Skip to main content
ARS Home » Research » Publications at this Location » Publication #123881
Title: CLASSIFICATION OF COMMERCIAL CEREAL PRODUCTS USING NIR-FT-RAMAN SPECTROSCOPY

Author
item Himmelsbach, David
item Kays, Sandra
item Barton Ii, Franklin

Submitted to: International Conference on Vibrational Spectroscopy
Publication Type: Abstract Only
Publication Acceptance Date: 8/22/2001
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: The classification of commercial cereal products was accomplished by their near-infrared Fourier-transform Raman (NIR-FT-Raman) spectra and the use of Soft Independent Modeling of Class Analogy (SICMA) classification. Ground cereal samples were placed in low-iron NMR tubes and irradiated with a 1064 nm laser source. Raman scatter was collected using the 180 degree reflective mode at 16 cm**-1 resolution with a Ge liquid nitrogen cooled detector. Data were collected over the Raman shift range of 200-3600 cm**-1. The data were preprocessed prior to modeling using locally developed Array Basic automated baseline correction and normalization algorithms. Separate principal component analysis (PCA) models were developed for each major nutritional component of the cereals. Comparisons of the results between models were accomplished by means of Cooman's plots. This provided a means by which samples could be grouped for classification purposes alone or as selection routine prior to quantitative prediction of the composition of major chemical components (e.g. fiber, sugar, fat or protein). It also enabled classification of cereals based on other added, natural or synthetic, components. The use of Raman spectroscopy in this manner represents a rapid and essentially non-destructive method to classify cereal products based on their nutritional components and provide for verification of food labels.