The ability of Fourier transform Raman spectroscopy to measure cyanogenic glycoside amygdalin was investigated using a standard addition series in which amygdalin (0-700 nmol/mg) was added to ground sweet almond. As an additional test, the method so developed, and further refined, was used to determine endogenous amygdalin in bitter almonds. Using partial least-squares regression (PLSR), the best correlation between the spectra of the standard addition series and amygdalin content exhibited a value of 0.999, while the root mean square error of cross validation (RMSECV) was found to be 4 nmol/mg using one principal component. The model was based on the first derivative of the aromatic C-H stretching band at 3060 cm<sup>-1</sup>. A less accurate but still excellent PLSR model could be developed on the vibrational band of the nitrile group, specific for the cyanogenic compound in the system. The best PLSR model obtained for endogenous amygdalin in 17 raw bitter almonds yielded a RMSECV = 13 nmol/mg and <i>r</i> = 0.937. In addition, the distribution of amygdalin in bitter almond was investigated <i>in situ</i> by obtaining spectra of an intact cross-section of the cotyledon of bitter almond using Raman microscopy. While amygdalin was found not to be present in measurable amounts in the center of bitter almond, the concentration increased towards the epidermis, but not linearly.
You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.
Contact your librarian or system administrator
Login to access OSA Member Subscription