AUTOMATED MULTIVARIATE CALIBRATION IN SEQUENTIAL INJECTION FOURIER-TRANSFORM INFRARED-SPECTROSCOPY FOR SUGAR ANALYSIS

Two approaches for automated preparation of a multi-component calibrat ion solution in sequential injection analysis (SIA) using one single s tandard for each component are proposed. The key to both approaches is the generation of highly reproducible injection sequences in the flow system. In the first approach, single standards of three different co mponents are stacked in the holding coil in well-defined sequences. Up on flow reversal and transport to the detector, the stacked zones disp erse into each other producing exactly defined mixtures of the three s tandards which can be recorded as a function of time. The second appro ach is based on complete mixing of the three single standards in diffe rent ratios prior to detection, For this purpose, repeated short seque nces of small volumes of standard and distilled water are aspirated in to the holding coil, which results in a homogenous calibration solutio n upon flow reversal and transport to the detector. The proposed appro aches have been applied to simultaneous determination of glucose, fruc tose, and sucrose in aqueous standards as well as in soft drinks using Fourier transform infrared spectroscopic detection and multivariate p artial least-squares data analysis. The obtained results were compared with external reference methods to confirm the results obtained by th e multivariate SIA-FTIR technique. Typical deviations from the results obtained by the reference method were in the order of 2.5, 4.4 and 2. 8 g/L (2.5, 4.1 and 2.1% of total sugar) for glucose, fructose, and su crose, respectively.