Nanoparticle model catalysts

Second, apart from single crystals, nanoparticle model catalysts should be employed to better mimic the complex properties of supported metals. Nevertheless, the metal nanoparticles should still exhibit well-defined surface facets to allow more reliable data interpretation and a comparison with single-crystal results. 

Recognition of the differences between single-crystal model catalysts and supported nanoparticles, mentioned above, stimulated the development of nanoparticle model catalysts (14,34,37,62,63,70-83,99). The most straightforward approach to their preparation is to grow metal nanoparticles on a single crystal of the support material. Figure 3c shows a HRTEM image of a Au/MgO model catalyst prepared... 

Fundamental investigations of the interactions of (reacting) gas molecules with single crystal and nanoparticle model catalysts have largely been carried out under UHV, with a number of surface-sensitive techniques, such as LEED, TPD, HREELS, IRAS, AES, XPS, UPS, and others being applied (3,33,34). Unfortunately, these methods typically cannot be used under catalytic reaction conditions (> 1 bar), for example, because of mean free path restrictions of the involved electrons, atoms, or ions. 

Before the introduction of STM, high-resolution (HR-)TEM was the primary technique for determination of the surface structures of nanoparticle model catalysts (14,54,74,77,197,198,211,226-230). For technological catalysts, it is still the only method that provides a direct atomic-scale characterization of metal nanoparticles and of the oxide support (211,231-238). Although TEM is unable to detect adsorbed molecules (in contrast to the methods discussed above), it is briefly mentioned here because HR-TEM was sometimes employed to corroborate STM data characterizing model catalysts and, in particular, to demonstrate the internal... 

C.2.6. PM-IRAS of CO on Nanoparticle Model Catalysts. For completeness, we mention that investigations of CO adsorption on palladium nanoparticle catalysts were also carried out by PM-IRAS. The observed adsorbate species essentially agree with those observed by SFG, and References 175,306,307,453) provide more information. 

These considerations were the incentive to carry out studies on nanoparticle model catalysts, which allow introducing select complex features of technological catalysts, but without having to deal with the full complexity of real systans. In this chapter, the development, properties and selected applications of nanoparticle model catalysts are discussed, with emphasis on combined nanoparticle characterization and kinetic tests. 

Ideally, a model catalyst should be prepared and characterized under clean conditions in order to avoid contamination. The catalyst should then be exposed to impurity-free reactive gases and a surface reaction should be studied by surface-sensitive methods at pressures approaching those of a technological process. These requirements can be met by ultrahigh vacuum (UHV)-grown nanoparticle model catalysts and in situ spectroscopy. 

Rupprechter G (2001) Surface vibrational spectroscopy from ultrahigh vacuum to atmospheric pressure Adsorption and reactions on single crystals and nanoparticle model catalysts monitored by sum frequency generation spectroscopy. Phys Chem Chem Phys 3 4621... 

The common underlying principle was shown in Figure 11.2. The electrochemical potential of electrons jl e(=Ep, the Fermi level) in the metal catalyst is fixed at that of the Fermi level of the support.37 This is valid both for electrochemically promoted model catalysts (left) and for seminconducting or ion-conducting-supported metal nanoparticles (right). 

Besides electronic effects, structure sensitivity phenomena can be understood on the basis of geometric effects. The shape of (metal) nanoparticles is determined by the minimization of the particles free surface energy. According to Wulffs law, this requirement is met if (on condition of thermodynamic equilibrium) for all surfaces that delimit the (crystalline) particle, the ratio between their corresponding energies cr, and their distance to the particle center hi is constant [153]. In (non-model) catalysts, the particles real structure however is furthermore determined by the interaction with the support [154] and by the formation of defects for which Figure 14 shows an example. 

Henry CR. 1998. Surface studies of supported model catalysts. Surf Sci Rep 31 235-325. Henry C. 2003. Adsorption and reaction at supported model catalysts. In Wieckowski A, Savinova ER, Vayenas CG. editors. Catalysis and Electrocatalysis at Nanoparticle Surfaces. New York Marcel Dekker. 

Johanek, V., Schauermann, S., Laurin, M. et al. (2004) On the role of different adsorption and reaction sites on supported nanoparticles during a catalytic reaction NO decomposition on a Pd/alumina model catalyst , J. Phys. Chem. B, 108, 14244. 

As we have seen in the previous chapter, the apparent topography and corrugation of thin oxide films as imaged by STM may vary drastically as a function of the sample bias. This will of course play an important role in the determination of cluster sizes with STM, which will be discussed in the following section. The determination of the size of the metallic nanoparticles on oxide films is a crucial issue in the investigation of model catalysts since the reactivity of the particles may be closely related to their size. Therefore, the investigation of reactions on model catalysts calls for a precise determination of the particle size. If the sizes of the metal particles on an oxidic support are measured by STM, two different effects, which distort the size measurement, have to be taken into account. 

Although model catalyst studies show the maximum possible activity obtainable, practical catalyst systems use nanoparticles of Pt or Pt alloys usually... 

Industrial heterogeneous catalysts and laboratory-scale model catalysts are commonly prepared by first impregnating a support with simple transition metal complexes. Catalytically active metal nanoparticles (NPs) are subsequently prepared through a series of high temperature calcination and / or reduction steps. These methods are relatively inexpensive and can be readily applied to numerous metals and supports however, the NPs are prepared in-situ on the support via processes that are not necessarily well understood. These inherent problems with standard catalyst preparation techniques are considerable drawbacks to studying and understanding complex organic reaction mechanisms over supported catalysts. (4)... 

Bowker M, Stone P, Morrall P, et al. Model catalyst studies of the strong metal-support interaction surface structure identified by STM on Pd nanoparticles on TiO2(110). J Catal. 2005 234 172-81. 

In the model catalysts described so far, the interparticle distances were more or less random, governed by the separation of substrate defects which control the nucleation and growth process of the metal particles (74,101). The position and separation of metal particles can be controlled accurately by electron beam lithography (EBL) (which has also been used to fabricate model catalysts), but the minimum size of the metal aggregates is currently still approximately lOnm. Figure 3g shows an example of a platinum nanoparticle array on Si02 (mean size 28 nm interparticle separation 200 nm) (53,106,107). 

With respect to the Pd/Al203 model catalysts described below, STM was used to examine the structure of the AI2O3 support and the nucleation and growth of metal deposits (e.g.. References (34,63,73,101,215) and references cited therein), providing information about the size, shape, and height of palladium nanoparticles. In some cases, even atomically resolved images of individual palladium nanoparticles were obtained (206). 

Fig. 12. HRTEM image and Fourier transform of a PtsSi nanoparticle observed after heating a Pt/Si02 model catalysts in H2 to 873 K for 1 h 247,248).

---