Application of Fourier Transform Infrared Spectroscopy for Monitoring Short-Chain Free Fatty Acids in Swiss Cheese

• Published in: Journal of dairy science 2007-08, Vol.90 (8), p.3596-3603
• Main Authors: Koca, N., Rodriguez-Saona, L.E., Harper, W.J., Alvarez, V.B.
• Format: Article
• Language: English
• Subjects: Animal productions; Animals; Biological and medical sciences; Cheese - analysis; Chromatography, Gas - methods; Dietary Fats - analysis; Fatty Acids, Nonesterified - analysis; Flame Ionization - methods; Food Analysis - instrumentation; Food Analysis - methods; Food industries; free fatty acid; Fundamental and applied biological sciences. Psychology; infrared spectroscopy; Least-Squares Analysis; Milk and cheese industries. Ice creams; multivariate analysis; Reference Values; Reproducibility of Results; Spectroscopy, Fourier Transform Infrared - methods; Statistics as Topic; Swiss cheese; Terrestrial animal productions; Vertebrates
• ISSN: 0022-0302; 1525-3198
• DOI: 10.3168/jds.2007-0063

Short-chain free fatty acids (FFA) are important sources of cheese flavor and have been reported to be indicators for assessing quality. The objective of this research was to develop a simple and rapid screening tool for monitoring the short-chain FFA contents in Swiss cheese by using Fourier transform infrared spectroscopy (FTIR). Forty-four Swiss cheese samples were evaluated by using a MIRacle three-reflection diamond attenuated total reflectance (ATR) accessory. Two different sampling techniques were used for FTIR/ATR measurement: direct measurement of Swiss cheese slices (∼0.5g) and measurement of a water-soluble fraction of cheese. The amounts of FFA (propionic, acetic, and butyric acids) in the water-soluble fraction of samples were analyzed by gas chromatography-flame ion-ization detection as a reference method. Calibration models for both direct measurement and the water-soluble fraction of cheese were developed based on a cross-validated (leave-one-out approach) partial least squares regression by using the regions of 3,000 to 2,800, 1,775 to 1,680, and 1,500 to 900cm−1 for short-chain FFA in cheese. Promising performance statistics were obtained for the calibration models of both direct measurement and the water-soluble fraction, with improved performance statistics obtained from the water-soluble extract, particularly for propionic acid. Partial least squares models generated from FTIR/ATR spectra by direct measurement of cheeses gave standard errors of cross-validation of 9.7 mg/100g of cheese for propionic acid, 9.3 mg/100g of cheese for acetic acid, and 5.5 mg/100g of cheese for butyric acid, and correlation coefficients >0.9. Standard error of cross-validation values for the water-soluble fraction were 4.4 mg/100g of cheese for propionic acid, 9.2 mg/100g of cheese for acetic acid, and 5.2 mg/100g of cheese for butyric acid with correlation coefficients of 0.98, 0.95, and 0.92, respectively. Infrared spectroscopy and chemometrics accurately and precisely predicted the short-chain FFA content in Swiss cheeses and in the water-soluble fraction of the cheese.