Under the optimized conditions, the four sets of standard solution
for eight triazines were determined for the calibration curves
corresponding to rice, wheat, chive, and carrot matrix (three replicate
injections for each concentration) and the four sets of blank
samples (11 per set) were tested for the limits of detection (LODs)
corresponding to these four sample matrix. The peak areas of the
eight triazines were used for quantification signals as ordinate
(Y) and their concentration (ng/g) as abscissa (X). The linearity of
calibration curves for eight triazines was checked in a wide range
of concentrations (0.1–100 ng/g) by F-test. Each standard solution
was injected in triplicate. The results showed that all of the calculated
F-values for eight triazines were higher than theoretical F-value
6.61 (P = 0.05) in these four sample matrix. Thus, the linearity
of the method was significant. The calibration curves, linear ranges
(LRs), determination coefficients (r2), LODs (S/N = 3) were summarized
in Table 1. The repeatability of the method was evaluated in
terms of intraday and interday precision. Intraday precision was
examined by five replicate injections of the sample spiked at
20 ng/g for each triazine on the same day. Interday precision was
assessed by injecting the same sample for three consecutive days
(five replicate injections per each day). The RSDs of the peak areas
in terms of intraday and interday precision were also shown in
Table 1. As shown in the table, an acceptable precision was
obtained with intraday RSD values below 1.7%, while interday
RSD values within 3.3%. These results indicated that this method
was repeatable and sensitive.