Theoretical modelling of the adsorption of thallium and element 113 atoms on gold using two-component density functional methods with effective core potentials

Estimating the adsorption energy of element 113 on a gold surface

“…It is worth emphasizing that these estimates are markedly lower than the previously reported value of 1.8 eV [3] and even slightly lower than that in Ref. [7] (1.3 eV).…”

“…The Au-Au distances were kept at their experimental crystal values, and the distance between the adatom and the surface was adjusted. Possible positions of the E113 atom included the hollow positions, adatom above the gold atom in the 2nd (hollow-2) and 3rd (hollow-3) layer, and the bridge position since other possibilities have been reported as less stable [7].…”

“…The very popular B3LYP functional however, should be used with caution for such systems since compounds of heavy and superheavy elements are not utilised for its calibration. The work [7] presents calculations on various species E113-Au , =4 -21 accounting for relaxation of the gold clusters upon addition of the E113 atom. Geometry optimisation of the adsorption complexes does not significantly affect the binding energies.…”

“…Recent work by Pershina et al reports the binding en-ergy of E113Au to be equal to 1.82 eV [3,4] based on four-component Kramers-restricted relativistic density functional theory (RDFT) calculations with the generalized gradient exchange-correlation functional by Becke and Perdew (B88P86 [5,6]). A fully unrestricted twocomponent (2c) RDFT/B3LYP study within the small core relativistic pseudopotential (RPP) model (shapeconsistent for E113 and energy-adjusted for Au and Tl) [7] predicts a binding energy of only 1.3 eV for the same molecule. The authors of [7] attribute this 0.5 eV discrepancy to different density functional approximations, though it is worth noting that the symmetry restrictions imposed in [3,4] imply the pure like coupling scheme (and therefore too high energies) for free atoms, thus giving rise to some overbinding.…”

“…A fully unrestricted twocomponent (2c) RDFT/B3LYP study within the small core relativistic pseudopotential (RPP) model (shapeconsistent for E113 and energy-adjusted for Au and Tl) [7] predicts a binding energy of only 1.3 eV for the same molecule. The authors of [7] attribute this 0.5 eV discrepancy to different density functional approximations, though it is worth noting that the symmetry restrictions imposed in [3,4] imply the pure like coupling scheme (and therefore too high energies) for free atoms, thus giving rise to some overbinding. The very popular B3LYP functional however, should be used with caution for such systems since compounds of heavy and superheavy elements are not utilised for its calibration.…”