3.3. Analytical characteristics
The linearity of the calibration graphs were studied by injecting in duplicate five concentrations (5–35 ng g1) of standard mixtures of the aflatoxins. Calibration plots were linear for the tested rangewith correlation coefficients >0.999 for all the aflatoxins. The limit of detection (LOD) was calculated to be 0.10 ng g1 for both aflatoxins G1 and G2 and 0.06 ng g1 for both aflatoxins B1 and B2.The LOD was obtained by using the equation:LOD ¼ ksb=m ð1Þ
where m is the calibration sensitivity, sb is the standard deviation of the blank, 3 was chosen as the factor k while the LOD was based on the signal-to-noise ratio of 3:1. LODs of 0.01 ng g1 and 0.5 ng g1 for each aflatoxin was reported by Ali et al. (2005) and Akiyama et al. (2001), respectively. The repeatability and reproducibility of the method was assessed by injecting five times each of the four standard mixtures on the same day (intra-day) and over 6 days (inter-
day), respectively. Good reproducibility of both the peak area (RSD 6 3.5%) and retention time (RSD 6 1.0%) were found.