A high performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) method was established for the determination of nonylphenol in milk powder. Samples were extracted with acetonitrile and cleaned up with a solid phase extraction column. The analytes were separated on a Waters ACQUITY UPLC BEH C18 column (2.1 mm × 50 mm, 1.7 μm), detected by electrospray ionization in the multiple reaction monitoring (MRM) mode, and quantified by an external standard method. The results showed that good linearity was observed in the concentration range of 1–200 ng/mL (R2 > 0.99). The limit of detection（LOD, RS/N = 3）was 2 μg/kg and the limit of quantitation (LOQ, RS/N = 10）was 5 μg/kg. The recoveries at different spiked levels were between 81.10% and 107.39% with relative standard deviation (RSD) ranging from 1.97% to 9.73%. This method has the advantages of accurate results, small background interference, simple operation, good stability and short pretreatment time, and is useful for the detection of nonylphenol in milk powder.

A method for the determination of nonylphenol in liquid milk by using ultra high performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) was developed. Samples were extracted using acetonitrile as extraction solvent and the organic phase was separated from the aqueous phase by adding saturated sodium chloride solution. The analyte was cleaned up using a ProElut PAEs Glass solid-phase extraction catridge and then separated on a Waters ACQUITY UPLC BEH C18 column (2.1 mm × 50 mm, 1.7 μm) using gradient elution with 0.1% ammonia water-methanol as mobile phase. The detection was performed by negative electrospray ionization under the multiple reaction monitoring mode and the quantitation was carried out by an internal standard method. The results showed a good linear relationship (R2 > 0.990) in then range from 1 to 200 ng/mL. The limit of detection（LOD, RS/N = 3）was 2.0 μg/kg and the limit of quantitation (LOQ, RS/N = 10）was 5 μg/kg. The recovery rate of the method for spiked samples was between 74.8% and 111.5% with relative standard deviation of 3.3%-9.0%. In general, the method was accurate, simple and reproducible and exhibited low background inference and it was thus suitable for the analysis of nonylphenol in liquid milk.