Skip to main content
ARS Home » Research » Publications at this Location » Publication #199226
Title: ATR-FOURIER TRANSFORM MID-INFRARED SPECTROSCOPY FOR DETERMINATION OF TRANS FATTY ACIDS IN GROUND CEREAL PRODUCTS WITHOUT OIL EXTRACTION

Author
item KIM, YOOKYUNG
item HIMMELSBACH, DAVID
item KAYS, SANDRA

Submitted to: Pittsburgh Conference
Publication Type: Abstract Only
Publication Acceptance Date: 10/17/2006
Publication Date: 2/25/2007
Citation: Kim, Y., Himmelsbach, D.S., Kays, S.E. 2007. Atr-fourier transform mid-infrared spectroscopy for determination of trans fatty acids in ground cereal products without oil extraction. 2006 Pittsburgh Conference.

Interpretive Summary:

Technical Abstract: Since FDA amended the rules for food labeling to include trans fatty acids (TFA), there has been a continued interest in development of optimized methods for determining TFA in foods. Traditional chromatographic methods are time-consuming and solvent based. Alternatively, spectroscopic methods involve FTIR spectroscopy with an ATR for direct use on fats or oils. They are based on the CH out of plane deformation band observed at 966 cm-1, which is characteristic of isolated double bonds with trans configuration. The aim of this study was to develop an ATR-FTIR spectroscopic method for analysis of TFA in ground cereal products without the need for prior extraction of oil. A wide range of cereal foods including cereal based snacks were purchased from retailers. Samples were ground and FTIR spectra obtained with an ATR-FTIR Spectrometer using ground samples pressed onto the diamond ATR crystal and from oily films (called residual films) remaining on the surface of the ATR after brushing away the samples. TFA were measured in ground samples by a hydrolytic extraction, GC method (AOAC 996.01). Using IR spectra of ground samples or residual films and analytical values for TFA, PLS regression models were developed for the prediction of TFA (n=69). The PLS models were evaluated by cross validation and by prediction of an independent validation set (n=22). Based on peaks related to lipids, three ranges were used to calculate TFA models: (whole range, 4000-650 cm-1; three selected functional groups range, 3025-2850, 1789-1695 and 1500-900 cm-1; and a fingerprint range, 1500-900 cm-1). In both ground sample and residual film models, the fingerprint region gave the best model performance compared to the other regions. Moreover, residual film models gave better performance than ground sample models. The ATR-FTIR technique can provide a rapid method for screening TFA in cereal products within minutes without oil extraction.