S. Vanini, Gabriela F. Cabeza
In this work, we review and analyze in detail several methods used to calculate the energy gap (Eg) without making prior assumptions about the nature of the optical transition (m). Based on absorption spectra obtained from ab initi o calculations within the DFT+U framework, we evaluate the electronic and optical properties using both the Tauc method and derivative-based approaches, including DASF and ILD. These methods are tested on four well-known photocatalytic materials: TiO₂ in its anatase and rutile phases, ZnO in wurtzite phase, ZrO₂ in its monoclinic, tetragonal, and cubic phases, and two oxyhalides (BiOBr and BiOI). For the eight semiconductors studied, the errors in the calculated optical band gaps relative to reported experimental values range from 4 % to 9 % using the Tauc method, and from 4 % to 20 % using methods derived from Cody’s model. Notably, the DASF method yields m values that more accurately reflect allowed transitions (direct or indirect) compared to the ILD method. Among all materials analyzed, ZnO shows the most consistent results for both Eg and m. The optical band gap values are also compared with those obtained from the electronic density of states. The main contribution of this work is the ability to reliably determine both the optical band gap and the nature of the electronic transition directly from absorption spectra calculated via DFT+U, with deviations not exceeding 20 % compared to experimental data. This methodology provides a robust and broadly applicable framework for the optical characterization of a wide variety of semiconducting and photocatalytic materials. • Band gap of TiO 2 , ZnO, ZrO 2 and BiOX were determined from DFT+U absorption spectra. • Errors relative to published values range from 4 % to 9 % (TAUC) and 4–20 % (DASF, ILD). • DASF gives m values closer to allowed transitions (direct/indirect) than ILD. • DFT-based methods offer an excellent alternative for determining Eg and transition type.