The Solution
Waters Torus DEA column’s stationary phase consists of ethylene bridged hybrid (BEH) particles with tri-functionally bonded diethylamine ligands. The combination of the hydrophilic surface and the anion exchange properties of the ligands provide chromatographic characteristics that are well suited for retention and separation of anionic compounds such as chlorate and perchlorate. The use of BEH particles provides mechanical robustness to the column, allowing for the use of this stationary phase in small particle size column chemistries (sub-2-µm), thus facilitating the benefits of UltraPerformance Liquid Chromatography.
(UPLC) for this application, such as increased chromatographic resolution and shorter run times. An example chromatogram showing the retention and separation for the two compounds is shown in Figure 1, which also demonstrates the column void time (t0) for this method. The t0 is equivalent to 2 times the column void volume specified in the SANTE guidelines 11813/20175 due to the 0.5 mL/min flow rate used, alternative flow rates would result in a different t0. The total run time of the method including column re-equilibration is 7 minutes.
Samples of various representative food commodities were extracted using the QuPPe method and 5 µl of the filtered extracts were injected onto a 2.1 x 50 mm Torus DEA Column. The samples were analyzed using an ACQUITY UPLC I-Class System coupled with the Xevo TQ-XS, by applying an ammonium formate gradient to induce elution from the Torus DEA Column, 130 Å, 1.7 µm, 2.1 mm x 50 mm (p/n: 186007614).
The LC-MS/MS method performance was assessed following the SANTE guidelines.⁵ Matrix-matched, bracketed calibration curves were run to test linearity. Example calibration curves in organic carrot are shown in Figure 2 for both compounds. Response was linear for both compounds over a range of 0.002 to 0.200 mg/kg (R²>0.995, residuals <20 %) in all tested matrices. Repeatability of the LC-MS/MS method was determined by injecting the 0.010 mg/kg level from the matrix matched curves (n=6). The repeatability (%RSD) from analysis of carrot, grape, and infant food can be seen in Figure 3.
A possible residue of perchlorate was quantified in infant food, in the absence of an isotopically labelled internal standard, using a standard addition calibration within TargetLynx XS Application Manager. The retention time and ion ratio were within the tolerances defined in the SANTE guidelines. An example of this standard addition plot is shown in Figure 4, where the possible incurred residue was quantified at 0.003 mg/kg in infant food. All residuals were back calculated against the known added concentration automatically within TargetLynx and all were <6 %.
