Abstract: Response surface methodology was adopted to optimize the extraction of patulin from an aqueous matrix. Meanwhile, the optimal solid-to-liquid ratios (V/V) for four different aqueous matrixes were selected. Optimization of the extraction conditions for improved recovery of patulin was done by the combined use of one-factor-at-a-time method and response surface methodology. Recovery experiments were conducted with spiked samples at solid-to-liquid ratios using four different aqueous matrixes. The optimal extraction conditions established were as follows: acetonitrile to aqueous sample ratio, 7:3; NaCl dosage, 1.4 g/3 mL sample; and extraction temperature, 34 ℃. Under these conditions, the recovery of patulin was 90.47%, and the optimal solid-to-liquid ratios were 7:3 for apple pulp matrix and 1:9 for concentrated apple juice, wine grape and wine, giving a recovery of 91.08%, 85.51%, 88.74% and 89.17% from spiked samples, respectively. At the same time, over the entry-exit inspection and quarantine industry standard method (SN/T 2008-2007), the pre-treatment method was advantageous in terms of less solvent consumption, simple operation and good reproducibility. Collectively, we concluded that the solvent induced phase transition extraction of patulin optimized by response surface methodology yielded reliable predictions and it had the potential to provide a new pretreatment method for the extraction of patulin from aqueous matrices.

Key words: high performance liquid chromatography (HPLC), response surface methodology, patulin, aqueous matrix, solvent induced phase transition extraction