Reports: AC5

Back to Table of Contents

44158-AC5
The Structure and Dynamics of Wet Electrons at Metal Oxide Surfaces

Hrvoje Petek, University of Pittsburgh

Research was performed on the unoccupied electronic states of atoms and molecules on metal and semiconductor surfaces. The unoccupied electronic structure was investigated by experiment (time-resolved two photon photoemission spectroscopy (TR-2PP), and scanning tunneling spectroscopy and theory). The PRF grant primarily supported the theoretical research of Dr. Jin Zhao.

Wet electrons on TiO2. By TR-2PP we discovered the unoccupied states of protic solvent covered metal oxide surfaces. By near UV excitation of H2O or CH3OH covered TiO2(110) rutile surface, we induce charge transfer from defect states in the substrate to the unoccupied states of the solvent. Because electrons are partially solvated by the dangling H atoms of protic solvent molecules, we call the observed resonances in two photon photoemission (2PP) spectra the wet electron states. We explored by theory how the availability of the dangling H atoms affects the energy of the wet electron states on TiO2(110) rutile and (101) anatase surfaces. Because the molecule-surface and intermolecule hydrogen bonding is unfavorable for the anatase surface, the acidity of water bound H atoms is larger and the wet electron state is predicted to be more stable than on the rutile surface. Moreover, excited state optimization methods suggest that the weaker molecule-surface binding makes the solvation of electrons through water reorganization more favorable than on rutile. Experiments to test these theoretical predictions are planned.

The structure of O atom vacancy defects on TiO2(110). By scanning tunneling microscopy, spectroscopy, and theory, we investigated the structure of O atom vacancy defects on TiO2(110) surfaces. Removal of O atoms from the bridging oxygen positions is the main defect on TiO2(110) surfaces that largely determines their chemical reactivity. The creation of a defect leaves two electrons per each O atom in a defect band below the conduction band of TiO2. The structure of these defect states has been controversial. While recent DFT theoretical calculations predict localized and unsymmetric DOS near the defect, our experiments and calculations show unambiguously that the excess charge is delocalized over ~10 proximate Ti sites near the vacancy site. Furthermore, we have investigated how different treatments of exchange and correlation in DFT can influence dramatically the structure of defects in metal oxides.

The universal electronic structure of alkali atoms on noble metals. The chemisorption of alkali atoms on metal surfaces is one of the oldest problems in surface science. Yet the electronic structure and the nature of the chemisorption bond are poorly understood, and have been debated in literature. By 2PP spectroscopy we have discovered that the unoccupied electronic structure of alkali atoms (Li – Cs) on Cu(111) and Ag(111) is independent of the period within the alkali atom group even though the sizes and ionization potentials differ significantly. We have investigated this telling observation by wave packet propagation methods and DFT. We find that the origin of the universal electronic structure is the compensation between the alkali atom size and the ionization potential. Small alkali atoms with a large ionization potential (Li) can approach closer to the metal surface than large alkali atoms with a low ionization potential (Cs). This compensation leads to the stronger destabilization of the s-state of Li than for Cs so that at the chemisorption distance their energies with respect to the reference vacuum level are essentially the same. The universal electronic structure of alkali atoms allows us for the first time to define the effective electronic potentials of alkali atom covered metal surfaces.

The electronic structure of molecular adsorbates on metal surfaces. Building on the success of explaining the unoccupied electronic structure of alkali atoms on noble metals, we have applied similar ideas to explain the molecular adsorption on metal surfaces. By STM dz/dV spectroscopy we investigated the electronic structure of C6F6 and C60 on Cu and Au surfaces. The electronic structure of the anion state of C6F6 strongly depends on the ionization potential of the substrate, the presence of a band gap, and the thickness of the overlayer. We can explain all the observed aspects of the unoccupied electronic structure of C6F6 based on the concepts used to define the electronic structure of alkali atoms, whereas attempts to use the dielectric continuum theory have failed.

Furthermore, we have investigated templated C60 single molecule wire growth on partially oxidized Cu(110 ) surfaces. Incomplete oxidation of Cu(110) surface produces added row oxide domains, which are occasionally cut by bare Cu troughs only one Cu atom wide. These nearly perfect templates assemble C60 molecules into wires that can approach 100 nm in length.

Back to top