Characterizing biomatrix materials using pyrolysis molecular beam mass spectrometer and pattern recognition

摘要

Fingerprinting techniques have been applied to pyrolysis molecular beam mass spectra of a complex biomatix material such as tobacco. Pyrolysis products of three major tobaccos (burley, bright and oriental) and their blends were scanned from m/z 25 to 500 by a molecular beam mass spectrometer (MBMS) in high vacuum and soft ionization (25 eV). The resultant complex mass spectra were further analyzed using multivariate data analysis (pattern recognition) to discover the underlying chemical differences that may not be obvious by comparison of such complex mass spectra. Using the MBMS and pattern recognition techniques, we were able to statistically distinguish among different major tobaccos and their blends and show their characteristic components by only looking at their primary pyrolysis products. Burley was signified by its nitrogen containing products, oriental by its abundant carbohydrate-derived products and bright by its intermediate behavior. Blended tobacco samples of burley and oriental with a ratio of 66:34 resulted in similar product composition to that of bright. Classes of pyrolysis products of a tobacco sample at different reaction conditions were also extracted by using the same analysis. These results indicate that this approach could be used to rapidly predict biomass pyrolysis product distribution.