Light scattering in strongly scattering media: Multiple scattering and weak localization

TLDR

Interest in interference effects in multiple elastic scattering has revived due to its connection with Anderson localization. The paper develops a rigorous scalar‑wave theory and a diffusion‑based model to analytically describe weak localization (enhanced backscattering) of light in finite slabs. The authors derive closed‑form expressions for angular‑dependent enhanced backscattering intensity using diffusion approximation and perform transmission and reflection experiments on polystyrene‑sphere and TiO₂ suspensions. Experiments measured scattering and transport mean free paths, with backscattering cone shapes agreeing with theory, but observed intensities were 15–20 % lower than the predicted twice‑background value.

Abstract

Recently the interest in interference effects in multiple (elastic) scattering of waves has undergone an important revival due to the discovered connection with Anderson localization. In this paper we discuss a rigorous scalar wave theory as a model to represent the enhanced backscattering (weak localization) of light for finite slabs. In addition, we discuss a general theory based on a diffusion approximation, and the resulting angular-dependent enhanced backscattering intensity will be presented in closed form for finite slabs and for general albedo. New transmission and reflection experiments for strongly scattering media are presented. Two types of liquid suspensions have been used as study object: polystyrene spheres in water and suspensions of ${\mathrm{TiO}}_{2}$ particles in 2-methylpentane-2,4-diol. From these experiments scattering mean free paths and transport mean free paths have been obtained. Relative values for the transport mean free paths could also independently be inferred from the observation of the angular dependence of enhanced (interference) backscattering. The observed shapes and widths of the enhanced backscattering cones are in very good agreement with the calculated values. A less satisfying feature is that the theory predicts a backscattering intensity of twice the background intensity, while the experimental value is some 15--20 % lower.