Surface diffraction study of the hydrated hematite ð1  102Þ surface Kunaljeet S. Tanwar a , Cynthia S. Lo a,b , Peter J. Eng c , Jeffrey G. Catalano d , Donald A. Walko e , Gordon E. Brown Jr. f,g , Glenn A. Waychunas h , Anne M. Chaka b , Thomas P. Trainor a, * a Department of Chemistry and Biochemistry, University of Alaska Fairbanks, 900 Yukon Dr., Fairbanks, AK 99775-6160, USA b National Institute of Standards and Technology, Gaithersburg, MD 20899, USA c Consortium for Advanced Radiation Sources, University of Chicago, Chicago, IL 60637, USA d Chemistry Division, Argonne National Laboratory, Argonne, IL 60439, USA e X-ray Science Division, Advanced Photon Source, Argonne National Laboratory, Argonne, IL 60439, USA f Department of Geological and Environmental Sciences, Stanford University, Stanford, CA 94305-2115, USA g Stanford Synchrotron Radiation Laboratory, SLAC, 2575 Sand Hill Road, MS 69, Menlo Park, CA 94025, USA h Earth Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA Received 4 August 2006; accepted for publication 9 October 2006 Available online 10 November 2006 Abstract The structure of the hydroxylated a-Fe 2 O 3 ð1  102Þ surface prepared via a wet chemical and mechanical polishing (CMP) procedure was determined using X-ray crystal truncation rod diffraction. The experimentally determined surface model was compared with theo- retical structures developed from density functional theory (DFT) calculations to identify the most likely protonation states of the sur- face (hydr)oxo moieties. The results show that the hydroxylated CMP-prepared surface differs from an ideal stoichiometric termination due to vacancies of the near surface bulk Fe sites. This result differs from previous ultra high vacuum studies where two stable termi- nations were observed: a stoichiometric (1 · 1) termination and a partially reduced (2 · 1) reconstructed surface. The complementary DFT studies suggest that hydroxylated surfaces are thermodynamically more stable than dehydroxylated surfaces in the presence of water. The results illustrate that the best fit surface model has predominantly three types of (hydr)oxo functional groups exposed at the surface at circumneutral pH: Fe–OH 2 , Fe 2 –OH, and Fe 3 –O and provide a structural basis for interpreting the reactivity of model iron-(hydr)oxide surfaces under aqueous conditions. Ó 2006 Elsevier B.V. All rights reserved. Keywords: Hematite ð1  102Þ; a-Fe 2 O 3 ; Surface X-ray diffraction; Crystal truncation rod, CTR; Density functional theory, DFT; Surface structure; Surface hydroxylation 1. Introduction Iron-(oxyhydr)oxides occur throughout natural aquatic and soil systems and play a key role in dictating the geo- chemistry of natural waters. Due to their high specific sur- face area and reactive (hydr)oxo surface functional groups, iron-(oxyhydr)oxides are often a dominant scavenger of aqueous trace metal(loids) through adsorption reactions. These minerals also act as key substrates in numerous het- erogeneous chemical transformations of importance in the degradation and transformation of environmental contam- inants [1–4]. Iron-(oxyhydr)oxides also play a key role in controlling the availability and geochemical cycling of iron, largely through the redox couple between aqueous Fe(II) and solid phase Fe(III) [5–7]. The macroscopic reactivity of a mineral–water interface system depends upon the coordination chemistry and topographic arrangement of the exposed surface functional groups, which in turn is a 0039-6028/$ - see front matter Ó 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.susc.2006.10.021 * Corresponding author. E-mail address: fftpt@uaf.edu (T.P. Trainor). www.elsevier.com/locate/susc Surface Science 601 (2007) 460–474