Probing the Relative Stability of Thiolate- and Dithiolate-Protected Au

Publication Date (Web): August 3, 2009. Copyright © 2009 American Chemical Society. *E-mail: [email protected]. Voice: 306-966-2017...
0 downloads 0 Views 2MB Size
Download PDF
pubs.acs.org/Langmuir © 2009 American Chemical Society

Probing the Relative Stability of Thiolate- and Dithiolate-Protected Au Monolayer-Protected Clusters Wenbo Hou, Mita Dasog, and Robert W. J. Scott* Department of Chemistry, University of Saskatchewan, 110 Science Place, Saskatoon, Saskatchewan, Canada Received May 20, 2009. Revised Manuscript Received July 13, 2009 Dithiolate ligands based on (()-R-lipoic acid derivatives have been investigated as ligands for both Au monolayerprotected clusters (MPCs) and mixed alkanethiol/dithiolate Au MPCs. The oxidative and thermal stability of the MPCs were investigated by a combination of UV-vis spectroscopy, TEM, and 1H NMR experiments. Results show that the dithiolate-protected MPCs are much more prone to oxidation by oxygen under ambient conditions than their alkanethiolate-protected MPC analogues; in addition, the Au core of the dithiolate-protected Au MPCs could be etched by KCN at much faster rates than both alkanethiolate-protected and mixed monolayer MPCs. These results suggest that strategies to increase ligand-metal interactions by incorporating more thiolate linkers into the ligand must also take into account the packing efficiency and/or stability of such ligands on the metal surface, which can make them much more prone to oxidation under ambient conditions.

Introduction Alkanethiol and functionalized thiol monolayers are the building blocks for both two-dimensional self-assembled monolayers (SAMs) on gold1,2 and three-dimensional monolayer-protected clusters (MPCs) of gold.3,4 Recently, complete structure solutions of Au25 and Au102 MPCs have been realized by single-crystal X-ray diffraction, which has shed light on the “staple” bonding motif between thiolate ligands and Au cluster surfaces.5-7 Nearly all applications of SAMs on Au and Au MPCs rely on the chemical inertness of the Au-thiol bond; however, a number of groups have noted that gold-thiolate bonds are prone to oxidation,8-12 ligand exchange,3,13,14 reductive desorption,15 and thermal desorption.13,16 SAMs and Au MPCs are prone to oxidation in the presence of ozone;8-10 in addition, earlier studies in our group,11 and by others,12 indicated that, while dodecanethiolateprotected gold MPCs are stable in air as pure solutions, they are *E-mail: [email protected]. Voice: 306-966-2017. Fax: 306-966-4730. (1) Laibinis, P. E.; Whitesides, G. M.; Allara, D. L.; Tao, Y. -T.; Parikh, A. N.; Nuzzo, R. G. J. Am. Chem. Soc. 1991, 113, 7152–7167. (2) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. Rev. 2005, 105, 1103–1169. (3) Templeton, A. C.; Wuelfing, W. P.; Murray, R. W. Acc. Chem. Res. 2000, 33, 27–36. (4) Daniel, M.-C.; Astruc, D. Chem. Rev. 2004, 104, 293–346. (5) Jadzinsky, P. D.; Calero, G.; Ackerson, C. J.; Bushnell, D. A.; Kornberg, R. D. Science 2007, 318, 430–433. (6) Walter, M.; Akola, J.; Lopez-Acevedo, O.; Jadzinsky, P. D.; Calero, G.; Ackerson, C. J.; Whetten, R. L.; Gr€onbeck, H.; H€akkinen, H. Proc. Nat. Acad. Sci. U.S.A. 2008, 105, 9157–9162. (7) Heaven, M. W.; Dass, A.; White, P. S.; Holt, K. M.; Murray, R. W. J. Am. Chem. Soc. 2008, 130, 3754–3755. (8) Schoenfisch, M. H.; Pemberton, J. E. J. Am. Chem. Soc. 1998, 120, 4502– 4513. (9) Lee, M.-T.; Hsueh, C.-C.; Freund, M. S.; Ferguson, G. S. Langmuir 1998, 14, 6419–6423. (10) Sandhyarani, N.; Pradeep, T. Chem. Phys. Lett. 2001, 338, 33–36. (11) Dasog, M.; Scott, R. W. J. Langmuir 2007, 23, 3381–3387. (12) Kanehara, M.; Sakurai, J.; Sugimura, H.; Teranishi, T. J. Am. Chem. Soc. 2009, 131, 1630–1631. (13) Schlenoff, J. B.; Li, M.; Ly, H. J. Am. Chem. Soc. 1995, 117, 12528–12536. (14) Hostetler, M. J.; Templeton, A. C.; Murray, R. W. Langmuir 1999, 15, 3782–3789. (15) Widrig, C. A.; Chung, C.; Porter, M. D. J. Electroanal. Chem. 1991, 310, 335–359. (16) Delemarche, E.; Michel, B.; Kang, H.; Gerber, C. Langmuir 1994, 10, 4103– 4108.

12954 DOI: 10.1021/la9018053

readily oxidized in the presence of halide salts. One strategy used in the literature to enhance gold-thiolate stability has been to replace simple monothiolate ligands with dithiolate or multithiolate ligands, which can potentially form multiple bonds with metal surfaces due to chelation effects.17-26 A number of groups have shown that dithiolate- and multithiolate-protected MPCs are more resistant to ligand exchange than monothiolate-protected MPCs.19,23,26 Herein we present results on the relative stability and surface protection of dithiolate-protected Au MPCs (based on (()-R-lipoic acid derivatives) compared with monothiolate-protected Au MPCs. Results show that the dithiolate MPCs are much more prone to surface oxidation and metal nanoparticle etching than their dodecanethiolate-protected MPC counterparts, likely due to poor packing of the dithiolate ligands on the MPC surface, leading to easier access of oxygen and cyanide anions to the metal surface. This work suggests that great care must be taken in attempting to make MPCs with multiple thiolate-metal linkages, as other factors such as packing efficiency can lead to oxidative instability. Several other groups have previously investigated the relative stability of a number of different thiolate ligands for MPCs. Murray and co-workers showed that the rate of ligand-exchange reactions of alkylamines with MPCs depended strongly on the steric bulk of the incoming ligands, and that the rate of cyanide etching of the Au MPC cores decreased with increasing (17) Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481–4483. (18) Garg, N.; Lee, T. R. Langmuir 1998, 14, 3815–3819. (19) Lin, S.-Y.; Tsai, Y.-T.; Chen, C.-C.; Lin, C.-M.; Chen, C. J. Phys. Chem. B 2004, 108, 2134–2139. (20) Langry, K. C.; Ratto, T. V.; Rudd, R. E.; McElfresh, M. W. Langmuir 2005, 21, 12064–12067. (21) Roux, S.; Garcia, B.; Bridot, J.-L.; Salome, M.; Marquette, C.; Lemelle, L.; Gillet, P.; Blum, L.; Perriat, P.; Tillement, O. Langmuir 2005, 21, 2526–2536. (22) Park, J.-S.; Vo, A. N.; Barriet, D.; Shon, Y.-S.; Lee, T. R. Langmuir 2005, 21, 2902–2911. (23) Wojczykowski, K.; Meissner, D.; Jutzi, P.; Ennen, I.; H€utten, A.; Fricke, M.; Volkmer, D. Chem. Commun. 2006, 3693–3695. (24) Choi, J.; Jun, Y.; Yeon, S.-I.; Kim, H. C.; Shin, J.-S.; Cheon, J. J. Am. Chem. Soc. 2006, 128, 15982–15983. (25) Lee, J.-S.; Lytton-Jean, A. K. R.; Hurst, S. J.; Mirkin, C. A. Nano Lett. 2007, 7, 2112–2115. (26) Agasti, S. S.; You, C.-C.; Arumugam, P.; Rotello, V. M. J. Mater. Chem. 2008, 18, 70–73.

Published on Web 08/03/2009

Langmuir 2009, 25(22), 12954–12961

Hou et al.

Article

alkanethiolate chain length and steric bulk.27 Recently, Rotello and co-workers showed that structural modifications of thiolate ligands can lead to significant improvements in both the thermal stability and stability of Au MPCs toward cyanide etching.26 Others have investigated dithiolate and multithiolate ligands as ligands for Au and Ag MPCs, on the basis that multiple thiolate linkages should increase the overall stability of the ligand-metal attachment. Roux et al. showed that Au MPCs stabilized by lipoic acid were much more resistant to ligand-exchange reactions with dithiothreitol than mercaptoundecanoic acid monolayers,21 while Mirkin and co-workers found that triple cyclic disulfide anchoring groups yielded Ag MPCs which were much more stable to oxidation than their monothiolate counterparts.25 While there has been a large amount of support for the enhanced stability of dithiolate and multithiolate ligands for SAMs and MPCs, Langry et al. have shown that monolayers based on lipoic acid are more easily removed than their monothiolate counterparts via tensile strength measurements.20 In this work, disulfide ligands based on (()-R-lipoic acid were synthesized as ligands that can bind to gold nanoparticle surfaces to form dithiolate stabilizers. Dithiolate-, 1-dodecanethiolate-, and mixed 1-dodecanethiolate/dithiolate-protected Au MPCs were synthesized and characterized by UV-vis spectroscopy, 1 H NMR, transmission electron microscopy (TEM), and thermal gravimetric analysis (TGA). TGA results showed that the dithiolate MPCs were more thermally stable than their monothiolate counterparts. The oxidative stability of each of these Au MPC samples was studied in the presence of both oxygen and cyanide anions, in order to determine their relative stabilities toward oxidation. Interestingly, 1H NMR results show that the dithiolate-stabilized MPCs are prone to oxidation in the presence of air under ambient conditions, whereas their dodecanethiolate-stabilized MPC counterparts are much more stable under these conditions. Preferential oxidation of the dithiolate ligands from mixed monolayers was also seen. Cyanide etching studies confirm that the cores of dithiolate-stabilized particles are much more accessible to substrates and etch at much higher rates than their monothiolate counterparts.

Experimental Section Materials. All solvents (high-performance liquid chromatography (HPLC)-grade toluene, chloroform, acetone, acetonitrile, tetrahydrofuran (THF), and ethanol) and KCN were purchased from EMD Chemicals Inc. and used as received. Deuterated solvents were purchased from Cambridge Isotope Laboratories. Tetraoctylammonium bromide (TOAB), 1-dodecanethiol, dodecylamine, (()-R-lipoic acid, 4-nitrophenol, and sodium borohydride were purchased from Aldrich, while hydrogen tetrachloroaurate(III) trihydrate and triethylamine were purchased from Alfa Aesar. 2-(1H-Benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU) and N-Hydroxybenzotriazole (HOBt) were purchased from Advanced ChemTech. All the chemicals were used without further purification. Synthesis. The disulfide ligand was synthesized via amide conjugation between (()-R-lipoic acid and amines, as shown in Scheme 1.28 A typical synthesis is as follows: 2.02 g of (()-R-lipoic acid, 5.84 g of HBTU, 2.21 g of HOBt, and 20 mL of triethylamine were stirred in 120 mL of chloroform for 1 h, followed by the addition of 2.40 g of dodecylamine. The solution was then stirred at room temperature for ∼72 h until it turned clear. The mixture was washed sequentially three times with saturated NaHCO3 solution and 10% citric acid solution. The organic layer was then (27) Templeton, A. C.; Hostetler, M. J.; Kraft, C. T.; Murray, R. W. J. Am. Chem. Soc. 1998, 120, 1906–1911. (28) K€onig, W.; Geiger, R. Chem. Ber. 1970, 103, 788–798.

Langmuir 2009, 25(22), 12954–12961

collected and concentrated. The crude product was subjected to silica gel column chromatography. Elution with 10% ethyl acetate and 90% chloroform gave compound 1 (0.804 g, 63.1% yield). 1-Dodecanethiolate-protected Au MPCs were synthesized via a modified Brust-Schiffrin method using standard literature procedures.29 All solvents were degassed with N2 (Praxair), and reactions were kept under an N2 atmosphere unless otherwise noted. A typical synthesis is as follows: 30 mL of an aqueous solution of HAuCl4 3 3H2O (0.40 g) was stirred with a solution of tetraoctylammonium bromide (TOAB, 0.34 g) in 80 mL of toluene until all the HAuCl4 was transferred to the organic layer and the water layer became colorless. 1-Dodecanethiol (0.17 g) was then added to the organic phase. After stirring for several minutes, 25.0 mL of a freshly prepared 0.40 M NaBH4 solution was added over 20 min, and the solution was stirred for the next 24 h. Excess thiol, free disulfide, and TOAB impurities were removed by sequential washing with ethanol, acetonitrile, and acetone. Dithiolate-protected Au MPCs (from compound 1) were synthesized via a one-phase method, which is similar to the methods reported by others.26,30 A typical synthesis is as follows: 5 mL of chloroform solution of HAuCl4 3 3H2O (0.039 g, 1.0  10-4 mol) was added into 50 mL of chloroform solution of disulfide (0.076 g of 1, 2.0  10-4 mol). Then 50 mL of a fresh aqueous solution of NaBH4 (0.077 g, 2.0  10-3 mol, 20 times excess) was added dropwise. The solution turned from bright orange to dark purple with the addition of NaBH4, indicating the formation of Au MPCs. After addition, the mixture was stirred for 2 h. The organic layer was extracted and washed twice with a 80% ethanol and 20% acetone mixture. Ligand-Exchange Reactions. Placed in a 25 mL roundbottom flask were 0.050 g of 1-dodecanethiolate- or dithiolateprotected Au MPCs (∼8.7 10-5 mol of ligands attached on the surface of Au MPCs) and 15 mL of chloroform. A 100 times excess (with respect to the number of ligands, 8.7 10-3 mol) of 1 or 1-dodecanethiol was added. The mixture was stirred at room temperature for 24 h. MPC Oxidation and Cyanide Etching Studies. Oxidation reactions were carried out by bubbling oxygen (Praxair) through 6.0 mL of solution of the 0.10 M Au MPCs (concentration with respect to gold) in d8-THF. One milliliter of this solution was removed for NMR measurements at certain time intervals. Only one detailed NMR time-study experiment was carried out for each MPC type; however, oxidation of dithiolate MPCs was observed in air on multiple occasions. For cyanide etching studies, 0.50 mL of a freshly prepared aqueous solution of KCN (10 mM) was added to a 3.0 mL solution of Au MPCs in THF (1.0 mM in Au). The decay in absorbance at 520 nm was monitored every minute by UV-vis spectroscopy until the solution became colorless. The decomposition rate data were fit to a general first-order kinetics equation, y=a e-bx, in which y is the absorption and x is the time (min).27 At least two experiments were run on each MPC type. Characterization. Absorption spectra were recorded on a Varian Cary 50 Bio UV-vis spectrometer with an optical path length of 1.0 cm. Transmission electron micrographs were obtained with a Philips 410 microscope operating at 100 keV. Samples were prepared by placing a drop of solution on a holey-carbon-coated Cu TEM grid (400 mesh) and allowing the solvent to evaporate in air. Two hundred nanoparticles were counted for each sample; particle diameters were measured manually using ImageJ software. 1H NMR and 2D COSY NMR were recorded on a Bruker 500 MHz Avance spectrometer; chemical shifts were referenced to the residual protons of the deuterated solvent. The relative amounts of dithiolate ligand as (29) Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. J. Chem. Soc., Chem. Commun. 1994, 801–802. (30) Zheng, M.; Davidson, F.; Huang, X. J. Am. Chem. Soc. 2003, 125, 7790– 7791.

DOI: 10.1021/la9018053

12955

Article

Hou et al. Scheme 1. Amide Conjugation between (()-r-Lipoic Acid and Dodecylamine to Form Disulfide Ligand 1

well as oxidized disulfide molecules (compound 1) were calculated from 1H NMR data using the integrated areas of methine R-H (proton 9 in Figure 3 below) signals from the disulfide species and methyl signals from both species. TGA was performed on a TGA Q5000IR (TA Instruments). The Au MPCs were placed in a ceramic TGA pan and heated from 25 to 600 °C at a rate of 10 °C min-1 under a nitrogen atmosphere, and the weight loss was recorded as a function of temperature.

Results and Discussion 1-Dodecanethiolate-protected Au MPCs were synthesized via a modified Brust-Schiffrin method.29 Figure 1 shows UV-vis spectra of the 1-dodecanethiolate-protected Au MPCs solution in chloroform; the very weak plasmon shoulder indicates the size of Au MPCs is quite small (below 2 nm).11,31 TEM measurements show the average particle size of 1-dodecanethiolate-protected Au MPCs is 1.8 ( 0.4 nm (Figure 2A). The disulfide ligand based on lipoic acid was synthesized according to Scheme 1. 1H NMR (CDCl3 500 MHz) analysis indicates the desired disulfide ligand was formed (see below). The Brust-Schiffrin method was not feasible for the synthesis of dithiolate-protected Au MPCs, as it was found that the phase-transfer agent, TOAB, attached strongly to the surface of the Au MPCs and could not be removed from the Au MPCs. Thus, dithiolate-protected Au MPCs were synthesized using a one-phase method, in which tetrachloroauric acid was reduced by sodium borohydride in the presence of the disulfide ligand in chloroform using a procedure similar to those reported by Huang et al.30 and Rotello et al.26 Roux et al. have previously noted that dithiolate-protected Au MPCs can be made from either the disulfide (lipoic/thioctic acid) or the dithiol (dihydrolipoic acid) and that the resulting Au MPCs are both stabilized by dithiolates.21 Figure 1 shows the UV-vis spectra of the dithiolate-protected Au MPCs. The plasmon band of the Au MPCs is near 523 nm, which is typically indicative of Au MPCs between 2 and 4 nm.32-34 This is in general agreement with TEM results, shown in Figure 2B, which indicate that the dithiolatestabilized Au MPCs have an average size of 3.0 ( 1.0 nm. The NMR spectra of free the disulfide ligand and corresponding dithiolate-stabilized Au MPCs are shown in Figure 3. It should be noted that the signal for the methine proton (proton 9) of the dithiolate ligand disappeared when the ligand was attached on the Au surface; in addition, the amide proton (proton 5) signal was shifted to about 6.6 ppm and became very broad (not shown in Figure 3), and the positions of the methylene protons (10-13) were shifted and broadened significantly. The disappearance of the methine proton (proton 9) upon ligand attachment to the Au surface is likely due to broadening effects due to its proximity to (31) Schaaff, T. G.; Shafigullin, M. N.; Khoury, J. T.; Vezmar, I.; Whetten, R. L.; Cullen, W. G.; First, P. N.; Gutierrez-Wing, C.; Ascensio, J.; Jose-Yacaman, M. J. J. Phys. Chem. B 1997, 101, 7885–7891. (32) Burda, C.; Chen, X.; Narayanan, R.; El-Sayed, M. A. Chem. Rev. 2005, 105, 1025–1102. (33) Link, S.; El-Sayed, M. A. Int. Rev. Phys. Chem. 2000, 19, 409–453. (34) 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, 17–30.

12956 DOI: 10.1021/la9018053

Figure 1. UV-vis spectra of 1-dodecanethiolate-protected Au MPCs (solid), dithiolate-stabilized Au MPCs (dashed), and mixed 1-dodecanethiolate/dithiolate-protected Au MPCs (dotted).

the gold surface, such that it cannot undergo rotational motion. Similar effects have been seen for other groups on thiols in close proximity to gold surfaces.35 We are not certain as to the reason for the NMR downfield shift of the amide proton; two possibilities include hydrogen bonding with the adjacent ligands on the particle surface or weak back-bonding of amide groups to the Au surface. In an effort to make mixed-ligand species that have both thiolate and dithiolate stabilizers, attempts were made to ligandexchange pure MPCs. 1-Dodecanethiolate-protected Au MPCs were found to be quite stable in the presence of excess amounts of the disulfide ligand 1 (even in the presence of NaBH4); after 24 h at room temperature, no ligand-exchange products were observed by 1H NMR. Attempts to heat up the dodecanethiolate-stabilized MPCs in the presence of disulfide ligands to yield Au MPCs stabilized by mixtures of ligands resulted in complete decomposition of the Au MPCs. Others have previously attempted ligand exchanges of alkanethiolate-stabilized MPCs with incoming disulfide ligands and found that either no ligand exchange occurred14 or that ligand-exchange reactions only occurred when the alkanethiolate chain length was short.36 On the other hand, dithiolate-stabilized Au MPCs were reactive in the presence of excess 1-dodecanethiol. After 24 h at room temperature, about 50% of the dithiolate ligands were replaced by 1-dodecanethiol (as observed by 1H NMR), leading to the formation of Au MPCs with mixed ligand stabilizers. One plausible explanation for this reactivity is that the dithiolate ligand cannot pack densely on the surface of Au MPCs due to steric constraints. This allows (35) Kohlmann, O.; Steinmetz, W. E.; Mao, X.-A.; Wuelfing, W. P.; Templeton, A. C.; Murray, R. W.; Johnson, C. S. J. Phys. Chem. B 2001, 105, 8801–8809. (36) Ionita, P.; Caragheorgheopol, A.; Gilbert, B. C.; Chechik, V. J. Am. Chem. Soc. 2002, 124, 9048–9049.

Langmuir 2009, 25(22), 12954–12961

Hou et al.

Article

Figure 2. TEM images of (A) 1-dodecanethiolate-protected Au MPCs, (B) dithiolate-stabilized Au MPCs, and (C) mixed 1-dodecanethiolate/dithiolate-protected Au MPCs. Histograms for each transmission electron micrograph can be found in the Supporting Information.

Figure 3. 1H NMR of disulfide ligand (top) and dithiolate-protected Au MPCs (bottom). Assignments were verified by 2D COSY NMR (See Supporting Information).

1-dodecanethiol ligands with a smaller volume to approach the surface and replace disulfide ligands via a ligand-exchange mechanism. Figure 1 shows the UV-vis spectrum of the mixed 1-dodecanethiolate/dithiolate-protected Au MPCs. The shape and position of the plasmon band is similar to that of dithiolate-protected Au MPCs, which indicates the size of mixed 1-dodecanethiolate/dithiolate-protected Au MPCs should be between 3 and 4 nm.32-34 The TEM image in Figure 2C shows that their average size is 4.1 ( 1.3 nm, which indicates that some particle size growth occurs during ligand exchange. We are not Langmuir 2009, 25(22), 12954–12961

certain as to why the MPC particle size is increasing during ligand exchange, but it may be due to the fact that the steric bulk of the incoming ligand is much smaller than the outgoing dithiolate ligand, thus leaving the surface incompletely protected during the ligand-exchange reaction. The thermal stability of the Au MPCs was studied by using thermogravimetric analysis (TGA). The mass change was plotted against temperature to generate the thermogravimetric profiles; such mass losses reflect the decomposition and removal of thiolate ligands from the gold surface.26 Figure 4A shows percentage DOI: 10.1021/la9018053

12957

Article

Hou et al.

weight loss curves for the three MPCs; the percentage mass of ligands is 28.0% in 1-dodecanethiolate-protected Au MPCs, 27.8% in dithiolate-protected Au MPCs, and 24.6% in 1:1 mixed 1-dodecanethiolate/dithiolate-protected Au MPCs. It should be noted that each of these samples did not contain free thiols/ disulfides or other impurities as determined by 1H NMR. Combining these data and the average particle sizes that were obtained by TEM, we can calculate, on average, how many gold atoms and ligand molecules constitute individual Au MPC in each sample. The results are summarized in Table 1. The number of Au atoms

in one cluster for three Au MPCs was calculated using eq 1 below, where n is the number of atoms per cluster, R is the nanoparticle radius, and Vg is the molar volume of Au, 10.2 cm3/mol. The number of ligands in one cluster for each of the three Au MPCs was estimated using eq 2 below, where n is the number of atoms per cluster, N is the number of ligands per cluster, AW is the atomic weight of Au, MW is the molecular weight of one thiolate ligand (eq 2 is derived from Weight % of Ligand =(N  MW)/ (N  MW þ n  AW), followed by rearrangement to solve for N). n ¼ 4πR3 =3V g N ¼ ðn  AW  wt%of ligandÞ=ðMW  wt%of AuÞ

Figure 4. (A) Weight loss versus temperature curves of 1-dodecanethiolate-, dithiolate-, and 1:1 mixed 1-dodecanethiolate/dithiolate-protected Au MPCs, and (B) derivative thermogravimetric curves of the Au MPCs. Table 1. Number of Au Atoms and Number of Ligands of a Single Nanoparticle for Each of the Three Au MPCs Based on TEM and TGA Results Au-MPC stabilizer

TEM D (nm)

TGA weight % of ligands

no. of Au atoms and ligands in one clustera

1-dodecanethiolate 1.8 ( 0.4 28.0 Au180Lm68 dithiolate 3.0 ( 1.0 27.8 Au834Ld170 1:1 mixed 14.1 ( 1.3 24.6 ∼Au2130Lm238Ld238 dodecanethiolate/ dithiolate a Lm: monothiolate (1-dodecanethiolate) ligands; Ld: dithiolate ligands. Relative amounts of each ligand in the mixed system was assumed to be 50/50, which was the ratio estimated from 1H NMR.

12958 DOI: 10.1021/la9018053

ð1Þ ð2Þ

The diameter of 1-dodecanethiolate-protected Au MPCs measured by TEM is 1.8 ( 0.4 nm, and the weight loss of adsorbed 1-dodecanethiolate ligands determined by TGA is 28.0%. This is consistent with the results obtained by Murray et al. (28.8% weight loss for 2.2 nm dodecanethiolate-protected Au MPCs).34 In addition, the weight loss associated with the 3.0 ( 1.0 nm dithiolate-protected MPCs (27.8%, ∼170 ligands) agrees well with earlier results by Murray for 2.8 nm particles covered by dodecanethiolate ligands (16.9% weight loss, ∼161 ligands). However, the weight losses seen for the 4.1 ( 1.3 nm mixed 1-dodecanethiolate/dithiolate-stabilized Au MPCs (24.6%, ∼ 476 ligands) are larger than earlier results obtained by Murray et al. (12.8% weight loss, ∼334 ligands, for 4.0 nm Au nanoparticles covered by dodecanethiolate),34 even when the difference of molecular weights of the different ligands has been considered. Similarly, the mixed thiolate/dithiolate-stabilized Au MPCs have a weight loss of 24.6%, which is higher than the value of 9.6% seen by Roux et al. for 5.5 nm Au nanoparticles stabilized by dihydrolipoic acid.21 Figure 4B shows the derivative curves for TGA mass losses for the three Au MPCs. From the positions of the derivative peaks, it is clear that the dithiolate-protected Au MPCs have the highest decomposition temperature. The positions of the derivative peaks reflect the thermal stability of the thiolate ligands. The dithiolate ligand is bulkier and has two thiolate attachments to the Au surface; thus, it is significantly more thermally stable than 1-dodecanethiolate-protected MPCs. The derivative peak of 1:1 mixed 1-dodecanethiolate/dithiolate-protected Au MPCs is quite broad, indicating decomposition of two kinds of thiolate ligands over a wide temperature range. Pure 1-dodecanethiolate-protected Au MPCs are fairly stable in the presence of oxygen over 72 h, a result that was observed previously in our group and by others, i.e., that dodecanethiolatestabilized Au MPCs are stable in air in the absence of halide impurities.11 Purified dithiolate-protected Au MPCs that were exposed to oxygen showed changes in their 1H NMR spectra (Figure 5) over time, as the anchored dithiolate ligands were oxidized to the free disulfide, 1. The peaks labeled by an asterisk in the bottom part of Figure 5 are attributed to free disulfide 1. Table 2 shows the percentage amounts of anchored dithiolate and/or 1-dodecanethiolate oxidized from the respective Au MPCs calculated from integration of 1H NMR spectra over different time intervals during the oxidation of each of the three Au MPCs. Compared with the oxidation of 1-dodecanethiolate-protected Au MPCs, the oxidative stability of the dithiolate-protected Au MPCs is poor, as 12.5% of the dithiolate ligands were oxidized to form free disulfides (compound 1) after only 2 h. The oxidation process slowed down after 2 h, and only 19% dithiol ligands were oxidized after 72 h. Figure 6A shows the UV-vis spectra of the Langmuir 2009, 25(22), 12954–12961

Hou et al.

Article

Figure 5. 1H NMR spectra of dithiolate-protected Au MPCs before (top) and after (bottom) exposure to oxygen for 21 h. Table 2. Percentage of Anchored Dithiolate or 1-Dodecanethiolate Oxidized at Different Time Intervals for Au MPCs, As Determined by 1H NMR percentage amounts of anchored ligands oxidized dithiolate-protected Au MPCs

1-dodecanethiolate-protected Au MPCs

time (h) 0 2 24 48 72

0% 12.5% 13.7% 16.1% 19.0%

0% 0% 0% 0% 0%

dithiolate-protected Au MPCs before and after exposure to oxygen; the plasmon band at 523 nm shifts to 526 nm after exposure to oxygen and increases significantly in intensity. This result suggests the average size of dithiolate-protected Au MPCs increases upon oxygen exposure. TEM results confirm the particle size growth; as shown in Figure 7A, the average nanoparticle size of the dithiolate-protected Au MPCs before oxidation is 3.0 ( 1.0 nm, while the size after oxidation is 4.6 ( 1.2 nm. The difference of the oxidative stabilities between 1-dodecanethiolate- and dithiolate-protected Au MPCs is surprising given that others have noted that dithiolate interactions with gold surfaces are more robust than those of thiolates. However, a possible explanation for these results is that 1-dodecanethiolateprotected Au MPCs have dense packing on their surface, thus making it difficult for oxygen to access the Au surface, while the dithiolate ligands on the dithiolate-protected Au MPCs do not Langmuir 2009, 25(22), 12954–12961

mixed 1-dodecanethiolate/dithiolate-protected Au MPCs dithiolate

1-dodecanethiolate

0% 6.5% 6.7% 9.4% 28.7%

0% 1.7% 1.5% 2.8% 4.0%

pack well, allowing for oxygen to access the surface. In addition, the dithiolate-to-disulfide oxidation is an intramolecular reaction (compared to the intermolecular reaction needed to form disulfides from 1-docanethiolate ligands), which may result in a kinetically favorable oxidation process. Alternatively, these results could suggest that the binding motifs for dodecanethiolate and dithiolate ligands could be fundamentally different, which could lead to large differences in the stabilities of the ligands and the gold cores. 1:1 mixed 1-dodecanethiolate/dithiolate-protected Au MPCs are also unstable in the presence of oxygen, and both anchored 1-dodecanethiolate and dithiolate were oxidized, as verified by 1 H NMR. However, only 4.0% of the 1-dodecanethiolate ligands were oxidized, while 28.7% of the dithiolate ligands were oxidized after 72 h. UV-vis spectra of 1:1 mixed 1-dodecanethoilate/ dithiolate-protected Au MPCs before and after exposure to DOI: 10.1021/la9018053

12959

Article

Figure 6. UV-vis spectra of (A) dithiolate-protected Au MPCs and (B) 1:1 mixed 1-dodecanethiolate/dithiolate-protected Au MPCs before (solid) and after (dotted) exposure to oxygen for 72 h.

Hou et al.

oxygen are shown in Figure 6B. The minimal change of plasmon band indicates little or no change in average particle size; however, TEM measurements indicate that the particle sizes increase slightly from 4.1 ( 1.3 nm to 4.8 ( 1.3 nm (see Figure 7B). The surface accessibility of 1-dodecanethiolate-, dithiolate-, partially oxidized dithiolate- (in which ∼10% attached dithiolate ligands were oxidized), and 1:1 mixed 1-dodecanethiolate/dithiolate-protected Au MPCs was further probed by examining cyanide-induced etching of the gold cores. The solutions of Au MPCs were all etched in the presence of KCN to give colorless solutions comprising Au(CN)2- complexes, disulfides, and alkyl cyanides.26 The reaction kinetics were obtained by monitoring the absorption changes of the MPC solutions (4.8 μM of nanoparticles for 1-dodecanethiolate-protected Au MPCs and 1.0 μM of nanoparticles for dithiolate-protected Au MPCs) at 520 nm every minute in the presence of an excess amount of an aqueous solution of KCN (1.43 mM). Figure 8A shows a typical etching experiment of dithiolate-stabilized Au MPCs; the Au plasmon band decreased gradually over time as the Au cores were etched. Representative etching profiles, as followed by UV-vis spectroscopy, for the four MPCs (1-dodecanethiolate-, dithiolate-, partially oxidized dithiolate-, and 1:1 mixed 1-dodecanethiolate/ dithiolate-protected MPCs) are depicted in Figure 8B. The concentration of KCN was treated as a constant during the reaction course, given that its concentration was much higher than the concentration of Au MPCs. The absorption data of all four MPCs follow first-order reaction kinetics. The pseudo firstorder and second-order rate constants for the KCN etching are summarized in Table 3; they reveal that partially oxidized dithiolate-protected Au MPCs have the highest Au core etching rate and 1-dodecanethiolate-protected Au MPCs are the most stable to KCN etching, despite their smaller particle size. Similar first-order kinetics of cyanide-induced etching of Au MPCs was previously reported by other groups.26,27 The relative etching rates of the Au MPCs are likely governed primarily by the surface accessibility of Au MPCs, as cyanide anions need first to access the surface to etch Au MPCs.26 Thus, the surfaces of dithiolateprotected Au MPCs are more accessible than 1-dodecanethiolateprotected Au MPCs, which agrees with oxidation data from the last section. The partially oxidized dithiolate-protected Au MPCs have the highest etching rate since they have the most accessible

Figure 7. TEM images of (A) dithiolate- and (B) mixed 1-dodecanethiolate/dithiolate-protected Au MPCs after exposure to oxygen for 72 h. Histograms for each TEM can be found in the Supporting Information. 12960 DOI: 10.1021/la9018053

Langmuir 2009, 25(22), 12954–12961

Hou et al.

Article

The increased etching rates of the partially oxidized dithiolateprotected MPCs indicate that such MPC oxidation strategies may be a viable strategy to produce MPCs with enhanced catalytic activities. Several groups have previously reported that thiolateprotected Pt or Pd nanoparticles do indeed have catalytic activities for hydrogenation and Suzuki cross-coupling reactions, although their catalytic activities were quite low.37,38 In addition, Murakami et al. demonstrated that dodecanethiolate-capped gold clusters enhanced the activity of a Mn-porphyrin catalyst for olefin oxidation reactions and postulated that partially oxidized thiolate-protected Au clusters are responsible for the improved activity of Mn-porphyrin catalyst.39 Other groups have successfully shown that bulky and/or dendritic thiols can lead to the synthesis of catalytically active MPCs.40,41 Thus partial thiolate oxidation may be another route to MPCs with more accessible surfaces.

Conclusions

Figure 8. (A) UV-vis spectra of dithiolate-protected Au MPCs upon exposure to KCN and (B) UV-vis absorption changes of Au MPCs at 520 nm in the presence of KCN at room temperature. Each plot represents a 1 min interval. Table 3. Pseudo-First-Order (k1) and Second-Order Rate Constants (k2) for the Decomposition of Au MPCs by KCN at Room Temperature Au-MPC stabilizer

k1/min-1

k2/dm3 mol-1 min-1

1-dodecanethiolate 9.16  10-3 ( 1.0  10-4 6.41 ( 0.07 dithiolate 0.166 ( 0.004 116.16 ( 2.80 partially oxidized 0.245 ( 0.007 171.12 ( 4.90 dithiolate 11.5 ( 0.168 1:1 mixed 1-dodecane- 1.64  10-2 ( 2.4  10-4 thiolate/dithiolate

surface, given that ∼10% of the anchored dithiolate ligands have been oxidized and detached from the surface of gold core. The 1:1 mixed 1-dodecanethiolate/dithiolate-protected Au MPCs have intermediate etching kinetics, as their surfaces are less accessible than those of pure dithiolate-protected Au MPCs.

Langmuir 2009, 25(22), 12954–12961

1-Dodecanethiolate-, dithiolate-, and 1:1 mixed 1-dodecanethiolate/dithiolate-protected Au MPCs have been synthesized, and their thermal and oxidative stability in the presence of oxygen and cyanide anions has been studied. Somewhat surprisingly, results revealed that dithiolate ligands were more susceptible to oxidation than their monothiolate counterparts, which suggests that the core MPC surfaces covered by dithiolate ligands are more accessible than those covered by thiolate monolayers. Cyanide etching experiments confirm this hypothesis, as the dithiolatestabilized Au MPCs were etched at much higher rates than the monothiolate- and mixed-monolayer-stabilized MPCs. These systematic investigations have revealed that the stability of Au MPCs can be tuned by choosing different thiolate-based ligands and oxidation conditions and indicate that great care must be taken when attempting to make more robust ligand attachments on metal surfaces by increasing the number of binding groups; increasing the number of attachments does not necessarily lead to more stable MPCs. Finally, results of cyanide etching experiments on partially oxidized MPCs suggest that this may be a viable route toward increasing the catalytic activity of MPCs by increasing the number of surface sites available for catalysis. Future work in our group is ongoing to see whether efficient MPC catalysts can be developed by such routes. Acknowledgment. We thank NSERC and the University of Saskatchewan for financial support of this project, the Saskatchewan Structural Science Centre for access to NMR facilities, and Sarah Caldwell (Western College of Veterinary Medicine) at University of Saskatchewan for providing assistance with TEM measurements. Supporting Information Available: TEM histograms of all Au MPCs and 2D COSY spectra of compound 1 and dithiolate-protected Au MPCs. This material is available free of charge via the Internet at http://pubs.acs.org. (37) (38) (39) (40) 1152. (41)

Lu, F.; Ruiz, J.; Astruc, D. Tetrahedron Lett. 2004, 45, 9443–9445. Eklund, S. E.; Cliffel, D. E. Langmuir 2004, 20, 6012–6018. Murakami, Y.; Konishi, K. J. Am. Chem. Soc. 2007, 129, 14401–14407. Alvarez, J.; Liu, J.; Roman, E.; Kaifer, A. E. Chem. Commun. 2000, 1151– Gopidas, K. R.; Whitesell, J. K.; Fox, M. A. Nano Lett. 2003, 3, 1757–1760.

DOI: 10.1021/la9018053

12961