Anal. Chem. 2004, 76, 2010-2016
Polydisperse Composition of Mixed Monolayer-Protected, Spin-Labeled Au Nanoparticles Helen Wellsted,† Esther Sitsen,† Agneta Caragheorgheopol,‡ and Victor Chechik*,†
Department of Chemistry, University of York, Heslington, York YO10 5DD, U.K, and Romanian Academy, Institute of Physical Chemistry “I. G. Murgulescu”, Spl. Independentei 202, 77208 Bucharest, Romania
Using EPR spectroscopy and spin-labeled ligands, we have studied product distribution in the ligand exchange reactions of Au nanoparticles. When the incoming ligand was used in excess, the exchanged nanoparticles had uniform composition of the organic shell. Preparative GPC was employed to fractionate these materials according to the diameter of the Au core, as confirmed by TEM, UV, and EPR measurements. When the exchange reaction was carried out with less than a stoichiometric amount of the incoming ligand, nanoparticles with polydisperse composition of the organic shell were formed. This was established by the EPR analysis of the exchange reaction mixtures following their GPC fractionation. The GPC separation of these mixtures is controlled by both Au core size and shell composition. Monolayer-protected gold nanoparticles can be conveniently prepared by the reduction of AuCl4- in the presence of an appropriate stabilizing ligand.1 The organic shell in such nanoparticles could be further functionalized by a ligand placeexchange reaction, in which the original protecting molecules on the Au surface are partially replaced by new ligands (usually thiols).2 These two methods make synthesis of functionalized Au nanoparticles extremely versatile and simple; a wide variety of gold nanoparticles protected by a functional organic ligand3 or by a mixture of ligands with finely tuned composition and functionality4 has been reported. The inherent problem of mixed monolayer-coated nanoparticles prepared by the above techniques is the dispersity of metal core size and variable composition of the organic shell. Indeed, asprepared samples of thiol-protected nanoparticles often have a size dispersity as high as 50%.5 Nanoparticle size is known to dramati* To whom correspondence should be addressed. E-mail:
[email protected]. † University of York. ‡ Institute of Physical Chemistry “I. G. Murgulescu”. (1) Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. Chem. Commun. 1994, 801-802. (2) Hostetler, M. J.; Templeton, A. C.; Murray, R. W. Langmuir 1999, 15, 37823789. (3) Templeton, A. C.; Wuelfing W. P.; Murray, R. W. Acc. Chem. Res. 2000, 33, 27-36. (4) Ingram, R. S.; Hostetler, M. J.; Murray, R. W. J. Am. Chem. Soc. 1997, 119, 9175-9178. (5) Chen, S.; Templeton, A. C.; Murray, R. W. Langmuir 2000, 16, 3543-3548.
2010 Analytical Chemistry, Vol. 76, No. 7, April 1, 2004
cally affect physical and chemical properties, and manipulation of particle size/dispersity is therefore important.6 Some control is possible during the initial synthesis and annealing;7 aging or ligand exchange can sometimes improve monodispersity.6,8 Recently, a very mild annealing procedure for obtaining highly monodisperse nanoparticles was reported by Murray’s group.9 An alternative way of improving monodispersity is size separation, for example, by fractional crystallization10 or gel permeation chromatography (GPC). GPC is well established as a method for the size separation of macromolecules, and it has been successfully used for size separation of both gold11 and semiconductor12 nanoparticles. Thiol-protected particles have also been separated by a size exclusion mechanism using HPLC.13 The technique is sensitive enough to detect changes of only two carbon atoms in the protecting agent or less than 4 Å in core size. Dispersity in the sample can be studied using the chromatographic peak width. In addition, HPLC has been used to separate gold nanoparticles protected by mixed monolayers of hexanethiolate and mercaptoundecanoic acid. The separation was controlled by the particle size; the relative amount of hexanethiolate and mercaptoundecanoic lignads present on the nanoparticle surface in each fraction was not, however, studied.14 Dispersity in the composition of the organic shell in mixed monolayer-protected nanoparticles is difficult to characterize. Although incomplete ligand exchange on the surface of Au nanoparticles should lead to a statistical distribution of new ligands (Scheme 1), the polydispersity of such samples has never been addressed, presumably due to the lack of efficient analysis (6) Bonnemann, H.; Richards, R. M. Eur. J. Inorg. Chem. 2001, 2455-2480. (7) Jana, N. R.; Gearheart, L.; Murphy, C. J. Langmuir 2001, 17, 6782-6786. (8) Maye, M. M.; Zhong, C.-J. J. Mater. Chem. 2000, 10, 1895-1901. (9) Hicks, J. F.; Miles, D. T.; Murray, R. W. J. Am. Chem. Soc. 2002, 124, 13322-13328. (10) Whetten, R. L.; Khoury, J. T.; Alvarez, M. M.; Murthy, S.; Vezmar, I.; Wang, Z. L.; Stephens, P. W.; Cleveland, C. L.; Luedtke, W. D.; Landman, U. Adv. Mater. 1996, 8, 428-433. (11) (a) Siebrands, T.; Giersig, M.; Mulvaney, P.; Fischer, Ch.-H. Langmuir 1993, 9, 2297-2300. (b) Wei, G.-T.; Liu, F.-K.; Wang, C. R. C. Anal. Chem. 1999, 71, 2085-2091. (12) Fischer, Ch.-H.; Kenndler, E. J. Chromatogr., A 1997, 773, 179-187. (13) Wilcoxon, J. P.; Martin, J. E.; Provencio, P. Langmuir 2000, 16, 99129920. (14) Limenez, V. L.; Leopold, M. C.; Mazzitelli, C.; Jorgenson, J. W.; Murray, R. W. Anal. Chem. 2003, 75, 199-206. 10.1021/ac035465x CCC: $27.50
© 2004 American Chemical Society Published on Web 02/27/2004
Scheme 1. Statistical Ligand Distribution during Place-Exchange Reaction Leading to Dispersity in the Composition of the Nanoparticle Shell
methods.2,15 We have recently reported that the line shape of electron paramagnetic resonance (EPR) spectra of spin-labeled Au nanoparticles is strongly affected by the spin-spin interactions between the spin labels adsorbed on the same particle.16,17 As spin-spin interactions are distance-dependent,18 the spectral line shape is very sensitive to the average distance between the spin labels and consequently depends on the average number of spin labels per particle. EPR can thus be conveniently used to extract qualitative information about the spin label coverage in nanoparticles. Here, we describe an EPR study of the composition of the organic shell in mixed monolayer-coated Au nanoparticles prepared by a place-exchange reaction. We used preparative GPC to fractionate place-exchanged samples of nanoparticles; the fractions were analyzed by EPR and TEM. EXPERIMENTAL SECTION Chemicals. All chemicals were purchased from Aldrich and used as received. All solvents were HPLC grade. TEMPO-modified spin label I was prepared as described earlier.17 Triphenylphosphine-protected gold nanoparticles, Au101(PPh3)21Cl5, were prepared according to a published procedure.19 Instruments. EPR spectra were recorded on a JEOL JESRE1X instrument with 100-kHz modulation frequency. TEM images were recorded on an FEI Technai G2 instrument. The core sizes were measured for 300 randomly picked particles using ImageJ software.20 GPC was performed using a 250-mm-long column (diameter 15 mm) packed with Biobeads S-X1 gel (200400 mesh) and DCM as the mobile phase. The gel was swollen in DCM overnight prior to packing. The eluent was collected from just before the first brown color (the nanoparticles) eluted from the column; 20-30 fractions of volume 0.5 mL were collected. Preparation of Spin-Labeled Nanoparticles. Spin-labeled nanoparticles were prepared via place exchange of TEMPOmodified disulfide I with triphenylphosphine-protected nanoparticles. Triphenylphosphine-protected Au nanoparticles (3 mg, 1.18 × 10-7 mol) were dissolved in DCM (0.5 g) and mixed with the solution of an appropriate amount of disulfide I in DCM (0.5 g). The formula Au101(PPh3)21Cl5 was used for calculating the stoichiometric amounts.19 The reaction mixture was left at room (15) Templeton, A. C.; Cliffel D. E. J. Am. Chem. Soc. 1999, 121, 7081-7089. (16) Ionita, P.; Caragheorgheopol, A.; Gilbert, B. C.; Chechik, V. J. Am. Chem. Soc. 2002, 124, 9048-9049. (17) Chechik, V.; Wellsted, H. J.; Korte, A.; Gilbert, B. C.; Caldararu, H.; Ionita, P.; Cagheorgheopol, A. Faraday Discuss. 2004, 125, 279-297. (18) Berliner, L. J., Eaton, S. S., Eaton, G. R., Eds. Distance Measurements in Biological Systems by EPR; Kluwer Academic/Plenum Publishers: New York, 2001. (19) Weare, W. W.; Reed, S. M.; Warner, M. G.; Hutchinson, J. E. J. Am. Chem. Soc. 2000, 122, 12890-12891. (20) Public domain software available for download at http://rsb.info.nih.gov/ ij/.
Scheme 2. Place-Exchange Reaction of Disulfide I with Triphenylphosphine-Protected Au Nanoparticles
temperature for 18 h and separated by GPC. The fractions were evaporated to dryness with a stream of nitrogen and redissolved in DCM (0.2 mL) before recording the EPR spectra. Analysis of EPR Spectra. Integration of the spectra was carried out using Bruker’s WIN-EPR software, with baseline correction obtained by fitting to a sixth-order polynomial. The amount of radical in the sample was then quantified by comparison of the second integral of its spectrum with that of a standard solution of disulfide I. To separate the contribution of broad line and triplet in the complex EPR spectra, the line shape was simulated using a standard curve-fitting procedure available within SPSS software. We assumed that the lines in the triplet component have purely Lorentzian shape and have equal double integral intensity. The broad line component was simulated as a mixture of a Lorentzian and a Gaussian line. Rotational correlation times τc were calculated from the triplet component of the EPR spectra using eq 1.21
τc(s) ) 6.51 × 10-10∆H(0)
(x
I(0) + I(+1)
x
I(0) -2 I(-1)
)
(1)
Here ∆H(0) is the peak-to-peak line width (gauss) of the central line and I(+1), I(0), and I(-1) are the peak-to-peak intensities of the low-, middle-, and high-field lines of the 14N hyperfine components, respectively. Au Analysis. The nanoparticles were digested by aqua regia (2 mL). After complete destruction of the particles, the mixture was rotary evaporated to dryness. The residue was then dissolved in water (1 mL) followed by addition of 1% aqueous potassium iodide (2 mL), and titrated with 1 mM aqueous sodium thiosulfate.22 RESULTS AND DISCUSSION Preparation of Mixed Monolayer-Protected, Spin-Labeled Au Nanoparticles. Spin-labeled nanoparticles were synthesized by a place-exchange reaction of TEMPO-modified disulfide I with triphenylphosphine-protected Au particles (Scheme 2). Although disulfides are much less reactive in exchange reactions than thiols, we recently showed that they could replace triphenylphosphine ligands from the Au surface.17 The extent of the exchange reaction depends on the stoichiometry and the concentration of the starting (21) Likhtenshtein, G. L. Spin Labeling Methods in Molecular Biology; Wiley: New York, 1976. (22) Beamish, F. E. The analytical chemistry of the noble metals; Pergamon Press: New York, 1966.
Analytical Chemistry, Vol. 76, No. 7, April 1, 2004
2011
Figure 1. EPR spectra of spin-labeled nanoparticles prepared by a place-exchange reaction of disulfide I with triphenylphosphineprotected Au nanoparticles using 1:1 (a) and 80:1 (b) spin label per particle stoichiometry.
materials. At low spin label-to-nanoparticle ratios (5 nm, it appears as shoulder in UV spectra of particles of 2-3 nm in diameter and completely disappears from the spectra for particles of
