We have applied dispersion corrected density functional theory to gauge the reactivity of the most common defects found in graphene. Specifically, we investigated single vacancies, 585 double vacancies, 555–777 reconstructed double vacancies, Stone–Wales defects, and hydrogenated zigzag and armchair edges. We found that the extent to which defects increase reactivity is strongly dependent on the (a) functional group to be attached and (b) number of functional groups attached. For the addition of one H, F, and phenyl groups to defective graphene, we found the following decreasing order of reactivity: single vacancy > hydrogenated zigzag edge > 585 double vacancy > 555–777 reconstructed double vacancy > Stone–Wales > hydrogenated armchair edge > perfect graphene. However, when two phenyl groups are attached, the Stone–Wales defect becomes more reactive than the 585 double vacancy and 555–777 reconstructed double vacancy. The largest increase of reactivity is observed for the functional groups whose binding energy onto perfect graphene is small. In contrast with recent experimental results, we determined that the reactivity of edges in comparison with perfect graphene is much higher than the reported value. When two groups are attached onto a 585, 555–777, or Stones–Wales defect, they prefer to be paired on the same CC bond on opposite sides of the sheet. However, for the single vacancy, this is not the observed behavior as the preferred addition sites are those carbon atoms that were previously bonded to the missing carbon.

The covalent functionalization of graphene with nitrene radicals has been investigated employing first principles calculations. Perfect graphene is very reactive against nitrene radicals and the binding energy per NH group is significantly increased when the nitrene groups are agglomerated. For bilayer graphene, we determined that the presence of a second layer does not affect the reactivity of the upper layer from a thermodynamical stand point. High levels of functionalization are needed to open a band gap in nitrene-modified graphene. At the LDA, GGA, LDA+U and HSE06 levels, we did not observe band gap opening even for the addition of one NH group per 32 carbons. This result is in contrast with a recent experimental study that attributed the band gap opening of epitaxial graphene to the adsorption of one nitrene radical per 53 carbon atoms. The small amount of adsorbed nitrene radicals is also in contrast with our results that predicted a large reactivity. We attribute this discrepancy to the large size of trimethylsilane that inhibited the agglomeration of nitrene radicals. Thus, it is possible to control the number of nitrene groups added just by varying the size of the functional group attached to nitrogen. For small functional groups like NH, it is feasible that the synthesis of 100% functionalized graphene is feasible because the binding energy per NH is duplicated with respect to the isolated addition. The addition of NH groups to graphene is more favorable when the functionalized CC bonds are broken. However, the structure with the disrupted CC bonds may prove difficult to synthesize because the break of CC bonds is not favorable at lower levels of functionalization.

No abstract

No abstract