Polyol Synthesis of Silver Nanoparticles: Use of Chloride and Oxygen

Single-crystal cubes and tetrahedrons of silver with truncated corners/edges have been prepared for the first time in high yields by reducing silver n...
0 downloads 8 Views 740KB Size
NANO LETTERS

Polyol Synthesis of Silver Nanoparticles: Use of Chloride and Oxygen to Promote the Formation of Single-Crystal, Truncated Cubes and Tetrahedrons

2004 Vol. 4, No. 9 1733-1739

Benjamin Wiley,† Thurston Herricks,‡ Yugang Sun,§ and Younan Xia*,§ Department of Chemical Engineering, Department of Materials Science and Engineering, Department of Chemistry, UniVersity of Washington, Seattle, Washington 98195-1700 Received July 9, 2004; Revised Manuscript Received July 22, 2004

ABSTRACT Single-crystal cubes and tetrahedrons of silver with truncated corners/edges have been prepared for the first time in high yields by reducing silver nitrate with ethylene glycol heated to 148 °C in the presence of poly(vinyl pyrrolidone) and a trace amount of sodium chloride. These nanoparticles were relatively monodisperse in size and shape, and their dimensions could be readily controlled in the range of 20 to 80 nm by varying the reaction time and other experimental parameters. We propose that the defects inherent in twinned nuclei of silver led to their selective etching and dissolution by chloride and oxygen (from air), leaving only the single crystalline ones to grow into nanoscale cubes and tetrahedrons.

Introduction. In recent years, intensive research has been devoted to the systematic control of the shape of metallic nanoparticles.1-3 Control of shape, in addition to size, has enabled tuning of the optical, optoelectronic, magnetic, and catalytic properties associated with various metallic nanostructures. To date, solution-based synthesis has been demonstrated as the most successful route to new nanoparticle morphologies. In many cases, however, the products may contain a mix of several shapes and sizes, making it difficult to investigate shape- and size-dependent properties of metallic nanoparticles.4 Furthermore, it is also well-known (though seldom reported in the literature) that the reproducibility of many chemical methods could be greatly affected by trace amounts of contaminants such as ionic species.5 Although there has been substantial study on how physical templates or capping agents (organic polymers or surfactants) control the shape of a nanoparticle, investigation is just beginning on the role(s) of inorganic species in shapecontrolled syntheses. For example, Murphy and co-workers have recently achieved synthesis of silver nanowires by controlling the amount of NaOH added to the reaction solution, without the use of any polymer or surfactant.6 Pileni * Corresponding author. E-mail: [email protected]. † Department of Chemical Engineering. ‡ Department of Materials Science and Engineering. § Department of Chemistry. 10.1021/nl048912c CCC: $27.50 Published on Web 08/07/2004

© 2004 American Chemical Society

and co-workers have also discovered that addition of inorganic salts could control the shape of copper nanostructures in a Cu(AOT)2-isooctane-water system, even though the micelle template in this system remained unchanged.7 These examples nicely demonstrate that introduction of inorganic species may provide a means as powerful as organic surfactants or polymers for controlling the shape of metallic nanoparticles. It is expected that inorganic ions should have a more pronounced influence on the nucleation process than organic surfactants or polymers because of the relatively small sizes associated with nuclei. We and other groups have recently demonstrated an effective approach based on the polyol process for the largescale synthesis of silver or gold nanostructures with controllable shapes.8 It involved the reduction of a precursor (such as silver nitrate for silver) by ethylene glycol at elevated temperatures in the presence of poly(vinyl pyrrolidone) (or PVP). To explore the role(s) of some common ionic species in the shape-controlled synthesis, here we added minute amounts of NaCl and other inorganic salts to the standard polyol synthesis.5 To our surprise, the trace amounts of additives had a dramatic effect on the synthetic pathways, as well as on the morphologies of both nuclei and products. More specifically, single crystalline nanoparticles of silver could be reproducibly obtained in a range of sizes by simply controlling the reaction time. These nanoparticles were

relatively monodisperse in size and were in the shape of cubes and tetrahedrons exhibiting truncated corners and/or edges. We note that it has been very difficult to synthesize single-crystal silver nanoparticles in high yields using solution-phase methods due to the irreversible and energetically favorable twinning of silver nanoparticles at small sizes. It is believed that the approach described here may provide a solution to this long-standing problem. Experimental Section. In each synthesis, 5 mL ethylene glycol (EG, J. T. Baker, 9300-01) was first heated in an oil bath at 148 °C for 4 h to remove trace amounts of water. A syringe pump (KDS-200, Stoelting, Wood Dale, IL) was then used to regulate the simultaneous injection of two 3-mL EG solutions into the hot EG at a rate of 45 mL per hour. One of the solutions contained 0.94 M silver nitrate (Aldrich, 209139-100G), and the other one contained 0.375 M poly(vinyl pyrrolidone) (PVP, Mw ≈ 55 000, Aldrich, 856568100G, the concentration was calculated in terms of the repeating unit) and 0.22 mM NaCl (Fisher, S271-500). Magnetic stirring was applied throughout the entire synthesis. The typical synthesis went through a number of color changes (see the next section) before the color became stable at approximately 46 h. A set of samples were taken in the course of each synthesis using a glass pipet. To minimize temperature perturbations during sampling, the glass pipet was held just above the solution and preheated for 30 s before immersion. The samples were washed with acetone and then with water to remove most of the EG and PVP. During the washing process, the suspension was centrifuged at 16 000 rpm for 10 min or 1 h (depending on whether acetone or water was used) to make sure that most of the silver particles taken from the reaction were recovered. Finally, the sample was dispersed in water for further characterization. A reference synthesis was also performed under the same conditions, except that no NaCl was added to the reaction system. Both TEM and SEM studies (Figure S1, Supporting Information) indicate that most of the nanoparticles in the initial (t ) 4 min) and final products (t ) 60 min) were irregular in shape and characterized by multiple twins. A drop of the aqueous suspension of particles was placed on a piece of silicon wafer (for SEM) or carbon-coated copper grid (Ted Pella, Redding, CA, for TEM) and dried in the fume hood. After that, the sample was transferred into a gravity-fed flow cell and washed for 1 h with deionized water to remove the remaining PVP. Finally, the sample was dried and stored in a vacuum. SEM images were taken using a FEI field-emission microscope (Sirion XL) operated at an accelerating voltage of 20 kV. TEM and electron diffraction studies were performed with a Phillips 420 microscope operated at 120 kV. Photographs were captured using a digital camera (Sony Cybershot). UV-visible extinction spectra were taken at room temperature on a Hewlett-Packard 8452 spectrometer (Palo Alto, CA) using a quartz cuvette with an optical path of 1 cm. All solutions were diluted by 120 times with water before spectral measurements. Results and Discussion. The reaction could be easily followed through its distinctive color changes. Within the first minute of injection, the solution became light yellow 1734

in color, indicating the formation of silver nanoparticles through polyol reduction. The yellow color kept increasing in intensity until t ) 10 min, after which it maintained its appearance for another 20 min. Figure 1A shows the yellowbrown color of the solution at t ) 10 min, and Figure 1B gives a TEM image of the silver particles present in this solution. Note that most of the particles were twinned at this time. As the reaction proceeded, the yellow color started to fade due to the dissolution of silver particles into the solution and the deposition of particles onto the inner wall of the flask. Figure 1C shows that the solution was nearly colorless at t ) 2 h; however, the stirring bar was barely visible due to the deposition of silver nanoparticles on the flask. Figure 1D shows a TEM image of the few particles that remained in the solution, implying that it still contained a mixture of twinned and single crystal particles. For this sample, very few particles were seen under TEM and the image shows the largest number (five) of particles that could be found together. Interestingly, even the particles on the wall of the flask were not safe from dissolution. At t ) 7 h (Figure 1E), only a small area of the flask (as indicated by an arrow) retained the gray coating seen at t ) 2 h. At this point, the reaction mixture returned to its initial colorless appearance. Figure 1F details the extinction spectra taken from solutions sampled between t ) 10 min and t ) 2 h. Note that the surface plasmon resonance (SPR) peak (centered at ∼400 nm) intrinsic to silver nanoparticles gradually decreased in intensity and eventually disappeared as more silver nanoparticles were dissolved into the solution and/or deposited onto the flask. The solution remained colorless (and there was no change in the extinction spectra) until a very light yellow color reappeared at t ) 24 h. This yellow color grew slowly in intensity over the next 20 h until it attained a slight redbrown tint (Figure 2A). Figure 2B indicates that the silver nanoparticles responsible for the renewed yellow color were nearly all single crystalline. Figure 2C shows an SEM image of the sample taken at t ) 45 h, establishing that the reaction mainly contained a mixture of nanoscale cubes and tetrahedrons with truncated corners and edges. The insets in Figure 2C show the electron diffraction patterns (convergent beam) that were recorded by directing the beam perpendicular to the (100) facet of a truncated cube (upper right) and (111) facet of a truncated tetrahedron (lower left), respectively. The spot patterns provide a piece of strong evidence to support the single crystallinity of these particles, even though they had grown to sizes as large as 60 nm. As the particles further grew in size up to ∼80 nm, their (111) and (100) facets could be imaged clearly by SEM (Figure 2D), making fully apparent the truncated nature of the nanoparticles. Figures 2C and 2D also confirm that the monodispersity of these single crystal nanoparticles was essentially retained as they grew in size. It is believed that the uniformity in size and shape can be further improved through optimization of the reaction conditions. To examine the reproducibility of this protocol, the same synthesis was performed more than 25 times, and single-crystal nanoparticles of similar size and Nano Lett., Vol. 4, No. 9, 2004

Figure 1. (A, C, E) Photographs of the reaction at 10 min, 2 h, and 7 h, respectively, showing the formation and subsequent dissolution of silver nanoparticles in the presence of 0.06 mM NaCl and air. The arrow in (E) indicates the remaining small piece of silver film deposited on the inner surface of the flask. TEM studies indicate that both single crystalline and twinned nanoparticles (labeled as sc and tw, respectively) were present in samples taken from a reaction at (B) t ) 10 min and (D) t ) 2 h. (F) UV-vis spectra of six samples taken from the same reaction between t ) 10 min and t ) 120 min, further confirming the disappearance of silver nanoparticles from the solution phase during this period of the reaction.

morphology (including the degree of truncation) were achieved at high yield in every case. Figure 3 shows the UV-vis extinction spectra associated with the silver nanoparticles. Along with the TEM images, these spectra suggest that the single-crystal nanoparticles were much more uniform in size and shape than the twinned ones initially formed. It is worth pointing out that the 20nm silver particles (see Figure 2B for a typical TEM image) exhibited an SPR peak significantly narrower than those of the particles obtained before t ) 120 min, with a decrease in the full width at half-maximum (fwhm) by as much as Nano Lett., Vol. 4, No. 9, 2004

23%. The SPR peak broadened as the particles further grew from 20 to 80 nm, and this change could be attributed to the increase in scattering from the larger particles.9 To elucidate the mechanism by which the high yield of single crystal particles was obtained, we started by examining the mechanism of twin formation. For a particle of relatively small size, the cost of an increase in internal strain is offset by the reduction in surface energy achieved through enlargement of the lower energy, (111) facet. TEM studies on small (