Designing Nanostructured Ferroelectric Films: Geometry, Polarization and Electronic Properties of the BaTiO₃/Hematite Interface from First Principles

By coupling together a junction-based photovoltaic system and a bulk photovoltaic effect-based material in a nanocomposite thin film device, the well-known limitations^1–3 of both technologies could be overcome. This idea relies on the proven ability of ferroelectrics to influence coupled materials, such as photocatalysts and organic photovoltaics. By using the recently-developed Electronic Lattice Strain procedure,⁴ we identified the BaTiO₃/hematite interface as a promising candidate for the proposed device. Screening was performed in terms of epitaxially-compatible interfaces (to minimize defects) and appropriate band alignment (to minimize charge transfer). To gain insights into the geometry, polarization and electronic properties of the aforementioned system, Density Functional Theory (DFT) modelling is employed. To this purpose, we assume the BaTiO₃(110) surface as substrate on which the hematite (100) surface grows epitaxially strained to it. Within the supercell approach, we are required to employ 5×5 surface unit cells of BaTiO₃(110) and 4×2 units of hematite (100) to reproduce the minimally strained interface (~2.6%).<br/> <br/>Our preliminary tests show that GGA-PBE functional with a Hubbard-type correction to the Ti 3<i>d</i> orbitals, where <i>U</i>_eff = 2.6 eV, can reproduce the spontaneous polarization in BaTiO_3.However, the bandgap is still poorly described: ~30% off with respect to experimental value. In the case of hematite, where <i>U</i>_eff = 3.5 eV is applied to the Fe 3<i>d</i> orbitals, the bandgap is in excellent agreement with experimental reports. Even though bandgaps can be accurately described by DFT via hybrid functionals, such as HSE06, their application for the above heterojunction is impractical due to the size of the interface model which consists of 855 atoms.<br/> <br/>In order to estimate a proper offset from the BaTiO₃/hematite heterojunction, we will employ bulk HSE06 values to correct the band alignments of this interface, which is computed with the PBE+<i>U </i>method. Our scheme is tested with heterojunctions that can be tackled with the HSE06 hybrid functional, ranging from covalent to ionic solids. This scheme might allow us to model not only this complex epitaxially-compatibles interface, but also other heterojunction for novel devices.<br/> <br/><b>REFERENCES</b><br/>1 A. Zenkevich, Y. Matveyev, K. Maksimova, R. Gaynutdinov, A. Tolstikhina and V. Fridkin, <i>Phys. Rev. B - Condens. Matter Mater. Phys.</i>, 2014, <b>90</b>, 161409.<br/>2 K. T. Butler, J. M. Frost and A. Walsh, <i>Energy Environ. Sci.</i>, 2015, 8, 838–848.<br/>3 J. E. Spanier, V. M. Fridkin, A. M. Rappe, A. R. Akbashev, A. Polemi, Y. Qi, Z. Gu, S. M. Young, C. J. Hawley, D. Imbrenda, G. Xiao, A. L. Bennett-Jackson and C. L. Johnson, <i>Nat. Photonics</i>, 2016, <b>10</b>, 611–616.<br/>4 K. T. Butler, Y. Kumagai, F. Oba and A. Walsh, <i>J. Mater. Chem. C</i>, 2016, <b>4</b>, 1149–1158.