Electron transport and visible light absorption in a plasmonic photocatalyst based on strontium niobate

Abstract

Semiconductor compounds are widely used for photocatalytic hydrogen production applications, where photogenerated electron-hole pairs are exploited to induce catalysis. Recently, powders of a metallic oxide (Sr_1-x NbO₃ , 0.03<x<0.20) were reported to show competitive photocatalytic efficiencies under visible light, which was attributed to interband absorption. This discovery expanded the range of materials available for optimized performance as photocatalysts. Here we study epitaxial thin films of SrNbO_3+δ and find that their bandgaps are ∼4.1 eV. Surprisingly, the carrier density of the conducting phase exceeds 10²² cm^-3 and the carrier mobility is only 2.47 cm² V^-1 s^-1 . Contrary to earlier reports, the visible light absorption at 1.8 eV (∼688 nm) is due to the plasmon resonance, arising from the large carrier density. We propose that the hot electron and hole carriers excited via Landau damping (during the plasmon decay) are responsible for the photocatalytic property of this material under visible light irradiation.

ARTICLE

Received 8 Feb 2016 | Accepted 27 Feb 2017 | Published 19 Apr 2017

D.Y. Wan1,2, Y.L. Zhao1, Y. Cai3, T.C. Asmara4, Z. Huang1, J.Q. Chen1, J. Hong5, S.M. Yin5, C.T. Nelson6, M.R. Motapothula1, B.X. Yan1,2, D. Xiang7, X. Chi2,4, H. Zheng3,6, W. Chen2,7, R. Xu5, Ariando1,2,A. Rusydi1,2,4, A.M. Minor3,6, M.B.H. Breese1,2,4, M. Sherburne3, M. Asta3, Q.-H. Xu1,7 & T. Venkatesan1,2,8,9,10

Semiconductor compounds are widely used for photocatalytic hydrogen production applications, where photogenerated electronhole pairs are exploited to induce catalysis. Recently, powders of a metallic oxide (Sr1 xNbO3, 0.03oxo0.20) were reported to show competitive photocatalytic efciencies under visible light, which was attributed to interband absorption. This discovery expanded the range of materials available for optimized performance as photocatalysts. Here we study epitaxial thin lms of SrNbO3 d and nd that their bandgaps are B4.1 eV. Surprisingly, the carrier density of the conducting phase exceeds 1022 cm 3 and the carrier mobility is only 2.47 cm2 V 1 s 1. Contrary to earlier reports, the visible light absorption at 1.8 eV (B688 nm) is due to the plasmon resonance, arising from the large carrier density. We propose that the hot electron and hole carriers excited via

Landau damping (during the plasmon decay) are responsible for the photocatalytic property of this material under visible light irradiation.

1 NUSNNI-NanoCore, National University of Singapore, Singapore 117411, Singapore. 2 Department of Physics, National University of Singapore, Singapore 117551, Singapore. 3 Department of Materials Science and Engineering, University of California, Berkeley, California 94720, USA. 4 Singapore Synchrotron Light Source, National University of Singapore, Singapore 117603, Singapore. 5 School of Chemical and Biomedical Engineering, Nanyang Technological University, Singapore 637459, Singapore. 6 Materials Science Division, Lawrence Berkeley National Laboratory, Berkeley, California 94720, USA.

7 Department of Chemistry, National University of Singapore, Singapore 117543, Singapore. 8 NUS Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore 117456, Singapore. 9 Department of Material Science and Engineering, National University of Singapore, Singapore 117575, Singapore. 10 Department of Electrical and Computer Engineering, National University of Singapore, Singapore 117583, Singapore. Correspondence and requests for materials should be addressed to Y.L.Z. (email: mailto:[email protected]

Web End [email protected] ) or to T.V. (email: mailto:[email protected]

Web End [email protected] ).

NATURE COMMUNICATIONS | 8:15070 | DOI: 10.1038/ncomms15070 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 1

DOI: 10.1038/ncomms15070 OPEN

Electron transport and visible light absorption in a plasmonic photocatalyst based on strontium niobate

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15070

Converting solar energy into chemical energy (for example, splitting water by sunlight) with the aid of photocatalysts is a promising way to reduce the increasing demand

for fossil fuels110. Very few oxide semiconductors have been used as photocatalysts, as they need to be chemically robust and their bandgap should be neither too wide nor narrow to absorb sunlight in the visible range efciently and also satisfy the minimum energy requirement (1.23 eV theoretically but 41.9 eV experimentally) for splitting water into hydrogen and oxygen1113. Large bandgap oxides (such as TiO2) are used in photocatalytic water splitting either by reducing their optical bandgap to absorb visible light or incorporating visible light absorbers such as organic dyes, low bandgap quantum absorbers or metal nanostructures14,15 in the host. In the former case, cationic or anionic doping, or a combination of both, is typically applied to narrow the bandgap1619. Unfortunately, in most cases, the bandgap change achieved is small mainly due to the fact that the discrete defect states are normally very close to the band edges2022. In the latter case, hot electrons in the visible light absorber inject into the conduction band of large bandgap oxides, which are subsequently used to reduce water to hydrogen gas. Among the visible light absorbers, metal (for example, Au and Ag) nanostructures are special as the hot electrons generated by the decay of visible light excitation of surface plasmon resonance can be injected into a large bandgap semiconductor such as TiO2 (refs 23,24). However, the materials used for enhancing the photocatalytic activity by using surface plasmon resonance are mainly Au, Ag nanostructures, which are not low-cost materials.

Recently a red metallic oxide Sr1 xNbO3 (0.03oxo0.20)

(in the form of powders) was used in photocatalytic hydrogen evolution (see Supplementary Note 1 and Supplementary Fig. 12) and the authors proposed a special band structure, in which the electronhole pairs come from the optical transition from the metallic conduction band (about 1.9 eV above the valence band) to a higher level unoccupied band25,26. In these reports, the visible light absorption was attributed to the electrons interband transitions and the electronhole pair separation was attributed to the assumption of high carrier mobility, although only temperature-dependent conductivity was measured. As both charge carrier density and mobility contribute to the conductivity of a sample, simply assuming a large mobility for a highly conductive material may lead to a wrong conclusion. It has also been reported by Oka et al.27 that the highest conducting SrNbO3 thin lm grown on KTaO3 substrate can have a high carrier density of 1022 cm 3 and relatively low mobility of 16 cm2 V 1 s 1 at room temperature. Besides carrier density and mobility, understanding the static optical properties (absorption, refractive index and loss function) and dynamics of the carriers would be crucial for understanding the details of the photo-catalytic process. The optical bandgap obtained from KubelkaMunk transformation of the reectance spectrum of the powder form Sr1 xNbO3 is inaccurate, as it neglects the

plasmonic absorption. The proper way to measure the optical bandgap is to obtain the complex dielectric function, by using KramersKronig-transformed reectivity or spectroscopic ellipsometry. Epitaxial thin lms are required for measuring the intrinsic mobility, as the grain boundaries in the lm are much less than in powders enabling accurate Hall measurement of carriers and hence mobility. Using thin lms also allows us to measure the transmittance, reectance and ellipsometry spectra accurately, which can give proper optical and plasmonic absorptions of this material. Furthermore, the dynamical process of the hot electron can be investigated by femtosecond time-resolved transient absorption (TA) spectroscopy.

Here we have prepared SrNbO3 d lms by pulsed laser

deposition (PLD) at various oxygen partial pressures on top of

insulating LaAlO3 substrates and compared their optical spectra, electronic transport and carrier dynamics properties. We found the optical bandgap to be around 4.1 eV, which was almost independent of the oxygen content, although the crystal structure changed from pseudo-tetragonal perovskite to orthorhombic with increasing oxygen partial pressure from 5 10 6 Torr

to 1 10 4 Torr. The plasmon peak for the lm grown at

5 10 6 Torr was found at about 1.8 ev (688 nm). The high

conductivity (B104 S cm 1) of the sample prepared at low

pressure is mainly contributed by the high charge carrier density (B1022 cm 3) rather than its mobility (2.47 cm2 V 1 s 1)

at room temperature. Thus, we believe that the photocatalytic activity of SrNbO3 d under visible to near-infrared (NIR)

irradiation is due to the hot electrons generated from the decay of the plasmon in SrNbO3 instead of interband absorption transition. Thus, SrNbO3 d represents an extraordinary material

system, which has a large bandgap of 4.1 eV but a degenerate conduction band with a large carrier density exceeding 1022 electrons per cm3, which leads to strong useful plasmonic effects.

ResultsCharacterization of strontium niobate thin lms. The X-ray diffraction spectrum of the lm deposited at oxygen partial pressure of 5 10 6 Torr is shown in Supplementary Fig. 1.

The y 2y scan indicates the lms lattice parameter along

out-of-plane [001] direction as 4.10 . The full width at half maximum of the rocking curve is measured to be 0.71, which is acceptable by considering the large lattice mismatch between the lm and the substrate (LaAlO3 forms in pseudocubic perovskite structure at room temperature with lattice constant 3.79 ). The reciprocal space maps (RSMs) of ( 103) and (103) [(013) and

(013)] are symmetric with respect to the out-of-plane [001] axis, suggesting orthogonality of [001] and [100] ([001] and [010]) axes. Despite their relatively broad RSM spots, the in-plane parameters are obtained as 4.04 equally. Hence, the lm forms in the tetragonal-like structure on LaAlO3 substrate with a large strain near the interface. The strain effect can be clearly seen in the local high-resolution transmission electron microscopy image of the lattice (Supplementary Fig. 1c). The highlighted open burgess circuit indicates an a[100]p-type edge dislocation core, where the extra plane on the LaAlO3 side indicates a compressive mist of the lm. As the oxygen partial pressure increases to 1 10 4 Torr, a small shift of the lm peaks towards low angles

can be observed in the y 2y scan and the full width at half

maximum of the rocking curve increases (Supplementary Fig. 2). Structural changes from tetragonal to orthorhombic were observed in the local high-resolution transmission electron microscopy images (Supplementary Fig. 3). The orthorhombic structure is close to the reported structure of Sr2Nb2O7 (bulk a 3.933 , b 26.726 , c 5.683 ), which has an equivalent

tetragonal structure with lattice parameters a b 3.901 and

c 3.933 . The decreasing of the out-of-plane and in-plane

lattice parameters with oxygen partial pressure was obtained from the electron diffraction pattern (Supplementary Fig. 3 and Supplementary Table 1). At the intermediate oxygen partial pressure, mixed structure exists in the lm.

It was reported that Sr content strongly determines the crystal structure of the non-stoichiometric SrNbO3 phase28,29. Here the elemental content of the lms deposited at different oxygen partial pressures are precisely studied (Supplementary Fig. 4 and Supplementary Table 2). Within the detection limit of proton-induced X-ray emissions and Rutherford backscattering spectroscopy, the lm deposited at 5 10 6 Torr has

Sr:Nb:O 1:1:3. Films deposited at higher oxygen pressures

show the same Sr/Nb ratio but increased oxygen content.

2 NATURE COMMUNICATIONS | 8:15070 | DOI: 10.1038/ncomms15070 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15070 ARTICLE

a b c

5106 Torr 1105 Torr

104

103

102

101

100

101

102

103

104

Carrier density (1021cm3)

10

Resisitivity ( cm)

Mobility (cm2 V1 s1)

100

101

1

102

0 0

50 100 100

150

200 200

250

300 300 0 100 200 300

Temperature (K) Temperature (K) Temperature (K)

Figure 1 | Electronic transport properties of SrNbO3 d lms. Temperature-dependent transport properties of the lms prepared at various oxygen

pressures: (a) resistivity, (b) mobile electron density obtained from Hall measurement and (c) electron mobility of the lms.

Electronic transport property of SrNbO3 d. The conducting

property of the lms show a transition from metallic to semiconductor transport behaviour when the deposition oxygen partial pressure is increased (Fig. 1a). The free electron density of the most conductive sample (prepared at 5 10 6 Torr) reaches

1022 cm 3, which agrees with the reported data of SrNbO3 thin lm grown on KTaO3 (ref. 27) (Fig. 1b) indicative of a degenerate Fermi level. In contrast, the electron mobility is only2.47 cm2 V 1 s 1 at room temperature, which is not signicant compared with other oxides (for example, TiO2 and BaSnO3) and semiconductors3033 (Fig. 1c). The room temperature mobility of our thin lm is smaller than that of SrNbO3 grown on KTaO3 (B16 cm2 V 1 s 1) and this can be attributed to strain in the lm as the lattice mismatch between SrNbO3 and KTaO3 (mismatch 0.85%) is smaller compared with that between

SrNbO3 and LaAlO3 (ref. 27) (mismatch 5.77%). However,

we can still conclude that the high conductivity of this material is mainly attributed to the high carrier density and not just the carrier mobility. Both the free electron density and mobility are almost independent of the measurement temperature, which indicates that the electronic transport property of SrNbO3 is

dominated by the free electrons in the conduction band and electronelectron scattering. The large free electron density and normal electron mobility of strontium niobate at room temperature imply the absence of signicant internal electric eld to avoid electronhole recombination and the interband transition model proposed in the previous report34 is not suitable. As the oxygen content increases in the lm, the sample becomes more insulating. Both the charge carrier density and the mobility decreases with oxygen partial pressure, which is consistent with the observed two crystal structures of the materials35. Metallic SrNbO3 forms in tetragonal perovskite structure at 5 10 6 Torr

and almost insulating SrNbO3.5 forms in the orthorhombic structure at 1 10 4 Torr. At the intermediate pressure, the

semiconductor forms in cermet structure. The TEM images show excess oxygen forming sheet-like defect structures leaving regions with 113 stoichiometry in the midst (Supplementary Fig. 3). These insulating oxygen sheets create a cermet-like structure with increasing oxygen uptake.

The photocatalytic hydrogen evolution performance is discussed in Supplementary Note 1 and shown in Supplementary Fig. 12 as the hydrogen evolution efciency is not the focus of this work.

UltravioletvisibleNIR spectra of SrNbO3 d. Figure 2a shows

the ultravioletvisibleNIR transmission spectra of SrNbO3 d

thin lms deposited at various oxygen pressure. A cutoff of the transmission edge is observed near 300 nm (Fig. 2a), which

indicates an optical bandgap of B4.1 eV (from Tauc plot indirect, Supplementary Fig. 5) and it is almost independent of the preparation oxygen pressure. The transmission above 600 nm of the lms continuously increases with oxygen partial pressure above 600 nm, which indicates absorption along with free carrier absorption in this wavelength range, where the latter is consistent with the metallic nature of the lms and powders25,36. Both the transmission and reection of the lm prepared at 5 10 6 Torr

were plotted, from which the accurate pure absorption spectrum could be obtained (Fig. 2b). As we can see in the reection spectrum, the reection above 700 nm is much larger than that below 600 nm, which is very similar to the reection behaviour of plasmonic materials. The minimum reection is located at around 600 nm, which could indicate the rough frequency of its plasmon resonance. Hence, the reection spectrum between 500 and 1,000 nm can be well tted by the DrudeLorentz Model and the corresponding plasmon frequency of the tting curve is1.65 eV (750 nm). The plasmon resonance in the strontium niobate is a result of its extremely large free electron density as measured in its electronic transport property. It should be noted that there is only one obvious absorption peak near 675 nm at the visible range in the absorbance spectrum.

Loss function of SrNbO3 and its plasmon energy position. The complex refractive index, o

n o

iko, and the loss

function, Im[e 1(o)], spectra of the 5 10 6 Torr sample

extracted from spectroscopic ellipsometry data are shown in Fig. 2c,d (for further details, see Supplementary Note 2, Supplementary Figs 68 and Supplementary Table 3). The extinction coefcient spectrum, k(o), of the sample (Fig. 2d)

shows that it has a Drude peak below 2 eV (typical of a metal) and a rst interband transition peak (indicating the bandgap of the lm) above 4.1 eV, consistent with its transmission spectrum (Fig. 2b). Between these two peaks, the k(o) is featureless, indicating the lack of major optical transitions within the 24.1 eV energy range. Meanwhile, the loss function spectrum of the sample (Fig. 2d) shows a large peak at B1.52.1 eV with a peak position of B1.8 eV (688 nm), indicating the energy position of the plasmon resonance37. The energy position of plasmon obtained from the ellipsometry spectroscopy is more accurate compared with that obtained by the tting of ultravioletvisible NIR reection spectrum. From n(o) and k(o) spectra, the normal-incident reectivity of the lm can be obtained using Fresnel equations, as shown in Fig. 2e. This reectivity is consistent with the spectrum measured by ultravioletvisible spectroscopy. From the reectivity, the KubelkaMunk function of the lm can be obtained (Fig. 2e) and the shape of the function resembles the previous reported results, with an apparent

NATURE COMMUNICATIONS | 8:15070 | DOI: 10.1038/ncomms15070 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 3

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15070

a b

100

0.8

3.02.5

90

Transmission (measured, %)

0.7

80

Absorbance (a.u.)

0.6

70

2.01.51.00.50.0

Percentage (%)

30

60

0.5

50

0.4

Absorbance (a.u)

40

0.3

0.2

20

10

0.1

400 600 800 1,000 1,200

Wavelength (nm)

1,400

1,400

0 400

600 800 1,000 1,200

0.0

Wavelength (nm)

c d e

3.0 Refractive index (n)

Extinction coefficient ([afii9837]) Loss function

0.80.60.40.2

4.5

2.5

[afii9837]

0.8

0.6 20

15 10 5 0

4.0

Normal-incident reflectivity

2.52.01.51.00.50.0

3.5

Loss function

3.0

Kubelka-munk function

2.0

1.5

1.0

0.5

0.0

6

n

Reflectivity

0.4

0.2

0.0

Kubelka-munk

0.0

1 6

5

4

3

2

1

Photon energy (eV)

6

2

3

4

5

1

2

3

4

5

Photon energy (eV) Photon energy (eV)

Figure 2 | The optical properties of SrNbO3 d lms. (a) UltravioletvisibleNIR spectra of SrNbO3d thin lms deposited at various oxygen partial

pressures, an absorption edge located at the wavelength of 300 nm can be observed. (b) The transmission, reection spectra of the lm deposited at 5 10 6 Torr, with the absorbance spectrum obtained as plot in red. Ellipsometry analysis of (c) the refraction index (n) and extinction coefcient (k).

(d) Loss function and (e) reectivity of SrNbO3 as a function of photon energy.

absorption edge at B2 eV. As there is no peak in the k(o) spectrum at around that energy, this absorption edge does not come from intra- or interband transition as previously reported. Instead, this absorption edge is plasmonic in origin, because it coincides with the plasmon peak at B1.8 eV in the loss function spectrum. The plasmon observed in ellipsometry spectra may be the surface Plasmon, because a similar peak near 675 nm can be observed in the ultravioletvisibleNIR absorption spectrum and in general, the volume plasmon is not possible to be directly coupled with the photon38,39. As the photocatalytic hydrogen evolution experiments were performed with powders, coupling of the photons to the plasmons is not an issue. As the calculated penetration depth of the plasmon is 80 nm, which is comparable to the particle size (B1001,000 nm), further experiments are needed to distinguish between surface versus bulk (Supplementary Fig. 9). Hereon, we will refer to this as just plasmon.

Band structure and density of states of SrNbO3 d. The energy band structures of SrNbO3, SrNbO3.4 and SrNbO3.5 are calculated using density functional theory (DFT) and shown in Fig. 3. In the calculations, the perovskite structure is assumed for the stoichiometric SrNbO3 compound and the extra oxygen atoms for the hyper stoichiometric compositions are assumed to order into planar defects, as illustrated by the structural gures in the left panels of Fig. 3. This structural model is consistent with electron microscopy data shown in Supplementary Fig. 3 (refs 40,41). The Fermi level of SrNbO3 is in the conduction band, which implies that this material is metallic even though there is a bandgap as large as 3.0 eV between the CB and the highest fully occupied band (B 1 band, reasonably close to our

ultravioletvisible spectrum considering that DFT calculations generally underestimate bandgap values). The Fermi level of SrNbO3.4 is located near the bottom of the conduction band; thus, the conductivity is poorer than that of SrNbO3, as there are fewer

states for the free carriers leading to a lower carrier density. These results are consistent with the experimental measurements. Unlike SrNbO3 and SrNbO3.4, the Fermi level of

SrNbO3.5 is located at the top of the valence band; thus, SrNbO3.5 is insulating and the lm should be transparent, consistent with experiments.

Ultrafast carrier dynamics in SrNbO3. To further understand the role of the plasmon in the catalytic process, time-resolved pumpprobe spectroscopy was used to characterize the carrier dynamics in SrNbO3. Figure 4a shows the various excitation wavelength-dependent differential reection (DR, DR/R) spectra with various time delay times. Two peaks located near 600 nm (positive) and 670 nm (negative) are observed in the DR spectra. It should be noted that the sign of DR signal would usually be opposite to the sign of differential transmission signal (Supplementary Fig. 10)42,43. Usually, the TA could be positive or negative because the pump pulse induces population or depopulation of particular states, which can decrease (photobleaching) or increase (photoinduced absorption) the absorption of the probe pulse43. However, this principle may not be applied to our DR spectra. As SrNbO3 is a metallic oxide, the heat capacity of electrons (Ce) is much smaller than that of the lattice (CL), for example, CeooCL and a short laser pulse can selectively heat the electrons as has been seen for gold nanoparticles44. It has been shown in this case that when the temperature of electrons increases, the intensity of plasmon band will decrease and its linewidth will increase, leading to bleaching of the plasmon band at the resonance and TA at the wings of the bleached spectrum45. Therefore, the negative peak near 670 nm can be attributed to the decrease (bleaching) of the plasmon band intensity as the 670 nm peak position is consistent with the plasmonic resonance peak measured by spectroscopic ellipsometry. The peak near 600 nm is the TA at the wings of the plasmon (derivative peak) due to the plasmon broadening

4 NATURE COMMUNICATIONS | 8:15070 | DOI: 10.1038/ncomms15070 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15070 ARTICLE

a b

2 1 0 1 2

Up

E-E f(eV)

E-E f(eV)

E-E f(eV)

Down sp

d

Total

Up

Down

3 4

2 1 0 1 2

SrNbO3

SrNbO3.4

SrNbO3.5

X S R

RT

T Z

T

Y

Y

20 10 0 10 20

DOS (states/eV)

c

d

Down sp

d

Total

Up

Every

5

layers

Every

4

layers

Up

Down

3 4

X

S

Z

T

20

40 0 20 40

DOS (states/eV)

e f

5 4 3 2 1

Up

Up

Down sp

d

Total

Down

0 1

X Z

T Y

S

RT

20

40 0 20 40

DOS (states/eV)

Figure 3 | Crystal structures and energy band structures with its density of state (DOS) of SrNbO(3d). (a) Distorted perovskite structure of SrNbO3.

O2 : small red sphere, Sr2 : large green sphere, Nb4 : small blue sphere. The unit cell is shown with the black solid line. (b) Band structure and DOS of SrNbO3, showing its metallic behaviour. (c) The layered structure of SrNbO3.4 with extra oxygen layers inserted every ve octahedral layers. The dash lines indicate where the extra oxygen layers are. Nb4 (small blue sphere) and Nb5 (small green sphere) are given different colours to show charge ordering in this composition. (d) Band structure and DOS of SrNbO3.4, with signicantly reduced carriers at the Fermi level. (e) The layered structure of SrNbO3.5, with an extra oxygen layer inserted every four octahedral layers. (f) Band structure and DOS of SrNbO3.5, showing its insulating behaviour. The calculation method is using DFT U (U 4 eV).

induced by increased temperature electrons (see Supplementary Note 3 and Supplementary Figs 10 and 11 for a more detailed discussion of the assignment). The other wing at the lower energies is not seen due to the limitation of our spectrometer.

After the conduction electrons are excited to the unoccupied states by the pump pulse, the electrons will form a transient broad non-thermalized (non-Fermi) distribution above the Fermi level44,46,47. These hot electrons have a large excess energy above the Fermi level, much larger than thermal excitations at ambient temperatures48,49. The hot electrons will exchange their energy with the lower-energy electrons via electronelectron scattering process and form a Fermi distribution characterized by the electron temperature (Te). The timescale of this process was reported to be around 100 fs to 1 ps50,51. The dynamics of the hot electrons within the rst 2.0 ps in SrNbO3 are shown in Supplementary Fig. 11. The excitation wavelength-dependent lifetimes of the hot electrons are around 120250 fs by tting them with a single exponential function. This is consistent with the reported timescale. However, it should be noted that the interaction between the pump pulse and probe pulse may still exist and affect the measurement results, as this lifetime is still on the order of the laser pulse duration (B150 fs). At this stage, the electron temperature (Te) is still much

larger than the lattice temperature (TL), for example, Te44TL. As the velocity of the hot electron is reduced, the interaction between electrons and lattice would increase via the electron-phonon coupling44. Thus, the lattice would be heated by the electrons and the phonons would be excited, which would take several picoseconds. At the nal step, the remaining heat energy in the lattice would be lost to the surroundings through the phonon relaxation processes. This would take over a long timescale from 100 ps to 1 ns depending on the material48. Figure 4b shows the dynamics of the hot electrons with a much larger time range of about 1 ns. These dynamic curves could be well tted by the biexponential function. The lifetimes of fast components are around 527 ps as shown in Fig. 4c and we assign this fast component to the process of the electronphonon coupling. The slow components have a relatively longer lifetime ranging from 200 to 500 ps as shown in Fig. 4d, which should be attributed to the phonon relaxation process.

DiscussionWe have discussed that the optical absorption and electronic transport properties have proved that the plasmon exists and it

NATURE COMMUNICATIONS | 8:15070 | DOI: 10.1038/ncomms15070 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 5

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15070

can be excited near 670 nm when under light irradiation in SrNbO3. Based on these experimental and theoretical results, we propose that the photocatalytic property of SrNbO3 can be attributed to the chemical property of the plasmon-induced hot

electronhole pairs. The expected working mechanism for H2 evolution from aqueous oxalic acid solution over Pt loaded

SrNbO3 sample is shown in Fig. 5. When the strontium niobate is under visible light irradiation, the plasmon would be excited near

a b

1.0

0.9

0.6

0.5

0.4

0.6

350 nm

Delay at 5.0 ps

Normalized intensity

0.3

0.2

0.1

0.0

0 200 400

Delay time (ps)

R/R(a.u.)

0.4

0.2

0.0

0.2

600

Wavelength (nm)

450

500

550

650

700

750

600 800 1,000

c d

700

35

30

25

20

15

600

500

Lifetime (ps)

Lifetime (ps)

400

300

200

10

5

100

0

0 350 400 450

Excitation wavelength (nm)

500 550 600 650 700 750 800 350 400 450

Excitation wavelength (nm)

500 550 600 650 700 750 800

Figure 4 | The TA spectra of SrNbO3 thin lms. (a) The differential reectance (DR/R) spectra for the SrNbO3 lm of the delay time at delays of 5.0 ps with pump light of various wavelengths and a white light continuum probe. (b) The excitation wavelength-dependent dynamic spectra with the probe wavelength at 670 nm with a measurement range of 1,100 ps. The excitation wavelength-dependent carrier lifetimes are shown for two processes with the probe pulse at 670 nm: (c) the fast process corresponding to the electronelectron scattering and (d) the slow process corresponding to thermal dissipation.

Potential versus NHE

(pH0, V)

Energy versus vacuum level (eV)

0.0

4.0

6.0

3.0

2.0

2.0

0.0

+2.0

Plasmon

excitation

Landau

Decay

H2

3.0

9.0

Pt H+

e

1.0

+3.0

+4.0

+5.0

EF CO2/H2C2O4 (0.43 V)

O2/H2O (1.23 V)

SrNbO3

H2C2O4

5.0

CB

h+

H+/H2 (0.0 V) H+, CO2

+1.0

7.0

8.0

B1 band

Figure 5 | Schematic of the photocatalytic hydrogen evolution reaction on Pt-loaded SrNbO3. CB, conduction band; B 1 band, the highest fully occupied

band below CB. The Fermi level (from UPS measurement) and the widths of CB and B 1 band shown (from DFT calculations) in this gure are for

illustration purpose.

6 NATURE COMMUNICATIONS | 8:15070 | DOI: 10.1038/ncomms15070 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15070 ARTICLE

670 nm. The plasmon resonance can be damped by the generation of hot electronhole pairs via Landau damping, which is a pure quantum mechanical process48,52. In Landau damping, a plasmon quantum can be transferred into a single electronhole pair excitation on a timescale ranging from one to few hundreds of femtoseconds. The electrical eld enhanced by the plasmon can also induce transitions of the conduction electrons from the occupied states (initial states below the Fermi level) to the unoccupied states (nal states above the Fermi level49. The position of the Fermi level has been measured using ultraviolet photoelectron spectroscopy (UPS) (Supplementary Fig. 13) and it is 3.70 eV below the vacuum level). The holes will be left below the Fermi level in the conduction band. A transient non-Fermi hot electron distribution will form with a timescale o100 fs48,51.

We assume that the hot carriers are mainly generated via the intraband transitions induced by plasmon decay, because the interband transition from CB to B1 is not allowed, as it is a d band to d band transition. One part of the hot electrons will redistribute their energies among many lower energy electrons via electronelectron scattering at a timescale of a few picoseconds and then cool down by electronphonon coupling and thermal dissipation48,53. The other part of the hot electrons will be injected into the Pt co-catalyst, which is an efcient electron trap centre and acts as the active site for the reduction reaction54,55. Thus, these hot electrons injected into Pt would reduce H to H2. As the energy distribution of the CB is very wide, the holes left in the electron-decient SrNbO3 particle would have enough energy to oxidize the oxalic acid to CO2.

In summary, we have demonstrated that epitaxial, single crystal SrNbO3 lm can be obtained by PLD. The electrons mobility of this material is very normal, only 2.47 cm2 V 1 s 1 at room temperature, and there are no band absorptions within the gap (B4 eV); thus, the interband transition model cannot be applied to explain the photocatalytic activity of this material. Further, the material has a degenerate conduction band with a gap of4.1 eV but the large carrier density leads to a large plasmon at1.61.8 eV, which simulates a mid-gap absorption. The hot electronhole pairs can be generated in the conduction band of SrNbO3 via the Landau damping during the decay of plasmons.

Therefore, we propose that the sunlight is absorbed by the plasmons whose decay leads to hot carriers responsible for the catalytic reaction. Understanding of this rst plasmonic metallic oxide and its use as a photocatalyst will open the doors for the design of a new family of photocatalytic materials.

Methods

Material preparation. The PLD target was prepared by solid reactions of the Sr4Nb2O9 precursor, Nb (Alfa Aesar, 99.99%, 325 meshes) and Nb2O5

(Alfa Aesar, 99.9985%, metals basis) powder mixtures in the proper molar ratio. The precursor was prepared by calcining SrCO3 (Alfa Aesar, 499.99%, metals basis) and Nb2O5 powder mixtures in a molar ratio of 4:1. The calcination and sintering were done in air and Ar gas environment for 20 h at a temperature of 1,200 and 1,400 C, respectively. The lms of SrNbO3 were deposited on LaAlO3(001) substrates from these targets by PLD where a Lambda Physik Excimer KrF ultraviolet laser with the wavelength of 248 nm was used. The lms were deposited at 750 C, laser energy density of 2 J cm 2, laser frequency of 5 Hz and oxygen partial pressure of 5 10 6 Torr to 1 10 6 Torr. Typically,

130 nm-thick lm could be obtained with half an hour deposition.

Physical characterization. The elemental compositions of the lms were studied by 2 MeV proton-induced X-ray emissions with Si (Li) detector. The 3.05 MeV helium ion beam is used for oxygen resonance Rutherford backscattering spectroscopy measurements to enhance the sensitivity for backscattering signal from oxygen element. SIMNRA56 simulation software is used to obtain the composition and thickness of the lms. The obtained spectra were precisely tted by SIMNRA simulation software. The crystal structure of the lm was studied by high resolution X-ray diffraction (Bruker D8 with Cu Ka1 radiation, l 1.5406 )

together with the RSMs. The local structure was measured by high-resolution scanning transmission electron microscopy (FEI Titan (Team0.5)@300 kV).

Optical bandgap of the lm was measured by a ultravioletvisible spectrophotometer (Shimadzu SolidSpec-3700). The transmissions of the lms were measured and the corresponding absorption coefcients at particular wavelengths were derived from BeerLambert Law. Physical Properties Measurement System (Quantum Design, Inc.) was used to measure the electronic transport properties.

Photocatalytic H2 production measurement. Light source used in the present experiment is 300 W Xe lamp with 410 nm long-pass lter. H2 evolution was measured by suspending 50 mg sample powders together with 1 wt.% Pt co-catalyst in 100 ml oxalic acid aqueous solution (0.025 M). The evolved gasses were collected and quantied by an online gas chromatograph (Agilent 6890N, Argon as a carrier gas, 5 molecular sieve column and thermal conductivity detector (TCD)).

TA measurement. The electronic band structure of the lms (all prepared at5 10 6 Torr, on LaAlO3 substrate) and the lifetimes of the photo exactions were

investigated by femtosecond time-resolved TA spectroscopy. The laser pulses were generated using a mode-locked Ti:sapphire oscillator seeded regenerative amplier with a pulse energy of 2 mJ at 800 nm and a repetition rate of 1 kHz. The laser beam was split into two portions. The larger portion of the beam passed through a Light Conversion TOPAS-C optical parametric amplier to generate 350 nm as the pump beam. The intensity of the pump beam was attenuated using a neutral density lter and modulated using an optical chopper at a frequency of 500 Hz. The smaller portion of the beam was used to generate white light by passing through a 1 mm sapphire plate, which acted as the probe beam. The white light beam was further split into two portions: one was used as the probe and the other was used as the reference to correct for the pulse-to-pulse intensity uctuation. The pump beam was focused onto the sample surface with a beam size of 300 mm and it fully covered the smaller probe beam (diameter: 100 mm). The reection of the probe beam from the sample surface was collected with a pair of lens and focused into a spectrometer. Very thick lm samples were used to minimize the signal contribution from the substrate. The delay between the pump and the probe pulses was controlled by a computer-controlled translation stage (Newport, ESP 300). Pumpprobe experiments were carried out at room temperature. During the measurements, the pump and the probe energies were kept low enough to minimize damage to the samples.

Theoretical calculations. The atomic and electronic structure of SrNbO3d

compounds were performed employing spin-polarized DFT calculations, using the PerdewBurkeErnzerhof57 exchange-correlation potential, and the projector-augment wave method58,59, as implemented in the Vienna ab-initio simulation programme60. In these calculations, Sr 4s4p5s, Nb 4p5s4d and O 2s2p orbitals were treated as valence states, employing the projector-augment wave potentials labelled Sr_sv, Nb_pv and O in the Vienna ab-initio simulation programme PerdewBurkeErnzerhof library. The cutoff energy for the plane-wave basis set was set to 450 eV and the DFT U approach due to Dudarev et al.61 was employed

to treat the Nb 4d orbitals occupied by the Nb4 ions present for SrNbO3 and SrNbO3.4 compositions with the value of UJ set to 4 eV. SrNbO3, SrNbO3.4 and SrNbO3.5 were modelled by supercells containing, respectively, 20 atoms with space group Pnam, 54 atoms with space group Pnnm and 44 atoms with space group Cmc2. For the cells with oxygen excess (that is, SrNbO3.4 and SrNbO3.5), the extra oxygen ions were placed in planar defects as illustrated in Fig. 3. In the structural relaxations, we employed using 8*8*4, 1*4*6 and 1*4*6 k-point meshes, for SrNbO3, SrNbO3.4 and SrNbO3.5, respectively. The density of states was calculated with 16*16*8, 2*8*12 and 2*8*12 k-point meshes for the same three structures, respectively. In systems with occupied Nb 4d orbitals (that is, SrNbO3 and

SrNbO3.4), we employed ferromagnetic ordering of the local moments on the Nb4 ions.

Data availability. The data that support the ndings of this study are available from the corresponding author on reasonable request.

References

1. Navarro, R. M., Alvarez-Galvan, M. C., de la Mano, J. A. V., Al-Zahrani, S. M. & Fierro, J. L. G. A framework for visible-light water splitting. Energy Enrion. Sci. 3, 18651882 (2010).

2. Augustynski, J., Alexander, B. D. & Solarska, R. Metal oxide photoanodes for water splitting. Top Curr. Chem. 303, 138 (2011).

3. Mayer, M. T., Lin, Y., Yuan, G. & Wang, D. Forming heterojunctions at the nanoscale for improved photoelectrochemical water splitting by semiconductor materials: case studies on hematite. Acc. Chem. Res. 46, 15581566 (2013).

4. Abdi, F. F. et al. Efcient solar water splitting by enhanced charge separation in a bismuth vanadate-silicon tandem photoelectrode. Nat. Commun. 4, 2195 (2013).

5. Brillet, J. et al. Highly efcient water splitting by a dual-absorber tandem cell. Nat. Photonics 6, 823827 (2012).

NATURE COMMUNICATIONS | 8:15070 | DOI: 10.1038/ncomms15070 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 7

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15070

6. Chen, Y. W. et al. Atomic layer-deposited tunnel oxide stabilizes silicon photoanodes for water oxidation. Nat. Mater. 10, 539544 (2011).

7. Li, Z. S., Luo, W. J., Zhang, M. L., Feng, J. Y. & Zou, Z. G. Photoelectrochemical cells for solar hydrogen production: current state of promising photoelectrodes, methods to improve their properties, and outlook. Energy Enrion. Sci. 6, 347370 (2013).

8. Esposito, D. V., Levin, I., Moffat, T. P. & Talin, A. A. H2 evolution at Si-based metal-insulator-semiconductor photoelectrodes enhanced by inversion channel charge collection and H spillover. Nat. Mater. 12, 562568 (2013).

9. Lin, F. D. & Boettcher, S. W. Adaptive semiconductor/electrocatalyst junctions in water-splitting photoanodes. Nat. Mater. 13, 8186 (2014).

10. Reece, S. Y. et al. Wireless solar water splitting using silicon-based semiconductors and earth-abundant catalysts. Science 334, 645648 (2011).

11. Kim, J. Y. et al. Single-crystalline, wormlike hematite photoanodes for efcient solar water splitting. Sci. Rep. 3, 2681 (2013).

12. Park, H. S., Lee, H. C., Leonard, K. C., Liu, G. J. & Bard, A. J. Unbiased photoelectrochemical water splitting in Z-scheme device using W/Mo-doped BiVO4 and ZnxCd1 xSe. ChemPhysChem. 14, 22772287 (2013).

13. Osterloh, F. E. & Parkinson, B. A. Recent developments in solar water-splitting photocatalysis. MRS Bull. 36, 1722 (2011).

14. Tanaka, A., Sakaguchi, S., Hashimoto, K. & Kominami, H. Preparation of Au/TiO2 with metal cocatalysts exhibiting strong surface plasmon resonance effective for photoinduced hydrogen formation under irradiation of visible light. ACS Catal. 3, 7985 (2013).

15. Pu, Y. C. et al. Au nanostructure-decorated TiO2 nanowires exhibiting photoactivity across entire UV-visible region for photoelectrochemical water splitting. Nano Lett. 13, 38173823 (2013).

16. Asahi, R., Morikawa, T., Ohwaki, T., Aoki, K. & Taga, Y. Visible-light photocatalysis in nitrogen-doped titanium oxides. Science 293, 269271 (2001).

17. Khan, S. U. M., Al-Shahry, M. & Ingler, W. B. Efcient photochemical water splitting by a chemically modied n-TiO2. Science 297, 22432245 (2002).

18. Liu, G. et al. A red anatase TiO2 photocatalyst for solar energy conversion. Energy Enrion. Sci. 5, 96039610 (2012).

19. Chen, X. B., Liu, L., Yu, P. Y. & Mao, S. S. Increasing solar absorption for photocatalysis with black hydrogenated titanium dioxide nanocrystals. Science 331, 746750 (2011).

20. Chen, X. & Mao, S. S. Titanium dioxide nanomaterials: synthesis, properties, modications, and applications. Chem. Rev. 107, 28912959 (2007).

21. Umebayashi, T., Yamaki, T., Itoh, H. & Asai, K. Analysis of electronic structures of 3d transition metal-doped TiO2 based on band calculations.

J. Phys. Chem. Solids 63, 19091920 (2002).22. Wang, J. et al. Origin of photocatalytic activity of nitrogen-doped TiO2 nanobelts. J. Am. Chem. Soc. 131, 1229012297 (2009).

23. Takai, A. & Kamat, P. V. Capture, store, and discharge. shuttling photogenerated electrons across TiO2silver interface. ACS Nano 5, 73697376 (2011).

24. Linic, S., Christopher, P. & Ingram, D. B. Plasmonic-metal nanostructures for efcient conversion of solar to chemical energy. Nat. Mater. 10, 911921 (2011).

25. Xu, X., Randorn, C., Efstathiou, P. & Irvine, J. T. A red metallic oxide photocatalyst. Nat. Mater. 11, 595598 (2012).

26. Efstathiou, P., Xu, X., Menard, H. & Irvine, J. T. An investigation of crystal structure, surface area and surface chemistry of strontium niobate andtheir inuence on photocatalytic performance. Dalton Trans. 42, 78807887 (2013).

27. Oka, D., Hirose, Y., Nakao, S., Fukumura, T. & Hasegawa, T. Intrinsic high electrical conductivity of stoichiometric SrNbO3 epitaxial thin lms. Phys. Rev. B 92, 205102 (2015).

28. Hessen, B., Sunshine, S. A., Siegrist, T. & Jimenez, R. Crystallization of reduced strontium and barium niobate perovskites from borate uxes. Mat. Res. Bull. 26, 8590 (1991).

29. Peng, N. H., Irvine, J. T. S. & Fitzgerald, A. G. Synthesis and crystal structure of the distorted perovskite Sr0.97NbO3 determined by high resolution powder neutron diffraction. J. Mater. Chem. 8, 10331038 (1998).

30. Li, S. S. & Thurber, W. R. The dopant density and temperature dependence of electron mobility and resistivity in n-type silicon. Solid State Electron. 20, 609616 (1977).

31. Kim, H. J. et al. High mobility in a stable transparent perovskite oxide. Appl. Phys. Express 5, 061102 (2012).

32. Chandiran, A. K., Abdi-Jalebi, M., Nazeeruddin, M. K. & Gratzel, M. Analysis of electron transfer properties of ZnO and TiO2 photoanodes for dye-sensitized solar cells. ACS Nano 8, 22612268 (2014).

33. Jeong, B. S., Norton, D. P. & Budai, J. D. Conductivity in transparent anatase TiO2 lms epitaxially grown by reactive sputtering deposition. Solid State Electron. 47, 22752278 (2003).

34. Xu, X., Randorn, C., Efstathiou, P. & Irvine, J. T. S. A red metallic oxide photocatalyst. Nat. Mater. 11, 595598 (2012).

35. Hammer, M. S., Rauh, D., Lorrmann, V., Deibel, C. & Dyakonov, V. Effect of doping- and eld-induced charge carrier density on the electron transport in nanocrystalline ZnO. Nanotechnology 19, 485701 (2008).

36. Isawa, K., Sugiyama, J., Matsuura, K., Nozaki, A. & Yamauchi, H. Synthesis and transport properties of SrxNbO3 (0.75rxr0.9). Phys. Rev. B 47, 28492853 (1993).

37. Avella, A. & Mancini, F. (eds) Strongly Correlated Systems: Experimental Techniques. Chapter 9 (Springer, 2015).

38. Ferrell, R. A. Predicted radiation of plasma oscillations in metal lms. Phys. Rev. 111, 12141222 (1958).

39. Xia, C., Yin, C. & Kresin, V. V. Photoabsorption by volume plasmons in metal nanoclusters. Phys. Rev. Lett. 102, 156802 (2009).

40. Lichtenberg, F., Herrnberger, A., Wiedenmann, K. & Mannhart, J. Synthesis of perovskite-related layered AnBnO3n2ABOX type niobates and titanates and

study of their structural, electric and magnetic properties. Prog. Solid. State 29, 170 (2001).41. Chen, C. et al. Atomic and electronic structure of the SrNbO3/SrNbO3.4 interface. Appl. Phys. Lett. 105, 221602 (2014).

42. Ruzicka, B. A., Wang, R., Lohrman, J., Ren, S. & Zhao, H. Exciton diffusion in semiconducting single-walled carbon nanotubes studied by transient absorption microscopy. Phys. Rev. B 86, 205417 (2012).

43. Price, M. B. et al. Hot-carrier cooling and photoinduced refractive index changes in organic-inorganic lead halide perovskites. Nat. Commun. 6, 8420 (2015).

44. Ahmadi, T. S., Logunov, S. L. & El-Sayed, M. A. Picosecond dynamics of colloidal gold nanoparticles. J. Phys. Chem. 100, 80538056 (1996).

45. Doremus, R. H. Optical properties of small gold particles. J. Phys. Chem. 40, 2389 (1964).

46. Heilweil, E. J. & Hochstrasser, R. M. Nonlinear spectroscopy and picosecond transient grating study of colloidal gold. J. Phys. Chem. 82, 4762 (1985).

47. Sun, C. K., Valle, F., Acioli, L. H., Ippen, E. P. & Fujimoto, J. G. Femtosecond-tunable measurement of electron thermalization in gold. Phys. Rev. B 50, 1533715348 (1994).

48. Brongersma, M. L., Halas, N. J. & Nordlander, P. Plasmon-induced hot carrier science and technology. Nat. Nanotechnol. 10, 2534 (2015).

49. Manjavacas, A., Liu, J. G., Kulkarni, V. & Nordlander, P. Plasmon-induced hot carriers in metallic nanoparticles. ACS Nano 8, 76307638 (2014).

50. Lisowski, M. et al. Ultra-fast dynamics of electron thermalization, cooling and transport effects in Ru(001). Appl. Phys. A 78, 165176 (2004).

51. Inouye, H., Tanaka, K., Tanahashi, I. & Hirao, K. Ultrafast dynamics of nonequilibrium electrons in a gold nanoparticle system. Phys. Rev. B 57, 1133411340 (1998).

52. Brown, A. M., Sundararaman, R., Narang, P., Goddard, III W. A. & Atwater, H. A. Nonradiative plasmon decay and hot carrier dynamics: effects of phonons, surfaces, and geometry. ACS Nano 10, 957966 (2016).

53. Watanabe, K., Menzel, D., Nilius, N. & Freund, H.-J. Photochemistry on metal nanoparticles. Chem. Rev. 106, 43014320 (2006).

54. Zhang, X., Chen, Y. L., Liu, R. S. & Tsai, D. P. Plasmonic photocatalysis. Rep. Prog. Phys. 76, 046401 (2013).

55. Yang, J., Wang, D., Han, H. & Li, C. Roles of cocatalysts in photocatalysis and photoelectrocatalysis. Acc. Chem. Res. 46, 19001909 (2013).

56. Mayer, M. SIMNRA, a simulation program for the analysis of NRA, RBS and ERDA. AIP Conf. Proc. 475, 541544 (1999).

57. Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865 (1996).

58. Blchl, P. E. Projector augmented-wave method. Phys. Rev. B 50, 17953 (1994).59. Kresse, G. & Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B 59, 1758 (1999).

60. Kresse, G. & Furthmller, J. Efcient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 54, 1116911186 (1996).

61. Dudarev, S., Botton, G., Savrasov, S., Humphreys, C. & Sutton, A. Electron-energy-loss spectra and the structural stability of nickel oxide: An LSDA U

study. Phys. Rev. B 57, 1505 (1998).

Acknowledgements

We appreciate Dr Kokkoris from National University of Athens, Greece, for the carbon

ions Rutherford backscattering spectroscopy measurement. Financial support from the

Singapore-Berkeley Research Initiative for Sustainable Energy (SinBerRISE) programme

is gratefully acknowledged. The NUS researchers acknowledge Singapore National

Research Foundation under its Competitive Research Funding Control of exotic

quantum phenomena at strategic interfaces and surfaces for novel functionality by in-situ

synchrotron radiation (NRF-CRP8-2011-06) and Oxide Electronics on silicon beyond

Moore (NRF-CRP15-2015-01), MOE-AcRF Tier-2 (MOE2015-T2-1-099) and FRC.

Author contributions

D.Y.W. and Y.L.Z. developed this project, designed the research strategy and analysed the

results. Y.L.Z., D.Y.W. and B.X.Y. prepared the samples and measured the electronic

8 NATURE COMMUNICATIONS | 8:15070 | DOI: 10.1038/ncomms15070 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms15070 ARTICLE

transport properties. D.Y.W. carried out the TA spectra measurements with instruction

by J.Q.C. Y.C. performed the DFT calculations. T.C.A. carried out ellipsometry

spectroscopy characterization. X.C. and T.C.A. discussed the ellipsometry spectra. Z.H.

measured the RSMs. J.H. carried out the photocatalytic experiments. C.T.N. performed

the TEM imaging. M.R.M. measured the Rutherford backscattering spectroscopy (RBS)

spectra. D.X. conducted and measured the UPS spectra supervised by W.C. R.X.

supervised the photocatalytic measurement. H.Z. analysed the TEM images.

A. supervised the RSMs experiments. A.R. supervised the ellipsometry spectra measure

ments. A.M. supervised the TEM imaging of the lms. M.B.H.B. supervised the RBS

experiments. M.A. supervised the DFT calculations. Q.-H.X. supervised the TA spectra

measurements. D.Y.W and Y.L.Z. prepared this manuscript. T.V. supervised and led the

project. All authors contributed to the scientic discussion and manuscript revisions.

Additional information

Supplementary Information accompanies this paper at http://www.nature.com/naturecommunications

Web End =http://www.nature.com/

http://www.nature.com/naturecommunications

Web End =naturecommunications Competing interests: The authors declare no competing nancial interests.

Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions/

Web End =http://npg.nature.com/

http://npg.nature.com/reprintsandpermissions/

Web End =reprintsandpermissions/

How to cite this article: Wan, D. Y. et al. Electron transport and visible light absorption

in a plasmonic photocatalyst based on strontium niobate. Nat. Commun. 8, 15070

doi: 10.1038/ncomms15070 (2017).

Publishers note: Springer Nature remains neutral with regard to jurisdictional claims in

published maps and institutional afliations.

This work is licensed under a Creative Commons Attribution 4.0

International License. The images or other third party material in this

article are included in the articles Creative Commons license, unless indicated otherwise

in the credit line; if the material is not included under the Creative Commons license,

users will need to obtain permission from the license holder to reproduce the material.

To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Web End =http://creativecommons.org/licenses/by/4.0/

r The Author(s) 2017

NATURE COMMUNICATIONS | 8:15070 | DOI: 10.1038/ncomms15070 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 9