This journal is c the Owner Societies 2011 Phys. Chem. Chem. Phys., 2011, 13, 923–926 923 Confinement effects for ionic carriers in SrTiO 3 ultrathin films: first-principles calculations of oxygen vacancies E. A. Kotomin,* ab V. Alexandrov, ac D. Gryaznov, ab R. A. Evarestov d and J. Maier a Received 2nd July 2010, Accepted 2nd November 2010 DOI: 10.1039/c0cp01060j One-dimensional confinement effects are modelled within the hybrid HF-DFT LCAO approach considering neutral and single-charged oxygen vacancies in SrTiO 3 ultrathin films. The calculations reveal that confinement effects are surprisingly short-range in this partly covalent perovskite; already for film thickness of 2–3 nm (and we believe, similar size nanoparticles) only the surface-plane defect properties differ from those in the bulk. This includes a pronounced decrease of the defect formation energy (by B1 eV), a much deeper defect band level and a noticeable change in the electronic density redistribution at the near-surface vacancy site with respect to that in the bulk. The results also show that the size effect pertains to the interactions between the oxygen vacancy and two neighboring titanium atoms and orientation (parallel or perpendicular to the surface) of the Ti–V O –Ti complex. In particular, we predict considerable oxygen vacancy segregation towards the surface. The role of ion transport, which is due to mobile ionic carriers (such as oxygen vacancies or interstitial ions in ionic crystals), is by no means of less importance than the role of electronic carriers as regards solid state electronics. The corresponding field of solid state ionics plays a central role in the context of energy research, more specifically for electrochemical devices that allow the transformation of chemical energy into electrical energy (fuel cells or sensors). In the case of batteries, energy can also be efficiently stored. In the case of electronics, both the trend to minimization as well as the possibility to exploit size effects have led to the development of a new field of nanoelectronics. Analogously, nanoionics deals with properties of ions (static and dynamic) in small systems. 1 The study of one-dimensional confinement is of interest for the understanding of interfacially-controlled ionic hetero- structures as has recently been demonstrated for the ionic conductivity in CaF 2 –BaF 2 films 2 where ionic accumulation and depletion layers overlap. This is a confinement effect on the configuration entropy. The phenomenological treatment is based on the core-space charge model 3–5 that assumes the defect formation energies remain constant when approaching the core except for a very core region. For even smaller sizes, one also expects effects on the formation energy of an individual charge carrier, and hence on the (standard) chemical potential. (In the case of nanoelectronics this is a well known fact.) The size effect on electrons, phonons, excitons or magnons has been the focus of some recent studies for nanomaterials. 6,7 For perovskite-based materials one of the most interesting issues concerns the ferroic (ferroelectric and ferromagnetic) properties. For instance, possible suppression of ferroelectricity in PbTiO 3 ultrathin films with decreasing thickness (reduced down to a single unit cell) has been questioned in the recent synchrotron X-ray study where no thickness limit was finally observed. 8 This fact was additionally confirmed by ab initio calculations which demonstrated the existence of a stable ferroelectricity in PbTiO 3 and BaTiO 3 films as thin as a few unit cells, 9 in contrast to intuitive expectations and the prediction of phenomenological Landau-Ginzburg theory. 10 In this Communication, we address the one-dimensional confinement and investigate how the electronic and energetic properties of an oxygen vacancy change in comparison with the bulk when confined in an ultrathin film. As ionic and electronic effects should be of different extents, the question of confinement of a not-completely ionized point defect, i.e., of a color center, is of particular interest. We have chosen SrTiO 3 as a prototypical crystal for a wide class of ABO 3 perovskite-structured materials with partly covalent chemical bonding and a neutral oxygen vacancy (also called the color F center) therein as a typical ionic carrier in this type of solids. Oxygen vacancies are common defects in non-stoichiometric ABO 3 perovskite family oxides and their behavior in nano-thin films or at a surface has been shown to be important for numerous applications including new high density electronic memories, 11 high-k nano-capacitors, 8,12 magnetoresistence at the interface of non-magnetic oxides, 13 spintronics, 14 solid oxide fuel cells, 15 optoelectronics, 16 etc. The F center is a combination of a missing oxygen ion (oxygen vacancy, i.e. an ionic defect) and two electrons trapped in the vacancy site (polaron or the electronic defect). Thus, energetic confinement effects could arise due to both restricted ionic relaxation around the vacancy and localization of an electronic wave function of the defect in ultrathin films. There were a number of recent studies of the F centers in ABO 3 perovskites (see ref. 17–19 and references therein). Note that in the color center notation the neutral F and charged F + centers correspond to two electrons and one electron trapped by anion vacancy and are denoted also as V O and V  O , respectively, in the traditional Kro¨ger-Vink nomenclature. In order to provide atomistic insight into the problem, we applied the hybrid HF-DFT ab initio method with the B3PW exchange–correlation functional 20 as implemented in the LCAO-based CRYSTAL06 computer package. 21 This approach permits us to correctly reproduce the experimental bandgap and describe the electronic localization as previously illustrated on a number of perovskites. 22 We used the LCAO a Max-Planck-Institut fu ¨r Festko ¨rperforschung, Heisenbergstrasse 1, D-70569, Stuttgart, Germany. E-mail: E.Kotomin@fkf.mpg.de b Institute for Solid State Physics, University of Latvia, 8 Kengaraga street, LV-1063, Riga, Latvia c Department of Chemical Engineering and Materials Science and NEAT ORU, University of California, Davis, California 95616, USA d Department of Quantum Chemistry, St. Petersburg State University, 26 Universitetsky Prospekt, Stary Peterhof 198504, Russia COMMUNICATION www.rsc.org/pccp | Physical Chemistry Chemical Physics Downloaded by Universitaet Osnabrueck on 20 January 2011 Published on 29 November 2010 on http://pubs.rsc.org | doi:10.1039/C0CP01060J View Online