Information-statistical approach to inverse optical problem solution for 3D disperse systems with nano and micro particles

Abstract

In nature, a class of objects called three-dimensional (3D) disperse systems (DS) is widely spread. For 3D DS the dispersive mediums are most often air or water, and the disperse phase consists from particles of different origin. In our study the particle sizes of 3D DS can vary from nanometers to about ten micrometers. Earlier [2-7] it was reported that a set of optical parameters of the so-called second class (obtained as a result of processing experimental data without involving any a priori information about 3D DS) is unique for each 3D DS and implicitly reflects characteristics of 3D DS: shape, refractive index of particles, distribution functions of number and mass of particles in size, etc. In other words, the characteristic of any 3D DS can be represented as ND vector in the N-dimensional space of optical parameters of the second class. In this set there are parameters specific to the 3D DS components. On the basis of theory and experiment it is possible to predict these specific parameters for a certain component. In the previous papers it was discussed the 3D DS polymodality problem, optical characterization of 3D DS mixtures, the use of unique optical vectors for monitoring the aggregation process . This part of study is connected with creation of algorithm based on the information-statistical methodology [8-11], which can help in search of the most informative data for particles of interest.