Determination of Polydimethylsiloxane in Vegetable Oil by Organic Phase Sampling-Inductively Coupled Plasma Atomic Emission Spectrometry

In order to solve the problem of high determination results caused by significant differences between the standard solution and the sample matrix, and the dissolution of silicon from plastic materials in the existing method of inductively coupled plasma atomic emission spectrometry (ICP-AES) for the determination of polydimethylsiloxane in vegetable oil, the method mentioned by title was proposed, and effects of the types and properties of diluents, dilution ratios, and container materials on the determination results were investigated. The range of diluent parameters which was beneficial for plasma stabilization was optimized by the test, with vapor pressure (20 ℃) from 0.032 kPa to 0.287 kPa, density (20 ℃) from 0.798 g·mL^-1 to 0.881 g·mL^-1, and kinematic viscosity (20 ℃) from 2.47 mm²·s^-1 to 15.8 mm²·s^-1. When the kinematic viscosity (20 ℃) of the solution to be tested was less than 6.35 mm²·s^-1, the correlation coefficient of the working curve was above 0.999 5. When the dilution ratio of vegetable oil to diluent was 1∶4 (mass ratio), the detection limit of the method was lower, using ICP solvent as the diluent. In the extraction process of polydimethylsiloxane, no matter whether the plastic centrifuge tube was heated or not, the content of the substance to be tested in the extraction solution increased to varying degrees, while there was no effect using the colorimetric tube made of glass material. The optimized experimental conditions were as follows. The sample (4 g) was taken and placed in a 50 mL-glass colorimetric tube. ICP solvent was added till to a total mass of 20 g, and then the mixture was mixed well and extracted by ultrasonic for 2 h at 70 ℃. 15 mL of 1.37 mol·L^-1 hydrochloric acid solution was added, following mixing well and settling to layer. The supernatant was taken and placed into a glass colorimetric tube, in which polydimethylsiloxane was determined by ICP-AES. As shown by the results, the matrix matching method effectively corrected the interference of the sample matrix on the determination results by adding an appropriate amount of blank vegetable oil. Linear relationship between values of the corresponding emission intensity of silicon element and mass fraction of polydimethylsiloxane was found in the range of 0.500-10.0 mg·kg^-1, with lower limit of determination of 0.30 mg·kg^-1. Test for recovery was made by standard addition method, giving results in the range of 95.6%-106%, with RSDs (n=7) of the determined values in the range of 1.1%-3.4%. This method and flame atomic absorption spectrometry of GB 5009.254-2016 second method were applied to determination of the same actual spiked sample, and the determination results obtained by two methods were consistent.