Determination of polystyrene microsphere suspension concentration by optical coherence tomography and artificial neural network

Abstract

In order to accurately determine the concentration of turbid samples, we compare two optical measurement systems at near-infrared wavelengths, one is optical coherence tomography(OCT) and the other is beam analyzer(BA). The back-scattering and transmitted light were measured and analyzed to determine the concentration of turbid solution prepared with polystyrene microsphere and water. The research shows that the concentration is strongly correlated with both depth-dependence coherent optical signal and beam expansion of transmitted light. By exploiting a suitable model of artificial neural network and dozens of training data, the concentration of sample was determined with an uncertainty of 8% and 7.5 % for OCT and BA respectively. After averaging the results of the two methods, the relative error can be reduced to 6% or better. Although the study focused on particular solution of polystyrene microsphere, the results of the two independent methods represent consistency and effectiveness for determination of optical properties.