Fast Characterization of Irregularly Shaped Particles Using Elastic Light Scattering

AIMABLE KALUME, Jessica Arnold, Chuji Wang, Gorden Videen, Yong-Le Pan, U.S. Army Research Laboratory

     Abstract Number: 189
     Working Group: Instrumentation and Methods

Abstract
Elastic light scattering (ELS) is a non-invasive diagnostic technique widely used to characterize small particles, capable of retrieving highly accurate information (size, shape, surface roughness, refractive index, etc.) for spherically shaped particles. However, this practice has had limited application for irregularly shaped particles due to the angular dependence of the scattered light intensity, resulting in an unlimited number of possible scattering patterns for a single particle. Working with synthetic data of ELS patterns, we demonstrated that machine-learning algorithms can classify different irregularly shaped particles fast and with great accuracy (up to 97%). To apply this method toward real world problems, experimental light-scattering patterns of various particles need to be collected. Through our recent works, we presented a novel method for actively controlling circular and/or spin rotational motion of optically trapped airborne particles, by manipulating the trapping laser beam using a cylindrical lens. We demonstrated a method for recording 2D forward-scattering patterns from optically trapped single airborne particles at multiple angular orientations, for a wide range of particle morphologies, paving the way towards the development of a fast characterization technique for irregularly shaped atmospheric particles, by combining ELS experimental data with machine-learning algorithms.