Transition metal clusters Fermi level

Of particular interest are the predicted magnetic moments and the question of whether or not isomers of transition metal clusters can be separated using inhomogeneous magnetic fields. To date, cluster isomers have only been detected via their different chemical reactivity (70-73). One would expect abnormally large magnetic moments for the Ih clusters if they had unusually high density of states at the Fermi level, as has been postulated for aluminum (74)-... 

The other mechanism involves atomic-size roughness (i.e., single adatoms or small adatom clusters), and is caused by electronic transitions between the metal and the adsorbate. One of the possible mechanisms, photoassisted metal to adsorbate charge transfer, is illustrated in Fig. 15.4. It depends on the presence of a vacant, broadened adsorbate orbital above the Fermi level of the metal (cf. Chapter 3). In this process the incident photon of frequency cjq excites an electron in the metal, which subsequently undergoes a virtual transition to the adsorbate orbital, where it excites a molecular vibration of frequency lj. When the electron returns to the Fermi level of the metal, a photon of frequency (u>o — us) is emitted. The presence of the metal adatoms enhances the metal-adsorbate interaction, and hence increases the cross... 

The work functions for low-index surfaces of the 4d transition metals have been calculated by a full-potential linear-muffin-tin-orbital method using a slab geometry (a periodic arrangement of 7-layer metal slabs and 10-layer vacuum slabs) (29), and the results (Fig. 8) agree well with experimental results. This is a considerable improvement with respect to extended Hiickel calculations for slab [30 (see footnote 21) ] or cluster [i0 (see Chapter 3)] geometries, which usually yield values closer to the atomic ionization energies. However, the shape of the DOS curves and the relative position of the Fermi level as found by the extended Hiickel calculations are reasonably similar to those obtained by the more sophisticated methods. Therefore, it seems that the major error is in the determination of the dipole layer potential. (Further analysis of this topic would lead us beyond the scope of this chapter.)... 

Electronic properties of transition metals can be described by a simplified model, which, essentially considers a narrow d sub-band having a somewhat localised, atomic-like nature, while the sp sub-band is wider and mostly delocalised.1 The Fermi level has a major contribution to the density of states from the d band in most typical active phases, such as those including noble metals. The limited size of clusters produces the so-called quantum or size confinement effects. These essentially arise from the presence of discrete, atomic-like electronic states. From the solid state point of view, the electronic states of clusters can be considered as being a superposition of bulk-like states with concomitant increases in oscillator strength.14 This separation of the states is visible in the valence band of metallic clusters synthesised by physical methods3 but is obscured in... 

It is believed that the catalytic activity on transition metal chalcogenides takes place through the interaction of the molecular oxygen with the transition metal d-states. The bands are filled with valence electrons of the atoms up to the Fermi level as shown for Mo4Ru2Seg cluster compounds (Fig. 14.2) [21]. 

The UPS spectra for these clusters reveal the general appearance of a 3-4 eV wide metal d band lying immediately below the Fermi level which is separated by 2-3 eV from an intense peak due to the CO 5intense peak at 3-4 eV below the 5a and l r features is due to the CO 4a levels. These spectra exhibit a remarkable qualitative similarity to the corresponding spectra of CO adsorbed on metal surfaces. More important for the present discussion is that they can be directly compared with the ionization energies obtained from theoretical calculations. For instance, the calculated spectra for a series of tetracobalt carbonyl clusters using the DV-Xa method have been compared with the He(I) UPS spectrum of [Co4(CO)i2]. [150] The transition state procedure [41] used in these calculations also accounted for the final state effects, that is, for the effects of orbital relaxation following the ionization process. This relaxation is typically about 2 eV for the valence orbitals of transition metal carbonyls and the calculated IP s compare well with the experimental ones. In a similar way, ruthenium, osmium, and rhenium carbonyl cluster UPS spectra have been calculated and reproduce the general trends and positions of the... 

---