Depth profiling of porcine adipose tissue by Raman spectroscopy

Raman spectroscopy was applied on a depth profile of porcine adipose tissue (from skin to meat) with the purpose of (1) discriminating between fat layers and (2) estimating the variation in fatty acid composition as a function of fat depth and fat layer: total degree of unsaturation (iodine value), fractions of saturated, and monounsaturated and polyunsaturated fatty acids. The thickness and composition of the outer layer of porcine adipose tissue influences the final quality of backfat. A too-thick outer layer is associated with problems such as oily appearance, rancidity development, and difficulties in separating muscle and adipose tissue when cutting. From principal component analysis on standard normal variate preprocessed Raman spectra (1800800?cm1), it was possible to discriminate between the outer and the inner backfat layer. Principal component analysis loadings showed that the separation of layer was mainly explained by variation in the bands originating from vibration of double bond C?=?C stretching plus?=?CH twisting and rocking. In the prediction of iodine value a three-component partial least squares regression model based on full range Raman spectra showed a root mean square error of cross validation of 2.00 and R2?=?0.69. Applying CauchyLorentz band fitting proved that information regarding fat unsaturation was found not only in band intensity, but also in band parameters such as location and width. The results suggest Raman spectroscopy as a potential measurement technique for rapid grading of pork carcasses. Copyright (c) 2011 John Wiley & Sons, Ltd.