No abstract

We investigate nitrogen-doping effects in a graphene sheet using a first-principles electronic-structure study in the framework of density-functional theory. As possible configurations of nitrogen impurities in graphene, substitutional nitrogen and pyridine-type defects around a monovacancy and around a divacancy are considered, and their energetics and electronic structures are discussed. The formation-energy calculations suggest that substitutional doping of the nitrogen atom into a graphene sheet is energetically the most favorable among the possible nitrogen-doping configurations. Furthermore, by comparison of the total energy of the pyridine-type defects with that of the substitutional nitrogen defect in graphene, it is revealed that formation of the pyridine-type defects becomes energetically favorable compared with formation of the substitutional nitrogen defect in the presence of a vacancy. From the results of electronic-band-structure calculations, it is found that the nitrogenimpurity states appear around the Fermi level as either acceptorlike or donorlike states, depending on the atomic geometries of the nitrogen impurities in graphene. We also calculate the scanning tunneling microscopy (STM) images associated with impurity-induced electronic states for future experimental identification of nitrogen impurities. The simulated STM images of the three N-doping configurations considered here are found to be strongly dependent on the local density of states around the nitrogen impurity, and therefore the doping configurations should be distinguishable from one another. The similarities and differences of the electronic structures and STM corrugations between N-doped and undoped graphenes are also discussed.

No abstract

No abstract

There is a mistake in the comparison between the jellium and the crystalline electrodes in the discussion of Sec. III. The quantity D denoted in Fig. 5 does not have arbitrary value but instead a definite one. With this proper value of D, the jellium model gives values similar to our full calculations. There should be no misunderstanding that the jellium model is inappropriate. The results and conclusions regarding the crystalline electrodes are unaffected. We thank N. D. Lang and M. Di Ventra 1 for pointing this out. 1 N.D. Lang and M. Di Ventra, Phys. Rev. B 68, 157301 ͑2003͒.