Published: January 12, 2011

r 2011 American Chemical Society 661 dx.doi.org/10.1021/ac102623r |Anal. Chem. 2011, 83, 661–665

LETTER

pubs.acs.org/ac

Electrogenerated Chemiluminescence of Au Nanoclusters for theDetection of DopamineLingling Li, Hongying Liu, Yuanyuan Shen, Jianrong Zhang, and Jun-Jie Zhu*

Key Laboratory of Analytical Chemistry for Life Science (Ministry of Education of China), School of Chemistry and ChemicalEngineering, Nanjing University, Nanjing 210093, P. R. China

bS Supporting Information

ABSTRACT: Electrogenerated chemiluminescence (ECL) emission wasobserved from the water-soluble, bovine serum albumin (BSA)-stabilizedAu nanoclusters for the first time. The possible ECL mechanism wasdiscussed according to the presented results and ascribed to the effectiveelectron transfer from the conduction-band of excited indium tin oxide(ITO) to Au nanoclusters (NCs). A simple label-free method for thedetection of dopamine has been developed based on the Au NCs ECL inaqueous media. The Au NCs could be an effective candidate for new typesof ECL biosensors in the future due to their fascinating features, such asgood water solubility, low toxicity, ease of labeling, and excellent stability.

Quantum-sized Au nanoparticles, also called Au nanoclusters(Au NCs), composed of very few atoms with a ultrasmall

size ranging from subnanometer to approximately 2 nm (corediameter),1 have drawn wide attention in recent years. Because ofthe strong quantum confinement of electrons in the ultrasmallsize regime, the Au NCs exhibit discrete size-dependent elec-tronic energy level and lose the bulklike electronic properties, thedisappearance of surface plasmon resonance, for instance.1-3

Hence, Au NCs show “molecule-like” properties, such as en-hanced catalytic activity, enhanced photoluminescence, intrinsicmagnetism, and unique charging properties,1,4,5 which render thematerial promising in various areas ranging from fundamentalstudies6-10 to potential applications from catalysis,11 sensing,2,12

to bioimaging.4,13 Furthermore, AuNCs exhibit other fascinatingfeatures, including ease of synthesis, good water solubility, lowtoxicity,2 surface functionalities, biocompatibility, and excellentstability, which also make them hold great promise in bioanalysis.

Electrogenerated chemiluminescence (ECL) is a light emis-sion that arises from the high-energy electron transfer reactionbetween electrogenerated species,14 which has special advantagesover photoluminescence techniques, such as low cost, wide rangeof analytes, and high sensitivity.15 Because of their uniquequantum size dependent optical and electrochemical properties,the ECL of quantum dots (QDs) has become more and morefascinating14-19 sinceDing and Bard reported the ECL of Si QDsin 2002.20 However the inherent toxicity inmost of the QDs usedin ECL studies, such as CdTe,14,15 CdS,17,19 and CdSe,18 wouldlimit their applications in a bioassay. Thus it is necessary fordeveloping novel low-toxicity or nontoxic ECL species.

Clearly, it is of considerable interest for the Au NCs ECL onthe basis of their low toxicity.2 To the best of our knowledge, noresearch has been reported involving the Au NCs ECL study.

Herein, we explored the ECL properties of Au NCs. Redfluorescent BSA-stabilized Au NCs, consisting of 25 gold atoms(Au25), were prepared with a simple, environmental friendlysynthetic route as reported previously (see the SupportingInformation).21 High-resolution transmission electron micro-scopy (HRTEM), X-ray photoelectron spectroscopy (XPS),photoluminescence (PL) spectra, and UV-vis spectra analyseswere performed and presented the consistent results with pre-vious reports (see the Supporting Information).2,21

To investigate the ECL behavior, the Au NCs were firstimmobilized onto hydroxylated indium tin oxide (ITO) elec-trodes through 3-aminopropyl-triethoxysilane (APTES) and gluta-ric dialdehyde (Figure S1 in the Supporting Information). TheBSA layer surrounding the Au NCs could afford abundantfunctionalities such as carboxyl and amine groups (Figure S2 inthe Supporting Information), facilitating the assembly of AuNCs. The thickness of the Au NCs film was characterized byAFM (See Figure S7 in the Supporting Information), whichrevealed a uniform Au NCs monolayer with a thickness ofabout 1.5 nm. The ECL of the Au NCs at the ITO electrodewas observed in the electrolyte containing K2S2O8, a coreac-tant commonly used in ECL systems, when the potential wascycled between þ2.0 and -1.5 V. Figure 1a shows the cyclicvoltammograms (CV) of clean ITO in 0.1 M PBS (pH 7.4)containing 0.1 M KCl with or without K2S2O8. The obviouscathodic peak at about -1.3 V (C1) represents the reduc-tion of ITO. Moreover, with the presence of K2S2O8, thecathodic current began to flow at -0.6 V, earlier than that

Received: October 5, 2010Accepted: January 4, 2011

662 dx.doi.org/10.1021/ac102623r |Anal. Chem. 2011, 83, 661–665

Analytical Chemistry LETTER

in the absence of K2S2O8, due to the reduction of peroxydi-sulfate on ITO.22 As for Au NC modified ITO electrodes, thecathodic ECL signal is about 5 times higher than that of bareITO, whereas the change of the anodic signal is negligible(Figure 1b). Note also that in the voltammograms, the AuNCsfilm blocks the electrochemical process on the modified ITOsurface as indicated by the decreased current compared to CVof bare ITO in Figure 1a, which could be ascribed to thenonconductive property of the BSA layer surrounding the AuNCs. The differential pulse voltammograms (DPV) inset inFigure 1b revealed two cathodic peaks at about -1.3 V (C1)and -1.0 V (C2), corresponding to the reduction of thereduction of ITO and peroxydisulfate, respectively.22 It wasreported that Au nanoparticles (Au NPs) exhibited a signifi-cant ECL amplification effect, which acted as a conductivewire to facilitate the electron transfer in the ECL reaction.19

To confirm that the obtained strong signal at the Au NCs

modified ITO electrode indeed originated from the Au NCs,trisodium citrate reduced Au NPs were used for the compar-ison. As shown in Figure 1c, no obvious enhanced signal couldbe observed in the presence of Au NPs, suggesting the materialis not ECL active. Furthermore, the Au NCs showed muchhigher electron-transfer resistance (Ret) value than that of AuNPs, due to the nonconductive property of the BSA layer inthe Au NCs (inset in Figure 1d). On the other hand, when AuNPs were assembled onto the Au NCs modified ITO elec-trode, significantly enhanced signal could be obtained due to asignal amplification effect (Figure 1d). Furthermore, wemeasured the ECL spectra by employing a series of opticalfilters, and a distinguished ECL peak at about 620 nm wasobserved, which was approximate to the PL peak (see theSupporting Information). All the obtained results showedstrong evidence that the observed emission was indeed theECL of Au NCs.

Figure 1. (a) CVs of the bare ITO electrode in the absence (1) or presence (2) of K2S2O8. Inset: DPV of the bare ITO in the absence of K2S2O8, thearrows indicate the scan direction. (b) ECL-potential curves of the bare ITO electrode (1) and Au NCs modified ITO electrode (2) with thecorresponding CV and DPV (inset). (c) ECL-potential curves of the bare ITO electrode (1) and Au NPs modified ITO electrode (2). (d)ECL-potential curves of the Au NCs modified ITO electrode (1) before and (2) after the assembly of Au NPs. Inset: Electrochemical impedancespectroscopy of (1) Au NCs and (2) Au NPs modified electrodes. (e) ECL-potential curves of the Au NCs modified ITO electrode with the potentialcycled between (1)þ2.0 and-1.5 V and (2) 0 and-1.5 V. (f) ECL-potential curves of the bare (dark) and Au NCs modified (red) Au electrodes (1,2) and glassy carbon (3, 4) electrodes. Scan rate: 100 mV s-1.

663 dx.doi.org/10.1021/ac102623r |Anal. Chem. 2011, 83, 661–665

Analytical Chemistry LETTER

In ECL system, there are two main mechanisms: one is ionannihilation ECL, the other is coreactant ECL.16 It was foundthat the ECL signal of the Au NCsmodified ITO electrode in theabsence of S2O8

2- was so weak to be distinguished from that ofbare ITO electrode, indicating that S2O8

2- played an importantrole in the ECL process as a coreactant. In addition, the cathodicECL of the Au NCs modified ITO electrode was also studied asthe potential was cycled between 0 and -1.5 V. Compared withthe ECL between þ2.0 and -1.5 V, the cathodic ECL justdecreased slightly (Figure 1e), indicating that the coreactantECL contributed to the whole ECL dominantly. Meanwhile, thesmall decrease suggested that the annihilation ECL via electron

transfer between redox species of Au NCs might also contributeto the overall ECL. According to our previous reports concerningthe cathodic ECL of QDs,18,19 it could be confirmed thatthe cathodic ECL herein was originated from the formation ofexcited-state Au NCs (Au25*) via electron-transfer (ET)annihilation of negatively charged Au25

•- and the stronglyoxidizing SO4

•- radicals produced by electroreduction ofS2O8

2-. Furthermore, the ECL behaviors of Au NCs modifiedglassy carbon and Au electrodes have also been investigated asshown in Figure 1f. Obviously, the ECL signal of thoseelectrodes was even lower than that of the bare glassy carbonand Au electrodes, revealing that ITO played an importantrole in ECL process. The ITO participated ECL behaviorshave also been reported previously, where the ITO acted as amedium for the electron transfer between QDs and dissolvedoxygen.14,15

It has been demonstrated that when metal nanoparticles, suchas Ag or Au, come in contact with a charged semiconductor, theFermi levels of the two materials equilibrate and the metalsaccept charges, such as the confirmed electron transfer from theconduction-band of excited TiO2 to Au NCs.23 Au25 NCsexhibits an electronic molecular-type HOMO-LUMO bandgapof 1.3 eV with the orbital energies of -10 and -8.7 eV,respectively.9,11,24 Meanwhile, similar to that of TiO2 (with a bandgap of 3.4 eV15), ITO has a band gap of 3.5 eV with the Fermi levelof -3.7 eV,14 which is much higher than the energy level of AuNCs. Furthermore, the conduction and valence band potentials forITOwere similar to those for In2O3, that were at-0.4 and 2.1 V (vsSCE) at pH 7, respectively.25 Thus in the cathodic ECL process,electrons could be injected into the conduction-band of ITO andthen transfer to the LUMO of Au NCs, resulting the generation ofAu25

•- (Figure 2a). In a word, ITO acted as an effective reductantfor Au25. The possible ECLmechanisms are described in Figure 2b

Figure 2. Schematic illustration of (a) electron transfer between ITOand Au NCs and (b) the ECL mechanisms of Au NCs.

Figure 3. (a) CV and DPV (inset) of Au NCs modified ITO electrode with 20 μM DA in the electrolyte. (b) Cathodic ECL profiles of the Au NCsmodified ITO electrode in the presence of different dopamine concentrations. The inset shows the correlating smoothed curves for ease of observation.(c) Linear calibration plot for DA detection. (d) Consecutive ECL intensity in the presence (dark) and absence (red) of interfering agents.

664 dx.doi.org/10.1021/ac102623r |Anal. Chem. 2011, 83, 661–665

Analytical Chemistry LETTER

with the following equations:

Au25 sfITO

Au25•- ð1Þ

S2O82- þ e- f S2O8

•2- ð2Þ

S2O8•2- f SO4

2- þ SO4•- ð3Þ

Au25•- þ SO4

•- f Au25/ þ SO4

2- ð4Þ

Au25/ f Au25

/þ hν ð5Þ

Actually, it was reported that silver nanoclusters (Ag NCs)exhibited cathodic hot electron-induced ECL with S2O8

2-

as a coreactant, and the ECL mechanisms of Ag NCs wereascribed to the ox-red excitation pathway,26 where the lumino-phore was oxidized by the same cathodically produced oxidiz-ing radical, SO4

•-, as described above. However, the reportedAg NCs ECL was obtained under a pulse voltage of-30 V thatwas unfavorable to its application owing to some unforeseenreaction under such draconian conditions. Herein, we couldobserve cathodic ECL of Au NCs, integrating with ITO andS2O8

2-, under attainable operating conditions, which couldprovide a fascinating method for both experimental andtheoretical studies.

As is well-known, dopamine (DA) is an important neuro-transmitter and plays a significant role in the function of thecentral nervous, renal, and hormonal systems.The detection ofDA has been reported based on its quenching effect on theanodic ECL of CdSe QDs resulting from an energy-transferprocess.15 In this work, when DA was injected into the electro-lyte, the cathodic ECL of the Au NCs modified ITO electrodeshowed an obvious increase. Also, the ECL increased linearlywith the increasing concentration of DA ranging from 2.5 to47.5 μM as shown in Figure 3b,c. It has been reported that theadsorption of DA on the TiO2 surface can form a charge transfercomplex, and electrons can transfer from the dopamine ligandsdirectly to the conduction band of TiO2.

27 Thus the ECLenhancement mechanism of DA here might be due to theformation of a similar charge transfer complex between DAand ITO, thus accelerating the electron injection into theconduction-band of ITO. Obviously, compared to the curvesin Figure 1b, well-defined CV and DPV could be obtained in thepresence of DA (Figure 3a), indicating that DA could indeedfacilitate the electrochemical behavior in this system.

The stability of the Au NCs modified electrodes was alsotested, which showed stable ECL signal under consecutivepotential scans from 0 to -1.5 V for 4 cycles with 40 μM DAin the electrolyte. An interference investigation was performed byusing the solution containing 30 μM of DA and 100 μM of uricacid and ascorbic acid respectively, which are two commoninterfering agents for DA detection. Also no obvious changecould be observed (Figure 3d), indicating an acceptable selec-tivity for DA detection.

In conclusion, ECL was first observed from the water-solubleand low-toxicity BSA-stabilized Au NCs. The ECL behavior andmechanism of Au NCs have been studied in detail and could beascribed to cathodic coreactant ECL based on electron transferbetween Au NCs and ITO. In addition, the ECL system showedacceptable selectivity for DA detection. The Au NCs couldbecome a promising material for ECL investigations.

’ASSOCIATED CONTENT

bS Supporting Information. Additional information asnoted in text. This material is available free of charge via theInternet at http://pubs.acs.org.

’AUTHOR INFORMATION

Corresponding Author*E-mail: jjzhu@nju.edu.cn. Phone and fax: þ86-25-83594976.

’ACKNOWLEDGMENT

L.L. and H.L. contributed equally to this work. This researchwas financially supported by the National Natural ScienceFoundation of China for the Key Program (Grants 20635020,50972058, and 20821063) and the National Basic ResearchProgram of China (Grant 2011CB933502). We also thankthe support from Discipline Crossing Foundation of NanjingUniversity, Analysis & Test Fund of Nanjing University andthe Foundation from State Key Laboratory of Bioelectronics,Southeast University.

’REFERENCES

(1) Zhu, M.; Lanni, E.; Garg, N.; Bier, M. E.; Jin, R. J. Am. Chem. Soc.2008, 130, 1138–1139.

(2) Liu, Y.; Ai, K.; Cheng, X.; Huo, L.; Lu, L.Adv. Funct. Mater. 2010,20, 951–956.

(3) Jin, R. Nanoscale 2010, 2, 343–362.(4) Muhammed, M. A. H.; Verma, P. K.; Pal, S. K.; Kumar, R. C. A.;

Paul, S.; Omkumar, R. V.; Pradeep, T. Chem.;Eur. J. 2009, 15, 10110–10120.

(5) Lin, C.-A. J.; Yang, T.-Y.; Lee, C.-H.; Huang, S. H.; Sperling,R. A.; Zanella, M.; Li, J. K.; Shen, J.-L.; Wang, H.-H.; Yeh, H.-I.; Parak,W. J.; Chang, W. H. ACS Nano 2009, 3, 395–401.

(6) Zhu, M.; Aikens, C. M.; Hollander, F. J.; Schatz, G. C.; Jin, R.J. Am. Chem. Soc. 2008, 130, 5883–5885.

(7) Negishi, Y.; Nobusada, K.; Tsukuda, T. J. Am. Chem. Soc. 2005,127, 5261–5270.

(8) Wu, Z.; Gayathri, C.; Gil, R. R.; Jin, R. J. Am. Chem. Soc. 2009,131, 6535–6542.

(9) Zhu, M.; Aikens, C. M.; Hendrich, M. P.; Gupta, R.; Qian, H.;Schatz, G. C.; Jin, R. J. Am. Chem. Soc. 2009, 131, 2490–2492.

(10) Tang, Z.; Xu, B.; Wu, B.; Germann, M. W.; Wang, G. J. Am.Chem. Soc. 2010, 132, 3367–3374.

(11) Zhu, Y.; Qian, H.; Zhu, M.; Jin, R. Adv. Mater. 2010, 22, 1915–1920.

(12) Xie, J.; Zheng, Y.; Ying, J. Y. Chem. Commun. 2010, 46, 961–963.

(13) Jao, Y.-C.; Chen, M.-K.; Lin, S.-Y. Chem. Commun. 2010, 46,2626–2628.

(14) Wang, Y.; Lu, J.; Tang, L. H.; Chang, H. X.; Li, J. H. Anal. Chem.2009, 81, 9710–9715.

(15) Liu, X.; Ju, H. X. Anal. Chem. 2007, 79, 8055–8060.(16) Miao, W. J. Chem. Rev. 2008, 108, 2506–2553.(17) Ding, S. N.; Xu, J. J.; Chen, H. Y. Chem. Commun. 2006, 3631–

3633.(18) Jie, G. F.; Zhang, J. J.; Wang, D. C.; Cheng, C.; Chen, H. Y.;

Zhu, J. J. Anal. Chem. 2008, 80, 4033–4039.(19) Jie, G. F.; Liu, B.; Pan, H. C.; Zhu, J. J.; Chen, H. Y. Anal. Chem.

2007, 79, 5574–5581.(20) Ding, Z. F.; Quinn, B.M.; Haram, S. K.; Pell, L. E.; Korgel, B. A.;

Bard, A. J. Science 2002, 296, 1293–1297.(21) Xie, J.; Zheng, Y.; Ying, J. Y. J. Am. Chem. Soc. 2009, 131, 888–889.(22) Bae, Y.; Lee, D. C.; Rhogojina, E. V.; Jurbergs, D. C.; Korgel,

B. A.; Bard, A. J. Nanotechnology 2006, 17, 3791–3797.

665 dx.doi.org/10.1021/ac102623r |Anal. Chem. 2011, 83, 661–665

Analytical Chemistry LETTER

(23) Kim, J.; Lee, D. J. Am. Chem. Soc. 2007, 129, 7706–7707.(24) Retnakumari, A.; Setua, S.;Menon,D.; Ravindran, P.;Muhammed,

H.; Pradeep, T.; Nair, S.; Koyakutty, M. Nanotechnology 2010, 21,055103.(25) Jin, S.; Song, N.; Lian, T. ACS Nano 2010, 4, 1545–1552.(26) Díez, I.; Pusa, M.; Kulmala, S.; Jiang, H.; Walther, A.; Goldmann,

A. S.; M€uller, A. H. E.; Ikkala, O.; Ras, R. H. A. Angew. Chem., Int. Ed. 2009,48, 2122–2125.(27) Wang, G.-L.; Xu, J.-J.; Chen, H.-Y. Biosens. Bioelectron. 2009, 24,

2494–2498.