SILAR-Deposited Hematite Films for Photoelectrochemical Water

Subscriber access provided by EASTERN KENTUCKY UNIVERSITY

Article

SILAR-Deposited Hematite Films for Photoelectrochemical Water Splitting: Effects of Sn, Ti, Thickness, and Nanostructuring Anthony J. Abel, Ivan Garcia-Torregrosa, Anjli M. Patel, Borirak Opasanont, and Jason B. Baxter J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/jp510027u • Publication Date (Web): 09 Feb 2015 Downloaded from http://pubs.acs.org on February 9, 2015

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.

The Journal of Physical Chemistry C 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 55

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

The Journal of Physical Chemistry

SILAR-Deposited Hematite Films for Photoelectrochemical Water Splitting: Effects of Sn, Ti, Thickness, and Nanostructuring

Anthony J. Abel,a Ivan Garcia-Torregrosa,a,† Anjli M. Patel, Borirak Opasanont, and Jason B. Baxter* Department of Chemical and Biological Engineering, Drexel University, Philadelphia, PA, 19104, USA a

Authors contributed equally to this work.

†

Present address: Department of Chemistry, Inorganic Chemistry and Catalysis group, Utrecht

University, Utrecht 3584 CG, The Netherlands *

Corresponding author. E-mail: [email protected]. Ph: 215-895-2240, Fax: 215-895-5837

1 ACS Paragon Plus Environment


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 55

Abstract Hematite (α-Fe2O3) has been widely investigated for photoelectrochemical (PEC) water splitting, but questions remain regarding the nature of improvements induced by different dopants. We report on facile annealing treatments to dope hematite with Ti and Sn, and we provide insight into the effects of the dopant concentration profiles on two key steps of PEC water oxidation: charge separation and interfacial hole transfer.

Hematite thin films were

deposited by successive ionic layer adsorption and reaction (SILAR), with and without the presence of a TiO2 underlayer on the F:SnO2 substrate, and annealed to drive diffusion of Ti and Sn from the underlying layers into the hematite. PEC measurements showed that Ti and Sn at the hematite surface increase hole injection efficiency from nearly zero to above 80%. Ti and Sn also slightly improve charge separation efficiency, although separation efficiency remains below 20% due to low hole mobility and high recombination rate.

To overcome the small hole

transport length, extremely thin hematite coatings were deposited on Sb:SnO2 monolayer inverse opal scaffolds. Photocurrent increased proportionately to the surface area of the scaffold. This study provides insight into the use of dopants and nanostructured architectures to improve PEC performance of hematite photoanodes.

Keywords: iron oxide, photocatalysis, solar fuel, hydrogen, tin, titanium


Page 3 of 55

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


Introduction The initial work on light-driven water splitting with TiO2 by Honda and Fujishima1 more than four decades ago has inspired much research in materials and devices capable of efficiently utilizing sunlight to produce hydrogen and other so-called solar fuels.

However, many

challenges must be overcome to compete with non-renewable sources of energy. Photoelectrochemistry (PEC) is a promising option to directly convert solar energy into energy stored in chemical bonds, provided that durable and abundant materials are employed, such as simple and multinary oxides of transition metals. Within this framework, hematite (α-Fe2O3) has been identified as a possible candidate photoanode material for PEC water splitting due to its good stability in aqueous media, abundance in the Earth’s crust, and bandgap (2 – 2.2 eV) that is within the optimal range for water splitting.2,3 Despite substantial research efforts over several decades, many challenges with hematite PEC water splitting have inhibited realization of solar-to-hydrogen efficiencies above 10% or photocurrents above 12 mAcm-2 that are theoretically possible for a material with band gap of 2.1 eV.4,5 Challenges include a conduction band position that is too low in energy to drive the hydrogen evolution reaction without bias, mismatch between optical absorption depth and carrier collection length, and sluggish kinetics of the oxygen evolution reaction (OER). In addition to challenges with hematite itself, poor quality interfaces between hematite and the current collector can considerably decrease photocurrent. Different doping strategies have been pursued to overcome hematite’s low intrinsic electron and hole mobility and high electron-hole recombination rate that result in short carrier collection lengths.6 Recent studies on the effects of doping 3 at% Ti into hematite grown by 3 ACS Paragon Plus Environment


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

atomic layer deposition show considerably enhanced PEC performance.7 The authors attribute the improvement to an increase in the hole collection length and also to increased OER efficiency that results from Ti atoms at the hematite surface. In another study, the incorporation of Ti in nanostructured hematite by a solid state reaction showed an 8-fold increase in photocurrent density.8 Alternatively, Sn can be incorporated into the hematite lattice by diffusing from the underlying fluorine-doped tin oxide (FTO, F:SnO2). This effect has resulted in drastic improvement of the onset potential and photocurrent density when hematite nanorod arrays deposited on FTO were annealed above 650 °C.9,10 The improvement was attributed to improved electron transport in the bulk hematite and to improvements of the FTO-hematite interface. While improvements in PEC performance with multiple different dopants have been reported, the mechanism of improvement and the nature of interactions between different dopants are often unclear. In addition to doping hematite to modify its bulk and surface properties, charge collection can also be improved by designing the hematite to have nanostructured architecture. Nanostructuring brings a larger fraction of the volume to within the hole collection length of the liquid junction. This strategy has been demonstrated by Sivula, Gratzel, and coworkers using Sidoped hematite in a nanocauliflower architecture.11,12 Alternatively, nanostructured conductive scaffolds can be coated with hematite in an extremely thin absorber or host-guest geometry. In these systems, the thickness of the semiconductor can be fine-tuned according to its minority carrier collection length and the interfacial area can be chosen to provide sufficient volume for light absorption.13,14 Examples of nanostructured host-guest systems like nanorods, nanonets, or inverse opal structures have been published in recent years,15,16,17,18,19,20 exhibiting photocurrents up to 3 mAcm-2 for undoped hematite without using sacrificial agents or co-catalysts.17 Scaffolds


Page 4 of 55

Page 5 of 55

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


must be chosen carefully to have appropriate band alignment with hematite as well as sufficient conductivity to ensure efficient charge transport. Herein, we report on the fabrication of hematite thin films prepared by successive ionic layer adsorption and reaction (SILAR), a solution-based layer-by-layer deposition method that provides tight control over the film thickness. PEC performance depends sensitively on film thickness, the presence of a TiO2 underlayer, and annealing conditions, which lead to different concentration profiles of Ti and Sn dopants throughout the film. We show that Ti and Sn dopants at small concentrations can increase photocurrent by improving both charge separation and, more importantly, interfacial hole transfer. In a complementary approach to further increase photocurrent, hematite was deposited on a monolayer inverse opal scaffold of Sb:SnO2 that resulted in 78% larger photocurrents than in a planar film.

Experimental Material Fabrication FTO/glass substrates (Hartford Glass, 12 Ω/sq) were cleaned by sequential sonication in Contrad detergent, 50/50 ethanol/acetone, and deionized (DI, 18 MΩcm-1) water. Thin films of iron (III) oxyhydroxide were deposited on the substrates by SILAR (nanoStrata, Strato Sequence VI). 0.05 M FeCl3 (97%, Sigma-Aldrich) and 0.1 M NaOH (98%, Sigma-Aldrich) were used as the iron and hydroxide precursor solutions, respectively. One deposition cycle was defined as immersion in (1) cation solution, (2) DI water rinse, (3) anion solution, (4) DI water rinse. Each immersion step lasted for 10 s. The anion and rinsing solutions were replaced every 30 cycles to prevent particle formation that could lead to non-planar films. To obtain fully oxidized α-Fe2O3



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

with desirable PEC performance, the samples were annealed in air using a box furnace. All samples were annealed at 450 °C for 10 min and then subsequently at either 600 °C for 29 min or 775 °C for 15 min. The average heating rate and cooling rates were ~35 and ~10 °C min-1, respectively. The temperature programs were selected so that all samples were heated for the same total amount of time at or above 450 °C (~50 min) and at or above 600 °C (~29 min). In most cases, an ultrathin (10-20 nm) film of TiO2 was deposited on the FTO before hematite deposition. A modified titanium butoxide sol-gel was prepared according to a previously described procedure.21 The sol-gel was diluted to 0.067 M in ethanol and was then deposited onto substrates by dip coating. Substrates were immersed for 17 s and withdrawn at a rate of 120 mm/min. Both immersion and withdrawal occurred at ambient conditions. The substrates were dried at room temperature for 30 min before sintering in air using a box furnace at 500 °C for 1 hr. In some samples, inverse-opal scaffolds were fabricated using methods from microsphere lithography. Close-packed monolayers of silica spheres were formed on substrates by selfassembly, infiltrated with Sb:SnO2 (ATO) sol-gel, and then etched away. The silica spheres had diameters of 1 µm and were prepared according to a recipe described in literature22 with waterto-ethanol volume ratio fixed at 0.65. The solution was stored at room temperature for 5 hr and centrifuged at 3500 rpm. The recovered powders were treated at 550 °C for 2 hr to remove remaining volatiles. A 20 wt% suspension of the spheres in ethanol was prepared by ultrasonication for 30 min. The ATO sol-gel for the scaffold was prepared from a 0.3 M aqueous solution of SnCl2·H2O (98 % pure, Alfa Aesar) and a 0.13 M aqueous solution of SbCl3, each of which were dissolved in ethanol and heated at 80 °C in a closed vessel for 2 hr under stirring. Both vessels


Page 6 of 55

Page 7 of 55

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


were then opened, allowing the solutions to evaporate for another 2 hr until dry powders were formed. After recovery of the powders, 0.35 M and 0.7 M mixtures containing 93 solids% SnCl2(EtO)2 and 7 solids% Sb(EtO)3 were prepared in ethanol at 50 °C for 2 hr under stirring. The resultant ATO sol-gels were allowed to age for 24 hr before use. To generate the inverse-opal scaffold, a hexagonal close-packed monolayer of silica microspheres was deposited onto the substrates by dip coating. Substrates were immersed for 3 min and withdrawn at a rate of 120 mm min-1. Immersion and withdrawal were performed at 40% humidity. The substrates were then dried at 120 °C for 1 hr. The resulting microsphere layer was infiltrated with 0.35 M ATO sol-gel by spin coating at 4000 rpm for 30 s, with a ramp of 1500 rpm/s. The films were dried at 100 °C for 30 min and sintered at 500 °C for 15 min. This process was repeated once more to obtain desired coverage, after which a final annealing step was performed at 600 °C for 1 hr. To remove the silica microspheres, the substrates were etched in 2 M NaOH at 50 °C for 15 min. A final dip coated layer of 0.7 M ATO sol-gel was deposited at 30% humidity and an 80 mm/min withdrawal rate, followed by annealing at 600 °C for 1 hr. This final step improves mechanical and electrical properties of the scaffold while avoiding excessive filling of the inverse opal structure.

Film Characterization Wavelength-dependent absorption spectra for each sample type was determined by measuring both transmission and diffuse reflection using a home-built UV-vis transmission spectrometer (Ocean Optics, USB 4000) and a 1.5-inch integrating sphere. Correction for reflectance is necessary because reflection varies with film thickness.



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

Grazing-incidence X-ray diffraction (GIXRD) was conducted on thin film samples at incidence angles ranging from 0.3° to 0.8° using a Rigaku SmartLab diffractometer equipped with Cu Kα X-ray source excited at 40 kV and 44 mA. Diffractograms were acquired using parallel beam in-plane optics with a step size of 0.016° and analyzed with Fityk software.23 Instrumental broadening for determination of crystallite size was obtained from (100) peak of ZnO nanowires with crystallite size >100 nm. X-ray photoelectron spectroscopy (XPS) measurements were performed on a Physical Electronics VersaProbe 5000 with a monochromatic Al Kα X-ray source operated at 15 kV and 25 W, yielding 100 µm spot size. Depth profiling was performed by argon sputtering at 2 kV and 2 µA within a 2 mm x 2 mm area. Data analysis and peak fitting were performed using CasaXPS software with the relative sensitivity factors provided by the equipment manufacturer. Metallic elements were assigned Shirley-type background while non-metals were assigned linear background. Peaks were fitted with 70% Gaussian - 30% Lorentzian line shape. All spectra were corrected by adjusting the adventitious C-C 1s peak to 285.0 eV.

Photoelectrochemical Characterization All samples were masked with electroplater’s tape that contained a 0.38 cm2 (7 mm diameter) hole to define the active area. The samples were examined in contact with a 1 M NaOH (pH 13.6) electrolyte. A saturated Ag/AgCl electrode (ThermoScientific, Orion 90-02) was used as a reference electrode, and a platinum coil was used as the counter electrode. All potentials were converted to the reversible hydrogen electrode (RHE) scale by: = ° / + 0.1976 + 0.0592


Page 8 of 55

Page 9 of 55

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


PEC measurements were performed with a Gamry potentiostat and electrochemical software. An increasingly anodic bias was applied at a rate of 50 mVs-1. The light source was a 300 W Xe arc lamp (Newport). An AM 1.5G filter (Newport) was used to simulate the solar irradiation spectrum, and the power was calibrated to 100 mWcm-2 (1 sun) using a thermopile detector. Light was incident on the front (electrolyte) side unless otherwise noted. A custom-made glass electrochemical cell (Starna Cells Inc.) with quartz windows housed the electrodes and electrolyte for all measurements.

Results and Discussion Characterization of Hematite Films Deposited by SILAR Hematite thin films for PEC water splitting require control over crystallinity, film thickness, conductivity, and surface chemistry.

Hematite thin films were fabricated by

depositing iron oxyhydroxide (FeOOH) using SILAR and then annealing in air. SILAR is a solution-phase alternative to atomic layer deposition that is carried out at ambient conditions and provides good control over film thickness and properties. In this work, FeCl3 was used as the iron source because of superior homogeneity and flatness compared to previously reported recipes that use Fe2SO4.24,25,26 This appears to be the first reported use of a FeCl3-based SILAR recipe for deposition of hematite, although a similar recipe without annealing was recently reported for deposition of FeOOH.27,28 Top-view and cross-sectional SEM images of films deposited from FeCl3, Figure 1(a,b), reveal relatively uniform, conformal coverage of the FTO substrate. A growth rate of ~5 Å/cycle is estimated from images of the thickest films. While the films have some slight nanostructuring and roughness, they appear to be of comparable flatness



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

a)

Page 10 of 55

Hematite FTO

b)

Figure 1. (a) Top view and (b) cross-sectional scanning electron micrographs Fe2O3/TiO2/FTO with 90 SILAR cycles of Fe2O3. The inset in (a) shows the bare FTO substrate. The ultrathin TiO2 layer is not visibly detectable in (b) at this magnification.

to films grown by atomic layer deposition29 and certainly do not have the nanorod structure often seen in hematite fabricated by chemical bath deposition.9 Photographs of the hematite films in Figure 2(a) show the characteristic rust color, which becomes increasingly dark with larger numbers of SILAR cycles. Deposition is uniform over areas larger than 1 cm2. Absorption was quantified by UV-visible absorption spectroscopy, as shown in Figure 2(b). The absorption onset at approximately 580 nm is consistent with the


Page 11 of 55

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


a) 30

60

90

120

150

180

210

Figure 2. (a) Photograph and (b) absorption spectra of SILAR Fe2O3 films of different thicknesses (numbers of SILAR cycles).

hematite band gap of ~2.1 eV.30 Absorption increases with increasing number of SILAR cycles. No difference in absorption was observed when an ultrathin underlayer of TiO2 was included. GIXRD, in Figure 3(a), confirms the presence of hematite, with characteristic diffraction peaks at 33.2° and 35.6° corresponding to (104) and (110) hematite reflections, respectively. Peak positions are similar for samples annealed at 600 °C and 775 °C, indicating that the difference in annealing temperature does not influence the crystallinity. The α-Fe2O3 peaks do not change position or relative intensity with grazing incidence angle, which indicates that the film has hematite modification throughout its thickness. The relative intensity of the two peaks also matches the JCPDS powder spectrum, confirming that there is no preferential alignment or texture of the film. The (104) diffraction peaks were fitted with pseudo-Voigt line shape, and 11 ACS Paragon Plus Environment


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 12 of 55

FWHM was used to determine crystallite size according to the Scherrer equation. The crystallite size of the 600 °C and 775 °C annealed samples, corrected for instrumental broadening, are 53 nm and 66 nm respectively.31 These values are close to the film thickness of 60 – 72 nm, indicating a high degree of crystallinity in films annealed at both temperatures. XPS detail spectra of the Fe 2p3/2 region, Figure 3(b), also confirm the hematite modification. Samples annealed at 600 °C and 775 °C each exhibit the double peak at 710.8 and 709.7 eV that is characteristic of hematite.32,33,34

Figure 3. (a) Grazing incidence X-ray diffractograms of samples annealed at 600 °C and 775 °C, normalized to (104) α-Fe2O3 peak height at 33.2°. Grazing incidence angle was 0.3°. Vertical bars represent JCPDS power diffraction peaks of α-Fe2O3 (hematite) (PDF# 01-072-0469) in orange, and tetragonal SnO2 (PDF# 01-088-0287) in blue. (b) XPS spectra of the Fe 2p3/2 region from the surfaces showing the double-peak at 710.8 eV and 709.7 eV characteristic of α-Fe2O3 for both annealed samples. 12 ACS Paragon Plus Environment

Page 13 of 55

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


With the combined evidence of XRD and XPS, we believe this is the first report of SILAR deposition of hematite on a transparent conducting oxide substrate. Importantly, we later show that SILAR is able to coat high-aspect ratio nanostructures, an important feature for possible PEC applications.

Effect of Underlayers, Dopants, and Film Thickness on PEC Performance Water splitting with hematite photoanodes was tested in a 3-electrode PEC cell to understand the influence of film thickness, annealing conditions, and underlayers on PEC performance. Figure 4 shows linear sweep voltammetry (LSV) curves under 1-sun illumination for four representative samples with 60 SILAR cycles (~30 nm) of Fe2O3 to compare the effects of annealing temperature (600 °C or 775 °C) and the presence or absence of the ultrathin TiO2 underlayer. Films without TiO2 that were annealed at 600 °C show negligible photocurrent, and their light curves are nominally identical to their dark curves.

Increasing the annealing

Figure 4. LSV curves for 60 SILAR cycles samples with and without the TiO2 underlayer and annealed at 600 °C and 775 °C 13 ACS Paragon Plus Environment


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 14 of 55

temperature to 775 °C results in increased photocurrent, reaching 0.1 mAcm-2 at 1.23 VRHE with an onset potential (Von) of 1.00 VRHE. Introduction of a TiO2 underlayer with 600 °C anneal results in a similar increase in photocurrent, but with a favorable shift in Von of 110 mV. Combining both the 775 °C anneal and a TiO2 underlayer results in further improvement of the photocurrent to 0.25 mAcm-2 at 1.23 VRHE; however, Von returns to 1.00 VRHE. While the improvement in onset potential with TiO2 is lost at higher annealing temperature, the TiO2 does delay the onset of dark current to 1.70 VRHE in each case. The TiO2 may increase the photocurrent by improving the interface between Fe2O3 and FTO, but Ti has also been reported to diffuse easily through Fe2O3 so bulk or surface effects of Ti are also possible.35 The shift in onset potential indicates a surface effect, despite the position of TiO2 under the Fe2O3 film.

Higher annealing temperature dramatically increases the

photocurrent, although XRD and XPS in Figure 3 showed negligible change in crystallinity. XPS depth profiles and PEC characterization of hematite films grown under different conditions were used to provide some insight into the role of dopants on PEC performance and are discussed below. XPS depth profiles were conducted to determine the chemical composition and dopant distribution within the film. Figure 5(a,b) shows atomic concentration as a function of sputter time (depth) for hematite films annealed at 600 °C and 775 °C, and Figure 5(c) shows the Sn 3d and Ti 2p detail spectra at 775 °C. Ti 2p peaks throughout the depth of the film are centered at 458.5 eV, which corresponds to the Ti4+ state.36,37 Sn 3d peaks throughout the film are centered at 468.4 eV, corresponding to the Sn4+ state.36,37 Each peak can be fitted with one component with small FWHM, supporting the interpretation of a single oxidation state for both Ti and Sn.


Page 15 of 55

c)

60

40

kcps

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


16 min 20

4 min

Sn 3d

0 510

500

490

Ti 2p 480

470

460

Surface 450

440

Binding Energy (eV)

Figure 5. Depth profile of the atomic concentrations of (a) 600 °C and (b) 775 °C annealed samples as determined by X-ray photoelectron spectroscopy (XPS). The hematite film was 120 cycles thick. Besides Fe, O, and C, the surface of 600 °C annealed sample has 0.5% Ti, while that of 775 °C annealed sample has 5% Ti, 5% Na, as well as 0.5% Sn. A sputter rate of ~4 nm min-1 is estimated from ref. 32. (c) XPS spectra highlighting peaks attributed to Sn 3d and Ti 2p. Increasing line height corresponds to increasing etch time.

Both annealing temperatures show elemental intermixing due to diffusion, with a more



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 16 of 55

distributed profile in the higher temperature sample. Ti from the TiO2 underlayer diffuses both toward the surface into Fe2O3 and toward the backside into FTO. Beyond just diffusing through the bulk Fe2O3, Ti also accumulates at the free surface. At the surface of the 600 °C sample, slight accumulation of ~0.5% Ti is found, which is a factor of two greater than the concentration at a 2 min sputter step below the surface. For the 775 °C sample, the surface concentration of Ti is an order of magnitude larger, ~5%, and maintains a surface accumulation profile similar to that of the 600 °C sample. The higher annealing temperature also enables Sn to diffuse from the FTO substrate through the hematite film. Sn concentration decreases monotonically to ~0.5% at the surface at 775 °C. At 600 °C, Sn diffuses slightly into the hematite, but its concentration is negligible in the near-surface region. With the 775 °C anneal only, Na is also initially present at the surface at up to 5% concentration, but was no longer detected after PEC testing (Figure S1). It may initially be present due to diffusion from the soda lime glass or contamination, but it does not appear to affect PEC performance, whose results are reproducible over many LSV sweeps. Incorporation of Ti and Sn into the hematite film at bulk and surface sites could impact both charge separation and surface reaction kinetics.38,39 To better understand the role of these dopants and limitations resulting from light absorption, charge separation, and OER kinetics, the effects of film thickness and hole scavengers were investigated. LSVs under 1-sun illumination are shown in Figure 6(a) for Fe2O3 samples with TiO2 underlayer (henceforth called Ti:Fe2O3) with thickness ranging from 30 to 210 SILAR cycles and annealed at 775 °C. Under these sample conditions, photocurrent increases monotonically with increasing thickness over the range studied as a higher fraction of incident photons is absorbed. Von was 1.0 VRHE, independent of thickness. With the 210-cycle film, the current


Page 17 of 55

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


density reaches 0.57 mAcm-2 at 1.23 VRHE, which is among the highest reported for planar hematite films of any thickness or fabrication method.40,41,42,43,44,45,46 Similar LSV curves were collected as a function of film thickness for Fe2O3 and Ti:Fe2O3 samples annealed at 600 °C and 775 °C to delineate the effects of annealing and the TiO2 underlayer. Figure 6(b) displays the current density sampled at 1.23 VRHE for each of the four sample types over a range of film thickness. Current increases approximately linearly for the 775

a) 210 180 150 120 90 60 30 cycles

b)

Figure 6.(a) Current density vs. voltage curves for Ti:Fe2O3 samples from 30 to 210 SILAR cycles annealed at 775 °C. A typical dark curve is shown by the dashed black line. (b) Current density vs. film thickness (in number of SILAR cycles) at 1.23 VRHE. Error bars represent 95% confidence intervals with sample size n ≥ 4. 17 ACS Paragon Plus Environment


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 18 of 55

°C Ti:Fe2O3 films shown in Figure 6(a). No photocurrent was observed for Fe2O3 films without TiO2 annealed at 600 °C regardless of thickness, confirming the need for thermal activation described previously.9,39 Films annealed at 775 °C without TiO2 had currents that were at least four times smaller than those with the TiO2 underlayer at the smallest thicknesses (60 cycles or less), but at larger thicknesses the currents were similar regardless of the presence of TiO2. This trend indicates that with 775 °C annealing, the primary role of TiO2 is to improve the FTO/Fe2O3 interface rather than to increase conductivity or passivate the surface. With this aggressive heat treatment, Sn from the FTO can diffuse throughout the film and also passivate the surface, as indicated by XPS in Figure 5(b), obviating the need for Ti to serve that same function. The increase in current with TiO2 underlayer at small thickness may result from improved nucleation and growth of Fe2O3 to form continuous and more crystalline films with smaller numbers of SILAR cycles. Ga2O3 and Nb2O5 underlayers have previously been reported to improve PEC performance by improving crystallinity of thin hematite films.15,40,46 While both Fe2O3 and Ti:Fe2O3 samples annealed at 775 °C have photocurrents that increase monotonically with film thickness, the Ti:Fe2O3 samples annealed at 600 °C show a maximum photocurrent at intermediate thickness, as shown in Figure 6(b). The initial increase in photocurrent with thickness is due to increased light absorption. The decline at larger thickness results from limitations in both charge separation and OER efficiency. The primary role of Ti appears to be surface passivation, such that surface recombination is suppressed and OER is favored, as will be shown in greater detail later. According to the XPS depth profiles in Figure 5, Sn does not diffuse all the way through the film at 600 °C, and improvements to bulk properties and surface reactions rely on Ti alone. Some Ti is found throughout the film in all samples measured, but at large thickness and low annealing temperature the amount of Ti


Page 19 of 55

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


appears insufficient to improve performance, and the additional Ti and Sn obtainable with higher annealing temperature is desirable. To reveal whether photocurrent is primarily limited by charge separation or hole injection, LSV curves for all sample variations at all thicknesses were performed in a 1 M NaOH – 0.5 M H2O2 electrolyte. H2O2 quickly scavenges photogenerated holes that have reached the hematite surface, eliminating the barrier to interfacial hole transfer present with OER.47 LSVs with H2O2 are shown in Figure 7(a) for each sample type with a thickness of 120 SILAR cycles. LSVs with H2O2 for other configurations are shown in Figure S2. Unlike the LSV curves in the standard electrolyte without H2O2, photocurrent is realized at all potentials, indicating that the injection barrier for holes to the electrolyte is removed. Fe2O3 annealed at 600 °C exhibits a rapid increase in photocurrent beginning at 0.53 VRHE, followed by a gradual rise until the onset of dark current at ~1.2 VRHE. No photocurrent was observed for this sample without H2O2, confirming that hole injection in OER is a significant barrier. With the TiO2 underlayer, Ti:Fe2O3 annealed at 600 °C shows similar behavior but Von that is delayed by ~100 mV. Although the onset is later, the “plateau” photocurrent is larger with the TiO2 underlayer. However, the opposite is observed in Ti:Fe2O3 and Fe2O3 annealed at 775 °C, where the presence of the TiO2 underlayer slightly reduced the photocurrent. Annealing at 775 °C also delayed the onset of dark current to ~1.4 VRHE.



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 20 of 55

Figure 7. (a) Current density vs. voltage curves for all sample variations in 1 M NaOH – 0.5 M H2O2 with thickness of 120 SILAR cycles, (b) Current density vs. voltage curves for 120 SILAR cycle thickness Ti:Fe2O3 at 775 °C (red curves) and 600 °C (blue curves) in 1 M NaOH – 0.5 M H2O2 (solid) and 1 M NaOH (dashed).

Because surface reaction limitations in this electrolyte may be neglected for all samples with H2O2, these data confirm that Ti and Sn play an important and complex role in bulk properties as well as hole injection kinetics. Figure 7(b) compares LSV curves for Ti:Fe2O3 annealed at 600 °C and 775 °C in 1 M NaOH – 0.5 M H2O2 and in 1 M NaOH. Without H2O2 in the electrolyte, photocurrent is significantly improved upon higher temperature annealing. However, when H2O2 is present, samples annealed at 600 °C exhibit a higher photocurrent than those annealed at 775 °C. 20 ACS Paragon Plus Environment

Page 21 of 55

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


In order to quantify the role that these dopants play in charge separation and injection, and to remove light harvesting as a convoluting factor, we utilize an approach first described by Dotan et. al. and used extensively elsewhere.47-50 The water-splitting photocurrent, Jph, can be expressed as = ×

! × "#$

where Jabs is the photon absorption rate expressed as current density, ηsep is the charge separation efficiency within the bulk film, and ηinj is the efficiency of hole injection from semiconductor to electrolyte. With the addition of 0.5 M H2O2, a known hole scavenger, to the electrolyte, ηinj can be treated as unity. Thus, ηsep can be calculated according to: &%

!

%

=

and ηinj is given by: &

"#$

=

%

&%

%

Figure 8(a) shows the potential-dependent ηsep as a function of film thickness for each of the four sample variations, and Figure 8(b) compares ηsep at 1.23 VRHE for each sample type across film thicknesses. In all cases, ηsep increases with applied potential because the depletion width increases. However, ηsep is less than 30% even in the best case, so charge separation is a significant limitation of photocurrent. In both Fe2O3 and Ti:Fe2O3 annealed at 600 °C, ηsep decreases as thickness increases. This trend is expected because electron-hole recombination competes with charge separation, and increasing the length that electrons and holes must travel to reach the interfaces will increase the probability that they recombine instead. At 1.23 VRHE, ηsep is only ~20% even for the thinnest films (~15 nm thick) due to low carrier mobility and short



a) Separation (%)

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 22 of 55

Fe2O3 600 °C

Ti:Fe2O3 600 °C

Fe2O3 775 °C

Ti:Fe2O3 775 °C

Potential (V vs RHE)

b) Figure 8. (a) Charge separation efficiency vs. potential for each of the four sample types. Color scheme for number of SILAR cycles is shown at right. (b) Charge separation efficiency as a function of sample thickness at 1.23 VRHE for each sample type. Error bars represent 95% confidence intervals with sample size n ≥ 4.

photoexcited carrier lifetimes. ηsep decreases to ~5% for film thicknesses above 100 nm. The reduction in ηsep with increasing thickness is much less pronounced in Ti:Fe2O3, which has higher electron density. The higher electron density may improve collection of electrons, which are generated farther from the FTO interface in thicker films since light is incident on the side of the liquid junction. In both Fe2O3 and Ti:Fe2O3 samples annealed at 775 °C, ηsep decreases only very slightly as thickness increases, indicating that Sn, too, improves charge separation, and may obviate the need for Ti to serve this purpose. The improvement in charge separation with either Ti or Sn may be due to improved electron collection lengths and also due to a larger electric field across a narrower space charge region that favors charge separation and reduces surface recombination.51 For all sample types, ηsep is significantly larger for the 30-cycle films than for thicker films. We later calculate a depletion width of 10 – 15 nm, which is similar to the thickness of the thinnest films. Therefore we attribute the increased ηsep to the fact that all


Page 23 of 55

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


charge generation occurs within the depletion region. That ηsep is less than 30% even in this extreme case indicates the fast recombination and poor hole transport in hematite. Potential-dependent ηinj is plotted in Figure 9(a) as a function of thickness for each of the three sample variations that have non-zero efficiency in standard electrolyte. ηinj varies over a much wider range than ηsep across different thicknesses, sample types, and potentials, reaching 85% in the best case. ηinj is negligible at low applied potentials and increases with anodic bias. The electron density decreases as the film surface becomes more depleted, which reduces electron-hole recombination and increases hole lifetime.52 For Ti:Fe2O3 annealed at 600 °C, ηinj is ~30% at 1.23 VRHE in the thinnest films and decreases to below 10% in thick films. While the decrease in ηsep with increasing thickness was expected because of the larger transport lengths, it was initially surprising that film thickness had such a dramatic effect on ηinj. We conclude that the amount of Ti or Sn dopant at the surface is critical to OER. Since all of the Ti and Sn diffuses from underlying layers during annealing, thicker films have lower dopant concentrations at the surface. Clearly too little Ti at the surface reduces ηinj for thick films. However, close examination of Figure 9 shows that ηinj also decreases at the smallest thicknesses, indicating an optimal range for Ti incorporation. The thickness-dependent surface concentration of Sn also affects ηinj. Figure 9 shows that ηinj increases with film thickness for Fe2O3 (without TiO2 underlayer) annealed at 775 °C, from less than 10% to above 60% at 1.23 VRHE over our thickness range. The large ηinj at 775 °C, coupled with ηinj near zero for 600 °C anneal when no Sn is present at the surface, confirms that Sn can enhance hole injection. However, the trend of decreasing ηinj with smaller thickness, and larger surface Sn concentration, indicates that too much Sn can severely inhibit the surface reaction. Use of both the TiO2 underlayer and the higher annealing temperature results in both



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

a)

Ti:Fe2O3 600 °C

Fe2O3 775 °C

Page 24 of 55

Ti:Fe2O3 775 °C

b) Figure 9. (a) Charge injection efficiency vs. potential for Ti:Fe2O3 annealed at 600 °C, Fe2O3 annealed at 775 °C, and Ti:Fe2O3 annealed at 775 °C. Injection efficiency was negligibly small for Fe2O3 annealed at 600 °C. Color scheme for number of SILAR cycles is shown at right. (b) Charge injection efficiency versus thickness at 1.23 VRHE for each sample type. Error bars represent 95% confidence intervals with sample size n ≥ 4. Sn and Ti at the surface in significant concentration. This scenario yields the largest ηinj at large anodic bias, achieving OER efficiencies of ~60% at 1.23 VRHE with only slight thickness dependence. Interestingly, this case also yields the worst onset potential of the three, indicating the complex dependence of interfacial hole transfer on near-surface composition. To summarize, for film thicknesses larger than the expected depletion width, doping with Sn or Ti increases bulk conductivity and improves charge separation. Small concentrations of Sn or Ti on the surface increase the efficiency of OER at the surface, but more aggressive annealing causes excessively large atomic concentration at the surface and decreases OER efficiency. Comparing the magnitudes of ηsep and ηinj in Figures 8(b) and 9(b), photocurrent at water oxidation conditions is more strongly limited by charge separation than by hole injection for all samples studied. However, since all samples of a given thickness have similar ηsep, their relative


Page 25 of 55

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


PEC performance is then governed by ηinj, as seen by comparing the trends in Figure 9(b) with those in Figure 6(b). Surprisingly, thickness affects ηinj much more than ηsep because of the sensitivity of surface recombination kinetics to the concentration of Sn and Ti dopant atoms occupying surface sites. The mechanism of improvement provided by these dopants is still ambiguous, but is likely related to passivation of surface defects. Mott-Schottky analysis and electrochemical impedance spectroscopy (EIS) were performed to gain further insight into the role of Sn and Ti in the PEC water splitting process. Figure 10(a) shows Mott-Schottky plots at 100 Hz with the standard electrolyte, and Table 1 shows the fitted flat band potential (Vfb) and carrier density (ND), for the Fe2O3 and Ti:Fe2O3 films of 120-cycle thickness that were annealed at 775 °C. Mott-Schottky plots of samples at 600 °C were not included because there was no obvious linear region. Vfb of 0.4 – 0.5 VRHE were obtained from the fit, which is consistent with the fairly large range of reported Vfb for hematite.7,12,47 A 0.09 V shift to less positive potential was observed in Vfb of Ti:Fe2O3 relative to Fe2O3 at 775 °C, and may partially explain the improved injection efficiency and Von of Ti:Fe2O3 samples. This shift in Vfb with Ti has been observed previously in Ti:Fe2O3 prepared by radio frequency magnetron sputtering.53 In addition to shifting Vfb, Ti diffusing from the underlying TiO2 layer also increases dopant density by about a factor of three. Carrier density in each case is on the order of 1019 cm-3, and the likely donor species are oxygen vacancies as well as Ti and Sn that have diffused from underlying layers. From these data, the depletion width is calculated to be 10 – 15 nm. Combined with the hole diffusion length of 2 – 4 nm,54 this gives a typical hole collection length that is comparable to the thickness of the 30-cycle SILAR films. With such a small hole collection length, it is somewhat surprising that the photocurrent continues to increase with film thickness. This increase may be due to slight nanostructuring of the films



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

Cbulk

c) Rs

Rct,ss Rtrap

Css

Figure 10. (a) Mott-Schottky plots of inverse square capacitance as a function of potential for 120-cycle samples annealed at 775 °C. Samples were tested in 1 M NaOH at 100 Hz. (b) EIS Nyquist plots of each sample variation with 120-cycle thickness. Plots were generated at 1.23 VRHE under 1 sun illumination. (c) Equivalent circuit used to fit EIS data.


Page 26 of 55

Page 27 of 55

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


Table 1. Fitted parameters from Mott-Schottky and EIS analysis. Vfb (V) Fe2O3, 600 °C Ti:Fe2O3, 600 °C Fe2O3, 775 °C Ti:Fe2O3, 775 °C

0.49 0.40

ND (1019 cm-3) 1.3 4.7

Rs (Ω) 28.6 26.6 50.6 43.0

Rtrap (kΩ) 3.7 2.8 1.1 0.8

Rct,ss (kΩ) 280.0 2.4 1.0 0.3

Cbulk (µF) 2.3 3.1 2.0 3.8

Css (µF) 7.4 10.3 16.5 28.7

upon annealing such that they are not planar, as been assumed here. We note that backside illumination gives similar LSVs to front side illumination for films with 60 SILAR cycles or less, Figure S2, but photocurrent is significantly reduced in thicker films with backside illumination as hole transport limitations dominate the behavior. Hole transport limitations are also confirmed by the spectrally-resolved internal quantum efficiency, Figure S2, which is 25% with 400 nm light that is absorbed near the front surface and monotonically decreases to 0% for longer wavelengths that penetrate more deeply into the film. Figure 10(b) shows Nyquist plots of EIS data at 1.23 VRHE for all sample variations of 120-cycle thickness.

EIS was performed under 1-sun illumination and with the standard

electrolyte. The spectra consist of two overlapping semicircles and were modeled with the equivalent circuit shown in Figure 10(c). This model was reported by Hamann and Bisquert for hematite deposited by atomic layer deposition.55,56 It considers charge transport through the bulk film and interfacial hole transfer from surface states. Fitted parameters are listed in Table 1. With 600 °C anneal, the presence of Ti reduces surface state charge transfer resistance, Rct,ss, by two orders of magnitude, which is in accordance with the dramatically improved ηinj in Figure 9(a). Rct,ss is further reduced by 3 – 10x with higher annealing temperature because of the presence of Sn (or possibly Na) at the surface, as well as Ti in the case of Ti:Fe2O3. The trend in Css runs opposite that of Rct,ss, with highest surface state capacitance in the presence of both Ti 27 ACS Paragon Plus Environment


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 28 of 55

and Sn. Rtrap follows the same trend as Rct,ss, although the change between the least and most doped samples is only a factor of five. These trends are consistent with the analysis in Steier et al.51 The increase in Rs at 775 °C is attributed to decreased conductivity of the FTO upon hightemperature annealing.

Use of Inverse Opal Scaffolds to Increase Photocurrent While the presence of Sn and Ti can have a pronounced effect on PEC conversion, in all cases Figure 8 shows that the charge separation efficiency is below 30% for the thinnest films studied here and declines further with increasing film thickness. Reducing film thickness to improve charge separation results in very little light absorption, while increasing film thickness improves light harvesting at the expense of charge separation.

One of the most effective

methods to decouple light absorption and charge collection is to design architectures consisting of extremely thin absorbers coated onto transparent, conductive scaffolds with large surface area.57,58,59 Here we fabricated an inverse-opal scaffold by self-assembly of a monolayer of SiO2 microspheres onto FTO substrates by dip coating, infiltrating the interstitial voids with ATO solgel by a multi-step spin coating process, and then etching away the SiO2 template. Figure 11(a) shows the ATO inverse opal scaffold at this stage. After etching, a final ATO spin coat step was applied to fill any nanopores in the scaffold, improving conductivity and adhesion to the substrate. The ATO scaffold was then coated with a conformal TiO2 interlayer and Fe2O3 absorber using the same procedures used for planar films, resulting in the structures shown in Figure 11(b). ATO was selected as the scaffold material for several reasons. Unlike FTO, ATO can be processed completely using sol gel chemistry. ATO has wide band gap, so is transparent to 28 ACS Paragon Plus Environment

Page 29 of 55

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


a)

500 nm

b)

200 nm

Figure 11. (a) Top-view SEM image of ATO inverse opal scaffold deposited by sol-gel, after etching away silica spheres but before final ATO treatment. (b) Cross sectional, colorized SEM image of hematite-coated inverse-opal scaffold. The distinct structure of hematite demonstrates conformal coating.

visible light. Since FTO and ATO are both variations of SnO2, the FTO/ATO and ATO/TiO2 interface are expected to be of good quality. The conductivity of ATO was reported to reach a maximum when Sb/(Sb+Sn)=0.07, which was the ratio used here. The sheet resistance of a 250 nm thick ATO film was 15x higher than FTO of similar thickness (15 ohms/sq), but more than 30x less than undoped SnO2 prepared by the same sol gel method. For these reasons, ATO provided a straightforward pathway to extend the work on planar films described above to nanostructured

scaffolds.16,18,19

Planar

ATO

films

were

inserted

to

form

an

Fe2O3/TiO2/ATO/FTO stack that was tested for PEC performance. For such stacks with a 50 nm ATO film, 60 SILAR cycles of Fe2O3, and 775 °C anneal, the LSVs under 1 sun illumination 29 ACS Paragon Plus Environment


were almost identical to control samples without ATO, as shown in Figure 12.

This test

indicates that ATO is an appropriate scaffold candidate with sufficient conductivity and interface quality. The 3D ATO scaffold increases the photocurrent compared to planar films due to improved light harvesting without sacrificing charge collection efficiency, as shown in Figure 12. Photocurrent increased by 78%, to 0.40 mAcm-2, at 1.23 VRHE. This improved current density is in good agreement with simple geometric arguments, which predict an 85% increase in surface area for a close-packed monolayer of spheres whose pores are filled to half their height. Similar improvements were also seen with thicker hematite coatings, as also shown in Figure 12 for 120-cycle coatings. The improvement in PEC performance with our solution-processed 3D ATO scaffold is consistent with that observed by Riha et al. for an ITO inverse-opal prepared by atomic layer deposition, although the magnitude of their plateau photocurrent was larger due to

-2

Current Density (mAcm )

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 30 of 55

Potential (V vs RHE) Figure 12. LSV curves of the 3D inverse opal Fe2O3/TiO2/ATO/FTO (black, solid), 2D planar Fe2O3/TiO2/ATO/FTO (black, dashed), and planar Fe2O3/TiO2/FTO (gray, dashed) structures under 1-sun illumination.


Page 31 of 55

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


use of an 8-layer inverse opal.57 We expect that further increases in PEC performance will arise from the design of scaffolds with increased surface area, as well as application of surface treatments to improve OER efficiency.

Conclusions Three key contributions to furthering PEC water splitting using hematite photoanodes were reported. First, a new FeCl3-containing recipe for hematite thin film fabrication by SILAR was presented. SILAR with FeCl3 allows deposition of uniform hematite films with wellcontrolled thickness. Second, TiO2 underlayers and different annealing temperatures were found to affect PEC performance through their influence on the Ti and Sn concentration profiles throughout the film. Their most significant role is passivation of surface recombination to enable more efficient interfacial hole transfer, and hence OER, at the hematite-liquid junction. OER was not measurable when no Ti or Sn was at the surface and increased with the presence of some dopants at the hematite surface. However, in some cases excessive quantities of dopant inhibited OER and reduced photocurrent.

Third, the use of ATO inverse opal scaffolds to support

extremely thin hematite films increased photocurrent in proportion to the increase in surface area. The small local coating thickness maintains charge separation efficiency, while the hostguest architecture enables enhanced light absorption through nanostructuring. These studies provide guidance for future work in designing dopant composition and nanostructured architectures for efficient PEC water splitting with hematite.

Acknowledgments



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 32 of 55

This work was supported by JBB’s NSF CAREER Award CBET-0846464 and NSF CMMI-1000111 and made use of Drexel University’s Core Facilities. IGT was supported by a European Erasmus Mundus programme MESC scholarship. AJA and AMP were partially supported by the Drexel STAR Scholars Program.

Supporting Information Available: XPS survey spectra before and after PEC testing for 120 SILAR cycle Ti:Fe2O3 annealed at 775 °C. Spectrally-resolved LHE, EQE and IQE for 120 SILAR cycle Ti:Fe2O3 annealed at 775 °C. LSV curves under front- and back-side illumination for 60 and 120 SILAR cycle Ti:Fe2O3 samples annealed at 600 °C. LSV curves for all sample variations with 1 M NaOH – 0.5 M H2O2 electrolyte. Dark curves for all sample variations. This information is available free of charge via the Internet at http://pubs.acs.org.


Page 33 of 55

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


References 1.

Fujishima, A.; Honda, K. Electrochemical Photolysis of Water at a Semiconductor Electrode. Nature 1972, 238, 37-38.

2.

Lin, Y.; Yuan, G.; Sheehan, S.; Zhou, S.; Wang, D. Hematite-Based Solar Water Splitting: Challenges and Opportunities. Energy Environ. Sci. 2011, 4, 4862.

3.

Murphy, A. B.; Barnesa, P. R. F.; Randeniyaa, L. K.; Plumba, I. C.; Grey;, I. E.; Horne, M. D.; Glasscock, J. A. Efficiency of Solar Water Splitting Using Semiconductor Electrodes. Int. J. Hydrogen Energy 2006, 31, 1999-2017.

4.

Sivula, K.; Le Formal, F.; Gratzel, M. Solar Water Splitting: Progress Using Hematite (Alpha-Fe2O3) Photoelectrodes. ChemSusChem 2011, 4, 432-449.

5.

Bassi, P. S.; Gurudayal; Wong, L. H.; Barber, J. Iron Based Photoanodes for Solar Fuel Production. Phys. Chem. Chem. Phys. 2014, 16, 11834-11842.

6.

Liao, P.; Toroker, M. C.; Carter, E. A. Electron Transport in Pure and Doped Hematite. Nano Lett. 2011, 11, 1775-1781.

7.

Zandi, O.; Klahr, B. M.; Hamann, T. W. Highly Photoactive Ti-Doped α-Fe2O3 Thin Film Electrodes: Resurrection of the Dead Layer. Energy Environ. Sci. 2013, 6, 634.

8.

Miao, C.; Shi, T.; Xu, G.; Ji, S.; Ye, C. Photocurrent Enhancement for Ti-Doped Fe2O3 Thin Film Photoanodes by an In Situ Solid-State Reaction Method. ACS Appl. Mater. Interfaces 2013, 5, 1310-1316.

9.

Morrish, R.; Rahman, M.; MacElroy, J. M. D.; Wolden, C. A. Activation of Hematite Nanorod Arrays for Photoelectrochemical Water Splitting. ChemSusChem 2011, 4, 474-479.



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 34 of 55

10. Bohn, C. D.; Agrawal, A. K.; Walter, E. C.; Vaudin, M. D.; Herzing, A. A.; Haney, P. M.; Talin, A. A.; Szalai, V. A. Effect of Tin Doping on α-Fe2O3 Photoanodes for Water Splitting. J. Phys. Chem. C 2012, 116, 15290-15296. 11. Kay, A.; Cesar, I.; Gratzel, M. New Benchmark for Water Photooxidation by Nanostructured α-Fe2O3 Films. J. Amer. Chem. Soc. 2006, 128, 15714-15721. 12. Cesar, I.; Sivula, K.; Kay, A.; Zboril, R.; Graetzel, M. Influence of Feature Size, Film Thickness, and Silicon Doping on the Performance of Nanostructured Hematite Photoanodes for Solar Water Splitting. J. Phys. Chem. C 2009, 113, 772-782. 13. Sivula, K.; Le Formal, F.; Gratzel, M. WO3-Fe2O3 Photoanodes for Water Splitting: A Host Scaffold, Guest Absorber Approach. Chem. Mater. 2009, 21, 2862-2867. 14. Stefik, M.; Cornuz, M.; Mathews, N.; Hisatomi, T.; Mhaisalkar, S.; Gratzel, M. Transparent, Conducting Nb:SnO2 for Host-Guest Photoelectrochemistry. Nano Lett. 2012, 12, 54315435. 15. Hisatomi, T.; Brillet, J.; Cornuz, M.; Le Formal, F.; Tetreault, N.; Sivula, K.; Gratzel, M. A Ga2O3 Underlayer as an Isomorphic Template for Ultrathin Hematite Films toward Efficient Photoelectrochemical Water Splitting. Faraday Discuss. 2012, 155, 223-232. 16. Moir, J.; Soheilnia, N.; O'Brien, P.; Jelle, A.; Grozea, C. M.; Faulkner, D.; Helander, M. G.; Ozin, G.A. Enhanced Hematite Water Electrolysis Using a 3d Antimony-Doped Tin Oxide Electrode. ACS Nano 2013, 7, 4261-4274. 17. Qiu, Y.; Leung, S. F.; Zhang, Q.; Hua, B.; Lin, Q.; Wei, Z.; Tsui, K. H.; Zhang, Y.; Yang, S.; Fan, Z. Efficient Photoelectrochemical Water Splitting with Ultrathin Films of Hematite on Three-Dimensional Nanophotonic Structures. Nano Lett. 2014, 14, 2123-2129.


Page 35 of 55

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


18. Wang, L.; Palacios-Padros, A.; Kirchgeorg, R.; Tighineanu, A.; Schmuki, P. Enhanced Photoelectrochemical Water Splitting Efficiency of a Hematite-Ordered Sb:SnO2 HostGuest System. ChemSusChem 2014, 7, 421-424. 19. Sun, Y.; Chemelewski, W. D.; Berglund, S. P.; Li, C.; He, H.; Shi, G.; Mullins, C. B. Antimony-Doped Tin Oxide Nanorods as a Transparent Conducting Electrode for Enhancing Photoelectrochemical Oxidation of Water by Hematite. ACS Appl. Mater. Interfaces 2014, 6, 5494-5499. 20. Han, H. S.; Shin, S.; Noh, J. H.; Cho, I. S.; Hong, K. S. Heterojunction Fe2O3-SnO2 Nanostructured Photoanode for Efficient Photoelectrochemical Water Splitting. 2014, 66, 664-669. 21. Yu, J.; Zhao, X.; Zhao, Q. Effect of Surface Structure on Photocatalytic Activity of TiO2 Thin Films Prepared by Sol-Gel Method. Thin Sol. Films 2000, 379, 7-14. 22. Zhang, H.; Wu, J.; Zhou, L.; Zhang, D.; Qi, L. Facile Synthesis of Monodisperse Microspheres and Gigantic Hollow Shells of Mesoporous Silica in Mixed Water-Ethanol Solvents. Langmuir 2007, 23, 1107-1113. 23. Wojdyr, M. Fityk: A General-Purpose Peak Fitting Program. J. Appl. Cryst. 2010, 43, 11261128. 24. Kulal, P. M.; Dubal, D. P.; Lokhande, C. D.; Fulari, V. J. Chemical Synthesis of Fe2O3 Thin Films for Supercapacitor Application. J. Alloy. Comp. 2011, 509, 2567-2571. 25. Niesen, T. P.; De Guire, M. R. Review: Deposition of Ceramic Thin Films at Low Temperatures from Aqueous Solutions. J. Electroceram. 2001, 6, 169-207. 26. Lokhande, C. D.; Dubal, D. P.; Joo, O.-S. Metal Oxide Thin Film Based Supercapacitors. Curr. Appl. Phys. 2011, 11, 255-270.



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 36 of 55

27. Cristino, V.; Berardi, S.; Caramori, S.; Argazzi, R.; Carli, S.; Meda, L.; Tacca, A.; Bignozzi, C. A. Efficient Solar Water Oxidation Using Photovoltaic Devices Functionalized with Earth-Abundant Oxygen Evolving Catalysts. Phys. Chem. Chem. Phys. 2013, 15, 1308313092. 28. Carbonare, N. D.; Cristino, V.; Berardi, S.; Carli, S.; Argazzi, R.; Caramori, S.; Meda, L.; Tacca, A.; Bignozzi, C. A. Hematite Photoanodes Modified with an Fe(III) Water Oxidation Catalyst. ChemPhysChem 2014, 15, 1164-1174. 29. Martinson, A. B. F.; DeVries, M. J.; Libera, J. A.; Christensen, S. T.; Hupp, J. T.; Pellin, M. J.; Elam, J. W. Atomic Layer Deposition of Fe2O3 Using Ferrocene and Ozone. J. Phys. Chem. C 2011, 115, 4333-4339. 30. Marusak, L. A.; Messier, R.; White, W. B. Optical Absorption Spectrum of Hematite, αFe2O3 near IR to UV. J. Phys. Chem. Solids 1980, 41, 981-984. 31. Warren, B. E., X-Ray Diffraction; Dover Publications, 1990. 32. McIntyre, N. S.; Zetaruk, D. G. X-Ray Photoelectron Spectroscopic Studies of Iron Oxides. Anal. Chem. 1977, 49, 1521-1529. 33. Grosvenor, A. P.; Kobe, B. A.; Biesinger, M. C.; McIntyre, N. S. Investigation of Multiplet Splitting of Fe 2p XPS Spectra and Bonding in Iron Compounds. Surf. Inter. Anal. 2004, 36, 1564-1574. 34. Biesinger, M. C.; Payne, B. P.; Grosvenor, A. P.; Lau, L. W. M.; Gerson, A. R.; Smart, R. S. C. Resolving Surface Chemical States in XPS Analysis of First Row Transition Metals, Oxides and Hydroxides: Cr, Mn, Fe, Co and Ni. Appl. Surf. Sci. 2011, 257, 2717-2730. 35. Kronawitter, C. X.; Zegkinoglou, I.; Shen, S. H.; Liao, P.; Cho, I. S.; Zandi, O.; Liu, Y. S.; Lashgari, K.; Westin, G.; Guo, J. H., et al. Titanium Incorporation into Hematite


Page 37 of 55

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


Photoelectrodes: Theoretical Considerations and Experimental Observations. Energy Environ. Sci. 2014, 7, 3100-3121. 36. Briggs, D.; Seah, M. P. Auger and X-Ray Photoelectron Spectroscopy: Practical Surface Analysis; Wiley: New York, 1990; Vol. 1. 37. Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D. Handbook of X-Ray Photoelectron Spectroscopy; Physical Electronics Division, Perkin-Elmer Corporation: Eden Praire, MN, 1992. 38. Wang, G. M.; Ling, Y. C.; Wheeler, D. A.; George, K. E. N.; Horsley, K.; Heske, C.; Zhang, J. Z.; Li, Y. Facile Synthesis of Highly Photoactive α-Fe2O3-Based Films for Water Oxidation. Nano Lett. 2011, 11, 3503-3509. 39. Ling, Y. C.; Wang, G. M.; Wheeler, D. A.; Zhang, J. Z.; Li, Y. Sn-Doped Hematite Nanostructures for Photoelectrochemical Water Splitting. Nano Lett. 2011, 11, 2119-2125. 40. Hisatomi, T.; Dotan, H.; Stefik, M.; Sivula, K.; Rothschild, A.; Gratzel, M.; Mathews, N. Enhancement in the Performance of Ultrathin Hematite Photoanode for Water Splitting by an Oxide Underlayer. Adv. Mater. 2012, 24, 2699-2702. 41. Kment, S.; Hubicka, Z.; Krysa, J.; Olejnicek, J.; Cada, M.; Gregora, I.; Zlamal, M.; Brunclikova, M.; Remes, Z.; Liu, N., et al. High-Power Pulsed Plasma Deposition of Hematite Photoanode for Pec Water Splitting. Catal. Today 2014, 230, 8-14. 42. Klahr, B. M.; Hamann, T. W. Water Oxidation on Hematite Photoelectrodes: Insight into the Nature of Surface States through In Situ Spectroelectrochemistry. J. Phys. Chem. C 2014, 118, 10393-10399.



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 38 of 55

43. Rioult, M.; Magnan, H.; Stanescu, D.; Barbier, A. Single Crystalline Hematite Films for Solar Water Splitting: Ti-Doping and Thickness Effects. J. Phys. Chem. C 2014, 118, 30073014. 44. Yang, X.; Liu, R.; Du, C.; Dai, P.; Zheng, Z.; Wang, D. Improving Hematite-Based Photoelectrochemical Water Splitting with Ultrathin TiO2 by Atomic Layer Deposition. ACS Appl. Mater. Interfaces 2014, 6, 12005-12011. 45. Zandi, O.; Hamann, T. W. Enhanced Water Splitting Efficiency through Selective Surface State Removal. J. Phys. Chem. Lett. 2014, 5, 1522-1526. 46. Zandi, O.; Beardslee, J. A.; Hamann, T. W. Substrate Dependent Water Splitting with Ultrathin α-Fe2O3 Electrodes. J. Phys. Chem. C 2014, 118, 16494-16503. 47. Dotan, H.; Sivula, K.; Gratzel, M.; Rothschild, A.; Warren, S. C. Probing the Photoelectrochemical Properties of Hematite (α-Fe2O3) Electrodes Using Hydrogen Peroxide as a Hole Scavenger. Energy Environ. Sci. 2011, 4, 958-964. 48. Abdi, F. F.; Han, L.; Smets, A. H. M.; Zeman, M.; Dam, B.; Van de Krol, R. Efficient Solar Water Splitting by Enhanced Charge Separation in a Bismuth Vanadate-Silicon Tandem Photoelectrode. Nature Comm. 2013, 4. 49. Han, J.; Zong, X.; Wang, Z.; Li, C. A Hematite Photoanode with Gradient Structure Shows an Unprecedentedly Low Onset Potential for Photoelectrochemical Water Oxidation. Phys. Chem. Chem. Phys. 2014, 16, 23544 - 23548. 50. Rao, P. M.; Cai, L.; Liu, C.; Cho, I. S.; Lee, C. H.; Weisse, J. M.; Yang, P.; Zheng, X. Simultaneously Efficient Light Absorption and Charge Separation in WO3/BiVO4 Core/Shell Nanowire Photoanode for Photoelectrochemical Water Oxidation. Nano Lett. 2014, 14, 1099-1105.


Page 39 of 55

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


51. Steier, L.; Herraiz-Cardona, I.; Gimenez, S.; Fabregat-Santiago, F.; Bisquert, J.; Tilley, S. D.; Gratzel, M. Understanding the Role of Underlayers and Overlayer in Thin Film Hematite Photoanodes. Adv. Funct. Mater. 2014, 24, 7681-7688. 52. Pendlebury, S. R.; Barroso, M.; Cowan, A. J.; Sivula, K.; Tang, J. W.; Gratzel, M.; Klug, D.; Durrant, J. R. Dynamics of Photogenerated Holes in Nanocrystalline α-Fe2O3 Electrodes for Water Oxidation Probed by Transient Absorption Spectroscopy. Chem. Commun. 2011, 47, 716-718. 53. Tang, H.; Matin, M. A.; Wang, H.; Deutsch, T.; Al-Jassim, M.; Turner, J.; Yan, Y. Synthesis

and

Characterization

of

Titanium-Alloyed

Hematite

Thin

Films

for

Photoelectrochemical Water Splitting. J. Appl. Phys. 2011, 110. 54. Kennedy, J. H.; Frese Jr., K. W. Photooxidation of Water at α-Fe2O3 Electrodesq. J. Electrochem. Soc. 1978, 125, 709-714. 55. Klahr, B. M.; Gimenez, S.; Fabregat-Santiago, F.; Bisquert, J.; Hamann, T. W. Electrochemical and Photoelectrochemical Investigation of Water Oxidation with Hematite Electrodes. Energy Environ. Sci. 2012, 5, 7626-7636. 56. Klahr, B. M.; Gimenez, S.; Fabregat-Santiago, F.; Hamann, T. W.; Bisquert, J. Water Oxidation at Hematite Photoelectrodes: The Role of Surface States. J. Amer. Chem. Soc. 2012, 134, 4294-4302. 57. Riha, S. C.; Klahr, B. M.; Tyo, E. C.; Seifert, S.; Vajda, S.; Pellin, M. J.; Hamann, T. W.; Martinson, A. B. F. Atomic Layer Deposition of a Submonolayer Catalyst for the Enhanced Photoelectrochemical Performance of Water Oxidation with Hematite. ACS Nano 2013, 7, 2396-2405.



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 40 of 55

58. Williams, V. O.; DeMarco, E. J.; Katz, M. J.; Libera, J. A.; Riha, S. C.; Kim, D. W.; Avila, J. R.; Martinson, A. B. F.; Elam, J. W.; Pellin, M. J., et al. Fabrication of TransparentConducting-Oxide-Coated

Inverse

Opals

as

Mesostructured

Architectures

for

Electrocatalysis Applications: A Case Study with NiO. ACS Appl. Mater. Interfaces 2014, 6, 12290-12294. 59. Majidi, H.; Edley, M. E.; Spangler, L. C.; Baxter, J. B. Tailoring Absorber Thickness and the Absorber-Scaffold Interface in CdSe-Coated ZnO Nanowire Extremely Thin Absorber Solar Cells. Electrochim. Acta 2014, 145, 291-299.


Page 41 of 55

TOC Figure H2O

O2

Photocurrent

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


600 C 775 C

Ti

3D


Photocurrent

O255 he JournalHof Page 42Chemist of 2OPhysical

1 2 3 4 5 6 Paragon Plus Environmen ACS 7 8 3D 9 600 C 775 C Ti

Hematite a) The Page 43 Journal of 55 of Physical Chemistry 1 FTO 2 3 4 5 6 7 8 9 10 11 12 13 b) 14 15 16 17 18 19 20 21 22 23 ACS Paragon Plus Environment 24 25 26

a)

The Journal of Physical Chemistry Page 44 of 55 30 60 90 120 150 180 210

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

ACS Paragon Plus Environment

Page The45Journal of 55 of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 ACS Paragon Plus Environment 22 23 24

The Journal of Physical Chemistry Page 46 of 55

1 2 3 4 5 6 7 8 9 10 11 12


Page The 47 Journal of 55 of Physical Chemistry

kcps

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 60 24 c) 25 26 27 40 28 29 16 min 30 31 20 4 min 32 33 Surface 34 Sn 3d Ti 2p 350 ACS Paragon Plus Environment 36510 500 490 480 470 460 450 440 Binding Energy (eV) 37

a)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 b)

15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

The Journal of Physical Chemistry Page 48 of 55 210 180 150 120 90 60 30 cycles


Page The49 Journal of 55 of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 ACS Paragon Plus Environment 25 26

Separation (%)

a)

1 2 3 4 5 6 7 8 9 b)

10 11 12 13 14 15 16 17 18 19 20 21 22

Fe2O3 600 °C

The Journal of Physical Chemistry Ti:Fe2O3 600 °C

Fe2O3 775 °C



Page 50 of 55 Ti:Fe2O3 775 °C

a)

Page 51 of 55 1 2 3 4 5 6 7 8 b) 9

10 11 12 13 14 15 16 17 18 19 20 21 22

The °C

Journal of Physical Chemistry Fe2O3 775 °C

Ti:Fe2O3 600


Ti:Fe2O3 775 °C

The Journal of Physical Chemistry Page 52 of 55

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

c)

Cbulk Rs

R

ct,ss ACS Paragon Plus Environment

Rtrap

Css

Page 53 of 55 Fe2O3, 600 °C Ti:Fe 1 2O3, 600 °C 2 Fe2O3, 775 °C 3 Ti:Fe 2O3, 775 °C

Vfb N Rs Rtrap Rct,ss The JournalDof Physical Chemistry (V) (1019 cm-3) (Ω) (kΩ) (kΩ) 28.6 3.7 280.0 - Paragon 2.8 2.4 ACS Plus26.6 Environment 0.49 1.3 50.6 1.1 1.0 0.40 4.7 43.0 0.8 0.3

Cbulk (µF) 2.3 3.1 2.0 3.8

Css (µF) 7.4 10.3 16.5 28.7

a) The Journal of Physical Chemistry Page 54 of 55 1 2 3 4 5 6 7 8500 nm 9 10 11 b) 12 13 14 15 16 17 18 19 ACS Paragon Plus Environment 20 200 nm 21

Current Density (mAcm-2)

PageThe 55 of Journal 55 of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12



SILAR-Deposited Hematite Films for Photoelectrochemical Water

Recommend Documents