Potential of infrared spectroscopy in combination with extended canonical variate analysis for identifying different paper types

Abstract

The increasing use of secondary fiber in papermaking has led to the production of paper containing a wide range of contaminants. Wastepaper mills need to develop quality control methods for evaluating the incoming wastepaper stock as well as testing the specifications of the final product. The goal of this work is to present a fast and successful methodology for identifying different paper types. In this way, undesirable paper types can be refused, thus improving the runnability of the paper machine and the quality of the paper manufactured. In this work we examine various types of paper using information obtained by an appropriate chemometric treatment of infrared spectral data. For this purpose, we studied a large number of paper sheets of three different types (namely coated, offset and cast-coated) supplied by several paper manufacturers. We recorded Fourier transform infrared (FTIR) spectra with the aid of an attenuated total reflectance (ATR) module and near-infrared (NIR) reflectance spectra by means of fiber optics. Both techniques proved expeditious and required no sample pretreatment. The primary objective of this work was to develop a methodology for the accurate identification of samples of different paper types. For this purpose, we used the chemometric discrimination technique extended canonical variate analysis (ECVA) in combination with the k nearest neighbor (kNN) method to classify samples in the prediction set. Use of the NIR and FTIR techniques under these conditions allowed paper types to be identified with 100% success in prediction samples.

Supervision of Ethylene Propylene Diene M-Class (EPDM) Rubber Vulcanization and Recovery Processes Using Attenuated Total Reflection Fourier Transform Infrared (ATR FT-IR) Spectroscopy and Multivariate Analysis

“…In contrast, chemometric multivariate analysis methods can be successfully applied to process control from spectral acquisitions in a nondestructive, fast, 17 and even real-time manner. Multivariate methods have been widely applied to determine the composition of polymer blends, 18–22 since they allow a more exhaustive interpretation of the infrared spectral data which allows determining and weighting the relative proportions of the components in an elastomeric blend.…”

A comprehensive study of protein–mesoporous–macroporous silica interactions by extended canonical variate analysis of Raman spectra

“…However, CVA cannot directly deal with data containing lesser samples than variables. These cases require a dimensionality reduction step (e.g., PCA) before applying CVA …”

“…These cases require a dimensionality reduction step (e.g., PCA) before applying CVA. [38] ECVA, like the classical CVA, is also a supervised method based on classes separation criteria. The ECVA algorithm is an improvement of CVA because ECVA permits processing data containing more variables than samples.…”

Identification of NR and EPDM Samples by Means of Thermogravimetric Analysis and Multivariate Methods

“…When dealing with datasets containing a large number of variables, it becomes imperative to deal with fast and efficient multivariate processing methods which allow concentrating the analytically significant information in a reduced set of latent variables [23], [25]. Multivariate feature extraction methods calculate a reduced set of latent variables from a large set of original variables, while eliminating most of the noise often present in the original signals.…”

“…Multivariate statistical methods have been successfully applied to post-process spectral acquisitions in many process control applications [23] and to determine the composition of blends of polymers [27]- [31] since they allow an exhaustive mathematical interpretation of the spectral information. In this paper this approach is extended to interpret the TG and DTG data.…”