- Email: [email protected]
Morphology-controlled α-Fe2O3 nanostructures on FTO substrates for photoelectrochemical water oxidation
Accepted Manuscript Morphology-controlled α-Fe2O3 nanostructures on FTO substrates for photoelectrochemical water oxidation Qinglong Liu, Changlong Chen, Guangzheng Yuan, Xing Huang, Xiaoxiao Lv, Yi Cao, Yaping Li, Anjie Hu, Xiangrong Lu, Peihua Zhu PII:
S0925-8388(17)31408-1
DOI:
10.1016/j.jallcom.2017.04.213
Reference:
JALCOM 41617
To appear in:
Journal of Alloys and Compounds
Received Date: 8 March 2017 Revised Date:
9 April 2017
Accepted Date: 11 April 2017
Please cite this article as: Q. Liu, C. Chen, G. Yuan, X. Huang, X. Lv, Y. Cao, Y. Li, A. Hu, X. Lu, P. Zhu, Morphology-controlled α-Fe2O3 nanostructures on FTO substrates for photoelectrochemical water oxidation, Journal of Alloys and Compounds (2017), doi: 10.1016/j.jallcom.2017.04.213. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Morphology-Controlled α-Fe2O3 Nanostructures on FTO Substrates for
Photoelectrochemical Water Oxidation Qinglong Liu, Changlong Chen,∗ Guangzheng Yuan, Xing Huang, Xiaoxiao Lv, Yi Cao, Yaping Li, Anjie Hu, Xiangrong Lu, Peihua Zhu Key Laboratory of Chemical Sensing & Analysis in Universities of Shandong, School of Chemistry and Chemical Engineering,
RI PT
University of Jinan, Jinan, 250022, Shandong, P. R. China.
Abstract: Hematite (α-Fe2O3) nanostructured films were prepared on fluorine doped tin oxide substrates by heat treating the hydrothermally prepared precursor films. By using different additives in
SC
the hydrothermal preparations, the final hematite nanostructures could be controlled to be square prisms, nanorods and polyhedral blocks, of which the square prisms and the nanorods were vertically
M AN U
grown on the substrates and possessed preferred orientation along [110] direction while the polyhedral blocks were composed of small isotropic nanoparticles. When used as photoanodes, the α-Fe2O3 nanorod films showed better photoelectrochemical water oxidation performance than the other two kinds of α-Fe2O3 films. Analyses based on XRD, SEM and electrochemical impedance spectroscopy revealed that the α-Fe2O3 nanorods had characteristics including small diameter, abundant
TE D
through-holes, growing along [110] axis vertical to the substrate and high donor density, which are believed to contribute to the good photoelectrochemical water oxidation performance. On the ground that α-Fe2O3 is very promising semiconductor material for photoelectrochemical water oxidation application, such morphology controlled, nanostructured hematite α-Fe2O3 films based on simple
EP
preparation method are envisioned to provide valuable platforms for supporting catalysts and co-catalysts to achieve efficient light-assisted oxidation of water.
AC C
Keywords: photoelectrochemical; water oxidation/splitting; photoanode; hematite; α-Fe2O3; morphology control 1. Introduction
The decreasing fossil energy and growing energy demand are becoming a global problem and has compelled scientists to focus on exploring new energy sources.[1] Hydrogen achieved by solar light-driven photoelectrochemical (PEC) water splitting is a kind of renewable, clean and environmentally friendly energy, which is expected to be one effective solution to the energy problem. ∗
Corresponding author. Tel: +86 531 82765475; Fax: +86 531 89631632. E-mail addresses: [email protected]. 1
Since the first report on PEC waterACCEPTED splitting with MANUSCRIPT TiO2 as photoanode by Fujishima in 1972,[2] lots of researches have been focused on this issue, of which many oxides, such as TiO2,[3,4] WO3,[5,6] BiVO4,[7,8] ZnO,[9] In2O3,[10] etc., were studied as potential photoanode materials. Generally speaking, the excellent photoanode material should meet such demands: proper band gap energy, strong visible light absorption, suitable band edge positions for oxidation of water, fast transport of the
environment and environmentally friendly, etc.[11,12]
RI PT
photo-generated electrons and holes in the semiconductor, high chemical stability in aqueous
As an n type semiconductor material, hematite (α-Fe2O3) has the suitable band gap (Eg= ~2.2 eV) and valence band edge position for solar light-driven PEC water oxidation.[13] Its theoretical
SC
solar-to-hydrogen (STH) conversion efficiency is 16.8%,[14] which is pretty high. Moreover, hematite also has other advantages such as excellent chemical stability in aqueous environment, being
M AN U
environmentally friendly, abundance and low cost,[15] which make it a very promising photoanode material for PEC water splitting. On the other hand, some inherent characteristics of hematite, including short life time for charge carriers (~10 ps), short hole diffusion length (2-4 nm), low electric conductivity, low water oxidation evolution kinetics and high charge recombination, hinder it to achieve high PEC water oxidation performance.[13,16] In order to improve the PEC water oxidation
TE D
performance of hematite, many attempts have been done, which include element doping,[17,18] fabricating nanostructures,[19] surface modification,[20,21,22] and so on. Among them, fabricating nanostructured α-Fe2O3 is especially popular because it can not only significantly increase its specific surface area but also decrease the diffusion length of the photo-excited charge carriers.[10] For instance,
EP
α-Fe2O3 nanostructures like nanosheet arrays,[23] nanoparticle-chains,[24] nanorod arrays,[18,25] and
AC C
tree-like structures[26] have been reported to exhibit much enhanced PEC water splitting performance as compared to their corresponding bulk materials. Many methods have been used to prepare nanostructured semiconductor materials exemplified by atmospheric pressure chemical vapor deposition (APVCD),[27] pyrolysis,[14] electrochemical deposition[28] and hydrothermal methods,[29,30] of which the hydrothermal method is especially popular due to its merits of simplicity, low cost and low equipment requirement. In this work, we prepare nanostructured α-Fe2O3 films on fluorine doped tin oxide (FTO) substrates by heat treating the hydrothermally grown precursor films. By changing the additives used in the growth solution, it is realized to control the morphologies of the hematite nanostructures to be square prism arrays, nanorod arrays and polyhedral blocks. It was found that the α-Fe2O3 nanorod arrays contained several 2
ACCEPTED MANUSCRIPT advantages including small diameter, abundant through-holes, growing along [110] axis vertical to the substrate and high donor density, which make it possess obviously enhanced PEC water oxidation performance in comparison with the other two kinds of α-Fe2O3 nanostructures. 2. Experimental Section 2.1 Materials
RI PT
Ferric(III) Chloride (FeCl3·6H2O), Sodium Fluoride (NaF) and concentrated sulfuric acid were purchased from Sinopharm Chemical Reagent Co. Sodium Nitrate (NaNO3) and Sodium Sulfate (Na2SO4) were purchased from Tianjin Dengke Chemical Reagent Co. Anhydrous ethanol, hydrogen peroxide (30%) and acetone were purchased from Tianjin Fuyu Fine Chemical Co. All materials were
SC
used without further purification. 2.2 Synthesis
M AN U
FTO substrates (Zhuhai Kaivo Optoelectronic Technology Co.) with dimensions of 25 mm × 20 mm × 1.6 mm were cleaned ultrasonically in a solution containing acetone and ethanol (1:1 in volume) followed by rinsing in deionized water. After that, they were soaked in piranha solution for 24 h and then were rinsed with water and dried with air flow. In a typical synthesis, 2.25 mmol of FeCl3·6H2O and a certain amount of additive were co-dissolved in deionized water to form 15 mL of solution. Then
TE D
the solution was transferred into a 20 mL of Teflon-lined autoclave and a piece of FTO substrate was immersed into the solution, keeping the substrate against the wall of the Teflon liner and making its FTO side facing down. To control the morphologies of the α-Fe2O3 precursor nanostructures to be
EP
square prism arrays, nanorod arrays and polyhedral blocks, 4.5 mmol NaF, 5.5 mmol NaNO3 and 5.5 mmol Na2SO4 as additives were introduced to the hydrothermal reaction solutions, respectively. The for 10 h before cooling to room temperature. After the reaction, a layer
AC C
autoclave was heated at 95
of uniform thin film with yellow color was observed on the FTO substrate, which was thoroughly rinsed with deionized water and then was dried at 70 structure, the film was annealed in air at 550
for 2 h. To achieve the final hematite crystal
for 1 h. After cooled to room temperature, the film on
FTO substrate was further rinsed with deionized water and then was heat treated at 550
for 1 h again
but in vacuum environment, which is envisioned to improve its electronic properties. 2.3 Characterization X-ray diffraction (XRD) patterns of the samples were recorded in the range of 2θ = 20-70° on an X-ray diffractometer (Bruker D8 Focus) with Cu Kα1 radiation (λ = 0.15406 nm) at room temperature while the voltage and electric current were held at 40 kV and 30 mA, respectively. The morphologies 3
ACCEPTED MANUSCRIPT of the products were observed by scanning electron microscopy (SEM, FEI Quanta FEG 250). During the SEM measurements, a small piece of sample was adhered onto a copper stub using double-sided carbon tape. 2.4 Photoelectrochemical water oxidation measurements A xenon lamp (AuLight, CEL-S500, 500 W) through an AM 1.5G filter (AuLight, PD 250) was
RI PT
used as a solar simulator to irradiate quartz photoelectrochemical cell. The light intensity at the place of working photoanode was calibrated to be 100 mW·cm-2 by using solar power meter (Thorlabs, PM100 USB) with an S302C detector. The PEC measurements were carried out using an electrochemical workstation (Bio-Logic SAS, SP-150) in typical three electrode configuration: the
SC
prepared α-Fe2O3 films on FTO substrates serve as work electrode, Pt wire as counter electrode and Ag/AgCl (in saturated KCl) as reference electrode. The electrolyte was the aqueous solution of 0.1 M
M AN U
NaOH. During the measurement, the photoanode was swept from 0 to 0.75 V (vs. Ag/AgCl) at a rate of 10 mV·s-1. The measured potential vs. Ag/AgCl was converted to the reversible hydrogen electrode (RHE) scale according to Nernst equation:
ERHE = EAg/AgCl + 0.059 pH + EθAg/AgCl
(1)
where ERHE is the calculated potential versus RHE, EAg/AgCl is the measured potential against the Ag/AgCl reference electrode, EθAg/AgCl = 0.197 V at 25
TE D
.
2.5 Electrochemical impedance spectroscopy (EIS) measurements EIS measurements were performed using the same three electrode configuration describes above.
EP
The measurements were done in darkness at the following conditions: an AC frequency of 1 kHz with amplitude of 10 mV, scanning from -0.75 to +0.65 V vs. Ag/AgCl (0.21-1.61 V vs. RHE), in a 0.1 M
AC C
NaOH solution.
2.6 Incident photon-to-current conversion efficiency (IPCE) measurements The IPCE measurements were performed at 1.60 V vs. RHE with a 300 W Xe lamp equipped with a monochromator (omni-λ 300). The IPCE can be calculated according to the following equation:
IPCE =
(2)
Where I (mA·cm-2) is the measured photocurrent density, λ (nm) is the wavelength of the incident light, Jlight (mW·cm-2) is the measured irradiance at a specific wavelength. 3. Results and discussion Fig. 1a is the optical photograph of the hydrothermally prepared samples. It shows that the films 4
MANUSCRIPT obtained with NaF, NaNO3 and NaACCEPTED 2SO4 as additives are uniform on the FTO substrates and possess yellow color. Generally speaking, such yellow color indicates that the films are not hematite α-Fe2O3, whose color usually is red. To analyze the crystalline structures of these hydrothermally prepared films, X-ray diffraction (XRD) measurements were done. As can be seen from Fig. 2a, except for the diffraction peaks originated from the substrate FTO, the three kinds of samples obtained with different
RI PT
additives all show their own XRD patterns, however, none of them can be ascribed to the α-Fe2O3 phase, indicating the hydrothermal products are intermediates. Specifically, the XRD patterns of the samples prepared with NaF (black) and NaNO3 (red) as additives show the crystal phase that can be indexed to the tetragonal β-FeOOH (JCPDS No. 34-1266), of which the (310) peaks at 2θ = 26.7° can
SC
be observed from the enlarged part of the XRD patterns (Fig. 2b, the black and red patterns). Particularly, for the intermediate obtained with NaF as additive, the (211) diffraction peak also occurs,
M AN U
which is even higher than the (310) peak, much different with the referenced XRD lines, indicating the possibility of existing preferred orientation. The XRD pattern of the intermediate prepared with Na2SO4 as additive is shown in Fig. 2a (the blue one), which can be indexed to rhombohedral NaFe3(SO4)2(OH)6 (JCPDS No. 36-0425). To convert these intermediates to the final hematite α-Fe2O3 phase, they were heat treated at 550°C first in air and then in vacuum. After heat treatment, the color of
TE D
these three kinds of films all changed from yellow to red, as shown in Fig. 1b. The X-ray patterns of these heat treated films are shown in Fig. 2c. As can be seen, ignoring the diffraction peaks from FTO, the XRD patterns of the three samples can be indexed to the hematite α-Fe2O3 (JCPDS No. 33-0664).
EP
Apart from the peaks from α-Fe2O3 and FTO, no other diffraction peaks are found from these heat-treated samples. Importantly, obvious preferred orientation can be observed from the top two
AC C
patterns (the black and the red), which correspond to the films obtained with NaF and NaNO3 as additives, respectively. Specifically, the both patterns only show a strong (110) peak and a weak (300) peak, indicating that the hematite crystals are highly oriented along [110] direction vertical to the FTO substrate.[25] The XRD peaks of the sample obtained with Na2SO4 as additive (Fig. 2c, blue pattern), however, does not show obvious preferred orientation, which match well with the powder XRD lines of the referenced hematite α-Fe2O3 (JCPDS No. 33-0664) both in position and in relative intensity, indicating the fine powder nature of this sample. Morphologies and microstructures of the hydrothermally prepared intermediates and their final heat treated hematite α-Fe2O3 were observed by using SEM. Fig. 3a shows the SEM image of the β-FeOOH obtained with NaF as additive. It can be seen that the film is composed of square prism-like short rods 5
ACCEPTED MANUSCRIPT vertically grown on the FTO substrate. These square prism-like rods have side length for the top surface from 40-110 nm and are distributed uniformly. The cross sectional view image of the film shown in the inset reveals that these square prism-like rods are ~ 250 nm long. After heat treatment, the obtained α-Fe2O3 is substantially maintained the short square prism morphology (Fig. 3b). It is interesting that although the heat treatment does not change the morphology and size of the rods, it
RI PT
leads to large amount of holes. From the inset of Fig. 3b it can be seen, the holes have a diameter of ~ 9 nm and most of them are distributed on the side surfaces, very few on the top surface. These holes are believed to be generated due to the dehydration of the β-FeOOH during the heat treatment according to the equation (3):
(3)
EP
TE D
M AN U
SC
2FeOOH → Fe2O3 + H2O
Fig. 1 Optical photographs of the films on FTO substrates before (a) and after (b) heat treatments.
AC C
Because the width of the square prism-like rods is much smaller than their length, the produced H2O would readily release from the side surfaces rather than from the top surfaces of the rods and thereby forming the abundant small blind holes on the side surfaces.[31] In addition, during the hydrothermal preparation, it is believed that the F- ions are the key factor that results in the square prism-like shape of the rods. According to the theory of hard and soft acids and bases, at the beginning of the hydrothermal reaction, the hard acid, Fe3+ ions, prefer to bind with F- to form FeF2+ complex ions rather than directly hydrolyze.[24] It has been reported that FeF2+ complex ions readily promoted the anisotropic growth of the formed crystals[31] and in this situation, with the reaction temperature increasing, the hydrolysis was initiated and square prism-like β-FeOOH rods formed. 6
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 2 (a) XRD patterns of the hydrothermally prepared precursor films on FTO substrates with additives of NaF (black), NaNO3 (red) and Na2SO4 (blue). The bottom are the reference XRD lines for β-FeOOH (JCPDS No. 34-1266, pink line) and NaFe3(SO4)2(OH)6 (JCPDS No.36-0425, green line ). (b) Same XRD patterns shown in (a) enlarged in the range of 2θ = 25-28°. (c) XRD patterns of the heat treated films on FTO substrates. The bottom is the reference XRD lines for α-Fe2O3 (JCPDS No. 33-0664). 7
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 3 (a) SEM image of the precursor film prepared with NaF as additive; the inset is the cross-sectional view. (b) SEM image of the heat treated film with NaF as additive; the inset is the enlargement of the part in the white frame. (c) SEM image of the precursor film prepared with NaNO3 as additive. (d) SEM image of the heat treated film with NaNO3 as additive; the inset is the enlargement of the part 8
in the white frame. (e) SEM image of the heatACCEPTED treated film with NaMANUSCRIPT 2SO4 as additive. (f) The enlargement of the part in the white frame in (e); the inset is the enlargement of the part in the white frame.
As for the hydrothermal prepared β-FeOOH with NaNO3 as the additive, its SEM image shown in Fig. 3c exhibits that the film is composed of nanorod arrays vertically grown on the FTO substrate. The nanorods are uniformly distributed on the substrate and possess width and length of 20-60 and ~
RI PT
800 nm, respectively. Obviously they are thinner but longer than the β-FeOOH square prisms that obtained with NaF as additive. After heat treatment, the achieved α-Fe2O3 also keeps the nanorod array morphology, as shown in Fig. 3d. Like the α-Fe2O3 square prisms, after heat treatment, the α-Fe2O3 nanorods have no obvious change in size but appear irregular holes. These holes (Fig. 3d, inset),
SC
however, appear to be through-holes and are fewer in number than that for the α-Fe2O3 square prisms (Fig. 3b, inset). It is approximately because the nanorods are much thinner than the square prisms and
M AN U
thus the produced H2O during the heat treatment would much readily escape and thereby making the left space connect together, forming the irregular through-holes.
Fig. 3e and f show the SEM images of the hematite α-Fe2O3 obtained by heat treating the hydrothermal intermediate with Na2SO4 as the additive. The SEM image taken at low magnification (Fig. 3e) shows that the film is composed of large polyhedral blocks. The long and short edges of these
TE D
polyhedral blocks are ~ 17 and ~ 6 µm, respectively. Observing under larger magnification (Fig. 3f), it can be found that the polyhedral blocks actually are composed of nanoparticles with size of 20-100 nm (inset). It has been confirmed by XRD technique in above discussion that the intermediate is
EP
NaFe3(SO4)2(OH)6 and after heat treatment it converts to hematite α-Fe2O3. In this process, although it is not sure what the byproducts are, the theoretical weight loss is 50.52%. Although the profile of the
AC C
polyhedral blocks is retained during the heat treatment, the weight loss is so large that it can be imaged that the internal crystal texture of the intermediate will be destroyed thoroughly during the heat treatment, resulting in the small isotropic nanoparticles rather than keeping the polyhedral blocks just with pores like the other two cases discussed above. This result is well consistent with the XRD patterns (Fig. 2c), where the polycrystalline nature was exhibited for the α-Fe2O3 obtained with Na2SO4 as the additive in the hydrothermal reaction and obvious preferred orientation was revealed for the other two samples that obtained with NaF and NaNO3 as additives, respectively. To evaluate the PEC water oxidation performance of these α-Fe2O3 films prepared on FTO substrates, they were used as photoanodes in typical three electrode configuration to conduct the PEC water splitting under simulated sunlight (100 mW·cm-2). Fig. 4a shows the linear sweep 9
ACCEPTED MANUSCRIPT voltammograms (LSV), in which the current densities of the three kinds of α-Fe2O3 films both in the dark and under illumination are illustrated. As can be seen, without the assistance of any oxygen evolution catalysts (OECs), all the α-Fe2O3 photoanodes exhibit considerable photo-response. In the dark, they show negligible current densities until the applied potential being increased to ~ 1.60 V vs. RHE, where the water electrolysis begins. By comparison, it can be seen that the α-Fe2O3 nanorods
RI PT
film obtained with NaNO3 as additive exhibits the best PEC water oxidation performance. For instance, at the potential of 1.50 V vs. RHE, its photocurrent density is ~ 0.31 mA·cm-2, whereas, the photocurrent densities at the same potential for the α-Fe2O3 polyhedral block film and the α-Fe2O3 square prism film are ~ 0.17 and ~ 0.10 mA·cm-2, respectively. On the other hand, the photocurrent
SC
turn-on potentials for these three kinds of films are also different, which are ~ 0.97, 1.10 and 1.15 V vs.
TE D
M AN U
RHE for the nanorod-, polyhedral block- and square prism-α-Fe2O3 films, respectively.
EP
Fig. 4 (a) LSV collected from α-Fe2O3 nanorod array film (red), polyhedral block film (blue) and square prism film (black) under AM1.5G illumination at 100 mW·cm-2 and in the dark. (b) IPCEs of α-Fe2O3 nanorod array film (red), polyhedral block film (blue) and
AC C
square prism film (black).
Together with the observed fine structures of the α-Fe2O3 films from the SEM images as well as their XRD analysis, it can be deduced that the high PEC water oxidation performance exhibited by the α-Fe2O3 nanorod film is attributed to its two advantages: the high specific surface area due to the small diameter as well as the through-holes; the strong preferential orientation of the [110] axis vertical to the substrate. The former will provide more active sites for the water oxidation reaction, while the latter increases its conductivity. The crystal lattice of hematite α-Fe2O3 can be described as an alternation of iron bilayers and oxygen layers parallel to the (001) basal plane. It has been verified that the electron conductivity is up to 4 orders of magnitude higher within the (001) basal planes (e.g., in [110] direction) than orthogonal to them.[26] Consequently, the α-Fe2O3 nanorod arrays with 10
ACCEPTED preferential orientation of the [110] axis vertical MANUSCRIPT to the substrate is very beneficial for the majority carriers (electrons for n type hematite α-Fe2O3) to transport to the FTO substrate and thereby decreasing the carrier recombination. For the hematite α-Fe2O3 film composed of square prisms, although its XRD pattern also shows preferential orientation of the [110] axis vertical to the substrate and at the same time there are many holes existed, its PEC water oxidation performance is not as good
RI PT
as that of the α-Fe2O3 nanorod array film. On the one hand, the width of the α-Fe2O3 square prisms is much larger than that of the α-Fe2O3 nanorods and on the other hand, the holes on the square prisms are blind, which would block the electrolyte to fill into the holes. The two reasons make the effective specific surface area of the α-Fe2O3 square prism film being much smaller than that of the α-Fe2O3
SC
nanorod film, which would partially offset the advantage of the preferred orientation of the crystals. As for the α-Fe2O3 polyhedral block film, although the polyhedral blocks don’t possess preferred
M AN U
orientation and at the same time they have much larger apparent size than the α-Fe2O3 square prisms, the nature that they are composed of small nanoparticles leads to their specific surface area being larger than that of the latter, which probably results in the slightly higher photocurrent density of the former.
To further investigate the photoresponse of the α-Fe2O3 films, incident photon to charge conversion
TE D
efficiency (IPCE) was measured. IPCE is usually more accurate in evaluating the photoactivity of materials because it is based on monochromatic light and is independent of the lamps used in different laboratories. As shown in Fig. 4b, the three kinds of films all exhibit photoactivity to the light in range
EP
of 325-580 nm. As a whole, the IPCEs for these three kinds of films increase in this order: square prism film < polyhedral block film < nanorod film, which is well in accordance with the PEC results
AC C
shown in Fig. 4a. For example, although to the 340 nm light the polyhedral block film shows comparable IPCE to that of the nanorod film, which is ~4.8%, to the light of 350-580 nm the IPCE of nanorod film is much higher than that of the polyhedral block film. It is believed that in addition to the size and crystalline orientation, other factor like electronic properties of the hematite α-Fe2O3 films might also affect the PEC water oxidation behavior. In order to investigate the electronic properties of the α-Fe2O3 films, electrochemical impedance spectroscopy (EIS) measurements were conducted. Fig. 5 shows the Mott-Schottky plot of the three kinds of α-Fe2O3 films drawn based on the EIS measurements. As can be seen, they all exhibit positive slopes under increased anodic bias, conforming to the behavior of n-type semiconductors. By comparison, it can be seen that the slopes of Mott-Schottky plots for the three kinds of samples 11
MANUSCRIPT increase in this order: nanorods < ACCEPTED polyhedral blocks < square prisms, meaning their donor densities decreasing in that order. The donor densities can be calculated from these slopes according to equation (4): = −
⁄ !
"
#
(4)
where Nd is the donor density (in cm-3), e is the elementary charge (1.60×10-19 C), ε is the relative
RI PT
dielectric constant of α-Fe2O3 (taken to be 80),[23] ε0 is the permittivity of free space (8.854×10-14 F·cm-1), C is the capacitance (in F·cm-2), V is the applied potential (vs. RHE). After calculation, it was found that the donor densities for the α-Fe2O3 films composed of nanorods, polyhedral blocks and square prisms were 2.83×1020, 1.79×1020 and 1.41×1020 cm-3, respectively. Clearly, among these
SC
samples, the nanorod array film possesses the highest donor density. Such results indicate that the charge transport in the α-Fe2O3 nanorod film will be faster than that in the α-Fe2O3 polyhedral block
M AN U
film and square prisms film. This result together with the particle size, the holes and the crystalline orientation may explain the PEC water oxidation performance of the three kinds of α-Fe2O3 films. In other words, the α-Fe2O3 nanorod arrays have advantages including small diameter, abundant through-holes, growing along [110] axis vertical to the substrate and highest donor density, which lead to the best PEC water oxidation behavior, whereas, for the other two kinds of α-Fe2O3 films, they are
TE D
either low in donor density or without the characteristic of preferred crystalline orientation and consequently their PEC water oxidation performance is not as good as that of the α-Fe2O3 nanorod
AC C
EP
arrays.
Fig. 5 Mott-Schottky plot collected at a frequency of 1 kHz in the dark for α-Fe2O3 nanorod array film (red), polyhedral block film (blue) and square prism film (black).
4. Conclusions: 12
MANUSCRIPT In summary, hematite α-Fe2O3ACCEPTED films on FTO substrates were prepared by heat treating the hydrothermally grown precursor films in air and vacuum. The nanostructures that compose the α-Fe2O3 films can be controlled to be nanorod arrays, square prism arrays and polyhedral blocks, of which the former two possess the characteristic of preferential orientation of the [110] axis vertical to the substrates while the polyhedral blocks are composed of small isotropic nanoparticles. Using these
RI PT
α-Fe2O3 films as photoanodes, PEC water splitting measurements were conducted, which revealed that the α-Fe2O3 nanorod array film exhibited the best PEC water oxidation performance. The several merits possessed by the α-Fe2O3 nanorod arrays, including small diameter, abundant through-holes, growing along [110] axis vertical to the substrate and high donor density, contribute to the good PEC
SC
performance. Such morphology controlled, nanostructured hematite α-Fe2O3 films are envisioned to achieve efficient light-assisted oxidation of water by further modification with oxygen evolution
M AN U
catalysts like Co-Pi, the subject of ongoing investigations in our laboratory. Acknowledgements
This work is financially supported by the Research Fund of University of Jinan (No. XKY1603) and the National Innovation and Entrepreneurship Training Program for Undergraduate Students in Local Colleges and Universities of China (No. 201610427011).
TE D
References
[1] M. Mishra, D.-M. Chun, Appl. Catal. A: Gen., 498 (2015) 126-141. [2] A. Fujishima, K. Honda, Nature, 238 (1972) 37-38.
EP
[3] D. Regonini, A.C. Teloeken, A.K. Alves, F.A. Berutti, K. Gajda-Schrantz, C.P. Bergmann, T. Graule, F. Clemens, ACS Appl. Mater. Interfaces, 5 (2013) 11747-11755.
AC C
[4] L. Liu, J. Hensel, R.C. Fitzmorris, Y. Li, J. Z. Zhang, J. Phys. Chem. Lett., 1 (2010) 155-160. [5] S.S. Kalanur, Y.J. Hwang, S.Y. Chae, O.S. Joo, J. Mater. Chem. A, 1 (2013) 3479-3488. [6] F. Ronconi, Z. Syrgiannis, A. Bonasera, M. Prato, R. Argazzi, S. Caramori, V. Cristino, C.A. Bignozzi, J. Am. Chem. Soc., 137 (2015) 4630-4633. [7] M. Nishikawa, S. Hiura, Y. Mitani, Y. Nosaka, J. Photochem. Photobiol. A: Chem., 262 (2013) 52-56. [8] L.-W. Shan, G.-L. Wang, J. Suriyaprakash, D. Li, L.-Z. Liu, L.-M. Dong, J. Alloys Compd., 636 (2015) 131-137. [9] Y. Qiu, K. Yan, H. Deng, S. Yang, Nano Lett., 12 (2012) 407-413. [10] C. Chen, J. Moir, N. Soheilnia, B. Mahler, L. Hoch, K. Liao, V. Hoepfner, P. O'Brien, C. Qian, L. 13
ACCEPTED MANUSCRIPT He, G.A. Ozin, Nanoscale, 7 (2015) 3683-3693. [11] Y. Ma, X. Wang, Y. Jia, X. Chen, H. Han, C. Chem. Rev., 114 (2014) 9987-10043. [12] M. G. Walter, E. L. Warren, J. R. McKome, S. W. Boettcher, Q. Mi, E. A. Santori, N. S. Lewis, Chem. Rev., 110 (2010) 6446-6473. [13] Y. Ling, G. Wang, D.A. Wheeler, J.Z. Zhang, Y. Li, Nano Lett., 11 (2011) 2119-2125.
RI PT
[14] J. Deng, X. Lv, J. Gao, A. Pu, M. Li, X. Sun, J. Zhong, Energy Environ. Sci., 6 (2013) 1965-1970. [15] B.M. Klahr, T.W. Hamann, J. Phys. Chem. C, 115 (2011) 8393-8399.
[16] H. Dotan, K. Sivula, M. Grätzel, A. Rothschild, S.C. Warren, Energy Environ. Sci., 4 (2011) 958-964.
Szalai, J. Phys. Chem. C, 116 (2012) 15290-15296.
SC
[17] C.D. Bohn, A.K. Agrawal, E.C. Walter, M.D. Vaudin, A.A. Herzing, P.M. Haney, A.A. Talin, V.A.
Mater. Interfaces, 6 (2014) 5852-5859.
M AN U
[18] Gurudayal, S.Y. Chiam, M.H. Kumar, P.S. Bassi, H.L. Seng, J. Barber, L.H. Wong, ACS Appl.
[19] J.Y. Kim, D.H. Youn, J.H. Kim, H.G. Kim, J.S. Lee, ACS Appl. Mater. Interfaces, 7 (2015) 14123-14129.
[20] X. Wang, K.Q. Peng, Y. Hu, F.Q. Zhang, B. Hu, L. Li, M. Wang, X.M. Meng, S.T. Lee, Nano Lett.,
TE D
14 (2014) 18-23.
[21] L. Xi, P.D. Tran, S.Y. Chiam, P.S. Bassi, W.F. Mak, H.K. Mulmudi, S.K. Batabyal, J. Barber, J.S.C. Loo, L.H. Wong, J. Phys. Chem. C, 116 (2012) 13884-13889.
EP
[22] L. Badia-Bou, E. Mas-Marza, P. Rodenas, E.M. Barea, F. Fabregat-Santiago, S. Gimenez, E. Peris, J. Bisquert, J. Phys. Chem. C, 117 (2013) 3826-3833.
AC C
[23] M. Ji, J. Cai, Y. Ma, L. Qi, ACS Appl. Mater. Interfaces, 8 (2016) 3651-3660. [24] S.A. Patil, D.V. Shinde, E.-K. Kim, J.K. Lee, R.S. Mane, S.-H. Han, RSC Adv., 2 (2012) 11808-11812.
[25] Y. Li, J. Feng, H. Li, X. Wei, R. Wang, A. Zhou, Int. J. Hydrogen Energy, 41 (2016) 4096-4105. [26] A. Kay, I. Cesar, M. Grӓtzel, J. Am. Chem. Soc., 128 (2006) 15714-15721. [27] P. Zhang, A. Kleiman-Shwarsctein, Y.-S. Hu, J. Lefton, S. Sharma, A.J. Forman, E. McFarland, Energy Environ. Sci., 4 (2011) 1020-1028. [28] G. Rahman, O.-S. Joo, Int. J. Hydrogen Energy, 37 (2012) 13989-13997. [29] L. Peng, T. Xie, Z. Fan, Q. Zhao, D. Wang, D. Zheng, Chem. Phys. Lett., 459 (2008) 159-163. [30] Y.-J. Chen, L.-Y. Chen, Appl. Energy, 164 (2016) 924-933. 14
AC C
EP
TE D
M AN U
SC
RI PT
MANUSCRIPT [31] C. Zheng, Z. Zhu, S. Wang, Y. ACCEPTED Hou, Appl. Surf. Sci., 359 (2015) 805-811.
15
ACCEPTED MANUSCRIPT
Highlights
α-Fe2O3 nanostructures with different morphologies are directly grown on FTO.
Additives can be used to control the morphology of the α-Fe2O3 nanostructures.
Good performance of α-Fe2O3 is due to the electronic and morphological features.
AC C
EP
TE D
M AN U
SC
RI PT
S0925-8388(17)31408-1
DOI:
10.1016/j.jallcom.2017.04.213
Reference:
JALCOM 41617
To appear in:
Journal of Alloys and Compounds
Received Date: 8 March 2017 Revised Date:
9 April 2017
Accepted Date: 11 April 2017
Please cite this article as: Q. Liu, C. Chen, G. Yuan, X. Huang, X. Lv, Y. Cao, Y. Li, A. Hu, X. Lu, P. Zhu, Morphology-controlled α-Fe2O3 nanostructures on FTO substrates for photoelectrochemical water oxidation, Journal of Alloys and Compounds (2017), doi: 10.1016/j.jallcom.2017.04.213. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Morphology-Controlled α-Fe2O3 Nanostructures on FTO Substrates for
Photoelectrochemical Water Oxidation Qinglong Liu, Changlong Chen,∗ Guangzheng Yuan, Xing Huang, Xiaoxiao Lv, Yi Cao, Yaping Li, Anjie Hu, Xiangrong Lu, Peihua Zhu Key Laboratory of Chemical Sensing & Analysis in Universities of Shandong, School of Chemistry and Chemical Engineering,
RI PT
University of Jinan, Jinan, 250022, Shandong, P. R. China.
Abstract: Hematite (α-Fe2O3) nanostructured films were prepared on fluorine doped tin oxide substrates by heat treating the hydrothermally prepared precursor films. By using different additives in
SC
the hydrothermal preparations, the final hematite nanostructures could be controlled to be square prisms, nanorods and polyhedral blocks, of which the square prisms and the nanorods were vertically
M AN U
grown on the substrates and possessed preferred orientation along [110] direction while the polyhedral blocks were composed of small isotropic nanoparticles. When used as photoanodes, the α-Fe2O3 nanorod films showed better photoelectrochemical water oxidation performance than the other two kinds of α-Fe2O3 films. Analyses based on XRD, SEM and electrochemical impedance spectroscopy revealed that the α-Fe2O3 nanorods had characteristics including small diameter, abundant
TE D
through-holes, growing along [110] axis vertical to the substrate and high donor density, which are believed to contribute to the good photoelectrochemical water oxidation performance. On the ground that α-Fe2O3 is very promising semiconductor material for photoelectrochemical water oxidation application, such morphology controlled, nanostructured hematite α-Fe2O3 films based on simple
EP
preparation method are envisioned to provide valuable platforms for supporting catalysts and co-catalysts to achieve efficient light-assisted oxidation of water.
AC C
Keywords: photoelectrochemical; water oxidation/splitting; photoanode; hematite; α-Fe2O3; morphology control 1. Introduction
The decreasing fossil energy and growing energy demand are becoming a global problem and has compelled scientists to focus on exploring new energy sources.[1] Hydrogen achieved by solar light-driven photoelectrochemical (PEC) water splitting is a kind of renewable, clean and environmentally friendly energy, which is expected to be one effective solution to the energy problem. ∗
Corresponding author. Tel: +86 531 82765475; Fax: +86 531 89631632. E-mail addresses: [email protected]. 1
Since the first report on PEC waterACCEPTED splitting with MANUSCRIPT TiO2 as photoanode by Fujishima in 1972,[2] lots of researches have been focused on this issue, of which many oxides, such as TiO2,[3,4] WO3,[5,6] BiVO4,[7,8] ZnO,[9] In2O3,[10] etc., were studied as potential photoanode materials. Generally speaking, the excellent photoanode material should meet such demands: proper band gap energy, strong visible light absorption, suitable band edge positions for oxidation of water, fast transport of the
environment and environmentally friendly, etc.[11,12]
RI PT
photo-generated electrons and holes in the semiconductor, high chemical stability in aqueous
As an n type semiconductor material, hematite (α-Fe2O3) has the suitable band gap (Eg= ~2.2 eV) and valence band edge position for solar light-driven PEC water oxidation.[13] Its theoretical
SC
solar-to-hydrogen (STH) conversion efficiency is 16.8%,[14] which is pretty high. Moreover, hematite also has other advantages such as excellent chemical stability in aqueous environment, being
M AN U
environmentally friendly, abundance and low cost,[15] which make it a very promising photoanode material for PEC water splitting. On the other hand, some inherent characteristics of hematite, including short life time for charge carriers (~10 ps), short hole diffusion length (2-4 nm), low electric conductivity, low water oxidation evolution kinetics and high charge recombination, hinder it to achieve high PEC water oxidation performance.[13,16] In order to improve the PEC water oxidation
TE D
performance of hematite, many attempts have been done, which include element doping,[17,18] fabricating nanostructures,[19] surface modification,[20,21,22] and so on. Among them, fabricating nanostructured α-Fe2O3 is especially popular because it can not only significantly increase its specific surface area but also decrease the diffusion length of the photo-excited charge carriers.[10] For instance,
EP
α-Fe2O3 nanostructures like nanosheet arrays,[23] nanoparticle-chains,[24] nanorod arrays,[18,25] and
AC C
tree-like structures[26] have been reported to exhibit much enhanced PEC water splitting performance as compared to their corresponding bulk materials. Many methods have been used to prepare nanostructured semiconductor materials exemplified by atmospheric pressure chemical vapor deposition (APVCD),[27] pyrolysis,[14] electrochemical deposition[28] and hydrothermal methods,[29,30] of which the hydrothermal method is especially popular due to its merits of simplicity, low cost and low equipment requirement. In this work, we prepare nanostructured α-Fe2O3 films on fluorine doped tin oxide (FTO) substrates by heat treating the hydrothermally grown precursor films. By changing the additives used in the growth solution, it is realized to control the morphologies of the hematite nanostructures to be square prism arrays, nanorod arrays and polyhedral blocks. It was found that the α-Fe2O3 nanorod arrays contained several 2
ACCEPTED MANUSCRIPT advantages including small diameter, abundant through-holes, growing along [110] axis vertical to the substrate and high donor density, which make it possess obviously enhanced PEC water oxidation performance in comparison with the other two kinds of α-Fe2O3 nanostructures. 2. Experimental Section 2.1 Materials
RI PT
Ferric(III) Chloride (FeCl3·6H2O), Sodium Fluoride (NaF) and concentrated sulfuric acid were purchased from Sinopharm Chemical Reagent Co. Sodium Nitrate (NaNO3) and Sodium Sulfate (Na2SO4) were purchased from Tianjin Dengke Chemical Reagent Co. Anhydrous ethanol, hydrogen peroxide (30%) and acetone were purchased from Tianjin Fuyu Fine Chemical Co. All materials were
SC
used without further purification. 2.2 Synthesis
M AN U
FTO substrates (Zhuhai Kaivo Optoelectronic Technology Co.) with dimensions of 25 mm × 20 mm × 1.6 mm were cleaned ultrasonically in a solution containing acetone and ethanol (1:1 in volume) followed by rinsing in deionized water. After that, they were soaked in piranha solution for 24 h and then were rinsed with water and dried with air flow. In a typical synthesis, 2.25 mmol of FeCl3·6H2O and a certain amount of additive were co-dissolved in deionized water to form 15 mL of solution. Then
TE D
the solution was transferred into a 20 mL of Teflon-lined autoclave and a piece of FTO substrate was immersed into the solution, keeping the substrate against the wall of the Teflon liner and making its FTO side facing down. To control the morphologies of the α-Fe2O3 precursor nanostructures to be
EP
square prism arrays, nanorod arrays and polyhedral blocks, 4.5 mmol NaF, 5.5 mmol NaNO3 and 5.5 mmol Na2SO4 as additives were introduced to the hydrothermal reaction solutions, respectively. The for 10 h before cooling to room temperature. After the reaction, a layer
AC C
autoclave was heated at 95
of uniform thin film with yellow color was observed on the FTO substrate, which was thoroughly rinsed with deionized water and then was dried at 70 structure, the film was annealed in air at 550
for 2 h. To achieve the final hematite crystal
for 1 h. After cooled to room temperature, the film on
FTO substrate was further rinsed with deionized water and then was heat treated at 550
for 1 h again
but in vacuum environment, which is envisioned to improve its electronic properties. 2.3 Characterization X-ray diffraction (XRD) patterns of the samples were recorded in the range of 2θ = 20-70° on an X-ray diffractometer (Bruker D8 Focus) with Cu Kα1 radiation (λ = 0.15406 nm) at room temperature while the voltage and electric current were held at 40 kV and 30 mA, respectively. The morphologies 3
ACCEPTED MANUSCRIPT of the products were observed by scanning electron microscopy (SEM, FEI Quanta FEG 250). During the SEM measurements, a small piece of sample was adhered onto a copper stub using double-sided carbon tape. 2.4 Photoelectrochemical water oxidation measurements A xenon lamp (AuLight, CEL-S500, 500 W) through an AM 1.5G filter (AuLight, PD 250) was
RI PT
used as a solar simulator to irradiate quartz photoelectrochemical cell. The light intensity at the place of working photoanode was calibrated to be 100 mW·cm-2 by using solar power meter (Thorlabs, PM100 USB) with an S302C detector. The PEC measurements were carried out using an electrochemical workstation (Bio-Logic SAS, SP-150) in typical three electrode configuration: the
SC
prepared α-Fe2O3 films on FTO substrates serve as work electrode, Pt wire as counter electrode and Ag/AgCl (in saturated KCl) as reference electrode. The electrolyte was the aqueous solution of 0.1 M
M AN U
NaOH. During the measurement, the photoanode was swept from 0 to 0.75 V (vs. Ag/AgCl) at a rate of 10 mV·s-1. The measured potential vs. Ag/AgCl was converted to the reversible hydrogen electrode (RHE) scale according to Nernst equation:
ERHE = EAg/AgCl + 0.059 pH + EθAg/AgCl
(1)
where ERHE is the calculated potential versus RHE, EAg/AgCl is the measured potential against the Ag/AgCl reference electrode, EθAg/AgCl = 0.197 V at 25
TE D
.
2.5 Electrochemical impedance spectroscopy (EIS) measurements EIS measurements were performed using the same three electrode configuration describes above.
EP
The measurements were done in darkness at the following conditions: an AC frequency of 1 kHz with amplitude of 10 mV, scanning from -0.75 to +0.65 V vs. Ag/AgCl (0.21-1.61 V vs. RHE), in a 0.1 M
AC C
NaOH solution.
2.6 Incident photon-to-current conversion efficiency (IPCE) measurements The IPCE measurements were performed at 1.60 V vs. RHE with a 300 W Xe lamp equipped with a monochromator (omni-λ 300). The IPCE can be calculated according to the following equation:
IPCE =
(2)
Where I (mA·cm-2) is the measured photocurrent density, λ (nm) is the wavelength of the incident light, Jlight (mW·cm-2) is the measured irradiance at a specific wavelength. 3. Results and discussion Fig. 1a is the optical photograph of the hydrothermally prepared samples. It shows that the films 4
MANUSCRIPT obtained with NaF, NaNO3 and NaACCEPTED 2SO4 as additives are uniform on the FTO substrates and possess yellow color. Generally speaking, such yellow color indicates that the films are not hematite α-Fe2O3, whose color usually is red. To analyze the crystalline structures of these hydrothermally prepared films, X-ray diffraction (XRD) measurements were done. As can be seen from Fig. 2a, except for the diffraction peaks originated from the substrate FTO, the three kinds of samples obtained with different
RI PT
additives all show their own XRD patterns, however, none of them can be ascribed to the α-Fe2O3 phase, indicating the hydrothermal products are intermediates. Specifically, the XRD patterns of the samples prepared with NaF (black) and NaNO3 (red) as additives show the crystal phase that can be indexed to the tetragonal β-FeOOH (JCPDS No. 34-1266), of which the (310) peaks at 2θ = 26.7° can
SC
be observed from the enlarged part of the XRD patterns (Fig. 2b, the black and red patterns). Particularly, for the intermediate obtained with NaF as additive, the (211) diffraction peak also occurs,
M AN U
which is even higher than the (310) peak, much different with the referenced XRD lines, indicating the possibility of existing preferred orientation. The XRD pattern of the intermediate prepared with Na2SO4 as additive is shown in Fig. 2a (the blue one), which can be indexed to rhombohedral NaFe3(SO4)2(OH)6 (JCPDS No. 36-0425). To convert these intermediates to the final hematite α-Fe2O3 phase, they were heat treated at 550°C first in air and then in vacuum. After heat treatment, the color of
TE D
these three kinds of films all changed from yellow to red, as shown in Fig. 1b. The X-ray patterns of these heat treated films are shown in Fig. 2c. As can be seen, ignoring the diffraction peaks from FTO, the XRD patterns of the three samples can be indexed to the hematite α-Fe2O3 (JCPDS No. 33-0664).
EP
Apart from the peaks from α-Fe2O3 and FTO, no other diffraction peaks are found from these heat-treated samples. Importantly, obvious preferred orientation can be observed from the top two
AC C
patterns (the black and the red), which correspond to the films obtained with NaF and NaNO3 as additives, respectively. Specifically, the both patterns only show a strong (110) peak and a weak (300) peak, indicating that the hematite crystals are highly oriented along [110] direction vertical to the FTO substrate.[25] The XRD peaks of the sample obtained with Na2SO4 as additive (Fig. 2c, blue pattern), however, does not show obvious preferred orientation, which match well with the powder XRD lines of the referenced hematite α-Fe2O3 (JCPDS No. 33-0664) both in position and in relative intensity, indicating the fine powder nature of this sample. Morphologies and microstructures of the hydrothermally prepared intermediates and their final heat treated hematite α-Fe2O3 were observed by using SEM. Fig. 3a shows the SEM image of the β-FeOOH obtained with NaF as additive. It can be seen that the film is composed of square prism-like short rods 5
ACCEPTED MANUSCRIPT vertically grown on the FTO substrate. These square prism-like rods have side length for the top surface from 40-110 nm and are distributed uniformly. The cross sectional view image of the film shown in the inset reveals that these square prism-like rods are ~ 250 nm long. After heat treatment, the obtained α-Fe2O3 is substantially maintained the short square prism morphology (Fig. 3b). It is interesting that although the heat treatment does not change the morphology and size of the rods, it
RI PT
leads to large amount of holes. From the inset of Fig. 3b it can be seen, the holes have a diameter of ~ 9 nm and most of them are distributed on the side surfaces, very few on the top surface. These holes are believed to be generated due to the dehydration of the β-FeOOH during the heat treatment according to the equation (3):
(3)
EP
TE D
M AN U
SC
2FeOOH → Fe2O3 + H2O
Fig. 1 Optical photographs of the films on FTO substrates before (a) and after (b) heat treatments.
AC C
Because the width of the square prism-like rods is much smaller than their length, the produced H2O would readily release from the side surfaces rather than from the top surfaces of the rods and thereby forming the abundant small blind holes on the side surfaces.[31] In addition, during the hydrothermal preparation, it is believed that the F- ions are the key factor that results in the square prism-like shape of the rods. According to the theory of hard and soft acids and bases, at the beginning of the hydrothermal reaction, the hard acid, Fe3+ ions, prefer to bind with F- to form FeF2+ complex ions rather than directly hydrolyze.[24] It has been reported that FeF2+ complex ions readily promoted the anisotropic growth of the formed crystals[31] and in this situation, with the reaction temperature increasing, the hydrolysis was initiated and square prism-like β-FeOOH rods formed. 6
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 2 (a) XRD patterns of the hydrothermally prepared precursor films on FTO substrates with additives of NaF (black), NaNO3 (red) and Na2SO4 (blue). The bottom are the reference XRD lines for β-FeOOH (JCPDS No. 34-1266, pink line) and NaFe3(SO4)2(OH)6 (JCPDS No.36-0425, green line ). (b) Same XRD patterns shown in (a) enlarged in the range of 2θ = 25-28°. (c) XRD patterns of the heat treated films on FTO substrates. The bottom is the reference XRD lines for α-Fe2O3 (JCPDS No. 33-0664). 7
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 3 (a) SEM image of the precursor film prepared with NaF as additive; the inset is the cross-sectional view. (b) SEM image of the heat treated film with NaF as additive; the inset is the enlargement of the part in the white frame. (c) SEM image of the precursor film prepared with NaNO3 as additive. (d) SEM image of the heat treated film with NaNO3 as additive; the inset is the enlargement of the part 8
in the white frame. (e) SEM image of the heatACCEPTED treated film with NaMANUSCRIPT 2SO4 as additive. (f) The enlargement of the part in the white frame in (e); the inset is the enlargement of the part in the white frame.
As for the hydrothermal prepared β-FeOOH with NaNO3 as the additive, its SEM image shown in Fig. 3c exhibits that the film is composed of nanorod arrays vertically grown on the FTO substrate. The nanorods are uniformly distributed on the substrate and possess width and length of 20-60 and ~
RI PT
800 nm, respectively. Obviously they are thinner but longer than the β-FeOOH square prisms that obtained with NaF as additive. After heat treatment, the achieved α-Fe2O3 also keeps the nanorod array morphology, as shown in Fig. 3d. Like the α-Fe2O3 square prisms, after heat treatment, the α-Fe2O3 nanorods have no obvious change in size but appear irregular holes. These holes (Fig. 3d, inset),
SC
however, appear to be through-holes and are fewer in number than that for the α-Fe2O3 square prisms (Fig. 3b, inset). It is approximately because the nanorods are much thinner than the square prisms and
M AN U
thus the produced H2O during the heat treatment would much readily escape and thereby making the left space connect together, forming the irregular through-holes.
Fig. 3e and f show the SEM images of the hematite α-Fe2O3 obtained by heat treating the hydrothermal intermediate with Na2SO4 as the additive. The SEM image taken at low magnification (Fig. 3e) shows that the film is composed of large polyhedral blocks. The long and short edges of these
TE D
polyhedral blocks are ~ 17 and ~ 6 µm, respectively. Observing under larger magnification (Fig. 3f), it can be found that the polyhedral blocks actually are composed of nanoparticles with size of 20-100 nm (inset). It has been confirmed by XRD technique in above discussion that the intermediate is
EP
NaFe3(SO4)2(OH)6 and after heat treatment it converts to hematite α-Fe2O3. In this process, although it is not sure what the byproducts are, the theoretical weight loss is 50.52%. Although the profile of the
AC C
polyhedral blocks is retained during the heat treatment, the weight loss is so large that it can be imaged that the internal crystal texture of the intermediate will be destroyed thoroughly during the heat treatment, resulting in the small isotropic nanoparticles rather than keeping the polyhedral blocks just with pores like the other two cases discussed above. This result is well consistent with the XRD patterns (Fig. 2c), where the polycrystalline nature was exhibited for the α-Fe2O3 obtained with Na2SO4 as the additive in the hydrothermal reaction and obvious preferred orientation was revealed for the other two samples that obtained with NaF and NaNO3 as additives, respectively. To evaluate the PEC water oxidation performance of these α-Fe2O3 films prepared on FTO substrates, they were used as photoanodes in typical three electrode configuration to conduct the PEC water splitting under simulated sunlight (100 mW·cm-2). Fig. 4a shows the linear sweep 9
ACCEPTED MANUSCRIPT voltammograms (LSV), in which the current densities of the three kinds of α-Fe2O3 films both in the dark and under illumination are illustrated. As can be seen, without the assistance of any oxygen evolution catalysts (OECs), all the α-Fe2O3 photoanodes exhibit considerable photo-response. In the dark, they show negligible current densities until the applied potential being increased to ~ 1.60 V vs. RHE, where the water electrolysis begins. By comparison, it can be seen that the α-Fe2O3 nanorods
RI PT
film obtained with NaNO3 as additive exhibits the best PEC water oxidation performance. For instance, at the potential of 1.50 V vs. RHE, its photocurrent density is ~ 0.31 mA·cm-2, whereas, the photocurrent densities at the same potential for the α-Fe2O3 polyhedral block film and the α-Fe2O3 square prism film are ~ 0.17 and ~ 0.10 mA·cm-2, respectively. On the other hand, the photocurrent
SC
turn-on potentials for these three kinds of films are also different, which are ~ 0.97, 1.10 and 1.15 V vs.
TE D
M AN U
RHE for the nanorod-, polyhedral block- and square prism-α-Fe2O3 films, respectively.
EP
Fig. 4 (a) LSV collected from α-Fe2O3 nanorod array film (red), polyhedral block film (blue) and square prism film (black) under AM1.5G illumination at 100 mW·cm-2 and in the dark. (b) IPCEs of α-Fe2O3 nanorod array film (red), polyhedral block film (blue) and
AC C
square prism film (black).
Together with the observed fine structures of the α-Fe2O3 films from the SEM images as well as their XRD analysis, it can be deduced that the high PEC water oxidation performance exhibited by the α-Fe2O3 nanorod film is attributed to its two advantages: the high specific surface area due to the small diameter as well as the through-holes; the strong preferential orientation of the [110] axis vertical to the substrate. The former will provide more active sites for the water oxidation reaction, while the latter increases its conductivity. The crystal lattice of hematite α-Fe2O3 can be described as an alternation of iron bilayers and oxygen layers parallel to the (001) basal plane. It has been verified that the electron conductivity is up to 4 orders of magnitude higher within the (001) basal planes (e.g., in [110] direction) than orthogonal to them.[26] Consequently, the α-Fe2O3 nanorod arrays with 10
ACCEPTED preferential orientation of the [110] axis vertical MANUSCRIPT to the substrate is very beneficial for the majority carriers (electrons for n type hematite α-Fe2O3) to transport to the FTO substrate and thereby decreasing the carrier recombination. For the hematite α-Fe2O3 film composed of square prisms, although its XRD pattern also shows preferential orientation of the [110] axis vertical to the substrate and at the same time there are many holes existed, its PEC water oxidation performance is not as good
RI PT
as that of the α-Fe2O3 nanorod array film. On the one hand, the width of the α-Fe2O3 square prisms is much larger than that of the α-Fe2O3 nanorods and on the other hand, the holes on the square prisms are blind, which would block the electrolyte to fill into the holes. The two reasons make the effective specific surface area of the α-Fe2O3 square prism film being much smaller than that of the α-Fe2O3
SC
nanorod film, which would partially offset the advantage of the preferred orientation of the crystals. As for the α-Fe2O3 polyhedral block film, although the polyhedral blocks don’t possess preferred
M AN U
orientation and at the same time they have much larger apparent size than the α-Fe2O3 square prisms, the nature that they are composed of small nanoparticles leads to their specific surface area being larger than that of the latter, which probably results in the slightly higher photocurrent density of the former.
To further investigate the photoresponse of the α-Fe2O3 films, incident photon to charge conversion
TE D
efficiency (IPCE) was measured. IPCE is usually more accurate in evaluating the photoactivity of materials because it is based on monochromatic light and is independent of the lamps used in different laboratories. As shown in Fig. 4b, the three kinds of films all exhibit photoactivity to the light in range
EP
of 325-580 nm. As a whole, the IPCEs for these three kinds of films increase in this order: square prism film < polyhedral block film < nanorod film, which is well in accordance with the PEC results
AC C
shown in Fig. 4a. For example, although to the 340 nm light the polyhedral block film shows comparable IPCE to that of the nanorod film, which is ~4.8%, to the light of 350-580 nm the IPCE of nanorod film is much higher than that of the polyhedral block film. It is believed that in addition to the size and crystalline orientation, other factor like electronic properties of the hematite α-Fe2O3 films might also affect the PEC water oxidation behavior. In order to investigate the electronic properties of the α-Fe2O3 films, electrochemical impedance spectroscopy (EIS) measurements were conducted. Fig. 5 shows the Mott-Schottky plot of the three kinds of α-Fe2O3 films drawn based on the EIS measurements. As can be seen, they all exhibit positive slopes under increased anodic bias, conforming to the behavior of n-type semiconductors. By comparison, it can be seen that the slopes of Mott-Schottky plots for the three kinds of samples 11
MANUSCRIPT increase in this order: nanorods < ACCEPTED polyhedral blocks < square prisms, meaning their donor densities decreasing in that order. The donor densities can be calculated from these slopes according to equation (4): = −
⁄ !
"
#
(4)
where Nd is the donor density (in cm-3), e is the elementary charge (1.60×10-19 C), ε is the relative
RI PT
dielectric constant of α-Fe2O3 (taken to be 80),[23] ε0 is the permittivity of free space (8.854×10-14 F·cm-1), C is the capacitance (in F·cm-2), V is the applied potential (vs. RHE). After calculation, it was found that the donor densities for the α-Fe2O3 films composed of nanorods, polyhedral blocks and square prisms were 2.83×1020, 1.79×1020 and 1.41×1020 cm-3, respectively. Clearly, among these
SC
samples, the nanorod array film possesses the highest donor density. Such results indicate that the charge transport in the α-Fe2O3 nanorod film will be faster than that in the α-Fe2O3 polyhedral block
M AN U
film and square prisms film. This result together with the particle size, the holes and the crystalline orientation may explain the PEC water oxidation performance of the three kinds of α-Fe2O3 films. In other words, the α-Fe2O3 nanorod arrays have advantages including small diameter, abundant through-holes, growing along [110] axis vertical to the substrate and highest donor density, which lead to the best PEC water oxidation behavior, whereas, for the other two kinds of α-Fe2O3 films, they are
TE D
either low in donor density or without the characteristic of preferred crystalline orientation and consequently their PEC water oxidation performance is not as good as that of the α-Fe2O3 nanorod
AC C
EP
arrays.
Fig. 5 Mott-Schottky plot collected at a frequency of 1 kHz in the dark for α-Fe2O3 nanorod array film (red), polyhedral block film (blue) and square prism film (black).
4. Conclusions: 12
MANUSCRIPT In summary, hematite α-Fe2O3ACCEPTED films on FTO substrates were prepared by heat treating the hydrothermally grown precursor films in air and vacuum. The nanostructures that compose the α-Fe2O3 films can be controlled to be nanorod arrays, square prism arrays and polyhedral blocks, of which the former two possess the characteristic of preferential orientation of the [110] axis vertical to the substrates while the polyhedral blocks are composed of small isotropic nanoparticles. Using these
RI PT
α-Fe2O3 films as photoanodes, PEC water splitting measurements were conducted, which revealed that the α-Fe2O3 nanorod array film exhibited the best PEC water oxidation performance. The several merits possessed by the α-Fe2O3 nanorod arrays, including small diameter, abundant through-holes, growing along [110] axis vertical to the substrate and high donor density, contribute to the good PEC
SC
performance. Such morphology controlled, nanostructured hematite α-Fe2O3 films are envisioned to achieve efficient light-assisted oxidation of water by further modification with oxygen evolution
M AN U
catalysts like Co-Pi, the subject of ongoing investigations in our laboratory. Acknowledgements
This work is financially supported by the Research Fund of University of Jinan (No. XKY1603) and the National Innovation and Entrepreneurship Training Program for Undergraduate Students in Local Colleges and Universities of China (No. 201610427011).
TE D
References
[1] M. Mishra, D.-M. Chun, Appl. Catal. A: Gen., 498 (2015) 126-141. [2] A. Fujishima, K. Honda, Nature, 238 (1972) 37-38.
EP
[3] D. Regonini, A.C. Teloeken, A.K. Alves, F.A. Berutti, K. Gajda-Schrantz, C.P. Bergmann, T. Graule, F. Clemens, ACS Appl. Mater. Interfaces, 5 (2013) 11747-11755.
AC C
[4] L. Liu, J. Hensel, R.C. Fitzmorris, Y. Li, J. Z. Zhang, J. Phys. Chem. Lett., 1 (2010) 155-160. [5] S.S. Kalanur, Y.J. Hwang, S.Y. Chae, O.S. Joo, J. Mater. Chem. A, 1 (2013) 3479-3488. [6] F. Ronconi, Z. Syrgiannis, A. Bonasera, M. Prato, R. Argazzi, S. Caramori, V. Cristino, C.A. Bignozzi, J. Am. Chem. Soc., 137 (2015) 4630-4633. [7] M. Nishikawa, S. Hiura, Y. Mitani, Y. Nosaka, J. Photochem. Photobiol. A: Chem., 262 (2013) 52-56. [8] L.-W. Shan, G.-L. Wang, J. Suriyaprakash, D. Li, L.-Z. Liu, L.-M. Dong, J. Alloys Compd., 636 (2015) 131-137. [9] Y. Qiu, K. Yan, H. Deng, S. Yang, Nano Lett., 12 (2012) 407-413. [10] C. Chen, J. Moir, N. Soheilnia, B. Mahler, L. Hoch, K. Liao, V. Hoepfner, P. O'Brien, C. Qian, L. 13
ACCEPTED MANUSCRIPT He, G.A. Ozin, Nanoscale, 7 (2015) 3683-3693. [11] Y. Ma, X. Wang, Y. Jia, X. Chen, H. Han, C. Chem. Rev., 114 (2014) 9987-10043. [12] M. G. Walter, E. L. Warren, J. R. McKome, S. W. Boettcher, Q. Mi, E. A. Santori, N. S. Lewis, Chem. Rev., 110 (2010) 6446-6473. [13] Y. Ling, G. Wang, D.A. Wheeler, J.Z. Zhang, Y. Li, Nano Lett., 11 (2011) 2119-2125.
RI PT
[14] J. Deng, X. Lv, J. Gao, A. Pu, M. Li, X. Sun, J. Zhong, Energy Environ. Sci., 6 (2013) 1965-1970. [15] B.M. Klahr, T.W. Hamann, J. Phys. Chem. C, 115 (2011) 8393-8399.
[16] H. Dotan, K. Sivula, M. Grätzel, A. Rothschild, S.C. Warren, Energy Environ. Sci., 4 (2011) 958-964.
Szalai, J. Phys. Chem. C, 116 (2012) 15290-15296.
SC
[17] C.D. Bohn, A.K. Agrawal, E.C. Walter, M.D. Vaudin, A.A. Herzing, P.M. Haney, A.A. Talin, V.A.
Mater. Interfaces, 6 (2014) 5852-5859.
M AN U
[18] Gurudayal, S.Y. Chiam, M.H. Kumar, P.S. Bassi, H.L. Seng, J. Barber, L.H. Wong, ACS Appl.
[19] J.Y. Kim, D.H. Youn, J.H. Kim, H.G. Kim, J.S. Lee, ACS Appl. Mater. Interfaces, 7 (2015) 14123-14129.
[20] X. Wang, K.Q. Peng, Y. Hu, F.Q. Zhang, B. Hu, L. Li, M. Wang, X.M. Meng, S.T. Lee, Nano Lett.,
TE D
14 (2014) 18-23.
[21] L. Xi, P.D. Tran, S.Y. Chiam, P.S. Bassi, W.F. Mak, H.K. Mulmudi, S.K. Batabyal, J. Barber, J.S.C. Loo, L.H. Wong, J. Phys. Chem. C, 116 (2012) 13884-13889.
EP
[22] L. Badia-Bou, E. Mas-Marza, P. Rodenas, E.M. Barea, F. Fabregat-Santiago, S. Gimenez, E. Peris, J. Bisquert, J. Phys. Chem. C, 117 (2013) 3826-3833.
AC C
[23] M. Ji, J. Cai, Y. Ma, L. Qi, ACS Appl. Mater. Interfaces, 8 (2016) 3651-3660. [24] S.A. Patil, D.V. Shinde, E.-K. Kim, J.K. Lee, R.S. Mane, S.-H. Han, RSC Adv., 2 (2012) 11808-11812.
[25] Y. Li, J. Feng, H. Li, X. Wei, R. Wang, A. Zhou, Int. J. Hydrogen Energy, 41 (2016) 4096-4105. [26] A. Kay, I. Cesar, M. Grӓtzel, J. Am. Chem. Soc., 128 (2006) 15714-15721. [27] P. Zhang, A. Kleiman-Shwarsctein, Y.-S. Hu, J. Lefton, S. Sharma, A.J. Forman, E. McFarland, Energy Environ. Sci., 4 (2011) 1020-1028. [28] G. Rahman, O.-S. Joo, Int. J. Hydrogen Energy, 37 (2012) 13989-13997. [29] L. Peng, T. Xie, Z. Fan, Q. Zhao, D. Wang, D. Zheng, Chem. Phys. Lett., 459 (2008) 159-163. [30] Y.-J. Chen, L.-Y. Chen, Appl. Energy, 164 (2016) 924-933. 14
AC C
EP
TE D
M AN U
SC
RI PT
MANUSCRIPT [31] C. Zheng, Z. Zhu, S. Wang, Y. ACCEPTED Hou, Appl. Surf. Sci., 359 (2015) 805-811.
15
ACCEPTED MANUSCRIPT
Highlights
α-Fe2O3 nanostructures with different morphologies are directly grown on FTO.
Additives can be used to control the morphology of the α-Fe2O3 nanostructures.
Good performance of α-Fe2O3 is due to the electronic and morphological features.
AC C
EP
TE D
M AN U
SC
RI PT