Electrodynamic interaction between two atoms in near-field contact.

A theoretical study of elements of the quantum electrodynamic interaction between two single-electron atoms in near-field contact is presented. The framework of the study is an electromagnetic propagator formalism that allows one to describe the near-field space-time interaction in such a manner that the Einstein causality and lack of photon localizability are manifest. First we set up the atom-field Hamiltonian in the so-called G-gauge, starting from the Coulomb Hamiltonian, and thereafter we show that this leads to a correct propagator description for the retarded part of the transverse electromagnetic field (operator). In the G-gauge approach, renormalization of the two-particle energy level structure stemming from the transverse self-field occurs. The intraparticle renormalization is calculated for a three-level atom (it is trivial for a two-level atom), and the interparticle renormalization, which depends on the atomic separation, is determined for two two-level atoms. The magnitude of the energy renormalization is always finite in our theory because we do not consider the atoms to be point-like objects from an electromagnetic point of view. Throughout, we relate our G-gauge formalism to the multipole theory often used in studies of interatomic electrodynamics.

Superluminal interactions in near-field optics.

The fact that photons emitted from an electric-dipole active atom cannot be spatially localized better than to the near-field zone of the atom is seen as the origin of genuine superluminality. By means of a simple model dipole current density the general theory is used to demonstrate numerically how superluminality enters the near-field dynamics, and how from a measurement one could be tempted to believe that superluminal propagation effects occur. Furthermore, it is shown how for source-detector distances larger than a pulse length one should be able to divide the pulse into two separate parts: one purely superluminal part arising solely in the non-local generation process of the field, and another part seemingly propagating with superluminal speed. We comment on different velocity analyses, and we argue that the only fundamental velocity entering the problem is the vacuum velocity of light, which in a measurement would appear as the velocity of the trailing edge of the pulse.