Materials containing supported metal nanoparticles (diameter » 1-50 nm) play critical roles in most consumer products through their uses in polymer (plastics) production, petroleum refining (to make gasoline), bulk and fine chemicals production, and food production [1]. Their technological and economic importance are likely to increase significantly in the future due to applications as emissions control catalysts (such as the automotive catalytic converter, the most widespread consumer application of a heterogeneous catalyst) and extensive use in fuel cells (many fuel cell configurations require 3 or more heterogeneous catalysts). The catalysts used in these applications generally contain two or more metals (thus are bimetallic) along with other dopants that are beneficial to catalyst performance. In spite of the technological importance of bimetallic catalysts, relatively little is known about how catalyst performance is affected by properties such as metal particle size and composition. This is largely due to the preparation methods commonly employed, which result in ill-defined catalysts with wide ranging metal particle sizes and unknown compositions.
In the long term, we are working to understand how factors such as particle size and composition affect catalytic performance with the goal of developing routes to make the next generation of catalysts. In order to do this, it is critical that we learn to prepare well defined and controllable catalysts. Consequently, our most immediate and pressing research goals are focused on the first (and arguably most critical) step: preparing supported bimetallic nanoparticles of known and controllable size and composition. Specifically, the objectives are to prepare a series of Pt-M (M = Ru, Au) particles containing a controlled number of metal atoms (e.g. 12, 25, 50, 100) at designated metal ratios (e.g. 100, 80, 60, 40, 20, 0% Pt).
Some
General Background Information on Supported Metal Catalysts
Supported
metal catalysts are generally comprised of small metal particles (1 – 100 nm
in diameter) dispersed on high surface area carrier (e.g. silica).
Modern catalyst formulations often include an “active” metal (e.g.
Pt) combined with other “less active” metals (e.g. Ru, Au).
The second metal can enhance catalyst performance in several ways, such
as promote desired reactions, prevent undesirable side reactions, or enhance
catalyst longevity.
Conventional and commercial bimetallic catalysts are generally prepared via wetness impregnation techniques. For example, metal salt precursors (e.g. H2PtCl6 and HAuCl4) are dissolved in just enough solvent (typically water) to fill the pores of the carrier material (the support). The water is then evaporated, leaving only the metal salts on the support. The metal salts are then activated with high temperature treatments under flowing air or hydrogen. These treatments serve to remove chlorine (a catalyst poison) and produce metal particles. The resulting metal particles may have anywhere from 10 – 100,000 atoms, depending on the metal(s), support, and activation step(s) used.
Because the metal particles are prepared on the support, there is relatively little control over particle properties (see Figure 1 for a cartoon) and evaluation of particle compositions is difficult at best. It is often difficult to consistently prepare catalysts with the same properties; indeed, the “lack of reproducibility of catalytic activity on a sample to sample basis using portions drawn from the same bottle is a humiliating reminder of the lack of control which we have over the complex chemistry involved” [1] . Further, the size and stoichiometry (metal:metal ratio) of individual particles often vary widely within a given sample. This means that there are literally thousands of different kinds of particles within a sample, each of which may have unique catalytic properties. The inhomogenaity inherent to traditional catalyst preparation routes dramatically complicates their characterization, which, in turn, poses a significant hurdle to understanding how reactions occur over these materials.
In
order to evaluate and understand what structural features are important in the
best catalysts, we must first learn how to consistently prepare catalysts with
known and reproducible particle sizes and compositions.
Using bimetallic molecular cluster compounds
(molecules that have several metal atoms with metal-metal bonds) as catalyst
precursors has allowed for the deposition of particles that are initially
consistent in both composition and structure
[2,3]
.
However, metal ratios are set by cluster stoichiometry and cannot be
systematically varied without an extensive molecular library.
Cluster preparation is time consuming and expensive (it can take months
to make prepare and characterize a single new cluster), so such molecular
libraries seldom are available and are not practical for this application.
Ideally, the best features of the traditional and cluster routes might be
combined: metal ratios must be readily varied; simultaneously, particle size and
composition must be both variable and controllable.
Dendrimers are hyperbranched polymers that emanate from a core and, like a south Texas live oak, ramify outward with each branching unit [4] . StarburstÒ PAMAM (polyamidoamine) dendrimers (see Figures 2 and 3) are a specific class of commercially available dendrimers that have repeating amine/amide branching units. PAMAM dendrimers are defined by their generation - the number of times the amine/amide branches have been added. Low generation dendrimers (G-0) are essentially simple organic molecules. Higher generation dendrimers (G-3 and larger) develop complicated three dimensional architectures as the branches intertwine. Similar to the older, broad live oaks on Trinity’s campus, higher generation dendrimers form an exterior chemical canopy while maintaining open spaces in the interior. This creates an ideal environment for trapping guest species [5-7] that can be prepared inside the dendrimer. (Note: for some cool dendrimer pictures that are much better than what I could ever draw, go to www.dendritech.com)
Recently, dendrimers have been used as templates to prepare and stabilize small, reduced metal particles [7,8] in solution. These materials, termed “nanocomposites” are prepared via a two-step synthesis (figure 3). First, metal ions (e.g. Cu2+, Pd2+, Pt2+, Au3+) are complexed to the interior amine groups (located wherever two branches diverge) of a PAMAM dendrimer. They are subsequently reduced to yield dendrimer encapsulated metal nanoparticles [7-9] . The resulting particle size distributions are narrow because individual dendrimer molecules act as templates for individual metal particles; i.e., the number of metal atoms in the particle is determined by the number of metal ions initially attached to the interior amine groups. Beyond their role as templates, the dendrimers continue to function as stabilizers for the resulting particles, preventing agglomeration. The metal ions can initially migrate to the dendrimer’s interior because they are relatively small (< 1 nm); however, the nanocomposites are stable for long periods of time because the metal particles (³ 1 nm) are trapped within the dendrimer’s organic framework [7-9] . In the absence of a dendrimer or other stabilizer, the addition of a reducing agent to a metal salt solution results in complete particle agglomeration ... to where the metal precipitates out as what’s called a “black” or forms a thin mirror on the inside of the glassware.
Figure 3. Scheme for Nanocomposite Preparation
The application of commercially available PAMAM dendrimers to catalyst preparation affords the opportunity for unprecedented control over supported bimetallic particle catalysts. Our initial efforts are focusing on the preparation and characterization of bimetallic Pt-Au/Ru nannocomposites. Initial studies will focus on preparing one size of nanoparticles (50 atoms) while varying the Pt:Au/Ru ratio. Appropriate metal precursors ([Pt(NH3)2(NO3)2], H2PtCl4, [Ru(NH3)5Cl]Cl, (NH4)3[Au(NH3)4], [Au(C4H4S)Cl]) have been identified and are either commercially available or readily prepared. Interactions between the precursors in the presence of the dendrimer are unknown and will be evaluated. Each step of the synthetic process will be monitored with UV-Visible spectroscopy in order to determine optimal preparative conditions.
Figure 4. Dendrimer Route to Supported Metal Particle Catalysts
Characterizing the resulting nanocomposites will also be critical and TEM is the only available means of directly measuring metal particle size. We are collaborating with several researchers in the Chemistry and Chemical Engineering Departments at the University of South Carolina in order to routinely perform Transmission Electron Microscopy (TEM) experiments. Further characterization of the metal nanoparticles requires removal of the dendrimer and will be performed at Trinity University. As depicted in figure 4, mono- and bimetallic nanocomposites will be prepared, supported onto silica, and the organic shell thermally removed. The resulting particles will have uniform sizes and known, reproducible synthetic histories. Consequently, characterization of the new materials will provide insight into the surface composition and morphology of the metal particles and assist in evaluating the methods used to prepare the nanocomposites. Suitable procedures for nanocomposite deposition and removal of the organic shell have already been identified [10,11] . Using the group’s newly acquired chemisorption-physisorption instrument, the supported nanoparticles will be characterized with BET surface area analysis and chemisorption techniques, which will allow for the evaluation of surface metal stoichiometries. These data will be compared to overall particle stoichiometry (determined by nanocomposite preparation) and TEM data to provide a complete picture of metal particle size, distribution, and morphology.
References
[1]
Ponec, V., Bond, G.C., "Catalysis by Metals and Alloys"
(B. Delmon, J.T. Yates, (Eds.)): Studies
in Surface Science and Catalysis, Vol. 95, Elsevier, Amsterdam, 1995.
[2] Braunstein, P., Rose, J., in "Heterometallic
Clusters in Catalysis" (I. Bernal (Ed.): Stereochemistry
of Organometallic and Inorganic Compounds, Vol. 3, Elsevier, Amsterdam,
1988, p. 3.
[3] Catalysis by Di- and Polynuclear Metal
Cluster Complexes (R. D. Adams and F. A.Cotton (Eds.)),
VCH Publishers, Inc., New York 1997.
[4] Fisher, M., Vogtle, F, "Dendrimers:
From Design to Application - A Progress Report" Angew.
Chem. Int. Ed. 38, 884 (1999).
[5] Tan, N.C.B., Balogh, L., Trevino, S.F.,
Tomalia, D.A., Li, J.S., "A small angle scattering study of dendrimer-copper
sulfide nanocomposites” Polymer 40,
2537-2545 (1999).
[6] Cooper, A.I., Londono, J.D., Wignall, G.,
McClain, J.B., Samulski, E.T., Lin, J.S., Dobrynin, A., Rubinstein, M., Burke,
A.L.C., Frechet, J.M.J., DeSimone, J.M., "Extraction of a hydrophilic
compound from water into liquid CO2 using dendritic surfactants" Nature
389, 368-371 (1997).
[7] Crooks, R.M., Zhao, M.Q., Sun, L., Chechik,
V., Yeung, L.K., "Dendrimer-encapsulated metal nanoparticles: Synthesis, characterization, and applications to catalysis"
Acc. Chem. Res. 34, 181-190 (2001).
[8] Crooks, R.M., Lemon, B.I., Sun, L., Yeung,
L.K., Zhao, M.Q., "Dendrimer-encapsulated metals and semiconductors:
Synthesis, characterization, and applications" Topics
in Current Chemistry 212, 81-135
(2001).
[9] Grohn, F., Bauer, B.J., Akpalu, Y.A.,
Jackson, C.L., J, A.E., "Dendrimer templates for the formation of gold
nanoclusters" Macromolecules 33, 6042-6050 (2000).
[10] Chandler, B.D., Amiridis, M.D., manuscript in preparation
(2001).
[11] Gates, B.C., in "Supported Metal Cluster
Catalysts" (G. Ertl, H. Knozinger, J. Weitkamp (Eds.)): Handbook of Heterogeneous Catalysis, Vol. 2, 1997, p.793.
Return to Chandler Research Group Page
Return to Dr. Chandler's Homepage
Return to Trinity Chemistry Homepage
