Multi-wavelength UV-detection in Capillary Hydrodynamic Fractionation. Data Treatment for an Absolute Estimate of the Particle Size Distribution

Abstract

A new approach is proposed for estimating the particle size distribution (PSD) of hydrophobic colloids by capillary hydrodynamic fractionation (CHDF) based on UV-detection at several wavelengths. At each elution time, the multi-wavelength UV signal is used to estimate the instantaneous PSD at the detector cell by solving the involved inverse problem through an artificial neural network. Then, the global PSD is obtained as a weighted sum of the estimated instantaneous PSDs along the entire elution time interval. With the current approach, the estimation procedure is absolute in the sense that no calibration of diameters is required and the instrumental broadening introduced by the fractionation capillary is automatically compensated for. The proposed method was evaluated on the basis of narrow polystyrene standards, as follows: i) a single standard, to emulate a narrow unimodal PSD; ii) a mixture of three standards of relatively close average diameters, to emulate a broad unimodal PSD; and iii) a mixture of two standards of quite different average diameters, to emulate a bimodal PSD. Experimental results indicate that the new approach is able to produce adequate PSD estimates provided that the particle refractive index is known with a relatively high accuracy.