Scattering by inhomogeneous particles: microwave analog experiments and comparison to effective medium theories

Abstract

It is common practice to use effective medium theories (EMT) to estimate average, “effective” optical constants of inhomogeneous materials. A variety of EMTs were developed for different internal structures of the medium and for a variety of shapes, size distributions and physical properties of the inhomogeneities. The most popular EMTs (Maxwell Garnett, Bruggeman, Looyenga, etc.) consider inhomogeneities that are much smaller than the wavelength. The so-called extended EMTs were developed to find effective optical constants in the case of inhomogeneities comparable and slightly larger than the wavelength. This paper compares angular distribution and wavelength dependence of intensity and polarization of scattered light obtained from calculations using the most popular EMTs and extended EMTs with the results of microwave analog measurements at the microwave facilities of the University of Florida. We simulated the light scattering by organic grains with silicate inclusions of size parameter x=0.075 (≈0.01 μm) , 0.60 (≈0.1 μm) , and 1.24 (≈0.2 μm ). The conclusion is that for inclusions of a small size and for a small volume fraction of them in the mixture all EMTs yield similar results and show reasonable agreement with experimental results. The accuracy is better for the angular dependencies of the intensity and of the polarization of the scattered light than for their wavelength dependencies. For inhomogeneities comparable and larger than the wavelength extended EMTs work better but for smaller inclusions non-extended EMTs show more accurate results. Large volume fractions of the inclusions in the mixture (>10%) essentially reduce the accuracy of the results obtained with EMTs. Based on our study we do not recommend to use EMTs in the back-scattering domain and at the scattering angles 30°< θ<70°.