat. energy 2016, 1, 16142. (20) Xiao, Z.; Huang, J. Energy-Efficient Hybrid Perovskite Memristors and Synaptic Devices. Adv. Electron. Mater. 2016, 2, 1600100. Dong, J. K.; Tak, Y. J.; Kim, W. G. Resistive Switching Properties through Iodine Migrations of a Hybrid Perovskite Insulating Layer. Adv. Mater. Interfaces. 2017, 4, 1601035.

(21)


19


Page 20 of 34

(22) Zhu, X.; Lu, W. D. Optogenetics-Inspired Tunable Synaptic Functions in Memristors. ACS Nano. 2018, 12, 1242−1249. (23) Ren, Y.; Milo, V.; Wang, Z. Analytical Modeling of Organic–Inorganic CH3NH3PbI3 Perovskite Resistive Switching and its Application for Neuromorphic Recognition. Adv. Theory Simul. 2018，1700035. Xiao, Z.; Yuan, Y.; Shao, Y.; Wang, Q.; Dong, Q.; Bi, C.; Sharma, P.; Gruverman, A.; Huang, J. Giant Switchable Photovoltaic Effect in Organometal Trihalide Perovskite Devices. Nat. Mater. 2015, 14, 193. (24)

(25) Sawa, A. Resistive Switching in Transition Metal Oxides. Mater. Today 2008, 11, 28-36. (26) Wu, X. L. S.; Turner, S.; Peng, H.; Lin, W.; Ding, J.; David, A.; Wang, B.; Tendeloo, G.; Wang, J.; Wu, T. Nonvolatile Resistive Switching in Pt/LaAlO3/SrTiO3 Heterostructures. Phys. Rev. X 2013, 3, 041027. (27) Du, A. K. Y.; Pan, H.; Zeng, K.; Wang, S.; Yang, P.; Wee, A. T. S. The Resistive Switching in TiO2 Films Studied by Conductive Atomic Force Microscopy and Kelvin Probe Force Microscopy. AIP Adv. 2013, 3, 082107. (28) Yan, Z. B.; Liu, J.-M. Coexistence of High Performance Resistance and Capacitance Memory Based on Multilayered Metal-Oxide Structures. Sci Rep. 2013, 3, 2482. (29) Choi, J.; Park, S.; Lee, J. Organolead Halide Perovskites for Low Operating Voltage Multilevel Resistive Switching. Adv. Mater. 2016, 28,6562-6567. (30)Chen, Y. C.; Chung, Y. L.; Chen, B. T. Revelation on the Interrelated Mechanism of Polarity-Dependent and Multilevel Resistive Switching in TaOx-Based Memory Devices. J. Phys. Chem. C. 2013, 117,5758–5764. (31) Niu, G.; Li, W.; Meng, F.; Wang, L.; Dong, H.; Qiu, Y. Study on the Stability of CH3NH3PbI3 Films and the Effect of Post-Modification by Aluminum Oxide in All-Solid-State Hybrid Solar Cells. J. Mater. Chem. A 2014, 2, 705-710. (32) Hu, L.; Shao, G.; Jiang, T.; Li, D.; Lv, X.; Wang, H.; Liu, X.; Song, H.; Tang, J.; Liu, H. Investigation of the Interaction between Perovskite Films with Moisture via in Situ Electrical Resistance Measurement. ACS Appl. Mater. Inter. 2015, 7, 25113-25120. (33) Leguy, A. M. A.; Hu, Y.; Campoy-Quiles, M.; Alonso, M. I.; Weber, O. J.; Azarhoosh, P.; Schilfgaarde, M. van; Weller, M. T.; Bein, T.; Nelson, J.; Docampo, P.; Barnes, P. R. F. Reversible Hydration of CH3NH3PbI3 in Films, Single Crystals, and Solar Cells. Chem. Mater. 2015, 27, 3397-3407. (34) Niu, G.; Guo, X.; Wang, L. Review of Recent Progress in Chemical Stability of Perovskite Solar Cells. J. Mater. Chem. A 2015, 3, 8970-8980.


20

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


(35) You, J.; Meng, L.; Song, T.-B.; Guo, T.-F.; Yang, Y. M.; Chang, W.-H.; Hong, Z.; Chen, H.; Zhou, H.; Chen, Q.; Liu, Y.; De Marco, N.; Yang, Y. Improved Air Stability of Perovskite Solar Cells via Solution-Processed Metal Oxide Transport Layers. Nat. Nanotechnol. 2016, 11, 75. (36) Zhang, L.; Ju, M.-G.; Liang, W. The Effect of Moisture on the Structures and Properties of Lead Halide Perovskites: A First-Principles Theoretical Investigation. Phys. Chem. Chem. Phys. 2016, 18, 23174-23183. (37) Liu, P. Z. Z.; Men,g Y.; Tian, H.; Li, J.; Pan, X.; Liang, X.; Chen, H. Z. D. Effect of TaOx Thickness on the Resistive Switching of Ta/Pr0.7Ca0.3MnO3/Pt Films. Appl. Phys. Lett. 2012, 100, 143506. (38) Jia, C.; Sun, X.; Li, G.; Chen, Y.; Zhang, W. Origin of Attendant Phenomena of Bipolar Resistive Switching and Negative Differential Resistance in SrTiO3:Nb/ZnO Heterojunctions. Appl. Phys. Lett. 2014, 104, 043501. (39) Yang, G.; Jia, C.; Chen, Y.; Chen, X.; Zhang, W. Negative Differential Resistance and Resistance Switching Behaviors in BaTiO3 Thin Films. J. Appl. Phys. 2014, 115, 204515. (40) Esaki, L. New Phenomenon in Narrow Germanium p-n Junctions. Phys. Rev. 1958, 109, 603. (41) Zhao, P.; Cui, H.; Woolard, D.; Jensen, K. L.; Buot, F. A. Simulation of Resonant Tunneling Structures: Origin of the I-V Hysteresis and Plateau-Like Structure. J. Appl. Phys. 2000, 87, 1337-1349. (42) Brown, E. R.; Söderström, J. R.; Parker, C. D.; Mahoney, L. J.; Molvar, K. M.; McGill, T. C. Oscillations Up to 712 GHz in InAs/AlSb Resonant-Tunneling Diodes. Appl. Phys. Lett. 1991, 58, 2291-2293. (43) Tom, B. B.; Broekaert, P. E.; Paul, A. van der Wagt, J.; Seabaugh, A. C.; Morris, F. J.; Moise, T. S.; Beam, E. A.; Frazier, G. A. A Monolithic 4-bit 2-Gsps Resonant Tunneling Analog-to-Digital Converter. IEEE J. Solid-State Circuits 1998, 33, 1342-1349. (44) Chen, M. A. R. J.; Rawlett, A. M.; Tour, J. M. Large On-Off Ratios and Negative Differential Resistance in a Molecular Electronic Device. Science 1999, 286, 1550-1552. (45) Mathews, R.H.; Sage, J.P.; Sollner, T.C.L.G.; Calawa, S.D.; Chen, C.-L.; Mahoney, L.J.; Maki, P.A.; Molvar, K.M. A New RTD-FET Logic Family. Proc. IEEE 1999, 87, 596-605. (46)Hasegawa, H.; Inabe, T. Electrical Properties of Organic-Inorganic Hybrid Tin Bromide Cubic Perovskites: Hole-Doping and Iodide Substitution Effects. New J. Chem. 2016 40,70437047. (47) Li, B.; Li, Y.; Zhen,g C.; Gao, D.; Huang, W. Advancements in the Stability of Perovskite Solar Cells: Degradation Mechanisms and Improvement Approaches. RSC Adv. 2016, 6, 3807938091.


21


Page 22 of 34

(48) Guo, X.; McCleese, C.; Kolodziej, C.; Samia, A. C.; Zhao, Y.; Burda, C. Identification and Characterization of the Intermediate Phase in Hybrid Organic-Inorganic MAPbI3 Perovskite. Dalton Trans. 2016, 45, 3806-3813. (49) McMahon, B. W.; Yu, J.; Boatz, J. A.; Anderson, S. L. Rapid Aluminum Nanoparticle Production by Milling in NH3 and CH3NH2 Atmospheres: an Experimental and Theoretical Study. ACS Appl. Mater. Inter. 2015, 7, 16101-16116. (50) Lindblad, R.; Bi, D.; Park, B.-W.; Oscarsson, J.; Gorgoi, M.; Siegbahn, H.; Odelius, M.; Johansson, E. M.; Rensmo, H. Electronic Structure of TiO2/CH3NH3PbI3 Perovskite Solar Cell Interfaces. J. Phys. Chem. Lett. 2014, 5, 648-653. (51) Philippe, B.; Park, B.-W.; Lindblad, R.; Oscarsson, J.; Ahmadi, S.; Johansson, E. M. J.; Rensmo, H. Chemical and Electronic Structure Characterization of Lead Halide Perovskites and Stability Behavior under Different Exposures-A Photoelectron Spectroscopy Investigation. Chem. Mater. 2015, 27, 1720-1731. (52) Wang, Q.; Shao, Y.; Xie, H.; Lyu, L.; Liu, X.; Gao, Y.; Huang, J. Qualifying Composition Dependent p and n Self-Doping in CH3NH3PbI3. Appl. Phys. Lett. 2014, 105, 163508. (53) Alberti, A.; Deretzis, I.; Pellegrino, G.; Bongiorno, C.; Smecca, E.; Mannino, G.; Giannazzo, F.; Condorelli, G. G.; Sakai, N.; Miyasaka, T.; Spinella, C.; Magna, A. L. Similar Structural Dynamics for the Degradation of CH3NH3PbI3 in Air and in Vacuum. ChemPhysChem 2015, 16, 3064-3071. (54) Shao, Y.; Xiao, Z.; Bi, C.; Yuan, Y.; Huang, J. Origin and Elimination of Photocurrent Hysteresis by Fullerene Passivation in CH3NH3PbI3 Planar Heterojunction Solar Cells. Nat. Commun. 2014, 5, 5784. (55) Rana, A.; Lu, H.; Bogle, K.; Zhang, Q.; Vasudevan, R.; Thakare, V.; Gruverman, A.; Ogale, S.; Valanoor, N. Scaling Behavior of Resistive Switching in Epitaxial Bismuth Ferrite Heterostructures. Adv. Funct. Mater. 2014, 24, 3962-3969. (56) Gu, C.; Lee, J.-S. Flexible Hybrid Organic-Inorganic Perovskite Memory. ACS Nano 2016, 10, 5413-5418. (57) Jiang, X.; Zhao, Y.; Chen, Y.; Li, D.; Luo, Y.; Zhao, D.; Sun, Z.; Sun, J.; Zhao, H. Characteristics of Different Types of Filaments in Resistive Switching Memories Investigated by Complex Impedance Spectroscopy. Appl. Phys. Lett. 2013, 102, 253507. (58) Yang, W.-Y.; Kim, W.-G.; Rhee, S.-W. Radio Frequency Sputter Deposition of Single Phase Cuprous Oxide Using Cu2O as A Target Material and Its Resistive Switching Properties. Thin Solid Films 2008, 517, 967-971. (59) Chen, A.; Haddad, S.; Wu, Y. C.; Fang, T. N.; Kaza, S.; Lan, Z. Erasing Characteristics of Cu2O Metal-Insulator-Metal Resistive Switching Memory. Appl. Phys. Lett. 2008, 92, 013503.


22

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


(60) Tavakoli, M. M.; Tsui, K.-H.; Zhan,g Q.; He, J.; Yao, Y.; Li, D.; Fan, Z. Highly Efficient Flexible Perovskite Solar Cells with Antireflection and Self-Cleaning Nanostructures. ACS Nano 2015, 9, 10287-10295


23


Page 24 of 34

Figure 1. (a) Structural diagram of the Au/MAPbI3/FTO memory cell. (b) SEM image of the surface morphology and cross-sectional SEM image of the as-prepared MAPbI3 film. (c) AFM image of the perovskite layer. (d) Typical XRD data of MAPbI3 thin films. Diffraction peaks at 2θ of 14.2, 28.6, and 32.0° represent the crystal faces (110), (220), and (310) of tetragonal perovskite structure.


24

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


Figure 2. Evolution of RS behavior with air-exposure time of MAPbI3 film. (a) No hysteresis behavior was observed in pristine device. (b-c) Typical bipolar RS and self-limited LRS after 3 and 5 days of air exposure. (d) Typical NDR effect after 7 days of air exposure.


25


Page 26 of 34

Figure 3. (a) and (b) The evolution of HRS and LRS states with time and 100 consecutive RS cycles. (c) Schematic diagram of a setup for RS-speed measurement, which consists of a pulse generator, an oscilloscope, a memory cell and a load resistor. (d) and (e)Transient responses of the device for set and reset process.


26

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


Figure 4. XPS spectra of (a) N1s, (b) Pb 4f, and (c) I 3d core levels of MAPbI3 film surface measured after different exposure times


27


Page 28 of 34

Figure 5. Evolution of I/Pb atomic ratio and work function with air-exposure time.


28

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 6. (a) Rectifying curve and (b) band alignment of pristine Au/MAPbI3/FTO device. (c-e) Rectifying curve of the device measured at different air-exposure times and (f-h) corresponding band alignment of the Au/MAPbI3 interface. Insets show the corresponding equivalent circuit diagrams. The work functions of Au, FTO and fresh CH3NH3PbI3 film are 5.1eV, 4.7eV and 5.3eV, respectively.


29


Page 30 of 34

Figure 7. (a) Dependence of HRS and LRS resistance on the diameter of Au electrode. (b-c) Complex impedance spectra of the HRS and LRS. Corresponding equivalent circuits are shown in the insets.


30

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 8. (a) Typical I-V curve of bipolar RS behavior accompanied with NDR effect on a loglog scale. (b) Switching RS behavior with different applied positive biases from 0 to 3 V.


31


Page 32 of 34

Figure 9. Schematic diagram illustrating the physical mechanism of RS and NDR behaviors. Left figure shows typical I-V curve of RS behavior accompanied with NDR effect. (I-VI) the energy band structure showingthe trapping / detrapping of electrons in bulk defects and interface states, bias-induced shift of MAPbI3 surface Femi-level, and each diagram corresponds to a specific point indicated in the I-V curve.


32

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 10. Average values of the HRS and LRS as a function of bending lengths (a) and bending cycles (b). (c) The I-V curves of the device before and after bending. (d) the retention test and consecutive NDR operations after bending.


33


Table of Contents Graphic 95x53mm (300 x 300 DPI)


Page 34 of 34

Interface State-Induced Negative Differential ... - ACS Publications

Recommend Documents