Epitaxial Bi2 FeCrO6 Multiferroic Thin Film as a New Visible Light Absorbing Photocathode Material.

full papers www.MaterialsViews.com

Photocathodes

Epitaxial Bi2FeCrO6 Multiferroic Thin Film as a New Visible Light Absorbing Photocathode Material Shun Li, Bandar AlOtaibi, Wei Huang, Zetian Mi, Nick Serpone, Riad Nechache,* and Federico Rosei*

Ferroelectric materials have been studied increasingly for solar energy conversion technologies due to the efficient charge separation driven by the polarization induced internal electric field. However, their insufficient conversion efficiency is still a major challenge. Here, a photocathode material of epitaxial double perovskite Bi2FeCrO6 multiferroic thin film is reported with a suitable conduction band position and small bandgap (1.9–2.1 eV), for visible-light-driven reduction of water to hydrogen. Photoelectrochemical measurements show that the highest photocurrent density up to −1.02 mA cm−2 at a potential of −0.97 V versus reversible hydrogen electrode is obtained in p-type Bi2FeCrO6 thin film photocathode grown on SrTiO3 substrate under AM 1.5G simulated sunlight. In addition, a twofold enhancement of photocurrent density is obtained after negatively poling the Bi2FeCrO6 thin film, as a result of modulation of the band structure by suitable control of the internal electric field gradient originating from the ferroelectric polarization in the Bi2FeCrO6 films. The findings validate the use of multiferroic Bi2FeCrO6 thin films as photocathode materials, and also prove that the manipulation of internal fields through polarization in ferroelectric materials is a promising strategy for the design of improved photoelectrodes and smart devices for solar energy conversion.

1. Introduction The production of solar fuels (e.g., hydrogen from water and methane/methanol from carbon dioxide) through artifi-

S. Li, W. Huang, Dr. R. Nechache, Prof. F. Rosei Institut National de la Recherche Scientiﬁque (INRS) Centre Énergie, Matériaux et Télécommunications 1650 Boulevard Lionel-Boulet, Varennes, Québec J3X 1S2, Canada E-mail: [email protected]; [email protected] B. AlOtaibi, Prof. Z. Mi Department of Electrical and Computer Engineering McGill University 3480 University Street, Montreal, Quebec H3A 0E9, Canada Prof. N. Serpone PhotoGreen Laboratory Dipartimento di Chimica Universita di Pavia Via Taramelli 12, Pavia 27100, Italy

cial photosynthesis is a promising and sustainable approach to overcome the limited supply of fossil fuels.[1] Photoelectrochemical (PEC) cells, in which energy collection and water electrolysis are combined into a semiconductor

Dr. R. Nechache NAST Center & Department of Chemical Science and Technology University of Rome Tor Vergata Via della Ricerca Scientiﬁca 1, 00133 Rome, Italy Prof. F. Rosei Centre for Self-Assembled Chemical Structures McGill University 801 Sherbrooke Street West Montreal, Quebec H3A 2K6, Canada

DOI: 10.1002/smll.201403206

4018 www.small-journal.com

© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

small 2015, 11, No. 32, 4018–4026

www.MaterialsViews.com

Figure 1. Possible pairing of an n-type semiconductor and a p-type semiconductor toward water splitting to generate hydrogen and oxygen. Pt and RuO2 act as co-catalysts for the reduction and oxidation, respectively.

photoelectrode, permit the decomposition of water into hydrogen and oxygen by direct use of sunlight.[2] Of great importance to a PEC cell design is the potential of pairing the photoanode (n-type) and the photocathode (p-type) for oxidation and reduction reactions, respectively, as originally reported by Nozik.[3] In this sense, Gerischer pointed out that there are no advantages in using the same semiconductor materials for the n-type and p-type photoelectrodes, since it would be improbable that the same material could serve as a good photocatalyst for both the oxygen and hydrogen reactions.[4] Nozik, however, also proposed a more favorable situation in which different materials are used as photoelectrodes, provided that their band edges are located in a range of energies relative to the electrolyte such that at each photoelectrode of the respective redox reaction can be carried out by the minority carriers.[3] The advantage of such a configuration is the potential of using small bandgap semiconductor materials that would absorb a greater portion of the solar spectrum.[4] Figure 1 shows one such pairing situation of a wide bandgap n-type and a small bandgap p-type semiconductor aided by co-catalysts (Pt and RuO2) for more efficient redox reactions. Among various semiconductor photoelectrodes, metal oxides have been identified as the most promising candidates in terms of facile implementation and long-term stability. While a plethora of research has focused on developing highly efficient n-type photoanodes,[6] efforts on metal-oxide p-type semiconducting photocathodes have been disproportionately scarce with only a few notable examples: Cu2O,[7] metal-doped Fe2O3,[8] CaFe2O4,[9] and Rh-doped SrTiO3.[10] The latter are more suitable for hydrogen generation because their band bending is favorable for electron transfer to the interface between electrode and electrolyte, which has led to an emerging interest in identifying and investigating novel cost-effective p-type oxide-based photocathode materials with suitable conduction band positions, small band gaps (1.6 eV < Eg < 2.2 eV) that absorb visible light, efficient charge transfer to the solid/liquid interface, and not least are chemically stable under illumination. Much interest has resurfaced recently in the use of ferroelectric (FE) materials to achieve conversion of solar energy to electricity because of their efficient FE polarization-driven carrier separation based on the bulk photovoltaic effect small 2015, 11, No. 32, 4018–4026

(BPVE).[11] In the BPVE, charge carriers are transported by the intrinsic polarization in the FE film without the presence of a gradient in the electrochemical potential, in contrast to traditional silicon-based bipolar junction solar cells.[12] In such materials, the internal electric field originating from FE polarization offers new approaches for exploring and controlling the photovoltaic (PV) effect, which may lead to the development of next generation PV cells or other novel optoelectronic devices.[13] By analogy, appropriate control of the polarization-induced internal electric fields may also enhance the PEC performance of FE material-based photoelectrodes. The behavior of PEC cells has been examined in several systems with internal electric fields;[14] such studies have focused mainly on a few FE oxides, including BaTiO3,[15,16] Pb(Zr,Ti) O3,[17,18] and multiferroic BiFeO3 (BFO).[19] The conversion efficiencies, however, were limited by the small photocurrent density generated (of the order of nA to mA cm−2), as well as by their wide bandgaps (typically 2.7–4 eV) that absorb only about 8%–20% of the solar spectrum. Accordingly, the real potential of using semiconducting FE oxides for PEC water splitting applications has yet to be demonstrated. Among the various FE materials, double perovskite multiferroic systems possess a magnetic order parameter in addition to the ferroelectric one; the electron–electron interaction that governs magnetic ordering leads to a smaller bandgap than that of other FE perovskites,[20,21] which offers an unique opportunity to investigate the PEC effects. As a typical double perovskite material, Bi2FeCrO6 (BFCO) has been reported to exhibit multiferroic properties well above room temperature.[22,23] However, the potential use of BFCO for solar energy conversion has only begun to be explored.[21,24] Our previous work indicated that BFCO is a promising candidate for the design of high efficiency PV devices with the highest power conversion efficiency of 8.1% in multilayer films, and the photocurrent is closely related with the polarization direction as well as Fe/Cr cationic ordering in the films.[24] Hence it is highly interesting to integrate it into a PEC cell for solar water splitting, and the investigation of the effect of ferroelectricity on the PEC performance will provide fundamental insights and useful guidelines for designing efficient PEC devices based on FE oxide materials. Here we demonstrate a p-type double perovskite BFCO multiferroic thin film photoelectrode deposited by pulsed laser deposition (PLD), exhibiting a small bandgap (1.9–2.1 eV) and a suitable conduction band position, and a high photocathodic current for PEC water splitting. The role of Fe/Cr cationic ordering on the photoabsorption and PEC performance is also investigated. Most importantly, we report an effective tuning of the photocurrent by modulating the internal electric field, which stems from the FE polarization in such visible-light absorbing thin film photoelectrodes.

2. Results and Discussion To study the influence of Fe/Cr cationic ordering on the PEC properties, a series of epitaxial BFCO thin films were deposited by PLD on several (100)-oriented substrates. Specifically, we used LaAlO3 (LAO), (LaAlO3)0.3(Sr2AlTaO6)0.7 (LSAT),


www.small-journal.com

4019


Figure 2. Crystal structure of the double perovskite BFCO showing different types of Fe and Cr octahedral stacking along a) [001] and b) [111] cubic directions. O and Bi atoms are denoted by the smallest ball with red and the biggest ball with purple, respectively. Fe and Cr octahedra are colored ochre and blue, respectively. c) X-ray diffraction patterns of BFCO thin films deposited on (100)-oriented LAO, LSAT, and STO substrates buffered with CRO layer, respectively. d) Corresponding asymmetrical XRD θ–2θ scans around (111) reflection recorded for all BFCO thin films, showing the superstructure reflections. The superlattice peaks of BFCO/CRO/LSAT contain a contribution from the substrate. The squares in the XRD figures correspond to Kβ peaks, while the triangles indicate tungsten contamination from the X-ray tube cathode.

and SrTiO3 (STO), with in-plane average parameters ranging from 3.792 Å for LAO to 3.905 Å for STO, buffered with a CaRuO3 (CRO) layer as the bottom electrode (≈ 15 nm), respectively. Cross-section transmission electron microscopy (TEM) and atomic force microscopy (AFM) images are displayed in Figures S1 and S2, Supporting Information, respectively. Figure 2c displays the X-ray diffraction (XRD) patterns of the heteroepitaxial BFCO/CRO thin films deposited on different substrates. Only the 00l (l = 1, 2, 3) pseudocubic reflections of the films involved in the BFCO/CRO heterostructures and in the substrates are visible, indicating that the films are highly (001)-oriented. No peaks related to impurity phases were observed, suggesting the crystallization of the BFCO films in single phase. XRD reciprocal space maps (RSM) of BFCO thin films around the (204) reflection of various substrates are reported in Figure S3, Supporting Information. The small condensed reciprocal point in all figures suggests a high quality epitaxial growth of the BFCO thin film on different substrates. The positions of the peaks are located close to those of the substrates along the in-plane reciprocal Qx axis for each map, suggesting that the in-plane lattice parameter of the heterostructure is close to that of the substrate. This reveals highly compressive-strained epitaxial layers throughout the whole heterostructure that originate from the lattice mismatch between the film and the substrate (Table S1, Supporting Information). The strain of CRO buffered layer leads to fully coherent epitaxial growth of the heterostructure and precludes strain relaxation, especially for the


BFCO film deposited on LAO substrate, in agreement with previous work.[25] Since the B site-cationic ordering in double perovskite materials of the type A2BB′O6 (displayed in Figure S4, Supporting Information) plays a crucial role in determining the magnetic properties and tuning the optical characteristics, evidence of its existence is an important step toward understanding and controlling the functionalities of such materials.[23,26,27] As observed from the different types of stacking in BFCO crystals (Figure 2a,b), the alternation of FeO6 and CrO6 planes is achieved only in the (111) cubic direction. Hence, we performed asymmetrical θ–2θ scans around the (111) substrate reflections (Figure 2d). Large θ–2θ scans reveal two periodic satellite peaks at 19.5° and 61.2°, corresponding to the superstructure reflections of (1/21/21/2) and (3/23/23/2), respectively, in addition to the main (111) cubic reflections of substrate and BFCO. In addition, the RSM measurements (Figure S5, Supporting Information) reveal the existence of extra spots. These superstructure reflections evidence the doubling of the BFCO unit cell due to the achievement of Fe/Cr ordering in the films along the [111] direction, in agreement with the predicted rhombohedral BFCO structure.[28,29] A careful analysis of the normalized intensity of the superstructure peaks was carried out to quantify the degree of Fe/Cr cationic ordering (R), which was estimated from the normalized ratio of the superlattice peak intensity to the main (111) reflection intensity of BFCO in the pseudocubic indexing[30] (Table 1). These results contrast


small 2015, 11, No. 32, 4018–4026

www.MaterialsViews.com Table 1. Summary of structural, physical characteristics and PEC performance of BFCO thin film photoelectrodes. Device structurea)

Rb) [%]

Band gapc) [eV] Ordered domain

Disordered domain

Jphd) [mA cm−2]

BFCO(120 nm)/CRO/LAO

0.66

2.12

2.68

0.18

BFCO(130 nm)/CRO/LSAT

0.72

2.05

2.61

0.40

BFCO(128 nm)/CRO/STO

1.20

1.94

2.58

1.02

a)

The thickness of the film was determined by X-ray reflectivity measurement; b)The degree of

Fe/Cr cationic ordering (R) is represented by the ratio between the superstructure reflection (½ ½ ½) and the main (111) perovskite peak of BFCO (I½½½/I111), measured from the asymmetrical XRD θ–2θ scans of the films around the (111) reflections; c)The band gaps of BFCO thin films were calculated from photoabsorption spectra measured by ellipsometry;

d)

The

photocurrent density ( Jph) was measured as the difference of current density between dark and illumination in Na2SO4 (1 mol L−1) at pH of 6.8 under simulated one-sun illumination (AM 1.5G) at a bias of −0.97 V.

those from our earlier study.[25] The discrepancy is attributed to the different growth temperature used in the present study, a crucial parameter that determines the cationic ordering.[24] X-ray photoelectron spectroscopy (XPS) was used to identify the chemical composition and oxidation states present in different BFCO thin films (Figure S6, Supporting Information). A survey of the XPS spectra reveals the presence of Bi, Fe, Cr, and O elements with a Bi/Fe/Cr atomic ratio of ca. 2:1:1 in all films, in agreement with its stoichiometric ratio. For transition metal ions, the 2p core level splits into 2p1/2 and 2p3/2 components. The binding energy of Fe 2p3/2 is expected at 710.7 eV for Fe3+ and 709.0 eV for Fe2+. We deduce that the oxidation state of Fe in the BFCO thin films is mainly Fe3+ and there is no evidence for Fe2+ within the resolution of a few atomic percent. In the case of Cr, the expected 2p3/2 values are around 576.3 and 575.2 eV for Cr3+ and Cr4+, respectively. An analysis of different valence states of Cr suggests that the 3+ state is predominant in all films. We used spectroscopic ellipsometry to elucidate the light absorption properties of the films under study. Figure 3a displays the absorption coefficient α of different BFCO films, together with that obtained for epitaxial BFO and BiCrO3 (BCO) films for comparison. Additional optical transition peaks appear in the range between 500 and 700 nm in the absorption spectra for all BFCO films, in contrast to that of the BFO and BCO films. For all samples, α varies between 2.7 × 105 and 6.0 × 105 cm−1 at 550 nm, much higher than that of the widely studied typical p-type oxide photocathode Cu2O (Eg ≈ 2.0 eV).[31] In addition, we observed a strong dependence (a factor of ≈ 2) of the absorption area on the R ratio (Figure S7, Supporting Information), demonstrating that Fe/ Cr cationic ordering is a crucial parameter that can be used to tune the amount of absorbed light in double perovskite BFCO films. This result accords with our previous report.[24] The direct optical transitions illustrated by the (αE)2 versus E plots (Figure 3b) feature two linear portions, suggesting the presence of two threshold gaps in the BFCO films. The first threshold gap observed in the range of 2.5–2.7 eV is close to that of BFO and BCO films; it is attributed to the disordered BFCO regions (d-BFCO). As for the second optical transition region, a linear extrapolation of (αE)2 to zero small 2015, 11, No. 32, 4018–4026

versus energy yields bandgaps of ca. 2.12, 2.05, and 1.94 eV in BFCO/CRO/LAO, BFCO/CRO/LSAT, and BFCO/CRO/ STO films, respectively. These bandgaps are thought to originate from the ordered domains (o-BFCO). In agreement with electronic structure calculations confirmed by absorption spectra of La2FeCrO6 with a similar double perovskite structure,[27] the band transitions at ca. 2.0 eV are ascribed to charge transfer excitations between Cr and Fe mixed d orbital Hubbard transitions that occur in o-BFCO. The Eg of BFCO is theoretically defined by the difference between the Cr 3d–O 2p hybrids in the valence band and the empty Fe 3d conduction band.[29,32] Altering Eg requires the modification of transition metal – oxygen (TM-O) bond lengths and their interaction energies; that is, the hybridization energy and the Coulomb repulsion.[33] Considering the inverse dependence of Eg with respect to the lattice parameter, i.e., the smaller the lattice parameter the larger the bandgap, we conclude that tuning Eg involves altering the lattice parameters, as examplified by epitaxial BFO films.[34] In our case, although such a variety of substrates allows for a continuous change of the compressive misfit strain

Figure 3. Spectroscopic ellipsometry of BFCO thin films: a) absorption coefficients and b) corresponding direct optical transitions. Spectroscopic data of epitaxial BFO and BCO films are also shown for comparison.



4021


(calculated lattice parameters shown in Table S2, Supporting Information), the shift in bandgap is 0.5 eV).[34] Evidently, tuning the bandgap is insensitive to the compressive strain in BFCO films. On the other hand, tunability of the bandgap is also related to the oxidation states in perovskite structures.[35] The XPS analysis shows that 3+ states are dominant for both Fe and Cr elements (Figure S6, Supporting Information). According to previous studies, the oxygen octahedra surrounding the TM cations in the (Fe3+)d5–(Cr3+)d3 system are rigid and will be less sensitive to strain due to the homogeneous distribution of the spins in Fe and Cr degenerate d-orbitals.[36] In addition, the size of ordered domains should not affect the bandgap, since they are quite similar in all BFCO films (Table S2, Supporting Information).[24] Accordingly, the modulation of the bandgap can be accommodated mostly by limited octahedral rotations (or tilts) and will only result in a small change ( 420 nm) of BFCO thin film grown on a) CRO/LAO and b) CRO/STO substrate, respectively.

BFCO is a promising new p-type metal oxide semiconductor that can find application in photoelectrode-mediated solar fuel production. The BFCO thin film used as photocathode is represented in Figure 4a. To evaluate the impact of Fe/Cr cationic ordering on the PEC performance, we carried out J–V measurements of the BFCO thin films grown on LAO and STO substrates (Figure 5) that have distinct R values (0.66 and 1.20, respectively) as well (curves for films grown on LSAT substrates are presented in Figure 4b). The photocurrent density (Jph) values of the studied samples are listed in Table 1. A significant Jph of −1.02 mA cm−2 was obtained at −0.97 V versus RHE for BFCO/CRO/STO, while BFCO/CRO/LAO only reached a current density of −0.18 mA cm−2 at the same bias under AM1.5G simulated sunlight illumination. Since STO is also active under UV light (Eg ≈ 3.2 eV), we also performed J–V measurements under visible light (λ > 420 nm) to rule out any contributions from the STO substrate. As for BFCO/ CRO/STO, a high Jph of −0.46 mA cm−2 was maintained under visible light irradiation at −0.97 V. By contrast, under otherwise identical conditions a negligible photoresponse was obtained from BFCO/CRO/LAO. The photocurrent density of the BFCO/CRO/LSAT sample (R ratio of 0.72) small 2015, 11, No. 32, 4018–4026

lies in between. The substantial differences in photocurrent density of different BFCO thin films can be ascribed to both enhanced light absorption and smaller bandgaps in the ordered BFCO regions. To absorb a greater portion of the solar spectrum and thereby enhance the PEC performance, future studies will focus on further narrowing the bandgap of the BFCO films (the bandgap can be reduced to 1.4 eV in highly ordered BFCO films).[24] To gain an insight into the effect of FE polarization on the PEC properties, J–V measurements were carried out for the BFCO/CRO/STO sample with two different FE polarization directions. To pole the BFCO thin film, silver electrodes were deposited on top of the film. An upward poling (Pup, 1 µs, −25 V pulse) resulted in an electric field (opposite to the polarization direction) driving the positive charges accumulated on the film’s surface and negative ones to the bottom electrode (Figure 6b, right). Subsequently, the Ag electrodes were removed by etching with HNO3 solution prior to the PEC measurements. The same procedure was repeated for the downward poled (Pdown, 1 µs, +25 V pulse) BFCO film with negative charges on the surface and positive charges on the bottom electrode (Figure 6c, right). The negatively poled (Pup) BFCO film exhibited a Jph up to ≈ −2.02 mA cm−2 at a potential of ca. −1.0 V versus RHE (Figure 6b), approximately twofold greater than that of the sample without a net FE polarization (Figure 6a). By contrast, for the positively poled (Pdown) BFCO film the Jph slightly decreased to ca. −0.85 mA cm−2 (Figure 6c) at the same bias. Thus, BFCO films with a positive polar surface (Pup) are advantageous for enhancing their PEC activity for water decomposition, whereas films with a negative polar surface (Pdown) are unfavorable. These results indicate that for FE semiconductors possessing a defined direction of the polarization vector, the behavior of the photogenerated carriers is strongly influenced by the polarization field, consistent with previous results on other FE perovskites.[15,17] Additional direct evidence of spontaneous FE polarization switching under an electric field was examined locally using piezoresponse force microscopy (PFM) on the BFCO/CRO/STO sample. The observed hysteresis loop confirms the FE character of the BFCO film (Figure S8, Supporting Information). On the basis of our results, a tentative schematic illustration of the energy band diagram of BFCO thin film based photocathode is depicted in Figure 7. The drawing on the left shows the ideal diagram for CRO/p-type BFCO/electrolyte junctions in the absence of a net FE polarization. After poling the FE BFCO thin film, the surface polarization charge causes band bending within the space charge region (SCR) associated with an electric field gradient. As the thickness of the films (ca. 120–130 nm) is of the same order of magnitude as the space charge layer (≈ 50–100 nm, detailed calculations are provided in the Supporting Information), the influence of band bending at the interface can no longer be neglected. For the negatively poled BFCO film, the energy band diagram is modified as illustrated in the middle of Figure 7. Under illumination, the photogenerated e−/h+ pairs will be separated by the driving force arising from an internal electric field, and the electrons in BFCO thin films will then be driven toward the film’s surface/electrolyte interface, while the holes will migrate



4023


gap (1.9–2.1 eV) and suitable conduction band position for the visible-light-driven reduction of water to hydrogen. We found two optical band transitions in the BFCO thin films resulting from the ordered and disordered domains, and the amount of absorbed light was closely related to the degree of Fe/Cr cationic ordering. PEC measurements showed that the highest photocurrent (up to −1.02 mA cm−2 at a potential of −0.97 V versus RHE) was obtained in the p-type BFCO thin film photocathode grown on CRO/STO substrate. In addition, the behavior of the photogenerated carriers was strongly influenced by the polarization field, such that the photocurrent density was enhanced by a factor of ≈ 2 after negatively poling the BFCO thin film, as a result of the modulation of the band structure by the appropriate control of the internal electric field gradient. These results indicate the potential of using multiferroic materials such as Figure 6. Variations of the current density with applied voltage (vs Ag/AgCl) in 1 mol L−1 Na2SO4 at pH 6.8 under chopped simulated sunlight illumination (AM1.5G) of BFCO/CRO/ BFCO with narrow bandgap in the field of STO sample: a) before, b) after negative (Pup, −25V), and c) positive poling (Pdown, +25V). solar water decomposition. Our study also provides a promising approach to improve Schematic illustrations are shown on the right of each figure. the PEC performance of FE materials to the bottom electrode. Consequently, the recombination of based PEC cells through the manipulation of polarization photogenerated electrons and holes is effectively hindered, induced internal fields. Further efforts are aimed at investithereby leading to enhanced PEC activity. On the other hand, gating the overall PEC efficiency, H2 production and photowhen the film is positively poled, the oppositely slanted band electrode stability under visible light illumination. causes electrons to drift into the bulk while the holes accumulate at the electrolyte (Figure 7, right). In this case, it is unfavorable for electron flow and reduces the PEC efficiency.

4. Experimental Section

3. Conclusions In conclusion, we established a novel photocathode material based on a BFCO epitaxial thin film with a narrow band

BFCO thin films were deposited epitaxially on (100)-oriented LAO, LSAT, and STO single crystalline substrates by PLD at 650 °C with a laser repetition rate of 8 Hz. The oxygen partial pressure was kept around 10 mTorr. For comparison, parent BFO and BCO thin films were deposited on STO substrates under the same conditions.

Figure 7. Simplified energy band diagrams of a PEC cell based on a p-type BFCO thin film in an electrolyte without polarization (left) and either negatively poled (middle) or positively poled (right).



small 2015, 11, No. 32, 4018–4026

www.MaterialsViews.com

The CRO thin layer was used as a bottom electrode, to promote epitaxial growth, and to apply the voltage needed to perform PEC measurements. The CRO film with thickness ≈ 15 nm was deposited by PLD at 650 ºC under an oxygen pressure of 20 mTorr. After deposition, the samples were cooled to room temperature under the same oxygen pressure used for deposition. XRD was used to analyze the crystal quality and film orientation using a high resolution Panalytical X’pert Pro diffractometer. Asymmetrical θ–2θ measurements around the (111) reflections were performed to qualitatively estimate the cationic ordering of Fe and Cr cations in the BFCO films. The sample was tilted by 54° (angle between [111] and [001] cubic directions) with respect to the surface normal. RSM measurements were carried out around the (204) reflections to identify the crystal structure of the films. The elemental composition and chemical state of the constituents of the BFCO films were examined by XPS using an ESCA Escalab 220i XL spectrometer with a monochromated Al Kα X-ray source (1486.6 eV). The optical absorption properties of BFCO films were characterized using spectroscopic ellipsometry performed at an incidence angle of 60° using a VASE ellipsometer (J.A. Woollam Company). A multilayer model consisting of air, thin film, bottom electrode, and substrate was applied to extract information on optical properties from the ellipsometric spectra. The absorption coefficient α of the films was deduced from the extinction coefficient k (α = 4πk/λ, where λ is the wavelength of the incident light). The bandgap was calculated using Tauc’s equation. The data were compared with those obtained for the BFO and BCO films grown under identical conditions. The absence of the characteristic shape of the (αE)1/2 versus E plot and the presence of clear linear slopes in (αE)2 versus E curves indicate a direct bandgap of the films. PFM measurements were performed using a Veeco Enviroscope AFM equipped with Pt/Ir coated ANSCM-PA probes from App Nano. Cross-section TEM measurements were carried out under angular dark field (G.L Botton). All the measurements were performed at room temperature. Photocurrent measurements were performed in a three-electrode configuration consisting of a Ag/AgCl reference electrode, a Pt counter electrode, and a BFCO thin film working electrode. A Wavenow electrochemical station (Pine Instrument Inc.) was used throughout this study. The CRO bottom electrode was connected with a Cu wire using silver paste. The entire sample, except the thin film surface, was then covered by insulating epoxy to eliminate leakage current. All PEC measurements were conducted in an electrolyte consisting of 1 mol L−1 Na2SO4 (pH = 6.8). The photoresponse was measured from a 300 W Xe light source equipped with an AM1.5G filter. A 420 nm long-pass filter was used to measure the photoresponse in the visible range. The measured photocurrent was normalized to the sample’s macroscopic area to obtain the photocurrent density (in mA cm−2) for comparison. All the current versus potential measurements were carried out at a 20 mV s−1 sweep rate. To pole the BFCO thin film, silver electrodes were deposited on top of the film. After electric poling, the sample was subsequently dipped in HNO3 solution (8 mol L−1) to remove the Ag electrodes prior to the PEC measurement. The same procedure was repeated for the sample with the opposite polarization direction. Mott–Schottky measurements were carried out using an electrochemical impedance method with AC amplitude of 10 mV at an applied frequency of 1 and 2 kHz, respectively.

small 2015, 11, No. 32, 4018–4026

Supporting Information Supporting Information is available from the Wiley Online Library or from the author.

Acknowledgements The authors acknowledge infrastructure support from the Canada Foundation for Innovation. F.R. is grateful to the Canada Research Chairs Program for partial salary support, to NSERC for a discovery grant and for an E.W.R. Steacie Memorial Fellowship, and to FQRNT for team grants. This study was also partly supported by an international collaboration grant (MDEIE) with the European Network WIROX. F.R. further acknowledges the Alexander von Humboldt Foundation for a F.W. Bessel Award. R.N. is grateful to NSERC for a personal postdoctoral fellowship for partial salary support. S.L. thanks FQRNT and CSC for graduate scholarships. F.R. is also grateful to Elsevier for a grant from Applied Surface Science.

[1] a) X. Chen, S. Shen, L. Guo, S. S. Mao, Chem. Rev. 2010, 110, 6503; b) F. E. Osterloh, Chem. Soc. Rev. 2013, 42, 2294. [2] a) A. Fujishima, Nature 1972, 238, 37; b) D. Scaife, Sol. Energy 1980, 25, 41; b) M. Grätzel, Nature 2001, 414, 338; c) M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. Mi, E. A. Santori, N. S. Lewis, Chem. Rev. 2010, 110, 6446; d) S. Sunkara, V. K. Vendra, J. B. Jasinski, T. Deutsch, A. N. Andriotis, K. Rajan, M. Menon, M. Sunkara, Adv. Mater. 2014, 26, 2878. [3] A. J. Nozik, Appl. Phys. Lett. 1976, 29, 150. [4] H. Gerischer, in Solar Power and Fuels (Ed: J. R. Bolton), Academic Press, New York 1977, Ch.4, p.77. [5] a) M. Woodhouse, B. A. Parkinson, Chem. Soc. Rev. 2009, 38, 197; b) N. Serpone, A. V. Emeline, J. Phys. Chem. Lett. 2012, 3, 673. [6] a) M. S. Wrighton, A. B. Ellis, P. T. Wolczanski, D. L. Morse, H. B. Abrahamson, D. S. Ginley, J. Am. Chem. Soc. 1976, 98, 2774; b) C. Santato, M. Ulmann, J. Augustynski, J. Phys. Chem. B 2001, 105, 936; c) A. Duret, M. Grätzel, J. Phys. Chem. B 2005, 109, 17184; d) S. P. Berglund, D. W. Flaherty, N. T. Hahn, A. J. Bard, C. B. Mullins, J. Phys. Chem. C 2011, 115, 3794. [7] a) M. Hara, T. Kondo, M. Komoda, S. Ikeda, J. N. Kondo, K. Domen, K. Shinohara, A. Tanaka, Chem. Commun. 1998, 3, 357; b) A. Paracchino, V. Laporte, K. Sivula, M. Grätzel, E. Thimsen, Nat. Mater. 2011, 10, 456. [8] W. B. Ingler Jr., J. P. Baltrus, S. U. M. Khan, J. Am. Chem. Soc. 2004, 126, 10238. [9] S. Ida, K. Yamada, T. Matsunaga, H. Hagiwara, Y. Matsumoto, T. Ishihara, J. Am. Chem. Soc. 2010, 132, 17343. [10] K. Iwashina, A. Kudo, J. Am. Chem. Soc. 2011, 133, 13272. [11] a) A. M. Glass, D. Von der Linde, D. Auston, T. J. Negran, J. Electron. Mater. 1975, 4, 915; b) V. Fridkin, Crystallogr. Rep. 2001, 46, 654. [12] a) A. M. Glass, D. Von der Linde, T. J. Negran, Appl. Phys. Lett. 1974, 25, 233; b) P. Brody, F. Crowne, J. Electron. Mater. 1975, 4, 955; c) M. Qin, K. Yao, Y. C. Liang, Appl. Phys. Lett. 2008, 93, 122904; d) H. Yi, T. Choi, S. Choi, Y. Oh, S. W. Cheong, Adv. Mater. 2011, 23, 3403; e) Y. Yuan, T. J. Reece, P. Sharma, S. Poddar, S. Ducharme, A. Gruverman, Y. Yang, J. Huang, Nat. Mater. 2011, 10, 296.



4025

full papers www.MaterialsViews.com [13] a) S. Y. Yang, J. Seidel, S. J. Byrnes, P. Shafer, C. H. Yang, M. D. Rossell, P. Yu, Y. H. Chu, J. F. Scott, J. W. Ager, L. W. Martin, R. Ramesh, Nat. Nanotechnol. 2010, 5, 143; b) Y. Yuan, Z. Xiao, B. Yang, J. Huang, J. Mater. Chem. A 2014, 2, 6027. [14] a) D. Tiwari, S. Dunn, J. Mater. Sci. 2009, 44, 5063; b) L. Li, P. A. Salvador, G. S. Rohrer, Nanoscale 2014, 6, 24. [15] J. L. Giocondi, G. S. Rohrer, J. Phys. Chem. B 2001, 105, 8275. [16] Y. Cui, J. Briscoe, S. Dunn, Chem.Mater. 2013, 25, 4215. [17] a) Y. Inoue, K. Sato, K. Sato, H. Miyama, J. Phys. Chem. 1986, 90, 2809; b) J. Chen, H. Lu, H.-J. Liu, Y.-H. Chu, S. Dunn, K. Ostrikov, A. Gruverman, N. Valanoor, Appl. Phys. Lett. 2013, 102, 182904. [18] a) Y. Inoue, Energy Environ. Sci. 2009, 2, 364; b) C. Wang, D. Cao, F. Zheng, W. Dong, L. Fang, X. Su, M. Shen, Chem. Commun. 2013, 49, 3769. [19] a) X. Chen, T. Yu, F. Gao, H. Zhang, L. Liu, Y. Wang, Z. Li, Z. Zou, J.-M. Liu, Appl. Phys. Lett. 2007, 91, 022114; b) U. A. Joshi, J. S. Jang, P. H. Borse, J. S. Lee, Appl. Phys. Lett. 2008, 92, 242106; c) W. Ji, K. Yao, Y.-F. Lim, Y. C. Liang, A. Suwardi, Appl. Phys. Lett. 2013, 103, 062901; d) S. J. A. Moniz, R. QuesadaCabrera, C. S. Blackman, J. Tang, P. Southern, P. M. Weaver, C. J. Carmalt, J. Mater. Chem. A 2014, 2, 2922. [20] a) R. Vidya, P. Ravindran, A. Kjekshus, H. Fjellvåg, Phys. Rev. B 2004, 70, 184414; b) R. F. Berger, J. B. Neaton, Phys. Rev. B 2012, 86, 165211. [21] R. Nechache, C. Harnagea, S. Licoccia, E. Traversa, A. Ruediger, A. Pignolet, F. Rosei, Appl. Phys. Lett. 2011, 98, 202902. [22] a) R. Nechache, C. Harnagea, A. Pignolet, F. Normandin, T. Veres, L.-P. Carignan, D. Ménard, Appl. Phys. Lett. 2006, 89, 102902; b) R. Nechache, C. V. Cojocaru, C. Harnagea, C. Nauenheim, M. Nicklaus, A. Ruediger, F. Rosei, A. Pignolet, Adv. Mater. 2011, 23, 1724. [23] R. Nechache, F. Rosei, J. Solid State Chem. 2012, 189, 13. [24] R. Nechache, C. Harnagea, S. Li, L. Cardenas, W. Huang, J. Chakrabartty, F. Rosei, Nat. Photon. 2015, 9, 61. [25] R. Nechache, C. Harnagea, A. Ruediger, F. Rosei, A. Pignolet, Funct. Mater. Lett. 2010, 3, 83.


[26] L. Balcells, J. Navarro, M. Bibes, A. Roig, J. Fontcuberta, Appl. Phys. Lett. 2001, 78, 781. [27] J. Andreasson, J. Holmlund, S. G. Singer, C. S. Knee, R. Rauer, B. Schulz, M. Käll, M. Rübhausen, S.-G. Eriksson, L. Börjesson, Phys. Rev. B 2009, 80, 075103. [28] P. Baettig, N. A. Spaldin, Appl. Phys. Lett. 2005, 86, 012505. [29] P. Baettig, C. Ederer, N. A. Spaldin, Phys. Rev. B 2005, 72, 214105. [30] For BFCO films grown on LSAT substrate, an asymmetrical θ–2θ scan was also performed on the bare substrate because LSAT itself possesses a superstructure along the [111] direction as well; the R ratio was calculated after subtracting the contribution from the substrate. [31] C. Malerba, F. Biccari, C. Leonor Azanza Ricardo, M. D’Incau, P. Scardi, A. Mittiga, Sol. Ener. Mat. Sol. C. 2011, 95, 2848. [32] Z.-W. Song, B.-G. Liu, Chinese Phys. B 2013, 22, 047506. [33] a) R. Zimmermann, P. Steiner, R. Claessen, F. Reinert, S. Hüfner, P. Blaha, P. Dufek, J. Phys.: Condens. Matter 1999, 11, 1657; b) T. Arima, Y. Tokura, J. B. Torrance, Phys. Rev. B 1993, 48, 17006. [34] a) H. L. Liu, M. K. Lin, Y. R. Cai, C. K. Tung, Y. H. Chu, Appl. Phys. Lett. 2013, 103, 181907; b) H. Dong, Z. Wu, S. Wang, W. Duan, J. Li, Appl. Phys. Lett. 2013, 102, 072905. [35] J. L. Fierro, Metal Oxides: Chemistry and Applications, CRC Press, Boca Raton, USA, 2005. [36] a) A. Vailionis, H. Boschker, W. Siemons, E. Houwman, D. Blank, G. Rijnders, G. Koster, Phys. Rev. B 2011, 83, 064101; b) A. Ohtomo, S. Chakraverty, H. Mashiko, T. Oshima, M. Kawasaki, J. Mater. Res. 2013, 28, 689. [37] a) K. Matsuzaki, K. Nomura, H. Yanagi, T. Kamiya, M. Hirano, H. Hosono, Appl. Phys. Lett. 2008, 93, 202107; b) L. Liao, B. Yan, Y. F. Hao, G. Z. Xing, J. P. Liu, B. C. Zhao, Z. X. Shen, T. Wu, L. Wang, J. T. L. Thong, C. M. Li, W. Huang, T. Yu, Appl. Phys. Lett. 2009, 94, 113106.


Received: October 29, 2014 Revised: April 8, 2015 Published online: May 19, 2015

small 2015, 11, No. 32, 4018–4026

Thin film multiferroic nanocomposites by ion implantation.

Visible Light-Driven Atom Transfer Radical Addition to Olefins using Bi2 O3 as Photocatalyst.

Epitaxial thin films of pyrochlore iridate Bi2+xIr2-yO7-δ: structure, defects and transport properties.

New visible light absorbing materials for solar fuels, Ga(Sbx)N1-x.

Development of a Thin Film Magnetic Moment Reference Material.

Selective visible-light-driven CO2 reduction on a p-type dye-sensitized NiO photocathode.

Epitaxial Bi5Ti3FeO15-CoFe2O4 pillar-matrix multiferroic nanostructures.

GeH: a novel material as a visible-light driven photocatalyst for hydrogen evolution.

Multi-Material Front Contact for 19% Thin Film Solar Cells.

Nanosecond light induced, thermally tunable transient dual absorption bands in a-Ge₅As₃₀Se₆₅ thin film.

Nanocolumnar structured porous Cu-Sn thin film as anode material for lithium-ion batteries.

Switching of both local ferroelectric and magnetic domains in multiferroic Bi0.9La0.1FeO3 thin film by mechanical force.

PLZT Multiferroic Thin Film Heterostructures.

Correlation of High Magnetoelectric Coupling with Oxygen Vacancy Superstructure in Epitaxial Multiferroic BaTiO₃-BiFeO₃ Composite Thin Films.

Water-Soluble Epitaxial NaCl Thin Film for Fabrication of Flexible Devices.

Optical properties of epitaxial BiFeO3 thin film grown on SrRuO3-buffered SrTiO3 substrate.

Butvar B-98 as a thin support film.

Localized photoelectrochemistry on a tungsten oxide-iron oxide thin film material library.

Ultra-Thin Pyrocarbon Films as a Versatile Coating Material.

Atomic layer deposited molybdenum nitride thin film: a promising anode material for Li ion batteries.

Holographic image generation with a thin-film resonance caused by chalcogenide phase-change material.

Hexaarylbiimidazoles as visible light thiol-ene photoinitiators.

Nanoscale electrical properties of epitaxial Cu3Ge film.

Photoemission performance of thin graded structure AlGaN photocathode.