Katherine Inzani1 David Scanlon2 Sverre Selbach1

1, NTNU Norwegian University of Science and Technology, Trondheim, , Norway
2, University College London, London, , United Kingdom

The intermediate band (IB) photovoltaic concept has the potential for large efficiency gains beyond single gap solar cells, but the development of suitable materials has so far proved challenging. One route to form a mid-gap band for sub-band gap absorption is to dope a wide gap semiconductor with a suitable dopant at high concentration. A proposed strategy to limit non-radiative recombination is to co-dope with both n-type and p-type species, effectively passivating the recombination sites. Additionally, incorporation of both types of dopant is thought to increase the solubility limits.

TiO2 is earth abundant and non-toxic, and anatase in particular has a wide band gap and high photocatalytic response, making it a promising candidate for an IB host. Here we explore chromium nitrogen co-doping of TiO2 using a computational approach to evaluate the electronic and optical properties. Density functional theory is used to calculate the effect of Cr and N substitution on the anatase and rutile phases of TiO2, using hybrid functionals to replicate experimental band gaps. Transition level diagrams are constructed to give an overview of the defect physics. The density of states and band structure are calculated across a range of dopant concentrations. Optical properties are examined by calculation of the dielectric function and absorption spectra, with local field effects included beyond the random phase approximation (RPA). Finally, the relative stabilities of the doped anatase and rutile phases are considered by including high-order correlations in the adiabatic connection fluctuation–dissipation theory with RPA.

This work provides a fundamental understanding of the effect of Cr-N co-doping on the optoelectronic properties of TiO2 and further to the viability of similar co-doping schemes to deliver an IB material.