Photochemical water splitting offers a useful solution for efficient energy conversion into hydrogen gas. Hematite has been focused on this purpose as the photoanode due to the advantages of low-cost, chemical stability, and suitable bandgap. The photocatalytic ability, however, is limited by the short-lived carriers and lack of photoresponse in the near infrared (NIR) region. As a solution, combining hematite with a noble metal can enhance the photocatalytic performance toward longer wavelength. Gold nanoparticles indicate characteristic absorption in the visible and NIR regions and photo-induced injection of electrons into the semiconductor. In this study, a hybrid material of hematite photoanodes with gold nanostructures was fabricated and the carrier dynamics under NIR excitation was elucidated by femtosecond transient absorption spectroscopy. The observed strong positive absorption under NIR excitation of Au nanorods (NRs) on the hematite anode indicated an increased electron density in hematite due to electron transfer from AuNRs, demonstrating efficient charge carrier generation in hematite by the decorated gold nanostructure.

In recent decades, the energy problem is an important issue in the world, and the clean technology to produce renewable energy has been requested to reduce the use of organic fossil fuel. Photoelectrochemical (PEC) water splitting utilizing inorganic semiconductors has been widely studied to harvest solar energy.1,2 Under solar illumination, water oxidation for oxygen generation is induced by photogenerated holes at the photoanode. In contrast, photogenerated electrons transfer to the counter electrode to produce hydrogen. Various metal oxide semiconductors can be utilized as a promising material for water oxidation due to the suitable band position. Many researchers have made efforts to develop efficient PEC systems. In particular, hematite (α-Fe2O3) has been studied as a photoanode material for solar water splitting due to its resource abundance and chemical stability.3–6 Moreover, hematite has an efficient bandgap (2.1 eV) and the high theoretical solar-to-hydrogen (STH) conversion efficiency of 12.9%, which gives a higher value than typical other semiconductors such as TiO2 and ZnO.7 However, several disadvantages are limiting the ability of photoanodes because of serious carrier recombination induced by poor conductivity and the short hole diffusion length (2–10 nm).8–10 In addition, the slow oxygen evolution kinetic rate needs high applied voltage.11,12 Therefore, several strategies such as nanostructure engineering and element doping have been studied to overcome these drawbacks.13–17 

On the other hand, the semiconductor material composited with noble metal nanoparticles can also enhance the photocatalytic performance by localized surface plasmon resonance (LSPR). This phenomenon induces ultrafast electron transfer and efficient carrier separation at the surface between the metal and the semiconductor, which is a beneficial method. In particular, gold nanoparticles (AuNPs) indicate chemical stability, characteristic absorption in the visible region, and injection of electrons into the semiconductor.18–22 In addition, AuNPs can be expected as a charge trapping state to escape from carrier recombination. Therefore, AuNPs are considered a suitable candidate for impressive enhancement of photocatalytic activity. Moreover, the shape of gold nanostructures affects their optical, electric, and catalytic natures. Gold nanorods (AuNRs) denote two plasmonic peaks associated with the oscillation of electrons along the longitudinal and transverse axes, and these positions depend on their aspect ratio.23,24 Some researchers have suggested the possibility that PEC performance of hematite photoanodes decorated with gold nanostructures exhibit high carrier conversion efficiency.25–28 

Transient absorption spectroscopy (TAS) is a very useful and powerful technique to directly monitor the photogenerated charge carriers (electrons and holes) in photocatalytic materials, which has allowed us to obtain the carrier dynamics from TAS data and effective material design guide.29–33 

Some hybrid materials that used gold nanoparticles for absorption of the visible region and high performance such as TiO2 were published.20,34 However, the reports of semiconductors that can absorb the near-infrared (NIR) region are less, and these ultrafast carrier dynamics have not been studied. In this work, we prepared hematite photoanodes decorated with gold nanostructures and revealed ultrafast carrier dynamics induced by surface plasmon resonance by transient absorption spectroscopy.

Synthesis of the hematite photoanode was made as follows according to a procedure reported previously.3 A FTO (fluorine-doped tin oxide) glass substrate was placed in the Teflon lined autoclave with the FTO side facing the wall of the liner. 19.5 ml growth solution containing 75 mM Fe2O3·6H2O, 0.5M NaNO3, and 0.5 ml ethanol was poured into the Teflon lined autoclave, followed by a heating process (4 h, 100 °C). The formed hematite anode was rinsed with deionized water and then finally annealed in N2 (1 h, 650 °C).

Preparation of the gold nanorods is described below.24 A solution of 0.1M hexadecyltrimethylammonium bromide (CTAB) dissolved in 5 ml of pure water was ultrasonicated at 40 °C until the solution becomes clear. After cooling down to room temperature, 5 ml of 0.5 mM tetrachloroauric acid (HAuCl4) solution was added to CTAB solution under intensive magnetic stirring at 27 °C. To obtain the gold seed, 0.6 ml of 0.01M sodium borohydride (NaBH4) solution was added to the CTAB solution, followed by stirring for 20 min. On the other hand, 5 ml of pure water containing 50 mM CTAB and 0.04M hydroquinone was dissolved at 40 °C using ultrasonication as a growth solution, followed by cooling down the solution to 27 °C. 0.2 ml of 4 mM silver nitrate (AgNO3) solution was added to growth solution, followed by the addition of 5 ml of HAuCl4 solution with magnetic stirring. Finally, 12 µl of gold seed solution was added to the growth solution with magnetic stirring for 30 min, followed by dividing the suspension and gold nanorods by centrifuging. Au decoration was carried out by adding gold nanorod solution drop wise on the hematite photoanode, followed by heating at 100 °C for 10 h.

The surface morphologies of the hematite nanomaterial were characterized by field-emission scanning electron microscopy (FE-SEM, HITACHI S4700). Steady state absorption spectra measurement was carried out with an absorption spectrophotometer (JASCO, V-670). Femtosecond pump–probe transient absorption spectroscopy was employed with an amplified Ti:sapphire laser (800 nm wavelength, 130 fs FWHM pulse width, 0.8 mJ/pulse intensity, 1 kHz repetition, Spectra Physics, Hurricane). We used the fundamental beam or second harmonic light from an amplified Ti:sapphire laser at a 500 Hz modulation frequency as a pump beam. A typical pump power of 1.0 mW corresponds to 1.0 mJ/cm2. In contrast, the white-light continuum generated by focusing the fundamental beam onto a sapphire plate (2 mm thick) was used as a probe beam. The probe beam was focused at the center of the pump light (∼0.3 mm diameter) on the specimen, and the transmitted probe beam was detected by a Si photodiode after passing through a monochromator (Acton Research, SpectraPro-150).

The absorption spectrum of AuNR solution indicates two plasmon peaks corresponding to the oscillation of electrons along the longitudinal and the transverse axes, as shown in Fig. 1(a). The stronger peak at 800 nm of AuNRs matches the fundamental beam of the Ti:sapphire laser for sufficient absorption and clear elucidation of ultrafast carrier dynamics in the hematite photoanode decorated with AuNRs (AuNRs/Hematite). As shown in Figs. 1(b) and 1(c), FE-SEM images of pristine hematite and AuNRs/Hematite reveal the modification of AuNRs on the hematite surface.

FIG. 1.

(a) Absorption spectrum of AuNR solution. FE-SEM images of (b) pristine hematite and (c) AuNRs/hematite.

FIG. 1.

(a) Absorption spectrum of AuNR solution. FE-SEM images of (b) pristine hematite and (c) AuNRs/hematite.

Close modal

Figures 2(a) and 2(b) show transient absorption (TA) decay profiles as a pump intensity dependence probed at 580 nm of (a) AuNRs/hematite excited at 800 nm and (b) hematite excited at 400 nm. Under both 400 nm and 800 nm pump wavelengths, TA decays indicate similar kinetics of photogenerated carriers. The similar decay processes at different pump intensities indicate that photogenerated carriers recombine through geminate electron–hole recombination so that second order bulk electron–hole recombination is not observed in our pump power range. In addition, the TA signal peak (y) vs pump intensity (x) was calculated for analysis of each normalized factor and fitted by the power function as follows:

y=Axpow.

We obtained a non-liner correlation of the TA peak intensity for AuNRs/hematite from pump intensity dependence, as shown in Fig. 2(c). On the other hand, the TA peak of the pristine hematite excited at 400 nm exhibited rather liner relation, as shown in Fig. 2(d). In the case of 400 nm excitation, hematite was excited through one-photon bandgap excitation due to the exponent of 0.9. In contrast, the exponent of 800 nm excitation was 1.39, which suggests including two-photon bandgap excitation of hematite and one-photon plasmon excitation. These results exhibit that different photoexcitation processes were observed for hematite photoanodes irradiated at the fundamental beam at 800 nm and the second harmonic beam at 400 nm from the laser.

FIG. 2.

TA decays as a pump intensity dependence probed at 580 nm of (a) AuNRs/hematite excited at 800 nm and (b) hematite excited at 400 nm. TA signal peak vs pump intensity for (c) AuNRs/hematite excited at 800 nm and (d) hematite excited at 400 nm.

FIG. 2.

TA decays as a pump intensity dependence probed at 580 nm of (a) AuNRs/hematite excited at 800 nm and (b) hematite excited at 400 nm. TA signal peak vs pump intensity for (c) AuNRs/hematite excited at 800 nm and (d) hematite excited at 400 nm.

Close modal

TA spectra of pristine hematite excited at 400 nm (0.7 mW) showed a transient absorption maximum at 580 nm (the shortest wavelength in the measured range) near the bandgap, as shown in Fig. 3(a). Similar spectra were observed in some previous fs-TAS study, and the broadband is assigned to the sum of photogenerated electrons and holes.33 In general, more mobile charges allow absorption in the longer wavelength region, so the observed decay behavior being faster at longer wavelength should include the charge trapping process. On the other hand, TA spectra of AuNRs/hematite excited at 800 nm (2.0 mW) indicated a broader absorption signal than those of pristine hematite. This expansion of absorption band was induced by modification of AuNRs to enable NIR region absorption. TA decay of pristine hematite indicates faster decay (<10 ps) and retardation (10–500 ps), which reflects mobile/short-lived and trapped/long-lived species, respectively. A large decrease in the amplitude to 10 ps suggests a significant loss of the activity of photogenerated carriers. In contrast, TA decay of AuNRs/hematite was observed as a rather monotonous decay, which includes the dynamics contributed by excited AuNRs/hematite at the longitudinal mode.

FIG. 3.

TA spectra of (a) hematite excited at 400 nm and (b) AuNRs/hematite excited at 800 nm. TA decays of (c) hematite excited at 400 nm and (d) AuNRs/hematite excited at 800 nm.

FIG. 3.

TA spectra of (a) hematite excited at 400 nm and (b) AuNRs/hematite excited at 800 nm. TA decays of (c) hematite excited at 400 nm and (d) AuNRs/hematite excited at 800 nm.

Close modal

In order to discuss electron transfer dynamics from gold nanorods to hematite, differences in spectra were calculated by subtracting TA spectra of pristine hematite excited at 400 nm from those of AuNRs/hematite excited at 800 nm, as shown in Figs. 3(a) and 3(b). Two spectra at 500 ps are similar in shape with each other, which suggests that the injected electron transfers backward from hematite to AuNRs and recombination is completed within 500 ps. In other words, despite the different excitation processes, the same dynamics contributed by the photogenerated carriers through bandgap excitation were observed. The factor used to normalize the spectra at 500 ps for the 580 nm TA intensities was applied to remove the effect of bandgap excitation of hematite in AuNRs/hematite at each delay time. Figure 4 shows the difference in TA spectra and indicates the absorption peak at 600 nm within 10 ps and broad shape. In addition, these spectra reached nearly zero after 100 ps. It is known that electrons in the semiconductor give broad absorption due to intra-band transition within the conduction band.20 These results denote that the electrons injected from AuNRs to hematite underwent a recombination with photogenerated holes in AuNRs within 100 ps. Electron injection time seems to be within 250 fs (the time-resolution), judging from the prompt rise observed at all probe wavelengths in Fig. 3(d). To our knowledge, this is the first observation of plasmon-induced electron transfer dynamics from AuNRs to the metal semiconductor under NIR excitation.

FIG. 4.

Difference of TA spectra between 400 nm excited hematite and 800 nm excited AuNRs/hematite after normalization at 500 ps delay time.

FIG. 4.

Difference of TA spectra between 400 nm excited hematite and 800 nm excited AuNRs/hematite after normalization at 500 ps delay time.

Close modal

Scheme 1 illustrates the ultrafast carrier dynamics of AuNRs/hematite excited at 800 nm. Both hematite and AuNRs could absorb the pump beam and generate the electron–hole pairs through two-photon bandgap excitation and one-photon plasmon excitation, respectively. Photogenerated electrons at AuNRs transfer to hematite within 250 fs, followed by a recombination with photogenerated holes at hematite as back electron transfer within 100 ps. Our result indicates that hematite photoanodes decorated with gold nanostructures can be expected to give additional PEC performance in the NIR region. In the case of an actual photoanode in the electrolyte, a band bending would form upward from the bulk to the surface since hematite is an n-type semiconductor. This would assist charge separation of the injected electrons from AuNR. Co-catalysts such as IrO2, Co-Pi, and CoOx can also help in charge separation.3 Actually, hole transfer to CoOx is reported to be within 1 ps.35 

SCHEME 1.

Schematic illustration of the ultrafast carrier dynamics of AuNRs/hematite excited at 800 nm.

SCHEME 1.

Schematic illustration of the ultrafast carrier dynamics of AuNRs/hematite excited at 800 nm.

Close modal

Ultrafast dynamics of hematite photoanodes decorated with gold nanostructures exhibited two mechanisms under the 800 nm excitation process. Electron–hole pairs of hematite were generated through two-photon bandgap excitation at 800 nm. AuNRs could also generate photogenerated carriers through the longitudinal mode plasmon excitation. The photogenerated electrons transfer to hematite within 1 ps and injected electrons return to AuNRs within 100 ps thorough carrier recombination. Hematite decorated with gold nanostructures can be expected to offer additional PEC performance in the NIR region as well as some applied technologies to utilize the wide-spectral range from the whole visible region to the NIR region.

1.
M.
Gratzel
,
Nature
414
,
338
(
2001
).
2.
M. G.
Walter
,
E. L.
Warren
,
J. R.
McKone
,
S. W.
Boettcher
,
Q.
Mi
,
E. A.
Santori
, and
N. S.
Lewis
,
Chem. Rev.
110
,
6446
(
2010
).
3.
Y.-J.
Chen
and
L.-Y.
Chen
,
Appl. Energy
164
,
924
(
2016
).
4.
Z.
Zhang
,
Md. F.
Hossain
, and
T.
Takahashi
,
Appl. Catal. B
95
,
423
(
2010
).
5.
B.
Klahr
and
T.
Hamann
,
J. Phys. Chem. C
118
,
10393
(
2014
).
6.
P.
Hiralal
,
S.
Saremi-Yarahmadi
,
B. C.
Bayer
,
H.
Wang
,
S.
Hofmann
,
K. G.
Upul Wijayantha
, and
G. A. J.
Amaratunga
,
Sol. Energy Mater. Sol. Cells
95
,
1819
(
2011
).
7.
A. B.
Murphy
,
P. R. F.
Barnes
,
L. K.
Randeniya
,
I. C.
Plumb
,
I. E.
Grey
,
M. D.
Horne
, and
J. A.
Glasscock
,
Int. J. Hydrog. Energy
31
,
1999
(
2006
).
8.
T.
Lindgren
,
H.
Wang
,
N.
Beermann
,
L.
Vayssieres
,
A.
Hagfeldt
, and
S.-E.
Lindquist
,
Sol. Energy Mater. Sol. Cells
71
,
231
(
2002
).
9.
H.
Dotan
,
K.
Sivula
,
M.
Grätzel
,
A.
Rothschild
, and
S. C.
Warren
,
Energy Environ. Sci.
4
,
958
(
2011
).
10.
U.
Bjoerksten
,
J.
Moser
, and
M.
Grätzel
,
Chem. Mater.
6
,
858
(
1994
).
11.
K.
Sivula
,
F.
Le Formal
, and
M.
Grätzel
,
ChemSusChem
4
,
432
(
2011
).
12.
K. G.
Upul Wijayantha
,
S.
Saremi-Yarahmadi
, and
L. M.
Peter
,
Phys. Chem. Chem. Phys.
13
,
5264
(
2011
).
13.
G. X.
Pei
,
J. H. J.
Wijten
, and
B. M.
Weckhuysen
,
Phys. Chem. Chem. Phys.
20
,
9806
(
2018
).
14.
B.
Klahr
,
S.
Gimenez
,
F.
Fabregat-Santiago
,
J.
Bisquert
, and
T. W.
Hamann
,
J. Am. Chem. Soc.
134
,
16693
(
2012
).
15.
M.
Barroso
,
A. J.
Cowan
,
S. R.
Pendlebury
,
M.
Grätzel
,
D. R.
Klug
, and
J. R.
Durrant
,
J. Am. Chem. Soc.
133
,
14868
(
2011
).
16.
K. J.
McDonald
and
K.-S.
Choi
,
Chem. Mater.
23
,
4863
(
2011
).
17.
C.
Miao
,
S.
Ji
,
G.
Xu
,
G.
Liu
,
L.
Zhang
, and
C.
Ye
,
ACS Appl. Mater. Interfaces
4
,
4428
(
2012
).
18.
S.
Li
,
J.
Zhang
,
Md. G.
Kibria
,
Z.
Mi
,
M.
Chaker
,
D.
Ma
,
R.
Nechache
, and
F.
Rosei
,
Chem. Commun.
49
,
5856
(
2013
).
19.
S.
Hashimoto
,
D.
Werner
, and
T.
Uwada
,
J. Photochem. Photobiol. C
13
,
28
(
2012
).
20.
L.
Du
,
A.
Furube
,
K.
Hara
,
R.
Katoh
, and
M.
Tachiya
,
J. Photochem. Photobiol. C
15
,
21
(
2013
).
21.
S.
Link
and
M. A.
El-Sayed
,
J. Phys. Chem. B
103
,
8410
(
1999
).
22.
C.
Bauer
,
J.-P.
Abid
,
D.
Fermin
, and
H. H.
Girault
,
J. Chem. Phys.
120
,
9302
(
2004
).
23.
S.
Picciolini
,
D.
Mehn
,
I.
Ojea-Jiménez
,
F.
Gramatica
, and
C.
Morasso
,
J. Vis. Exp.
114
,
e54319
(
2016
).
24.
L.
Vigderman
and
E. R.
Zubarev
,
Chem. Mater.
25
,
1450
(
2013
).
25.
L.
Wang
,
H.
Hu
,
N. T.
Nguyen
,
Y.
Zhang
,
P.
Schmuki
, and
Y.
Bi
,
Nano Energy
35
,
171
(
2017
).
26.
J.
Zhang
,
G.
Chen
,
M.
Chaker
,
F.
Rosei
, and
D.
Ma
,
Appl. Catal. B
132-133
,
107
(
2013
).
27.
P. S.
Shinde
,
Su. Y.
Lee
,
J.
Ryu
,
S. H.
Choi
, and
J. S.
Jang
,
Chem. Commun.
53
,
4278
(
2017
).
28.
M. G.
Koren
,
H.
Dotan
, and
A.
Rothschild
,
J. Phys. Chem. C
120
,
15042
(
2016
).
29.
S. R.
Pendlebury
,
A. J.
Cowan
,
M.
Barroso
,
K.
Sivula
,
J.
Ye
,
M.
Grätzel
,
D. R.
Klug
,
J.
Tang
, and
J. R.
Durrant
,
Energy Environ. Sci.
5
,
6304
(
2012
).
30.
F.
Le Formal
,
S. R.
Pendlebury
,
M.
Cornuz
,
S. D.
Tilley
,
M.
Graẗzel
, and
J. R.
Durrant
,
J. Am. Chem. Soc.
136
,
2564
(
2014
).
31.
M.
Barroso
,
S. R.
Pendlebury
,
A. J.
Cowan
, and
J. R.
Durrant
,
Chem. Sci.
4
,
2724
(
2013
).
32.
S. R.
Pendlebury
,
M.
Barroso
,
A. J.
Cowan
,
K.
Sivula
,
J.
Tang
,
M.
Grätzel
,
D.
Klug
, and
J. R.
Durrant
,
Chem. Commun.
47
,
716
(
2011
).
33.
S. R.
Pendlebury
,
X.
Wang
,
F.
Le Formal
,
M.
Cornuz
,
A.
Kafizas
,
S. D.
Tilley
,
M.
Grätzel
, and
J. R.
Durrant
,
J. Am. Chem. Soc.
136
,
9854
(
2014
).
34.
K.
Ueno
,
T.
Oshikiri
, and
H.
Misawa
,
ChemPhysChem
17
,
199
(
2016
).
35.
R. B.
Singh
,
H.
Matsuzaki
,
Y.
Suzuki
,
K.
Seki
,
T.
Minegishi
,
T.
Hisatomi
,
K.
Domen
, and
A.
Furube
,
J. Am. Chem. Soc.
136
,
17324
(
2014
).