Desenvolvimento e validação de metodologia cromatográfica para determinação de bisfenol A em simulantes de alimentos de ensaios de migração

Abstract

This study aimed to develop and chromatographic method for the determination of bisphenol A in four different food simulants using high performance liquid chromatography and diode array detector (PDA) to support in enasios migration. The simulating used were A (purified water), B (solution aquosade 3% acetic acid), C (aqueous 10% ethanol) and D (aqueous 95% ethanol). Preliminary assessment of the analytical conditions was conducted in order to establish those that confer the best results, which were the system of mobile phase acetonitrile-water in the ratio 70:30 by volume, 40 ° C oven and length of the column wavelength of 201 nm. The retention time of the bisphenol A was found to be 3.25 minutes. For simulating the method of treatment consists of evaporation of sample 1 ml of sample at 60 ° C with duration of 1 hour and 45 minutes (A and C), 2 hours (B) and 1 hour (D). In sequence, the method was evaluated with respect to the parameters of analytical validation, including linearity, limit of detection and quantification, repeatability, intermediate precision between different analysts and between different days, recovery, selectivity and robustness. The method for simulating the coefficient for determining (r ²) or greater than 0.9988 in the range tested (50 to 2000 ppb); detection limit and quantitation, respectively, 14.7 ppb and 48.9 ppb; coefficient variation in repeatability of 2.07% for the inter-day precision of 1.05% and the accuracy of inter-Analyst 1.76%, recovery of 97.9% (CV = 1.3%) (50 ppb ), 99.2% (CV = 0.7%) (1000 ppb) and 103.6% (CV = 0.4%). The results for the simulant B includes coefficient of determination (r ²) 0.9972 or higher; limit of detection and quantification, respectively, 7.8 ppb and 26.0 ppb; coefficients of variation for repeatability of 3.74%, inter day precision, accuracy and 0.14% cross-Analyst 0.2%, 94.8% recovery (CV = 4.8%) (50 ppb) and 99.8% (CV = 0.2% ) (1000 ppb) and 92.3% (CV = 1.2%) (2000 ppb). For the simulant C results include a coefficient of determination (r ²) 0.9988 or higher; limit of detection and quantification, respectively, 4.8 ppb and 16.1 ppb, coefficient of variation for repeatability of 2.84%, accuracy inter-day precision of 1.37% and inter-analysts of 0.14%; de104 recovery, 6% (CV = 2.5%) (50 ppb), 99.8% (CV = 1.3%) ( 1000 ppb) to 102.5% (CV = 2.9%) (2000 ppb). The results for simulating D was 0.9984 r ² or more, detection limit and quantitation, respectively, 13.7 ppb and 45.6 ppb; coefficient of variation of 2.0% to repeatability, accuracy of an inter-day 9% and accurate inter-de1 analysts, 2%; recuperção 92.6% (CV = 1.76%) (50 ppb) and 99.9% (CV = 0.9%) (1000 ppb) and 104 8% (CV = 2.7%) (2000 ppb). There were no significant interferences to the method tested and proved to be robust.