C
Nanoscale
Dynamic Article Links <
Cite this: Nanoscale, 2012, 4, 4091
COMMUNICATION
www.rsc.org/nanoscale
Ultrafast, 2 min synthesis of monolayer-protected gold nanoclusters (d < 2 nm)† Matthew N. Martin,a Dawei Li,a Amala Dassb and Sang-Kee Eah*a Received 13th April 2012, Accepted 28th May 2012 DOI: 10.1039/c2nr30890h
An ultrafast synthesis method is presented for hexanethiolate-coated gold nanoclusters (d < 2 nm, <250 atoms per nanocluster), which takes only 2 min and can be easily reproduced. With two immiscible solvents, gold nanoclusters are separated from the reaction byproducts fast and easily without any need for post-synthesis cleaning. Gold nanoclusters (d < 2 nm, <250 atoms per nanocluster) and nanoparticles (d > 2 nm) have a wide range of applications.1 Due to the strong size-dependent properties, some applications specifically require nanoclusters. For example, smaller gold nanoclusters have much higher catalytic activities than larger nanoparticles, where the critical diameter is 2 nm.2 DNA or drug delivery through cell membranes using gold nanoparticles is more efficient with decreasing size.3 Single-electron charging at room temperature also requires nanoclusters smaller than 2 nm.4 The surface plasmon resonance near 520 nm for gold nanoparticles completely disappears for nanoclusters.5 The most popular synthesis method by Brust et al. generates a polydisperse mixture of nanoclusters and nanoparticles in the diameter range of 1–4 nm.6 A major modification for the Brust method, adjusting the ratio of thiol molecules to gold atoms to 3, was reported by the Whetten group7 for synthesis of gold nanoclusters only.5 Recently, atomically monodisperse gold nanoclusters have been of great interest, especially after atomic precision X-ray characterization of a crystal of Au102 nanoclusters, followed by crystallization of Au25 nanoclusters.8 Since then, many groups have reported refined synthesis methods for generating single populations of atomically monodisperse gold nanoclusters such as Au25, Au38, Au102, Au144, and others.8,9 All these synthesis methods have adopted the thiol : gold ratio of 3 and generate Au(I)-thiolate polymers before the final product of nanoclusters. Even though it takes <1 s for a strong reducer like BH4 to reduce gold ions in water, formation of the polymers and reduction
a Department of Physics, Applied Physics, and Astronomy, Rensselaer Polytechnic Institute, Troy, New York 12180, USA. E-mail:
[email protected]; Fax: +1-518-276-6680; Tel: +1-518-276-3972 b Department of Chemistry and Biochemistry, University of Mississippi, Mississippi 38677, USA † Electronic supplementary information (ESI) available: Experimental details of gold nanocluster synthesis and mass-spectrometry. See DOI: 10.1039/c2nr30890h
This journal is ª The Royal Society of Chemistry 2012
of bound Au(I) ions take at least several hours or days.5–9 Details on how to form Au(I)-thiolate polymers are not fully disclosed in the literature, even with best efforts by the authors: recently the Kornberg group used 96 wells to find the right conditions to control the diameter in an extremely small range of 2 to 3 nm.10 The first Au102 crystal was selected from a polydisperse solution of gold nanoclusters and it took 4 years for the same group to increase the yield from <1% to >50%.11 Even worse is the very slow size-focusing9,12 (or thioletching), taking at least several hours or days, of polydisperse nanoclusters into a single group of stable (or magic numbered) nanoclusters. As a result, different groups like the Jin group for Au144 nanoclusters and the Kornberg group for Au102 nanoclusters cannot reproduce others’ results yet. Therefore, we aimed to develop a new synthesis method for gold nanoclusters by removing the step of forming Au(I)-thiolate polymers so that it can be easily reproduced and readily adopted by any other interested group. Here, we present such a method for hexanethiolatecoated gold nanoclusters, which takes only 2 minutes and does not require any post-synthesis cleaning. The reactions among gold ions, reducing ions, gold nanoclusters, and alkanethiol ligand molecules are performed in a single phase of mostly methanol. Then, alkanethiolate-coated gold nanoclusters and the remaining free thiol molecules are extracted into hexane via fast phase separation. Contrary to all other synthesis methods, minimizing the reaction of Au(III) ions with thiol molecules as much as possible is key to controlling the diameter (d) <2 nm. Recently, we discovered that nearly monodisperse gold nanoparticles in mildly basic water after reaction can be made simply by mixing AuCl4/2H+/Cl ions with BH4/OH/2Na+ ions at room temperature. Without using any stabilizer like citrate, the diameter of these ‘‘naked’’ (or stabilized by adsorbed hydroxide anions and/or boron anions) nanoparticles in water can be reproducibly tuned in a very narrow range of 3.2 to 5.2 nm.13 For the smallest reproducible diameter of 3.2 nm, the aqueous solution’s color changes from light yellow for gold ions before adding the reducing chemical to initially orange in <1 s and finally red in 1 min. This implies that all gold ions are fast reduced to gold nanoclusters (d < 2 nm) and nanoclusters grow into nanoparticles (d > 2 nm) by coagulation, separating the initial fast nucleation from the later slow growth.14 We speculated that the diameter could be controlled <2 nm, if stabilizing molecules are present before adding a strong reducer, like BH4. However, adding citrate ions, a weak stabilizer and poor reducer at room temperature, did not change the diameter. Nanoscale, 2012, 4, 4091–4094 | 4091
Fig. 1 Hexanethiolate-coated gold nanoclusters (d < 2 nm) in hexane were made by fast reduction of all gold ions with borohydride, coating nanoclusters with hexanethiolate, and phase-transferring them from mostly methanol to hexane. The whole process took <2 min without any need for postsynthesis cleaning, since all of the reaction byproduct ions remained in the mostly methanol bottom phase.
With strong stabilizers, such as water-soluble thiols 4-mercaptobenzoic acid or 3-mercaptopropionic acid, we could decrease d < 3 nm but we could not find the right conditions for d < 2 nm.15 The aqueous solution’s color changed from light yellow to cloudy white in <10 s due to the rapid reaction of Au(III) ions with thiol molecules in water, which could be avoided by adding the water-soluble thiol and the reducer at the same time for the smallest size of d < 3 nm.16 The dilemma for controlling d < 2 nm was that the rapid growth of gold nanoclusters into nanoparticles must be stopped by organothiolatecoating but at the same time the reaction of Au(III) ions with thiol molecules must be minimized as much as possible. We solved this dilemma by reducing Au(III) ions with water-insoluble alkanethiol molecules in mostly methanol. This size control, d < 2 nm, is a result of a very delicate balancing act. In water, water-soluble thiols reacted too fast with gold ions and water-insoluble alkanethiols were too slow in stopping the growth of nanoclusters. Both cases resulted in larger nanoclusters/nanoparticles (d < 3 nm or larger). In a methanol–water mixture, gold nanoclusters grew too fast into nanoparticles (d > 3 nm) without a strong stabilizer but stopping this growth with alkanethiols was fast enough for d < 2 nm. To 5 g of methanol, 6.7 mg of hexane with 5.0 mmol of 1-hexanethiol molecules and 100 mg of water with 5.0 mmol of AuCl4/ 2H+/Cl ions were added. Then, 300 mg of water containing 15 mmol of BH4/OH/2Na+ ions was added all at once on a shaker for fast and uniform mixing. As shown in Fig. 1, gold nanoclusters were immediately formed and coated by hexanethiolate in the mostly methanol mixture. Then, they rapidly precipitated to the bottom (ESI, Fig. 1S†). These precipitated gold clusters were phase-transferred to the top phase of hexane by shaking for 30 seconds, where they are completely redispersed. Since hexane is immiscible with methanol at 0.25–0.75 mole fractions,17 all the reaction byproduct ions remain in the bottom methanol–water phase. The small amount of water further confines the reaction byproduct ions to the bottom mostly methanol phase away from the top hexane phase. Therefore, a clean (except for the only impurity of free, unreacted alkanethiol molecules) hexane solution of alkanethiolate-coated gold nanoclusters (d < 2 nm) could be prepared in <2 min without any need for post-synthesis cleaning, since the reagent stock solutions were prepared beforehand. All steps were done at room temperature in air. In the mostly methanol mixture, the reaction of Au(III) ions with water-insoluble hexanethiol molecules is much slower than with water-soluble thiol molecules in water (ESI, Fig. 2S†). Reaction for 5 4092 | Nanoscale, 2012, 4, 4091–4094
minutes was too short to change the solution’s color from light yellow to cloudy white but long enough to make d < 3 nm. When this reaction time was minimized to <5 s, the diameter could be controlled <2 nm. As shown in Fig. 2, the UV-Vis spectrum of gold nanoclusters (d < 2 nm) is featurelessly rising from the visible to the UV,5 while the spectrum of gold nanoclusters/nanoparticles (d < 3 nm) has the shoulder around 520 nm due to the surface plasmon resonance of nanoparticles. The gold nanoclusters were dried by evaporating hexane, redispersed in hexane, and characterized by mass-spectrometry. Interestingly, the matrix-assisted laser desorption ionization mass spectrum in the time-of-flight mode (MALDI-TOF)18 in Fig. 3 shows two sharp peaks for Au25 and Au38 nanoclusters and three broad peaks for Au68, Au102, and Au144 nanoclusters, even though they were formed very rapidly in <1 s and there was no slow size-focusing9,12 for hours or days. We found out that the same ‘‘sweet zone’’13 applies to both ‘‘naked’’ gold nanoparticle in water and alkanethiolate-coated gold nanoclusters in mostly methanol. As shown in Fig. 4, nanoclusters (d < 2 nm) were made with BH4/OH/2Na+ ions equal to or more than 300% molar amount of gold atoms, while polydisperse
Fig. 2 UV-Vis spectra of hexanethiolate-coated gold nanoclusters in hexane. Smaller nanoclusters (d < 2 nm, bottom) were made when the reaction of Au(III) ions with 1-hexanethiol molecules was minimized (<5 s) before adding the reducer, borohydride. Waiting for 5 min resulted in larger nanoclusters/nanoparticles (d < 3 nm, top).
This journal is ª The Royal Society of Chemistry 2012
Fig. 3 MALDI-TOF mass spectrum of magic numbered hexanethiolatecoated gold nanoclusters (d < 2 nm, <250 atoms per nanocluster).
nanoparticles/nanoclusters were made with less than 300%. In the same ‘‘sweet zone’’, nearly monodisperse nanoparticles (tunable d ¼ 3.2–5.2 nm) were made in water.13 From this observation, we infer that the same fastest reaction of nanocluster formation in <1 s in both water and mostly methanol is followed by growth into nanoparticles in water and no growth due to alkanethiolate-coating in mostly methanol, respectively. We discovered that controlling d < 2 nm requires careful optimization of the ultrafast reaction conditions. The addition of BH4/ OH/2Na+ ions must be done all at once instead of dropwise, while the mostly methanol solution must be on a shaker for the fastest and most uniform mixing possible. Alkanethiol molecules >100% and <75% molar amounts of gold atoms may result in too fast reaction of thiol molecules with gold ions and too slow stopping the growth of gold nanoclusters into nanoparticles, respectively. Using hexanethiol molecules more than 100% molar amount of gold atoms resulted in more trapping of gold nanoclusters/nanoparticles
at the methanol–hexane interface due to their larger size (ESI, Fig. 3S†). Also, hexanethiol molecules less than 75% resulted in d < 3 nm. We found no difference between 1-hexanethiol and a longer chain thiol, 1-dodecanethiol. However, with a benzene ring thiol, 4-methylbenzenethiol, we could not find the right conditions for controlling d < 2 nm and efficient phase-transfer to hexane or toluene. The solubility difference in the mostly methanol mixture between alkanethiol and benzenethiol may be the reason. With amines and phosphines, weaker stabilizers than thiol, we could not stop the growth of gold nanoclusters into nanoparticles fast enough. This synthesis method does not generate a single group of atomically monodisperse8,9 gold nanoclusters. Instead, the mixture of polydisperse gold nanoclusters could be separated into the 5 groups of more monodisperse nanoclusters via ultracentrifugation19 or sizeexclusion chromatography.20 Or, the smallest Au25 nanoclusters could be separated by the size-dependent solubility differences in various solvents such as acetonitrile and acetone.21 This approach of combining ultrafast and reproducible synthesis of polydisperse nanoclusters with post-synthesis size-filtering might be more efficient and available than directly synthesizing atomically monodisperse nanoclusters via the very slow size-focusing for hours or days, which is difficult to be reproduced by other groups independently at this moment. We tried to scale up this synthesis by increasing all the concentrations of the reagents 10 times but found out that the diameter becomes larger, d < 3 nm. Previously, we discovered that ‘‘naked’’ gold nanoclusters grow quickly into larger nanoparticles in water without a strong thiol stabilizer.13 Even with more stabilizing alkanethiol molecules, nanoclusters may grow too fast at 10 times higher concentrations of all agents than ones for d < 2 nm. Or, alkanethiols may react with gold ions too fast in mostly methanol at higher concentrations. Alternatively, scaling up could be done via a continuous flow process with a micromixer22 for better and faster mixing, taking advantage of the ultrafast reaction, the fast and easy phase separation, no need for post-synthesis cleaning, and convenient recovery of the very volatile solvent, hexane.
Fig. 4 For ultrafast (<1 s) and complete reduction of all gold ions, the ratio of BH4/OH/2Na+ ions to AuCl4/2H+/Cl ions must be equal to or larger than 300%. In the ‘‘sweet zone’’ ($300%), the same UV-Vis spectrum was obtained for gold nanoclusters (d < 2 nm) in hexane.
This journal is ª The Royal Society of Chemistry 2012
Nanoscale, 2012, 4, 4091–4094 | 4093
In conclusion, we present a new synthesis method for alkanethiolate-coated gold nanoclusters (d < 2 nm, <250 atoms per nanocluster), which takes only 2 minutes and can be easily reproduced by any other interested group. This ultrafast synthesis is made possible by minimizing the reaction of Au(III) ions with thiol molecules before their reduction to nanoclusters, which is also critical for controlling d < 2 nm. In addition, no post-synthesis cleaning is needed, since two immiscible solvents of methanol and hexane are used for easy and fast separation of the final product of gold nanoclusters in hexane from the reaction byproducts in mostly methanol. The only impurity in the hexane solution of gold nanoclusters is free, unreacted alkanethiol molecules at a very low level. Post-synthesis size-filtering and scaling up could be achieved via ultracentrifugation and a continuous flow process, respectively.
Acknowledgements This work was supported by Rensselaer Polytechnic Institute’s startup fund. A.D. acknowledges support from University of Mississippi (UM) start-up fund, NSF 0903787, and UM’s College of Liberal Arts for MALDI-TOF acquisition.
Notes and references 1 M.-C. Daniel and D. Astruc, Chem. Rev., 2004, 104, 293. 2 M. Haruta, Gold Bull., 2004, 37, 27; J. Huang, T. Akita, J. Faye, T. Fujitani, T. Takei and M. Haruta, Angew. Chem., Int. Ed., 2009, 48, 7862; J. K. Edwards, B. Solsona, N. E. Ntainjua, A. F. Carley, A. A. Herzing, C. J. Kiely and G. J. Hutchings, Science, 2009, 323, 1037. 3 W. Jiang, B. Y. S. Kim, J. T. Rutka and W. C. Chan, Nat. Nanotechnol., 2008, 3, 145; C. K. Kim, P. Ghosh, C. Pagliuca, Z. J. Zhu, S. Menichetti and V. M. Rotello, J. Am. Chem. Soc., 2009, 131, 1360; A. Verma, O. Uzun, Y. Hu, Y. Hu, H. S. Hna, N. Watson, S. Chen, D. J. Irvine and F. Stellacci, Nat. Mater., 2008, 7, 588. 4 R. W. Murray, Chem. Rev., 2008, 108, 2688; S. W. Boettcher, N. C. Strandwitz, M. Schierhorn, N. Lock, M. C. Lonergan and G. D. Stucky, Nat. Mater., 2007, 6, 592. 5 M. M. Alvarez, J. T. Khoury, T. G. Schaaff, M. N. Shafigullin, I. Vezmar and R. L. Whetten, J. Phys. Chem. B, 1997, 101, 3706;
4094 | Nanoscale, 2012, 4, 4091–4094
6 7 8
9
10 11 12
13 14 15 16 17 18 19 20 21 22
M. J. Hostetler, J. E. Wingate, C.-J. Zhong, J. E. Harris, R. W. Vachet, M. R. Clark, J. D. Londono, S. J. Green, J. J. Stokes, G. D. Wignall, G. L. Glish, M. D. Porter, N. D. Evans and R. W. Murray, Langmuir, 1998, 14, 17. M. Brust, M. Walker, D. Bethell, D. J. Schiffrin and R. Whyman, J. Chem. Soc., Chem. Commun., 1994, 801. M. M. Alvarez, J. T. Khoury, T. G. Schaaff, M. N. Shafigullin, I. Vezmar and R. L. Whetten, Chem. Phys. Lett., 1997, 266, 91. P. D. Jadzinsky, G. Calero, C. J. Ackerson, D. A. Bushnell and R. D. Kornberg, Science, 2007, 318, 430; M. W. Heaven, A. Dass, P. S. White, K. M. Holt and R. W. Murray, J. Am. Chem. Soc., 2008, 130, 3754; M. Zhu, E. Lanni, N. Garg, M. E. Bier and R. Jin, J. Am. Chem. Soc., 2008, 130, 1138. J. F. Parker, J. E. F. Weaver, F. McCallum, C. A. Fields-Zinna and R. W. Murray, Langmuir, 2010, 26, 13650; R. Jin, H. Qian, Z. Wu, Y. Zhu, M. Zhu, A. Mohanty and N. Garg, J. Phys. Chem. Lett., 2010, 1, 2903. C. J. Ackerson, P. D. Jadzinsky, J. Z. Sexton, D. A. Bushnell and R. D. Kornberg, Bioconjugate Chem., 2010, 21, 214. Y. Levi-Kalisman, P. D. Jadzinsky, N. Kalisman, H. Tsunoyama, T. Tsukuda, D. A. Bushnell and R. D. Kornberg, J. Am. Chem. Soc., 2011, 133, 2976. Y. Shichibu, Y. Negishi, H. Tsunoyama, M. Kanehara, T. Teranishi and T. Tsukuda, Small, 2007, 3, 835; A. C. Dharmaratne, T. Krick and A. Dass, J. Am. Chem. Soc., 2009, 131, 13604. M. N. Martin, J. I. Basham, P. Chando and S.-K. Eah, Langmuir, 2010, 26, 7410. J. Polte, R. Erler, A. F. Th€ unemann, S. Sokolov, T. T. Ahner, K. Rademann, F. Emmerling and R. Kraehnert, ACS Nano, 2010, 4, 1076. C. J. Ackerson, P. D. Jadzinsky and R. D. Kornberg, J. Am. Chem. Soc., 2005, 127, 6550. T. Yonezawa, M. Sutoh and T. Kunitake, Chem. Lett., 1997, 619. J.-J. Max and C. Chapados, J. Chem. Phys., 2008, 128, 224512. A. Dass, A. Stevenson, G. R. Dubay, J. B. Tracy and R. W. Murray, J. Am. Chem. Soc., 2008, 130, 5940. R. P. Carney, J. Y. Kim, H. F. Qian, R. Jin, H. Mehenni, F. Stellacci and O. M. Bakr, Nat. Commun., 2011, 2, 335. S. Knoppe, J. Boudon, I. Dolamic, A. Dass and T. B€ urgi, Anal. Chem., 2011, 83, 5056. J. Kim, K. Lema, M. Ukaigwe and D. Lee, Langmuir, 2007, 23, 7853; Y. Negishi, N. K. Chaki, Y. Shichibu, R. L. Whetten and T. Tsukuda, J. Am. Chem. Soc., 2007, 129, 11322. H. Tsunoyama, N. Ichikuni and T. Tsukuda, Langmuir, 2008, 24, 11327.
This journal is ª The Royal Society of Chemistry 2012
Electronic Supplementary Material (ESI) for Nanoscale This journal is © The Royal Society of Chemistry 2012
Electronic Supplementary Information Ultrafast, 2-minute Synthesis of Monolayer-Protected Gold Nanoclusters (d<2 nm) Matthew N. Martin,a Dawei Li,a Amala Dass,b and Sang-Kee Eah*a Department of Physics, Applied Physics, and Astronomy, Rensselaer Polytechnic Institute, Troy, New York 12180,USA,Department of Chemistry and Biochemistry, University of Mississippi, University, Mississippi 38677, USA * Corresponding author:
[email protected]; phone, (518) 276-3972; fax, (518) 276-6680. Rensselaer Polytechnic Institute. b University of Mississippi.
a
Experimental Details Materials: the following chemicals were used as obtained: HAuCl4·3H2O, NaBH4 granules, 1.0 N HCl solution, 1.0 N NaOH solution, 1-hexanethiol, 1-dodecanethiol, 4-methylbenzenethiol, 3mercaptopropionic acid, and 4-mercaptobenzoic acid from Sigma-Aldrich; methanol, hexane, and toluene from VWR. Deionized (DI) water with resistivity of 18.2 MΩcm was used. 40 mL KG-33 borosilicate glass vials from Kimble-Kontes were used after cleaning with DI water, acetone, hexane, and being heated at 540 °C. No rigorous (or dangerous) cleaning material such as aqua regia or a hydrofluoric/nitric acid mixture was used for glasswares. Preparation of the reagent stock solutions: an aqueous stock solution of 50 mM AuCl4-/2H+/Clions in a glass vial was made by dissolving HAuCl4·3H2O powder with the same molar amount of HCl using a 1.0 N HCl solution, which was stable for more than several months. An aqueous stock solution of 50 mM BH4-/OH-/2Na+ ions in a plastic beaker or a glass vial was made by dissolving NaBH4 granules with the same molar amount of NaOH using a 1.0 N NaOH solution, which was stable for several hours at room temperature and several days if the container was sealed from CO2 molecules in the air. [S1] Therefore, cooling the borohydride solution or using a freshly made borohydride solution was not required. A hexane stock solution of 1-hexanethiol, 1-dodecanethiol, or 4-methylbenzenethiol was made at the concentration of 5.0 µmol per 6.7 mg, or 500 mM. By using these stable stock solutions, ultrafast synthesis could be repeated as many times as possible to check the reproducibility by getting an identical UV-VIS spectrum. Finding the right conditions for controlling the diameter of gold nanoclusters <2 nm could be done fast and easily via many reproducible repetitions. Ultrafast, 2-minute synthesis of gold nanoclusters: First, 6.7 mg of hexane containing 5.0 µmol of 1-hexanethiol was added to 5 g of methanol in a glass vial as shown in Figure 1 of the main text. Next, 100 µL of water containing 5.0 µmol of AuCl4-/2H+/Cl- ions was added to the vial, changing its color to light yellow. Then, 300 µL of water containing 15 µmol of BH4-/OH-/2Na+ ions was added all at once. The solution’s color was changed from light yellow to brown in <1 second, indicating formation of gold nanoclusters and capping (growth stop) by thiol. The solution’s color changed into black within a few tens of seconds due to precipitation of hexanethiolate-coated gold nanoclusters in mostly methanol. After adding 5 g of hexane, the vial was shaken for 30 seconds to extract only hexanethiolate-protected gold nanoclusters to the hexane top phase with nearly 100% efficiency and leave all the reaction byproduct ions in the mostly methanol bottom phase. Thanks to the two immiscible liquids of methanol and hexane, no post-synthesis cleaning was necessary with just one impurity of free, unreacted alkanethiol molecules in the hexane solution of gold nanoclusters. The whole process took <2 minutes. S1
Electronic Supplementary Material (ESI) for Nanoscale This journal is © The Royal Society of Chemistry 2012
The stably dispersed hexanethiolate-coated gold nanoclusters in hexane were checked for their diameter with a UV-Vis spectrophotometer (Shimadzu, double beam). MALDI-TOF mass spectrometry: the hexanethiolate-coated gold nanoclusters were dried by evaporating hexane, sent from RPI to Mississippi, redispersed in hexane, and characterized by MALDITOF mass spectrometry.[S2] Mass-spectra were obtained using a Bruker Autoflex equipped with a nitrogen laser. 3-(4-tert-butylphenyl)-2-methyl-2-propenylidene]malononitrile (DCTB) was used as the matrix with 1:1000 analyte:matrix ratio. 2 μL of the mixture was applied to the target and air dried. Poor solubility of hexanethiolate-protected gold nanoclusters in methanol The mostly methanol solution changes its color from initially brown to black in a few minutes, since hexanethiolate-protected gold nanoclusters are not soluble in methanol. The precipitated gold nanoclusters at the bottom of the glass vial are shown in Fig. 1S. Once “naked” gold nanoclusters are coated with a monolayer of alkanethiolate molecules, the gold cores do not touch each other and therefore the aggregation is reversible. Therefore, they are completely redispersed in a non-polar solvent such as hexane or toluene. We choose the pair of methanol and hexane, since these two solvents are immiscible with each other. Note that methanol is miscible with toluene. Instead of considering methanol as a bad solvent due to the precipitation, we exploited its property as an intermediate solvent between water and non-polar solvents due to reversible aggregation and re-dispersion of monolayerprotected gold nanoclusters. All reactions of gold ions’ reduction, nanocluster formation, and thiolatecoating happen very fast in <1 second, which must be challenging to characterize and understand. A theoretical study for the ultrafast and complex processes might be better than an experimental one for understanding how thiol ligands react with gold ions, neutral gold atoms, and “naked” gold nanoclusters in the mixture of mostly methanol, little water, and very little hexane.
Figure 1S. Photographs (two viewing angles) of precipitated hexanethiolate-protected gold nanoclusters in the mostly methanol mixture several minutes after their formation. These aggregated nanoclusters are completely redispersed in hexane after phase-transfer. Slow reaction of gold ions with 1-hexanethiol in methanol The mixture of 5 g of methanol, 100 mg of water, and 6.7 mg of hexane containing AuCl4-/2H+/Clions and 1-hexanethiol molecules, is of light yellow color due to Au(III) ions as shown in Fig. 2S. However, the same solution changes its color to cloudy white after overnight aging, implying that Au(III) ions react with 1-hexanethiol molecules very slowly. Note that Au(III) ions in water react with water-soluble thiol molecules such as 3-mercaptopropionic acid or 4-mercaptobenzoic acid very fast, changing the aqueous solution’s color from light yellow to cloudy white in <10 seconds. After slow reaction for 5 min, the mostly methanol solution’s light yellow color does not noticeably change. S2
Electronic Supplementary Material (ESI) for Nanoscale This journal is © The Royal Society of Chemistry 2012
However, the diameter of hexanethiolate-coated gold nanocluster/nanoparticles becomes larger, d<3 nm in comparison to d<2 nm for the minimum reaction time, as shown in Fig. 2 of the main text. We also checked that gold nanoclusters (d< 2 nm) don’t form with the cloudy white solution. Unlike other synthesis methods for gold nanoclusters in the literature, minimizing this reaction is key to controlling d<2 nm in our new ultrafast synthesis method. We speculate that “naked” gold nanoclusters are first formed and then later hexanethiolate-coating follows, even though we cannot observe this separation of nucleation from stabilization in time during the ultrafast reaction in <1 sec.
Figure 2S. Photographs of 1-hexanethiol molecules and AuCl4-/2H+/Cl- ions in the mixture of 5 g of methanol, 100 mg of water, and 6.7 mg of hexane right after the addition (left, light yellow) and after overnight reaction (right, cloudy white). On the contrary, the reaction of Au(III) ions with water-soluble thiol molecules in water is much faster and takes <10 sec. “Sweet zone” for complete reduction of all the gold ions As with “naked” gold nanoparticles in water without any stabilizer, we discovered the same “sweet zone” for hexanethiolate-coated gold nanoclusters. In water, the ratio of BH4-/OH-/2Na+ ions to AuCl4/2H+/Cl- ions in water must be equal to or larger than 300 % molar amount of gold ions for simultaneous reduction of all gold ions in <1 second and generating nearly monodisperse “naked” gold nanoparticles without any stabilizer.[S1] The size uniformity may be due to the separation of the initial fast nucleation from the later slow growth via coagulation. As shown in Fig. 4 of the main text, UV-Vis spectra clearly show that the boundary exists between 200 and 300 %, above which we obtained an identical spectrum at both 300 and 400 %. Each borohydride ion (BH4-) with 4 hydride (H-) arms can ideally reduce one Au3+ ion completely. It is tempting to predict that 100 % BH4- ions can reduce all the gold ions. However, the actual reaction stoichiometry is quite complex due to the presence of excess H+ and OHions, which we don’t fully understand, especially the dynamic progression as a function of time. We don’t have a clear understanding why the boundary must be between 200 and 300 %. The photograph of the 100 % case shows that the methanol phase is not completely colorless but faintly yellow, meaning that not all gold ions are reduced to neutral gold atoms and eventually “naked” gold nanoclusters. The UV-Vis spectrum for the 200 % case shows the absorbance at wavelengths >~600 nm becomes larger than that of the identical 300 and 400 % case. This means that there are polydisperse gold nanoclusters/nanoparticles in the hexane phase. In water, polydisperse “naked” gold nanoparticles are obtained for a ratio below the boundary, while nearly monodisperse ones within the “sweet zone”.[S1] We note that ”naked” gold nanoparticles without thiol molecules grow and precipitate very rapidly in a mixture of methanol and water. From this observation of the same “sweet zone”, we infer that the same fastest reaction of nanocluster formation in <1 sec is followed by growth into nanoparticles in water and alkanethiolate-coating with no growth in mostly methanol respectively.
S3
Electronic Supplementary Material (ESI) for Nanoscale This journal is © The Royal Society of Chemistry 2012
Trapped gold nanoclusters/nanoparticles at the methanol-hexane interface due to excess thiols Using hexanethiol molecules more than 100 % molar amount of gold atoms resulted in more trapping of gold nanoclusters/nanoparticles at the methanol-hexane interface, as shown in Fig. 3S. Also, hexanethiol molecules less than 75 % resulted in d<3 nm. Currently, we don’t fully understand why more 1-hexanethiol molecules make the phase-transfer less efficient due to the liquid-liquid interface trapping.
Figure 3S. Using 1-hexanethiol (HT) more than 100 % molar amount of gold atoms causes more gold nanoclusters/nanoparticles to be trapped at the liquid-liquid interface. References [S1] “Charged Gold Nanoparticles in Non-Polar Solvents: 10-min Synthesis and 2D Self-Assembly”, Martin, M. N.; Basham, J. I.; Chando, P.; Eah, S.-K. Langmuir 2010, 26, 7410-7417. http://dx.doi.org/10.1021/la100591h
[S2] “Nanoparticle MALDI-TOF Mass Spectrometry without Fragmentation: Au25(SCH2CH2Ph)18 and Mixed Monolayer Au25(SCH2CH2Ph)18−x(L)x”, Dass, A.; Stevenson, A.; Dubay, G. R.; Tracy, J. B.; Murray, R. W., J. Am. Chem. Soc. 2008, 130, 5940-5946. http://dx.doi.org/10.1021/ja710323t
S4
