In this paper we studied the possibility of quantification of sugars in aqueous solution by use of Fourier transform infrared (FT-IR) spectroscopy coupled with an attenuated total reflectance (ATR) accessory. Seven kinds of sugar solution (glucose, maltose, sucrose, fructose, lactose, galactose, mannose) were prepared as food models and their infrared spectrum was obtained for different concentrations. It was shown that the shape of spectrum in the fingerprint region for each sugar was quite different. In this region, all peak areas except that of C-OH vibration band, plotted against the concentration of the sugar solution, showed excellent linearity in non-ideal solution like maltose, sucrose, fructose and lactose, while all peak areas showed excellent linearity in ideal solution like glucose. The calibration curve of the peak with non-linearity was fitted well by use of non-linear function of activity coefficient of each sugar solution. Curve-fit of the fingerprint region was attempted to find the most independent wave number, and it was shown that the curve fit spectra with the maximum absorbance of all IR bands was the most independent and the most suitable for quantitative analysis of sugar concentration.
View full abstract