Microstructures in Rough Metal Surfaces: Electromagnetic Mechanism in Surface-Enhanced Raman Spectroscopy

Abstract

One of the two fundamental mechanisms underlying Surface-Enhanced Raman Scattering (SERS) is the existence of large electromagnetic (EM) fields in the vicinity of the rough metal substrates that are used as substrates. Surface roughness below the micron scale plays a relevant role in this process, due to the roughness-induced excitation of surface-plasmon polaritons. Since in many scattering configurations dipolar and/or electrostatic approximations cannot be employed, we study this EM mechanism from the rigorous standpoint of classical Maxwell equations. By means of numerical simulation calculations based on the Green’s theorem integral equation formulation, the linearly polarized electro- magnetic field scattered from one-dimensional, randomly rough metal surfaces is obtained. In particular, Ag, Au, and Cu surfaces are considered possessing fractal properties analogous to those observed in colloidal aggregates or coldly deposited films commonly used in SERS experiments. We analyze the influence of the roughness parameters on the near field intensity. The enhancement factor of the SERS signal is assumed to be proportional to the square of that of the near field intensity at the pump frequency; in light of the random nature of the roughness, the analysis is performed on the probability density function of the enhancement factor. The optimum pump frequency is obtained from the spectral dependence of both the average field enhancement and the absorption. In addition to the near field intensity calculations, the far field scattered from such random self-affine fractals is studied, revealing interesting features in the angular distribution, such as incoherent peaks at the specular direction, and weak backscattering peaks for the rougher surfaces that also yield large near field enhancements.