Structure and properties of surface and subsurface defects in graphite accounting for van der Waals and spin-polarization effects

Abstract

The geometries, formation energies, and diffusion barriers of surface and subsurface intrinsic defects in graphite are calculated using spin-polarized density-functional theory and the generalized gradient approximation with a semiempirical van der Waals ͑vdW͒ correction for dispersion interactions. The calculated formation energies and diffusion barriers of subsurface interstitial ͑I͒ atoms deviate qualitatively and quantitatively from those of surface adatoms. The same trend is found also for subsurface and adatom clusters ͑I 2 ,I 3 ͒. In spite of the semiquantitative agreement on the optimized geometries, the formation energies and diffusion barriers of surface and subsurface vacancies ͑V͒, divacancies ͑VV͒, and intimate ͑I-V͒ Frenkel pairs differ significantly from the values for the analogous defects in the bulk of graphite. This suggests limited transferability of the bulk and subsurface defect models to the surface of graphite. These findings are rationalized in terms of the balance between the covalent and vdW interaction terms at the surface, subsurface, and bulk of graphite. Finally, pairing of individual defects ͑adatoms, I and V͒ is calculated to be energetically advantageous both on the surface and in the subsurface regions. This process is shown to either saturate residual dangling bonds or produce singlet spin states, thus contributing to the quenching of residual spin polarization from damaged graphite surfaces.

The contribution made by lattice vacancies to the Wigner effect in radiation-damaged graphite

“…In their absence, a practical approach to correct the interlayer binding of graphite for GGAs is to modify the energy functional either on an empirical [49,[92][93][94][95], or an ab initio [50,96,97] basis, so that the experimentally observed lattice parameters are reproduced. However, this is based on assumptions which, it has been seen here, are not necessarily correct.…”

Modeling chemisorption of carbon dimer at (8, 0) nanotube

“…"Grafted interlayer bridge" di-interstitial [9], later marked I 2 (7557) [10], is similar to ISW in graphene. However, the alternative structure -splitted di-interstitial -was found [9], later marked I 2 (αβ) [10]. Suppose that the dimer is situated immediately above C-C bond of hexagonal net (basal atomic plane of graphite), so that it forms a 4-member cycle, perpendicular to basal plane.…”

“…Our findings show that b-grafting is ∼ 0.9 eV/dimer lower than h-grafting. In graphite, this is reversed: h-grafting is lower by ∼ 1.4 eV/dimer [9] or ∼ 2.0 eV/dimer [10].…”

“…Although topologically similar, carbon nets of graphene and small-diameter SWNTs have different many-electron properties. The results of calculations of ad-dimer grafting in graphene and graphite [1,9,10] may be extrapolated to large-diameter SWNTs [2,3], but the case of large curvature should especially be considered.…”

“…According to the total-energy theory, in 3-D graphite [9,10], both h-and b-graftings are possible, as having lower energy than independent substrate and dimer, that is also true for SWNT (8,0). But which of them has lower total energy?…”