Facile Preparative Route to Alkanethiolate-Coated Au38 Nanoparticles

Langmuir 2007, 23, 7853-7858

7853

Facile Preparative Route to Alkanethiolate-Coated Au38 Nanoparticles: Postsynthesis Core Size Evolution Junhyung Kim, Kesta Lema,† Macmillan Ukaigwe,‡ and Dongil Lee* Department of Chemistry, Western Michigan UniVersity, Kalamazoo, Michigan 49008-5413 ReceiVed March 14, 2007. In Final Form: April 17, 2007 A facile preparative route to alkanethiolate (Cn) Au38 nanoparticles, where n ) 4, 6, 8, 10, and 12, is described. Subnanometer-sized nanoparticles are initially produced by a modified Brust synthesis, which undergo core-size evolution upon removal of reaction impurities that have served as additional protecting layers. C4-C12 Au38 nanoparticles are prepared in ∼300 mg quantities by the selective removal of reaction impurities with dimethyl sulfoxide. The prepared nanoparticles are 1.1-1.2 nm in core size, and all exhibit optical and electrochemical characteristics of Au38 nanoparticles. Voltammetry of these Au38 nanoparticles reveals that the energy gap between the first one-electron oxidation and the first reduction is rather insensitive to the ligand employed. By contrast, the energy gaps between the first and second oxidations and between the second and third oxidations are ligand-dependent; both substantially increase with ligand thickness. The charging energetics of alkanethiolate-coated Au38 nanoparticles can thus be described as a sum of electron addition energies and the discrete electronic energy levels of the Au38 core.

Introduction The advent of improved synthetic methods for metal nanoparticles has generated a widespread research effort on their optical, electronic, and chemical properties.1-8 Monolayerprotected metal clusters (MPCs) containing a few to a few hundred core atoms are of particular interest because they represent the bridge between bulk and molecular behavior. The core-sizedependent electrochemical and optical properties of these smallcore MPCs have been described.9-15 A significant challenge in this research is the development of synthetic methods capable of producing monodisperse nanoparticles in useful quantities with controlled size and shape for many applications in biological detection4-6 and catalysis.16-20 * Corresponding author. E-mail: [email protected]. † Present address: Kellogg Community College, Battle Creek, Michigan. ‡ Present address: Department of Chemistry, Chicago State University, Chicago, Illinois. (1) Hayat, M. A. In Colloidal Gold: Principles, Methods, and Applications; Hayat, M. A., Ed.; Academic Press: San Diego, CA, 1991. (2) Sampaio, J. F.; Beverly, K. C.; Heath, J. R. J. Phys. Chem. B 2001, 105, 8797. (3) Feldheim, D. L.; Keating, C. D. Chem. Soc. ReV. 1998, 27, 1. (4) Zanchet, D.; Micheel, C. M.; Parak, W. J.; Gerion, D.; Alivisatos, A. P. Nano Lett. 2001, 1, 32. (5) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin, C. A. Science 1997, 277, 1078. (6) Nam, J.-M.; Thaxon, C. S.; Mirkin, C. A. Science 2003, 301, 1884. (7) Templeton, A. C.; Wuelfing, W. P.; Murray, R. W. Acc. Chem. Res. 2000, 33, 27. (8) Wood, A.; Giersig, M.; Mulvaney, P. J. Phys. Chem. B 2001, 105, 8810. (9) Chen, S.; Ingram, R. S.; Hostetler, M. J.; Pietron, J. J.; Murray, R. W.; Schaaff, T. G.; Khoury, J. T.; Alvarez, M. M.; Whetten, R. L. Science 1998, 280, 2098. (10) Link, S.; Beeby, A.; FitzGerald, S.; El-Sayed, M. A.; Schaaff, T. G.; Whetten, R. L. J. Phys. Chem. B 2002, 106, 3410. (11) Yang, Y.; Chen, S. Nano Lett. 2003, 3, 75. (12) Lee, D.; Donkers, R. L.; Harper, A. S.; Wang, G.; Murray, R. W. J. Am. Chem. Soc. 2004, 126, 6193. (13) Balasubramanian, R.; Guo, R.; Mills, A. J.; Murray, R. W. J. Am. Chem. Soc. 2005, 127, 8126. (14) Menard, L. D.; Gao, S.-P.; Xu, H.; Twesten, R. D.; Harper, A. S.; Song, Y.; Wang, G.; Douglas, A. D.; Yang, J. C.; Frenkel, A. I.; Nuzzo, R. G.; Murray, R. W. J. Phys. Chem. B 2006, 110, 12874. (15) Woehrle, G. H.; Warner, M. G.; Hutchison, J. E. J. Phys. Chem. B 2002, 106, 9979. (16) Ahmadi, T. S.; Wang, Z. L.; Green, T. C.; Henglein, A.; El-Sayed, M. A. Science 1996, 272, 1924. (17) Burda, C.; Chen, X.; Narayanan, R.; El-Sayed, M. A. Chem. ReV. 2005, 105, 1025.

The Brust synthesis21 has been a convenient method for preparing small-core gold nanoparticles coated with thiolate ligands. The synthesis is a two-step process that typically leads to polydisperse (in core size) Au MPCs with an average core diameter of 1-5 nm:

AuCl4-(toluene) + RSH f (-AuISR-)n (polymer) (1) (-AuISR-)n + BH4- f Aux(SR)y

(2)

The behavior of reactions 1 and 2 is consistent with a nucleation-growth-passivation process;7,22,23 namely, high thiol/ gold mole ratios, chilling the reaction, and quenching reaction 2 at short times generally produce higher proportions of MPCs with very small core sizes. For example, by conducting the Brust reaction at very low temperature (-78 °C) to suppress the core nucleation-growth process, Jimenez et al.24 synthesized hexanethiolate Au38 nanoparticles. They also prepared the Au38 nanoparticles by employing a hyperexcess (300:1 mole ratio) of thiol to gold atoms. A more facile preparative route to Au38 nanoparticles was reported by Donkers and co-workers.25 Employing phenylethanethiolate (PhC2) as a protecting ligand in the Brust reaction (3:1 thiol/gold mole ratio, 0 °C), they produced small-core MPCs containing a higher proportion of Au38 nanoparticles. The Au38 nanoparticles were then isolated from the larger nanoparticles using solvent fractionation. The acetonitrile solubility of the PhC2Au38 nanoparticle was a key factor allowing its dissolution from the larger reaction product mass and isolation in quite monodisperse form. (18) Haruta, M. Gold Bull. 2004, 37, 27. (19) Valden, M.; Lai, X.; Goodman, D. W. Science 1998, 281, 1647. (20) Hughes, M. D.; Xu, Y.-J.; Jenkins, P.; McMorn, P.; Landon, P.; Enache, D.; Carley, A. F.; Attard, G. A.; Hutchings, G. J.; King, F.; Stitt, E. H.; Johnston, P.; Griffin, K.; Kiely, C. J. Nature 2005, 437, 1132. (21) Brust, M.; Walker, M.; Bethell, D.; Schriffrin, D. J.; Whyman, R. J. Chem. Soc., Chem. Commun. 1994, 801. (22) Chen, S.; Templeton, A. C.; Murray, R. W. Langmuir 2000, 16, 3543. (23) Hostetler, M. J.; Wingate, J. E.; Zhong, C.-J.; Harris, J. E.; Vachet, R. W.; Clark, M. R.; Londono, J. D.; Green, S. J.; Stokes, J. J.; Wignall, G. D.; Glish, G. L.; Porter, M. D.; Evans, N. D.; Murray, R. W. Langmuir 1998, 14, 17. (24) Jimenez, V. L.; Georganopoulou, D. G.; White, R. J.; Harper, A. S.; Mills, A. J.; Lee, D.; Murray, R. W. Langmuir 2004, 20, 6864. (25) Donkers, R. L.; Lee, D.; Murray, R. W. Langmuir 2004, 20, 1945.

10.1021/la700753u CCC: $37.00 © 2007 American Chemical Society Published on Web 06/02/2007

7854 Langmuir, Vol. 23, No. 14, 2007

However, the isolation procedure for the PhC2-Au38 nanoparticles failed for hexanethiolate,24 and the preparation of alkanethiolate Au38 nanoparticles in useful quantities remains inaccessible. In this article, we describe a facile preparative route to alkanethiolate (Cn) Au38 nanoparticles, where n ) 4, 6, 8, 10, and 12. We report that the core size is additionally determined by the purification procedure of the synthesized MPC product. C4-C12 Au38 MPCs are prepared in ∼300 mg quantities by selective cleaning of reaction impurities that would otherwise serve as additional protecting layers. The prepared C4-C12 MPCs are 1.1-1.2 nm in core diameter and exhibit the optical and electrochemical characteristics of Au38. The improved preparative route to C4-C12 Au38 MPCs has allowed us to investigate the ligand-thickness dependence of the Au38 charging energetics. The voltammetry of C4-C12 Au38 MPCs reveals that the charging energetics can be described as a sum of electron addition energies26,27 and the discrete electronic energy levels of the Au38 core. Experimental Section Chemicals. 1-Butanethiol (99%), 1-hexanethiol (98%), 1-octanethiol (98.5%), 1-decanethiol (98%), 1-dodecanethiol (98%), tetran-octylammonium bromide (Oct4NBr, 98%), sodium borohydride (NaBH4, 99%), and hydrogen tetrachloroaurate trihydrate (HAuCl4‚ 3H2O, ACS reagent grade) were used as received from Aldrich. HPLC-grade toluene and acetone, reagent-grade ethanol, acetonitrile, dichloromethane, and dimethyl sulfoxide (DMSO) were purchased from Aldrich and used as received. Water was purified using a Millipore Milli-Q system (18.2 MΩ·cm). Measurements. Transmission electron microscopy (TEM) images of MPC samples were obtained with a JEOL transmission electron microscope (JEM-1230). MPC samples were prepared by quickly dipping/withdrawing a Formvar/carbon-coated copper grid (01814F, Ted Pella) in 1 mg/mL MPC in CH2Cl2 and drying in air for at least 1 h before imaging. Three typical regions of each sample were imaged at 500 000× or 600 000× magnification. Core-size histograms were read from digitized photographic images using Scion Image Alpha 4.0.3.2 (www.scioncorp.com). Values were confirmed by manual reading. Optical absorbance spectra were collected with a Perkin-Elmer spectrometer (Lambda 40). Mass spectra of extracted reaction impurities were obtained using positive/negative ion electrospray mass spectrometry on a Q-TOF Ultima (Waters) or a LCMS-2010EV (Shimadzu) HPLC/mass spectrometer. Voltammetry was conducted with an electrochemical workstation (CHI 660B) in 0.1 M Bu4NClO4 solutions that were degassed and blanketed with a high-purity Ar atmosphere during the experimental procedure. The working electrode was a 0.4-mm-diameter Pt disk. The working electrode was polished with 0.05-µm Al2O3 slurries and cleaned electrochemically by potential cycling in 0.1 M H2SO4 solution.28 A Pt wire counter electrode and a Ag wire quasi-reference electrode (AgQRE) were used. Sublimed ferrocene (Fc) was added as an internal reference for AgQRE. In this paper, potentials are reported versus AgQRE or Fc/Fc+. Reduced temperature voltammetry was carried out using cold baths of acetone/dry ice (-78 °C). Au38 MPC Synthesis. Alkanethiolate (Cn) Au38 nanoparticles, where n ) 4, 6, 8, 10, and 12, were prepared using a modified Brust two-phase procedure in which the NaBH4 reduction reaction was carried out with a 5:1 thiol/gold molar ratio at 0°C before being quenched at 30 min. Briefly, for the preparation of C6-MPCs, to a vigorously stirred solution of 3.60 g (6.58 mmol) of Oct4NBr in 160 mL of toluene was added 2.00 g (5.08mmol) of HAuCl4‚3H2O in 40 mL of Milli-Q water. The water phase was quickly cleared, and the toluene solution became dark red as AuCl4- was transferred into it. The aqueous phase was removed, and 3.6 mL (26mmol) of (26) Franceschetti, A.; Zunger, A. Phys. ReV. B 2000, 62, 2614. (27) Franceschetti, A.; Zunger, A. Appl. Phys. Lett. 2000, 76, 1731. (28) Woods, R. In Electroanalytical Chemistry; Bard, A. J., Ed.; Marcel Dekker: New York, 1980; Vol. 9, p 1.

Kim et al. 1-hexanethiol was added and stirred until the solution became colorless, indicating AuI-thiol polymer formation (reaction 1). The toluene solution was cooled to 0 °C in the ice bath, and 1.94 g (51.3mmol) of NaBH4 that had been dissolved in 40 mL of Milli-Q water and cooled to 0 °C was quickly added to the toluene solution with vigorous stirring. The solution mixture immediately became black, indicative of MPC formation. Stirring was continued for 30 min at 0 °C. After removing the aqueous phase, the black organic phase was collected and washed four times with 200 mL of Milli-Q water and rotary evaporated to produce a black product. (as-prepared MPC). DMSO (200 mL) was slowly added to the as-prepared MPC and allowed to stand overnight. The black product, which was mostly insoluble in DMSO, was then collected and transferred to a flask. Acetone (30 mL) was added to the flask to extract Au38 MPCs. The acetone solution was rotary evaporated and subsequently washed with copious amounts of acetonitrile and ethanol, providing 250300 mg of a dark-brown product (C6-Au38 MPCs, I).

Results and Discussion MPC Synthesis. In the present work, we modified the Brust procedure21 to increase the yield of smaller-core MPCs; namely, reaction 1 was carried out with a 5:1 thiol/gold molar ratio to enhance passivation, and reaction 2 was carried out at 0 °C and quenched at 30 min to suppress particle growth. These synthesis conditions led to a large fraction of ethanol-soluble MPCs containing two dominant cores, 1.2 and 1.7 nm, which are consistent with the core sizes of previously reported Au38 and Au140 MPCs, respectively.12,24,29,30 The ethanol solubility of these small-core MPCs seems to be related to an elevated polarity of smaller-core Au MPCs; larger core sizes are soluble only in nonpolar solvents. Because both MPCs were soluble in ethanol,31 other solvents were investigated to separate these two cores. For PhC2-coated MPCs, the acetonitrile solubility of the PhC2-Au38 MPCs was the enabling factor in isolating them from the larger portion of Au140 MPCs.25 However, the isolation procedure for the PhC2-Au38 MPCs failed for hexanethiolate.24 Neither C6Au38 nor C6-Au140 MPCs were soluble in acetonitrile. This is not surprising considering the fact that the protecting ligand has a strong influence on the solubility properties, as evidenced by the fact that we often observe dramatic solubility changes resulting from ligand place-exchange reactions.32,33 Among other solvents examined for the isolation of C6-Au38 MPCs, we have identified DMSO as a key cleaning solvent for the preparation of Au38 MPC. That is, the C6-Au38 MPC could be isolated by washing the crude, as-prepared MPC product with DMSO and extracting it into acetone. The acetone solution was then rotary evaporated and subsequently washed with acetonitrile and ethanol to give a dark-brown product (I). However, washing the same as-prepared MPC product initially with acetonitrile instead of DMSO produced a rather polydisperse Au140-like MPCs product (II), as discussed next. The 1H NMR spectrum of I (Figure S1, Supporting Information) confirms the absence of most reaction impurities, although it shows the presence of Oct4N+ (ca. one Oct4N+ per MPC, 3.25 ppm) in the ligand shell. This is typical for MPCs synthesized by the Brust procedure.25,34 Isolated MPC products I and II were characterized by voltammetry, absorbance spectroscopy, and TEM. Electrochemi(29) Hicks, J. F.; Templeton, A. C.; Chen, S.; Sheran, K. M.; Jasti, R.; Murray, R. W.; Debord, J.; Schaaff, T. G.; Whetten, R. L. Anal. Chem. 1999, 71, 3703. (30) Hicks, J. F.; Miles, D. T.; Murray, R. W. J. Am. Chem. Soc. 2002, 124, 13322. (31) We suspect that the presence of reaction impurities may boost ethanol solubility. The purified Au38 and Au140 MPCs exhibit much lower solubility in ethanol. (32) Hostetler, M. J.; Templeton, A. C.; Murray, R. W. Langmuir 1999, 15, 3782. (33) Song, Y.; Murray, R. W. J. Am. Chem. Soc. 2002, 124, 7096. (34) Waters, C. A.; Mills, A. J.; Johnson, K. A.; Schiffrin, D. J. Chem. Commun. 2003, 540.

Alkanethiolate-Coated Au38 Nanoparticles

Figure 1. Differential pulse voltammograms (DPVs) at 25 °C and 0.02 V/s of (a) I and (b) II in 0.1 M Bu4NClO4 in CH2Cl2 and (c) the as-prepared MPC product in 0.1 M Bu4NClO4 in tetrahydrofuran at a 0.4-mm-diameter Pt working electrode, Ag wire quasi-reference electrode (AgQRE), and Pt wire counter electrode.

cal measurements provide a powerful means for exploring the size-dependent electronic charging properties and the electronic structure of small-core MPCs near the Fermi level, for example, the opening of an electrochemical energy gap between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO).9,11-13 The differential pulse voltammetry (DPV) of I (Figure 1a) exhibits the characteristics of Au38 nanoparticles;12,24 the pattern of current peaks lying at the formal potentials of the nanoparticle charge state couples consists of an electrochemical HOMO-LUMO gap, as measured using the first oxidation (ox1) and reduction (re1) peaks and unevenly spaced second (ox2) and third (ox3) steps. This voltammetric pattern reflects the discretization and spacing of electronic levels of the Au38 core, confirming the isolation of Au38 MPCs. By contrast, the DPV peaks of II in Figure 1b display a distinctly different pattern; peaks are much less resolved and roughly evenspaced, characterizing it as a larger Au140-like MPC.29,30 The toluene solution absorbance spectrum (black line, Figure 2) of I also displays the characteristic steplike structure of Au38 MPCs, indicative of moleculelike properties and transitions between discrete electronic levels.12,24,35 The spectrum of II, however, displays (blue dash, Figure 2) a steep decay absorbance from low to higher wavelength. This is typical for Au MPCs with a core of 1.6-2.0 nm.23 The TEM images in Figure 3a,b parallel the voltammetry and absorbance results. That is, the average core size of I is 1.2 nm, consistent with the core diameter reported12,24 for Au38 MPCs, whereas that of II (Figure 3b) is substantially larger (1.7 ( 0.3 nm). These analytical results clearly indicate that the isolated MPC I and II are distinctly different in core size even though they are isolated from the same synthetic product (as-prepared MPC). What caused the size difference between I and II? Core-Size Evolution during Purification. To investigate the factor(s) determining the product size, we have characterized the (35) Schaaff, T. G.; Shafigullin, M. N.; Khoury, J. T.; Vezmar, I.; Whetten, R. L.; Cullen, W. G.; First, P. N.; Gutie´rrez-Wing, C.; Ascensio, J.; Jose-Yacama´n, M. J. J. Phys. Chem. B 1997, 101, 7885.

Langmuir, Vol. 23, No. 14, 2007 7855

Figure 2. Absorbance spectra of I (black line), II (blue dash), and the as-prepared MPC product (red line) in toluene. All spectra have been normalized at 300 nm and offset for clarity.

MPC core size and properties at each stage of isolation. Figure 3c shows a representative TEM image of the as-prepared MPC resulting from the modified Brust synthesis as described in the Experimental Section. The gold particles in Figure 3c are relatively monodisperse, and the average core size is 0.8 nm, which is distinctly smaller than that of Au38 or Au140. The DPV of the as-prepared MPC in Figure 1c also displays a qualitatively different charging pattern that distinguishes itself from Au38 or Au140; there is a fairly reversible oxidation peak at 0.7 V and a less-resolved reduction peak at ca. -1.5 V comprising an electrochemical HOMO-LUMO gap of 2.2 V. This charging pattern is reminiscent of small-core MPCs previously reported by Yang et al.11 and Menard et al.14 for respectively Au11 and Au13 clusters. Because the as-prepared MPC sample is unpurified, we were unable to analyze it further. Attempts made to purify the as-prepared MPC resulted only in rather polydisperse, larger MPCs (vide infra). The absorbance spectrum of the as-prepared MPC in toluene (red line, Figure 2) is distinctly different from that of Au38 but rather similar to that of Au140. This is somewhat surprising because one would expect to observe more distinct molecular features for the subnanometer-sized MPCs. We note, however, that the absorbance characteristics are also sensitive to the media surrounding the metal cores. In the ligand-place exchange reactions of subnanometer, thiol-stabilized Au11 nanoparticles, Woehrle and co-workers reported15 that welldefined optical transitions became featureless for MPC products exchanged with charged ligands, presumably because of some kind of electrostatic agglomeration of MPCs. A similar effect could be seen in the unpurified MPC solution. The analytical evidence described above suggests that the modified Brust method initially produces subnanometer-sized MPCs that turn into Au38 or Au140 MPCs during the purification step, depending on the cleaning procedure used. This implies that the core size is determined not only by the synthetic condition used but also by the purification procedure following the synthesis step. Before the cleaning step, there may be some reaction impurities such as Oct4N+, excess thiol, and disulfide in the ligand shell that could play an important role in preventing core aggregation. As revealed for Au38 MPCs,12 the ligand shell of small-core MPCs is somewhat open to intrusion by other species. The penetration of reaction impurities into the ligand shell is thus highly likely considering the sharp curvature of the very

7856 Langmuir, Vol. 23, No. 14, 2007

Kim et al. Scheme 1. MPC Core-Size Evolution during Purification

Figure 3. TEM images and corresponding histograms of the core diameters of (a) I, (b) II, and (c) the as-prepared MPC. Scale bar ) 50 nm.

small core and strong hydrophobic interactions between alkanethiolate ligands and these impurities. When the impurities are removed from the ligand shell, the subnanometer-sized MPCs are prone to aggregate to form more stable MPCs. It is noteworthy that the core of the subnanometer-sized MPC remains unchanged during repeated dissolving/drying cycles as long as these reaction impurities are present, supporting their core-protecting role. We conclude therefore that the difference in the product size was caused by the difference in the purification procedure; that is, DMSO cleaning predominantly led to the formation of Au38 MPCs whereas acetonitrile cleaning resulted in Au140-like MPCs. Positive/negative ion electrospray mass spectrometry was employed to identify the extracted impurities in each solvent. Although the fragmentation patterns of acetonitrile extracts are very complicated because of the presence of various types of impurities (Figure S2a and b, Supporting Information), the mass peaks clearly show the presence of C6H13S (m/z 117), unreacted gold-thiol complexes Au(C6H13SH) (m/z 315) and Au(C6H13S)2 (m/z 431), and Oct4N+ (m/z 466). By contrast, only Oct4N+ (m/z 466) is predominantly observed in Figure S2c (Supporting Information) from DMSO extracts. These results indicate that acetonitrile effectively breaks up the ligand shell and removes reaction impurities from it, whereas highly polar DMSO can remove only some of the charged species from the ligand shell.

Gravimetric analyses of the extracts additionally revealed that the mass of the acetonitrile extracts (7.76 mg/mL) was more than 30-fold larger than that of the DMSO extracts (0.21 mg/mL), confirming the vastly different cleaning power between these two solvents. From these observations, we propose a core-size evolution mechanism as illustrated in Scheme 1. The as-prepared MPCs (core diameter