J. Phys. Chem. B 2005, 109, 3099-3103
3099
Chemiluminescent Reactions Induced by Gold Nanoparticles Hua Cui,* Zhi-Feng Zhang, and Ming-Juan Shi Department of Chemistry, UniVersity of Science & Technology of China, Hefei, Anhui, P.R. China 230026 ReceiVed: October 29, 2004; In Final Form: December 19, 2004
The reaction of gold nanoparticles with a potassium periodate-sodium hydroxide-carbonate system undergoes chemiluminescence with three emission bands at 380-390, 430-450, and 490-500 nm, respectively. It was found that the light intensity increased linearly with the concentration of the gold nanoparticles, and the CL intensity increased dramatically when the citrate ions on the nanoparticle surface were replaced by SCN-. The shape, size, and oxidation state of gold nanoparticles after the chemiluminescent reaction were characterized by UV-visible absorption spectrometry, transmission electron microscopy (TEM), and X-ray photoelectron spectrometry (XPS). Gold nanoparticles are supposed to function as a nanosized platform for the observed chemiluminescent reactions. A chemiluminescent mechanism has been proposed in which the interaction between free CO3•- and O2•- radicals generated by a KIO4-NaOH-Na2CO3 system and gold nanoparticles results in the formation of emissive intermediate gold(I) complexes, carbon dioxide dimers, and singlet oxygen molecular pairs on the surface of the gold nanoparticles. This work is not only of great importance for gaining a better understanding of the unique optical and surface properties and chemical reactivity of nanoparticles but also of great potential for developing new biosensing and immunolabeling technologies.
Introduction
Experimental Section
Driven by the research for new materials with interesting and unique electronic, catalytic, and chemical properties, the field of nanoparticle research has grown immensely in recent decades.1,2 The application of the chemical reactivities of gold nanoparticles in catalysis is now an expanding area. For example, gold nanoparticles (in particular, the very active oxidesupported ones) have been frequently investigated in gas-phase catalysis.3 Gold nanoparticles also represent a useful alternative to platinum group metals for catalysis in liquid-phase oxidation.4,5 Furthermore, the size-dependent chemical and electrochemical reactivities for gold nanoparticles have been welldocumented.6-10 For example, Murray’s group reported that monolayer-protected clusters (MPCs) could store charges and then be used as quantitative redox reagents in chemical reactions.11,12 Dawson et al.13 demonstrated that short-lived radicals such as (SCN)2•- could chemically interact with gold nanoparticles and induce the oxidation of the gold surface. Chemiluminescence, which combines chemical reactions with electromagnetic radiation, has been observed for semiconductor nanoparticles in chemical or electrochemical reactions.14-16 Recently, our group reported that gold nanoparticles of 16 nm could greatly enhance two anodic and one cathodic electrogenerated chemiluminescence (ECL) peaks of luminol in neutral and alkaline media and initiate one new cathodic ECL peak because of their unique catalytic property and electrochemical reactivity.17 However, to the best of our knowledge, there is no report regarding liquid-phase chemiluminescence induced by gold nanoparticles by virtue of their catalytic and surface properties and chemical reactivity. Here, we show that gold nanoparticles can react with a KIO4-NaOH-Na2CO3 solution and generate multichannel chemiluminescence.
For experiments reported here, all the reagents used in these experiments were of analytical grade, and triple-distilled water was used throughout. Colloidal gold nanoparticles of 6 nm in diameter were synthesized by the hydroborate reduction method,18 while colloidal gold nanoparticles of 16-, 25-, 38-, 68-, and 99nm diameters were synthesized by the citrate reduction method.19 The size and shape of the synthesized gold nanoparticles were characterized by a model H-800 transmission electron microscope (Hitachi, Japan). Statistical analysis of TEM data revealed that the average diameters of the gold colloids were about 6 ( 3.5, 16 ( 1, 25 ( 2, 38 ( 2.8, 68 ( 5.1, and 99 ( 7 nm, respectively. Chemiluminescence detection was conducted on a lab-built flow-injection chemiluminescence system, consisting of a model IFIS-C flow-injection system (Ruimai Electronic Science Co., China), a model CR105 photomultiplier tube (PMT) (Bingsong Electronics Co., China), a model GD-1 luminometer (Ruimai Electronic Science Co., China), and a computer, as shown in Figure 1. The chemiluminescence signals were monitored by the PMT adjacent to the flow CL cell. The concentrations and flow rates of all the solutions were optimized. CCD images of the CL reaction were taken from a lab-built static CL reaction cell of 2 mL with a Roper Scientific CCD (Princeton Instruments, U.S.A.). The CCD camera was cooled to -35 °C, and the exposure time was 15 s. The beam was finally focused on a CCD matrix of a 512 × 512 photodetector. The CL spectrum of this system was measured using highenergy cutoff filters of various wavelengths because the light intensity was not high enough to be detected by a refitted fluorimeter.20 Fluorescence spectra and UV-visible absorption spectra were measured on a model LS-55 spectrofluorimeter (Perkin Elmer, USA) and a model UV-2401 PC spectrophotometer (Shimadzu, Japan), respectively. X-ray photoelectron spectra were acquired on a model ESCALAB MK II electron spectrograph (VG, England).
* Author to whom correspondence should be addressed. E-mail: hcui@ ustc.edu.cn. Fax: +86-551-3601592.
10.1021/jp045057c CCC: $30.25 © 2005 American Chemical Society Published on Web 02/03/2005
3100 J. Phys. Chem. B, Vol. 109, No. 8, 2005
Cui et al.
Figure 1. Schematic diagram of lab-built flow-injection chemiluminescence detection system. Figure 3. Chemiluminescence spectra for KIO4-NaOH-Na2CO3gold colloids system with and without the addition of KSCN. The solution conditions are as in Figure 2.
Figure 2. (A) CL intensity profiles via time in flow-injection mode. Solutions of 0.1 mmol/L KIO4, 0.04 mol/L NaOH, and 0.02 mol/L Na2CO3 were mixed with gold colloids, from top to bottom: colloidal gold nanoparticles of 68, 99, 38, 16, and 6 nm, 2.94 × 10-5 mol/L HAuCl4, and the supernatant solution of 68-nm gold colloids. (Inset) CCD image taken from a lab-built static CL reaction cell; the bright dots in the CCD image indicate light emission. (B) Linear relationship between CL intensity and gold concentration of 68-nm gold colloids.
Results and Discussion Size-Dependent Chemiluminescence. When a solution containing 0.04 mol/L of NaOH and 0.02 mol/L of Na2CO3 was mixed with a solution of gold colloids and then reacted with 0.1 mmol/L of KIO4, stable, reproducible, and sizedependent chemiluminescent signals were generated from larger gold nanoparticles with a diameter of 16, 38, 68, and 99 nm (Figure 2). The most intensive CL occurred for the colloidal solution of 68-nm gold nanoparticles. However, no evident chemiluminescence was observed for gold colloids with a diameter of 6 nm. The solution of precursor ionic chemicals and the colorless supernatant separated by centrifuging a colloidal solution of 68-nm gold nanoparticles were also reacted with a KIO4-NaOH-Na2CO3 system, and there was no significant CL signal, showing that gold nanoparticles were necessary for the generation of chemiluminescence for this system. For 68-nm gold nanoparticles, the CL intensity was found to be linear with the concentration of gold nanoparticles, as shown in Figure 2B. The CL spectrum for 68-nm gold colloids was measured using high-energy cutoff filters of various wavelengths as described elsewhere20 and shown in Figure 3. It was clearly indicated that there were three emission bands around 380-390, 430-450, and 490-500 nm. It is known that the fluorescent spectra of the stable emitting species should be identical with the CL spectra. However, no evident fluorescence signal was found for either the gold colloids, which were chemiluminescently active, or the reacted solution mixtures. Therefore, the emissive species for the observed chemiluminescence must be a kind of shortlived intermediate that was generated in situ during the chemical reactions in the presence of gold nanoparticles. The effect of a protecting layer on the surface of gold nanoparticles on the CL was investigated. KSCN (2.35 × 10-4 mol/L) (the same concentration as sodium citrate) was added to the 68-nm gold colloids before the reaction to replace the citrate ions on the surface of gold nanoparticles. TEM studies
showed no evident changes in the shape and size of the gold nanoparticles (Figure 4), but the citrate ions on the particle surface were supposed to be replaced by SCN- via placeexchange reactions.6,21,22 Such surface-modified 68-nm gold colloids were used to react with KIO4-NaOH-Na2CO3, and the CL intensity increased dramatically. However, no evident chemiluminescent signal could be observed for KSCN solutions. The CL spectrum for SCN--modified 68-nm gold colloids still shows three emission bands (Figure 3B). It was notable in Figure 3 that the emission band around 430-450 nm had about an 8-nm shift compared with citrate ion-protected gold nanoparticles, while the other two bands were almost identical. Characterization of Gold Nanoparticles. Since chemiluminescence could only be generated in the presence of gold nanoparticles, the gold nanoparticles would play an important role during the chemical reactions. UV-visible absorption spectra and TEM studies before and after the reactions demonstrated that the size, shape, and the maximum surface plasma resonance (SPR) absorption band for 68-nm gold nanoparticles (Figure 5A) did not change after oxidation by the KIO4-NaOH-Na2CO3 system. However, for 6-nm gold nanoparticles, UV-visible absorption spectra (Figure 5B) showed that the SPR band was evidently red-shifted and dampened after the reaction, and precipitates were visible in the resulting solution. Further efforts were made to clarify the changes occurring on the surface of gold nanoparticles. Gold colloids before and after the CL reactions were separated from the aqueous matrix by centrifuging. Both of the supernatant aqueous solutions and the solid powders were dried at room temperature under vacuum and then used for XPS studies. As shown in Figure 6, an Au signal was observed not for the partition of the supernatant aqueous solution but for the solid powders. Therefore, no dissolution of gold nanoparticles was supposed to take place during the chemical reactions. However, the oxidation states of the surface gold atoms of the nanoparticles were changed with the emergence of new peaks of Au4f binding energies as shown in Figure 6. Using Au4f5/2 as a standard, the binding energy for 68-nm gold nanoparticles shifted positively about 2.2 eV after the chemical reactions, while for SCN--modified 68-nm gold nanoparticles, it shifted positively about 3.0 eV. Therefore, the addition of SCN- was advantageous for the oxidization of the gold nanoparticles. An Au(III) complex was supposed to be generated at last on the surface of the gold nanoparticles after the chemical reactions according to the large shifts of the Au4f binding energies.23 Figure 5 shows that there are two ligand-to-metal charge-transfer absorption bands around 220 and 300 nm for Au(III) complex in solution.29 However, the charge-transfer absorption bands in the reacted mixture (Figure 5) were difficult to be resolved because strong background absorption occurred.
Chemiluminescence Induced by Gold Nanoparticles
J. Phys. Chem. B, Vol. 109, No. 8, 2005 3101
Figure 4. TEM photos for 68-nm gold nanoparticles before (A) and after (B) the addition of 2.35 × 10-4 mol/L KSCN.
Figure 6. X-ray photoelectron spectra of the Au4f regions. The inserted figure demonstrates the new peaks for 68-nm Au after the CL reaction 1 and 68-nm Au + KSCN after the CL reaction 2.
Figure 5. UV-visible absorption spectra for the CL reaction systems. (A) Au colloids(68 nm)-KIO4-NaOH-Na2CO3 system. The inserted TEM photos are the unreacted (up) and reacted (down) 68-nm gold nanoparticles; (B) Au colloids(6 nm)-KIO4-NaOH-Na2CO3 system. The solution conditions are as in Figure 2.
Possible Mechanism of the Chemiluminescent Reactions. Lin et al.24 investigated the CL caused by the redox reaction between periodate and polyhydroxyl compounds in alkaline media that could be enhanced by carbonate ions. They demonstrated the generation of free CO3•- and O2•- radicals from the KIO4-NaOH-Na2CO3 system and assigned the observed two CL bands of 436-446 and 471-478 nm to carbon dioxide dimer and singlet oxygen molecular pair emissions, respectively. Oxyanions such as CO32-, H2PO4-, and SO42- were capable of displacing the citrate-passivating ligands on the surface of the gold nanoparticles.25 In this case, when the KIO4-NaOHNa2CO3 solution was mixed with the gold colloidal solutions,
free CO3•- and O2•- radicals generated by the KIO4-NaOHNa2CO3 system might displace the citrate ions on the surface of the gold nanoparticles25 and adjacent adsorbed radicals would react to form molecular dimers as in the case of exothermic surface reactions between an atom or molecule and a metal surface.26-27 Accordingly, it was deduced that the presence of gold nanoparticles would greatly stimulate the generation of the excited states of the carbon dioxide dimer and singlet oxygen molecular pair, which contributed to the 430-450 nm and 490500 nm emission bands, respectively (Scheme 1). The emission wavelengths were slightly red-shifted compared to those of Lin’s results.24 Zhang et al.28 reported there were interactions between free radicals and gold nanoparticles. The interaction between the free radicals and the gold cores in this work was supposed to be the origin of the red-shifting of these two CL bands. Dawson and co-workers reported that pulse radiolytically generated (SCN)2•- bonded strongly to the gold nanoparticle surface and yielded an oxidation product [AuI(SCN)2]-.13 It was postulated that similar interactions between the gold cores and the bonded CO3•- and O2•- radicals may also take place, leading to the oxidation of the surface gold atoms to form similar gold(I) complexes as the intermediates. These Au(I) complexes were emissive when exposed to ultraviolet light at room temperature.30 The CL emission around the 380-390 nm band and partial
3102 J. Phys. Chem. B, Vol. 109, No. 8, 2005
Cui et al.
SCHEME 1: Possible Mechanism for the Chemiluminescence Involving Carbon Dioxide Dimer and Singlet Oxygen Molecular Pair
SCHEME 2: Possible Mechanism for the Chemiluminescence Involving the Oxidation of Surface Gold Atoms
emission around the 430-450 nm band were probably attributed to such emissive Au(I) intermediates since they assembled the reported fluorescence spectra of the Au(I) complex.29 Therefore, the 430-450 nm CL band was supposed to be composed of two components, i.e., emission from carbon dioxide dimers and emission from Au(I) intermediates. That was why this band was more susceptible to the surface state of the gold nanoparticles than the other two bands as shown in Figure 3. However, no evident fluorescence signal was found for the mixed solution after the CL reactions. Therefore, Au(I) intermediates were supposed to be readily oxidized to Au(III) on the surface of the gold nanoparticles, which was consistent with the XPS results for the reacted solution (Figure 6). When KSCN was added to the gold colloids and oxidized by KIO4-NaOH/Na2CO3, (SCN)2•- radicals were supposed to be generated, which would enhance the oxidation of the surface gold atoms,13 leading to the increase of CL intensity observed in the experiments. The proposed CL reaction pathways were shown in Scheme 2. For 6-nm colloidal gold nanoparticles, the CL intensity was extremely low, probably because the reactions taking place on the more active surface of the smaller gold nanoparticles would directly result in the generation of Au(III) sites. The red-shifted absorption band of UV-visible absorption spectra (Figure 5B) and visible precipitates in the mixed solution suggested that the 6-nm gold nanoparticles took part in the chemical reaction and aggregated after the reaction, which was different from the case for 68-nm gold nanoparticles, and could not form Au(I) intermediates. Therefore, only larger gold nanoparticles could induce such multichannel CL reactions. Conclusion The reaction of gold nanoparticles with a KIO4-NaOHNa2CO3 solution gave rise to chemiluminescence with three
emission bands at 380-390, 430-450, and 490-500 nm, probably arising from emissive intermediates such as gold(I) complexes, carbon dioxide dimers, and singlet oxygen molecular pairs on the surface of the gold nanoparticles. Chemiluminescence occurring on the surface of gold nanoparticles presented a new view of the chemical reactivity and catalytic effects of gold nanoparticles. The linear relationship between the CL intensity and the concentration of gold nanoparticles implies that gold nanoparticles may be sensitively detected by the chemiluminescence method, which can probably find future use in new biosensing and immunolabeling technologies. Acknowledgment. The support of this research by the National Natural Science Foundation of the P.R. of China (Grant No. 20375037) and the Overseas Outstanding Young Scientist Program of the Chinese Academy of Sciences is gratefully acknowledged. We also thank Prof. X. M. Liu for taking TEM photos and Prof. C. Gao and Mr. J. J. Ding for taking CCD images for us. Acknowledgments are also granted to Prof. M. R. Ji for detailed discussion about the X-ray photoelectron spectra. References and Notes (1) Link, S.; El-Sayed, M. A. Annu. ReV. Phys. Chem. 2003, 54, 331. (2) McConnell, W. P.; Novak, J. P.; Brousseau, L. C., III; Fuierer, R. R.; Tenent, R. C.; Feldheim, D. L. J. Phys. Chem. B 2000, 104, 8925. (3) Haruta, M. Catal. Today 1997, 36, 153. (4) Prati, L.; Rossi, M. Stud. Surf. Sci. Catal. 1997, 110, 509. (5) Prati, L.; Rossi, M. J. Catal. 1998, 176, 552. (6) Chen, S.; Ingram, R. S.; Hostetler, M. J.; Pietron, J. J.; Murray, R. W.; Schaaff, T. G.; Khoury, J. T.; Alvarez, M. M.; Whetten, R. L. Science 1998, 280, 2098. (7) Weiz, D. A.; Lin, M. Y.; Sandroff, C. J. Surf. Sci. 1985, 158, 147. (8) Haruta, A. Chem. ReV. 2003, 3, 75.
Chemiluminescence Induced by Gold Nanoparticles (9) Jana, N. R.; Gearheart, L.; Obare, S. O.; Murphy, C. J. Langmuir 2002, 18, 922. (10) Chen, S.; Murray, R. W.; Feldberg, S. W. J. Phys. Chem. B 1998, 102, 9898. (11) Templeton, A. C.; Wuelfing, W. P.; Murray, R. W. Acc. Chem. Res. 2000, 33, 27. (12) Pietron, J. J.; Hicks, J. F.; Murray, R. W. J. Am. Chem. Soc. 1999, 121, 5565. (13) Dawson, A.; Kamat, P. V. J. Phys. Chem. B 2000, 104, 11842. (14) Ding, Z.; Quinn, B. M.; Haram, S. K.; Pell, L. E.; Korgel, B. A.; Bard, A. J. Science 2002, 296, 1293. (15) Myung, N.; Ding, Z.; Bard, A. J. Nano Lett. 2002, 2, 1315. (16) Poznyak, S. K.; Talapin, D. V.; Shevchenko, E. V.; Weller, H. Nano Lett. 2004, 4, 693. (17) Cui, H.; Xu, Y.; Zhang, Z. F. Anal. Chem. 2004, 76, 4002. (18) Brown, K. R.; Fox, A. P.; Natan, M. J. J. Am. Chem. Soc. 1996, 118, 1154. (19) Frens, G. Nat. Phys. Sci. 1973, 241, 20.
J. Phys. Chem. B, Vol. 109, No. 8, 2005 3103 (20) Cui, H.; Zou, G. Z.; Lin, X. Q. Anal. Chem. 2003, 75, 324. (21) Hickman, J. J.; Ofer, D.; Laibinis, P. E.; Whitesides, G. M.; Wrighton, M. S. Science 1991, 252, 688. (22) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin, C. A. Science 1997, 277, 1078. (23) Kitagawa H.; Kojima N.; Nakajima T. J. Chem. Soc., Dalton Trans. 1991, 3121. (24) Lin, J. M.; Yamada, M. Anal. Chem. 1999, 71, 1760. (25) Cumberland, S. L.; Strouse, G. F. Langmuir 2002, 18, 269. (26) Greber, T. Surf. Sci. Rep. 1997, 28, 3. (27) Gergen, B.; Nienhaus, H.; Weinberg, W. H.; McFarland, E. W. Science 2001, 294, 2521. (28) Zhang, Z. Y.; Berg, A.; Levanon, H.; Fessenden, R. W.; Meisel, D. J. Am. Chem. Soc. 2003, 125, 7959. (29) Rawashdeh-Omary, M. A.; Omary, M. A.; Patterson, H. H.; Fackler, J. P. J. Am. Chem. Soc. 2001, 123, 11237. (30) Gangopadhayay, A. K.; Chakravorty, A. J. Chem. Phys. 1961, 35, 2206.
