Article pubs.acs.org/ac
New Insights into Electrocatalysis Based on Plasmon Resonance for the Real-Time Monitoring of Catalytic Events on Single Gold Nanorods Chao Jing,† Frankie James Rawson,‡ Hao Zhou,† Xin Shi,† Wen-Hui Li,† Da-Wei Li,† and Yi-Tao Long*,† †
Key Laboratory for Advanced Materials and Department of Chemistry, East China University of Science and Technology, Shanghai, 200237 P. R. China ‡ Laboratory of Biophysics and Surface Analysis, School of Pharmacy, University of Nottingham, University park, Nottingham, NG7 2NR United Kingdom S Supporting Information *
ABSTRACT: Gold nanoparticles (GNPs) have been widely applied in industrial catalysis and electrocatalysis. Owing to their wide variety of shapes, sizes, and compositions, a range of different catalytic properties is possible. Thus, it is important to monitor catalytic processes and their mechanisms on single GNP surfaces to avoid averaging effects in bulk systems. Therefore, a novel method based on dark-field scattering spectroscopy was developed to monitor, in real-time, the electrocatalytic oxidation of hydrogen peroxide on a single gold nanoparticle surface. The catalytic mechanism was revealed via the plasmon resonance scattering spectral shift of single gold nanorod with the elimination of bulk effect. Moreover, we found that the presence of chloride ions could block the catalytic activity of nanorods for the oxidation of H2O2. Most importantly, it was discovered that individual nanoparticles have variable properties with different spectra shifts during the catalytic process. The obtained optical signals from individual nanorods not only offer versatile information regarding the reaction but also improve the understanding of electrochemistry and the catalysis mechanism of single nanoparticles.
C
an urgent requirement to develop devices and techniques capable of in situ monitoring electrocatalytic events in real-time taking place at single catalyst nanoparticles. Contributions have emerged that attempt to monitor electrocatalytic reaction on single nanoparticles, such as surface plasmon resonance (SPR) imaging, in which the electrochemical current on single Pt nanoparticles has been imaged. This technique provided an efficient method to monitor the electrochemical reactions on single nanoparticles.16−18 Recently, dark-field microscopy (DFM), which enables the observation of scattering spectroscopy of single plasmonic nanoparticles, has attracted more and more attention.19−24 The scattering light of plasmonics (diameter d, 3−100 nm ≪ wavelength of incident light λ, 400−900 nm) originating from localized surface plasmon resonance is highly sensitive to surface morphology, composition, surroundings, and surface electron density of the particles.25−29 Notably, DFM integrated electrochemistry offers a unique platform that affords the possibility of monitoring reaction process and electron charging on the single nanoparticles via plasmon resonance band shifts and addresses the key challenges highlighted thus far.30−34 The aim of these investigations was to overcome the problem associated with low detection limits due to small changes in
atalytic reactions, especially noble metal catalysis, are crucial for the development of processes used within science and industry including photovoltaic conversion, gas treatment, and biosensing.1−4 Plasmonic gold nanoparticles (GNPs) have been widely applied in electrochemical catalysis as functional substrates for construction of sensitive sensors due to their unique physical and chemical properties.5,6 Commonly for studies of electrocatalysis, investigations are performed in which the results represent averages obtained across a population of nanoparticles. Particularly, for heterogeneous crystal catalysis, such as that which takes place at GNPs, the activity for every individual nanoparticle is different depending on its size, shape, and composition.7−9 This can make it difficult to elucidate a detailed mechanism of the various interactions taken place that underpins the electrocatalytic mechanism. Therefore, in order to truly understand the controlling effects of such catalytic events, it is useful to study reactions occurring at individual nanoparticles.10,11 Thereby, we can eliminate average effects and better understand the factors inducing catalysis containing the shape, quantum effect, and surface functions. Technology capable of monitoring redox events at a nanoparticle will enable for the efficient development and improvement of reaction rates and efficiency.12,13 The redox state of the surface of a catalyst is a crucial component in controlling the catalytic process.14,15 However, our ability to peer into the surface change on nanoparticles via electrochemical investigations is extremely difficult due to the magnitude of electric signals generated. Consequently, there is © 2014 American Chemical Society
Received: February 27, 2014 Accepted: April 26, 2014 Published: April 28, 2014 5513
dx.doi.org/10.1021/ac500785u | Anal. Chem. 2014, 86, 5513−5518
Analytical Chemistry
Article
preparation of all the solutions. The indium tin oxide (ITO) slides were purchased from Geao Co. Ltd. (Wuhan, China). Preparation of the Samples. Gold nanoparticles used in the experiments were immobilized on the (ITO) slides. The surfaces of ITO slides were cleaned in an ultrasonic bath. The slides were first ultrasonic treated in ethanol and acetone to remove oily matter and then treated in water to remove watersoluble matter. The samples were cleaned in each solvent for more than 1 h. The cleaned ITO slides were then modified with GNPs via electrostatic adsorption by placing them in the diluted gold colloid solution (50 times) for 15 min. The GNPfunctionalized ITO slides were rinsed with water and dried under a stream of ultrapure nitrogen prior to the dark-field measurements. The nanoparticles adsorbed on ITO slides were stable during the electrochemical scanning as shown in Figure S11, Supporting Information. Dark-Field Microscopy and Scattering Spectroscopy. The dark-field measurements were carried out on an inverted microscope (eclipse Ti−U, Nikon, Japan) that was equipped with a dark-field condenser (0.8 < NA < 0.95) and a 40× objective lens (NA = 0.6). The white light source was a 100 W halogen lamp. The GNP-functionalized slides were immobilized on a platform, and the white light source was used to excite the GNPs and generate plasmon resonance scattering light. A true-color digital camera (Nikon DS-fi, Japan) was used to capture the dark-field color images. The scattering light of gold nanoparticle was split by a monochromator (Acton SP2300i, Princeton Instruments, USA) that was equipped with a grating (grating density: 300 lines/mm; blazed wavelength: 500 nm) and recorded by a spectrometer CCD (Pixis 400, Princeton Instruments, USA) to obtain the scattering spectra. The exposure time for every scattering spectrum was 10 s. Additional details of the dark-field microscopy and spectroscopy setup have been reported in a previous work.20 Electrochemistry. All electrochemical measurements were performed using a CHI 660 (Chenhua, Shanghai, China) electrochemistry station. Here, a homemade microelectrochemical cell has been developed; the ITO slide acts as the working electrode, and two Pt wires function as the counter and reference electrode, respectively. Here, the Pt wire was selected as reference electrode to eliminate the influence of Cl− ions dissociated from AgCl. All measurements were performed at 25 °C. Calibration of Pt Reference Electrode. The potential of Pt electrode was calibrated via potassiumferricyanide as shown in Figure S1, Supporting Information. The E01/2 of 1.00 mM K3[Fe(CN)6] solution using Pt quasi-reference electrode and saturated calomel electrode (SCE) was −0.127 and 0.211 V, respectively.
surface redox state which cannot be detected by simple electrochemical investigation using techniques such as cyclic voltammetry on single gold nanoparticles. We devise a novel method to monitor electrochemical redox events on single gold nanoparticles from the optical insight using a developed spectroelectrochemistry technique which combines DFM and electrochemistry (Figure 1a). We show that single GNPs
Figure 1. (a) Setup of dark-field microscopy integrated with an electrochemical workstation. (b) Scheme of electrocatalytic oxidation of H2O2 on the surface of gold nanorods in KNO3 and KCl solutions, respectively. (c−e): Simultaneous plasmonic scattering spectra peak shift of single nanorod (c) and electrochemical current of entire samples (d) under the applied triangular wave potential (e) on the GNRs.
electrochemically catalyze the oxidation of H2O2 in KNO3 and KCl solutions as shown in Figure 1b. Indium tin oxide (ITO) was modified by electrostatically adhering gold nanorods (GNR), and this acted as working electrode. Two Pt wires were immersed in the electrolyte and acted as counter electrode and quasi-reference electrode, respectively. The applied potential was calibrated via potassium ferricyanide solution (Figure S1, Supporting Information). The time dependent dark-field scattering spectra of the single GNR during the triangular wave potential scanning reveal the information about surface of a single GNR during the electrocatalytic process (Figure 1c−e). We demonstrate for the first time that the electrochemical catalytic oxidation of H2O2 at a single GNR can be investigated, and it was determined that different nanoparticles exhibited various catalytic activities. Moreover, this indicates the importance of developing techniques capable of monitoring in situ events occurring at single nanoparticles to shed new light on the mechanisms involved.
■
RESULTS AND DISCUSSION Scattering Spectra of Single Gold Nanorod. Single GNR (∼40 nm × 65 nm) exhibited spectra peak wavelength λmax of 599 nm at an open circuit corresponding to the red color as shown in Figure S3, Supporting Information. Nanorods were selected since their scattering spectra were more sensitive to the surface changes than nanospheres.35 After applying 1.00 V potential, the nanospheres only showed ca. 2 nm red shift as shown in Figure S2, Supporting Information. The scattering cross section of plasmonic nanoparticles was shown as eqs S1 and S2, Supporting Information. The Δλmax induced by the surrounding medium and electron density can be calculated from eqs 1 and 2.36−38
■
EXPERIMENTAL SECTION Materials. All reagents were of analytical grade. Absolute ethanol (AR), acetone (AR), potassium chloride (AR), and potassium nitrate (AR) were purchased from Sigma (USA). Gold nanorods (40 nm × 65 nm; 40 nm × 84 nm; 10 O.D.) were purchased from Nanoseedz (Hong Kong, China). Ultrapure water with a resistivity of 18.2 MΩ·cm was produced using a Milli-Q apparatus (Millipore, USA) and used in the 5514
dx.doi.org/10.1021/ac500785u | Anal. Chem. 2014, 86, 5513−5518
Analytical Chemistry
Article
Δλmax ≈ m(Δn)(1 − exp(− 2d /ld))
(1)
⎛1 − L ⎞ ΔN ⎟ε λ ε+⎜ ⎝ L ⎠m 2N
(2)
Δλmax = −
Information). Interestingly, after the addition of H2O2 (0.10 mM), an obvious oxidation peak at ca. 0.78 V was observed as illustrated in Figure 2c. As the concentration of H2O2 increased, the oxidation peaks increased gradually indicating that the peak current originated from the catalytic oxidation of H2O2. On the contrary, no current peak was observed on bare ITO electrode in H2O2 solution (see Figure S6, Supporting Information). To investigate the reaction process and explore the catalytic mechanism on the surface of a single GNR, the simultaneous dark-field scattering spectra of single GNR were recorded during the CV scanning which was performed at a scan rate of 10 mVs−1. Figure 2b,d (III) displays the scattering spectra Δλmax of single GNR (∼40 nm × 65 nm) obtained in the absence of H2O2. Figure 2d (I) displays the scattering spectra Δλmax of single GNR obtained in the presence of H2O2 which showed less shifts in Δλmax when compared with spectra in Figure 2d (III). In detail, for spectra obtained in the absence of H2O2, on applying a negative potential, a blue shift in the λmax of 1−2 nm occurred, which can be explained by the negative potential increasing the electron density of the GNR surface. As the potential increased from 0.00 to 0.70 V, the scattering peak was red-shifted which occurred due to double layer charging and electron discharge caused by the applied positive potential. Between the potential of 0.75 V on the forward scan and 0.75 V on the reverse scan, a plateau in the Δλmax is observed. This can be attributed to the surface of the GNR being oxidized over this potential range, and also the electron discharge. When the potential was below 0.40 V in the reverse scan, GNR was reduced gradually and the λmax blue-shifted to the initial position indicating good redox reversibility of GNRs. It could be found that before the oxidation potential, the nanorods showed an average peak shift rate of ca. 0.04 nms−1 due to the double layer charging and ions adsorption. After the oxidation potential, the nanorods showed a fast peak shift rate of ca. 0.3 nms−1 as well as the similar blue peak shift rate starting from the reduction potentials as shown in Figure S7, Supporting Information. Scheme 1 illustrates the reaction mechanism on a single GNR in the presence and absence of H2O2. During CVs performed without H2O2, the surface of GNR was oxidized into gold hydroxide and gold oxide (see eqs S3 and S4, Supporting Information).39 Gold hydroxide/oxide was represented as a compact α-oxide film on the surface of GNR (Scheme 1a).40 The compact film was assumed to be 0.23 nm in thickness and 3.3−1.35j in refractive index. The data we use originates from a second cycle of CVs, and therefore, the thickness of the hydroxide/oxide film on the gold may be more than 0.23 nm.
Here, m is the sensitivity factor of GNPs (Δλmax per refractive index unit (RIU) changing, nm/RIU), Δn is the refractive index changes of surrounding medium, d is the effective thickness of the adsorption layer, ld is the electromagnetic field decay length, εm is the dielectric constant of the surrounding environment, ε is the dielectric constant of the GNP, N is the electron density of GNPs, and L is the shape factor. As the surrounding refractive index n increased, the scattering spectra of GNPs would red shift. Additionally, the electron density change is proportional to the spectral peak shift. Thus, Δλmax of a single GNR offers information such as electron transfer and substance exchange of the redox process which is relevant to the reaction mechanism. Electrocatalytic Process of H2O2 on Single GNR in KNO3 Solution. The electrochemical behavior of GNRs deposited on ITO was investigated using cyclic voltammetry. Cyclic voltammograms (CVs) were performed at GNR modified ITO in KNO3, beginning from a start potential of −0.40 V, switching the scan at 1.00 V, and an end potential of −0.40 V (Figure 2a). At the scan rate of 0.1 V s−1, CVs
Figure 2. (a) CVs of GNRs in 0.10 M KNO3 solution, scan rate of (a1) to (a7): 30, 50, 80, 100, 300, 500, and 800 mVs−1, Pt quasireference electrode. (b) Scattering spectra of single GNR (∼40 nm × 65 nm) under the cyclic triangle wave scanning. (c) CVs on GNR modified ITO electrodes of H2O2 in 0.10 M KNO3 solution, the concentration of H2O2 from (c1) to (c7): 0.10, 0.20, 0.30, 0.50, 0.70, 0.90, and 1.00 mM; scan rate: 100 mVs−1, Pt quasi-reference electrode. (d) Scattering spectra Δλmax of two types of single GNR: 40 nm × 65 nm (I, III) and 40 nm × 84 nm (II, IV) in the presence (I, II) and absence (III, IV) of 1.00 mM H2O2 in 0.10 M KNO3 solution under the applied potential (V) from −0.10 to 1.00 V.
H 2O2 ⇔ O2 + 2H+ + 2e−
(3)
Au(OH)x + x H+ + x e− ⇔ Au + x H 2O
(4)
AuOx + 2x H+ + 2x e− ⇔ Au + x H 2O
(5)
The formation of hydroxide/oxide film induced the obvious red shift of scattering spectra λmax as a result of their high refractive index. In the presence of H2O2, GNR surface was first oxidized by applied high potential (more than 0.70 V). At the same time, GNR with hydroxide/oxide film catalyzed the oxidation of H2O2 and participated in the redox reaction. As eqs 3−5 depicted, H2O2 was oxidized into O2 and protons while gold hydroxide/oxide was reduced into gold atoms, leading to less red shift compared with the condition without H2O2. For the oxidation of H2O2, we presumed the production
exhibited an oxidation peak at 0.76 V and a reduction peak at 0.18 V. Both the anodic and cathodic peak currents were proportional to scan rate indicative of a surface controlled process (see Figure S4, Supporting Information). Control CVs were performed at bare ITO electrode (not modified with GNR), and no redox peaks were observed. This confirms that the peaks seen in Figure 2a resulted from redox events occurring at the GNRs (see Figure S5, Supporting 5515
dx.doi.org/10.1021/ac500785u | Anal. Chem. 2014, 86, 5513−5518
Analytical Chemistry
Article
and catalytic mechanism of the single GNR and provide more comprehensive and detailed information compared with electrochemical current signals which are averages of all GNRs on the surface of ITO. Electrocatalytic Process of H2O2 on Single GNR in KCl Solution. The reaction condition is also vital for the catalytic efficiency of gold nanoparticles, especially the influence of chloride ions. Utilizing DFM coupled the electrochemistry; the effect of Cl− ions on the electrocatalysis of H2O2 at single GNRs was investigated. 0.10 M KCl solution caused strong corrosion of gold on GNRs (see Figures S8−10, Supporting Information). This occurs due to the formation of the goldchloride complex, which leads to the dissolution of gold atoms.42 This was also confirmed by performing consecutive CVs in 0.10 M KCl at GNR modified ITO electrodes (Figure 3a) which showed that the oxidation peak current associated
Scheme 1. Reaction Mechanism of the Electrocatalytic Oxidation of H2O2 on Single GNR Surfacea
a (a) GNR with hydroxide/oxide film was partially reduced into Au atom, and H2O2 was oxidized into O2 after treatment of H2O2 in KNO3 solution. (b) GNR was corroded by the formation of [AuClx]−, and gold chloride blocked the catalytic activity of GNR for oxidation of H2O2 in KCl solution.
as O2 and H+ for that oxygen is the only element enabled to be oxidized. To note, the λmax was still red-shifted in the presence of H2O2 compared to the initial spectral peak at 599 nm. These results may be due to that only a proportion of the GNR surface atoms was involved in the reaction, and the positive potential induced electron discharge. In this catalysis reaction, GNR acted as a reagent and not as conducting wires or substrate as reported in previous literature examples of electrocatalytic processes.41 It is important to note that every gold nanorod exhibited various spectral shifts and catalytic activity during the redox reaction that the common electrochemistry method could not discover. As shown in Figure 2d (II, IV), the scattering spectra of single GNR (40 nm × 84 nm) during the CV scanning in the absence and presence of H2O2 showed about 14 and 6 nm red shift, respectively, which are more than GNR of 40 nm × 65 nm. These results may be due to the larger aspect ratio GNRs being more sensitive to the medium changes. Also, the selected GNR exhibited stronger catalytic ability, displaying more spectral shift between Figure 2d (II and IV). In addition, the scattering peak shifts of different single gold nanorods during the CV scanning were investigated. As shown in Figure S7, Supporting Information, the nanorods with initial scattering peak wavelength from 645 to 720 nm showed various peak red shifts after their oxidation and catalysis process. These results show that the heterogeneity in size and shape of individual GNRs caused their differing ability to electrochemically catalyze the oxidation of H2O2. Nanorods with similar scattering spectra peak wavelength may have very different results under CV scanning as shown in Figure S7, Supporting Information (λ0 = 665 nm and λ0 = 667 nm). This novel developed technique provides a tool to highlight which nanoparticles provide the highest catalytic efficiency and thus may be useful in highthroughput screening of catalysts. This method therefore eliminates the average effect of monitoring nanoparticles in the bulk system. That is, the scattering spectra reveal the redox
Figure 3. (a) Time dependent CVs of GNRs in 0.10 M KCl solutions; scan rate: 100 mVs−1, Pt quasi-reference electrode. (b) Scattering spectra of single GNR under the cyclic triangle wave scanning. (c) CVs of H2O2 on GNR modified ITO electrodes in 0.02 M KCl solutions; the concentration of H2O2 from (c1) to (c6): 0.10, 0.30, 0.50, 0.70, 0.90, and 1.00 mM; scan rate: 100 mVs−1. (d) Scattering spectra Δλmax of single GNR in the presence (I, III) and absence (II, IV) of 1.00 mM H2O2 in 0.10 M KNO3 (I, II) solution and 0.02 M KCl solution (III, IV) under the applied potential (V).
with the GNR becomes smaller with time as the gold corrosion increases. Therefore, to avoid rapid dissolution of gold atoms, 0.02 M KCl solution was used as electrolyte. As continuous CVs performing at GNR modified electrodes in the presence of H2O2, the electrochemical current was similar to the bare ITO electrode for the formation of gold chloride. From the nanoscale view, after 0.70 V in the forward scan, the scattering spectra Δλmax without H2O2 in Figure 3d (IV) red-shifted distinctly compared with Figure 3d (II). This result is explained by the formation of gold chloride. As previously explained, at lower potential (less than 600 mV vs Ag/AgCl), the gold-chloride complex mainly has two types as depicted below:42,43 Au + 2Cl(aq)− ⇔ [AuCl 2](ad)− + e− 5516
(6)
dx.doi.org/10.1021/ac500785u | Anal. Chem. 2014, 86, 5513−5518
Analytical Chemistry Au + 4Cl(aq)− ⇔ [AuCl4](ad)− + 3e−
(7)
CONCLUSIONS
■
ASSOCIATED CONTENT
■
REFERENCES
(1) Das, J.; Aziz, M. A.; Yang, H. J. Am. Chem. Soc. 2006, 128, 16022−16023. (2) Catchpole, K. R.; Polman, A. Opt. Express 2008, 16, 21793− 21800. (3) Liu, N.; Tang, M. L.; Hentschel, M.; Giessen, H.; Alivisatos, A. P. Nat. Mater. 2011, 10, 631−636. (4) Hashmi, A. S.; Hutchings, G. J. Angew. Chem., Int. Ed. 2006, 45, 7896−7936. (5) Cui, H.; Xu, Y.; Zhang, Z. F. Anal. Chem. 2004, 76, 4002−4010. (6) Guerrette, J. P.; Percival, S. J.; Zhang, B. J. Am. Chem. Soc. 2013, 135, 855−861. (7) Langille, M. R.; Personick, M. L.; Zhang, J.; Mirkin, C. A. J. Am. Chem. Soc. 2012, 134, 14542−14554. (8) Eustis, S.; El-Sayed, M. A. Chem. Soc. Rev. 2006, 35, 209−217. (9) Pyykko, P. Angew. Chem., Int. Ed. 2004, 43, 4412−4456. (10) Chen, P.; Zhou, X. C.; Andoy, N. M.; Han, K.-S.; Choudhary, E.; Zou, N. M.; Chen, G. Q.; Shen, S. Chem. Soc. Rev. 2014, 43, 1107− 1117. (11) Han, K. S.; Liu, G. K.; Zhou, X. C.; Medina, R. E.; Chen, P. Nano Lett. 2012, 12, 1253−1259. (12) Fan, F.-R. F.; Bard, A. J. Nano Lett. 2008, 8, 1746−1749. (13) Xiao, X. Y.; Fan, F.-R. F.; Zhou, J. P.; Bard, A. J. J. Am. Chem. Soc. 2008, 130, 16669−16677. (14) Lopez, N.; Nørskov, J. K. J. Am. Chem. Soc. 2002, 124, 11262− 11263. (15) Ghosh, A.; Basak, S.; Wunsch, B. H.; Kumar, R.; Stellacci, F. Angew. Chem., Int. Ed. 2011, 50, 7900−7905. (16) Wang, W.; Tao, N. Anal. Chem. 2014, 86, 2−14. (17) Brust, M.; Gordillo, G. J. J. Am. Chem. Soc. 2012, 134, 3318− 3321. (18) Shan, X.; Díez-Pérez, I.; Wang, L.; Wiktor, P.; Gu, Y.; Zhang, L.; Wang, W.; Lu, J.; Wang, S.; Gong, Q.; Li, J.; Tao, N. Nat. Nanotechnol. 2012, 7, 668−672. (19) Jing, C.; Gu, Z.; Ying, Y.-L.; Li, D.-W.; Zhang, L.; Long, Y.-T. Anal. Chem. 2012, 84, 4284−4291. (20) Zhang, L.; Li, Y.; Li, D.-W.; Jing, C.; Chen, X.; Lv, M.; Huang, Q.; Long, Y.-T.; Willner, I. Angew. Chem., Int. Ed. 2011, 50, 6789− 6792. (21) Shi, L.; Jing, C.; Ma, W.; Li, D. W.; Halls, J. E.; Marken, F.; Long, Y. T. Angew. Chem., Int. Ed. 2013, 52, 6011−6014. (22) Cheng, J.; Liu, Y.; Cheng, X. D.; He, Y.; Yeung, E. S. Anal. Chem. 2010, 82, 8744−8749. (23) Xiao, L.; Wei, L.; Cheng, X.; He, Y.; Yeung, E. S. Anal. Chem. 2011, 83, 7340−7347. (24) Li, K.; Qin, W.; Li, F.; Zhao, X.; Jiang, B.; Wang, K.; Deng, S.; Fan, C.; Li, D. Angew. Chem., Int. Ed. 2013, 52, 11542−11545. (25) Li, Y.; Jing, C.; Zhang, L.; Long, Y.-T. Chem. Soc. Rev. 2012, 41, 632. (26) Zheng, X.; Liu, Q.; Jing, C.; Li, Y.; Li, D.; Luo, W.; Wen, Y.; He, Y.; Huang, Q.; Long, Y. T.; Fan, C. Angew. Chem., Int. Ed. 2011, 50, 11994−11998. (27) Kwon, M. J.; Lee, J.; Wark, A. W.; Lee, H. J. Anal. Chem. 2012, 84, 1702−1707. (28) Hoggard, A.; Wang, L. Y.; Ma, L. L.; Fang, Y.; You, G.; Olson, J.; Liu, Z.; Chang, W. S.; Ajayan, P. M.; Link, S. ACS Nano 2013, 7, 11209−11217. (29) Solis, D., Jr.; Paul, A.; Olson, J.; Slaughter, L. S.; Swanglap, P.; Chang, W. S.; Link, S. Nano Lett. 2013, 13, 4779−4784. (30) Novo, C.; Funston, A. M.; Gooding, A. K.; Mulvaney, P. J. Am. Chem. Soc. 2009, 131, 14664−14666.
In conclusion, we monitored the electrocatalytic oxidation of H2O2 on the surface of single gold nanorod via an optical and visible method for the first time. From the plasmon resonance scattering signals, we found that bare GNRs could catalyze the oxidation of H2O2 and the gold atoms on the surface were involved in the redox reaction via the formation of gold hydroxide/oxide. Notably, from the single-particle spectra, it was confirmed that every individual nanoparticle showed various spectral shifts due to its unique sensitivity, catalytic ability, and conductivity. The approach provides effective guidance for the screening of optimal catalyst parameters. Furthermore, Cl− ions have intensive interaction with Au inducing the formation of metal-ion complex and could block the catalytic activity for H2O2. This method offers a unique way for monitoring electrochemical processes and catalysis reaction at single particle level which would improve the understanding of the mechanism of electrochemistry and catalysis of metal nanoplasmonics including electron transfer, substance exchanging, and catalyst poisoning.
S Supporting Information *
Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.
■
ACKNOWLEDGMENTS
This research was supported by 973 Program (2013CB733700), National Science Fund for Distinguished Young Scholars (21125522), and the National Natural Science Foundation of China (21327807).
Both AuCl2− and AuCl4− could be adsorbed on the surface of gold as illustrated in Scheme 1b. The formed gold chloride may be an incompact structure which coats the surface of GNR. This may expand the complex film thickness and enhance the medium refractive index greatly inducing significant red shift of the scattering spectra. Besides, after one triangle wave scanning, the scattering peak (Figure. 3d, IV) could not be shifted back to the initial position (red shift about 5 nm). The gold-chloride complex began to dissolve upon the potential of 0.70 V which caused the irregular shape of GNR. In Figure 3d (III), no obvious scattering peak changes were obtained compared with line (IV), suggesting the catalytic activity was blocked due to the fact that gold chloride could not be reduced into gold atoms by H2O2. This subsequently means hydroxide/oxide formation is critical for causing the catalytic oxidation of H2O2. Most reference electrodes such as Ag/AgCl, saturated calomel electrode, have Cl− ions inside. Thus, electrochemical experiments that include gold should not operate for a long time to reduce the effect of Cl− ions in reference electrodes. In addition, it is appropriate to use a salt bridge to slow down the diffusion of Cl− ions.
■
■
Article
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Author Contributions
C.J, D.-W.L., and Y.-T.L. designed the research; C.J., H.Z., X.S., and W.-H.L. performed research and analyzed data; C.J., F.J.R., and Y.-T.L. wrote the paper. Notes
The authors declare no competing financial interest. 5517
dx.doi.org/10.1021/ac500785u | Anal. Chem. 2014, 86, 5513−5518
Analytical Chemistry
Article
(31) Long, Y.-T.; Jing, C. Localized Surface Plasmon Resonance Based Nanobiosensors, 1st ed.; Springer: Heidelberg, 2014. (32) Dondapati, S. K.; Ludemann, M.; Müller, R.; Schwieger, S.; Schwemer, A.; Händel, B.; Kwiatkowski, D.; Djiango, M.; Runge, E.; Klar, T. A. Nano Lett. 2012, 12, 1247−1252. (33) Hill, C. M.; Pan, S. L. J. Am. Chem. Soc. 2013, 135, 17250− 17253. (34) Hill, C. M.; Clayton, D. A.; Pan, S. L. Phys. Chem. Chem. Phys. 2013, 15, 20797−20807. (35) Novo, C.; Funston, A. M.; Mulvaney, P. Nat. Nanotechnol. 2008, 3, 598−602. (36) Mie, G. Ann. Phys. 1908, 330, 377−445. (37) Novo, C.; Mulvaney, P. Nano Lett. 2007, 7, 520−524. (38) Anker, J. N.; Hall, W. P.; Lyandres, O.; Shah, N. C.; Zhao, J.; Van Duyne, R. P. Nat. Mater. 2008, 7, 442−453. (39) Li, Y.; Cox, J. T.; Zhang, B. J. Am. Chem. Soc. 2010, 132, 3047− 3054. (40) Xia, S. J.; Birss, V. I. J. Electroanal. Chem. 2001, 500, 562−573. (41) Pingarrón, J. M.; Yáñez-Sedeño, P.; González-Cortés, A. Electrochim. Acta 2008, 53, 5848−5866. (42) Dahlin, A. B.; Zahn, R.; Vörös, J. Nanoscale 2012, 4, 2339. (43) Sannomiya, T. Langmuir 2010, 26, 7619−7626.
5518
dx.doi.org/10.1021/ac500785u | Anal. Chem. 2014, 86, 5513−5518
New Insights into Electrocatalysis Based on Plasmon Resonance for the Real-Time Monitoring of Catalytic Events on Single Gold Nanorods Chao Jing,† Frankie James Rawson,‡ Hao Zhou,† Xin Shi,† Wen-Hui Li,† Da-Wei Li,† and Yi-Tao Long*,† †
Key Laboratory for Advanced Materials and Department of Chemistry, East China University of Science and Technology, Shanghai, 200237 P. R. China ‡ Laboratory of Biophysics and Surface Analysis, School of Pharmacy, University of Nottingham, University park, Nottingham, NG7 2NR United Kingdom S Supporting Information *
ABSTRACT: Gold nanoparticles (GNPs) have been widely applied in industrial catalysis and electrocatalysis. Owing to their wide variety of shapes, sizes, and compositions, a range of different catalytic properties is possible. Thus, it is important to monitor catalytic processes and their mechanisms on single GNP surfaces to avoid averaging effects in bulk systems. Therefore, a novel method based on dark-field scattering spectroscopy was developed to monitor, in real-time, the electrocatalytic oxidation of hydrogen peroxide on a single gold nanoparticle surface. The catalytic mechanism was revealed via the plasmon resonance scattering spectral shift of single gold nanorod with the elimination of bulk effect. Moreover, we found that the presence of chloride ions could block the catalytic activity of nanorods for the oxidation of H2O2. Most importantly, it was discovered that individual nanoparticles have variable properties with different spectra shifts during the catalytic process. The obtained optical signals from individual nanorods not only offer versatile information regarding the reaction but also improve the understanding of electrochemistry and the catalysis mechanism of single nanoparticles.
C
an urgent requirement to develop devices and techniques capable of in situ monitoring electrocatalytic events in real-time taking place at single catalyst nanoparticles. Contributions have emerged that attempt to monitor electrocatalytic reaction on single nanoparticles, such as surface plasmon resonance (SPR) imaging, in which the electrochemical current on single Pt nanoparticles has been imaged. This technique provided an efficient method to monitor the electrochemical reactions on single nanoparticles.16−18 Recently, dark-field microscopy (DFM), which enables the observation of scattering spectroscopy of single plasmonic nanoparticles, has attracted more and more attention.19−24 The scattering light of plasmonics (diameter d, 3−100 nm ≪ wavelength of incident light λ, 400−900 nm) originating from localized surface plasmon resonance is highly sensitive to surface morphology, composition, surroundings, and surface electron density of the particles.25−29 Notably, DFM integrated electrochemistry offers a unique platform that affords the possibility of monitoring reaction process and electron charging on the single nanoparticles via plasmon resonance band shifts and addresses the key challenges highlighted thus far.30−34 The aim of these investigations was to overcome the problem associated with low detection limits due to small changes in
atalytic reactions, especially noble metal catalysis, are crucial for the development of processes used within science and industry including photovoltaic conversion, gas treatment, and biosensing.1−4 Plasmonic gold nanoparticles (GNPs) have been widely applied in electrochemical catalysis as functional substrates for construction of sensitive sensors due to their unique physical and chemical properties.5,6 Commonly for studies of electrocatalysis, investigations are performed in which the results represent averages obtained across a population of nanoparticles. Particularly, for heterogeneous crystal catalysis, such as that which takes place at GNPs, the activity for every individual nanoparticle is different depending on its size, shape, and composition.7−9 This can make it difficult to elucidate a detailed mechanism of the various interactions taken place that underpins the electrocatalytic mechanism. Therefore, in order to truly understand the controlling effects of such catalytic events, it is useful to study reactions occurring at individual nanoparticles.10,11 Thereby, we can eliminate average effects and better understand the factors inducing catalysis containing the shape, quantum effect, and surface functions. Technology capable of monitoring redox events at a nanoparticle will enable for the efficient development and improvement of reaction rates and efficiency.12,13 The redox state of the surface of a catalyst is a crucial component in controlling the catalytic process.14,15 However, our ability to peer into the surface change on nanoparticles via electrochemical investigations is extremely difficult due to the magnitude of electric signals generated. Consequently, there is © 2014 American Chemical Society
Received: February 27, 2014 Accepted: April 26, 2014 Published: April 28, 2014 5513
dx.doi.org/10.1021/ac500785u | Anal. Chem. 2014, 86, 5513−5518
Analytical Chemistry
Article
preparation of all the solutions. The indium tin oxide (ITO) slides were purchased from Geao Co. Ltd. (Wuhan, China). Preparation of the Samples. Gold nanoparticles used in the experiments were immobilized on the (ITO) slides. The surfaces of ITO slides were cleaned in an ultrasonic bath. The slides were first ultrasonic treated in ethanol and acetone to remove oily matter and then treated in water to remove watersoluble matter. The samples were cleaned in each solvent for more than 1 h. The cleaned ITO slides were then modified with GNPs via electrostatic adsorption by placing them in the diluted gold colloid solution (50 times) for 15 min. The GNPfunctionalized ITO slides were rinsed with water and dried under a stream of ultrapure nitrogen prior to the dark-field measurements. The nanoparticles adsorbed on ITO slides were stable during the electrochemical scanning as shown in Figure S11, Supporting Information. Dark-Field Microscopy and Scattering Spectroscopy. The dark-field measurements were carried out on an inverted microscope (eclipse Ti−U, Nikon, Japan) that was equipped with a dark-field condenser (0.8 < NA < 0.95) and a 40× objective lens (NA = 0.6). The white light source was a 100 W halogen lamp. The GNP-functionalized slides were immobilized on a platform, and the white light source was used to excite the GNPs and generate plasmon resonance scattering light. A true-color digital camera (Nikon DS-fi, Japan) was used to capture the dark-field color images. The scattering light of gold nanoparticle was split by a monochromator (Acton SP2300i, Princeton Instruments, USA) that was equipped with a grating (grating density: 300 lines/mm; blazed wavelength: 500 nm) and recorded by a spectrometer CCD (Pixis 400, Princeton Instruments, USA) to obtain the scattering spectra. The exposure time for every scattering spectrum was 10 s. Additional details of the dark-field microscopy and spectroscopy setup have been reported in a previous work.20 Electrochemistry. All electrochemical measurements were performed using a CHI 660 (Chenhua, Shanghai, China) electrochemistry station. Here, a homemade microelectrochemical cell has been developed; the ITO slide acts as the working electrode, and two Pt wires function as the counter and reference electrode, respectively. Here, the Pt wire was selected as reference electrode to eliminate the influence of Cl− ions dissociated from AgCl. All measurements were performed at 25 °C. Calibration of Pt Reference Electrode. The potential of Pt electrode was calibrated via potassiumferricyanide as shown in Figure S1, Supporting Information. The E01/2 of 1.00 mM K3[Fe(CN)6] solution using Pt quasi-reference electrode and saturated calomel electrode (SCE) was −0.127 and 0.211 V, respectively.
surface redox state which cannot be detected by simple electrochemical investigation using techniques such as cyclic voltammetry on single gold nanoparticles. We devise a novel method to monitor electrochemical redox events on single gold nanoparticles from the optical insight using a developed spectroelectrochemistry technique which combines DFM and electrochemistry (Figure 1a). We show that single GNPs
Figure 1. (a) Setup of dark-field microscopy integrated with an electrochemical workstation. (b) Scheme of electrocatalytic oxidation of H2O2 on the surface of gold nanorods in KNO3 and KCl solutions, respectively. (c−e): Simultaneous plasmonic scattering spectra peak shift of single nanorod (c) and electrochemical current of entire samples (d) under the applied triangular wave potential (e) on the GNRs.
electrochemically catalyze the oxidation of H2O2 in KNO3 and KCl solutions as shown in Figure 1b. Indium tin oxide (ITO) was modified by electrostatically adhering gold nanorods (GNR), and this acted as working electrode. Two Pt wires were immersed in the electrolyte and acted as counter electrode and quasi-reference electrode, respectively. The applied potential was calibrated via potassium ferricyanide solution (Figure S1, Supporting Information). The time dependent dark-field scattering spectra of the single GNR during the triangular wave potential scanning reveal the information about surface of a single GNR during the electrocatalytic process (Figure 1c−e). We demonstrate for the first time that the electrochemical catalytic oxidation of H2O2 at a single GNR can be investigated, and it was determined that different nanoparticles exhibited various catalytic activities. Moreover, this indicates the importance of developing techniques capable of monitoring in situ events occurring at single nanoparticles to shed new light on the mechanisms involved.
■
RESULTS AND DISCUSSION Scattering Spectra of Single Gold Nanorod. Single GNR (∼40 nm × 65 nm) exhibited spectra peak wavelength λmax of 599 nm at an open circuit corresponding to the red color as shown in Figure S3, Supporting Information. Nanorods were selected since their scattering spectra were more sensitive to the surface changes than nanospheres.35 After applying 1.00 V potential, the nanospheres only showed ca. 2 nm red shift as shown in Figure S2, Supporting Information. The scattering cross section of plasmonic nanoparticles was shown as eqs S1 and S2, Supporting Information. The Δλmax induced by the surrounding medium and electron density can be calculated from eqs 1 and 2.36−38
■
EXPERIMENTAL SECTION Materials. All reagents were of analytical grade. Absolute ethanol (AR), acetone (AR), potassium chloride (AR), and potassium nitrate (AR) were purchased from Sigma (USA). Gold nanorods (40 nm × 65 nm; 40 nm × 84 nm; 10 O.D.) were purchased from Nanoseedz (Hong Kong, China). Ultrapure water with a resistivity of 18.2 MΩ·cm was produced using a Milli-Q apparatus (Millipore, USA) and used in the 5514
dx.doi.org/10.1021/ac500785u | Anal. Chem. 2014, 86, 5513−5518
Analytical Chemistry
Article
Δλmax ≈ m(Δn)(1 − exp(− 2d /ld))
(1)
⎛1 − L ⎞ ΔN ⎟ε λ ε+⎜ ⎝ L ⎠m 2N
(2)
Δλmax = −
Information). Interestingly, after the addition of H2O2 (0.10 mM), an obvious oxidation peak at ca. 0.78 V was observed as illustrated in Figure 2c. As the concentration of H2O2 increased, the oxidation peaks increased gradually indicating that the peak current originated from the catalytic oxidation of H2O2. On the contrary, no current peak was observed on bare ITO electrode in H2O2 solution (see Figure S6, Supporting Information). To investigate the reaction process and explore the catalytic mechanism on the surface of a single GNR, the simultaneous dark-field scattering spectra of single GNR were recorded during the CV scanning which was performed at a scan rate of 10 mVs−1. Figure 2b,d (III) displays the scattering spectra Δλmax of single GNR (∼40 nm × 65 nm) obtained in the absence of H2O2. Figure 2d (I) displays the scattering spectra Δλmax of single GNR obtained in the presence of H2O2 which showed less shifts in Δλmax when compared with spectra in Figure 2d (III). In detail, for spectra obtained in the absence of H2O2, on applying a negative potential, a blue shift in the λmax of 1−2 nm occurred, which can be explained by the negative potential increasing the electron density of the GNR surface. As the potential increased from 0.00 to 0.70 V, the scattering peak was red-shifted which occurred due to double layer charging and electron discharge caused by the applied positive potential. Between the potential of 0.75 V on the forward scan and 0.75 V on the reverse scan, a plateau in the Δλmax is observed. This can be attributed to the surface of the GNR being oxidized over this potential range, and also the electron discharge. When the potential was below 0.40 V in the reverse scan, GNR was reduced gradually and the λmax blue-shifted to the initial position indicating good redox reversibility of GNRs. It could be found that before the oxidation potential, the nanorods showed an average peak shift rate of ca. 0.04 nms−1 due to the double layer charging and ions adsorption. After the oxidation potential, the nanorods showed a fast peak shift rate of ca. 0.3 nms−1 as well as the similar blue peak shift rate starting from the reduction potentials as shown in Figure S7, Supporting Information. Scheme 1 illustrates the reaction mechanism on a single GNR in the presence and absence of H2O2. During CVs performed without H2O2, the surface of GNR was oxidized into gold hydroxide and gold oxide (see eqs S3 and S4, Supporting Information).39 Gold hydroxide/oxide was represented as a compact α-oxide film on the surface of GNR (Scheme 1a).40 The compact film was assumed to be 0.23 nm in thickness and 3.3−1.35j in refractive index. The data we use originates from a second cycle of CVs, and therefore, the thickness of the hydroxide/oxide film on the gold may be more than 0.23 nm.
Here, m is the sensitivity factor of GNPs (Δλmax per refractive index unit (RIU) changing, nm/RIU), Δn is the refractive index changes of surrounding medium, d is the effective thickness of the adsorption layer, ld is the electromagnetic field decay length, εm is the dielectric constant of the surrounding environment, ε is the dielectric constant of the GNP, N is the electron density of GNPs, and L is the shape factor. As the surrounding refractive index n increased, the scattering spectra of GNPs would red shift. Additionally, the electron density change is proportional to the spectral peak shift. Thus, Δλmax of a single GNR offers information such as electron transfer and substance exchange of the redox process which is relevant to the reaction mechanism. Electrocatalytic Process of H2O2 on Single GNR in KNO3 Solution. The electrochemical behavior of GNRs deposited on ITO was investigated using cyclic voltammetry. Cyclic voltammograms (CVs) were performed at GNR modified ITO in KNO3, beginning from a start potential of −0.40 V, switching the scan at 1.00 V, and an end potential of −0.40 V (Figure 2a). At the scan rate of 0.1 V s−1, CVs
Figure 2. (a) CVs of GNRs in 0.10 M KNO3 solution, scan rate of (a1) to (a7): 30, 50, 80, 100, 300, 500, and 800 mVs−1, Pt quasireference electrode. (b) Scattering spectra of single GNR (∼40 nm × 65 nm) under the cyclic triangle wave scanning. (c) CVs on GNR modified ITO electrodes of H2O2 in 0.10 M KNO3 solution, the concentration of H2O2 from (c1) to (c7): 0.10, 0.20, 0.30, 0.50, 0.70, 0.90, and 1.00 mM; scan rate: 100 mVs−1, Pt quasi-reference electrode. (d) Scattering spectra Δλmax of two types of single GNR: 40 nm × 65 nm (I, III) and 40 nm × 84 nm (II, IV) in the presence (I, II) and absence (III, IV) of 1.00 mM H2O2 in 0.10 M KNO3 solution under the applied potential (V) from −0.10 to 1.00 V.
H 2O2 ⇔ O2 + 2H+ + 2e−
(3)
Au(OH)x + x H+ + x e− ⇔ Au + x H 2O
(4)
AuOx + 2x H+ + 2x e− ⇔ Au + x H 2O
(5)
The formation of hydroxide/oxide film induced the obvious red shift of scattering spectra λmax as a result of their high refractive index. In the presence of H2O2, GNR surface was first oxidized by applied high potential (more than 0.70 V). At the same time, GNR with hydroxide/oxide film catalyzed the oxidation of H2O2 and participated in the redox reaction. As eqs 3−5 depicted, H2O2 was oxidized into O2 and protons while gold hydroxide/oxide was reduced into gold atoms, leading to less red shift compared with the condition without H2O2. For the oxidation of H2O2, we presumed the production
exhibited an oxidation peak at 0.76 V and a reduction peak at 0.18 V. Both the anodic and cathodic peak currents were proportional to scan rate indicative of a surface controlled process (see Figure S4, Supporting Information). Control CVs were performed at bare ITO electrode (not modified with GNR), and no redox peaks were observed. This confirms that the peaks seen in Figure 2a resulted from redox events occurring at the GNRs (see Figure S5, Supporting 5515
dx.doi.org/10.1021/ac500785u | Anal. Chem. 2014, 86, 5513−5518
Analytical Chemistry
Article
and catalytic mechanism of the single GNR and provide more comprehensive and detailed information compared with electrochemical current signals which are averages of all GNRs on the surface of ITO. Electrocatalytic Process of H2O2 on Single GNR in KCl Solution. The reaction condition is also vital for the catalytic efficiency of gold nanoparticles, especially the influence of chloride ions. Utilizing DFM coupled the electrochemistry; the effect of Cl− ions on the electrocatalysis of H2O2 at single GNRs was investigated. 0.10 M KCl solution caused strong corrosion of gold on GNRs (see Figures S8−10, Supporting Information). This occurs due to the formation of the goldchloride complex, which leads to the dissolution of gold atoms.42 This was also confirmed by performing consecutive CVs in 0.10 M KCl at GNR modified ITO electrodes (Figure 3a) which showed that the oxidation peak current associated
Scheme 1. Reaction Mechanism of the Electrocatalytic Oxidation of H2O2 on Single GNR Surfacea
a (a) GNR with hydroxide/oxide film was partially reduced into Au atom, and H2O2 was oxidized into O2 after treatment of H2O2 in KNO3 solution. (b) GNR was corroded by the formation of [AuClx]−, and gold chloride blocked the catalytic activity of GNR for oxidation of H2O2 in KCl solution.
as O2 and H+ for that oxygen is the only element enabled to be oxidized. To note, the λmax was still red-shifted in the presence of H2O2 compared to the initial spectral peak at 599 nm. These results may be due to that only a proportion of the GNR surface atoms was involved in the reaction, and the positive potential induced electron discharge. In this catalysis reaction, GNR acted as a reagent and not as conducting wires or substrate as reported in previous literature examples of electrocatalytic processes.41 It is important to note that every gold nanorod exhibited various spectral shifts and catalytic activity during the redox reaction that the common electrochemistry method could not discover. As shown in Figure 2d (II, IV), the scattering spectra of single GNR (40 nm × 84 nm) during the CV scanning in the absence and presence of H2O2 showed about 14 and 6 nm red shift, respectively, which are more than GNR of 40 nm × 65 nm. These results may be due to the larger aspect ratio GNRs being more sensitive to the medium changes. Also, the selected GNR exhibited stronger catalytic ability, displaying more spectral shift between Figure 2d (II and IV). In addition, the scattering peak shifts of different single gold nanorods during the CV scanning were investigated. As shown in Figure S7, Supporting Information, the nanorods with initial scattering peak wavelength from 645 to 720 nm showed various peak red shifts after their oxidation and catalysis process. These results show that the heterogeneity in size and shape of individual GNRs caused their differing ability to electrochemically catalyze the oxidation of H2O2. Nanorods with similar scattering spectra peak wavelength may have very different results under CV scanning as shown in Figure S7, Supporting Information (λ0 = 665 nm and λ0 = 667 nm). This novel developed technique provides a tool to highlight which nanoparticles provide the highest catalytic efficiency and thus may be useful in highthroughput screening of catalysts. This method therefore eliminates the average effect of monitoring nanoparticles in the bulk system. That is, the scattering spectra reveal the redox
Figure 3. (a) Time dependent CVs of GNRs in 0.10 M KCl solutions; scan rate: 100 mVs−1, Pt quasi-reference electrode. (b) Scattering spectra of single GNR under the cyclic triangle wave scanning. (c) CVs of H2O2 on GNR modified ITO electrodes in 0.02 M KCl solutions; the concentration of H2O2 from (c1) to (c6): 0.10, 0.30, 0.50, 0.70, 0.90, and 1.00 mM; scan rate: 100 mVs−1. (d) Scattering spectra Δλmax of single GNR in the presence (I, III) and absence (II, IV) of 1.00 mM H2O2 in 0.10 M KNO3 (I, II) solution and 0.02 M KCl solution (III, IV) under the applied potential (V).
with the GNR becomes smaller with time as the gold corrosion increases. Therefore, to avoid rapid dissolution of gold atoms, 0.02 M KCl solution was used as electrolyte. As continuous CVs performing at GNR modified electrodes in the presence of H2O2, the electrochemical current was similar to the bare ITO electrode for the formation of gold chloride. From the nanoscale view, after 0.70 V in the forward scan, the scattering spectra Δλmax without H2O2 in Figure 3d (IV) red-shifted distinctly compared with Figure 3d (II). This result is explained by the formation of gold chloride. As previously explained, at lower potential (less than 600 mV vs Ag/AgCl), the gold-chloride complex mainly has two types as depicted below:42,43 Au + 2Cl(aq)− ⇔ [AuCl 2](ad)− + e− 5516
(6)
dx.doi.org/10.1021/ac500785u | Anal. Chem. 2014, 86, 5513−5518
Analytical Chemistry Au + 4Cl(aq)− ⇔ [AuCl4](ad)− + 3e−
(7)
CONCLUSIONS
■
ASSOCIATED CONTENT
■
REFERENCES
(1) Das, J.; Aziz, M. A.; Yang, H. J. Am. Chem. Soc. 2006, 128, 16022−16023. (2) Catchpole, K. R.; Polman, A. Opt. Express 2008, 16, 21793− 21800. (3) Liu, N.; Tang, M. L.; Hentschel, M.; Giessen, H.; Alivisatos, A. P. Nat. Mater. 2011, 10, 631−636. (4) Hashmi, A. S.; Hutchings, G. J. Angew. Chem., Int. Ed. 2006, 45, 7896−7936. (5) Cui, H.; Xu, Y.; Zhang, Z. F. Anal. Chem. 2004, 76, 4002−4010. (6) Guerrette, J. P.; Percival, S. J.; Zhang, B. J. Am. Chem. Soc. 2013, 135, 855−861. (7) Langille, M. R.; Personick, M. L.; Zhang, J.; Mirkin, C. A. J. Am. Chem. Soc. 2012, 134, 14542−14554. (8) Eustis, S.; El-Sayed, M. A. Chem. Soc. Rev. 2006, 35, 209−217. (9) Pyykko, P. Angew. Chem., Int. Ed. 2004, 43, 4412−4456. (10) Chen, P.; Zhou, X. C.; Andoy, N. M.; Han, K.-S.; Choudhary, E.; Zou, N. M.; Chen, G. Q.; Shen, S. Chem. Soc. Rev. 2014, 43, 1107− 1117. (11) Han, K. S.; Liu, G. K.; Zhou, X. C.; Medina, R. E.; Chen, P. Nano Lett. 2012, 12, 1253−1259. (12) Fan, F.-R. F.; Bard, A. J. Nano Lett. 2008, 8, 1746−1749. (13) Xiao, X. Y.; Fan, F.-R. F.; Zhou, J. P.; Bard, A. J. J. Am. Chem. Soc. 2008, 130, 16669−16677. (14) Lopez, N.; Nørskov, J. K. J. Am. Chem. Soc. 2002, 124, 11262− 11263. (15) Ghosh, A.; Basak, S.; Wunsch, B. H.; Kumar, R.; Stellacci, F. Angew. Chem., Int. Ed. 2011, 50, 7900−7905. (16) Wang, W.; Tao, N. Anal. Chem. 2014, 86, 2−14. (17) Brust, M.; Gordillo, G. J. J. Am. Chem. Soc. 2012, 134, 3318− 3321. (18) Shan, X.; Díez-Pérez, I.; Wang, L.; Wiktor, P.; Gu, Y.; Zhang, L.; Wang, W.; Lu, J.; Wang, S.; Gong, Q.; Li, J.; Tao, N. Nat. Nanotechnol. 2012, 7, 668−672. (19) Jing, C.; Gu, Z.; Ying, Y.-L.; Li, D.-W.; Zhang, L.; Long, Y.-T. Anal. Chem. 2012, 84, 4284−4291. (20) Zhang, L.; Li, Y.; Li, D.-W.; Jing, C.; Chen, X.; Lv, M.; Huang, Q.; Long, Y.-T.; Willner, I. Angew. Chem., Int. Ed. 2011, 50, 6789− 6792. (21) Shi, L.; Jing, C.; Ma, W.; Li, D. W.; Halls, J. E.; Marken, F.; Long, Y. T. Angew. Chem., Int. Ed. 2013, 52, 6011−6014. (22) Cheng, J.; Liu, Y.; Cheng, X. D.; He, Y.; Yeung, E. S. Anal. Chem. 2010, 82, 8744−8749. (23) Xiao, L.; Wei, L.; Cheng, X.; He, Y.; Yeung, E. S. Anal. Chem. 2011, 83, 7340−7347. (24) Li, K.; Qin, W.; Li, F.; Zhao, X.; Jiang, B.; Wang, K.; Deng, S.; Fan, C.; Li, D. Angew. Chem., Int. Ed. 2013, 52, 11542−11545. (25) Li, Y.; Jing, C.; Zhang, L.; Long, Y.-T. Chem. Soc. Rev. 2012, 41, 632. (26) Zheng, X.; Liu, Q.; Jing, C.; Li, Y.; Li, D.; Luo, W.; Wen, Y.; He, Y.; Huang, Q.; Long, Y. T.; Fan, C. Angew. Chem., Int. Ed. 2011, 50, 11994−11998. (27) Kwon, M. J.; Lee, J.; Wark, A. W.; Lee, H. J. Anal. Chem. 2012, 84, 1702−1707. (28) Hoggard, A.; Wang, L. Y.; Ma, L. L.; Fang, Y.; You, G.; Olson, J.; Liu, Z.; Chang, W. S.; Ajayan, P. M.; Link, S. ACS Nano 2013, 7, 11209−11217. (29) Solis, D., Jr.; Paul, A.; Olson, J.; Slaughter, L. S.; Swanglap, P.; Chang, W. S.; Link, S. Nano Lett. 2013, 13, 4779−4784. (30) Novo, C.; Funston, A. M.; Gooding, A. K.; Mulvaney, P. J. Am. Chem. Soc. 2009, 131, 14664−14666.
In conclusion, we monitored the electrocatalytic oxidation of H2O2 on the surface of single gold nanorod via an optical and visible method for the first time. From the plasmon resonance scattering signals, we found that bare GNRs could catalyze the oxidation of H2O2 and the gold atoms on the surface were involved in the redox reaction via the formation of gold hydroxide/oxide. Notably, from the single-particle spectra, it was confirmed that every individual nanoparticle showed various spectral shifts due to its unique sensitivity, catalytic ability, and conductivity. The approach provides effective guidance for the screening of optimal catalyst parameters. Furthermore, Cl− ions have intensive interaction with Au inducing the formation of metal-ion complex and could block the catalytic activity for H2O2. This method offers a unique way for monitoring electrochemical processes and catalysis reaction at single particle level which would improve the understanding of the mechanism of electrochemistry and catalysis of metal nanoplasmonics including electron transfer, substance exchanging, and catalyst poisoning.
S Supporting Information *
Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.
■
ACKNOWLEDGMENTS
This research was supported by 973 Program (2013CB733700), National Science Fund for Distinguished Young Scholars (21125522), and the National Natural Science Foundation of China (21327807).
Both AuCl2− and AuCl4− could be adsorbed on the surface of gold as illustrated in Scheme 1b. The formed gold chloride may be an incompact structure which coats the surface of GNR. This may expand the complex film thickness and enhance the medium refractive index greatly inducing significant red shift of the scattering spectra. Besides, after one triangle wave scanning, the scattering peak (Figure. 3d, IV) could not be shifted back to the initial position (red shift about 5 nm). The gold-chloride complex began to dissolve upon the potential of 0.70 V which caused the irregular shape of GNR. In Figure 3d (III), no obvious scattering peak changes were obtained compared with line (IV), suggesting the catalytic activity was blocked due to the fact that gold chloride could not be reduced into gold atoms by H2O2. This subsequently means hydroxide/oxide formation is critical for causing the catalytic oxidation of H2O2. Most reference electrodes such as Ag/AgCl, saturated calomel electrode, have Cl− ions inside. Thus, electrochemical experiments that include gold should not operate for a long time to reduce the effect of Cl− ions in reference electrodes. In addition, it is appropriate to use a salt bridge to slow down the diffusion of Cl− ions.
■
■
Article
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Author Contributions
C.J, D.-W.L., and Y.-T.L. designed the research; C.J., H.Z., X.S., and W.-H.L. performed research and analyzed data; C.J., F.J.R., and Y.-T.L. wrote the paper. Notes
The authors declare no competing financial interest. 5517
dx.doi.org/10.1021/ac500785u | Anal. Chem. 2014, 86, 5513−5518
Analytical Chemistry
Article
(31) Long, Y.-T.; Jing, C. Localized Surface Plasmon Resonance Based Nanobiosensors, 1st ed.; Springer: Heidelberg, 2014. (32) Dondapati, S. K.; Ludemann, M.; Müller, R.; Schwieger, S.; Schwemer, A.; Händel, B.; Kwiatkowski, D.; Djiango, M.; Runge, E.; Klar, T. A. Nano Lett. 2012, 12, 1247−1252. (33) Hill, C. M.; Pan, S. L. J. Am. Chem. Soc. 2013, 135, 17250− 17253. (34) Hill, C. M.; Clayton, D. A.; Pan, S. L. Phys. Chem. Chem. Phys. 2013, 15, 20797−20807. (35) Novo, C.; Funston, A. M.; Mulvaney, P. Nat. Nanotechnol. 2008, 3, 598−602. (36) Mie, G. Ann. Phys. 1908, 330, 377−445. (37) Novo, C.; Mulvaney, P. Nano Lett. 2007, 7, 520−524. (38) Anker, J. N.; Hall, W. P.; Lyandres, O.; Shah, N. C.; Zhao, J.; Van Duyne, R. P. Nat. Mater. 2008, 7, 442−453. (39) Li, Y.; Cox, J. T.; Zhang, B. J. Am. Chem. Soc. 2010, 132, 3047− 3054. (40) Xia, S. J.; Birss, V. I. J. Electroanal. Chem. 2001, 500, 562−573. (41) Pingarrón, J. M.; Yáñez-Sedeño, P.; González-Cortés, A. Electrochim. Acta 2008, 53, 5848−5866. (42) Dahlin, A. B.; Zahn, R.; Vörös, J. Nanoscale 2012, 4, 2339. (43) Sannomiya, T. Langmuir 2010, 26, 7619−7626.
5518
dx.doi.org/10.1021/ac500785u | Anal. Chem. 2014, 86, 5513−5518
