The Case of Hematite - ACS Publications - American Chemical Society

Subscriber access provided by UNIV OF CAMBRIDGE

Article

Enhanced Bulk and Interfacial Charge Transfer Dynamics for Efficient Photoelectrochemical Water Splitting: The Case of Hematite Nanorod Arrays Jian Wang, Bo Feng, Jinzhan Su, and Liejin Guo ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b07723 • Publication Date (Web): 10 Aug 2016 Downloaded from http://pubs.acs.org on August 13, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27

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

ACS Applied Materials & Interfaces

Enhanced Bulk and Interfacial Charge Transfer Dynamics for Efficient Photoelectrochemical Water Splitting: The Case of Hematite Nanorod Arrays Jian Wang, Bo Feng, Jinzhan Su,* Liejin Guo International Research Centre for Renewable Energy (IRCRE), State Key Laboratory of Multiphase Flow in Power Engineering (MFPE), Xi’an Jiaotong University (XJTU), Xi’an, Shaanxi 710049, China. KEYWORDS hematite nanorods, titanium dioxide, underlayer, overlayer, charge transfer, photoelectrochemical

ABSTRACT Charge transport in the bulk and across the semiconductor/electrolyte interface is one of the major issues that limit the photoelectrochemical (PEC) performance in hematite photoelectrodes. Efficient charge transport in the entire hematite is of great importance to obtain high photoelectrochemical property. Herein, to reach this goal, we employed both TiO2 underlayer and overlayer deposition on hematite nanorod films, followed with a fast annealing treatment. The TiO2 underlayer and overlayer not only serve as dopant sources for carrier density increasing, but also reduce charge recombination at fluorine-doped tin oxide (FTO)/hematite

ACS Paragon Plus Environment

1


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

Page 2 of 27

interface and accelerate charge transfer across the hematite/electrolyte interface, respectively. This synergistic doping and interface modifying effects give rise to an enhanced photoelectrochemical water oxidation performance of hematite nanorod arrays, generating an impressive photocurrent density of 1.49 mA cm-2 at 1.23 V vs. RHE. This is the first report on using both underlayer and overlayer modification with same material to improve charge transport through the entire electron transport path in hematite, which provides a novel way to manipulation of charge transfer across semiconductor interface for a high-performance photoelectrode.

Introduction Hematite (α-Fe2O3) is regarded as a promising candidate of photoanode material for photoelectrochemical (PEC) solar water splitting due to its abundance, low cost and favourable band gap (2.0~2.2 eV).1-2 However, hematite suffers from poor electrical conductivity, a short hole diffusion length of 2-4 nm and sluggish water oxidation kinetics, giving rise to low solar to hydrogen (STH) conversion efficiency.3-5 To improve the PEC performance, electron-hole recombination in bulk hematite and at hematite/electrolyte interface are two key issues that must be addressed.6 Doping is regarded as one of the most common strategies to improve electron transport or reduce bulk recombination in hematite.7 Various dopants such as Ti4+, Si4+ and Sn4+ have been used to enhance the conductivity in hematite.8-10 High charge recombination rate at FTO/hematite interface usually induces low charge collection efficiency in hematite photoelectrode.6 One method to address this issue is employing functional underlayers such as TiO2, SiOx, Ga2O3 and Nb2O5 between fluorine-doped tin oxide (FTO) substrate and hematite.1114

These ultrathin underlayers have proper conduction band edge positions which are beneficial


2

Page 3 of 27

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


to reduce electron-hole recombination at the FTO/hematite interface.6 TiO2 is one of the most commonly used underlayer materials for hematite films.11, 15-16 The effect of TiO2 underlayers on the PEC performances of thin hematite films was first investigated by Hisatomi et al.11 It was found that the photocurrent increased with a thin TiO2 underlayer between FTO substrate and hematite and decreased significantly when increasing the thickness of TiO2 underlayers, which was explained by the fact that the offset of the conduction band edges at TiO2/Fe2O3 was relatively large and electrons need to tunnel through the TiO2 underlayer to reach the FTO. However, the exact nature of TiO2 underlayer for improved photocurrent had not been investigated in their studies. In a recent report, TiO2 underlayer was spin-coated on FTO substrates before growing hematite nanorods.15 After high temperature sintering, Ti was doped into Fe2O3 from the TiO2 underlayer. Thus the TiO2 underlayer acts as both a source for titanium dopants which improved electron conductivity and also as a charge recombination barrier by suppressing the electron back transfer from FTO to hematite. However, the role of TiO2 underlayer that acts as a blocking layer was not investigated in detail in this study. Another main factor limiting the PEC efficiency of hematite photoanodes is slow water oxidation reaction kinetics.17 Thus large overpotential is often required to decrease surface recombination which is accompanied with hole accumulation at the surface. Two main strategies are usually adopted to address this problem, which include modifying the hematite surface with an oxygen evolution catalyst (OEC) that possesses fast water oxidation kinetics and passivating surface states of hematite using a functional overlayer of metal oxides.18-20 The surface modification of hematite electrodes with OECs (e.g. Co-Pi, FeOOH, Ni(OH)2) has been demonstrated to decrease the overpotential required.19, 21-22 Surface passivation using functional overlayers such as Ga2O3 and Al2O3 has been demonstrated to eliminate surface states thus


3


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

Page 4 of 27

reduce surface recombination.18, 23 TiO2 is an ideal semiconductor loaded on hematite which can improve the PEC performance of hematite greatly.24-25 However, the mechanisms for enhancement were divergent which include Ti doping from the TiO2 overlayers upon high temperature annealing that promoted band bending and increased carrier density and constructing a tandem heterojunction structure between Fe2O3 and TiO2 which improves the charge separation efficiency at the interface.25-26 A recent report showed that TiO2 overlayer is reacted with hematite to form Fe2TiO5 after high annealing process, which is confirmed by x-ray photoelectron spectroscopy (XPS) and soft X-ray absorption spectroscopy (XAS).24 The author hypothesized that some Ti-treated hematite was actually Fe2TiO5 in hematite rather than typically Ti doping. As discussed above, the exact nature of TiO2 underlayer and overlayer for improved PEC performance is still unclear and needs further detailed studies. Furthermore, to achieve a better performance it is desirable to fabricate hematite photoelectrode with both TiO2 underlayer and overlayer for a synergistic enhancement. Herein, to increase charge transfer and separation efficiency in the entire hematite photoanodes including bulk hematite and hematite/electrolyte interface simultaneously and to further investigate the exact role of TiO2 underlayer and overlayer, we employed both TiO2 underlayer and overlayer deposition on hematite nanorod films by a facile and cost-effective method. The following short-time annealing treatment induce Ti doping and high crystallinity without damaging the conductivity of substrates. The functions of TiO2 underlayer and overlayer were also investigated and it revealed that both TiO2 underlayer and overlayer served as dopant sources for carrier density increase while simultaneously TiO2 underlayer could reduce charge recombination at FTO/hematite interface by acting as a block layer for hole transport and TiO2 overlayer effectively accelerated charge


4

Page 5 of 27

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


transfer across the hematite/electrolyte interface, respectively. This synergistic doping and interface modifying effects gave rise to increased carrier density and PEC water oxidation performance. Experimental details Preparation of hematite photoanodes. Hematite nanorod arrays were synthesized using a simple hydrothermal method followed by two-step annealing.27 In a typical synthesis process, one piece of FTO substrate (14 ohm/sq, 3 × 4 cm2, Nippon Sheet Glass Co. Ltd.) was placed in an aqueous solution containing 0.15 M FeCl3·6H2O and 1 M NaNO3 at pH 1.5 (set by HCl) in a Teflon vessel. Then the vessel was sealed and maintained at 100 oC for 24 h in an oven. After reaction ceasing and cooling down naturally, the yellow thin film was annealed at 550 oC for 2 h followed by 90 s-short annealing at 800 oC. For preparing TiO2 underlayer, 0.01 M tetrabutyl titanate (TBOT) solution in absolute ethanol was spin-coated (2000 rpm for 20s) on FTO substrates twice followed by annealing at 300 oC for 1h. For loading TiO2 overlayer on the surface of hematite film, before two-step annealing the substrates coated with FeOOH nanorods was placed in 0.01 M TBOT solution in ethanol at 60 oC for 2h then followed by annealing treatment. In a remainder of this paper, we denote the as-prepared bare hematite photoanode as bare Fe2O3, hematite with a TiO2 underlayer as UL-Fe2O3, hematite with a TiO2 overlayer as Fe2O3-OL and hematite with both TiO2 underlayer and overlayer as UL-Fe2O3-OL photoelectrode, respectively. Photoelectrochemical measurements. The photocurrent density, Mott-Schottky (MS) curves and electrochemical impedance spectroscopy (EIS) were conducted in a three-electrode cell with hematite film as working anode, Ag/AgCl as reference electrode and a Pt slice as counter


5


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

Page 6 of 27

electrode. The electrolyte was an aqueous solution of 1 M NaOH (pH 13.6) or a mixed solution of 0.5 M NaI and 1 M NaOH (pH 13.6). The simulated sunlight (light intensity: 100 mW cm-2) was generated by a 500 W Xe lamp (PerfectLight Co. Ltd., China) equipped with an AM 1.5 G filter. All measurements were carried out on an electrochemical station (CHI 760D). To measure the J-V curves, the voltage scan rate is 0.01 V s-1. MS data were conducted with frequency of 1 kHz under dark condition. EIS spectra were collected with potential frequency range of 100 kHz to 0.01 Hz and potential range of 0.8 to 1.6 V vs. RHE, and ZSimpWin software was adopted to fit the data. The applied potential was converted into a reversible hydrogen electrode (RHE) potential using following equation: ERHE = EAg/AgCl + 0.059 × pH + E0Ag/AgCl where E0Ag/AgCl is 0.1976 V at 25 oC. The incident photon-to-current efficiency (IPCE) was conducted using the same cell and the monochromatic light was generated using a 150 W Xe lamp (LSP-X150, Zolix Instruments Co. Ltd., China) together with a spectrometer (Omni-λ1805i, Zolix Instruments Co. Ltd., China). The IPCE value was obtained by applying an external potential of 1.23 V vs. RHE and was calculated according to the following formula: IPCE = (hc/e)(Iph,λ/Pλλ) where Iph,λ is the photocurrent density, Pλ is the power intensity of the light, λ is the wavelength of light, and h, c , and e are Planck’s constant, speed of light in vacuum, and elementary charge, respectively.


6

Page 7 of 27

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


Material characterization. The morphologies of all samples were characterized by field emission scanning electron microscope (SEM, 7800F, JEOL, Japan) and transmission electron microscope (TEM, G2F30, FEI, USA). X-ray diffraction spectra (XRD) were carried out with an X-ray diffractometer (X’pert PRO MPD, PANalytical, the Netherlands) using Cu Kα radiation (λ ~ 0.154056 nm). X-ray photoelectron spectra (XPS, AXIS Ultra DLD, Shimadzu/Kratos Analytical) were conducted to investigate the element composition and their chemical states. The absorption properties were carried out on the UV-Vis spectrophotometer (UV4100, HITACHI, Japan). Results Morphology and structure. The morphology of hematite nanorod arrays was examined by scanning electron microscope (SEM). As exhibited in Figure 1, all the films show nanorod arrays standing perpendicularly on the FTO substrates, which is similar as the report in literature.15 The cross-section SEM images show that the hematite films without TiO2 underlayer (bare Fe2O3 and Fe2O3-OL) have a thickness of around 720 nm, and with a thin TiO2 underlayer, the films (ULFe2O3 and UL-Fe2O3-OL) exhibit smaller thickness of around 620 nm. This is due to different nucleation process between FTO and TiO2. TiO2 overlayer does not change the thickness of hematite photoelectrodes owing to its ultrathin characteristics. X-ray diffraction (XRD) spectra (shown in Figure S1) of all the samples exhibit that all the films show similar crystal structure containing two phases of hematite (JCPDS No. 33-0664) and cassiterite and no peaks of impurity is detected, indicating that TiO2 overlayer with a low amount is amorphous or completely incorporates into hematite nanorods after high temperature annealing.26 The morphology and structure as well as Ti distribution in hematite nanorods were further investigated using transmission electron microscope (TEM) and high-resolution TEM (HRTEM). All samples show


7


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

Page 8 of 27

irregular shapes (Figure S2) owing to recrystallization during the high temperature annealing. The observed distinct crystal lattices of 0.265 nm and 0.247 nm are in accordance with (104) and (110) crystal planes of hematite, respectively. Furthermore, any other crystal phases, e.g. Fe2TiO5, are not found on the surface of Fe2O3-OL electrode, suggesting that Ti element serves as doping source other than reacts with hematite to form new phase. The scanning transmission electron microscopy (STEM) mode-energy dispersive spectroscopy (EDS) mapping images show that Ti diffuses into hematite from both underlayer and overlayer and Ti atoms can be found throughout the entire nanorods (shown in Figure 2). The different signal intensities suggest that Ti concentration in Ti underlayer sample is lower than in TiO2 overlayer sample because the deposited TiO2 underlayer is ultrathin and the amount of Ti diffused from the bottom of nanorods is small. As we adopted a fast annealing process, the Sn signal intensities in all samples which are introduced from FTO substrates are low. Further, the thicknesses of TiO2 underlayer and overlayer which were detected by SEM and TEM images (shown in Figure S3), were found to be around 4 nm and 3-7 nm, respectively. X-ray photoelectron spectroscopy (XPS) survey spectra was conducted to further investigate Ti doping in hematite nanorods (Figure 3a), while no Ti peak was found for the bare hematite film, as expected. All the samples show distinct Fe 2p3/2 peak at around 710.8 eV which is typical value observed for Fe2O3 (Figure 3b). The O 1s spectrum for Fe2O3-OL photoanode (Figure S4a in Supporting Information) shows that main peak at 529.9 eV matches well with the reported value for Fe2O3,28 and the shoulder peak at 531.7 eV is due to the surface hydroxyl groups.29 Of note, as shown in Figure 3c, the binding energy of O 1s main line for Fe2O3-OL and UL-Fe2O3-OL electrodes (~529.9 eV) shift slightly to higher value compared with bare Fe2O3 and UL-Fe2O3 electrodes (~529.8 eV). This shift may be induced by surface Ti doping according


8

Page 9 of 27

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


to previous reports.28 Figure 3d reveals that Ti peaks are evident in Fe2O3-OL and UL-Fe2O3-OL photoanodes after annealing at 800 oC, while Ti peaks are relatively weak for UL-Fe2O3 photoanode (see inset of Figure 3d), because only small amount of Ti atoms diffuse from the bottom of films to the top layer. By observing detailed Ti XPS spectrum and fitting curves for Fe2O3-OL photoanode (Figure S4b), two distinct peaks at 458.3 and 464.0 eV corresponding to Ti 2p3/2 and Ti 2p1/2, respectively, are fully consistent with Ti4+ state in the Ti-O-Fe structure.30 The shoulder at band energy of 459.5 eV is characteristic of Ti4+ sate in TiO2 lattice.31 This result indicates that both Ti elements as a dopant in α-Fe2O3 lattice as TiO2 as overlayer can be detected in the Fe2O3-OL photoelectrode. The concentrations for each element in hematite were also deduced from XPS analysis, as summarized in Table 1. The sample with overlayer treatment (Fe2O3-OL) gives about 7.95% Ti atoms detected at the surface, which is much higher than 0.33% of sample with underlayer treatment (UL-Fe2O3), while sample with both overlayer and underlayer treatment (UL-Fe2O3-OL) gives the highest surface Ti atom ratio of 8.68%, which accounts for the synergistic doping effect. This high Ti concentration will induce high donor density in hematite which is beneficial for PEC property. Sn concentration at the surface diffused from the substrate is relatively low in all samples owing to the short-time annealing treatment, in consistent with signal intensity in EDS data. PEC performance. The photocurrent responses of hematite nanorod films were measured in 1 M NaOH electrolyte under AM 1.5 G illumination (100 mW cm-2). As shown in Figure 4a, bare Fe2O3 film shows negligible photocurrent density of 0.005 mA cm-2 at 1.23 V vs. RHE. With a thin TiO2 underlayer (UL-Fe2O3), the photocurrent increased around 57 times reaching 0.29 mA cm-2 at 1.23 V vs. RHE. While surface treated hematite nanorods with a TiO2 overlayer (Fe2O3OL) gives a 183 times higher photocurrent of 0.92 mA cm-2 at 1.23 V vs. RHE than the bare


9


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

Page 10 of 27

hematite. With both TiO2 underlayer and overlayer treatment on hematite nanorods, the photocurrent of UL-Fe2O3-OL photoanode is further increased to 1.49 mA cm-2 (297 times higher than bare Fe2O3) at 1.23 V vs. RHE and 2.86 mA cm-2 at 1.6 V vs. RHE, demonstrating the synergistic enhancing effect of TiO2 underlayer and overlayer. UV-Vis absorptance spectra show that all samples have similar absorption and same absorption edge at around 600 nm (see Figure S5), which indicate that the indirect band gaps of all hematite photoanodes are around 2.08 eV and Ti doping doesn’t contribute to the light absorption. Incident photon-to-current efficiency (IPCE) for hematite photoelectrodes were conducted by measuring photocurrent at 1.23 V vs. RHE as a function of incident light wavelength (Figure 4b). In consistent with the J-V curves, Fe2O3-OL and UL-Fe2O3 photoelectrodes exhibit substantially enhanced IPCE values compared to bare hematite nanorods. With both TiO2 underlayer and overlayer, UL-Fe2O3-OL photoelectrode generates an IPCE value of 32.2% at 300 nm and 19.8% at 340 nm, respectively. The IPCE value dropped to zero at wavelengths over 600 nm, in accordance with the bandgap of hematite. The observed high efficiency for TiO2 underlayer or overlayer treated sample can be attributed to increased charge transport and improved donor density by Ti doping. We also calculated the photocurrent density by integrating the IPCE spectra with a standard AM 1.5 G (100 mW cm-2) solar spectrum using the following equation:

700

I ph = ∫300

λIPCE ( λ )P( λ )d ( λ ) 1240

where P(λ) is the solar spectra irradiance at a particular wavelength (λ) and IPCE(λ) is the obtained IPCE data as a function of wavelengths (λ) at 1.23 V vs. RHE. The calculated photocurrents were 0.02, 0.17, 0.63 and 0.81 mA cm-2 for bare Fe2O3, UL-Fe2O3, Fe2O3-OL and UL-Fe2O3-OL samples, respectively. It shows that the calculated photocurrent for bare Fe2O3


10

Page 11 of 27

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


photoelectrode is higher than the measured photocurrents at 1.23 V vs. RHE, while for TiO2 treated sample the calculated values are lower. This discrepancy is believed to be caused by the mismatch between the light sources adopted in photocurrent measurement and IPCE measurement. Discussion. A recent report showed that TiO2 underlayer played two roles in hematite nanorod films by acting as Ti doping source and functional blocking layer.15 The bifunctional role of TiO2 results in improved charge transfer and separation efficiency in hematite films. To investigate the effect of TiO2 underlayer, we compared the photocurrents under illumination from different sides, e. g. the frontside (film side) or the backside (FTO side). As illustrated in Figure 4c, at low applied potential (

The Case of Hematite - ACS Publications - American Chemical Society

Recommend Documents