RAPID METHODS FOR ANALYSIS OF EDIBLE OILS AND FATS BY FOURIER TRANSFORM INFRARED SPECTROSCOPY

Analysis using Fourier transform infrared (FTIR) spectroscopy techniques on edible fats and oils extracted from palm fruit, groundnut, sesame seed, cottonseed and animal body fats rendered from cow, chicken, lamb and lard were investigated. The studies included development and applications of rapid FTIR techniques to determine some quality parameters such as moisture content in crude palm oil (CPO), soap and hexane residues in refined palm oil and groundnut oil, malondialdehyde (MDA) as a secondary oxidation product in refined palm oil, minor components such as sesamol and gossypol in sesame and cottonseed oils, and aflatoxins in groundnut and groundnut cake. The detection of lard in different mixtures with other animals' body fats such as cow, chicken and lamb was also investigated. Different sample handling techniques were used such as transmission cells of NaCl, BF2, KBr and attenuated total reflectance (ATR) using internal reflectance element (IRE) of ZnSe. Partial Least Square (PLS) and Principle Component Analysis (PCA) statistical methods were used to drive calibrations from FTIR versus actual or chemical values. In this study the frequency of 3700-3072 cm⁻¹ was used to determine moisture content in CPO as it indicates the absorption of compounds containing hydroxyl groups (OH). The frequency at 1675- 1500 cm⁻¹ was used to determine soap residues in refined edible oils. For the determination of hexane residue in oils, the frequency used included all the data from 2935-2817 cm⁻¹, 1 490-1333 cm⁻¹ and 1200-1000 cm⁻¹ for -CH₃ and -CH₂, and in-plane -CH bending. In the determination of MDA as a secondary oxidation product, the correlation and variance spectra were used to select the best regions (2900-2800 and 1800-1600 cm⁻¹) to derive calibration from FTIR versus values obtained by chemical methods with SEC of 1.49. The spectral regions included the data from 3650-3000, 1600- 1450 and 1200-900 cm⁻¹ that were used to determine sesamol in sesame seed oil. The study also included a qualitative and semi quantitative determination of palm and groundnut oils as adulterants in sesame seed oil using the spectral regions from 1504- 1503, 1400- 1397 and 917-914 cm⁻¹. The gossypol was also determined as an important quality factor in cottonseed oil and cakes using the spectral regions from 3600-2520 and 1900-800 cm⁻¹. The study also covered the detection of lard in mixture of body fats of chicken, lamb and cow by using changes in frequency and absorbance of spectral regions 3009-3000, 1418-1417, 1385-1370, 1126- 1085 and 966-967 cm⁻¹. The simple Beer-Lambert law was used to develop equations for the determination of mixtures. Aflatoxins exhibit characteristic absorption bands at wavelengths of 3004-2969 cm⁻¹ for CH₂, aromatic =CH, -C-H, C=C and phenyls, 1744-1720 cm⁻¹ for C=O, 1364-369 cm⁻¹ for methyl adjacent to epoxy ring, 1217-1220 cm⁻¹ for in plane -CH bending of phenyl, 1035-1037 cm⁻¹ for symmetric stretching of =C-O-C or symmetric bending of phenyl, and 900-902 cm⁻¹ which may be for isolated H. In this calibration set the spectral regions that showed the highest correlation between concentration information and spectral response were set to include the data from 3000-2932, 1832-1693, 1400-1329 and 1250-1187 cm⁻¹ for aflatoxins B₁, with standard errors of calibration (SEC) of 1.80 parts per million (ppm). All of the results were in good correlation and of comparable accuracy to the classical wet chemical methods such as the American Oil Chemists Society (AOCS), Association of Official Analytical Chemists (AOAC) and International Union of Pure and Applied Chemistry (IUPAC) methods. This study represents the use of FTIR spectroscopy as a new rapid analytical technique developed for determination of some quality parameters of fats and oils, together with the detection of adulterants and contaminants. The FTIR spectroscopic technique has the potential to replace the timeand effort-consuming chemical methods for fast analysis of fats and oils. This can also eliminate the use of toxic chemicals that are hazardous to the analysts as well as to the environment in the analysis.