272883 First Principles Modeling of TiO2 Rutile/Anatase Interfaces
Metal oxides, such as TiO2, are often used as photocatalysts in pollutant removal, water-splitting, or chemical synthesis. Multi-component photocatalyst materials can be more photoactive than single-component catalysts, such as for hydrocarbon decomposition. Mixed-phase TiO2 is an important multi-component photocatalyst, and there are several explanations for its increased activity. One hypothesis is that photoexcited electrons prefer one phase over the other due to a conduction band offset, which separates the electrons and decreases the recombination rate of holes and electrons. Another possibility is that sites at or near the interface between rutile and anatase may serve to trap charge and slow down the recombination process. Four-coordinated Ti (Ti4c) atoms at the interface have recently been suggested as key to the catalytic activity.
Motivated to explain the rationale of mixed-phase TiO2, we used molecular dynamics simulations and density functional theory (DFT) to model interfaces between rutile and anatase. Empirical models with molecular dynamics were used to optimize feasible interface structures between several different low-index surfaces. These interfaces involve thousands of atoms, so modeling at a quantum level is time-consuming, and use of empirical potentials saves computational time for the optimization. Several potential interfacial structures have been identified, and the structure of the interfacial region, which enables the adhesion of the two phases, is elucidated. With viable interfaces identified from the molecular dynamics simulations, we modeled the interfaces at the DFT and DFT+U level in order to obtain information on the electronic bands. We computed the valence and conduction band offsets for an interface with approximately 1500 atoms. Several alignment techniques were used, and all show that rutile’s conduction band lies higher than anatase’s conduction band, in agreement with previous theoretical estimates. We observed the formation of Ti4c atoms at the interface, and these Ti4c atoms can trap electrons within gap states, suggesting an important role of Ti4c atoms during photocatalysis. We further modeled nanoscale interfacial systems with a high concentration of Ti4c in order to assess how increasing the Ti4c concentration may affect photocatalysis. Such systems were created by placing metastable phases of TiO2, with Ti4c as building blocks, in contact with rutile and anatase. Our results provide important insights on TiO2 composite photocatalysts that may guide experimental synthesis of new composite photocatalyst systems.
 T. Ohno, K. Sarukawa, K. Tokieda, and M. Matsumura. “Morphology of a TiO2 Photocatalyst (Degussa, P-25) Consisting of Anatase and Rutile Crystalline Phases.” Journal of Catalysis 203, 2001, 82-86.
 C. Wu, Y. Yue, X. Deng, W. Hua, and Z. Gao. “Investigation on the synergetic effect between anatase and rutile nanoparticles in gas-phase photocatalytic oxidations.” Catalysis Today 93-95, 2004, 863-869.
G. Li, and K. Gray. “The solid–solid interface: Explaining the high and
photocatalytic reactivity of TiO2-based nanocomposite materials.” Chemical
Physics 339, 2007, 173-187.
 G. Li, N. M. Dimitrijevic, L. Chen, J. M. Nichols, T. Rajh, and K. Gray. “The important role of tetrahedral Ti4+ sites in the phase transformation and photocatalytic activity of TiO2 nanocomposites.” Journal of the American Chemical Society 130, 2008, 5402-3.
 P. Deák, B. Aradi, and T. Frauenheim. “Band Lineup and Charge Carrier Separation in Mixed Rutile-Anatase Systems.” The Journal of Physical Chemistry C 115, 2011, 3443-3446.