Classification of chemical compound classes in slow pyrolysis liquids from brown coal using comprehensive gas-chromatography mass-spectrometry

Abstract

The depletion of fossil fuels like crude oil or natural gas in foreseeable future urges the search for alternative fuels. Alternative resources for the production of fuels are biomass or coal, which have already been the feedstock for the chemical industry decades ago. One way for the production of fuels from theses feedstocks is pyrolysis and current research focuses on the influence of process parameters on composition of liquids from pyrolysis and the optimization of the properties necessary for the proposed utilization. To unravel the chemical composition of these oils, high performance instrumental analytical methods like comprehensive gas-chromatography mass-spectrometry (GC×GC–MS) are highly beneficial. Unfortunately obtained data sets are very complex and dedicated interpretation methods are needed. In this study, the classification of about thousand compounds in a GC×GC–MS chromatogram of a brown coal pyrolysis oil is demonstrated by means of linear discriminant analysis. Based on a reference compound training set, the compound classes alkanes, alkenes, thiophenes, and benzothiophenes could be assigned with low classification error. This will help in the understanding of the influence of process parameters and feedstocks on the composition of pyrolysis oils.