NANO LETTERS
Ultrasmall Copper Nanoparticles from a Hydrophobically Immobilized Surfactant Template
2009 Vol. 9, No. 6 2239-2242
Jonathan J. Brege, Christopher E. Hamilton, Christopher A. Crouse, and Andrew R. Barron* Richard E. Smalley Institute for Nanoscale Science and Technology, and Department of Chemistry, Rice UniVersity, Houston, Texas 77005 Received January 9, 2009; Revised Manuscript Received April 14, 2009
ABSTRACT Ultrasmall copper nanoparticles are produced by N,N,N′,N′-tetramethyl-p-phenylenediamine (TMPDA) reduction of aqueous Cu2+ on a hydrophobically immobilized sodium dodecylbenzenesulfonate (SDBS) surfactant template in the presence of sodium citrate at room temperature. Single-walled carbon nanotubes (SWNTs) act as a scaffold controlling the size of the SDBS micelle, which in turn confines a limited number of copper ions near the nanotube surface. TMPDA reduction forms copper nanoparticles as confirmed by X-ray photoelectron spectroscopy and electron diffraction, whose size was determined by atomic force microscopy and transmission electron microscopy to be approximately 2 nm. Particles formed in the absence of the SWNT immobilizer range from 2 to 150 nm.
The ability to work at the nanometer level to purposefully engineer structures with size-dependent properties and functions is the key to nanotechnology. For many applications, small nanoparticles are better than larger ones because they have a higher surface area to volume ratio; however, the synthesis of nanoparticles in the sub-5-nm range offers particular challenges. As the diameter of particles decreases, the surface energy increases which favors the aggregation of small particles or their ripening into larger particles. Although small nanoparticles can be separated from a polydispersed sample, the overall yield is low and size control is limited to the efficacy of separation.1 There is a need therefore for new synthetic methods for the controlled synthesis of uniformly sized nanoparticles that does not require complicated separation techniques. One such approach is the use of micelle systems. While we have found this to be successful for larger nanoparticles,2 there is significant size dispersity when attempting to grow ultrasmall particles due to the dynamic nature of surfactants in aqueous solutions, i.e., different size micelles produces different size particles. Using an organic solvent, and forming a reverse micelle system, can restrict the dynamic nature of the micelles, but the particles thus formed are relatively large.3 We have recently reported that transition metal ions (e.g., Cu2+, Co2+, and Ni2+) are more effective at quenching the fluorescence of sodium dodecyl sulfate (SDS) surfacted single-walled carbon nanotubes (SWNTs) than their group * To whom correspondence should be addressed. E-mail:
[email protected]. URL: www.rice.edu/barron. Tel: (713) 348-5610. 10.1021/nl900080f CCC: $40.75 Published on Web 05/11/2009
2009 American Chemical Society
2 or 12 counterparts.4,5 We have ascribed this trend to a stronger interaction of the transition metal ions with the anionic head groups of the SDS and hence localization of the metal at the SWNT surface. We have also observed that the use of sodium dodecylbenzenesulfonate (SDBS) as the surfactant results in a concentration of metal ions at the surface of the SWNT. These results suggest that if the SDBS-SWNT conjugate could be used as a template for the aggregation of metal ions, then nanoparticles could be formed where the size is controlled by the structure of the SDBS/SWNT complex. Copper and copper oxide nanoparticles are of great interest because they have potential in a wide variety of areas including optical, catalytic, electronic, and antifouling applications.6 Copper nanoparticles have been produced using many methods including using supercritical carbon dioxide,3,7 water-in-oil microemulsions,8 high-temperature decomposition of organometallic precursors,9 and a polyol reduction method.10 Polymer composites containing copper nanoparticles have been shown to possess antibacterial and antifungal properties.11 Copper particles can also be used to fabricate other nanomaterials; e.g., they are used as catalysts for the growth of single-walled carbon nanotubes.12 It is for these reasons that we have investigated the synthesis of copper nanoparticles using SDBS-SWNTs as templating moieties. Single-walled carbon nanotubes are individualized in aqueous solution with SDBS using a modification of previously published procedures13 to obtain a concentration of 24 mg L-1 as determined by UV-vis spectroscopy. To a
Figure 2. TEM image of ultrasmall copper nanoparticles including the selected area electron diffraction (inset).
Figure 1. Representative AFM image (2 µm × 2 µm) and associated height profile of copper nanoparticles formed using SDBS-SWNT templates spin-coated on mica. Figure 3. High-resolution Cu 2p3/2 X-ray photoelectron spectra of the Cu nanoparticles confirming the presence of Cu0.
1% SDBS-SWNT solution was added CuCl2 (0.08 mmol) followed by sodium citrate (0.08 mmol). After equilibrium was reached, a 2-fold excess of N,N,N′,N′-tetramethyl-pphenylenediamine (TMPDA) was used to reduce the copper.3,14 Extraction and purification of the particles was accomplished by the addition of 1-butanol followed by centrifugation at 4400 rpm for 5 min. Removal of the aqueous layer allowed for isolation of the nanoparticles by addition of ethanol and further centrifugation. The copper nanoparticles were characterized by X-ray photoelectron spectroscopy, atomic force microscopy (AFM), transmission electron microscopy (TEM), and UV-vis spectroscopy. A representative AFM image of the resulting nanoparticles is shown in Figure 1. Analysis of the particle heights indicates that all the particles are less than 2 nm in diameter with an average of 0.97 ( 0.50 nm. The size of the nanoparticles is confirmed from TEM images, which show distinct, nonaggregated, crystalline nanoparticles (Figure 2). On the basis of TEM, the sizes all appear to be under 2 nm. The crystallinity of the particles is clear based upon the observed lattice fringes at high resolution as well as the selected area electron diffraction pattern consistent with metallic copper (Figure 2). Small diameter nanoparticles tend to aggregate in solution to lower their surface energy; however, in the present case the addition of sodium citrate stabilizes the particles and prevents them from aggregating 2240
together since no evidence of particle aggregation is observed in either AFM or TEM images. The chemical composition of the nanoparticles was confirmed by high-resolution X-ray photoelectron spectroscopy on a dried sample (Figure 3). The major contribution of the Cu 2p3/2 peak (932.4 eV) is that associated with zerovalent copper,15 consistent with the electron diffraction data. However, the smaller contribution in Figure 3 (934.2 eV) along with additional peaks at 941.6 and 943.7 eV are consistent with CuO. The indicated Cu:CuO ratio of 2.8:1 suggests the Cu nanoparticles have a CuO shell. Attempts to minimize the CuO component through the use of excess reducing agent were not successful suggesting the oxide shell is formed upon exposure to the atmosphere. It has been reported that copper nanoparticles have a characteristic absorption plasmon peak at ca. 570 nm.16 Interestingly, the UV-visible spectrum of Cu nanoparticles prepared herein does not show the presence of such a plasmon; however, this is not inconsistent with the formation of copper nanoparticles for two reasons. First, it has been shown that when a copper chloride or copper oxide layer is present on copper nanoparticles, the plasmon absorption is not observed.16 Second, it has been shown that as the particle size decreases, the prominent band at 570 diminishes. In fact, Nano Lett., Vol. 9, No. 6, 2009
Figure 4. TEM image of copper nanoparticles produced in the absence of SWNTs.
it is predicted that for particles of less than 2 nm no peak should be observed. Thus, the lack of plasmon absorption for our particles is due to a combination of small particle size and a thin copper oxide layer around the particles (cf. Figures 2 and 3). A number of control experiments were performed to confirm the interdependency of the SWNT/SDBS/Cu2+ system. It has been reported that carbon nanotubes are capable of spontaneously reducing metal ions on their sidewalls to produce particles.17 A control experiment was therefore conducted with SDBS-SWNTs, sodium citrate, and CuCl2 (in the absence of reducing agent). No particle formation was observed indicating that the SWNTs do not act as a reducing source but as a template for the surfactant···metal interaction. To confirm the templating role of the SWNT, a control experiment was conducted where Cu2+ was reduced with TMPDA in the presence of SDBS surfactant and sodium citrate, but without the SWNTs. The TEM image in Figure 4 shows that although a few small nanoparticles are produced, a plethora of other diameter particles are also present. Without the SWNT template present to immobilize the surfactant and restrict the number of copper ions confined to an area, uncontrollable reduction proceeds forming a wide dispersity sample. AFM confirms this very wide particle size distribution of 2-150 nm. On the basis of the forgoing, we propose that in the present process, the SWNTs provide a hydrophobic surface onto which the SDBS surfactant forms micelles of a size limited by the diameter of the SWNTs (ca. 0.7-2.0 nm for HiPco samples). The aggregation of Cu2+ ions and/or the growth of the copper nanoparticle are thus limited by the hydrophobically immobilized surfactant. We note that Wang et al. have used SWNTs dispersed in poly(styrenealt-maleic acid) (PSMA) to form a nanotube···metal nanoparticle assembly.18 The particles (3-6 nm) are formed on the PSMA that is adhered to the nanotube Nano Lett., Vol. 9, No. 6, 2009
forming a rigid assembly that is stable for weeks; however, there the particles are not separated from the SWNTs. Multiwalled carbon nanotubes (MWNTs) have been used as templates for the synthesis of Cu and CuO nanoparticles.19,20 The largest size was limited by the inside diameter of the MWNTs; however, a wide range of particle sizes were formed in each case. In conclusion, we have developed a method to make “static” micelles in aqueous solution for the synthesis of metal nanoparticles. We have demonstrated that the formation of a SWNT-surfactant conjugate modifies the size and templating effect of the surfactant as compared with the surfactant alone. As a result ultrasmall (ca. 2 nm) copper nanoparticles may be prepared. The SWNT template technique could be applicable to other nanoparticle syntheses where a reduction in size and narrowing of dispersity are desired. We propose that the concept of using a hydrophobically immobilized SWNT template may be extended to other carbon materials with the potential for creating a range of size and shaped nanoparticles with a high degree of control and unlike those made by other methods. Acknowledgment. Financial support for this work was provided by the Robert A. Welch Foundation and the Air Force Office Scientific Research (FA8650-05-D-5807). We gratefully thank Denise Benoit for assistance in obtaining TEM images. Supporting Information Available: Size distribution of copper nanoparticles, TEM and AFM images, and UV-vis and fluorescence spectra. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Jolivet, J. P.; Barron, A. R. Nanoparticle fabrication. In EnVironmental Nanotechnology: Applications and Impacts of Nanotechnology; Wiesner, M. R., Bottero, J.-Y., Eds.; McGraw-Hill: New York, 2007. (2) Whitsitt, E.; Barron, A. R. J. Colloid Interface Sci. 2005, 287, 318. (3) Williams, G. L.; Vohs, J. K.; Brege, J. J.; Fahlman, B. D. J. Chem. Educ. 2005, 82, 771. (4) Brege, J, J.; Gallaway, C.; Barron, A. R. J. Phys. Chem. C 2007, 111, 17812. (5) Brege, J, J.; Gallaway, C.; Barron, A. R. J. Phys. Chem. C 2009, 113, 4270. (6) Anyaogu, K. C.; Fedorov, A. V.; Neckers, D. C. Langmuir 2008, 24, 4340. (7) Ohde, H.; Hunt, F.; Wai, C. M. Chem. Mater. 2001, 13, 4130. (8) Qi, L.; Ma, J.; Shen, J. J. Colloid Interface Sci. 1997, 186, 498. (9) Crouse, C.; Barron, A. R. J. Mater. Chem 2008, 18, 4146. (10) Park, B. K.; Jeong, S.; Kim, D.; Moon, J.; Lim, S.; Kim, J. S. J. Colloid Interface Sci. 2007, 311, 417. (11) (a) Cioffi, N.; Ditaranto, N.; Torsi, L.; Picca, R. A.; De Giglio, E.; Sabbatini, L.; Novello, L.; Tantillo, G.; Bleve-Zacheo, T.; Zambonin, P. G. Anal. Bioanal. Chem. 2005, 382, 1912. (b) Esteban-Cubillo, A.; Pecharroma′n, C.; Aguilar, E.; Santare′n, J.; Moya, J. S. J. Mater. Sci. 2006, 41, 5208. (12) Zhou, W.; Han, Z.; Wang, J.; Zhang, Y.; Jin, Z.; Sun, X.; Zhang, Y.; Yan, C.; Li, Y. Nano Lett. 2006, 6, 2987. (13) O’Connell, M. J.; Bachilo, S. M.; Huffman, C. B.; Moore, V. C.; Strano, M. S.; Haroz, E. H.; Rialon, K. L.; Boul, P. J.; Noon, W. H.; Kittrell, C.; Ma, J.; Hauge, R. H.; Weisman, R. B.; Smalley, R. E. Science 2002, 297, 593. (14) Similar to other phenylenediamine reducing agents, copper nanoparticle formation occurs from a single electron donation from each of the amine groups of TMPDA leaving TMPDA+-Cl- as the oxidized byproduct. Creighton, J. A.; Eadon, D. G. J. Chem. Soc., Faraday Trans. 1991, 87, 3881. 2241
(15) Park, B. K.; Jeong, S.; Kim, D.; Moon, J.; Lim, S.; Kim, J. S. J. Colloid Interface Sci. 2007, 311, 417. (16) Anno, E.; Tanimoto, M.; Yamaguchi, T. Phys. ReV. B 1988, 38, 3521. (17) Choi, H. C.; Shim, M.; Bangsaruntip, S.; Dai, H. J. Am. Chem. Soc. 2002, 124, 9058. (18) Wang, D.; Li, Z.-C.; Chen, L. J. Am. Chem. Soc. 2006, 128, 15078.
2242
(19) Chen, P.; Wu, X.; Lin, J.; Tan, K. L. J. Phys. Chem. B 1999, 103, 4559. (20) Wu, H.-Q.; Wei, X.-W.; Shao, M.-W.; Gu, J.-S.; Qu, M.-Z. Chem. Phys. Lett. 2002, 364, 152.
NL900080F
Nano Lett., Vol. 9, No. 6, 2009
