Controllable Photovoltaic E ff ect of Microarray Derived from Epitaxial Tetragonal BiFeO

Controllable Photovoltaic E ff ect of Microarray Derived from Epitaxial Tetragonal BiFeO

₃

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

*^S Supporting Information

ABSTRACT: Recently, the 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 a controllable photovoltaic process in ferroelectric- based materials. In this work, we prepared well-ordered microarrays based on epitaxially tetragonal BiFeO₃ (T-BFO) films by the pulsed laser deposition technique. The polarization- dependent photocurrent image was directly observed by a

conductive atomic force microscope under ultraviolet illumination. By choosing a suitable buffer electrode layer and controlling the ferroelectric polarization in the T-BFO layer, we realized the manipulation of the photovoltaic process. Moreover, based on the analysis of the 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 depolarizationfield induced by the ferroelectric polarization in T-BFO. This work is very meaningful for deeply understanding the photovoltaic process of BiFeO₃-based devices at the microscale and provides us a feasible avenue for developing data storage or logic switching microdevices based on the FePV effect.

KEYWORDS: ferroelectric photovoltaic effect, tetragonal BiFeO₃film, microarray, conductive atomic force microscope, heterostructure, depolarizationfield

■

INTRODUCTION

BiFeO₃(BFO) is a free-lead multiferroic material in which the ferroelectric (P), antiferromagnetic (M), and ferroelastic (ε) properties coexist at room temperature.¹⁻³Due to the coupling effect between P, M, and ε, BFO shows great potential for developing multifunctional devices such as magneto−elec- tric,⁴⁻⁶ magneto-elastic,⁷ and even magneto−electro−elastic energy conversion devices.² In recent years, the photovoltaic effect has also been observed in the BFO.⁸⁻¹²Compared 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 band gap (∼2.8 eV) which is helpful for inducing more light absorption and larger photocurrent for FePV effect.^1,7,16 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 photogenerated 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 the short-circuit current density (J_SC) and open-circuit voltage (V_OC) depend on the direction of ferroelectric polarization.^19,20 Also, some groups have found that the electrodes and defects such as oxygen vacancies (V_Os) have an influence on the FePV effect in BFO.²¹⁻²³In addition, it has been proved that there are three possible types of domains (71°, 109°, and 180°) in the rhombohedral BFOfilm (R-BFO).^3,24,25Many groups have found the domain-depend- ent PV effect in the R-BFOfilms.²⁶⁻²⁹They observe a linearly increased V_OC with increasing 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 the microscale

Received: May 9, 2017 Accepted: July 26, 2017 Published: July 26, 2017

www.acsami.org

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 the microscale are still a big technical challenge.

Different from R-BFO film, the tetragonal-phase BFO (T- BFO) film only has a single-domain structure, which can exclude the influence of other types 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 others,^2,30,31 which can induce larger depolarization fields to separate the photogenerated carriers. In this work, based on epitaxial T-BFOfilms, we realize the direct observation of the FePV effect at the microscale using a combined system integrating a conductive atomic force microscope (CAFM) and ultraviolet light lamp. Moreover, the photovoltaic process is well manipulated in such T-BFO-based microarray by the combined changing of buffer electrode layer and ferroelectric polarization direction. We reveal that the manipulation of the photovoltaic process in the present microarray 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 microarray 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 the 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 microarray, a 10 nm thick La_0.67Sr_0.33MnO₃ (LSMO) layer was first deposited on the

single-crystal (001)-LaAlO₃wafer; then a 40 nm thick BFOfilm 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 BFOfilm is down shifted to 2θ≈ 38.4°in comparison with the LAO peak 2θ ≈ 48°. This reveals compressively strained growth of the BFOfilm on LAO with an out-of-plane lattice parameter of 4.67 Å. The correspondingc/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) measure- ments confirm that the sample has a good ferroelectric property. From Figure 1b, one observes that the sample exhibits well-defined PFM hysteresis loops with coercive voltages of−3.5 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 the buffer electrode layer and the BFO layer can be controlled so small, which is beneficial for the separation of photogenerated carriers. We patterned the quadrate array of 3 nm thick Ti or Cr buffer layer on the T-BFOfilm and then deposited a 5 nm thick Au layer to prevent Ti or Cr buffer layer from oxidizing.

In the microarray, each cell was designed as a square with various side lengths (l= 2.5, 5, 10, and 20μm) and gap widths (d= 1.5, 5, 5, and 5μm), as shown inFigure 1c. Finally, two series of devices with different buffer electrode layer, i.e., Au/

Ti/T-BFO/LSMO and Au/Cr/T-BFO/LSMO, were obtained.

The UV−vis spectroscopy shown inFigure S1bshows 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.

Figure 1.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 microarray derived from Au/Ti (or Cr)/T-BFO/LSMO. Side length of each square electrode isl= 2.5, 5, 10, and 20μm, and the gap between squares isd= 1.5, 5, 5, and 5μm. (d) Setup for measuring the local photovoltaic effect.

ACS Applied Materials & Interfaces

For studying the FePV effect at the microscale, we built a combined measurement system integrating the scan probe microscope (SPM) with an ultraviolet lamp and source meter, as sketched inFigure 1d. The positive voltage was applied to the tip, and the bottom electrodes of the samples were grounded. Wefirst measured the photovoltaic response of the samples with different electrode sizes. The results showed that the photovoltaic parameters (J_SC and V_OC) decreased with increasing electrode size (see detailed results in Figure S2).

This indicates that the sample exhibits a 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 microarray with the cell parameters ofl= 2.5μm andd= 1.5μm for the following investigations. The measurements were performed in a 3 ×3 array with an area of 12×12μm².Figure 2presents the local ferroelectric switching behavior and photocurrent image of two microcell arrays under light illumination. The wavelength of incident light is 365 nm, and its density is 200 mW/cm². For the Au/Ti/T-BFO/LSMO microarray, 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 (P_up) domains pointing to the top electrode, while the bright regions are in downward-polarized (P_down) state.

Figure 2b shows the photocurrent image of the Au/Ti/T- BFO/LSMO microarray, 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 microarray exhibits evident photovoltaic response, producing negative V_OC and positive J_SCwhether it is in P_upor P_down state. Nevertheless, the cell in P_downstate (e.g., cell A) produces smaller|V_OC|and|J_SC|values than in the P_upstate (e.g., cell B). The above results indicate that the polarities of bothJ_SCandV_OCare hard to be reversed

by changing the ferroelectric polarization direction in the T- BFO layer when the Ti buffer electrode layer is used.

The photovoltaic behaviors become different when the buffer electrode layer is replaced by Cr.Figure 2d presents the 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 V_OC and J_SCof the Au/Cr/T- BFO/LSMO change with the ferroelectric polarization reversal of the T-BFO layer. As shown inFigure 2d and2f, the cell in the P_downstate (e.g., cell A′) produces positiveV_OC= +0.09 V and negativeJ_SC=−1.05 mA/cm², while the cell in the P_upstate (e.g., cell B′) shows negativeV_OC=−0.15 V and positiveJ_SC= +0.75 mA/cm². These indicate that with a proper buffer electrode the FePV effect can be reversed under the assistance of the polarization switching. It is worth mentioning that in this work we first observe the photocurrent image in a T-BFO- based microcell array by using the CAFM technique. It is a visualized and reliable evidence for the FePV effect, which is significant for concisely 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 inFigure 3. In order to determine the energy band alignment of T-BFO, we carried out ultraviolet photoelectron spectroscopy (UPS) measurement (the results is shown in Figure S4). The interface band structure and ferroelectric polarization in T-BFO are crucial for the separation of photogenerated carriers. Under illumination, the photogenerated carriers are excited from the T-BFO layer of the heterostructure and then separated under the action of net built-in electricfield (E_bi) in the whole heterostructure. The E_bican be described as

= ±

E_bi E_in E_{dp (1)}

Figure 2.Local ferroelectric switching and photovoltaic characteristics for a 3×3 microarray: (a−c) Au/Ti/T-BFO/LSMO and (d−f) Au/Cr/T- BFO/LSMO. Side length of each square electrode isl= 2.5μm, and gap between squares isd= 1.5 μm. (a and d) Ferroelectric phase images recorded by PFM. Circle-fork A (A′) represents polarized-down (Pdown) state, and circle-dot B (B′) represents polarized-up (Pup) state. (b and e) Photocurrent images scanned by CAFM. Light intensity is 200 mW/cm². (c and f) 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.

ACS Applied Materials & Interfaces

where E_dp represents the depolarization field caused by the ferroelectric polarization in T-BFO and E_in is the internal electricfield induced by the interfacial interaction between the electrode layer and the T-BFO layer. TheE_inis irreversible and can be deduced as follows

= −

E_in (W_top W_bottom)/qd ₍₂₎

where W_top and W_bottom are the work functions of the top electrode (Ti or Cr) and bottom electrode (LSMO),

respectively,q is the positive elementary charge, and dis the thickness of the T-BFOfilm. Taking typical parametersW_Ti= 4.33 eV,W_Cr= 4.50 eV,W_LSMO= 4.96 eV,^33,34andq= 1.6× 10⁻¹⁹C,d= 40 nm intoeq 2, we obtain the electricfieldE_in≈ 1.6×10⁷V/m for the Au/Ti/T-BFO/LSMO andE_in≈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 theE_inpoints downward.

Now we need to distinguish the competitive contribution of E_in and E_dp to the separation of photogenerated carriers. We first analyze the situation that the T-BFO layer is in P_upstate, as shown in Figure 3a and 3c. For both Au/Ti/T-BFO/LSMO and Au/Cr/T-BFO/LSMO, theE_dpin the T-BFO layer has the same direction as theE_in, pointing downward. Thus, the built-in field of the whole heterostructure is E_bi = E_in + E_dp, consequently inducing a positive J_SC. When the T-BFO layer is in the P_down state, however, the E_dp in the T-BFO layer is antiparallel to theE_in, so the net built-infield isE_bi=E_in−E_dp. If E_in > E_dp, the E_bi has the same direction as the E_in, i.e., pointing downward, so the J_SC is recorded positively. This exactly corresponds to the Au/Ti/T-BFO/LSMO (Figure 3b).

On the contrary, ifE_in<E_dp, theE_bipoints upward, it causes a negative J_SC, exactly corresponding to the Au/Cr/T-BFO/

LSMO (Figure 3d). According to the above analysis, we suggest that the manipulation of the photovoltaic process in the present T-BFO-based array is dominated by the combined effect of two factors, i.e., theE_dporiginated from the ferroelectric polarization in T-BFO and the E_in caused by the asymmetric metal/

ferroelectric/metal (MFM) structure. By choosing a suitable buffer electrode layer and controlling the polarization direction of the ferroelectric T-BFO layer, these two factors can be well tuned and then cause the expected changes in V_OC and J_SC. Similarly, if the ferroelectric polarization in the film changes (e.g., during preparation the T-BFO is replaced by the R-BFO), Figure 3. Schematic diagram of energy levels and photogenerated

carriers transfer process for Au/Ti/T-BFO/LSMO in Pupstate (a) and Pdownstate (b) and Au/Cr/T-BFO/LSMO in Pupstate (c) and Pdown

state (d).

Figure 4.Time-dependent (a)V_OCand (b)J_SCplots with light-on and -offresponses under different polarization state for a cell in the Au/Cr/T- BFO/LSMO microarray. Light intensity is 200 mW/cm². P_downand P_upstates are obtained by applying the pulse voltage of +6.0 and−6.0 V/0.1 s, respectively. (c) Data storage and (d) logic switching microdevice prototype designed based on the Au/Cr/T-BFO/LSMO microarray. Yellow bars represents electrodes (LSMO and Cr); red and blue columns represents the P_downstate and P_upstate, respectively, of the T-BFOfilms.

ACS Applied Materials & Interfaces

the combined effect of E_in and E_dp will have a change, also causing a different PV performance.

Such controllable photovoltaic parameters can exhibit rapid and reliable response to both incident light and polarization reversal in the T-BFO layer. Here, we take an example of Au/

Cr/T-BFO/LSMO microarray.Figure 4a and 4b presents the time (t)-dependent V_OC and J_SCplots with light-on and -off responses under different polarization state. One can observe that the V_OC shows switchable photoresponse depending on the polarization direction (Figure 4a). When the light is turned on, theV_OCrapidly rises to−0.15 and +0.08 V from zero for the P_upand P_down states, respectively. Also, such response can be well repeated. In the meantime, a similar response can also be achieved for the J_SC, as shown in Figure 4b. The rapid photovoltaic response to incident light and close dependence on the polarization state provides us a feasible way to develop a unique technology for data storage or logical switching. Herein, we propose a prototype based on the Au/Cr/T-BFO/LSMO microarray to demonstrate our idea, as shown inFigure 4c. In this prototype, “0” state and “1” state correspond to the P_up state and P_downstate in the T-BFO layer. The writing of“0”or

“1” state in each cell can be easily realized by changing the polarization direction of the T-BFO layer. During reading the data, a light is irradiated to the whole microarray; meanwhile, the induced V_OC or J_SCis directly detected for each cell. As mentioned before, a different polarization state produces different V_OC or J_SC; the data state in each cell is thus read out. This data storage mode can be called“electrical writing and optical reading”, which consumes less power than conventional modes. We can also develop a logical switching device using the present tunable FePV effect of Au/Cr/T-BFO/LSMO, as shown inFigure 4d, for a given cell (e.g., cell B) with the P_down state, its logic state will be alternately switched to the on or off state when the incident laser is on or offand then be read out by detecting the corresponding V_OC (J_SC). In addition, these kinds of electronic devices have much room for improvement.

For example, optimizing the MFM structure could get more reliable readoutV_OCwhich can be larger and/or switchable and is more beneficial for the data storage and data readout. As suggested in mich literature,^9,10,19,20 building symmetrical structures can obtain the switchable V_OC and/or using oxide electrodes with longer screening lengths can enhance the depolarization field and then increase the V_OC. Finally, considering that the performance of the above microdevices is related to the response speed, device density, and so on, further relative investigation needs to be carried out in the future.

■

^CONCLUSION

With the assistance of a combined system integrating SPM, UV light lamp, and source meter, we demonstrated a controllable photovoltaic process in the epitaxial T-BFO-based hetero- structures at the microscale. The polarization-dependent photovoltaic effect was directly observed by means of mapping photocurrent in a microarray. Detailed analysis reveals that the competition between two factors, i.e., the depolarizationfield originated from the ferroelectric polarization in T-BFO and the internal electricfield caused by the asymmetric MFM structure, plays a key role in determining the modulation of the photovoltaic process. The rapid and controllable photovoltaic process provides us a possible way to design desirable optoelectronic microdevices such as ferroelectric−optical memorizer and logical switching units.

■

EXPERIMENTAL METHODS

Sample Preparation. T-BFO/La0.67Sr0.33MnO3(T-BFO/LSMO) bilayers were epitaxially grown on the single-crystal LaAlO₃ (LAO) substrates by pulsed laser deposition (PLD) using a KrF (λ= 248 nm) excimer laser with a laser energy density of 1.1 J/cm²and a repetition rate of 8 Hz under an oxygen pressure of 15 Pa. A 10 nm thick LSMO layer and a 40 nm thick T-BFO layer were in sequence deposited on the LAO substrate at 680°C. Then the bilayers were annealed for 10 min and cooled to room temperature with a cooling rate of 5°C/min under an oxygen pressure of one-half atmospheric pressure. After that a 3 nm thick Ti (and Cr) layer and a 5 nm thick Au layer with various sizes were in sequence deposited on the T-BFO/LSMO bilayer using the markless photolithography (SF 100, Intelligent Micro Pattering.

LLC) and electron beam evaporation (EBM,PVD 75, Kurt J. Lesker) technique. Finally, the Au/Ti(Cr)/T-BFO/LSMO heterostructures were obtained.

Structural Characterization. The information on the phase purity, crystal structure, and epitaxial quality were examined by X-ray diffraction (XRD) scan (X’Pert PRO, Pan-Analyzer). Ultraviolet photoelectron spectroscopy (UPS, UV40A-PS, PREVAC) was used to measure the Fermi energy and the valence band edge of thefilms.

Electrical Characterization. The piezoelectric and conduction properties were characterized using scanning probe microscopy (SPM, Asylum Research Cypher Inc.) with a Pt/Ir-coated tip. Current density (J) vs voltage (V) curves were measured using a Keithley 6430A source connected to the SPM. During the measurements, the diamond-coated tip was used. The positive voltage was applied to the tip, and the bottom electrodes of the samples were grounded. A UV light source with a wavelength of 365 nm and light intensity of up to 200 mW/cm²was used. The angle between the sample surface and incident light wasfixed at 45°.

■

ASSOCIATED CONTENT

*^S Supporting Information

The Supporting Information is available free of charge on the ACS Publications websiteat DOI:10.1021/acsami.7b06535.

Transmittance of the top electrodes; as-grown PV effect of the Au/Cr/T-BFO/LSMO heterostructure; dark current mapping matching the written phases of two heterostrctures with zero scanning bias under dark circumstance; energy band aligment measured by UPS (PDF)

■

AUTHOR INFORMATION Corresponding Authors

*E-mail: wanjg@nju.edu.cn.

*E-mail: xingsengao@scnu.edu.cn.

ORCID

Peilian Li:0000-0003-0902-9689

Jian-guo Wan:0000-0002-9673-576X

Zhen Fan:0000-0002-1756-641X

Xingsen Gao:0000-0002-2725-0785

Jun-ming Liu:0000-0001-8988-8429 Author Contributions

Z.L. conducted the data acquisition and drafted the manuscript.

P.L., G.T., and Z.H. participated in sample fabrication. P.L.

carried out the PFM and CAFM measurement. J.-g.W., D.P., Z.F., and X.G. contributed to data interpretation. J.-g.W., X.G., and J.-m.L. contributed to manuscript writing.

Notes

The authors declare no competingfinancial interest.

ACS Applied Materials & Interfaces

■

ACKNOWLEDGMENTS

This work was supported by the National Key Research Programme of China (Grant Nos. 2016YFA02010004 and 2016YFA02010002), the National Natural Science Foundation of China (Grant No. 51472113), and the National Key Projects for Basic Research of China (Grant No. 2015CB921203).

■

(1) Alexe, M.; Hesse, D. Tip-enhanced Photovoltaic Effects in^REFERENCES Bismuth Ferrite.Nat. Commun.2011,2, 256.

(2) Sando, D.; Barthelé my, A.; Bibes, M. BiFeÓ ₃Epitaxial Thin Films and Devices: Past, Present and Future.J. Phys.: Condens. Matter2014, 26, 473201.

(3) Baek, S. H.; Jang, H. W.; Folkman, C. M.; Li, Y. L.; Winchester, B.; Zhang, J. X.; He, Q.; Chu, Y. H.; Nelson, C. T.; Rzchowski, M. S.;

Pan, X. Q.; Ramesh, R.; Chen, L. Q.; Eom, C. B. Ferroelastic Switching for Nanoscale Non-volatile Magnetoelectric Devices.Nat. Mater.2010, 9, 309−314.

(4) Lebeugle, D.; Colson, D.; Forget, A.; Viret, M.; Bataille, A. M.;

Gukasov, A. Electric-field-induced Spin Flop in BiFeO₃Single Crystals at Room Temperature.Phys. Rev. Lett.2008,100, 227602.

(5) Lee, S.; Ratcliff, W.; Cheong, S. W.; Kiryukhin, V. Electric Field Control of the Magnetic State in BiFeO₃Single Crystals.Appl. Phys.

Lett.2008,92, 192906.

(6) Zhao, T.; Scholl, A.; Zavaliche, F.; Lee, K.; Barry, M.; Doran, A.;

Cruz, M. P.; Chu, Y. H.; Ederer, C.; Spaldin, N. A.; Das, R. R.; Kim, D.

M.; Baek, S. H.; Eom, C. B.; Ramesh, R. Electrical Control of Antiferromagnetic Domains in Multiferroic BiFeO₃ Films at Room Temperature.Nat. Mater.2006,5, 823−829.

(7) Sando, D.; Agbelele, A.; Rahmedov, D.; Liu, J.; Rovillain, P.;

Toulouse, C.; Infante, I. C.; Pyatakov, A. P.; Fusil, S.; Jacquet, E.;

Carreté ro, C.; Deranlot, C.; Lisenkov, S.; Wang, D.; Le Breton, J.-M.;́ Cazayous, M.; Sacuto, A.; Juraszek, J.; Zvezdin, A. K.; Bellaiche, L.;

Dkhil, B.; Barthelé my, A.; Bibes, M. Crafting the Magnonic and́ Spintronic Response of BiFeO₃Films by Epitaxial Strain.Nat. Mater.

2013,12, 641−646.

(8) Choi, T.; Lee, S.; Choi, Y. J.; Kiryukhin, V.; Cheong, S.-W.

Switchable Ferroelectric Diode and Photovoltaic Effect in BiFeO₃. Science2009,324, 63−66.

(9) Guo, R.; You, L.; Zhou, Y.; Lim, Z. S.; Zou, X.; Chen, L.; Ramesh, R.; Wang, J. Non-volatile Memory Based on the Ferroelectric Photovoltaic Effect.Nat. Commun.2013,4, 1990.

(10) Hu, W. J.; Wang, Z. H.; Yu, W. L.; Wu, T. Optically Controlled Electroresistance and Electrically Controlled Photovoltage in Ferro- electric Tunnel Junctions.Nat. Commun.2016,7, 10808.

(11) Zhang, W.; Yang, M.-M.; Liang, X.; Zheng, H.-W.; Wang, Y.;

Gao, W.-X.; Yuan, G.-L.; Zhang, W.-F.; Li, X.-G.; Luo, H.-S.; Zheng, R.-K. Piezostrain-enhanced Photovoltaic Effects in BiFeO₃/ La_0.7Sr_0.3MnO₃/PMN−PT Heterostructures. Nano Energy 2015,18, 315−324.

(12) 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.

(13) Schröder, M.; Haußmann, A.; Thiessen, A.; Soergel, E.; Woike, T.; Eng, L. M. Conducting Domain Walls in Lithium Niobate Single Crystals.Adv. Funct. Mater.2012,22, 3936−3944.

(14) Glass, A. M.; von der Linde, D.; Negran, T. J. High-voltage Bulk Photovoltaic Effect and the Photorefractive Process in LiNbO₃.Appl.

Phys. Lett.1974,25, 233.

(15) Chakrabartty, J.; Nechache, R.; Harnagea, C.; Li, S.; Rosei, F.

Enhanced Photovoltaic Properties in Bilayer BiFeO₃/Bi-Mn-O Thin Films.Nanotechnology2016,27, 215402.

(16) Himcinschi, C.; Bhatnagar, A.; Talkenberger, A.; Barchuk, M.;

Zahn, D. R. T.; Rafaja, D.; Kortus, J.; Alexe, M. Optical Properties of Epitaxial BiFeO₃Thin Films Grown on LaAlO₃.Appl. Phys. Lett.2015, 106, 012908.

(17) Bhatnagar, A.; Kim, Y. H.; Hesse, D.; Alexe, M. Persistent Photoconductivity in Strained Epitaxial BiFeO₃Thin Films.Nano Lett.

2014,14, 5224−5228.

(18) Katiyar, R. K.; Sharma, Y.; Misra, P.; Puli, V. S.; Sahoo, S.;

Kumar, A.; Scott, J. F.; Morell, G.; Weiner, B. R.; Katiyar, R. S. Studies of the Switchable Photovoltaic Effect in Co-substituted BiFeO₃Thin Films.Appl. Phys. Lett.2014,105, 172904.

(19) Lopez-Varo, P.; Bertoluzzi, L.; Bisquert, J.; Alexe, M.; Coll, M.;

Huang, J.; Jimenez-Tejada, J. A.; Kirchartz, T.; Nechache, R.; Rosei, F.;

Yuan, Y. Physical Aspects of Ferroelectric Semiconductors for Photovoltaic Solar Energy Conversion.Phys. Rep.2016,653, 1−40.

(20) Yuan, Y.; Xiao, Z.; Yang, B.; Huang, J. Arising Applications of Ferroelectric Materials in Photovoltaic Devices. J. Mater. Chem. A 2014,2, 6027−6041.

(21) Gao, R. L.; Yang, H. W.; Chen, Y. S.; Sun, J. R.; Zhao, Y. G.;

Shen, B. G. Oxygen Vacancies Induced Switchable and Nonswitchable Photovoltaic Effects in Ag/Bi_0.9La_0.1FeO₃/La_0.7Sr_0.3MnO₃Sandwiched Capacitors.Appl. Phys. Lett.2014,104, 031906.

(22) 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: Condens. Matter Mater. Phys.2011, 84, 125305.

(23) Ge, C.; Jin, K.-J.; Zhang, Q.-H.; Du, J.-Y.; Gu, L.; Guo, H.-Z.;

Yang, J.-T.; Gu, J.-X.; He, M.; Xing, J.; Wang, C.; Lu, H.-B.; Yang, G.- Z. Toward Switchable Photovoltaic Effect via Tailoring Mobile Oxygen Vacancies in Perovskite Oxide Films. ACS Appl. Mater.

Interfaces2016,8, 34590−34597.

(24) Baek, S.-H.; Folkman, C. M.; Park, J.-W.; Lee, S.; Bark, C.-W.;

Tybell, T.; Eom, C.-B. The Nature of Polarization Fatigue in BiFeO₃. Adv. Mater.2011,23, 1621−1625.

(25) Balke, N.; Choudhury, S.; Jesse, S.; Huijben, M.; Chu, Y. H.;

Baddorf, A. P.; Chen, L. Q.; Ramesh, R.; Kalinin, S. V. Deterministic Control of Ferroelastic Switching in Multiferroic Materials. Nat.

Nanotechnol.2009,4, 868−875.

(26) Yang, S. Y.; Seidel, J.; Byrnes, S. J.; Shafer, P.; Yang, C.-H.;

Rossell, M. D.; Yu, P.; Chu, Y.-H.; Scott, J. F.; Ager, J. W., III; Martin, L. W.; Ramesh, R. Above-bandgap Voltages from Ferroelectric Photovoltaic Devices.Nat. Nanotechnol.2010,5, 143−147.

(27) Guo, R.; You, L.; Chen, L.; Wu, D.; Wang, J. Photovoltaic Property of BiFeO₃Thin Films with 109°Domains.Appl. Phys. Lett.

2011,99, 122902.

(28) Seidel, J.; Fu, D.; Yang, S.-Y.; Alarcón-Lladó, E.; Wu, J.; Ramesh, R.; Ager, J. W., III Efficient Photovoltaic Current Generation at Ferroelectric Domain Walls.Phys. Rev. Lett.2011,107, 126805.

(29) Bhatnagar, A.; Chaudhuri, A. R.; Kim, Y. H.; Hesse, D.; Alexe, M. Role of Domain Walls in the Abnormal Photovoltaic Effect in BiFeO₃.Nat. Commun.2013,4, 3835.

(30) Ricinschi, D.; Yun, K. Y.; Okuyama, M. A Mechanism for the 150μC cm⁻²Polarization of BiFeO₃Films Based on First-principles Calculations and New Structural Data.J. Phys.: Condens. Matter2006, 18, L97−L105.

(31) Ederer, C.; Spaldin, N. A. Effect of Epitaxial Strain on the Spontaneous Polarization of Thin Film Ferroelectrics.Phys. Rev. Lett.

2005,95, 257601.

(32) Singh, A. V.; Althammer, M.; Rott, K.; Reiss, G.; Gupta, A.

Junction Size Dependence of Ferroelectric Properties in E-beam Patterned BaTiO₃ Ferroelectric Tunnel Junctions. Appl. Phys. Lett.

2015,107, 122903.

(33) Michaelson, H. B. The Work Function of the Elements and its Periodicity.J. Appl. Phys.1977,48, 4729.

(34) Miao, Q.; Zeng, M.; Zhang, Z.; Lu, X.; Dai, J.; Gao, X.; Liu, J.-M.

Self-assembled Nanoscale Capacitor Cells Based on Ultrathin BiFeO₃ Films.Appl. Phys. Lett.2014,104, 182903.

ACS Applied Materials & Interfaces