An offline silver-impregnated silica gel solid-phase extraction (Ag-SPE) approach combined with a programmed temperature vaporization-gas chromatography-flame ionization detector (PTV-GC-FID) was proposed for routine analysis of mineral oil saturated hydrocarbons (MOSH) in chocolates. The MOSH in chocolates were extracted by n-hexane and 1 mL of extract was purified by offline Ag-SPE column. The SPE columns packed with 0.3％ Ag-activated silica gel were used to separate MOSH from triglycerides and olefins in chocolates. The eluent of MOSH fraction was only 5 mL and then concentrated to 0.2 mL through nitrogen blowing with little evaporation loss. The PTV parameters were as follows: the initial temperature was set at 45℃ and held for 1 min (split ratio was 200∶1), then warmed up to 360℃ at linear gradient of 250℃ minSymbolm_1 and held for 27 min (split valve was closed for 2 min followed by split ratio of 100∶1). The GC injection volume was 40 μL. The GC column was heated from 35℃ (3 min) to 350℃ at 25℃/min, and then raised to 370℃ (10 min) at 5℃/min. The flow rate of the carrier gas was 1.3 mL/min (and pressure was 60 kPa), FID temperature was set at 380℃. The limit of quantification (LOQ) and the recoveries of the method were 0.5 mg/kg and 84.9％-108.6％, respectively, with the relative standard deviations (RSD) of 0.2％-1.5％. Twenty-five commercial chocolate samples were analyzed with the proposed method, and it was found that the MOSH in three samples were not detected, and the concentrations of MOSH in other 22 samples were 1.09-8.15 mg/kg (the concentrations of MOSH with C16-C35 component were 0.56-4.43 mg/kg). The results suggested that it was necessary to routinely detect mineral oil contamination in chocolates for food safety.

An alternative method based on an off-line solid phase extraction ( SPE ) combined with programmable temperature vaporizer-based ( PTV) large volume injection-gas chromatography-flame ionization detection ( LVI-GC-FID ) was developed. The goal of this study was to determine mineral oil saturated hydrocarbons ( MOSH ) in camellia seed oils. The purification condition of SPE columns with silver impregnated the activated silica gel and activated aluminum oxide was optimized. The optimal SPE cartridge was loaded with 10 g of Ag-activated silica gel per 10 g of activated aluminum oxide. The PTV initial temperature was set at 75℃ for 1 min (split 200:1), and heated from 75℃ to 370℃ at 250℃/min. Then the diverter valve was closed for 1 min and opened again with the split flow ratio changing to 50:1 . The injection volume was 40μL. The calibration curve of paraffin oil was liner in the range of 5-500 mg/kg with correlation coefficient of 0. 998. The detection limit (LOD) and the quantification limit (LOQ) of paraffin oils in hexane were 0. 26 mg/kg and 0. 80 mg/kg, respectively. The recoveries from spiked oil samples were between 93 . 3% and 112 . 7%, with relative standard deviation ( RSD ) of 1 . 8%-5 . 2%, the RSD of intra-day and inter-day were less than 2 . 6% . This procedure was applied to analyze the MOSH in 11 commercial camellia seed oils and the contamination was found to range from 6. 8 mg/kg to 76. 7 mg/kg. The method is simple in operation with high sensitivity, good reproducibility and low cost, and suitable for determination of MOSH in vegetable oils.