| Abstract:In this paper, the styrene-butadiene latex (SB-latex) contents in the coating layer of light weight coated (LWC) paper were measured by using an attenuated total reflectance-Fourier transform infrared spectroscopy (ATR-FTIR). The peak areas under the specific absorption peaks at 2922 cm-1, 1028 cm-1, 756 cm-1, and 699 cm-1 were calculated, by using the relative software. So the relation between different contents of SB-latex and the peak areas under those four peaks was obtained. Three different models for predicting the SB-latex content in the coating layer of LWC paper were established by applying the polynomial curve fitting (PCF), partial least-squares regression (PLS regression), and generalized regression neural network (GRNN), respectively. The results showed that the attenuated total reflectance-Fourier transform infrared spectral information of the coating layer of LWC paper can be used to determine the SB-latex content in the coating layer. Among these three methods of predicting the content of SB-latex in the coating layer, the GRNN model has the best predicting effect since it has the minimum error, i.e., the maximum error is -3.39%. |
