Determination of Residual Amounts of 12 Quinolone Antibiotics in Evironmental Water by Magnetic Solid Phase Extraction-High Performame Liquid Chromatography-Tandem Mass Spectrometry

A magnetic material (GB@nZVI) was prepared from graphene oxide (GO) and zero valency nano iron (nZVI) by the authors and was used as magnetic adsorbent in SPE of 12 quinolone antibiotics (QNLO) from water sample. A portion of the magnetic material (15 mg) was placed in a small beaker, washed and activated with methanol and de-ionized water successively. 10 mL of the environmental water, which has been filtered through 0.22 μm filtering membrane to remove the suspended particles, was then added to the activated adsorbent in the small beaker, and its acidity was adjested to pH 7.0 with acetic acid and aq.ammonia. The mixture was then shaked for 5 min to have the QNLOs in water fully adsorbed by the magnetic adsorbent, which was pulled and kept at the bottom of the beaker by the action of a magnet at outside beaker bottom. The supernatant was discarded, and the magnetic adsorbent was ultrasonicated thrice with 1.0 mL of aq.ammonia each. The ammoniacal eluates were combined and evaporated to near-dryness by N₂-blowing. The residue was taken up with 100 μL of water, and the solution was filtered through 0.22 μm filtering membrane. The filtrate was used for HPLC-MS/MS analysis under the working condition of the instrument. In chromatographic separation, ZORBAX Eclipse Plus C₁₈ column was used as stationary phase, and mixtures of (A) methanol and (B) 1% (volum fraction) formic acid solution in various ratios were used as mobile phase in gradient elution. And the 12 QNLOs were determined by MS/MS under the conditions of ESI⁺ and MRM modes. Linearity ranges of the standard curves for the 12 QNLOs found were same between 1.0-50 μg·L^-1. Their detection limits (3S/N) were found in the range of 1.6-9.6 ng·L^-1. Values of RSDs (n=5) found were ranged from 3.4%-6.1% (intraday) and from 7.2%-13% (interday). Recovery was tested by standard addition method, giving results of recovery in the range of 90.3%-103%. The magnetic material was characterized by SEM and FTIR, showing that GO was tightly adsorbed on the surface of nZVI; and that the SPE recovery given by the GO@nZVI after using for 10 times was about 10% lower than that given by the sample portion of GO@nZVI using for the first time. Hence it could be applied in repeated usage.