AVS 53rd International Symposium
    Surface Science Thursday Sessions
       Session SS1-ThM

Paper SS1-ThM7
Point Defect and Adsorbate Identification on TiO@sub 2@(110) by Non-Contact AFM

Thursday, November 16, 2006, 10:00 am, Room 2002

Session: Reactivity of Oxide Surfaces I
Presenter: J.V. Lauritsen, University of Aarhus, Denmark
Authors: J.V. Lauritsen, University of Aarhus, Denmark
G.H. Olesen, University of Aarhus, Denmark
M.C. Christensen, University of Aarhus, Denmark
A.S. Foster, Helsinki University of Technology, Finland
M. Reichling, University of Osnabruck, Germany
F. Besenbacher, University of Aarhus, Denmark
Correspondent: Click to Email
The AFM can provide atomic-scale insight into the surface structure of in principle any type of oxide. In order to apply AFM to study surface chemistry it is essential to understand the contrast patterns associated with surface defects and adsorbates. This is not trivial, since the atomic contrast is determined by interactions between the surface and the tip which is mostly in an undefined state. In this study, we successfully identify defects and adsorbates present on a reduced TiO@sub 2@(110) surface under UHV conditions by means of atom-resolved non-contact AFM images and theoretical simulations. We make use of the fact that we can image the TiO@sub 2@ surface with two well-defined tip states characterized by different short-range interactions reflecting a negatively or a positively charged tip, respectively. The details of the complementary images for the two types of tip allow us to unambiguously identify the Ti@super 4+@ and O@super 2-@ sub-lattices and hence investigate the exact structure and location of defects on the metal oxide surface. A detailed statistical analysis of atom-resolved AFM images shows that three different types of defects can be seen in the first hours after sample preparation. Interestingly, the same defects are imaged either as holes in the oxygen rows or as protrusion located between the Ti rows depending on the tip charge. By comparing the characteristic contrast patterns with detailed AFM simulations for both tips, we provide a qualitative and quantitative interpretation of the AFM images, including the identification of the three types of defects as bridging oxygen vacancies and two types of bridging hydroxyls (OH) resulting from water dissociation, respectively. The observation that the AFM tip can be functionalized in situ to discriminate between chemically different elements of an oxide surface is interesting and may provide an attractive method to perform adsorbate identification on other interesting metal-oxide systems.