Fig. 3a shows the response of a gold electrode modified by electrodeposition

of palladium for successive injections of 150 μL ascorbic acid from (A) 10 μmol L−1 to (E) 50 μmol L−1 and seven samples of honey (A1–A7). Plot calibration (Fig. 3b) implies that the proportionality between the amperometric current and the concentrations of the analyte is Current (μA) = −0.068 + 0.028 [AA] (μmol L−1); correlation coefficient, 0.9938. The proposed system presents a good reproducibility and has a very favourable signal-to noise ratio, demonstrated by the very stable baseline obtained for these low concentrations. The detection limit was calculated on the basis of 3σ (σ being the residual standard deviation of the intercept), yielding Caspase inhibitor a value of 0.14 μmol L−1 and the quantification PLX4032 solubility dmso limit was calculated on the basis of 10σ, yielding a value of 0.49 μmol L−1. Under the optimum conditions, the FIA-amperometric system applied for the determination of ascorbic acid in seven honey samples was based on two steps involving the injection of: (1) the sample and the standards solutions in the channel without the reactor and (2) the sample in the channel containing the enzymatic

reactor with ascorbate oxidase immobilised. Table 1 and Fig. 4 compare the results of the analyses performed by amperometric method, developed in this work, and iodometric titration method (Suntornsuk, Gritsanapun, Nilkamhank, & Paochom, 2002) for seven different samples (in triplicates). The comparison of the amperometry with gold/palladium electrode and the iodometry gives a slope and intercept close to unity and zero, respectively. A good correlation (r2 = 0.9998) between the amperometric and titration methods was found. The confidence interval for the slope and intercept are (0.77 ± 0.04) and (0.63 ± 0.15) mg (100 g−1),

respectively, for a 95% confidence level. A paired Student’s t-test showed that the mean values (texp < tcrit; diglyceride 0.5 < 2.5, n = 7, P = 0.95) not significantly differ. Taking into account of these results, do no significant differences between the three methods were observed, which strongly indicates the absence of systematic errors. Recovery experiments on honey solutions spiked with different amounts of ascorbic acid were also carried out. The method recoveries obtained for the ascorbic acid ranged from 92% to 107%; such values confirm the accuracy of the proposed method. The main disadvantage of the present method is the fact that the honey inactivates the ascorbate oxidase after 50 injections, requiring the construction of a new reactor. This work demonstrated the potentiality of the amperometric method using gold electrodes modified with palladium coupled with flow injection analysis techniques for the determination of ascorbic acid in honey using the enzyme ascorbate oxidase immobilised in a tubular reactor.