TECHNICAL NOTE pubs.acs.org/ac
Electrochemiluminescent Metallopolymer-Nanoparticle Composites: Nanoparticle Size Effects Anitha Devadoss,† Calum Dickinson,‡ Tia E. Keyes,† and Robert J. Forster*,† †
Biomedical Diagnostics Institute, National Centre for Sensor Research, School of Chemical Sciences, Dublin City University, Dublin 9, Ireland ‡ Materials and Surface Science Institute, University of Limerick, Limerick, Ireland ABSTRACT: Metallopolymer-gold nanocomposites have been synthesized in which the metal complex-Au nanoparticle (NP) mole ratio is systematically varied by mixing solutions of 4-(dimethylamino) pyridine protected gold nanoparticles and a [Ru(bpy)2PVP10]2þ metallopolymer; bpy is 2,20 -bipyridyl and PVP is poly-(4-vinylpyridine). The impact of changing the gold nanoparticle diameter ranging from 4.0 ( 0.5 to 12.5 ( 1 nm has been investigated. The photo induced emission of the metallopolymer undergoes static quenching by the metal nanoparticles irrespective of their size. When the volume ratio of Au NP-Ru is 1, the quenching efficiency increases from 38% to 93% on going from 4.0 ( 0.5 to 12.5 ( 1 nm diameter nanoparticles while the radius of the quenching sphere remains unaffected at 75 ( 5 Å. The conductivity of thin films is initially unaffected by nanoparticle incorporation until a percolation threshold is reached at a mole ratio of 4.95 10-2 after which the conductivity increases before reaching a maximum. For thin films of the nanocomposites on electrodes, the electrochemiluminescence intensity of the nanocomposite initially increases as nanoparticles are added before decreasing for the highest loadings. The electrochemiluminescence intensity increases with increasing nanoparticle diameter. The electrochemiluminescence (ECL) emission intensity of the nanocomposite formed using 12.5 nm particles at mole ratios between 5 10-3 and 10 10-3 is approximately 7-fold higher than that found for the parent metallopolymer. The application of these materials for low cost ECL-based point of care devices is discussed.
E
lectrochemiluminescence (ECL) is an attractive detection technique for both quantitative and qualitative analysis due to its high sensitivity and wide dynamic range. Metal complexes,1-4 such as [Ru(bpy)3]2þ, are increasingly being used as biomolecule labels and as solution phase ECL reagents.5-7 However, polymeric materials offer enhanced performance; e.g., the biomolecule “label” can contain multiple metal centers, and thin films can be formed which simplifies the assay design and minimizes reagent consumption. In particular, ruthenium-based metallopolymers of the form [M(bpy)2PVP10]2þ, where bpy is 2,20 bipyridyl and PVP is poly(4-vinylpyridine), exhibit useful ECL properties.8,9 For example, immobilization of the [Ru(bpy)2PVP10]2þ metallopolymer increases the overall efficiency of ECL emission by almost a factor of 4 in comparison with solution phase measurements.8 However, while the redox response associated with the ruthenium center is close to ideal, the rate of homogeneous charge transport is generally rather slow and it typically takes tens of seconds to electrolyze a film that is a few hundred nanometers thick.10 Therefore, in order to increase the ECL intensity by immobilizing more material on the electrode surface, the rate of charge transport must be increased if a reasonable response time is to be maintained. One attractive strategy to achieving this goal is to incorporate metallic nanoparticles that could simultaneously increase the film conductivity leading to the electrogeneration of more luminophores per unit time. For example, incorporating r 2011 American Chemical Society
4-dimethylaminopyridine (DMAP) protected gold nanoparticles in polyelectrolyte films improves the conductivity and the electron transfer characteristics of the films.11 Moreover, these nanoparticles could act as platforms for the immobilization of capture agents such as antibodies and DNA as well as perhaps enhancing the emission intensity through metal enhanced fluorescence effects. In particular, gold nanoparticles are widely used in biosensors,11 immunoassays,12 labeling biomolecules,13,14 optical imaging,15,16 and optoelectronic devices.17 The metal nanoparticles can be synthesized with various sizes and shapes by changing either the rate of reduction18 or concentration of the reducing agent.19-22 Tailoring the size and shape at the nanometer scale is important since it allows the surface plasmon to be tuned into or out of resonance with molecular luminophores.23 We have recently demonstrated that the electrochemiluminescence intensity of [Ru(bpy)2PVP10]2þ metallopolymer can be enhanced by a factor of nearly three by incorporating monodispersed 4-nm DMAP-protected gold nanoparticles because of an enhanced charge transport rate.24 It has been found that the rate of homogeneous charge transport increases only when a network of communicating nanoparticles has been formed.25,26 In this Technical Note, we report on the effect of Received: October 12, 2010 Accepted: February 1, 2011 Published: February 25, 2011 2383
dx.doi.org/10.1021/ac102697c | Anal. Chem. 2011, 83, 2383–2387
Analytical Chemistry
TECHNICAL NOTE
Figure 1. TEM images of DMAP-protected gold nanoparticles prepared with various concentrations of NaBH4 as reducing agent: (a) 0.4, (b) 0.3, (c) 0.2, and (d) 0.1 M. The concentration of the HAuCl4 was 30 mM. The inset figure shows the size distribution obtained by averaging over at least 500 individual particles.
changing the DMAP-protected gold nanoparticle diameter (4.0-12.5 nm) on the optically and electrochemically driven emission of the [Ru(bpy)2PVP10]2þ metallopolymer.
’ EXPERIMENTAL SECTION Materials. All reagents were purchased from Sigma-Aldrich and used as received. Synthesis of DMAP-Protected Gold Nanoparticles. DMAP-protected gold nanoparticles were synthesized using the phase transfer procedure reported by Gittins.27 Briefly, 30 cm3 of a 30 mM aqueous solution of HAuCl4 were added to 80 cm3 of 25 mM tetraoctylammonium bromide (TOAB) in toluene. Then, 25 cm3 of aqueous NaBH4 were added to the mixture with stirring causing an immediate reduction to occur. The concentration of NaBH4 was varied from 0.4 to 0.3, 0.2, and 0.1 M in order to achieve different sized gold nanoparticles. After 30 min, the two phases separated and the toluene phase was washed with 0.1 M H2SO4, 0.1 M NaOH, and H2O (three times) and then dried over anhydrous Na2SO4. An equal volume of 0.1 M aqueous 4-(dimethylamino) pyridine (DMAP) was then added. The phase transfer was clearly visible, as the dark pink colored solution transferred from toluene to water following addition of the DMAP, and was complete within 1 h. Synthesis of [Ru(bpy)2PVP10](ClO4)2. The metallopolymer was synthesized using a modified literature method.28 First, cis-[Ru(bpy)2Cl2] was synthesized by a standard procedure.29
Then, 60 mg (0.13 mmol) of cis-[Ru(bpy)2Cl2] was added to 39.2 mg (0.23 mmol) of AgNO3 in 10 cm3 of reagent grade methanol followed by vigorous stirring for 1 h.30 The resulting precipitate of AgCl was removed by vacuum filtration, and the filtrate, [Ru(bpy)2(H2O)Cl], evaporated to dryness. A 10-fold molar excess of poly (4-vinylpyridine) (weight average molecular weight = 160 000 g mol-1) was dissolved in 50 cm3 of 80:20 (v/v) water/ethanol and added to the filtrate. The mixture was refluxed in the dark and was monitored using cyclic voltammetry until complete. The volume was then reduced to a minimum, and NaClO4 was added to precipitate the product. The typical yield of [Ru(bpy)2PVP10] (ClO4)2 was 83%. The product was characterized using cyclic voltammetry, as well as UV-visible and emission spectroscopies. Apparatus and Procedures. Absorbance and photoluminescence spectra were recorded using a Shimadzu UV-240 spectrophotometer and a PerkinElmer LS-50 luminescence spectrometer, respectively. Transmission electron microscopy (TEM) was carried out on a JEOL JEM-2011 electron microscope operated at an accelerating voltage of 200 kV. Images were recorded using a Gatan dual Vision 600t CCD camera. The samples for nanoparticle imaging were prepared by drop-casting the colloidal solution of gold nanoparticles onto a carbon coated copper grid, and the solvent was evaporated. The particle size of approximately 500 individual nanoparticles was used to determine the particle size distribution. The zeta potential measurements were performed in water using a Malvern Zetasizer Nano 2384
dx.doi.org/10.1021/ac102697c |Anal. Chem. 2011, 83, 2383–2387
Analytical Chemistry
TECHNICAL NOTE
Table 1. Zeta Potential Values Obtained from the Electrophoretic Mobility Measurements for the DMAP-Protected Gold Nanoparticles with Various Diametersa
a
diameter of DMAP-Au nanoparticles, nm
zeta potential, mV
mobility, μm cm/Vs
conductivity, S/m
KSV, M-1
quenching efficiency, %
4
17.3 ( 0.8
1.353
0.0152
0.5 ( 0.2 10
5 7.5
19.6 ( 0.7 14.8 ( 0.5
1.540 1.160
0.0148 0.0799
0.63 ( 0.15 106 5.4 ( 0.34 106
71.8 91.2
12.5
19.3 ( 0.3
1.516
0.0251
5.53 ( 0.3 107
93
6
38
Stern-Volmer constants for luminescence quenching are also presented.
instrument and are the average of at least five independent measurements. Cyclic voltammetry was performed using a CH Instrument Model 720b electrochemical workstation using a conventional three-electrode cell. The working and counter electrodes were a glassy carbon electrode (3 mm diameter) and a platinum wire, respectively. All potentials were quoted versus an Ag/AgCl reference electrode, and all the measurements were recorded at room temperature, 25 ( 0.1 °C. For the conductivity measurements, platinum interdigitated array (IDA) electrodes were used along with a CH Instrument Model 720b electrochemical workstation. The IDA electrodes had 50 digit pairs, a digit width and interdigit gap width of 15 μm, and a digit height of 180 nm. The nanocomposite solution with the required Au nanoparticle (NP)-Ru mole ratio was drop-cast onto the IDA electrode to cover the IDA fingers and gaps between the fingers. The ECL measurements involving the simultaneous detection of light and current were performed using a CH instrument model 720b connected to an Oriel 70680 photomultiplier tube (PMT). The PMT was biased at -850 V by a high voltage power supply (Oriel, model 70705) and an amplifier/recorder (Oriel, model 70701). During the experiments, the cell was kept in a light-tight box in a specially designed holder so that the working electrode was always positioned opposite to the fiber optic bundle, the other end of which was coupled to the PMT. Glassy carbon electrodes were cleaned by successive polishing using 3 and 0.05 μm alumina slurry, followed by sonication in acetone, ethanol, and water, respectively, for 15 min. Films of the [Ru(bpy)2PVP10](ClO4)2 were obtained by drop-casting 100 μL of a 20 μM solution of the Ru metallopolymer in ethanol on the glassy carbon electrode and then evaporating to dryness in the dark overnight. Nanocomposite films were prepared by thoroughly mixing solutions of the metallopolymer and nanoparticles (4-12.5 nm diameter) at the required mole ratio. Specifically, the polymer was dissolved in a N,N-dimethylformamide (DMF)/ethanol mixture (v/v 50:50), and the required volume of the aqueous suspension of gold nanoparticles was then added to give the required mole ratio.
’ RESULTS AND DISCUSSION
4.0, 5.0, 7.5, and 12.5 nm, respectively. The insets of Figure 1 show the corresponding particle size distribution obtained by measuring over 500 individual particles. It is observed that the nanoparticle size depends approximately linearly on the mole ratio of gold salt to reducing agent. Table 1 contains the zeta potentials and electrophoretic mobilities for the DMAP-protected gold nanoparticles and shows that nanoparticles are positively charged most likely due to partial protonation of the exocyclic nitrogen.31 The DMAPprotected gold nanoparticles exhibit an intense surface plasmon resonance (SPR)32 around 524 nm. When the nanoparticle size is tuned, the overlap integral between the metallopolymer luminescence or electrochemiluminescence and the surface plasmon of the nanoparticles can be systematically varied. Photoluminescence Quenching. Mixtures of the ruthenium containing metallopolymer and DMAP-protected gold nanoparticles were dissolved in water, and their photoluminescence was measured. To prepare micromolar aqueous solutions of the metallopolymer, a concentrated solution was first prepared in a N,N-dimethylformamide (DMF)/ethanol mixture (v/v 50:50) and diluted to the required concentration such that the acetonitrile content was less than 1%. Figure 2a illustrates a typical emission spectrum and reveals that the parent metallopolymer emits at approximately 610 nm, which overlaps significantly with the plasmon of the nanoparticles (Figure 2a). DMAP-protected gold nanoparticles quench the emission of the Ru metallopolymer. Significantly, Figure 2b shows that the ratio of the emission observed in the absence, I0, and presence, I, of the nanoparticles increases nonlinearly with increasing concentration of 5.0, 7.5, and 12.5 nm nanoparticles. It is important to note that data have been corrected for the absorbance of the nanoparticles at the excitation wavelength, which depends on their size and concentration.24 Time-correlated single photon counting experiments reveal that the excited state lifetimes of ruthenium metallopolymer are unaffected by the addition of DMAP-protected gold nanoparticles. This strongly suggests that static quenching is the dominant mechanism and that it can be described using the Perrin model in which the radius of an effective quenching sphere can be calculated using eq 1: I0 ¼ expðVq Na ½Q Þ I
Gold Nanoparticle Synthesis. DMAP-protected gold nano-
particles with different core diameters were synthesized by varying the reducing agent concentration in the phase transfer (toluene/water) procedure established by Gittins.27 As the particle nucleation density depends on the mole ratio of the reducing agent, the size of the nanoparticles can be tuned by systematically varying the concentration of the reducing agent from 0.4 to 0.1 M. When the mole ratio of the reducing agent is high, nucleation proceeds rapidly and smaller nanoparticles are expected to be formed.19 Figure 1a-d shows TEM images of DMAP-protected gold nanoparticles with average core sizes of
ð1Þ
where Vq is volume of the sphere centered on the luminophore in which quenching occurs, Na is Avogadro’s constant, and [Q] is the concentration of quencher. The radius of the quenching sphere is found to be 75 ( 5 Å and is independent of the size of the quencher nanoparticles. At a 1:1 volume ratio of AuNP-Ru, the quenching efficiency of the nanoparticles increases from 38% to 93% as the nanoparticle radius is increased from 4.0 to 12.5 nm. Significantly, for the [Ru(bpy)2PVP10]2þ metallopolymer used here, the ruthenium centers are separated by approximately 30 Å 2385
dx.doi.org/10.1021/ac102697c |Anal. Chem. 2011, 83, 2383–2387
Analytical Chemistry
TECHNICAL NOTE
Figure 3. Dependence of the conductivity on AuNP-Ru mole ratio for a metallopolymer nanocomposite incorporating 4 nm DMAP-protected gold nanoparticles.
Figure 2. (a) Spectral overlap between the normalized extinction spectrum of DMAP-protected gold nanoparticles with the emission spectrum of the metallopolymer in units of quantum yield. (b) Variation of I0/I with the nanoparticle concentration in metallopolymer-nanoparticle composites developed by incorporating 5(9), 7.5 (2), and 12.5 (() nm DMAP-protected gold nanoparticles, respectively. The concentration of the metallopolymer was 20 10-6 M.
suggesting that a nanoparticle can quench more than one ruthenium center. Electrical Conductivity. The electrochemiluminescence of metallopolymers can be enhanced by adding metal nanoparticles through either metal enhanced emission effects or an increased conductivity since this will produce a greater number of luminophores per unit time. The electronic conductivity of the nanocomposite films incorporating 4 nm AuNPs were measured using IDA electrodes at different AuNP-Ru mole ratios to correlate the conductivity and ECL properties. The film conductivity was determined using eq 2: σ ¼
dG Di ATotal 3 DE
ð2Þ
where σ is the conductivity and (dG/ATotal) is Zaretsky cell constant (0.04 cm-1). ∂i/∂E was obtained from the slope of the linear plots of current versus voltage far from the Ru2þ oxidation potential. Figure 3 shows the dependence of the conductivity of the nanocomposite films with different mole ratios of AuNPRu. This figure reveals that the conductivity of the nanocomposite is independent of the nanoparticle mole ratio until approximately 5 10-2, after which the conductivity increases significantly before becoming independent of the nanoparticle loading. The conductivity for the pure metallopolymer was found to be 0.14 10-10 Ω-1 cm-1 where the conductivity increased to 8 10-10 Ω-1 cm-1 after reaching the percolation threshold (mole ratio of 4.95 10-2). This enhanced conductivity is attractive for electrochemiluminescence since it increases the rate of luminophore production. However, incorporating nanoparticles also causes quenching of the emission. Therefore, a central
Figure 4. Potential dependence of the electrochemiluminescence intensity obtained from (a) nanoparticle free metallopolymer and the metallopolymer-gold nanocomposite films developed at the optimal AuNP-Ru molar ratio by incorporating (b) 4 nm, (c) 5 nm, and (d) 7.5 nm, respectively. The inset shows the ECL response for the nanocomposite formed by incorporating 12.5 nm DMAP-protected gold nanoparticles; 0.1 M TPA was used as the coreactant.
objective is to determine which of these factors, increased conductivity or quenching, dictates the ECL properties of the nanocomposite. Electrochemiluminescence. The electrochemiluminescence emission intensity for thin films of the nanocomposites was measured as a function of the mole ratio of gold nanoparticles using voltammetry at a scan rate of 100 mV s-1 in the presence of 0.1 M TPA as the coreactant. Figure 4 shows the change in the electrochemiluminescence intensity obtained for thin films of nanocomposites at the optimal AuNP-Ru molar ratio with mean diameters of (b) 4, (c) 5, and (d) 7.5 nm as well as for the nanoparticle free metallopolymer (curve a). The inset of Figure 4 shows the ECL emission obtained for the composite based on 12.5 nm nanoparticles (dotted line). The maximum enhancement for the nanocomposite incorporating 4, 5, and 7.5 nm particles was obtained at a AuNP-Ru mole ratio of 5.4 ( 0.8 10-2, which corresponds to approximately 20 ruthenium metal centers per nanoparticle. Significantly, the electrochemiluminescence intensity of the nanocomposite incorporating 12.5 nm DMAP-protected gold nanoparticles (AuNP-Ru molar ratio of 6.1 10-3) is approximately seven times higher that the parent metallopolymer. These enhanced ECL intensities indicate that the increased rate of luminophore production due to the higher conductivity of the nanocomposites overcomes the nanoparticle quenching effects. These results confirm that the ECL intensity can be enhanced by incorporating plasmonic 2386
dx.doi.org/10.1021/ac102697c |Anal. Chem. 2011, 83, 2383–2387
Analytical Chemistry metal nanoparticles even under conditions where the nanoparticles quench the photo driven emission. Significantly, the enhancement in the ECL intensity observed for the nanocomposites is lower than the increase in conductivity. For example, the conductivity of the nanocomposite incorporating 4 nm particles increases by a factor of more than 50 when the loading exceeds the percolation threshold, but the ECL intensity increases by a factor of less than 4. This result suggests that, despite both the nanoparticles and the metallopolymer being positively charged, the separation of the nanoparticles and the ruthenium centers is not controlled by the loading but that association/aggregation occurs. This result is important since it suggests that for this system the nanoparticle-luminophore cannot be simply tuned through the loading so as to achieve metal enhanced fluorescence. Thus, while metallopolymernanoparticle composites open up the possibility of detecting analytes at lower concentrations, exploitation of both metal enhanced fluorescence and enhanced conductivity will most likely require direct functionalization of the nanoparticle surface so as to control the nanoparticle-luminophore separation. Another significant observation is that quenching assays could be developed using (electrochemi-)luminescent metallopolymers labeled with primary secondary capture antibodies whose ECL/emission is quenched by the binding of a nanoparticle labeled with secondary antibodies in the presence of the antigen.
’ CONCLUSIONS Nanocomposites developed by dispersing DMAP-protected gold nanoparticles within a ruthenium containing metallopolymer exhibit interesting size dependent emission properties. The photoluminescence emission of the nanocomposites shows an enhanced quenching efficiency with the nanoparticle size. The electronic conductivity and the electrochemiluminescence intensity is enhanced by the loading of DMAP-protected gold nanoparticles. The enhanced ECL emission has been achieved at the intermediate loading due to the enhanced quenching at higher nanoparticle loading, and the enhancement factor increases with the size of the nanoparticles.
TECHNICAL NOTE
(7) Forster, R. J.; Bertoncello, P.; Keyes, T. E. Annu. Rev. Anal. Chem. 2009, 2, 359. (8) Dennany, L.; Hogan, C. F.; Keyes, T. E.; Forster, R. J. Anal. Chem. 2006, 78, 1412. (9) Hogan, C. F.; Forster, R. J. Anal. Chim. Acta 1999, 396, 13. (10) Huang, T.; Murray, R. W. Langmuir 2002, 18, 7077. (11) Yu, A.; Liang, Z.; Cho, J.; Caruso, F. Nano Lett. 2003, 3, 1203. (12) Tian, D.; Duan, C.; Wang., W.; Li, N.; Zhang, C.; Cui, Z.; Lu, Y. Talanta 2009, 78, 399. (13) Jailswail, J. K.; Mattoussi, H.; Mauro, J. M.; Simon, S. M. Nat. Biotechnol. 2003, 21, 47. (14) Di Pasqua, A. J.; Mishler, R. E.; Ship, Y.; Dabrowiak, J. C.; Asefa, T. Mater. Lett. 2009, 63, 1876. (15) Schultz, D. A. Curr. Opin. Biotechnol. 2003, 14, 13. (16) Hashmi, A. S. K Chem. Rev. 2007, 107, 3180. (17) Niemeyer, C. M. Angew. Chem., Int. Ed. 2001, 40, 4128. (18) Iwamoto, M.; Kuroda, K.; Kanzow, J.; Hayashi, S.; Faupel, F. Adv. Powder Technol. 2005, 16, 137. (19) Hostetler, M. J. Langmuir 1998, 14, 17. (20) Rance, G. A.; Marsh, D. H.; Khlobystov, A. N. Chem. Phys. Lett. 2008, 460, 230. (21) Chen, H. M.; Liu, R.; Tsai, D. P. Cryst. Growth Des. 2009, 5, 2079. (22) Sohn, K.; Kim, F.; Pradel, K. C.; Wu, J.; Peng, Y.; Zhou, F.; Huang, J. ACS Nano 2009, 3, 2191. (23) Willets, K. A.; Van Duyne, R. P. Annu. Rev. Phys. Chem. 2007, 58, 267. (24) Devadoss, A.; Spehar-Deleze, A.; Tanner, D.; Bertoncello, P.; Marthi, R.; Keyes, T. E.; Forster, R. J. Langmuir 2010, 26, 2130. (25) Chen, Z.; Brokken-Zijp, J. C. M.; Huinink, H. P.; Loos, J.; de With, G.; Michels, M. A. J. Macromolecules 2006, 39, 6115. (26) Chen, Z.; Brokken-Zijp, J. C. M.; Michels, M. A. J. J. Polym. Sci. B: Polym. Phys. 2006, 44, 33. (27) Gittins, D. I.; Caruso, F. Angew. Chem., Int. 2001, 40, 3001. (28) Hogan, C. F.; Forster, R. J. Anal. Chim. Acta 1999, 396, 13. (29) Sullivan, B. P.; Salmon, D. J.; Meyer, T. J. Inorg. Chem. 1978, 17, 3334. (30) Pellegrin, Y.; Berg, K. E.; Blondin, G.; Anxolabehere-Mallart, E.; Leibl, W.; Aukauloo, A. Eur. J. Inorg. Chem. 2003, 1900. (31) Gandubert, V. J.; Lennox, B. Langmuir 2005, 21, 6532. (32) Jain, P. K.; Lee, K. S.; El-Sayed, I. H.; El-Sayed, M. A. J. Phys. Chem. B 2006, 110, 7238.
’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected].
’ ACKNOWLEDGMENT This material is based upon works supported by the Science Foundation Ireland under Grant No. 10/CE/B1821. The assistance of Dr. Dermot Brougham in the acquisition of the zeta potential values is gratefully acknowledged. ’ REFERENCES (1) Bertoncello, P.; Forster, R. J. Biosens. Bioelectron. 2010, 24, 3191. (2) Richter, M. Chem. Rev. 2004, 104, 3003. (3) Dennany, L.; Forster, R. J.; Rusling, J. F. J. Am. Chem. Soc. 2003, 125, 5213. (4) Dennany, L.; O’Reilly, E. J.; Keyes, T. E.; Forster, R. J. Electrochem. Commun. 2006, 8, 1588. (5) Hun, X.; Zhang, Z. Electroanalysis 2008, 20, 874. (6) Deiss, F.; LaFratta, C. N.; Symer, M.; Blichaz, T. M.; Sojic, N.; Walt, D. R. J. Am. Chem. Soc. 2009, 131, 6088. 2387
dx.doi.org/10.1021/ac102697c |Anal. Chem. 2011, 83, 2383–2387
