An Unusual Mechanism for Negative Differential Resistance in

Subscriber access provided by UNIV OF NEWCASTLE

Article

An unusual mechanism for negative differential resistance in ferroelectric nanocapacitors: polarization switchinginduced charge injection followed by charge trapping Peilian Li, Zhifeng Huang, Zhen Fan, Hua Fan, Qiuyuan Luo, Chao Chen, Deyang Chen, Min Zeng, Minghui Qin, Zhang Zhang, Xubing Lu, Xingsen Gao, and Jun-Ming Liu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b05634 • Publication Date (Web): 25 Jul 2017 Downloaded from http://pubs.acs.org on July 27, 2017

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 free 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 accessible to all readers and 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.

ACS Applied Materials & Interfaces 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 23

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

An Unusual Mechanism for Negative Differential Resistance in Ferroelectric Nanocapacitors: Polarization Switching-Induced Charge Injection Followed by Charge Trapping Peilian Li,† Zhifeng Huang,† Zhen Fan,*,† Hua Fan,† Qiuyuan Luo,† Chao Chen,† Deyang Chen,† Min Zeng,† Minghui Qin,† Zhang Zhang,† Xubing Lu,† Xingsen Gao,*,† and Jun-Ming Liu.†, ‡ †Institute for Advanced Materials and Guangdong Provincial Key Laboratory of Quantum Engineering and Quantum Materials, South China Academy of Advanced Optoelectronics, South China Normal University, Guangzhou 510006, China ‡Laboratory of Solid State Microstructures and Innovation Center of Advanced Microstructures, Nanjing University, Nanjing 21009, China KEYWORDS: Negative differential resistance, Ferroelectric nanocapacitors, Polarization switching, Charge injection, Charge trapping ABSTRACT: Negative differential resistance (NDR) has been extensively investigated for its wide device applications. However, a major barrier ahead is the low reliability. To address the reliability issues, we consider ferroelectrics and propose an alternative mechanism for realizing the NDR with deterministic current peak positions, in which the NDR results from the polarization switching-induced charge injection and subsequent charge trapping at the metal/ferroelectric

interface.

In

this

work,

ferroelectric

Au/BiFe0.6Ga0.4O3

(BFGO)/Ca0.96Ce0.04MnO3 (CCMO) nanocapacitors are prepared, and their ferroelectricity and NDR behaviors are studied concurrently. It is observed that the NDR current peaks are located at the vicinity of coercive voltages (Vc) of the ferroelectric nanocapacitors, thus evidencing the proposed mechanism. In addition, the NDR effect is reproducible and robust with good endurance and long retention time. This study therefore demonstrates a ferroelectric-based NDR device, which may facilitate the development of highly reliable NDR devices.

ACS Paragon Plus Environment

1


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 23

1. INTRODUCTION Negative differential resistance (NDR), a nonlinear transport phenomenon where the current decreases with increasing applied voltage, has attracted considerable attention owing to its numerous device applications, such as diodes, oscillators, amplifiers, and analog-to-digital converters.1-4 It has been found that a wide variety of materials could exhibit the NDR, including semiconductor

quantum

wells,5

metal-oxide

heterostructures,6

polymer-nanoparticle

composites,7 and biological molecules.8 For most of them, however, the greatest challenge is to achieve a reproducible and stable NDR effect. For example, the voltage corresponding to the current peak of NDR (Vpeak) may vary from sample to sample, and it may also shift successively and even disappear during the cyclic voltage sweeps. These reliability issues may be caused by the uncertainty and complexity associated with the conventional mechanisms of NDR, such as charge trapping and detrapping,6,9,10 filament formation and rupture,11,12 and redox reaction.8,13 An innovative and alternative mechanism for generating the NDR effect is therefore of great need for the design of highly reliable NDR devices. Ferroelectrics, a class of materials whose spontaneous polarization can be electrically switched,14,15 may offer a unique mechanism for realizing the NDR with high reliability. For a semiconducting ferroelectric material, the switching of polarization direction can easily modulate the carrier injection through the metal/ferroelectric interface.16 Therefore, the coercive voltage (Vc) of the polarization switching that triggers the carrier injection, may result in a deterministic Vpeak for the NDR (Figure 1a). Note that this work deals with the current-voltage (I-V) characteristics measured in the low frequency regime whose time scale is much longer than the polarization switching time. Therefore, the current of the NDR shown in Figure 1a does not represent the displacement current of polarization switching. Instead, it is formed by the charge injection which is triggered by the polarization switching and subsequently suppressed by the charge trapping (see the detailed descriptions below).


2

Page 3 of 23

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 1. Schematic illustrations of the mechanism for NDR in ferroelectrics. (a) Typical NDR characteristics in a ferroelectric capacitor. Energy band diagrams of a metal electrode (M)/dead layer (DL)/ferroelectrics (FE) structure at different voltage regimes: (b) Regime i, (c) Regime ii, and (d) Regime iii. The three voltage regimes are indicated in Panel a with different colors, and their boundaries are drawn schematically. Although the exact mechanism for the electron injection is unknown, it may be realized via thermionic emission (green arrows), field emission (brown arrows) and trap-assisted tunneling (blue arrows), all of which are strongly dependent on the interface barrier.

A ferroelectric capacitor can be considered as a series connection of two back-to-back Schottky barriers formed at the metal/ferroelectric interfaces and a bulk resistance (taking the ntype junction as an example). When a positive voltage is applied to the top electrode, the bottom Schottky barrier which is reverse-biased mainly controls the current flow. (Note: the voltage


3


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 4 of 23

sweep rate is sufficiently low so that the displacement current is insignificant). Let us take a further look at the bottom barrier, in which a dead layer (DL) (a thin dielectric layer) is assumed to exist between the ferroelectric layer (FE) and the metal electrode (M), as shown in Figure 1bd.17 If the polarization is initially oriented upward, the negative polarization charge at the bottom barrier can induce an upward band bending, suppressing the electron injection (Figure 1b). Therefore, the current flowing through the barrier is low when the applied positive voltage is much lower than Vc. When the applied voltage approaches Vc, the polarization switching occurs. The positive polarization charge can induce a large electric field (Ed) within the dead layer:17 Ed ≈

P

ε 0ε d

,

(1)

where P is the polarization, ε0 is the vacuum permittivity, and εd is the static dielectric constant of the dead layer. Given P = 60 µC/cm2 and εd = 100, typical values for normal ferroelectrics, the resultant Ed can be as large as ~6.8 MV/cm. This large Ed leads to a significant barrier lowering and in turn triggers an intense electron injection (Figure 1c). As the applied voltage further increases, the electron injection continues but will be suppressed gradually. This is because the barrier height increases with the applied voltage due to the electron trapping. The injected electrons have a certain probability to be trapped inside the dead layer or the depletion region of ferroelectric layer. These trapped electrons can compensate the positive polarization charge, resuming the Schottky barrier gradually and making the subsequent electron injection more difficult (Figure 1d). Therefore, the current decreases with increasing applied voltage. As described above, the proposed mechanism for the NDR in ferroelectrics consists of two aspects: i) current rise caused by the polarization switching-induced charge injection, and ii) current drop due to the charge trapping (hereafter this mechanism is termed as “PSCICT”). It is worth noting that the NDR effects in ferroelectrics were reported previously, but those studies did not show evident correlations between the Vpeak and Vc.18-20 In addition, the reported mechanisms, such as polarization relaxation,18 charge trapping and detrapping,19 diffusionlimited conduction,20 and interband tunneling due to the band overlap,21 were completely different from the mechanism proposed here. The merit of the PSCICT mechanism is that the coercive voltage Vc determines the position of current peak, i.e., Vpeak, which is a core ingredient of physics for ensuring the high reliability of the NDR effect.


4

Page 5 of 23

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


To realize the PSCICT mechanism, the carrier transport through the extrinsic conduction channels, e.g., domain walls, grain boundaries, and dislocations, should be avoided. Otherwise, the NDR will be overwhelmed by the extrinsic leakage currents whose magnitudes increase monotonically with increasing applied voltage. As a result, high-quality epitaxial ferroelectric thin films capped by the nano-sized electrodes are demanded. Additionally, according to Eq. (1), a larger polarization is favorable for a more significant NDR. In this regard, ferroelectric BiFe0.6Ga0.4O3 (BFGO), which exhibits a stable super-tetragonal phase, a giant polarization of ~150 µC/cm2,22,23 and a narrow bandgap (~2.95 eV, see Figure S1), could be a preferred candidate material. In this work, we fabricate the ferroelectric Au/BFGO (~20 nm)/Ca0.96Ce0.04MnO3 (CCMO) nanocapacitors (~0.1 µm2 in area) to check the validity of the NDR mechanism proposed above. We choose the CCMO as the bottom electrode for i) the lattice matching (the in-plane lattice constants of BFGO and CCMO stable phases are 3.77 Å and 3.74 Å, respectively);22,24 and ii) constructing an n-type Schottky contact with the BFGO (the work function of CCMO is ~5.2 eV25,26 and the BFGO may be an n-type semiconductor with an electron affinity of ~3.3 eV27,28 ). With these nanocapacitors, we find that the NDR current peaks are located at the vicinity of Vc, thus validating the PSCICT mechanism. It is further shown that the NDR effect of the Au/BFGO/CCMO nanocapacitors has good reproducibility, high endurance, and long retention.

2. EXPERIMENTAL PROCEDURE The BFGO thin films of ~20 nm in thickness together with the ~5-nm CCMO buffer layers were grown on the (001)-oriented LaAlO3 (LAO) substrates by pulsed laser deposition with a KrF excimer laser (λ = 248 nm). During the deposition, the substrate temperature was kept at 650 °C and the oxygen pressure was 15 Pa, respectively. After the deposition, the samples were cooled down to room temperature at a rate of 5 °C/min in oxygen ambient of 1.0 atm. Then, the Au top electrodes with a lateral size of ~300 nm and a thickness of ~12 nm were ex situ grown on the film surfaces using polystyrene spheres as the templates.29 The crystal structures were examined by X-ray diffraction (XRD; PANalytical X’Pert PRO). Scanning probe microscopy (SPM) including atomic force microscopy (AFM), piezoresponse force microscopy (PFM), conductive atomic force microscopy (C-AFM), and scanning Kelvin probe microscopy (SKPM) were performed using a commercial atomic force microscope (Cypher, Asylum Research), to study the


5


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 6 of 23

topography, ferroelectricity, conductance, and surface potential, respectively. For the currentvoltage (I-V) measurements with various voltage sweep rates, the Keithley 6430 SourceMeter was used.

3. RESULTS AND DISCUSSION Figure 2a shows the XRD θ-2θ scan of the BFGO/CCMO/LAO epitaxial thin film. Only the (00l) diffraction peaks from the BFGO layer and LAO substrate are observed with no detectable impurity phases. The diffraction peaks of CCMO are not observed, probably due to i) the very small thickness of the CCMO layer (~5 nm) and ii) the lattice constants of CCMO being close to those of LAO (see Figure S2 for details). It is noteworthy that the BFGO film exhibits sharp diffraction peaks and thickness fringes, indicating a good crystallinity. According to the 2θ value of the BFGO (001) peak (~19.1°), the out-of-plane lattice constant of BFGO is determined to be as large as 4.61 Å, which is a typical feature of the super-tetragonal phase.


6

Page 7 of 23

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. Topography, crystal structure and ferroelectricity. (a) XRD θ-2θ scan of the BFGO film (20 nm) grown on the CCMO-buffered LAO substrate. (b) Topographic image of wellordered Au electrodes grown on the BFGO film. Inset shows the device structure of the Au/BFGO/CCMO nanocapacitors. (c) Local PFM hysteresis loops of phase (pink) and amplitude (blue) signals. (d) PFM out-of-plane phase image obtained after poling the outer and inner regions with -5 V and +5 V, respectively.

The AFM was performed after the deposition of Au top electrodes onto the BFGO film surface. As clearly shown in Figure 2b, the BFGO film surface is quite flat, and the Au electrode arrays are well-ordered and of uniform size (~300 nm in later size and ~12 nm in thickness). The combined XRD and AFM results demonstrate that our Au/BFGO/CCMO nanocapacitors are of high structural quality. To probe the ferroelectricity, the PFM hysteresis loops were measured for the BFGO film. Figure 2c presents the butterfly-like amplitude loop and the rectangular phase loop with 180o switching, demonstrating the ferroelectricity of the BFGO film. The asymmetry of ±Vc is due to the internal bias field in the BFGO film, which may be caused by the asymmetric built-in fields at the ferroelectric/electrode interfaces30,31

and/or asymmetric distributions of trapped

charges.32,33 To gain more evidence for ferroelectricity, the PFM imaging was performed after poling the outer and inner square regions with -5 V and +5 V, respectively. As seen in Figure 2d, the ±5 V poled regions show sharp phase contrast of ~180°, indicating that the domains in the two regions are aligned in the opposite directions. In addition, it is deducible that the as-grown region has a downward polarization, because this region has the same color as the +5 V poled region. The self-polarization of the as-grown BFGO film may be associated with the electrostatic boundary conditions34,35 and/or the strain effects36-38 (see detailed discussion in Figure S3). Note that the PFM phase contrast after the poling may be caused by the electrostatic interactions between the tip and sample surface.39 To clarify this, the surface potential was monitored by the SKPM technique. Figure S3a-f show that the SKPM contrast decays gradually as the waiting time increases from 0 to 1 h, whereas the PFM phase contrast is persisted. Therefore, the PFM signals are mainly contributed from the electromechanical responses of the ferroelectric domains, further confirming the ferroelectric nature of the BFGO film.


7


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 8 of 23

Figure 3a presents the typical I-V curves of the nanocapacitors measured with the sequence of 0 V → 4 V → -4 V → 0 V using different voltage sweep rates of 0.15, 0.25, 0.3 and 0.47 V/s. In prior to the measurement, a preset pulse (-4 V, 1 s) was applied. The NDR characteristics are clearly observed with two current peaks located at +2.3 V and –1.8 V respectively. These current peaks are not from the displacement current of polarization switching, whose magnitude is estimated to be on the order of only ~10 pA at the applied voltage sweep rates. Instead, the current peaks are probably due to the charge injection and subsequent trapping.9,10 One can further distinguish the nature of the current peaks by calculating the integral of current over time (to be shown later). It is also observed from Figure 3a that the Vpeak values agree well with the coercive voltages (Vc) of the BFGO films, indicating that the current jump (i.e. charge injection) is triggered by the polarization switching. This is consistent with the PSCICT mechanism proposed above in the Introduction section. In addition, the observations of relatively symmetric I-V curves and NDR behaviors in both voltage polarities suggest that both the Au/BFGO and BFGO/CCMO interfaces form Schottky barriers and the charge injection and trapping occur at both interfaces (see Figure S4-S7 for more evidence). Figure 3b also reveals that the current peak and valley values (i.e., Ipeak and Ivalley) in the NDR region become larger as the sweep rate increases. This is because the charge trapping is a relatively slow process compared with the charge injection, and thus the overshoot of the current beyond its steady-state value is larger when the voltage is swept faster.9


8

Page 9 of 23

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 3. Characterization of the NDR behaviors. (a) Logarithmic I-V curves measured with different voltage sweep rates and PFM phase loop. (b) Current peak and valley values (Ipeak and Ivalley), and injected charge density (QI/A) as function of voltage sweep rate. The solid and hollow

symbols indicate the values obtained with 0 V → 4 V and 0 V → -4 V, respectively. I-V curves measured with (c) 0 V → +4 V → 0 V and (d) 0 V → -4 V → 0 V for three sequential cycles. Insets in Panel a, c and d show the sequences of applied voltages.

Note that without the ferroelectric BFGO layer, the Au/CCMO-only heterostructure does not show any NDR behaviors (Figure S8). Also note the measured current actually corresponds to the injected charge (QI) rather than switched polarization charge. If one integrates the measured current over time in the peak region, the injected charge density (QI/A, where A is the electrode area) much larger than the ferroelectric polarization of BFGO can be yielded (see Figure S9 for the detailed calculation methods). The QI/A values are 98, 61, 51, and 48 mC/cm2 for the positive


9


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 10 of 23

voltage sweeps with the rates of 0.15, 0.25, 0.3, and 0.47 V/s, respectively (Figure 3b). For the negative voltage sweeps with the rates of 0.15, 0.25, 0.3, and 0.47 V/s, the respective QI/A values are -77, -72, -66, and -60 mC/cm2 (Figure 3b). These QI/A values are two orders of magnitude larger than the ferroelectric polarization of BFGO (~150 µC/cm2). Hence, one can rule out the displacement current of the polarization switching as a major contributor to the measured current. As the PSCICT mechanism states, the injected charge carriers have a certain probability (k) to be trapped, and therefore, the trapped charge (QT) is proportional to the QI, as QT = kQI. Our SKPM results have suggested that the trapped charge density (QT/A) can be comparable or even larger than the ferroelectric polarization. The k value may therefore be estimated as ~0.01, consistent with those reported previously.9 Another observation from Figure 3b is that the QI (and consequently QT) increases as the sweep rate decreases, indicating that more charge carriers can be injected and subsequently trapped as the sweep rate becomes lower. To further support that the NDR in the Au/BFGO/CCMO nanocapacitors is caused by the PSCICT mechanism, the I-V characteristics were measured using the unipolar voltage sweeps. Figure 3c shows the I-V curves measured with 0 V → 4 V → 0 V for three sequential cycles. In prior to the first cycle, a preset pulse (-4 V, 1 s) was applied. The NDR behavior is observed only in the first cycle while it disappears in the following two cycles. (Note: the NDR effects refer to the regions near the current peaks at ~Vc rather than the satellite peaks. Those satellite peaks may be caused by the trap emission or tip vibration.) This observation can be well explained by the PSCICT mechanism as follows. The polarization is switched from upward to downward at ~Vc in the first cycle, and the induced electron injection and subsequent electron trapping at the BFGO/CCMO interface (an n-type Schottky contact) give rise to the NDR current peak (Figure 1). In the following two cycles, however, neither polarization switching nor electron detrapping occurs because the voltages with the same polarity are applied. The trapped electrons therefore make the Schottky barrier relatively high, suppressing the conduction in the following cycles. Similar NDR behaviors are observed in the case of unipolar negative voltage sweeps (Figure 3d), where the polarization is switched upward and the electron injection and trapping occur at the Au/BFGO interface. Note that the slight inconsistence between the Vpeak values in Figure 3a and Figure 3c,d may be associated with the device-to-device variation. Finally, the reproducibility, fatigue, and retention performances of the Au/BFGO/CCMO NDR structures were tested. The NDR behaviors were observed in more than 35 devices with


10

Page 11 of 23

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


various electrode areas, and the QI values derived from the NDR current peaks are shown statistically in Figure 4a. The QI scales in approximate linearity with the electrode area, which not only demonstrates a good reproducibility but also rules out the filament-type resistive switching as the origin of the NDR.43 Figure 4b shows the I-V curves measured with the cyclic voltage sweeps, where the NDR current peaks can still be observed after 1000 cycles. Such fatigue performance is better than that of the TiO2-based NDR devices reported recently.6 In terms of the retention measurement, the delay time between the preset pulse and the measurement pulse was varied. As shown in Figure 4c,d, the NDR behaviors can be well retained for a delay time up to 30 min, which may promise non-volatile memory applications. We believe that better fatigue and retention data can be obtained if the issue of tip drift during the measurement can be addressed.


11


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 12 of 23

Figure 4. Reproducibility, fatigue and retention performances of the Au/BFGO/CCMO NDR device. (a) Injected charge (QI) as a function of electrode area. Solid lines are linear fits to the data. (b) I-V curves measured in the cyclic test. I-V curves measured with (c) 0 V → 4 V → 0 V and (d) 0 V → -4 V → 0 V for different delay times. The delay time is defined as the interval between the preset pulse and the measurement pulse, as shown in the insets.

4. CONCLUSIONS In summary, we have proposed a mechanism for realizing the NDR based on ferroelectrics, in which the NDR is caused by the polarization switching-induced charge injection and subsequent charge trapping at the metal/ferroelectric interface. This mechanism is then verified by the experimental observation of NDR current peaks at the vicinity of Vc in the ferroelectric Au/BFGO (~20 nm)/CCMO nanocapacitors. Moreover, the NDR in these nanocapacitors is shown to be reproducible and stable with an endurance of ~1000 cycles and a retention of ~30 min. Further optimizing the ferroelectricity and the interface properties (e.g., Schottky barrier height, dead layer thickness, and trap density) may advance the NDR characteristics. This study therefore demonstrates an alternative type of NDR devices based on ferroelectrics, which may benefit the development of highly reliable NDR devices.

ASSOCIATED CONTENT Supporting Information. The following files are available free of charge on the ACS Publications website at (to be inserted by the publisher). UV-visible absorption spectrum of the BFGO film (Figure S1), XRD θ-2θ scan of the ~50nm CCMO grown on the LAO substrate (Figure S2), PFM and SKPM images taken after different waiting times in the air (Figure S3), Schematics showing the theoretical band alignment (Figure S4), I-V characteristics of the Ti/BFGO/CCMO heterostructures (Figure S5), SKPM images taken after different waiting times in the dry Ar gas (Figure S6), C-f characteristics of the Ti/BFGO/CCMO heterostructures (Figure S7), I-V curves of the Au/CCMO nanocapacitors (Figure S8), methods to calculate the QI (Figure S9). (PDF) AUTHOR INFORMATION


12

Page 13 of 23

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


Corresponding Authors *Email: [email protected] *Email: [email protected] Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT The authors would like to thank the National Key Research Program of China (Nos. 2016YFA0201002 & 2016YFA0300101), the State Key Program for Basic Researches of China (No. 2015CB921202), National Natural Science Foundation of China (Nos. 51602110, 11674108, 51272078, and 51431006), the Project for Guangdong Province Universities and Colleges Pearl River Scholar Funded Scheme (2014), the Science and Technology Planning Project of Guangdong Province (No. 2015B090927006), the Natural Science Foundation of Guangdong Province (No. 2016A030308019).

REFERENCES 1. Sollner, T. C. L. G.; Goodhue, W. D.; Tannenwald, P. E.; Parker, C. D.; Peck, D. D. Resonant Tunneling through Quantum Wells at Frequencies up to 2.5 THz. Appl. Phys. Lett. 1983, 43, 588-590. 2. 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. 3. 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.


13


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 14 of 23

4. Broekaert, T. P. E.; Brar, B.; van der Wagt, J. P. A.; Seabaugh, A. C.; Morris, F. J.; Moise, T. S.; Beam, E. A.; Frazier, G. A. A Monolithic 4-bit 2-Gsps Resonant Tunneling Analog-toDigital Converter. IEEE J. Solid-State Circuits 1998, 33, 1342-1349. 5. Chang, L. L.; Esaki, L.; Tsu, R. Resonant Tunneling in Semiconductor Double Barriers. Appl. Phys. Lett. 1974, 24, 593-595.

6. Du, Y.; Pan, H.; Wang, S.; Wu, T.; Feng, Y. P.; Pan, J.; Wee, A. T. S. Symmetrical Negative Differential Resistance Behavior of a Resistive Switching Device. ACS Nano 2012, 6, 25172523. 7. Tseng, R. J.; Ouyang, J.; Chu, C.-W.; Huang, J.; Yang, Y. Nanoparticle-induced Negative Differential Resistance and Memory Effect in Polymer Bistable Light-emitting Device. Appl. Phys. Lett. 2006, 88, 123506.

8. Jangjian, P.-C.; Liu, T.-F.; Li, M.-Y.; Tsai, M.-S.; Chang, C.-C. Room Temperature Negative Differential Resistance in DNA-based Molecular Devices. Appl. Phys. Lett. 2009, 94, 043105. 9. Kiehl, R. A.; Le, J. D.; Candra, P.; Hoye, R. C.; Hoye, T. R. Charge Storage Model for Hysteretic Negative-differential Resistance in Metal-Molecule-Metal Junctions. Appl. Phys. Lett. 2006, 88, 172102.

10. Xie, X. N.; Gao, X.; Wang, Y.; Wang, J.; Loh, K. P.; Wee, A. T. S. Negative Differential Resistance Based on Electron Injection/Extraction in Conducting Organic Films. Appl. Phys. Lett. 2009, 95, 063301.

11. Bernede, J. C.; Abachi, T. Differential Negative Resistance in Metal/Insulator/Metal Structures with an Upper Bilayer Electrode. Thin Solid Films 1985, 131, L61-L64. 12. Lin, J.; Ma, D. Origin of Negative Differential Resistance and Memory Characteristics in Organic Devices Based on Tris(8-hydroxyquinoline) Aluminum. J. Appl. Phys. 2008, 103, 124505.


14

Page 15 of 23

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


13. Tang, Q.; Moon, H. K.; Lee, Y.; Yoon, S. M.; Song, H. J.; Lim, H.; Choi, H. C. RedoxMediated Negative Differential Resistance Behavior from Metalloproteins Connected through Carbon Nanotube Nanogap Electrodes. J. Am. Chem. Soc. 2007, 129, 11018-11019. 14. Ratcliff, W.; Lynn, J. W.; Kiryukhin, V.; Jain, P.; Fitzsimmons, M. R. Magnetic Structures and Dynamics of Multiferroic Systems Obtained with Neutron Scattering. npj Quantum Mater. 2016, 1, 16003.

15. Jain, P.; Stroppa, A.; Nabok, D.; Marino, A.; Rubano, A.; Paparo, D.; Matsubara, M.; Nakotte, H.; Fiebig, M.; Picozzi, S.; Choi, E. S.; Cheetham, A. K.; Draxl, C.; Dalal, N. S.; Zapf, V. S. Switchable Electric Polarization and Ferroelectric Domains in a Metal-OrganicFramework. npj Quantum Mater. 2016, 1, 16012. 16. Lee, D.; Baek, S. H.; Kim, T. H.; Yoon, J. G.; Folkman, C. M.; Eom, C. B.; Noh, T. W. Polarity Control of Carrier Injection at Ferroelectric/Metal Interfaces for Electrically Switchable Diode and Photovoltaic Effects. Phys. Rev. B 2011, 84, 125305. 17. Tagantsev, A. K.; Gerra, G. Interface-induced Phenomena in Polarization Response of Ferroelectric Thin Films. J. Appl. Phys. 2006, 100, 051607. 18. Scott, J. F.; Melnick, B. M.; Cuchiaro, J. D.; Zuleeg, R.; Araujo, C. A.; McMillan, L. D.; Scott, M. C. Negative Differential Resistivity in Ferroelectric Thin-Film Current-Voltage Relationships. Integr. Ferroelectr. 1994, 4, 85-92. 19. Chen, H. D.; Udayakumar, K. R.; Li, K. K.; Gaskey, C. J.; Cross, L. E. Dielectric Breakdown Strength in Sol-gel Derived PZT Thick Films. Integr. Ferroelectr. 1997, 15, 89-98. 20. Dawber, M.; Scott, J. F. Negative Differential Resistivity and Positive Temperature Coefficient of Resistivity Effect in the Diffusion-limited Current of Ferroelectric Thin-film Capacitors. J. Phys.: Condens. Matter 2004, 16, L515-L521. 21. Maity, A. K.; Lee, J. Y.-m.; Sen, A.; Maiti, H. S. Negative Differential Resistance in Ferroelectric Lead Zirconate Titanate Thin Films: Influence of Interband Tunneling on Leakage Current. Jpn. J. Appl. Phys. 2004, 43, 7155-7158.


15


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 16 of 23

22. Fan, Z.; Xiao, J.; Liu, H.; Yang, P.; Ke, Q.; Ji, W.; Yao, K.; Ong, K. P.; Zeng, K.; Wang, J. Stable Ferroelectric Perovskite Structure with Giant Axial Ratio and Polarization in Epitaxial BiFe0.6Ga0.4O3 Thin Films. ACS Appl. Mater. Interfaces 2015, 7, 2648-2653. 23. Yan, J.; Gomi, M.; Yokota, T.; Song, H. Phase Transition and Huge Ferroelectric Polarization Observed in BiFe1−xGaxO3 Thin Films. Appl. Phys. Lett. 2013, 102, 222906. 24. Zeng, Z.; Greenblatt, M.; Croft, M. Charge Ordering and Magnetoresistance of Ca1−xCexMnO3. Phys. Rev. B 2001, 63, 224410. 25. Lee, H. S.; Park, H. H.; Rozenberg, M. J. Manganite-based Memristive Heterojunction with Tunable Non-linear I-V Characteristics. Nanoscale 2015, 7, 6444-6450. 26. 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. 27. Yi, H. T.; Choi, T.; Choi, S. G.; Oh, Y. S.; Cheong, S. W. Mechanism of the Switchable Photovoltaic Effect in Ferroelectric BiFeO3. Adv. Mater. 2011, 23, 3403-3407. 28. Clark, S. J.; Robertson, J. Band Gap and Schottky Barrier Heights of Multiferroic BiFeO3. Appl. Phys. Lett. 2007, 90, 132903.

29. Fan, H.; Fan, Z.; Li, P.; Zhang, F.; Tian, G.; Yao, J.; Li, Z.; Song, X.; Chen, D.; Han, B.; Zeng, M.; Wu, S.; Zhang, Z.; Qin, M.; Lu, X.; Gao, J.; Lu, Z.; Zhang, Z.; Dai, J.; Gao, X.; Liu, J.-M. Large Electroresistance and Tunable Photovoltaic Properties of Ferroelectric Nanoscale Capacitors Based on Ultrathin Super-tetragonal BiFeO3 Films. J. Mater. Chem. C 2017, 5, 3323-3329. 30. Lee, J.; Choi, C. H.; Park, B. H.; Noh, T. W.; Lee, J. K. Built-in Voltages and Asymmetric Polarization Switching in Pb(Zr,Ti)O3 Thin Film Capacitors. Appl. Phys. Lett. 1998, 72, 3380-3382.


16

Page 17 of 23

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


31. Gruverman, A.; Kholkin, A.; Kingon, A.; Tokumoto, H. Asymmetric Nanoscale Switching in Ferroelectric Thin Films by Scanning Force Microscopy. Appl. Phys. Lett. 2001, 78, 27512753. 32. Warren, W. L.; Dimos, D.; Pike, G. E.; Tuttle, B. A.; Raymond, M. V.; Ramesh, R.; Evans, J. T. Voltage Shifts and Imprint in Ferroelectric Capacitors. Appl. Phys. Lett. 1995, 67, 866868. 33. Tagantsev, A. K.; Stolichnov, I.; Setter, N.; Cross, J. S. Nature of Nonlinear Imprint in Ferroelectric Films and Long-Term Prediction of Polarization Loss in Ferroelectric Memories. J. Appl. Phys. 2004, 96, 6616-6623. 34. Chu, Y. H.; He, Q.; Yang, C. H.; Yu, P.; Martin, L. W.; Shafer, P.; Ramesh, R. Nanoscale Control of Domain Architectures in BiFeO3 Thin Films. Nano Lett. 2009, 9, 1726-1730. 35. Chu, Y. H.; Cruz, M. P.; Yang, C. H.; Martin, L. W.; Yang, P. L.; Zhang, J. X.; Lee, K.; Yu, P.; Chen, L. Q.; Ramesh, R. Domain Control in Multiferroic BiFeO3 through Substrate Vicinality. Adv. Mater. 2007, 19, 2662-2666. 36. Miao, P.; Zhao, Y.; Luo, N.; Zhao, D.; Chen, A.; Sun, Z.; Guo, M.; Zhu, M.; Zhang, H.; Li, Q. Ferroelectricity and Self-Polarization in Ultrathin Relaxor Ferroelectric Films. Sci. Rep. 2016, 6, 19965. 37. Chen, J.; Luo, Y.; Ou, X.; Yuan, G.; Wang, Y.; Yang, Y.; Yin, J.; Liu, Z. Upward Ferroelectric Self-polarization Induced by Compressive Epitaxial Strain in (001) BaTiO3 Films. J. Appl. Phys. 2013, 113, 204105. 38. Luo, Y.; Li, X.; Chang, L.; Gao, W.; Yuan, G.; Yin, J.; Liu, Z. Upward Ferroelectric Selfpoling in (001) Oriented PbZr0.2Ti0.8O3 Epitaxial Films with Compressive Strain. AIP Adv. 2013, 3, 122101. 39. Balke, N.; Maksymovych, P.; Jesse, S.; Herklotz, A.; Tselev, A.; Eom, C. B.; Kravchenko, II; Yu, P.; Kalinin, S. V. Differentiating Ferroelectric and Nonferroelectric Electromechanical Effects with Scanning Probe Microscopy. ACS Nano 2015, 9, 6484-6492.


17


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 18 of 23

40. Khan, A. I.; Chatterjee, K.; Wang, B.; Drapcho, S.; You, L.; Serrao, C.; Bakaul, S. R.; Ramesh, R.; Salahuddin, S. Negative Capacitance in a Ferroelectric Capacitor. Nat. Mater. 2015, 14, 182-186. 41. Appleby, D. J.; Ponon, N. K.; Kwa, K. S.; Zou, B.; Petrov, P. K.; Wang, T.; Alford, N. M.; O'Neill, A. Experimental Observation of Negative Capacitance in Ferroelectrics at Room Temperature. Nano Lett. 2014, 14, 3864-3868. 42. Islam Khan, A.; Bhowmik, D.; Yu, P.; Joo Kim, S.; Pan, X.; Ramesh, R.; Salahuddin, S. Experimental Evidence of Ferroelectric Negative Capacitance in Nanoscale Heterostructures. Appl. Phys. Lett. 2011, 99, 113501.

43. Yan, Z. B.; Liu, J. M. Resistance Switching Memory in Perovskite Oxides. Ann. Phys. (Amsterdam, Neth.) 2015, 358, 206-224.


18

Page 19 of 23

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


Schematic illustrations of the mechanism for NDR in ferroelectrics. (a) Typical NDR characteristics in a ferroelectric capacitor. Energy band diagrams of a metal electrode (M)/dead layer (DL)/ferroelectrics (FE) structure at different voltage regimes: (b) Regime i, (c) Regime ii, and (d) Regime iii. The three voltage regimes are indicated in Panel a. Although the exact mechanism for the electron injection is unknown, it may be realized via thermionic emission (green arrows), field emission (brown arrows) and trap-assisted tunneling (blue arrows), all of which are strongly dependent on the interface barrier. 170x128mm (300 x 300 DPI)



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

Topography, crystal structure and ferroelectricity. (a) XRD θ-2θ scan of the BFGO film (20 nm) grown on the CCMO-buffered LAO substrate. (b) Topographic image of well-ordered Au electrodes grown on the BFGO film. Inset shows the device structure of the Au/BFGO/CCMO nanocapacitors. (c) Local PFM hysteresis loops of phase (pink) and amplitude (blue) signals. (d) PFM out-of-plane phase image obtained after poling the outer and inner regions with -5 V and +5 V, respectively. 170x137mm (300 x 300 DPI)


Page 20 of 23

Page 21 of 23

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


Characterization of the NDR behaviors. (a) Logarithmic I-V curves measured with different voltage sweep rates and PFM phase loop. (b) Current peak and valley values (Ipeak and Ivalley), and injected charge density (QI/A) as function of voltage sweep rate. The solid and hollow symbols indicate the values obtained with 0 V → 4 V and 0 V → -4 V, respectively. I-V curves measured with (c) 0 V → +4 V → 0 V and (d) 0 V → -4 V → 0 V for three sequential cycles. Insets in Panel a, c and d show the sequences of applied voltages. 170x130mm (300 x 300 DPI)



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

Reproducibility, fatigue and retention performances of the Au/BFGO/CCMO NDR device. (a) Injected charge (QI) as a function of electrode area. Solid lines are linear fits to the data. (b) I-V curves measured in the cyclic test. I-V curves measured with (c) 0 V → 4 V → 0 V and (d) 0 V → -4 V → 0 V for different delay times. The delay time is defined as the interval between the preset pulse and the measurement pulse, as shown in the insets. 170x143mm (300 x 300 DPI)


Page 22 of 23

Page 23 of 23

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


82x44mm (300 x 300 DPI)


An Unusual Mechanism for Negative Differential Resistance in

Recommend Documents