Optical Scattering and Machine Learning for Microplastics Detection

The presence of microplastics has been identified as a considerable form of pollution in oceans, rivers, and even soil. Considering their high durability to natural decomposition and possible effects on the food chain and human beings, there is an immediate requirement for detecting methods that can produce precise data. Traditional techniques such as spectroscopy and microscopy have created accurate protocols for analysis of composition. But the limitations posed by these methods make them impractical for widespread applications and on-site testing. The purpose of this chapter is to analyze the theory and practice aspects of the optical scattering technology, as a possible tool for detection of microplastics. The interaction between light and particles provides the information concerning such characteristics of particles as size, refractive index, form, environment, and excitation sources. This allows for using non-invasive and readily accessible technologies. Technical development and introduction of machine learning algorithms in recognizing scattered pattern data will be analyzed. Some case studies will provide an example of portable devices utilizing different light sources and machine learning algorithms to deal with some problems of traditional techniques. At the same time, the review will be concentrated on the issue of combination of methods for solving such problems as mixing of materials and verification of results in terms of existing reference chemical techniques.