This journal is c The Royal Society of Chemistry 2013 Chem. Commun., 2013, 49, 7917--7919 7917

Cite this: Chem. Commun.,2013,49, 7917

Plasmon-enhanced near-infrared-active materials inphotoelectrochemical water splitting†

Chih Kai Chen, Hao Ming Chen, Chih-Jung Chen and Ru-Shi Liu*

We report the near-infrared-driven photoelectrochemical water

splitting using a ZnO nanorod-array decorated with CdTe quantum

dots and plasmon-enhanced upconversion nanoparticles. The plasmon

enhanced the intensity of the upconversion emission, which improved

the photocurrent and the gas evolution rate of the photoelectro-

chemical reaction greatly.

Hydrogen has been hailed as the clean energy form of the futurebecause of its reaction with oxygen without producing pollutants andgreenhouse gases, only water. Photoelectrochemical (PEC) and photo-catalytic water splitting, which directly convert solar energy intohydrogen and oxygen, have become two of the most promisingmethods in renewable energy research since Fujishima and Honda’sfirst demonstration in 1972.1 The use of metal oxide electrodematerials, such as TiO2, ZnO, Fe2O3, and WO3, in water splittinghas been progressing because of their unique nanostructure-derivedproperties and effectiveness.2–7 However, some of the semiconductorshave large band gaps (>3 eV, except Fe2O3, and WO3), which limitsthe absorption of visible light and thus the improvement of solarenergy conversion efficiency.6 Numerous methods,4,8 such as dopingheteroatoms2 and the use of photosensitive dyes9 or quantum dots(QDs),10,11 have been adopted to harvest visible light. Currently,sensitizers do not use near-infrared (NIR) and infrared (IR) lightactive materials because the NIR and IR active sensitizers commonlyabsorb sub-band gap energy, which indicates that irradiation withenergy below 1.8 eV (B688 nm) cannot provide sufficient energypotential to excite electron–hole pairs with enough energy to drive theoverall water splitting reaction.4 Using upconversion nanoparticles(UCNs) is an alternative method of converting NIR and IR into usableenergy and improving the overall energy conversion efficiency.UCNs have attracted significant attention in biological imaging, datastorage, and solar energy conversion.7,12 Previously, gold nanoparticle(Au NP)-modified UCNs have been found to enhance the upconver-sion emission through plasmonic modulation.13,14 Few reports have

suggested the use of UCNs to convert NIR light to visible light in solarcells, PEC, or photoelectrochemical applications.15

In the present work, we constructed a PEC cell that couldup-convert the NIR light into visible light and further generatechemical fuels through light-driven water splitting. For this purpose,we attempted to embed UCNs, Er3+/Yb3+ co-doped NaYF4, in the CdTeQD-sensitized ZnO nanorod array to compose a NIR-driven PEC cell(Fig. 1a). UCNs were employed to harvest and convert NIR light tohigh-energy photons. The high-energy photons excited the CdTe QDsto generate high-energy electron–hole pairs. These excited electronstransferred to the conduction band (CB) of ZnO nanorods, and thenconducted to the Pt foil through the external circuit and reacted withwater to generate hydrogen, thereby producing the photocurrent. Thephotogenerated holes oxidized water to produce oxygen (Fig. 1b).When appropriate amounts of Au NPs were modified on the UCNsurface, the induced surface plasma resonance (SPR) enhanced theintensity of upconversion emission, which improved the photocurrentand the gas evolution rate of the photoelectrochemical reaction.

In a typical procedure, the ZnO nanorod-array was grown onfluorine-doped tin oxide (F–SnO2) substrates via a modified

Fig. 1 Sketch of the (a) Au-UCNs–CdTe–ZnO photoelectrode and (b) the mecha-nism of energy conversion from NIR to chemical fuel.

Department of Chemistry, National Taiwan University, Sec. 4, Roosevelt Road,

Taipei 106, Taiwan. E-mail: rsliu@ntu.edu.tw; Fax: +886-2-33668671;

Tel: +886-2-33661169

† Electronic supplementary information (ESI) available: Detailed synthesis,sample preparation, and experimental methods. See DOI: 10.1039/c3cc42567c

Received 9th April 2013,Accepted 4th July 2013

DOI: 10.1039/c3cc42567c

www.rsc.org/chemcomm

ChemComm

COMMUNICATION

Publ

ishe

d on

04

July

201

3. D

ownl

oade

d by

Nat

iona

l Tai

wan

Uni

vers

ity o

n 25

/04/

2014

09:

05:2

8.

View Article OnlineView Journal | View Issue

7918 Chem. Commun., 2013, 49, 7917--7919 This journal is c The Royal Society of Chemistry 2013

hydrothermal reaction (experiment details are provided in the ESI†).The one-dimensional ZnO nanorods facilitated the charge transpor-tation to reduce the possibility of recombination. The synthesizedZnO nanorods were about 5 mm in length and about 150 nm indiameter with a high surface density, as shown in Fig. S1 (ESI†). Thetransmission electron microscopy (TEM) and high-resolution trans-mission electron microscopy (HRTEM) images of the ZnO nanorodsdecorated with CdTe QDs (Fig. 2a and b) provide evidence that QDsare directly attached to the ZnO nanorod surface. An abrupt transi-tion was found between the (0002) lattice planes of the ZnO nanorodand the (311) lattice planes of the CdTe QDs. The lattice spacingbetween the (311) planes is 0.20 nm, which is also in agreement withthat of the CdTe bulk crystal (JCPDS No. 89-3053). The structuralproperties of the ZnO were examined by X-ray diffraction (XRD)pattern studies as shown in Fig. S2a (ESI†).

In the present study UCNs (Er3+/Yb3+ co-doped NaYF4) weresynthesized via a facile wet chemical technology (ESI†). The watersoluble UCNs were modified with the desired weight percent(0.2, 0.4, 1 and 2 wt%) of Au NPs, in which the controlled UCNconcentration was 10 mg mL�1. The UCNs typically have a cubicstructure with a uniform size of approximately 30 nm to 35 nm(Fig. S3, ESI†). The (111) lattice planes of UCNs are clearly illustratedin the HRTEM image. The d-spacing of the (111) lattice plane is0.32 nm, which is in agreement with the XRD pattern (Fig. S2b, ESI†)and the bulk value of NaYF4 (JCPDS 77-2042) with a lattice constantof a = 5.470 Å. The Au NPs with diameters of 6 nm to 8 nm could beclearly observed on the surface of UCNs (Au-UCNs) as revealed byTEM (Fig. 2c–f). Under low Au NP concentration, only a smallamount of Au NPs was modified on the surface of the UCNs(Fig. 2c). When the Au NP concentration increased to 0.4 wt%, thenumber of attached Au NPs was found to be more than that of AuNPs prepared under low concentration conditions. As the Au NPconcentration further increased to 2 wt%, the Au NPs graduallyaggregated on the surface of the UCNs (Fig. 2f).

The ultraviolet-visible absorption spectra of UCNs with variousamounts of Au NPs presented an evident increase in absorbance atapproximately 520 nm, which originated from the SPR of Au NPs(Fig. 3a) revealing that Au NPs were successfully attached to the

surface of the UCNs with variable concentrations. The resultsdemonstrate that the amount of Au NPs could be effectivelycontrolled by the weight percentage, which allows the optimizationof the Au NP modification. The Au-UCN samples showed a slightredshift and a broader SPR absorption peak with a considerableincrease in the amount of Au NPs as compared with the pure Au NPsolution. The redshift peak can be attributed to the decreased inter-particle distance of Au NPs on the Au-UCNs.5,7 The broader peakresulted from both the aggregation of Au NPs as the Au NP contentincreased and the electron transfer between the Au NPs and UCNs.16

The electron transfer between the Au NPs and UCNs decreased theelectron density on the Au NPs, thereby broadening the SPR peak.17

To absorb the high-energy photons and generate electron–holepairs for the PEC reaction, the overlap of the upconversion emis-sion peak of UCNs and the excitation peak of the CdTe QDs isessential. The first excitonic peak of the CdTe QDs was at 690 nm(Fig. 3b), which could overlap the emission spectra of the UCNs(Fig. 3c). Thus, the upconversion emission could be absorbed bythe CdTe QDs and generated the high-energy electron–hole pairs.The upconversion emission intensity of the Au-x-UCNs wasenhanced most significantly in the case of x = 1.0 wt%. Theenhancement of the emission intensity at 650 nm and 530 nmwas more than threefold and double that of the pure UCNs,respectively. The electric field generated by plasmonic resonancemight cause local field enhancement or surface plasmon-coupledemission (SPCE) to enhance the upconversion emission inten-sity.13,18 The SPCE could enhance the emission rate at 530 nm,which overlapped with the SPR peak, hence, enhancing the upcon-version emission, even absorbed by the Au NPs. The enhancementof the upconversion emission at 650 nm is attributed to the localfield enhancement. The plasmonic field influenced the localelectric field of the UCNs, which consequently generated moreelectrons in the upconversion progress and thus enhanced theemission intensity. By contrast, the emission spectra intensitydecreased when the weight percentage of Au NPs increased to2.0 wt%, which revealed that the NIR light was scattered by the AuNPs and thus quenched the upconversion emission. Owing to theemission at 530 nm being absorbed by the Au NPs, the enhance-ment of the upconversion emission intensity at 650 nm was greaterthan that at 530 nm of the Au-UCNs. To discuss further about theplasmon enhanced upconversion emission, the quantum yields ofthe Au-x-UCNs suspended in water (x = 0, 0.2, 0.4, 1.0 and 2.0 wt%)were calculated and are listed in Table S1 (ESI†).19 The quantumyield of the Au-1.0 wt%-UCNs (0.40%) was enhanced by plasmonmore than twofold compared to the pure UCNs (0.15%).

Fig. 2 TEM image. (a) CdTe–ZnO and (b) HRTEM image of the CdTe–ZnO;TEM images of the UCNs modified with (c) 0.2 wt%; (d) 0.4 wt%. (e) 1.0 wt%.(f) 2.0 wt% Au nanoparticles; (g) TEM image of the Au-1.0%UCNs–CdTe–ZnO.

Fig. 3 UV-vis spectra. (a) Au-x-UCNs suspension in water (x = 0, 0.2, 0.4, 1.0 and2.0 wt%); (b) CdTe QDs and Au NP suspension in water, and (c) upconversionemission spectra of the Au-x-UCNs (x = 0, 0.2, 0.4, 1.0 and 2.0 wt%) suspended inwater at a controlled concentration of 10 mg mL�1 of UCNs.

Communication ChemComm

Publ

ishe

d on

04

July

201

3. D

ownl

oade

d by

Nat

iona

l Tai

wan

Uni

vers

ity o

n 25

/04/

2014

09:

05:2

8.

View Article Online

This journal is c The Royal Society of Chemistry 2013 Chem. Commun., 2013, 49, 7917--7919 7919

To prevent the electron transfer between the Au, ZnO and CdTQD interface, the prepared Au-UCNs were deposited on the ZnOnanorod surface after coating with a thin layer of ZnS as shown inFig. S5 (ESI†). After coating with ZnS, the Au-UCN surface wasinvestigated with an amorphous ZnS layer. The TEM image(Fig. 2g) shows that the Au-1.0%-UCNs–CdTe–ZnO structure pro-vides more compelling evidence that the Au-UCNs have attachedonto the ZnO nanorod surface as expected. As a proof of concept, weperformed a set of linear-sweep voltammogram measurements(Fig. 4a) for the PEC, which was composed of an Au-UCNs–CdTe–ZnO nanocomposite as the photoanode, a platinum plate as thecounter electrode, and Ag/AgCl as the reference electrode under980 nm laser illumination with a power density of 50 mW cm�2

(experiment details are provided in the ESI†). The generated photo-current demonstrated that via the up-conversion process the UCNsconverting the 980 nm light to the visible light were absorbed by theCdTe QDs to excite electron–hole pairs to drive the photoelectro-chemical reaction. For the other sets, the photocurrent intensity ofthe Au-UCNs increased as the Au NP weight percentage increased tothe appropriate Au NP concentration (1.0 wt%) but it decreased athigh Au NP concentrations, a trend similar to that observed for theupconversion emission spectrum. The quenching of the upconver-sion emission under high Au NP concentration caused the lowerphoto-response. The UCNs–ZnO, Au-UCNs–ZnO and Au–CdTe–ZnOphotoeletrode showed negligible photocurrent intensity (Fig. S6,ESI†), confirming that the enhancement of the photocurrent wasfrom the NIR up-conversion mechanism, neither from the plasmoneffect of the Au nanoparticle nor the CdTe QDs modification. Thephotocurrent density of the Au-1.0%-UCNs–CdTe–ZnO photo-electrode can reach up to 0.036 mA cm�2 at 1.0 V. Fig. S6 (ESI†)shows plots of the photocurrent that was generated by the Au-1.0%-UCNs–CdTe–ZnO photoelectrode. This result verifies the rapidtransportation of the electrons and the stability of the photoelec-trode. The enhancement ratio is almost 40-fold of the pristine ZnOnanorod photoelectrode, and threefold greater than the UCNs–CdTe–ZnO photoelectrode. To demonstrate the water splittingthe reaction was conducted under NIR illumination, and the gasevolution (Fig. 4b) of the PEC was measured with 0.5 V bias under980 nm laser illumination with a power density of 100 mW cm�2.The Au-1.0%-UCNs–CdTe–ZnO and UCNs–CdTe–ZnO photo-electrodes show that about 0.18 and 0.045 mmol h�1 of H2 wereproduced (0.08 and 0.017 mmol h�1 of O2), indicating that water hadbeen split under NIR illumination. The slight decrease in the oxygenevolution after two hours of measurement might contribute to thelower faraday efficiency.8 We estimated the highest overall conver-sion efficiency from the NIR to chemical fuel to be around 0.006%.

These findings show that the Au NP-modified UCNs can enhancethe upconversion emission intensity, thereby increasing NIR lightconversion into high-energy photons, which are then absorbed bythe CdTe QDs to generate sufficient potential electron–hole pairs todrive the photoelectrochemical water splitting.

The band positions of the samples were determined by theelectrochemical methods as shown in Fig. S7 (ESI†).

In summary, we have demonstrated a process of utilizing NIR todrive the photoelectrochemical water splitting reaction. TheAu-UCNs can significantly enhance the upconversion emissionintensity by a plasmonic effect. Thus, the enhancement of photo-current and gas evolution was achieved. The results offer aconvincing demonstration that energy can be converted from NIRto chemical fuel. We believe that our strategy is fundamental to thedesign of solar energy devices and should become an acceptedtechnique for solar energy utilization in the NIR and IR regions.

We thank the financial support from National ScienceCouncil of Taiwan (Contract No. NSC 101-2113-M-002-014-MY3) and Academia Sinica (Contract No. AS-103-TP-A06).

Notes and references1 A. Fujishima and K. Honda, Nature, 1972, 238, 37.2 H. M. Chen, C. K. Chen, R. S. Liu, C. C. Wu, W. S. Chang, K. H. Chen,

T. S. Chan, J. F. Lee and D. P. Tsai, Adv. Energy Mater., 2011, 1, 742.3 S. D. Tilley, M. Cornuz, K. Sivula and M. Gratzel, Angew. Chem., Int.

Ed., 2010, 49, 6405.4 H. M. Chen, C. K. Chen, R. S. Liu, L. Zhang, J. Zhang and

D. P. Wilkinson, Chem. Soc. Rev., 2012, 41, 5654.5 K. L. Kelly, E. Coronado, L. L. Zhao and G. C. Schatz, J. Phys.

Chem. B, 2003, 107, 668.6 Y. Lin, G. Yuan, R. Liu, S. Zhou, S. W. Sheehan and D. Wang, Chem.

Phys. Lett., 2011, 507, 209.7 Y. Wang, G. Chen, M. Yang, G. Silber, S. Xing, L. H. Tan, F. Wang,

Y. Feng, X. Liu, S. Li and H. Chen, Nat. Commun., 2010, 1, 87.8 H. M. Chen, C. K. Chen, C. J. Chen, L. C. Cheng, P. C. Wu,

B. H. Cheng, Y. Z. Ho, M. L. Tseng, Y. Y. Hsu, T. S. Chan,J. F. Lee, R. S. Liu and D. P. Tsai, ACS Nano, 2012, 6, 7362.

9 R. Abe, K. Shinmei, K. Hara and B. Ohtani, Chem. Commun., 2009,3577.

10 H. M. Chen, C. K. Chen, Y. C. Chang, C. W. Tsai, R. S. Liu, S. F. Hu,W. S. Chang and K. H. Chen, Angew. Chem., Int. Ed., 2010, 49, 5966.

11 N. Chouhan, C. L. Yeh, S. F. Hu, R. S. Liu, W. S. Chang andK. H. Chen, Chem. Commun., 2011, 47, 3493; X. P. Qi, G. W. She,Y. Y. Liu, L. X. Mu and W. S. Shi, Chem. Commun., 2012, 48, 242;H. M. Chen, C. K. Chen, C. C. Lin, R. S. Liu, H. Yang, W. S. Chang,K. H. Chen, T. S. Chan, J. F. Lee and D. P. Tsai, J. Phys. Chem. C,2011, 115, 21971.

12 F. Wang and X. Liu, Chem. Soc. Rev., 2009, 38, 976.13 H. Zhang, Y. Li, I. A. Ivanov, Y. Qu, Y. Huang and X. Duan, Angew.

Chem., Int. Ed., 2010, 49, 2865.14 D. R. Baker and P. V. Kamat, Adv. Funct. Mater., 2009, 19, 805;

W. Zhang, F. Ding and S. Y. Chou, Adv. Mater., 2012, 24, OP236;N. Liu, W. P. Qin, G. S. Qin, T. Jiang and D. Zhao, Chem. Commun.,2011, 47, 7671; L. D. Sun, W. Feng and C. H. Yan, Chem. Commun.,2009, 4393.

15 H.-Q. Wang, M. Batentschuk, A. Osvet, L. Pinna and C. J. Brabec,Adv. Mater., 2011, 23, 2675; W. Qin, D. Zhang, D. Zhao, L. Wang andK. Zheng, Chem. Commun., 2010, 46, 2304; M. Zhang, Y. Lin,T. J. Mullen, W. F. Lin, L. D. Sun, C. H. Yan, T. E. Patten,D. Wang and G. Y. Liu, J. Phys. Chem. Lett., 2012, 3, 3188.

16 G. Bajaj and R. K. Soni, Appl. Surf. Sci., 2010, 256, 6399.17 G. Shan, M. Zhong, S. Wang, Y. Li and Y. Liu, J. Colloid Interface Sci.,

2008, 326, 392.18 B. Messinger, K. von Raben, R. Chang and P. Barber, Phys. Rev. B,

1981, 24, 649.19 Q. Liu, T. Yang, W. Feng and F. Li, J. Am. Chem. Soc., 2012, 134, 5390;

W. Q. Zou, C. Visser, J. A. Maduro, M. S. Pshenichnikov andJ. C. Hummelen, Nat. Photonics, 2012, 6, 560.

Fig. 4 (a) Linear-sweep voltammograms and (b) gas evolution of the Au-x-UCNs–CdTe–ZnO (x = 0, 0.2, 0.4, 1.0 and 2.0 wt%) and ZnO as a photoanode in 0.5 MNa2SO4 solution.

ChemComm Communication

Publ

ishe

d on

04

July

201

3. D

ownl

oade

d by

Nat

iona

l Tai

wan

Uni

vers

ity o

n 25

/04/

2014

09:

05:2

8.

View Article Online