A propagator of the dynamically screened Coulomb interaction in the vicinity of a graphene monolayer is calculated using ground-state Kohn-Sham orbitals, and the imaginary part of this propagator is used to calculate the energy-loss rate of a static blinking point charge due to excitation of electronic modes in graphene. Energy loss calculated for all (Q,$\omega${}) modes gives intensities of electronic excitations, including plasmon dispersions in graphene, with low-energy two-dimensional (2D) and high-energy $\pi${}1, $\pi${}2, and $\pi${}+$\sigma${} plasmons. Plasmon energies are in good agreement with experimental results. This spectral analysis also enables us to study the contribution of each plasmon mode to the stopping power and potential induced by a point charge moving parallel to the graphene. We find the bow waves that in pristine graphene appear for higher velocities (v$\ge${}2vF) and predominantly originate from excitation of $\pi${} plasmons. Doping induces extra features which appear for lower v$\approx${}vF velocities and predominantly originate from the excitation of 2D or Drude plasmons.
The electron excitation spectra of the Li-intercalated graphene () are obtained using a high-accuracy ab initio density function theory calculation within the random phase approximation. The low-energy parts of the spectra are analyzed to predict the existence of two quasi-two-dimensional collective modes: a Dirac plasmon similar to the one in the doped graphene but significantly stronger, and an acoustic plasmon.
A transparent and compact method for the calculation of the electromagnetic-field propagator in presence of a thin metallic slab is developed. Electron wave functions for the slab are obtained numerically by using density-functional theory within the local density approximation, and used to construct the slab conductivity tensor. Expressions for the free-photon Green’s function and photon self-energy (i.e., slab conductivity tensor) in terms of electronic wave functions are derived analytically, taking advantage of the symmetry of the problem and separating it in and polarizations. Dyson equation for the polariton propagator is analytically prepared to be solved in two steps: first, solving the Dyson equation with only paramagnetic (nonlocal) part of photon self-energy included, and second, renormalizing such propagator because of its interaction with diamagnetic (local) polarizations. Such approach allows us to take both polarization mechanisms into account as well as their mutual influence. Long-wavelength and quasistatic limits of our results are derived and compared with previous results. Finally, the method is used to calculate spectra of polaritons, i.e., electromagnetic excitations produced by an oscillating dipole placed in the vicinity or inside a metallic slab.
The problem of dynamical screening and local potentials for electrons tunneling in the metal-insulator-metal system is treated in the self-consistent GW approximation. The self-energy is evaluated for electron coupling to the long-wavelength charge-density oscillations. The analytic results for dynamical image potentials, obtained in the semiclassical (local) limit, contain explicitly the competing influences of screening dynamics, tunneling electron energy, and recoil. The role of dynamical image potentials on the tunneling rates for thin barriers is discussed.
The interband and plasmons in pristine graphene and the Dirac plasmon in doped graphene are not applicable, since they are broad or weak, and weakly couple to an external longitudinal or electromagnetic probe. Therefore, the ab initio density functional theory is used to demonstrate that the chemical doping of the graphene by the alkali or alkaline-earth atoms dramatically changes the poor graphene excitation spectrum in the ultraviolet frequency range (4–10 eV). Four prominent modes are detected. Two of them are the intralayer plasmons with square-root dispersion, characteristic of the two-dimensional modes. The remaining two are the interlayer plasmons, very strong in the long-wavelength limit but damped for larger wave vectors. The optical absorption calculations show that the interlayer plasmons are both optically active, which makes these materials suitable for small-organic-molecule sensing. This is particularly intriguing because the optically active two-dimensional plasmons have not been detected in other materials.