Discrimination of three tobacco types (Burley, Virginia and Oriental) by pyrolysis single-photon ionisation–time-of-flight mass spectrometry and advanced statistical methods

Pyrolysis single-photon ionisation (SPI)–time-of-flight mass spectrometry (TOFMS) and statistical analysis techniques have been applied to differentiate three major tobacco types, Burley, Virginia and Oriental, by means of the gas phase. SPI is known as a soft ionisation technique that allows fast a...

Full description

Saved in:
Bibliographic Details
Published in:Analytical and bioanalytical chemistry 2005, Vol.381 (2), p.487-499
Main Authors: Adam, T, Ferge, T, Mitschke, S, Streibel, T, Baker, R. R, Zimmermann, R
Format: Article
Language:English
Subjects:
chemical composition
data collection
discriminant analysis
gases
ionization
ions
mass spectrometry
Mass Spectrometry - methods
monitoring
Multivariate Analysis
Nicotiana - classification
nitrogen
principal component analysis
pyrolysis
smoke
Species Specificity
tobacco
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Pyrolysis single-photon ionisation (SPI)–time-of-flight mass spectrometry (TOFMS) and statistical analysis techniques have been applied to differentiate three major tobacco types, Burley, Virginia and Oriental, by means of the gas phase. SPI is known as a soft ionisation technique that allows fast and comprehensive on-line monitoring of a large variety of aliphatic and aromatic substances without fragmentation of the molecule ions. The tobacco samples were pyrolysed at 800°C in a nitrogen atmosphere. The resulting pyrolysis gas contained signals from more than 70 masses between m/z 5 and 170. Mass spectra obtained were analysed by principal component analysis (PCA) and linear discriminant analysis (LDA) to distinguish between different tobacco types. Prior variable reduction of the data set was carried out by calculation of the Fisher ratios. Results achieved give information about chemical composition and characteristics of the smoke derived from each tobacco type and enable conclusions on plant cultivation to be drawn. Based on LDA, a model for tobacco type recognition of unknown samples was established, which was cross-checked by additional measurements of each tobacco type. Furthermore, first results on the recognition of tobacco mixtures based on principal component regression (PCR) are presented.
ISSN:1618-2642
1618-2650
DOI:10.1007/s00216-004-2935-0