A selection of fourteen common and commercially-available table-top artificial sweeteners was considered. The samples contained aspartame and saccharine as high-intensity sweeteners, and dextrose, sorbitol, sucrose and maltodextrin as low-intensity sweeteners. These were all examined both in powder form and as aqueous solutions. Raman spectra, excited at 1064 nm, were acquired using a compact dispersive scheme. These spectra provided fluorescence-free Raman signatures from which to identify the most significant peaks of the various sweeteners. These peaks were also compared with ones obtained by means of computational analysis, in order to show the effect of the entire sweetener matrix. The spectroscopic data were then processed by means of chemometric analysis for distinguishing what kind of sweetener was present in a given sample. First, Principal Component Analysis was applied for the purpose of data dimensionality reduction and explorative investigation, and provided good clustering depending on the type of sweetener. Next, the K-nearest neighbor method was used in order to assign the samples to pre-defined classes. An excellent identification in accordance with the type of high or low-power sweetener was thus obtained. These results confirm the success of Raman spectroscopy in attaining a straightforward analysis of intact food, with high potentials for its use as a non-destructive and "green" analytical method for quality control in the food industry.
Characterization of Artificial Sweeteners Using Raman Spectroscopy
AG Mignani;L Ciaccheri;G Manca;AA Mencaglia;
2016
Abstract
A selection of fourteen common and commercially-available table-top artificial sweeteners was considered. The samples contained aspartame and saccharine as high-intensity sweeteners, and dextrose, sorbitol, sucrose and maltodextrin as low-intensity sweeteners. These were all examined both in powder form and as aqueous solutions. Raman spectra, excited at 1064 nm, were acquired using a compact dispersive scheme. These spectra provided fluorescence-free Raman signatures from which to identify the most significant peaks of the various sweeteners. These peaks were also compared with ones obtained by means of computational analysis, in order to show the effect of the entire sweetener matrix. The spectroscopic data were then processed by means of chemometric analysis for distinguishing what kind of sweetener was present in a given sample. First, Principal Component Analysis was applied for the purpose of data dimensionality reduction and explorative investigation, and provided good clustering depending on the type of sweetener. Next, the K-nearest neighbor method was used in order to assign the samples to pre-defined classes. An excellent identification in accordance with the type of high or low-power sweetener was thus obtained. These results confirm the success of Raman spectroscopy in attaining a straightforward analysis of intact food, with high potentials for its use as a non-destructive and "green" analytical method for quality control in the food industry.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.