Abstract
The renormalized adiabatic PBE (rAPBE) method has recently been shown to comprise a significant improvement over the random phase approximation (RPA) for total energy calculations of simple solids and molecules. Here we consider the formation energies of 19 group I and II metal oxides and a few transition-metal oxides. The mean absolute error relative to experiments is 0.21 eV and 0.38 eV per oxygen atom for rAPBE and RPA, respectively, and thus the rAPBE method greatly improves the description of metal-oxygen bonds across a wide range of oxides. The failure of the RPA can be partly attributed to the lack of error cancellation between the correlation energy of the oxide on the one hand and the bulk metal and oxygen molecule on the other hand, which are all separately predicted much too negative by the RPA. We ascribe the improved performance of the rAPBE to its significantly better description of absolute correlation energies which reduces the need for error cancellation. The rAPBE is just one out of an entire class of renormalized exchange-correlation kernels which should be further investigated.
Highlights
Metal-oxides constitute an important class of inorganic materials that find use in a variety of established and emergent technologies including transparent electrodes, superconductors, microelectronics, batteries, catalysis, photovoltaics, piezoelectrics, and much more
In Refs. 20 and 21 we showed that the renormalized adiabatic LDA and PBE kernels provide in general a vast improvement over the random phase approximation (RPA)
The BEEF-vdW was included here to compare the performance of the ab initio RPA and renormalized adiabatic PBE (rAPBE) methods to a semi-empirical method that includes long range interactions
Summary
Metal-oxides constitute an important class of inorganic materials that find use in a variety of established and emergent technologies including transparent electrodes, superconductors, microelectronics, batteries, catalysis, photovoltaics, piezoelectrics, and much more. This makes the oxides a very interesting and topical class of materials and drives the need for developing more accurate methods for prediction of their properties. As the results are highly sensitive to U, which is typically not calculated ab-initio but fitted to experiments, the LDA+U method does not quality as a predictive theory. In order to achieve better accuracy it is necessary to go beyond these “standard” DFT approaches
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
