Controllable Photovoltaic Effect of Microarray Derived from Epitaxial

Jul 26, 2017 - Recently, the ferroelectric photovoltaic (FePV) effect has attracted great interest due to its potential in developing optoelectronic d...
0 downloads 3 Views 1MB Size
Subscriber access provided by UNIV OF NEWCASTLE

Article

Controllable Photovoltaic Effect of Micro-array Derived from Epitaxial Tetragonal BiFeO3 Films Zengxing Lu, Peilian Li, Jianguo Wan, Zhifeng Huang, Guo Tian, Danfeng Pan, Zhen Fan, Xingsen Gao, and Jun-Ming Liu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b06535 • Publication Date (Web): 26 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 18

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

Controllable Photovoltaic Effect of Micro-array Derived from Epitaxial Tetragonal BiFeO3 Films

Zengxing Lu,† Peilian Li,§Jian-guo Wan,*,† Zhifeng Huang,§Guo Tian,§Danfeng Pan,† Zhen Fan,§ Xingsen Gao,*,§ and Jun-ming Liu †



Laboratory of Solid State Microstructures and Innovation Center of Advanced Microstructures,

Nanjing University, Nanjing 210093, China §

Institute for Advanced Materials and Laboratory of Quantum Engineering and Quantum

Materials, South China Normal University, Guangzhou 510006, China

*

Address correspondent to E-mail: [email protected]

*

Address correspondent to E-mail: [email protected] 1

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

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

Page 2 of 18

ABSTRACT: Recently, ferroelectric photovoltaic (FePV) effect has attracted great interest due to its potential in developing optoelectronic devices such as solar cell and electric-optical sensors. It is important for actual applications to realize controllable photovoltaic process in ferroelectricbased materials. In this work, we have prepared well-ordered micro-arrays based on epitaxially tetragonal BiFeO3 (T-BFO) films by pulsed laser deposition technique. Polarization-dependent photocurrent image was directly observed by conductive atomic force microscope under ultraviolet illumination. By choosing suitable buffer electrode layer and controlling the ferroelectric polarization in T-BFO layer, we realized the manipulation of the photovoltaic process. Moreover, based on the analysis of band structure, we revealed the mechanism of manipulating the photovoltaic process, and attributed it to the competition between two key factors, i.e., the internal electric field caused by energy band alignments at interfaces, and the depolarization field induced by the ferroelectric polarization in T-BFO. This work is very meaningful for deeply understanding the photovoltaic process of BiFeO3–based devices at microscale, and provides us a feasible avenue for developing data storage or logic switching micro-devices based on the FePV effect.

Keywords: ferroelectric photovoltaic effect, tetragonal BiFeO 3 film, micro-array, conductive atomic force microscope, heterostructure, depolarization field

2

ACS Paragon Plus Environment

Page 3 of 18

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

INTRODUCTION BiFeO3 (BFO) is a free-lead multiferroic material, in which the ferroelectric (P), antiferromagnetic (M) and ferroelastic (ɛ) properties co-exist at room temperature1-3. Due to the coupling effect between P, M and ɛ, BFO shows great potentials for developing multifunctional devices such as magneto-electric4-6, magneto-elastic7 and even magneto-electro-elastic energy conversion devices2. In recent years, photovoltaic effect has also been observed in the BFO8-12. Compared with conventional ferroelectric photovoltaic (FePV) materials (e.g., BaTiO3, Pb(Zr,Ti)O3 and LiNbO3)13-18, BFO is highly promising for actual applications because of its narrower optical bandgap (~2.8 eV) which is helpful for inducing more light absorption and larger photocurrent for FePV effect1, 7, 16. Moreover, the coexistence of multiferroic and photovoltaic properties in BFO and the cross-coupling between them endow the BFO with a new connotation of physics and new degrees of freedom for multifunctional applications. In the FePV process, the photo-generated electron-hole pairs are separated by the depolarization field induced by the ferroelectric polarization, so the photocurrent and photovoltage can change with the variation of ferroelectric polarization strength, which further influences the photoelectric conversion efficiency. Previous investigations have demonstrated that the polarity and/or magnitude of short circuit current density (JSC) and open circuit voltage (VOC) depend on the direction of ferroelectric polarization19, 20. And some groups have found that the electrodes and defects such as oxygen vacancies (V Os) have an influence on the FePV effect in BFO21-23. In addition, it has been proved that there are three possible types of domains (71°, 109°and 180°) in the rhombohedral BFO film (R-BFO)3, 24, 25. Many groups have found the domain-dependent PV effect in the R-BFO films26-29. They observe a linearly increased VOC with increasing the number of the domain walls, which are mainly related to the 71°- and 109°-domains. From the point of actual applications, it is important to control these factors so as to manipulate the FePV process simply. Unfortunately, the complexity of microstructures in BFO-based materials makes it difficult. To realize this purpose, it may be a feasible way to study the FePV process at microscale since various influences, such as defects and grain boundaries, can be avoided to a great degree. Nevertheless, so far few relative investigations have been carried out because direct observation and manipulation of the FePV process at microscale are still a big technical challenge. 3

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

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

Page 4 of 18

Different from R-BFO film, the tetragonal-phase BFO (T-BFO) film only has single domain structure, which can exclude the influence of other type of domains to a great degree, thus a clear FePV process can be expected and its manipulation can also become facile. Moreover, the T-BFO has larger remanent polarization (~150 μC/cm2) than the others2, 30, 31, which can induce larger depolarization fields to separate the photo-generated carriers. In this work, based on epitaxial tetragonal-phase BFO (T-BFO) films, we realize the direct observation of the FePV effect at microscale using a combined system integrating conductive atomic force microscope (CAFM) and ultraviolet light lamp. Moreover, the photovoltaic process is well manipulated in such T-BFObased micro-array by the combined changing of buffer electrode layer and ferroelectric polarization direction. We reveal that the manipulation of photovoltaic process in the present micro-array is dominated by the competition between two factors, i.e., the depolarization field induced by the ferroelectric polarization in T-BFO, and the internal electric field caused by the band alignments at interfaces. Accordingly, a prototype device based on the present T-BFO-based micro-array is proposed for electric-optical data storage and logic switching, and its feasibility is also demonstrated. This work is meaningful for understanding the FePV process in BFO-based materials at microscale, and provides us a feasible avenue for developing high-performance electric-optical integrating devices based on the FePV effect.

RESULTS AND DISCUSSION For preparing T-BFO-based micro-array, a 10-nm-thick La0.67Sr0.33MnO3 (LSMO) layer was first deposited on the single-crystal (001)-LaAlO3 wafer, then a 40 nm thick BFO film was epitaxially grown on it by pulsed laser deposition (see details in the Experimental Methods). The X-ray diffraction (XRD) patterns shown in Figure 1a exhibit the (001) and (002) reflections of the BFO film and LAO substrate. The (002) peak of the BFO film is down shifted to 2 θ~38.4° in comparison with the LAO peak 2θ~48°. This reveals compressively strained growth of the BFO film on LAO with an out-of-plane lattice parameter of 4.67 Å. The corresponding c/a ratio is approximately 1.23, which is very close to the calculated value ~1.27 in Ref. 31, indicating that the film is tetragonal phase. The piezoresponse force microscopy (PFM) measurements confirm that the sample has a good ferroelectric property. From Figure 1b, one observes that the 4

ACS Paragon Plus Environment

Page 5 of 18

sample exhibits well-defined PFM hysteresis loops with coercive voltages of -3.5 V and +5.0 V. To study the influence of the interfacial band structure on the FePV effect, we chose two kinds of metal materials, i.e., Ti and Cr, for preparing the top buffer electrode layer. Ti and Cr are chosen because their work functions are close to that of BFO. This indicates that the barrier height between buffer electrode layer and BFO layer can be controlled so small, which is beneficial for the separation of photo-generated carriers. We patterned the quadrate array of 3nm-thick Ti or Cr buffer layer on the T-BFO film, then deposited a 5-nm-thick Au layer to prevent Ti or Cr buffer layer from oxidizing. In the micro-array, each cell was designed as a square with various side lengths (l = 2.5, 5, 10 and 20 μm) and gap widths (d = 1.5, 5, 5 and 5 μm), as shown in Figure 1c. Finally, two series of devices with different buffer electrode layer,

20

30

40

100 150 50

100

50

Phase ()

150

(b)200 Amplitude (pm)

LAO (002)

T-BFO (002)

LAO (001)

(a)

T-BFO (001)

i.e., Au/Ti/T-BFO/LSMO and Au/Cr/T-BFO/LSMO, were obtained. The UV-Vis spectroscopy

Intensity (a.u.)

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

0

50 -10

-5

0

5

10

Voltage (V)

2 Theta ()

(d)

(c)

~ d

o

45 Cr(Ti)

Au T-BFO

l

LSMO LAO

Figure 1. (color online) X-ray diffraction pattern (a), and PFM hysteresis loops (b) of the T-BFO/LSMO grown on the (001)-LAO single-crystal wafer. (c) Optical image of the micro-array derived from Au/Ti (or Cr)/TBFO/LSMO. The side length of each square electrode is l = 2.5, 5, 10, 20 μm, and the gap between squares is d = 1.5, 5, 5 and 5 μm. (d) The setup for measuring the local photovoltaic effect.

5

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

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

Page 6 of 18

shown in Figure S1 shows that the transmittance of Au (5 nm)/Ti (3 nm) and Au (5 nm)/Cr (3 nm) are high enough for the subsequent FePV measurements. For studying the FePV effect at microscale, we built a combined measurement system integrating the scan probe microscope (SPM) with ultraviolet lamp and source meter, as sketched in Figure 1d. The positive voltage was applied to the tip and the bottom electrodes of the samples were grounded. We first measured the photovoltaic response of the samples with different electrode sizes. The results showed that the photovoltaic parameters ( JSC and VOC) decreased with increasing the electrode size (see detailed results in Figure S2). This indicates that the sample exhibits better photovoltaic effect under smaller electrode size. A possible reason lies in fewer grain boundaries in epitaxial T-BFO and less leakage current pathways induced by the defects under smaller electrodes 32. Accordingly, we made the micro-array with the cell parameters of l = 2.5 μm and d = 1.5 μm for the following investigations. The measurements were performed in a 3×3 array with an area of 12×12 μm 2. Figure 2 presents local ferroelectric switching behavior and photocurrent image of two micro-cell arrays under the light illumination. The wavelength of incident light is 365 nm and its density is 200 mW/cm2. For the Au/Ti/T-BFO/LSMO micro-array, it can be well electrically poled upon the application of ±8 V DC bias, as shown in Figure 2a. The ~180°contrast of phase clearly reveals that the polarization direction of the T-BFO domain can be completely reversed by applying an appropriate bias. Note that we define that the dark regions poled by -8 V correspond to the upward-polarized (Pup) domains pointing to the top electrode, while the bright regions are in downward-polarized (Pdown) state. Figure 2b shows the photocurrent image of the Au/Ti/T-BFO/LSMO micro-array, and Figure 2c plots the current density (J) vs. voltage (V) curves for two typical cells A and B measured under light illumination and dark circumstance. It is clear that each cell in the micro -array exhibits evident photovoltaic response, producing negative VOC and positive JSC whether it is in Pup or Pdown state. Nevertheless, the cell in P down state (e.g. cell A) produces smaller |VOC| and |JSC| values than in Pup state (e.g. cell B). The above results indicate that the polarities of both JSC and VOC are hard to be reversed by changing the ferroelectric polarization direction in the T-BFO layer when the Ti buffer electrode layer is used.

6

ACS Paragon Plus Environment

Page 7 of 18

The photovoltaic behaviors become different when the buffer electrode layer is replaced by Cr. Figure 2d presents local PFM image of the Au/Cr/T-BFO/LSMO measured in a 3×3 array. The alternative bright and dark patterns indicate that the ferroelectric polarization reversal in each cell can also be controlled well. However, different from the Au/Ti/T-BFO/LSMO, the polarities of both VOC and JSC of the Au/Cr/T-BFO/LSMO change with the ferroelectric polarization reversal of the T-BFO layer. As shown in Figure 2d and 2f, the cell in P down state (e.g., cell A’) produces positive VOC = +0.09 V and negative JSC = -1.05 mA/cm2, while the cell in Pup state (e.g., cell B’) shows negative VOC = -0.15 V and positive JSC = +0.75 mA/cm2. These indicate that with a proper buffer electrode, the FePV effect can be reversed under the assist of the polarization switching. It is worth mentioning that in this work, we firstly observe the photocurrent image in a T-BFO-based micro-cell array by using the CAFM technique. It is a visualized and reliable evidence for the FePV effect, which is significant for concisely

-50 220 (e)

Deg

B'

J (mA/cm2)

pA

-15 -30 -45

-1.0 1.0 (f)

0.30

A-dark 0.15 A-light B-dark 0.00 B-light -0.3 -0.2 -0.1 0.0 -1.0 -0.5 0.0 0.5 1.0

Voltage (V) 50

Au/Cr/T-BFO/LSMO

0 0.9

-50

-100 -150

3 μm

Au/Ti/T-BFO/LSMO

0

2

A'

Current

(d)

(c) 15

J (mA/cm )

3 μm

2.5

pA

B

Deg

Phase

A

Current

220 (b)

(a)

Phase

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

-50

-2.5

A'-dark 0.0 A'-light -0.9 B'-dark B'-light -1.8 -0.15 0.00 -1.0 -0.5 0.0 0.5

0.15

1.0

Voltage (V)

Figure 2. (color online) Local ferroelectric switching and photovoltaic characteristics for a 3×3 micro-array: (a)~(c) Au/Ti/T-BFO/LSMO, and (d)~(f) Au/Cr/T-BFO/LSMO. The side length of each square electrode is l = 2.5 μm and the gap between squares is d = 1.5 μm. (a) and (d) are the ferroelectric phase images recorded by PFM. The circle-fork A (A’) represents polarized-down (Pdown) state, and circle-dot B (B’) represents polarizedup (Pup) state. (b) and (e) are the photocurrent images scanned by CAFM. The light intensity is 200 mW/cm 2. (c) and (f) are the current density (J) vs. voltage (V) curves, measured under light illumination and dark circumstance, for the typical cells with different polarization states extracted from (a) and (d), respectively.

7

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

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

Page 8 of 18

clarifying the relationship between the FePV effect and the polarization, and deeply understanding the photovoltaic process in BFO-based materials. We then explore the mechanism of manipulating the photovoltaic process for the present samples. We drew simplified schematic illustrations of band structures for both kinds of the heterostructures, as shown in Figure 3. In order to determine the energy band alignment of TBFO, we carried out ultraviolet photoelectron spectroscopy (UPS) measurement (the results i s shown in Figure S4). The interface band structure and ferroelectric polarization in T-BFO are crucial for the separation of photo-generated carriers. Under illumination, the photo-generated carriers are excited from the T-BFO layer of the heterostructure, and then separated under the action of net built-in electric field (Ebi) in the whole heterostructure. The Ebi can be described as: Ebi = Ein ± Edp

(1)

where Edp represents the depolarization field caused by the ferroelectric polarization in T-BFO, Ein is the internal electric field induced by the interfacial interaction between the electrode layer and T-BFO layer. The Ein is irreversible, and can be deduced as follows: Ein = (Wtop - Wbottom)/qd

(2)

where Wtop and Wbottom are the work functions of top electrode (Ti or Cr) and bottom electrode (LSMO), respectively, q is the positive elementary charge, and d is the thickness of the T-BFO film. Taking typical parameters WTi = 4.33 eV, WCr = 4.50 eV, WLSMO = 4.96 eV

33,34

and q =

1.6×10-19 C, d = 40 nm into Eq.(2), we obtain the electric field Ein ≈ 1.6×107 V/m for the Au/Ti/T-BFO/LSMO and Ein ≈ 1.2×107 V/m for the Au/Cr/T-BFO/LSMO. For both kinds of devices, the potential on the top interface is higher than that of the bottom, so the Ein points downwards.

8

ACS Paragon Plus Environment

Page 9 of 18

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

Now we need to distinguish the competitive contribution of Ein and Edp to the separation of photo-generated carriers. We first analyze the situation that the T-BFO layer is in P up state, as shown in Figure 3a and 3c. For both Au/Ti/T-BFO/LSMO and Au/Cr/T-BFO/LSMO, the Edp in T-BFO layer has the same direction as the Ein, pointing downwards. So the built-in field of the whole heterostructure is Ebi = Ein + Edp, consequently inducing a positive JSC. When the T-BFO layer is in Pdown state, however, the Edp in T-BFO layer is antiparallel to the Ein, so the net builtin field is Ebi = Ein - Edp. If Ein > Edp, the Ebi has the same direction as the Ein, i.e. pointing downwards, so the JSC is recorded positively. This exactly corresponds to the Au/Ti/TBFO/LSMO (Figure 3b). On the contrary, if Ein < Edp, the Ebi points upwards, then causes a negative JSC, exactly corresponding to the Au/Cr/T-BFO/LSMO (Figure 3d). According to the above analysis, we suggest that the manipulation of photovoltaic process in the present T-BFObased array is dominated by the combined effect of two factors, i.e., the Edp originated from the ferroelectric

polarization

in

T-BFO

and

the

Ein

caused

by

the

asymmetric

metal/ferroelectric/metal (MFM) structure. By choosing suitable buffer electrode layer and Au/Ti/T-BFO/LSMO: |Ein|>|Edp| (b)

Ebi

(a)

Ebi

h -

-

P

P

+ Ti

T-BFO

+ LSMO

Ti

T-BFO

LSMO

Au/Cr/T-BFO/LSMO: |Ein|