Classification of waste materials using Fourier transform infrared spectroscopy and soft independent modeling of class analogy

Fourier transform infrared (FTIR) spectroscopy combined with multivariate data analysis is under development as a method to classify waste materials. The chemical composition of the sample is reflected by a series of regions of the infrared spectrum which are used as variables for multivariate data analysis. In this study, separated biowaste collection, mechanically-biologically treated waste (MBT-waste), and old landfill materials were collected to provide materials representing different stages of decomposition. A total of 819 FTIR absorbance spectra were recorded. Principal component analyses (PCA) were performed followed by soft independent modeling of class analogy (SIMCA) for classification of waste materials. Strong classification occurred for an analysis where spectral carbonate regions were included, and for another analysis when they were not. The SIMCA model enabled the differentiation and the classification of unknown samples according to the three categories in both cases. The classification methods developed here provide an assessment tool that regulatory authorities may wish to explore when assessing whether a treated waste from an uncertain process can be classed as compost or MBT-waste.