Wet Chemical Synthesis of Monodisperse Colloidal Silver Nanocrystals Using Digestive Ripening † Ravi Shankar, ‡ Bin Bin Wu, and Terry P. Bigioni* Department of Chemistry, UniVersity of Toledo, Toledo, Ohio 43606 ReceiVed: NoVember 29, 2009; ReVised Manuscript ReceiVed: January 27, 2010 Silver metal nanocrystals with passivating ligand shells of dodecanethiol were synthesized by a one-phase wet chemical method. Postsynthesis thermochemical processing, known as digestive ripening, was used to focus the size distribution from a very broad distribution to 5.6 ( 0.4 nm. Chemical analysis of the final Ag nanocrystals showed a metallic core with a densely packed ligand shell. No oxidation was observed despite the entirely aerobic conditions. The size distribution was found to change most dramatically after the two steps proceeding digestive ripening, namely, ligand exchange and precipitation. Imaging also showed that the digestive ripening mechanism of Ag nanocrystals could be significantly different than that for Au, and involved a competitive silver thiolate side product. This synthesis was optimized to produce Ag nanocrystals with little to no silver thiolate materials. Introduction The study of colloidal metal nanoparticles began with the celebrated experiments of Michael Faraday in the mid 1800s. 1 It has only been in the last few decades, however, that their synthesis has been extensively studied. 2-25 Both aqueous and nonaqueous methods have been developed, primarily focused on Au and Ag. Producing narrow size distributions has been an ongoing challenge; however, there have been some notable successes. The most successful aqueous syntheses are largely based on the work of Turkevich and Frens. 2-4 These original experiments used low concentrations of reactive species to promote a brief nucleation burst followed by diffusion-controlled growth kinet- ics, while suppressing secondary nucleation. 26 This led to uniform growth and produced narrow size distributions over a wide range of sizes (16-150 nm). Approaches to synthesizing monodisperse nanoparticles in nonaqueous syntheses have been quite different. Wilcoxon and Brust developed the first methods, 5,6 which used a two-phase approach in order to satisfy the different solvation requirements for the reagents and products. These reactions necessarily had heterogeneous reagent distributions and uncontrolled nucleation, and therefore produced polydisperse nanoparticles. However, the Brust method produced very small (<4 nm) particles with multiply peaked size distributions. 8 Whetten et al. refined this method by using a polymeric Au:alkane thiol precursor as both the source of metal and ligands, which also produced multiply peaked size distributions but with even smaller particle sizes (<2 nm). 9 Each peak in the size distribution could subsequently be separated into monodisperse fractions by size-selective precipitation. 8 It is worth noting that the monodispersity of each fraction is likely due to the thermodynamic stability of particular sizes of nanocrystals. 8 Such size preferences should be most pronounced for the smallest particle sizes. 9,27,28 The emergence of such thermodynamically stable sizes for small oil-soluble nanocrystals provides an interesting contrast with the determinant role of kinetics in controlling the sizes of larger aqueous nanoparticles. More recently, single-phase methods have been developed for synthesizing oil-soluble metal nanocrystals, which produce narrow size distributions without the need for fractionation. 20-22 Like the work of Turkevich and Frens, these methods used different strategies to control the kinetics of nucleation and growth in a homogeneous one-phase system. These strategies consistently produced size distributions with <15% polydisper- sity, although results as good as ∼5% have been reported. 22 An alternative method has been developed that involves a thermochemical postprocessing step to narrow the size distribu- tion, with impressive polydispersities of Au nanocrystals as low as ∼3.6%. 16 First, Au nanocrystals were synthesized with the inverse micelle technique, giving a poor size distribution. Next, a series of processing steps, including refluxing in ligand-rich solvent known as “digestive ripening”, dramatically narrowed the size distribution. 16,29 Although the mechanism for digestive ripening is not understood, it is thought that large particles dissolve and small particles grow, in sharp contrast to Ostwald ripening. 30 A vacuum evaporation technique 17 has also been adopted to produce Ag nanocrystals as a starting material, followed by the same processing approach to reduce the size distribution. In this case, however, the polydispersity was ∼15%. 18 Although this result could be improved, the technique requires specialized vacuum evaporation equipment that is not amenable to wide- spread adoption. To further explore the use of digestive ripening for producing oil-soluble Ag nanocrystals with a narrow size distribution, we have developed an all-wet-chemical synthesis. The general strategy for this synthesis was to prepare spherical nanocrystals by the inverse micelle method, exchange the ionic surfactant for an alkanethiol ligand, and then use digestive ripening to focus the size distribution to ∼7% polydispersity. This strategy was modeled after the successful method developed for Au. 16,31 Our results provide an alternate route to synthesizing mono- disperse Ag nanocrystals and provide some insight into the mechanism of digestive ripening in the Ag system. They also † Part of the “Protected Metallic Clusters, Quantum Wells and Metal- Nanocrystal Molecules Symposium” special issue. ‡ Current Address: Department of Nanoscience and Nanoengineering, South Dakota School of Mines and Technology, 501 East Saint Joseph Street, Rapid City, SD 57701. J. Phys. Chem. C 2010, 114, 15916–15923 15916 10.1021/jp911316e  2010 American Chemical Society Published on Web 02/18/2010