Electromagnetically induced modification of gold optical properties.

Abstract

The reflection of light from a metal film, i.e., a mirror, is among the most fundamental and well-understood effects in optics. If the film thickness is greater than the wavelength, reflection is strong and is explained in simple terms by the Fresnel equations. For film thickness much less than the wavelength, reflection is far weaker and more exotic effects become possible. This is especially so if the light illuminating the film is pulsed at the femtosecond time scale. In this work, a phenomenon is proposed where few-femtosecond laser pulses temporarily modify a thin metal film's optical properties via processes that appear linear and classical in nature. By casting a pulsed standing-wave pattern across the metal surface, we consider the possibility that conduction electrons are redistributed to create temporary regions of partly enhanced or reduced density without the excitation of inter-band transitions. The process would constitute a temporary change to the conductivity of the metal, and thus, may be observable as changes to the metal's transmittance and reflectance. In regions where the density is enhanced (reduced), the transmittance is decreased (increased). The concept is termed Electromagnetically Induced Modification (EIM) and is premised on the fact that the pulse length is shorter than the relaxation time of the conduction electrons. An experiment is conducted to test the concept by measuring the change in reflectance and transmittance of gold films with thickness ranging from 20-300 Angstrom. The results show that the film's transmittance decreases only when the standing-wave pattern is present. As the pulse length is increased, or as the film thickness is increased, the changes disappear. The changes show little dependence on the pulse intensity as it is varied by a factor of two. To gain further insight, the Drude theory is used to develop a simplified model for EIM, which qualitatively agrees with the observations. However, neither the experiment nor the model can prove the validity of the EIM concept. As such, an assessment is made for the potential of alternative well-known processes to explain the observations.