Oxygen-rich tetrahedral surface phase on high-temperature rutile VO2(110)T single crystals

First principles studies of the electronic and structural properties of the rutile VO$_{2}$(110) surface and its oxygen-rich terminations

“…As visible from the surface free energy graphs above, the difference between the effective adsorption energy of the V 5 O 15 ring phase and the reaction enthalpy for a V 2 O 5 formation by oxidizing VO 2 is below 0.1 eV in all cases, while the absolute errors in calculated oxidation enthalpies range from 0.1 eV to 0.79 eV. These facts indicate, that the ring terminations are related to a monolayer of V 2 O 5 (001) in agreement with our previous work [18]. This finding also implies that a correct evaluation of the stability for both phases is important to properly describe the energetics of the oxygen rich surface terminations, a criterion which seems to be matched best by spin-polarized PBE+U, SCAN and SCAN+rVV calculations.…”

“…In order to shed some more light on these findings, we present a detailed study of various oxygen adsorption phases and surface reconstructions on rutile VO 2 surfaces, and evaluate the performance of common DFT functionals, including GGA, GGA+U,meta-GGA and meta-GGA+U. In agreement with the experimental findings of Wagner et al [18], we show that the simple regular oxygen adsorption phases considered in the previous calculations [15,16] are less stable than surface terminations with tetrahedral V coordination polyhedra that are structurally and electronically related to a V 2 O 5 (001) monolayer. This findings impose another important constraint to the DFT functional used for the description of off-stoichiometric VO 2 surface terminations, namely a correct description of the stability of the insulating V 2 O 5 phase with respect to VO 2 .…”

“…Since the experimentally observed (2 × 2) superstructure [18] cannot be explained by the surface buckling only, we investigated the stability of a large set of surface reconstructions obtained by optimizing preselected random structures for a number of surface stoichiometries and each employed functional and spin configuration. The detailed procedure is outlined in the Supplementary Information.…”

“…The aforementioned DFT studies using the PBE(+U) approach on the rutile VO 2 (110) surfaces under oxygen-rich conditions [16,15] were done only for regular oxygen adsorption phases. However, a very recent study by Wagner et al [18] indicates the presence of (2 × 2) surface reconstructions with tetrahedrally coordinated V atoms. In order to shed some more light on these findings, we present a detailed study of various oxygen adsorption phases and surface reconstructions on rutile VO 2 surfaces, and evaluate the performance of common DFT functionals, including GGA, GGA+U,meta-GGA and meta-GGA+U.…”

“…Despite its good performance on both bulk phases, the hybrid sc-PBE0 functional surprisingly fails to describe the surface energies of rutile VO 2 which do not converge upon increasing the thickness of the slab and even yield negative values for thicker slabs. On the experimental side, recent studies show the presence of a (2 × 2) reconstruction on the VO 2 (110) surface, accompanied by a change in the surface oxygen content of single crystals [18] and thin films [19,16].…”