Desenvolvimento e otimização de metodologia para análise de edulcorantes por eletroforese capilar

Abstract

An innovative methodology was developed to simultaneously separate and quantify aspartame, cyclamate, saccharin and acesulfame-K through capillary electrophoresis with direct and indirect detection of UV, analysis time of roughly 6 minutes and with no sample derivatization. As the cyclamate has low absorptivity we used a system of electrolyte added to a chromophore, which allowed the indirect detection of this analyte. The wavelength was chosen taking into account an intermediary absorptivity among chromophore electrolyte, aspartame, saccharin and acesulfame-K. After testing some chromophores, the benzoic acid was selected as the most appropriate. A point to be considered is that food matrices have benzoic acid in its constitution, used as a preservative. Although it is still detected in the electrolyte system and the proposed analytical conditions, its quantification cannot be attained due to its inclusion in the electrolyte, showing signs of lack of adjustments of the linear model with 95% confidence. The ideal condition for electrophoretic separation was obtained by applying 32 factorial design. The optimization consisted of 20.0 mmol L-1 of sodium tetraborate and 15.0 mmol L-1 Tris-Aben (pH 9.15 ± 0.03), voltage of +20 kV, injection of 4 seconds at 50 mbar, 215 nm, using levofloxacin as internal standard. A sample of a solid prepared for drink lemon tea with the four sweeteners in its constitution was analyzed, and its pre-treatment consisting only of filtering the sample, after reconstitution, without any derivatization. Even though the analytical methodology was not validated, some important figures were evaluated: linearity, selectivity, limits of detection and quantification, precision, recovery and strength. The parameters assessed showed a regression coefficient greater than 0.98 and did not show lack of adjustment in the concentration range tested, evaluated as 95% granted for all analytes. Selectivity was tested through standard addition curve, which makes possible the observation of parallelism between the slopes of the patterns and the sample, indicating selectivity of the methodology. The RSD (%) for migration time and area using ratio as internal standard was less than 5% for the area and less than 2% for migration time for the four analytes, considering sample and standard. The detection limits found were 6.8 mg L-1 for aspartame, 12 mg L-1 for cyclamate, 0.50 mg L-1 for saccharin and 3.3 mg L-1 for acesulfame-K. These values are more significant than those found in the literature for determination by CE and they are very competitive, if not better, than the determination by other techniques. The accuracy was found to be within acceptable limits, and the recovery values in the range of 91.9 to 102.2%.