In-line monitoring of filler content in polyethylene terephthalate extrusion using near infrared spectroscopy

Viability of using NIR spectroscopy for in-line composition monitoring of polymeric system was demonstrated. Spectroscopic measurements were made in-line, at the exit port of a single screw extruder and a gear pump. System of polyethylene terephathalate (PET) filled with an inorganic filler, titanium dioxide (TiO₂), was the subject of study. Various mixtures of filled and unfilled PET resins, in increments of 10%, were run through the extrusion system. The spectral data, in the wavelength range of 1200 nm to 2400 nm, was collected on the flowing polymer stream. Presence of inorganic filler caused the scattering of light. The scattering of light resulted in a systematic baseline shift. Changes in the baseline were correlated to the Ti₂ concentration data. Multivariate techniques involving the use of singular value decomposition (SVD) to perform principal component regression (PGR) and partial least squares regression (PLS) were applied to quantitatively determine Ti₂ content in the PET melt stream. The results for both PLS and PGR were almost similar. Qualitative interpretation of loadings and scores for various factors showed meaningful contribution from the first two factors only. Different types of cross validation also suggested two factors as optimal. The standard error of prediction (SEP) values of about 1% were obtained for calibration model using two factors.