Fourier transform Raman Spectrometry for the quantitative analysis of oil content and humidity in olives

A method for the direct and fast determination of oil content and humidity in olives using Fourier transform Raman spectrometry is reported. The only sample preparation step required was crushing of the olives using a hammer mill. The crushed olives were placed in a dedicated sample cup, which was rotated excentrically to the horizontal laser beam during spectrum acquisition. This allowed us to sample an increased volume and thus compensate for sample inhomogeneities. In this way the reproducibility of Raman spectra taken from crushed olives was significantly improved. Partial least-squares (PLS) regression was used for the chemometric evaluation of the Raman spectra. Standard errors of prediction for the validation set of 0.81% for oil content (in the range 19.68-35.71%) and 1.54% for humidity (in the range 29.23-51.49%), both expressed as weight percentage referred to fresh matter, were obtained.