PCCP View Article Online

Published on 05 January 2016. Downloaded by Gazi Universitesi on 23/01/2016 05:00:55.

PAPER

Cite this: DOI: 10.1039/c5cp07479g

View Journal

Solution growth of Ta-doped hematite nanorods for efficient photoelectrochemical water splitting: a tradeoff between electronic structure and nanostructure evolution† Yanming Fu,a Chung-Li Dong,b Zhaohui Zhou,a Wan-Yi Lee,c Jie Chen,a Penghui Guo,a Liang Zhaoa and Shaohua Shen*a Ta-doped hematite (a-Fe2O3) nanorod array films were successfully prepared on fluorine-doped tin dioxide (FTO) coated glass substrates via a facile solution growth process with TaCl5 as a Ta doping precursor. Under 1 sun illumination and at an applied potential of 1.0 V vs. Ag/AgCl, the Ta-doped a-Fe2O3 photoanode with optimized dopant concentration showed a photocurrent density as high as 0.53 mA cm 2, which was about 3.5 times higher than that of the undoped sample. As demonstrated by Mott–Schottky and X-ray absorption spectroscopy measurements, considerable increase in photoelectrochemical (PEC) performance achieved for Ta-doped a-Fe2O3 nanorod films should be mainly attributed to the increased electron donor density induced by Ta doping. However, with superfluous Ta doping, the [110]-oriented nanorod structure was destroyed, which caused greatly restrained photoinduced holes transferring to the surface and retarded surface water

Received 4th December 2015, Accepted 4th January 2016

oxidation reaction, leading to decreased PEC water splitting activity. This study clearly demonstrated that

DOI: 10.1039/c5cp07479g

great necessity to balance the trade-off between the electronic structure and nanostructure evolution by

doping could be effective to enhance the PEC activity of a-Fe2O3 nanorods as photoanodes, while it is of optimizing the dopant concentration, for increased donor density and meanwhile with the nanorod

www.rsc.org/pccp

nanostructure well preserved for directed charge transfer.

Introduction Solar hydrogen conversion using a photoelectrochemical (PEC) water splitting system has been considered as one of the strategic ways to solve the global energy problem.1 As the most important component in a PEC system, semiconductor photoelectrodes which absorb solar light and on which water splitting reaction happens have attracted numerous attention. To date, different kinds of semiconductors, such as metal oxides,2–4 (oxy) nitrides5 and sulfides,6 have been developed as photoelectrodes for solar water splitting. Considered as a very promising photoanode a

International Research Center for Renewable Energy, State Key Laboratory of Multiphase Flow in Power Engineering, Xi’an Jiaotong University, Shaanxi 710049, China. E-mail: [email protected] b Department of Physics, Tamkang University, Tamsui 25137, Taiwan c National Synchrotron Radiation Research Center, 101 Hsin-Ann Road, Hsinchu Science Park, Hsinchu 30076, Taiwan † Electronic supplementary information (ESI) available: The experimental section including the experimental method for the fabrication of Ta-doped a-Fe2O3 films, the characterization instruments, the photoelectrochemical measurement details, and the electronic structure calculation method. Additionally, the XRD and Raman patterns with related discussion and the Ta4f XPS spectra of the pure and Ta-doped films are included. See DOI: 10.1039/c5cp07479g

This journal is © the Owner Societies 2016

candidate for solar water oxidation, a-Fe2O3 (hematite) owns a combination of remarkable advantages such as nontoxicity, low cost, and chemical stability. And particularly the narrow band gap of about 2.1 eV enables a-Fe2O3 to absorb a very large portion of visible light in the solar spectrum.7–9 However, its low conductivity, high electron–hole recombination rate and slow surface reaction kinetics lead to very low efficiency for PEC water splitting over a-Fe2O3 photoanodes, which greatly restricts its practical application for solar energy conversion.10,11 To this end, many strategies aiming at resolving these intrinsic drawbacks have been proposed to enhance the PEC activities of a-Fe2O3 films and significant achievements have been made in the past few decades.10 In order to solve the problem that photocurrent density at low bias voltage is always limited by the four-electron water oxidation reaction, some effective oxygen evolution catalysts (OEC), e.g., Co–Pi,12 IrO2,13 CoOx,14 etc., have been used to modify a-Fe2O3 photoanodes. For example, Zhong et al. reported a stable 5-fold enhancement in the photocurrent density at 1.0 V vs. reversible hydrogen electrode (RHE) with the Co–Pi/a-Fe2O3 composite photoanodes compared to a-Fe2O3 photoanodes.15 Recently, Shen et al. created a Ag-doped a-Fe2O3 ultrathin layer on the surface a-Fe2O3 nanorods, and found that the Ag-doped overlayer not only increased the

Phys. Chem. Chem. Phys.

View Article Online

Published on 05 January 2016. Downloaded by Gazi Universitesi on 23/01/2016 05:00:55.

Paper

carrier density in the near-surface region but also accelerated the surface oxidation reaction kinetics, synergistically contributing to improved PEC performances.16 It has been well evidenced that doping is very effective to improve the charge carrier transport ability and hence PEC activities of a-Fe2O3 films by increasing the electrical conductivity.10 A large number of doping agents, such as Sb,17 Si,18 Sn19 and W20 have been reported to effectively improve the PEC activity for solar water splitting over a-Fe2O3 films. Cesar et al. obtained Si-doped a-Fe2O3 films by the atmospheric pressure chemical vapor deposition (APCVD) method and the samples gave a photocurrent up to 1.45 mA cm 2 at 1.23 V vs. RHE.21 In our previous studies, it was revealed that optimal W and Cr-doped a-Fe2O3 nanorod films exhibited much higher PEC activity than the undoped films under the irradiation of solar light, which should be due to the improved charge transfer ability and reduced charge recombination.20,22 Another effective approach to enhanced PEC activity of a-Fe2O3 films is to design specific nanostructures, e.g., nanowires,23 nanotubes,24 and nanobelts,25 for directed and promoted charge carrier transport. Vayssieres and co-workers fabricated a-Fe2O3 nanorod arrays by a simple solution method for the first time and investigated the PEC activity for water splitting.26,27 They found that a-Fe2O3 nanorod arrays possessed much higher PEC efficiency than nanoparticulate a-Fe2O3, acting as photoanodes for solar water oxidation, mainly due to their nanorod-diameters shortening the required hole transport length. To date, a-Fe2O3 nanostructures can be fabricated via different physical and chemical methods, including thermal oxidation,28 APCVD,29 electrodeposition,30 ultrasonic spray pyrolysis (USP),31 and atomic layer deposition (ALD).32 Compared to methods aforementioned, the solution-based growth method is attracting increasing attention since its first report,33 due to its unique advantages such as combining of facile and economic synthetic conditions and equipment, easy morphology control, and potential large scale production.34 Recently, metal ion doped a-Fe2O3 nanostructures have been successfully obtained via the aqueous solution growth method with the dopant precursor directly dissolved in the reaction solution. Cao et al. prepared Ti4+ doped a-Fe2O3 films by a facile hydrothermal reaction and found that the photocurrents of the doped a-Fe2O3 electrodes were much higher than those of the pure sample.35 In our previous work, Ti-doped and Zr-doped a-Fe2O3 nanostructured films were prepared via the aqueous solution growth method, producing photocurrent densities 7–14 fold higher than those of undoped a-Fe2O3 films.36,37 This remarkable enhancement in PEC performances should be due to the increase in charge carrier density or the reduction in charge recombination induced by Zr and Ti doping. In this study, a-Fe2O3 nanorod arrays were fabricated and doped with Ta5+ via the aqueous solution growth method with tantalic chloride (TaCl5) directly added in the precursor solution as the doping agent. The obtained Ta-doped a-Fe2O3 nanorod films showed considerable enhancement in PEC water splitting activity related to the undoped a-Fe2O3 nanorod film. Although Ta-doped a-Fe2O3 films prepared by high temperature annealing or drop casting have been demonstrated to show improved PEC performances,38–40 there is a rare report on the

Phys. Chem. Chem. Phys.

PCCP

direct growth of Ta-doped a-Fe2O3 nanorod arrays on conductive substrates for PEC water splitting. Moreover, the Ta-doping induced electronic structure and nanostructure evolution of a-Fe2O3 nanorods are also deserved to be investigated in detail, given the effects of pentavalent (5+) Ta dopants on the morphology and electronic properties of a-Fe2O3 nanorods intrinsically differing from those of reported effective Ti and Zr dopants in the chemical states of tetravalence (4+). Together with our previous achievements,36,37 the present study provides a general and alternative approach to low-cost and large-scale fabrication and modification of a-Fe2O3 films for efficient solar water splitting as well as other applications, such as gas sensing, lithium battery, etc.

Results and discussion The Ta-doped a-Fe2O3 films were grown onto fluorine-doped tin dioxide (FTO, Nippon Sheet Glass) coated glass substrates via the aqueous solution growth method described by Vayssieres,41 with minor modification (see Experimental in the ESI†). Fig. 1 shows the SEM images of undoped and Ta-doped a-Fe2O3 films to investigate the morphology change of nanorods induced by Ta doping. The undoped a-Fe2O3 sample (TaFe-0) clearly shows well-ordered nanorods vertically grown on the FTO substrates with a diameter of around 50 nm and a thickness of about 600 nm (Fig. 1A and B), which is quite similar to the previous reports.41,42 After Ta doping, all samples intimately grew on the FTO substrates and the film thickness did not change a lot (insets in Fig. 1C–H). At relatively low doping concentrations, the obtained Ta-doped a-Fe2O3 films (TaFe-x, x = 2, 5, 10, 20) kept their nanorod morphology with the diameter of nanorods gradually increasing (Fig. 1C–F). With further increasing Ta doping concentration, apparent morphology change happened to Ta-doped a-Fe2O3 films (TaFe-x, x = 50, 100), and the well-ordered nanorod structure was largely destroyed (Fig. 1G and H). This change in nanorod morphology means that high Ta dopant concentration will disrupt the [110]-oriental growth of the a-Fe2O3 nanorods perpendicular to the FTO substrates, evidencing the remarkable effects of Ta dopant concentration on the nanostructure of those obtained Ta-doped a-Fe2O3 films. Obvious morphology change also happened to a-Fe2O3 nanorod arrays after doped with Ti4+ and Zr4+, which was proposed to greatly depend on the amounts of dopant precursors added in the synthetic aqueous solution.36,37 As further investigated by XRD patterns and Raman spectra (see Fig. S1, S2 and related discussion in the ESI†), it was indicated that at a low doping level, the introduced Ta dopants will not change the crystal structure of a-Fe2O3 nanorods, with selective orientation of the (110) planes parallel to the substrate,43,44 while the high doping level would destroy the [110]-oriented growth of the a-Fe2O3 nanorods and result in the emergence of Ta2O5 species in a-Fe2O3. The morphology and structure of pristine (TaFe-0) and Ta-doped a-Fe2O3 (TaFe-10) samples were further characterized by TEM as shown in Fig. 2A and B. Both of them show similar morphology of nanorods with clean and smooth surface (Fig. 2A and B), meaning moderate Ta doping will not apparently change the morphology.

This journal is © the Owner Societies 2016

View Article Online

Published on 05 January 2016. Downloaded by Gazi Universitesi on 23/01/2016 05:00:55.

PCCP

Paper

Fig. 2 TEM image of (A) pristine a-Fe2O3 (TaFe-0), (B) Ta-doped a-Fe2O3 (TaFe-10), and (C) STEM elemental mapping images in the selected area for the (D) Fe element, (E) O element, and (F) Ta element distribution of TaFe-10. The insets in (A) and (B) show the corresponding HRTEM images, respectively. Defects are indicated by black arrows in the inset of (B).

Fig. 1 SEM images of undoped and Ta-doped a-Fe2O3 films. (A) Top-view SEM image and (B) cross-sectional SEM image of TaFe-0, (C) TaFe-2, (D) TaFe-5, (E) TaFe-10, (F) TaFe-20, (G) TaFe-50, and (H) TaFe-100. The insets are the cross-sectional SEM images of the same films.

To reveal the different structures between pristine and Ta-doped a-Fe2O3 samples, HRTEM images of TaFe-0 and TaFe-10 are also obtained as shown in Fig. 2 (insets in Fig. 2A and B). Lattice images show both of them are well crystallized, with the same lattice fringe spacing of 0.25 nm, corresponding to the interlayer spacing of the (110) planes of hematite. These results illustrate that low-level Ta doping will not change the lattice structure of hematite significantly. However, when compared to the perfect lattices of TaFe-0 (inset in Fig. 2A), some defects were clearly observed in the lattice of hematite for TaFe-10 (inset in Fig. 2B), which may be related to the disorder phase induced by Ta doping. STEM elemental mapping was further performed to confirm the existence and distribution of Ta element in TaFe-10. As shown in Fig. 2C–F, a substantial and clear Ta signal that correlates spatially with Fe and O could be observed, indicating a homogenous lateral distribution of Ta dopants over the entire a-Fe2O3 nanorods. These observations indicate that at moderate doping level, Ta atoms could be homogeneously doped into the lattice of hematite during the aqueous solution growth process without greatly altering the nanorod structure.

This journal is © the Owner Societies 2016

The surface elemental composition and the chemical states of Ta dopants were examined by XPS analysis. As shown in Fig. S3 in the ESI,† the binding energies of Ta4f7/2 and Ta4f5/2 for those Ta-doped a-Fe2O3 films were observed at 25.8 eV and 27.7 eV, respectively, with energy difference to be ca. 1.9 eV. These values are close to the previously reported values,38 confirming that Ta exists as Ta5+.45,46 It could be found in Fig. 3A that with the amounts of TaCl5 in solution precursor increasing, the surface molar ratio of Ta:Fe increased gradually from TaFe-0 to TaFe-100, meaning that the Ta dopant concentration in a-Fe2O3 films could be facilely tuned. The depth distribution of Ta dopants in the nanorod films was examined by XPS measurements during controlled Ar+ etching for 920 s, with XPS etching profiles for the TaFe-10 and TaFe-100 films shown in Fig. 3B. For the TaFe-10 film, during the initial etching period the molar ratio of Ta:Fe rapidly decreased from 0.020, and then to be leveled at B0.010 for longer than 80 seconds. This indicates that the Ta dopants concentrated at the surface of the a-Fe2O3 nanorods, which might be attributed to the absorption of Ta ions on the surface of the b-FeOOH nanorods during growth in the precursor solution. The molar ratio of Ta : Fe kept relatively constant in the bulk of a-Fe2O3 nanorods, signifying the homogeneous distribution of Ta dopants in the interior of the film along the nanorods (as supported by the STEM elemental mapping), and the successful incorporation of Ta into the b-FeOOH lattice during the growth of the nanorods in aqueous solution.37 Similar dopant distribution was also observed for other doped a-Fe2O3 films grown in aqueous solution.36,37 For the TaFe-100 film with high Ta dopant concentration, the molar ratio decreased very quickly (from 0.046 to 0.020) as similarly observed for TaFe-10 film during the initial etching for 80 seconds. Moreover, the molar ratio of Ta : Fe further decreased gently from 0.020 to 0.005 with the Ar+ etching proceeding to 900 seconds. Such a gradual distribution of Ta dopants in the bulk of the a-Fe2O3 film should be resulted from the large amount of TaCl5 in the precursor solution, as the excessive TaCl5 possibly led to the changed thermodynamic growth conditions and the formation of Ta2O5

Phys. Chem. Chem. Phys.

View Article Online

Published on 05 January 2016. Downloaded by Gazi Universitesi on 23/01/2016 05:00:55.

Paper

PCCP

Fig. 4 Mott–Schottky plots of TaFe-0, TaFe-10 and TaFe-100 in 0.10 M NaOH; frequency: 1 kHz.

Fig. 3 (A) Molar ratio of Ta : Fe in the doped films and (B) etching profile of TaFe-10 and TaFe-100 samples.

species in the near-surface region of the a-Fe2O3 film by overall consideration on the Raman spectra and XPS etching profiles. Based on the previous analysis and discussion, it has been convincingly evidenced that Ta dopants at a moderate concentration could be successfully incorporated into the lattice of a-Fe2O3 nanorods during the aqueous solution growth process without destroying the nanorod structure. Then, it will be of great necessity to investigate the effects of Ta dopants on the optical and electronic properties of these Ta-doped a-Fe2O3 nanorods and hence their PEC activities for solar water splitting, as discussed in detail in the following sections. Unsurprisingly, in good accordance with the previous work on the solution-based a-Fe2O3 films doped with other metal ions like W6+ and Cr2+,20,22 all the Ta-doped a-Fe2O3 samples obtained in this study did not show significant differences in optical absorption profiles (UV-Vis spectra not shown), with similar band gaps determined to be ca. 2.15 eV by the Kubelka– Munk equation, indicating the very similar optical properties of the Ta-doped a-Fe2O3 films would not play an essential role in enhanced PEC performances.38 In order to understand the effects of different Ta doping levels on the photoelectrode/electrolyte interfacial properties in electrolyte solution, Mott–Schottky (M–S) measurements were carried out at 1000 Hz in the dark for the pristine and Ta-doped a-Fe2O3 films (TaFe-10 and TaFe-100), with M–S plots shown in Fig. 4. The positive slopes indicate that both pristine and Ta-doped a-Fe2O3 samples are n-type semiconductors with electrons as the majority charge carriers. Nd was calculated to

Phys. Chem. Chem. Phys.

be 5.59  1019, 1.18  1020, and 1.03  1020 cm 3 from the slopes of the M–S plots for pristine a-Fe2O3, TaFe-10 and TaFe-100, respectively (see Experimental in the ESI†).36 Note that both TaFe-10 and TaFe-100 samples showed substantially increased donor density compared to the pristine a-Fe2O3 film. These values provide solid evidence to confirm that the Ta5+ ions act as donor dopants increasing the electron donor density in a-Fe2O3. Given the intercept of the M–S plots, the flat-band potentials (Vfb) were acquired to be 542, 550 and 425 mV vs. Ag/AgCl for a-Fe2O3, TaFe-10 and TaFe-100, respectively. When compared to TaFe-0, the Vfb shift for TaFe-10 is ignorable, while the Vfb of TaFe-100 shows apparent positive shift of B120 mV. Such anodic Vfb shift has been observed in Ti-treated a-Fe2O3, which might be caused by the morphology change in TaFe-100 (Fig. 1H) or new surface states introduced by Ta2O5 species in the near-surface region of the film with high concentration Ta dopants.30,47 As demonstrated by M–S measurements, Ta doping gave rise to a significant increase in the majority carrier concentration in the a-Fe2O3 film. In order to justify this conclusion, the DFT+U calculation was performed to elucidate the role of Ta impurity in modifying the electronic structure of a-Fe2O3. As seen from Fig. 5A, a Ta atom was incorporated to replace the Fe-04 atom in the crystallographic unit cell, where the direction of magnetic moments on Fe atoms was labeled. Two peaks in the projected density of states (PDOS) were observed below the Fermi level in the band gap of a-Fe2O3, as shown in Fig. 5B. It was indicated that such two impurity levels were occupied by two electrons which were localized on two Fe atoms. When a Ta atom with 5 valence electrons was substituted for a Fe atom with 8 valence electrons, three of five electrons would be taken by the surrounding O atoms, leading to two electrons left on the Ta atom. Ta 5d levels were higher in energy than Fe 3d levels, inducing residual two Ta 5d electrons donated to two Fe atoms (Fe-02 and Fe-07 atoms). As shown in Fig. 5C for Fe 3d states in the up-spin channel, the electron on the lower energy impurity level was localized on the Fe-02 3d t2g orbital and the other electron was localized on the Fe-07 3d t2g orbital according to ligand field theory.48 Meanwhile, the occupied 3d down-spin states of Fe-02 and Fe-07 atoms formed the top of the valence band (Fig. 5D). The reason why remaining Ta 5d up-spin electrons were transferred to Fe-02 and

This journal is © the Owner Societies 2016

View Article Online

Published on 05 January 2016. Downloaded by Gazi Universitesi on 23/01/2016 05:00:55.

PCCP

Fig. 5 (A) Crystallographic unit cell, (B) projected density of states (PDOS) of Ta 5d and O 2p orbitals, (C) PDOS of spin-up Fe 3d orbitals and (D) PDOS of spin-down Fe 3d orbitals for Ta-doped a-Fe2O3. Small red balls and big brown and green balls represent O, Fe and Ta atoms, respectively. The vertical dashed line denotes the Fermi level.

Fe-07 atoms rather than the Fe-05 atom was that the 3d states in the up-spin channel were fully filled for the Fe-05 atom but empty for Fe-02 and Fe-07 atoms.49 These electrons would migrate in the form of polarons and contribute to the conductivity of a-Fe2O3. Therefore, incorporation of a Ta element into a-Fe2O3 could increase the concentration of electron carriers and thus the conductivity, contributing to enhanced PEC performance for solar water splitting, as discussed in the following sections. In order to gain insight into the electronic structure of these Ta-doped a-Fe2O3 nanorod films, the in situ X-ray absorption spectra (XAS) were conducted and are shown in Fig. 6. Fig. 6A presents the O K-edge XAS of pristine (TaFe-0) and Ta-doped (TaFe-10 and TaFe-100) a-Fe2O3 films. For TaFe-0, the first two peaks at around 530 eV are attributed to the electron transition from O 2p to hybridized O 2p-Fe 3d states, which is consistent with the previous study.16 Due to the crystal field effect, the Fe 3d state is further split into lower energy 3d t2g and higher energy 3d eg states. It is clear that there is a significant spectral change on increasing the Ta concentration. In TaFe-10, the first peak is decreased in the intensity, which implies that some electrons enter into Fe 3d t2g states when doping with Ta. As the doping level is further increased, the spectrum varies substantially for TaFe-100. The first two peaks strongly reduce in their intensities and energies of these peaks shift to higher energies. The spectrum of reference Ta2O5 is also included at the bottom of Fig. 6A. As comparison, it could be deduced that the TaFe-100 may contain Ta oxides since the first peak reduces strongly and the energy of the second peak is in agreement with that of Ta2O5 (as detected by Raman spectra). Fe L-edge XAS probes the electron transition from Fe 2p to Fe 3d states, and thus can be used to investigate the Fe 3d states directly. As presented in Fig. 6B, no substantial difference of the spectral profiles of the Fe L-edge can be observed, indicating the Ta doping does not affect the Fe site significantly. When Ta

This journal is © the Owner Societies 2016

Paper

Fig. 6 X-ray absorption spectra (XAS) of pristine and Ta-doped a-Fe2O3 films. (A) O K-edge, (B) Fe L-edge, insets are the in situ XAS performed at the O K-edge and Fe L-edge under the light illuminated by an AM 1.5 solar simulator.

with 5 valence electrons substitutes the Fe atom with 3 valence electrons in a-Fe2O3, three of five valence electrons of Ta would be attracted by surrounding O atoms, giving rise to two electrons left behind. The surplus electrons may increase the electron donor density in a-Fe2O3 and may transfer to the oxygen site (as revealed by the reduction of peak intensity of the O 2p-Fe 3d t2g state in O K-edge XAS), lowering the conduction band, which is beneficial for the electron conductivity. In situ XAS were further performed at the O K-edge and the Fe L-edge under the light illumination of an AM 1.5 solar simulator. The insets of Fig. 6A display the O K-edge XAS of TaFe-10 and TaFe-100 and no spectral difference can be distinguished with and without light illumination. This suggests that O 2p is not the active site for the PEC performance. The insets of Fig. 6B show the Fe L-edge XAS of TaFe-10 and TaFe-100. No significant variation can be observed in TaFe-100, whereas, the increased intensity of the Fe L-edge in TaFe-10 in the illuminated condition is found when compared to that in the dark. The increase in peak intensity indicates that there are more available unoccupied Fe 3d states (i.e., more 3d holes) induced by the light, which implies that the Fe is the active site for the PEC activity. The more photoexcited holes can be ascribed to the fact that the band structure is modified by Ta doping, increasing electron donor density in the O 2p band and lowering in the conduction band, which could benefit PEC water splitting. As the doping level is increased in TaFe-100, there are possible Ta-oxide species formed in a-Fe2O3 (revealed by the O K-edge XAS in Fig. 6A) and these Ta-oxide species with a large bandgap may block photoinduced holes migrating to the surface, which could lead to lowered PEC ability for a-Fe2O3 with high Ta concentration. The effects of Ta doping on the PEC activity of a-Fe2O3 films will be discussed in detail in the following sections. Fig. 7 shows a set of light chopped current-potential plots for the pristine and Ta-doped a-Fe2O3 films measured in a 0.5 M Na2SO4 electrolyte with illumination of a solar simulator (AM1.5, 100 mW cm 2). For all the films the photocurrent densities

Phys. Chem. Chem. Phys.

View Article Online

Published on 05 January 2016. Downloaded by Gazi Universitesi on 23/01/2016 05:00:55.

Paper

Fig. 7 Current–potential plots for undoped and Ta-doped a-Fe2O3 films under solar light (AM 1.5, 100 mW cm 2). The inset shows the photocurrent at 1.0 V vs. Ag/AgCl plotted as a function of the weight of TaCl5 in the precursor solution.

increased gradually with the increased bias voltage. The photocurrent density of the undoped a-Fe2O3 film (TaFe-0) was about 0.15 mA cm 2 at 1.0 V vs. Ag/AgCl. In comparison to TaFe-0, Ta doping obviously enhanced the PEC performances for water splitting. With the increasing amounts of TaCl5 added in the precursor solution, the obtained Ta-doped a-Fe2O3 films showed gradually increasing PEC water splitting performances, with the highest photocurrent density reaching B0.53 mA cm 2 at 1.0 V vs. Ag/AgCl for the TaFe-10 film (inset in Fig. 7), which is 3.5 times as high as that of the pristine a-Fe2O3 film. However, further increase in the amounts of TaCl5 added in the precursor solution gave rise to a decrease in photocurrent densities from B0.53 mA cm 2 for TaFe-10 to B0.09 mA cm 2 for TaFe-100. Therefore, it is undoubtedly that Ta doping with moderate dopant concentration is necessary and also very effective to improve the PEC water splitting performances of a-Fe2O3 films. The incident-photon-to-current efficiency (IPCE) was further evaluated to confirm the improvement in the PEC performance of a-Fe2O3 films induced by Ta doping. As shown in Fig. 8, the IPCE profiles for undoped and Ta-doped a-Fe2O3 films show a similar trend depending on the wavelength of incident light. At wavelengths longer than 610 nm, the IPCE values drop to zero, in accordance with the band gap of a-Fe2O3, as frequently reported for doped a-Fe2O3.16,22 In comparison to the undoped a-Fe2O3 film (TaFe-0), the TaFe-10 film shows remarkable enhancement in IPCE values over the entire wavelength range of 350–610 nm. Most notably, the IPCE value at 360 nm was increased from 11% to 23% at 1.0 V vs. Ag/AgCl after Ta doping. However, the TaFe-100 film shows an IPCE value of 15% at 360 nm, much lower than that of the TaFe-10 film. Overall these results indicate the positive effects of appropriate Ta doping on the PEC activities for solar water splitting over a-Fe2O3 films, whereas superfluous Ta dopants would instead lower the PEC performance. As previously analyzed, the M–S measurement results showed that compared to the undoped film, the carrier density of the Ta-doped a-Fe2O3 films at a low doping level (TaFe-10 as the representative film) was dramatically enhanced by two fold.

Phys. Chem. Chem. Phys.

PCCP

Fig. 8 IPCE profiles of TaFe-0, TaFe-10 and TaFe-100 with applied potential set at 1.0 V vs. Ag/AgCl.

It has been widely reported that doping of foreign metal ions such as Ti4+, Zr4+ and Sn4+ could effectively modify the electronic properties of a-Fe2O3 films, resulting in the improved conductivity and carrier migration ability for the improved PEC activities of the a-Fe2O3 films as photoanodes.36,37,43 For TaFe-100, though the carrier density was also increased by Ta doping, the flat-band potential had 120 mV anodic shift, which may be related to the near-surface localized Ta2O5 species induced by superfluous Ta dopants (as detected by XAS in Fig. 6, Raman spectra in Fig. S2 and the related discussion in the ESI†) blocking the hole transfer from a-Fe2O3 to the electrolyte because of the mismatching band structure and the large band gap of Ta2O5.50 Then the increased overpotential for the water oxidation reaction leads to hole accumulation and more serious charge recombination at the photoanode surface, resulting in reduced PEC performances. Given the non-ignorable change of film morphologies induced by high concentration Ta doping, it is also necessary to take into consideration the possibility that the destroyed [100]-oriented growth of the nanorod structure would affect the PEC activities of the Ta-doped a-Fe2O3 films (e.g., TaFe-100) in a negative manner. The previous characterization results have revealed that a strong preferential orientation of the [110]-axis perpendicular to the FTO substrate could be kept for those Ta-doped a-Fe2O3 films with low doping concentrations (TaFe-x, x = 2, 5, 10, 20). Such a [110] preferential orientation has also been observed for Si-doped a-Fe2O3 films fabricated by APCVD.51 It has been well evidenced that the electron conductivity is up to 4 orders of magnitude higher in the [110] direction than the orthogonal in a-Fe2O3, as the classical explanation for the strongly anisotropic conductivity of a-Fe2O3.43,44,51–53 Then, another believable reason for the decreased PEC performances of Ta-doped a-Fe2O3 films with high doping concentrations (TaFe-x, x = 50, 100) may lie on the destroying of the [110]oriented nanorod structure retarding majority carrier (electron) conduction to the back contact. From the detailed discussion above, we can infer the possible mechanism of Ta dopants in changing the PEC performance of a-Fe2O3 films in different doping levels. As shown in Fig. 9, at low doping concentration, the Ta-doped a-Fe2O3 films possessed

This journal is © the Owner Societies 2016

View Article Online

PCCP

Paper

evolution should be balanced by optimizing the dopant concentration, for increased donor density and meanwhile with the unique nanostructure well preserved for efficient charge transfer.

Published on 05 January 2016. Downloaded by Gazi Universitesi on 23/01/2016 05:00:55.

Acknowledgements

Fig. 9 Charge transfer, band structure and nanostructure evolution of Ta-doped a-Fe2O3 films with different Ta dopant contents.

remarkably increased charge carrier density for increased electronic conductivity and hence the reduced charge carrier recombination, meanwhile with the [110]-oriented nanorod structure kept for directed electron conduction, synergistically contributing to the great improvement in PEC performances. However, at high doping concentration, despite increased charge carrier density, the formed Ta2O5 species blocked photoinduced hole transferring to the surface due to band structure mismatching between a-Fe2O3 and Ta-oxide species (Ta2O5), and moreover the [110]-oriented nanorod structure favoring electron conduction was destroyed, leading to the lowered PEC performance of the obtained Ta-doped a-Fe2O3 films. Thus, Ta doping could be very effective to enhance the PEC activity of a-Fe2O3 nanorods as photoanodes, while it is of great necessity to balance the trade-off between the electronic structure and nanostructure evolution by optimizing the Ta dopant concentration, for the increased donor density as well as the well preserved nanorod nanostructure for efficient charge transfer.

Conclusions In summary, Ta5+ ions could be doped into a-Fe2O3 nanorod arrays by a mild aqueous solution method using TaCl5 as the dopant precursor. The concentrations of Ta dopants could be easily controlled by varying the weights of TaCl5 added in the precursor solution. The photoelectrochemical (PEC) performances of the Ta-doped a-Fe2O3 films greatly depend on the Ta dopant concentrations, which increase first and then decrease with the increasing Ta dopant concentration. The Ta-doped a-Fe2O3 film with optimized Ta dopant concentration showed a photocurrent density of about 350% higher than the pristine one under solar irradiation at 1.0 V vs. Ag/AgCl. It is proved that the PEC enhancement induced by moderate Ta doping should be related to the increased charge carrier density. However, superfluous Ta dopants lead to blocked hole transfer and destroyed [110]-oriented growth of a-Fe2O3 nanorods, resulting in more serious surface charge recombination as well as poor electron conduction, and hence relatively poor PEC water splitting activity. This study clearly demonstrates that doping could be very effective to enhance the PEC activity of a-Fe2O3 nanostructures, while it is noteworthy that the trade-off between the electronic structure and nanostructure

This journal is © the Owner Societies 2016

The authors gratefully acknowledge financial support of the National Natural Science Foundation of China (No. 51323011, 51236007), the Program for New Century Excellent Talents in University (No. NCET-13-0455), the Natural Science Foundation of Shaanxi Province (No. 2014KW07-02), the Natural Science Foundation of Jiangsu Province (No. BK 20141212) and the Nano Research Program of Suzhou City (No. ZXG201442, No. ZXG 2013003). S. Shen is supported by the Foundation for the Author of National Excellent Doctoral Dissertation of P. R. China (No. 201335) and the ‘‘Fundamental Research Funds for the Central Universities’’. C. L. Dong is grateful to Ministry of Science and Technology (MoST) for financial support through No. 101-2112-M213-004-MY3 and thankful to National Synchrotron Radiation Research Center (NSRRC) for providing beamtime.

Notes and references 1 X. Chen, S. Shen, L. Guo and S. S. Mao, Chem. Rev., 2010, 110, 6503–6570. 2 A. Fujishima and K. Honda, Nature, 1972, 238, 37–38. 3 Y. Park, K. J. McDonald and K. S. Choi, Chem. Soc. Rev., 2013, 42, 2321–2337. 4 J. Su, X. Feng, J. D. Sloppy, L. Guo and C. A. Grimes, Nano Lett., 2011, 11, 203–208. 5 J. Hou, H. Cheng, O. Takeda and H. Zhu, Energy Environ. Sci., 2015, 8, 1348–1357. 6 R. Raja, P. Sudhagar, A. Devadoss, C. Terashima, L. K. Shrestha, K. Nakata, R. Jayavel, K. Ariga and A. Fujishima, Chem. Commun., 2015, 51, 522–525. 7 M. J. Katz, S. C. Riha, N. C. Jeong, A. B. F. Martinson, O. K. Farha and J. T. Hupp, Coord. Chem. Rev., 2012, 256, 2521–2529. 8 Gurudayal, P. M. Chee, P. P. Boix, H. Ge, Y. Fang, J. Barber and L. H. Wong, ACS Appl. Mater. Interfaces, 2015, 7, 6852–6859. 9 A. G. Tamirat, W. N. Su, A. A. Dubale, H. M. Chen and B. J. Hwang, J. Mater. Chem. A, 2015, 3, 5949–5961. 10 S. Shen, J. Mater. Res., 2014, 29, 29–46. 11 H. G. Cha, M. J. Kang, I. C. Hwang, H. Kim, K. B. Yoon and Y. S. Kang, Chem. Commun., 2015, 51, 6407–6410. 12 G. M. Carroll, D. K. Zhong and D. R. Gamelin, Energy Environ. Sci., 2015, 8, 577–584. 13 S. D. Tilley, M. Cornuz, K. Sivula and M. Gratzel, Angew. Chem., 2010, 49, 6405–6408. 14 Y. Surendranath, D. A. Lutterman, Y. Liu and D. G. Nocera, J. Am. Chem. Soc., 2012, 134, 6326–6336. 15 D. K. Zhong and D. R. Gamelin, J. Am. Chem. Soc., 2010, 132, 4202–4207. 16 S. Shen, J. Zhou, C. Dong, Y. Hu, E. N. Tseng, P. Guo, L. Guo and S. S. Mao, Sci. Rep., 2014, 4, 6627.

Phys. Chem. Chem. Phys.

View Article Online

Published on 05 January 2016. Downloaded by Gazi Universitesi on 23/01/2016 05:00:55.

Paper

17 J. Moir, N. Soheilnia, P. O’Brien, A. Jelle, C. M. Grozea, D. Faulkner, M. G. Helander and G. A. Ozin, ACS Nano, 2013, 7, 4261–4274. 18 A. Annamalai, P. S. Shinde, A. Subramanian, J. Y. Kim, J. H. Kim, S. H. Choi, J. S. Lee and J. S. Jang, J. Mater. Chem. A, 2015, 3, 5007–5013. 19 D. D. Qin, Y. L. Li, T. Wang, Y. Li, X. Q. Lu, J. Gu, Y. X. Zhao, Y. M. Song and C. L. Tao, J. Mater. Chem. A, 2015, 3, 6751–6755. 20 S. Shen, C. X. Kronawitter, J. Jiang, S. S. Mao and L. Guo, Nano Res., 2012, 5, 327–336. 21 I. Cesar, K. Sivula, A. Kay, R. Zboril and M. Graetzel, J. Phys. Chem. C, 2009, 113, 772–782. 22 S. Shen, J. Jiang, P. Guo, C. X. Kronawitter, S. S. Mao and L. Guo, Nano Energy, 2012, 1, 732–741. 23 M. T. Mayer, C. Du and D. Wang, J. Am. Chem. Soc., 2012, 134, 12406–12409. 24 T. J. LaTempa, X. Feng, M. Paulose and C. A. Grimes, J. Phys. Chem. C, 2009, 113, 16293–16298. 25 M. Marelli, A. Naldoni, A. Minguzzi, M. Allieta, T. Virgili, G. Scavia, S. Recchia, R. Psaro and V. D. Santo, ACS Appl. Mater. Interfaces, 2014, 6, 11997–12004. 26 T. Lindgren, H. Wang, N. Beermann, L. Vayssieres, A. Hagfeldt and S.-E. Lindquist, Sol. Energy Mater. Sol. Cells, 2002, 71, 231–243. 27 N. Beermann, L. Vayssieres, S. E. Lindquist and A. Hagfeldt, J. Electrochem. Soc., 2000, 147, 2456–2461. 28 L. Wang, C. Y. Lee, R. Kirchgeorg, H. Hildebrand, J. Mulle, E. Spiecker and P. Schmuki, Mater. Horiz., 2014, 1, 344–347. ¨tzel and 29 D. K. Zhong, M. Cornuz, K. Sivula, M. Gra D. R. Gamelin, Energy Environ. Sci., 2011, 4, 1759–1764. 30 N. Mrbagheri, D. Wang, C. Peng, J. Wang, Q. Huang, C. Fan and E. E. Ferapontova, ACS Catal., 2014, 4, 2006–2015. ¨tzel, J. Phys. Chem. B, 2005, 109, 31 A. Duret and M. Gra 17184–17191. 32 T. Wang, Z. Luo, C. Li and J. Gong, Chem. Soc. Rev., 2014, 43, 7469–7484. 33 L. Vayssieres, L. Rabenberg and A. Manthiram, Nano Lett., 2002, 2, 1393–1395. 34 L. Vayssieres, A. Hagfeldt and S. E. Lindquist, Pure Appl. Chem., 2000, 72, 47–52. 35 D. Cao, W. Luo, M. Li, J. Feng, Z. Li and Z. Zou, CrystEngComm, 2013, 15, 2386–2391.

Phys. Chem. Chem. Phys.

PCCP

36 S. Shen, C. X. Kronawitter, D. A. Wheeler, P. Guo, S. A. Lindley, J. Jiang, J. Z. Zhang, L. Guo and S. S. Mao, J. Mater. Chem. A, 2013, 1, 14498–14506. 37 S. Shen, P. Guo, D. A. Wheeler, J. Jiang, S. A. Lindley, C. X. Kronawitter, J. Z. Zhang, L. Guo and S. S. Mao, Nanoscale, 2013, 5, 9867–9874. 38 X. Zhang, H. Li, S. Wang, F. R. F. Fan and A. J. Bard, J. Phys. Chem. C, 2014, 118, 16842–16850. 39 Y. Cong, M. Chen, T. Xu, Y. Zhang and Q. Wang, Appl. Catal., B, 2014, 147, 733–740. 40 V. M. Aroutiounian, V. M. Arakelyan, G. E. Shahnazaryan, G. M. Stepanyan, J. A. Turner and O. Khaselev, Int. J. Hydrogen Energy, 2002, 27, 33–38. 41 L. Vayssieres, N. Beermann, S. E. Lindquist and A. Hagfeldt, Chem. Mater., 2001, 13, 233–235. 42 S. Shen, J. Jiang, P. Guo and L. Guo, Int. J. Photoenergy, 2013, 8, 174982. 43 Y. Ling, G. Wang, D. A. Wheeler, J. Z. Zhang and Y. Li, Nano Lett., 2011, 11, 2119–2125. ¨tzel, 44 I. Cesar, A. Kay, J. A. Gonzalez Martinez and M. Gra J. Am. Chem. Soc., 2006, 128, 4582–4583. 45 R. S. Devan, W. D. Ho, C. H. Chen, H. W. Shiu, C. H. Ho, C. L. Cheng, S. Y. Wu, Y. Liou and Y. R. Ma, Nanotechnology, 2009, 20, 445708. 46 M. T. Seman, J. J. Robbins, D. Leonhardt, S. Agarwal and C. Wolden, J. Electrochem. Soc., 2008, 155, J168–J174. 47 R. Franking, L. Li, M. A. Lukowski, F. Meng, Y. Tan, R. J. Hamers and S. Jin, Energy Environ. Sci., 2013, 6, 500–512. 48 Z. D. Pozun and G. Henkelman, J. Chem. Phys., 2011, 134, 224706. 49 A. Bandyopadhyay, J. Velev, W. H. Butler, S. K. Sarker and O. Bengone, Phys. Rev. B: Condens. Matter Mater. Phys., 2004, 69, 174429. 50 W. J. Chun, A. Ishikawa, H. Fujisawa, T. Takata, J. N. Kondo, M. Hara, M. Kawai, Y. Matsumoto and K. Domen, J. Phys. Chem. B, 2003, 107, 1798–1803. ¨tzel, J. Am. Chem. Soc., 2006, 128, 51 A. Kay, I. Cesar and M. Gra 15714–15721. 52 K. M. Rosso, D. M. A. Smith and M. Dupuis, J. Chem. Phys., 2003, 118, 6455–6466. 53 C. X. Kronawitter, I. Zegkinoglou, S. Shen, J. Guo, F. J. Himpsel, S. S. Mao and L. Vayssieres, Phys. Chem. Chem. Phys., 2013, 15, 13483–13488.

This journal is © the Owner Societies 2016

Solution growth of Ta-doped hematite nanorods for efficient photoelectrochemical water splitting: a tradeoff between electronic structure and nanostructure evolution.

Ta-doped hematite (α-Fe2O3) nanorod array films were successfully prepared on fluorine-doped tin dioxide (FTO) coated glass substrates via a facile so...
566B Sizes 1 Downloads 12 Views