Dynamical polarizability of graphene with spatial dispersion

We perform a detailed analysis of electronic polarizability of graphene with different theoretical approaches. From Kubo's linear response formalism, we give a general expression of frequency and wave-vector dependent polarizability within the random phase approximation. Four theoretical approaches have been applied to the single-layer graphene and their differences are on the band overlap of wave functions. By comparing with the ab initio calculation, we discuss the validity of methods used in literature. Our results show that the tightbinding method is as good as the time-demanding ab initio approach in calculating the polarizability of graphene. Moreover, due to the special Dirac-cone band structure of graphene, the Dirac model reproduces results of the tight-binding method for energy smaller than 3 eV. For doped graphene, the intraband transitions dominate at low energies and can be described by the Lindhard formula for two-dimensional electron gases. At zero temperature and long-wavelength limit, with the relaxation time approximation, all theoretical methods reduce to a long-wave analytical formula and the intraband contributions agree to the Drude polarizability of graphene. Effects of electrical doping and temperature are also discussed. This work may provide a solid reference for researches and applications of the screening effect of graphene.

Automated summary

In this work, different theoretical approaches were used to analyze the electronic polarizability of graphene. It was found that the tight-binding method is as effective as the ab initio approach in calculating the polarizability of graphene. The special Dirac-cone band structure of graphene allows the Dirac model to reproduce results of the tight-binding method for energy smaller than 3 eV.