Discrimination of Beer Flavours by Analysis of Volatiles Using the Mass Spectrometer as an Electronic Nose
Ewa Sikorska1,6, Jaroslaw Chmielewski1, Tomasz Górecki2, Igor Khmelinskii3, Marek Sikorski4 and Denis De Keukeleire5
1 Faculty of Commodity Science, The Poznan University of Economics, al. Niepodleglosci 10, 60-967 Poznan, Poland.
2 Faculty of Mathematics and Computer Science, A. Mickiewicz University, ul. Umultowska 87, 61-614 Poznan, Poland.
3 Universidade do Algarve, DQB, FCT, Campus de Gambelas, 8005- 139 Faro, Portugal.
4 Faculty of Chemistry, A. Mickiewicz University, Grunwaldzka 60-780 Poznan, Poland.
5 Ghent University, Faculty of Pharmaceutical Sciences, Laboratory of Pharmacognosy and Phytochemistry, Harelbekestraat 72, B-9000 Ghent, Belgium.
6 Corresponding author. Email: ewa.sikorska@ae.poznan.pl
J. Inst. Brew. 113(1), 110–116, 2007 | VIEW ARTICLE
ABSTRACT
Entire mass spectra of beer headspace components were used as fingerprints for beer brand classification and differentiation of beer samples stored under various conditions. Chemometric analysis of the mass spectra allowed for the discrimination of beer brands and for the detection of beer aging and photodegradation. The numeric methods used include unsupervised PCA modelling and discrimination using kNN, LDA, and D-PLS methods.
Key words:
Publication no. G-2007-0424-499 ©2007 The Institute & Guild of Brewing