Electrodeposited nanostructured a-Fe2O3 thin films for solar water splitting: Influence of Pt doping on photoelectrochemical performance

at SciVerse ScienceDirect

Materials Chemistry and Physics 140 (2013) 316e322

Contents lists available

Materials Chemistry and Physics

journal homepage: www.elsevier .com/locate/matchemphys

Electrodeposited nanostructured a-Fe2O3 thin films for solar water splitting:Influence of Pt doping on photoelectrochemical performance

Gul Rahman a,b, Oh-Shim Joo a,*

aClean Energy Research Center, Korea Institute of Science and Technology (KIST), Seongbuk-gu, Seoul 130-650, Republic of Koreab School of Science, University of Science and Technology, 52 Eoeun dong, Yuseong-gu, Daejeon 305-333, Republic of Korea

h i g h l i g h t s

� Un-doped and Pt doped a-Fe2O3 thin films were synthesized by simple electrodeposition.� The surface morphology of a-Fe2O3 thin films changed with Pt % in the film.� A high photocurrent for water splitting was observed on Pt doped films.� Pt doping also enhanced the catalytic activity of a-Fe2O3 thin films for water oxidation.

a r t i c l e i n f o

Article history:Received 19 April 2012Received in revised form26 February 2013Accepted 15 March 2013

Keywords:SemiconductorsElectrochemical techniquesNanostructuresThin filmsElectrochemical properties

* Corresponding author. Tel.: þ82 2 958 5215; fax:E-mail addresses: [email protected], joocat@kis

0254-0584/$ e see front matter � 2013 Elsevier B.V.http://dx.doi.org/10.1016/j.matchemphys.2013.03.042

a b s t r a c t

Electrochemically deposited a-Fe2O3 thin films, whose composition was tuned by Pt doping, wereinvestigated as photoanode for photoelectrochemical water splitting. Morphological and structuralcharacteristics of the nanostructured a-Fe2O3 thin films were studied by scanning electron microscopyand X-ray diffraction techniques. The films were characterized by Raman spectroscopy and X-rayphotoelectron spectroscopy to determine the effect of Pt doping on the a-Fe2O3 structure. The photo-electrochemical performance of the films was examined by linear sweep voltammetry and electro-chemical impedance spectroscopy. Results of these studies showed that Pt doping increased the densityof small-sized nanoparticles in a-Fe2O3 thin films. The Pt doped films exhibited higher photo-electrochemical activity by a factor of 1.4 over un-doped a-Fe2O3 films. The highest photocurrent densityof 0.56 mA cm�2 was registered for 3% pt doped film at 0.4 V versus Ag/AgCl in 1 M NaOH electrolyte andunder standard illumination conditions (AM 1.5 G, 100 mW cm�2). This high photoactivity can beattributed to the high active surface area and increased donor density caused by Pt doping in the film.Electrochemical impedance analysis also revealed significantly low charge transfer resistance of Pt dopedfilms, indicating its superior electrocatalytic activity for water splitting reaction compared to un-dopeda-Fe2O3 thin films.

� 2013 Elsevier B.V. All rights reserved.

1. Introduction

In the search of new sources to cope with energy crisis related tofossil fuels, efficient storage of solar energy in the form of hydrogenusing cost effective and stable semiconductor material, has been asubject of interest since 1972 [1]. Metal oxide semiconductors suchas TiO2, WO3, ZnO, BiVO4, and a-Fe2O3 have been investigatedextensively for photoelectrochemical (PEC) water splitting [2e10].Among them, a-Fe2O3 is a promising material for water oxidationwith suitable bandgap (Eg w 2.1 eV), stability in aqueous solution,

þ82 2 958 5807.t.re.kr (O.-S. Joo).

All rights reserved.

ample abundance, non-toxic and environmentally friendly [9,11e13]. However, the practical use of this material is limited byseveral problems such as lowelectronmobility (w10�2 cm2V�1 s�1)[14,15], short hole-diffusion length (w2e4 nm) [16], low absorptioncoefficient due to the indirect bandgap and short life time of chargecarriers (w10 ps) [17]. Also, the conduction band edge of hematitedoes not straddle the reversible hydrogenpotential and require highoverpotential (external bias) for water reduction [18].

More challenges with the use of a-Fe2O3 photoelectrode forefficient water splitting are its slow water oxidation kinetics andpoor charge transport properties [12,13,19]. Significant improve-ments have been achieved by many research groups to elevate theslow rate of water oxidation by applying water oxidation catalystssuch as oxides of Ru, Ir, and cobalt [20e22]. For instance, the surface

G. Rahman, O.-S. Joo / Materials Chemistry and Physics 140 (2013) 316e322 317

modification of a-Fe2O3 with IrO2 was reported to remarkably in-crease the rate of oxygen evolution reaction and overall watersplitting efficiency [21]. The poor charge transport properties of a-Fe2O3 can be enhanced by improving the quality of nanostructures(crystallinity, crystallographic orientation) and by changing thecomposition of film by doping. Several research groups havefocused their research on hematite thin films to improve the lowelectronic conductivity by tuning the composition (e.g., doping)[23]. Both metals and non-metals have been utilized as dopants toenhance the carrier density and hence the conductivity of a-Fe2O3[18,23,24].

Nanostructured hematites have been synthesized by severaltechniques, including hydrothermal synthesis [13,25], spray py-rolysis [11,26], colloidal and magnetite colloidal solution ap-proaches [27], atomic layer deposition (ALD) [28], atmosphericpressure chemical vapor deposition (APCVD) [29], and electro-chemical methods [30]. Hematite films prepared by APCVD, ALDand magnetite colloidal solution methods have shown particularlypromising results. However, these approaches require a specialinstrument configuration or the use of toxic organic solvents andmetaleorganic precursors.

In this study, we demonstrated the use of cathodic electrode-position to synthesize Pt doped nanostructured a-Fe2O3 thin films.This method is cost-effective and safe, and it can be performedwitha simple apparatus. The size, film thickness, and morphology of thea-Fe2O3 nanoparticles can be tailored by simply tuning the depo-sition conditions. McFarland co-workers have extensively utilizedelectrodeposition technique to synthesize doped a-Fe2O3 films andstudied the effect of doping on the photoactivity and overall PECperformance of a-Fe2O3 [8,30]. In their particular studies of Ptdoped a-Fe2O3 films [30], no evidence of enhanced electrocatalyticactivity was observed due to the dopant. Herein, we found that Ptdoping not only elevate the photoactivity of a-Fe2O3 by improvingthe charge transport properties, but also significantly decreases thecharge transfer resistance of the film for water oxidation and hence,improve its electrocatalytic activity. A thorough investigation of thephysical and photoelectrochemical properties of un-doped and Ptdoped electrochemically deposited a-Fe2O3 thin films is presentedin this study.

2. Experimental

2.1. Synthesis of a-Fe2O3 thin films

Nanostructured a-Fe2O3 thin films were obtained after anneal-ing of the electrochemically deposited iron films on a fluorine-doped tin oxide (FTO, TEC 8, Pilkington glass) glass substrate. Theelectrodeposition bath consisted of an aqueous solution of 10 mMiron(II) sulfate heptahydrate (FeSO4$7H2O, Samchun, 98e102%). ForPt doped films, H2PtCl6. 5.7(H2O) was added into deposition bathranging from 1 to 5wt. % of Pt/(Ptþ Fe). For the electrodeposition ofthe hematite films, a typical three-electrode electrochemical cellwas used that comprised the FTO substrate (1 cm � 1.5 cm), plat-inum (2 cm � 2 cm) and Ag/AgCl/NaCl (3 M) as the working,counter and reference electrodes, respectively. The FTO substratewas ultrasonically pre-cleaned by sequential rinses with acetone,distilled water and isopropanol. The cell was connected to apotentiostat (IviumStat technologies, Netherlands) that was usedfor iron oxide film preparation and other electrochemical mea-surements. Iron films were deposited on the FTO substrate from aniron precursor solution by applying a constant potential of �1.0 V(versus Ag/AgCl electrode) for 10 min. After each deposition, thefilm was thoroughly rinsed with de-ionized water and then driedwith a gentle stream of argon. To obtain crystalline nanostructures,the as-deposited films were annealed at 700 �C (reached at a rate of

2 �C min�1) for 2 h in air to obtain highly activated a-Fe2O3nanostructures.

2.2. Structural and morphological characterization

Scanning electron microscopy images were collected with a fieldemission scanning electron microscope (NOVA NanoSEM200- FEICompany). The crystalline phaseswere identified byXRD (XRD-6000,Shimadzu, Japan) with Ka radiations (l ¼ 1.542 �A). Diffraction pat-ternswere recorded from 20 to 80 2qwith a sampling pitch of 0.020�.

The phase analysis was additionally performed using RamanMicroscope (Nicolet ALMEGA XR Dispersive Raman). The laserbeam (l ¼ 633 nm) was focused on the sample by a lens to producea spot. The spectra were measured from 100 to 2000 cm�1 in 10 sacquisition times.

2.3. X-ray photoelectron spectroscopy (XPS)

XPS spectra of the filmwere acquired with PHI 5000 VersaProbe(Ulvac-PHI) under high vacuum condition (6.8 � 10�8 pa), using amonochromatic Al Ka X-ray source (1486.6 eV). The data werecollected from a spot size of 100 mm � 100 mm. Carbon 1s peak(284.6 eV) was used for internal calibration.

2.4. Photoelectrochemical characterization

The photoelectrochemical characterization of the a-Fe2O3 wasperformed using an IviumStat potentiostat with a three-electrodeconfiguration: a working electrode (the hematite film), aplatinum-wire counter electrode and an Ag/AgCl (in 3 M NaCl)reference electrode. A copper wire was soldered on the exposedportion of the FTO substrate (to establish a connection), and anepoxy resin was used to seal all exposed portions of the FTO exceptfor the well-defined working area of the hematite electrodes. Forphotocurrent measurements, the electrodes were immersed in asolution of 1 M NaOH (pH 13.6). The hematite electrode wasscanned from �400 to 700 mV (versus Ag/AgCl electrode) at a rateof 50 mV s�1. The samples were illuminated from the front sidewith simulated sunlight from a 150 W short-arc xenon lamp(Portable solar simulator, PEC-L01) equipped with an air mass filter(A.M. 1.5 G) with a corrected intensity of 1 sun (100 mW cm�2) atthe sample surface. The electrolyte was purged with nitrogen gasbefore the experiments to prevent any possible reaction with dis-solved oxygen at the counter electrode.

2.5. Electrochemical impedance spectroscopy (EIS)

Electrochemical impedance spectroscopy (EIS) was performedin 1 M NaOH using an IviumStat with a three-electrode configu-ration. A sinusoidal perturbation with an amplitude of 10 mV andfrequencies ranging from 100 kHz to 10 Hz was superimposed on abias voltage ranging from �200 to 400 mV (versus Ag/AgCl elec-trode). EIS was performed in the dark, and a Nyquist plot wasconstructed to simulate the equivalent circuit to obtain the spacecharge layer capacitance of the a-Fe2O3 thin films.

3. Results and discussions

3.1. SEM analysis

Top-view scanning electronmicroscopy images of the un-dopedand different wt.% Pt doped a-Fe2O3 thin films are shown in panelaee of Fig. 1. The un-doped film (image a) shows vertically grownsegregated islands of nanostructures made up of small nano-particles. These islands are separated by empty channels. Inset to

Fig. 1. SEM images of un-doped and Pt doped a-Fe2O3 films; (a) un-doped film, (b) 1% Pt, (c) 2% Pt, (d) 3% Pt, and (e) 5% Pt.

G. Rahman, O.-S. Joo / Materials Chemistry and Physics 140 (2013) 316e322318

image (a) shows the cross sectional SEM image of the un-dopedfilm whose thickness was estimated w250 nm. The morphologyof 1 and 2% Pt doped films did not show significant variation ascompared to un-doped a-Fe2O3 film. However, further increase inPt amount changed the surface morphology of Fe2O3 deposits.Image (d) shows the surface morphology of 3% Pt doped film inwhich the particles are regularly distributed with average diameterof w100 nm. The segregated islands consisted of small-sizednanoparticles seems to be separated from each other. The filmappeared more dense and uniform than that of un-doped film. Asimilar but less prominent effect was observed in 5% Pt doped filmas shown by image (e). These results predict that Pt doping controlsthe growth mechanism during cathodic electrodeposition of a-Fe2O3 thin films and affects the size and overall morphology ofnanostructures.

3.2. Structural and chemical analysis

The XRD patterns of high temperature annealed un-doped andPt doped a-Fe2O3 thin films are presented in Fig. 2. Results show thepresence of two representative peaks of a-Fe2O3 for (104) and (110)planes respectively (consistent with the powder standards (PDF #01-089-0599)). The intensities of these peaks indicate the presenceof well-crystalline hematite phase. The intensity of (104) peakwhich is slightly larger than (110) peak, indicate that the mostconductive plane (001) of the a-Fe2O3 is not aligned vertically toFTO substrate [27]. When Pt was introduced in the film, no signif-icant change was observed in the position and intensity of a-Fe2O3representative peaks. However, a detectable peak broadening wasseen in 5% Pt doped film. As shown in SEM image, 5% doped filmcontained small-sized nanoparticles on its surface. The observed

Fig. 2. XRD patterns of un-doped and Pt doped a-Fe2O3 films; (a) un-doped film, (b) 1%Pt, (c) 2% Pt, (d) 3% Pt, and (e) 5% Pt. The representative peaks of a-Fe2O3 are designatedby asterisk (*).

G. Rahman, O.-S. Joo / Materials Chemistry and Physics 140 (2013) 316e322 319

peak broadening could therefore be related to these small nano-particles caused by Pt doping in a-Fe2O3 thin films.

To investigate the effect of Pt doping on the crystal phases of a-Fe2O3 thin films, the samples were analyzed by Raman spectros-copy. Fig. 3 shows the Raman spectra of the un-doped and Pt-dopeda-Fe2O3 thin films. The spectra exhibited typical bands of a-Fe2O3phase, showing peaks at 224, 243, 292, 409, 495, and 610 cm�1 andmatched well with the reported Raman data for hematite [31]. Theband around 660 cm�1 was observed in all samples, which wasattributed to presence of Fe3O4 or disorder phase within the Fe2O3crystal lattice [32]. The intensity of this peak was observed to in-crease with Pt concentration which could be attributed to thechange in the surface structure and grain boundary disorder of a-Fe2O3 as previously described [31]. This result agrees well with thePt and Ti doped hematite thin films [30,33].

The chemical composition of the un-doped and Pt-doped a-Fe2O3 films was investigated using X-ray photoelectron

Fig. 3. Raman spectra of un-doped and Pt-doped a-Fe2O3 films; (a) un-doped film, (b)1% Pt, (c) 2% Pt, (d) 3% Pt, and (e) 5% Pt.

spectroscopy (XPS). The Fe 2p spectra (Fig. 4(A)) of un-doped and5 wt.% Pt-doped films exhibited the typical 2p1/2 and 2p3/2 peaks ofFe3þ at w724.4 and 710.8 eV, respectively [34]. A characteristicsatellite peak was also observed at 718.8 eV, suggesting the a-Fe2O3phase in both un-doped and Pt-doped films. Fig. 4(B) depicts the Pt4f spectra obtained from the a-Fe2O3 films with 0 and 5wt.% Pt. Theresult revealed that Pt is in the form of Ptþ4 and not Pt0 as evi-denced by the Pt 4f5/2 and 4f7/2 peaks at w78.0 and w74.0 eV [35].The surface concentration of Pt calculated from XPS analysis(Fig. 4(C)) was 0.04, 0.08, 0.11, and 0.29% for 1, 2, 3, and 5 wt. % Pt infilm deposition solution, respectively. The doped Pt concentrationis substantially smaller than in deposition solution which suggeststhat a small amount of Pt is introduced in the lattice of a-Fe2O3. Ingeneral, the Mþ4 substitution in hematite cause the reduction ofFeþ3 to Feþ2 due to the extra electron on dopant. However, no Feþ2

peaks were observed in the Fe 2p spectra of Pt doped films. Thismay be because of the quite low amount of Pt in the films or highoxidizing annealing conditions employed in our experiment.

3.3. Photoelectrochemical performance

The PEC performance of the a-Fe2O3 photoanode films wasstudied in a 1 M NaOH solution using a three-electrode electro-chemical cell connected to a potentiostat and a solar simulator. Thepotentials were measured relative to the Ag/AgCl (3 M NaCl) elec-trode. Fig. 5(A) depicts the currentepotential (IeV) curves of un-doped and Pt doped a-Fe2O3 films in 1 M NaOH solution in thedark and under illumination. In case of un-doped film, photocur-rent onset at w�0.1 V and increased with applied potential untilw0.3 V. However, little change was observed above 0.3 V and thecurrent attained the shape of plateau that can be attributed to theelectron transport limitations in a-Fe2O3 thin films. Pt doped filmsshowed higher photoactivity than un-doped film. For instance, thephotocurrent density of un-doped film was 0.39 mA cm�2 at 0.4 V,and 0.56 mA cm�2 for 3% Pt doped sample. This increase inphotocurrent is attributed to the increased Pt % as well as to greatersurface area of doped films. Positive shift of photocurrent onsetpotential was observed for doped films from w�0.1 to 0.08 Vversus Ag/AgCl, compared to un-doped film. This positive shift in-dicates the water oxidation kinetics is limited due to the increasedsurface states as a result of increased surface area of Pt doped films.The shift can be overcome by coupling Pt doped films with oxygenevolving catalysts. Similar effect of dopant on the photocurrentonset potential has been observed for other a-Fe2O3 films [11,25].Another important observation was the change of dark currentwith Pt doping of a-Fe2O3 thin films. Fig. 5(B) illustrates theextended form of dark curves of un-doped and Pt doped films. Theonset potential of dark current decreases significantly with Ptdoping which indicate that the overpotential required for wateroxidation is reduced on Pt doped films. A maximum of w100 mVreduction of overpotential was observed on 5% Pt doped film overun-doped film. Such a reduction in overpotential of water oxidationmeans that the Pt doped film surface became more catalytic foroxygen evolution than the un-doped a-Fe2O3 film surface. Thiscould be related to the high active surface area of Pt doped filmswhich contain small-sized nanoparticles as compared to the un-doped film (shown by SEM analysis).

3.4. EIS analysis

To obtain a better understanding of the charge transport prop-erties of un-doped and Pt doped films, electrochemical impedancemeasurements were performed. According to the depletion layermodel, the semiconductor space charge layer capacitance (C)

Fig. 4. X-ray photoelectron spectra of: (A) Fe 2p and (B) Pt 4f recorded from un-doped and 5% Pt-doped a-Fe2O3 films (C). Atomic % of Pt in the deposited film measured by XPSanalysis.

G. Rahman, O.-S. Joo / Materials Chemistry and Physics 140 (2013) 316e322320

depends on the applied potential (V) and is given by the MotteSchottky equation:

1=C2 ¼�2=εrεoeND

��V � Vfb � kBT=e

�;

where εr is the dielectric constant of the semiconductor (εr ¼ 80for hematite), εo is the vacuum permittivity, e is the charge of theelectron, ND is the donor density, Vfb is the flat band potential, kBis the Boltzmann constant and T is the absolute temperature. TheMotteSchottky plots were generated from the capacitance valuesmeasured at 10 kHz in the dark as shown by Fig. 6. The positiveslopes of the plot indicate the presence of a characteristic n-typesemiconductor and that electrons are the majority charge carriers[23]. From the slopes of MotteSchottky plots, donor density wascalculated while the flat band potential was estimated from theintercept of potential axis. These values are tabulated in Table 1.The values of flat band potential of Pt doped films are morepositive compared to un-doped film and are in the range reportedfor a-Fe2O3 [36]. Overall, these values are more cathodic than thephotocurrent onset potentials shown in Fig. 5(A). This differencein flat band potentials and photocurrent onset potentials ismainly attributed to a high electronehole recombination as aresult of surface states [33]. From the slope of MotteSchottkyplots, a donor density of 7.9 � 1018 cm�3 was calculated for un-doped film which is increased up to 9.8 � 1018 cm�3 for 5% Ptdoped film. These results provide direct evidence to support thatthe Pt doping serve as electron donor and the donor densityincreased with Pt doping in a-Fe2O3 due to the substitution ofFeþ3 by Ptþ4 in the hematite lattice. The increased donor densitycauses shrinking of the space-charge layer width and,

consequently, strengthen the electric field near the film electro-lyte interface. The separation and transport of electrons and holesare thus enhanced, thereby improving the photoelectrochemicalperformance of the film. To investigate it further, the Debyelength of un-doped and Pt doped films was calculated using theformula:

LD ¼�εoεrkBT=2e

2ND

�1=2

where LD is the Debye length, εr is the dielectric constant of thesemiconductor, εo is the vacuum permittivity, e is the charge of theelectron,ND is the donor density, kB is the Boltzmann constant and Tis the absolute temperature. According to the Schottky barriermodel, the transit time through the depletion layer is proportionalto the square of the Debye length [37]. As can be seen in Table 1, ahigher Pt concentration corresponds to a shorter Debye length,which would decrease the transit time and be helpful for electronehole separation, thus a higher photocurrent can be expected [38].However, we noted that increasing the Pt concentration more than3% did not improve the photocurrent significantly. This could beattributed to the fact that a narrower depletion layer is deleteriousfor suppressing the recombination rate. These observationstogether suggest that the strategy of doping hematite should beto include appropriate amount of dopant in the structure tobalance the competing effects between charge separation andrecombination.

Furthermore, the water oxidation kinetics at the a-Fe2O3 film/electrolyte interface was investigated by constructing Nyquist plotsof the photoanodes at 0.75 V in the dark. The Nyquist plots of un-doped and Pt doped a-Fe2O3 thin films and fitted curves are

Fig. 5. (A) Current vs. potential curves of un-doped and Pt doped a-Fe2O3 films; (a) un-doped film, (b) 1% Pt, (c) 2% Pt, (d) 3% Pt, and (e) 5% Pt. The photocurrent was measuredunder standard illumination conditions (AM 1.5 G 100 mW cm�2). (B) Extended formof dark currents of (A).

Fig. 6. MotteSchottky plots of un-doped and Pt doped a-Fe2O3 films analyzed in thedark; (a) un-doped film, (b) 1% Pt, (c) 2% Pt, (d) 3% Pt, and (e) 5% Pt. The electro-chemical impedance analysis was performed in 1 M NaOH, and the MotteSchottkyanalysis was performed at 10 kHz.

Table 1Various parameters obtained from MotteSchottky plots of un-doped and Pt dopeda-Fe2O3 films analyzed in the dark.

Electrode Donor density(ND)/E18.cm�3

Flat bandpotential (Vfb)/V

Debyelength LD/nm

Un-doped 7.9 �0.66 2.671% Pt 3.9 �0.46 3.802% Pt 8.7 �0.55 2.553% Pt 9.2 �0.47 2.485% Pt 9.8 �0.56 2.40

Fig. 7. Nyquist plots un-doped and Pt doped a-Fe2O3 films showing the imaginaryversus the real component of the impedance at �0.75 vs Ag/AgCl and the fitted plotsobtained in the dark.

G. Rahman, O.-S. Joo / Materials Chemistry and Physics 140 (2013) 316e322 321

shown in Fig. 7. The symbols represent the experimental results,and the solid lines are the fitting results of the calculated data. Insetto the figure shows the corresponding equivalent circuit and thefitting results are summarized in Table 2. For all films, semi-circleshaped Nyquist plots were obtained. The equivalent circuit repre-sents solution resistance (Rs), a-Fe2O3 film/electrolyte interfaceresistance (Rct) and its corresponding capacitive counterpart(CPEct). The lower resistance and higher constant phase elementvalues represent better charge transport at a-Fe2O3 film/electrolyteinterface. According to the fitting results in Table 2, Pt dopingcauses great difference in charge transfer resistance of a-Fe2O3

films. Un-doped film showed high charge transfer resistance towater oxidation reaction but as the Pt % increased in the film, asignificant decline was observed. The lowest Rct observed was43.85 U for 5% Pt doped film which is w4.2 fold smaller than un-doped film (Rct ¼ 183.7 U). Such a lower charge transfer resis-tance is attributed to the high active surface area of Pt doped films

Table 2Equivalent circuit parameters obtained from fitting of Nyquist plots of un-doped andPt doped a-Fe2O3 films.

Electrode RS/U Rct/U CPEct/F

Un-doped 30.41 183.7 4.218E-51% Pt 27.3 151.1 4.247E-52% Pt 26.32 122.0 3.760E-53% Pt 31.56 66.96 7.030E-55% Pt 25.3 43.85 3.405E-6

G. Rahman, O.-S. Joo / Materials Chemistry and Physics 140 (2013) 316e322322

that catalyze water oxidation effectively. The value of constantphase element also increased significantly for 3 and 5% doped film,indicating yet again the better electrocatalytic activity of Pt dopeda-Fe2O3 films over un-doped film.

4. Conclusions

In summary, a-Fe2O3 thin films were prepared and doped withPt using cathodic electrodeposition. SEM analysis showed that Ptdoping changed the surface morphology of a-Fe2O3 films byreducing the size and increasing the particle density. These small-sized nanoparticles exhibited high photocurrent and overall PECperformance. Electrochemical impedance spectroscopy revealedthat Pt doped films have higher donor density than that of un-doped film which is responsible for high photoactivity. Moreover,a significantly low charge transfer resistance was observed forwater oxidation on Pt doped a-Fe2O3 films, indicating its potentialfor photoelectrochemical applications.

Acknowledgments

The authors gratefully acknowledge financial support from theMinistry of Science and Technology of Korea that supported theresearch at the Hydrogen R & D Center, which is a 21st CenturyFrontier R & D program. The authors also acknowledge the researchprogram of the Korea Institute of Science and Technology (KIST).

References

[1] A. Fujishima, K. Honda, Nature 238 (1972) 37e38.[2] S.C. Moon, H. Mametsuka, S. Tabata, E. Suzuki, Catal. Today 58 (2000)

125e132.[3] R. Liu, Y. Lin, L.Y. Chou, S.W. Sheehan, W. He, F. Zhang, H.J.M. Hou, D. Wang,

Angew. Chem. Int. Ed. 50 (2011) 499e502.[4] N. Chouhan, C.L. Yeh, S.F. Hu, R.S. Liu, W.S. Chang, K.H. Chen, Chem. Commun.

47 (2011) 3493e3495.[5] Y.H. Ng, A. Iwase, A. Kudo, R. Amal, J. Phys. Chem. Lett. 1 (2010) 2607e2612.[6] P. Chatchai, Y. Murakami, S.Y. Kishioka, A.Y. Nosaka, Y. Nosaka, Electrochim.

Acta 54 (2009) 1147e1152.[7] V.R. Satsangi, S. Kumari, A.P. Singh, R. Shrivastav, S. Dass, Int. J. Hydrogen

Energy 33 (2008) 312e318.

[8] A. Kleiman-Shwarsctein, Y.S. Hu, A.J. Forman, G.D. Stucky, E.W. McFarland,J. Phys. Chem. C 112 (2008) 15900e15907.

[9] R. van de Krol, Y. Liang, J. Schoonman, J. Mater. Chem. 18 (2008) 2311e2320.[10] J. Hensel, G. Wang, Y. Li, J.Z. Zhang, Nano Lett. 10 (2010) 478e483.[11] F. Le Formal, M. Grätzel, K. Sivula, Adv. Funct. Mater. 20 (2010) 1099e1107.[12] A.A. Tahir, K.G.U. Wijayantha, S. Saremi-Yarahmadi, M. Mazhar, V. McKee,

Chem. Mater. 21 (2009) 3763e3772.[13] H.G. Cha, J. Song, H.S. Kim, W. Shin, K.B. Yoon, Y.S. Kang, Chem. Commun. 47

(2011) 2441e2443.[14] T. Lindgren, H. Wang, N. Beermann, L. Vayssieres, A. Hagfeldt, S.E. Lindquist,

Sol. Energy Mater. Sol. Cells 71 (2002) 231e243.[15] F.J. Morin, Phys. Rev. 93 (1954) 1195e1199.[16] K. Sivula, R. Zboril, F. Le Formal, R. Robert, A. Weidenkaff, J. Tucek, J. Frydrych,

M. Grätzel, J. Am. Chem. Soc. 132 (2010) 7436e7444.[17] A.G. Joly, J.R. Williams, S.A. Chambers, G. Xiong, W.P. Hess, J. Appl. Phys. 99

(2006) 053521.[18] G. Wang, Y. Ling, D.A. Wheeler, K.E.N. George, K. Horsley, C. Heske, J.Z. Zhang,

Y. Li, Nano Lett. 11 (2011) 3503e3509.[19] D.K. Zhong, J. Sun, H. Inumaru, D.R. Gamelin, J. Am. Chem. Soc. 131 (2009)

6086e6087.[20] D.K. Zhong, M. Cornuz, K. Sivula, M. Grätzel, D.R. Gamelin, Energy Environ. Sci.

4 (2011) 1759e1764.[21] S.D. Tilley, M. Cornuz, K. Sivula, M. Grätzel, Angew. Chem. Int. Ed. 49 (2010)

6405e6408.[22] C.J. Sartoretti, B.D. Alexander, R. Solarska, I.A. Rutkowska, J. Augustynski,

J. Phys. Chem. B 109 (2005) 13685e13692.[23] I. Cesar, K. Sivula, A. Kay, R. Zboril, M. Grätzel, J. Phys. Chem. C 113 (2008)

772e782.[24] G. Yan, M. Zhang, J. Hou, J. Yang, Mater. Chem. Phys. 129 (2011) 553e557.[25] Y. Ling, G. Wang, D.A. Wheeler, J.Z. Zhang, Y. Li, Nano Lett. 11 (2011) 2119e

2125.[26] A. Duret, M. Grätzel, J. Phys. Chem. B 109 (2005) 17184e17191.[27] R.H. Goncalves, B.H.R. Lima, E.R. Leite, J. Am. Chem. Soc. 133 (2011) 6012e

6019.[28] Y. Lin, S. Zhou, S.W. Sheehan, D. Wang, J. Am. Chem. Soc. 133 (2011) 2398e

2401.[29] A. Kay, I. Cesar, M. Grätzel, J. Am. Chem. Soc. 128 (2006) 15714e15721.[30] Y.S. Hu, A. Kleiman-Shwarsctein, A.J. Forman, D. Hazen, J.N. Park,

E.W. McFarland, Chem. Mater. 20 (2008) 3803e3805.[31] D. Bersani, P.P. Lottici, A. Montenero, J. Raman Spectrosc. 30 (1999) 355e360.[32] D.L.A. de Faria, S.V. Silva, M.T. de Oliveira, J. Raman Spectrosc. 28 (1997)

873e878.[33] J.A. Glasscock, P.R.F. Barnes, I.C. Plumb, N. Savvides, J. Phys. Chem. C 111

(2007) 16477e16488.[34] T. Fujii, F.M.F. de Groot, G.A. Sawatzky, Phys. Rev. B 59 (1999) 3195e3202.[35] C.R. Parkinson, M. Walker, C.F. McConville, Surf. Sci. 545 (2003) 19e33.[36] F. Le Formal, N. Tétreault, M. Cornuz, T. Moehl, M. Grätzel, K. Sivula, Chem. Sci.

2 (2011) 737e743.[37] H.P. Maruska, A.K. Ghosh, Sol. Energy 20 (1978) 443e458.[38] M. Zhang, W. Luo, Z. Li, T. Yu, Z. Zou, Appl. Phys. Lett. 97 (2010) 042105.