Detection for Different Adulterants of Ziziphi Spinosae Semen Based on Near Infrared Spectroscopy

Abstract

In this paper, near infrared spectroscopy technology was used to study the qualitative and quantitative detection of Ziziphi Spinosae Semen and its three kinds of counterfeits, Ziziphus mauritiana lam, Hovenia dulcis Thunb. and Lens culinaris. Different single-adulterant samples were prepared with adulterant concentration in range of 1%~90%. Multiple-adulterants samples were also prepared by adding the three kinds of counterfeits simultaneously. Near-infrared spectroscopy data in the range of 800~2500 nm were acquired. Principal component analysis (PCA) was used firstly to qualitatively identify Ziziphi Spinosae Semen and three kinds of counterfeits. For single-adulterant samples, five different pretreatment methods were applied to denoising. Partial least squares regression was used to establish PLS1 models to quantitatively predict concentrations of the adulterants. Successive projection algorithm (SPA) was used to select the optimal wavelength to optimize the PLS1 models. Three-wavelength prediction model for the adulteration detection of Ziziphus mauritiana Lam was established, with determination coefficient R2p for prediction set of 0.9659 and root mean square error (RMSEP) of 6.1910%. Eight-wavelength prediction model for the adulteration detection of Hovenia dulcis Thunb. was established, with R2p of 0.9491 and RMSEP of 7.6232%. Five-wavelength prediction model for the Lens culinaris adulteration detection was established, with R2p of 0.9666 and RMSEP of 6.1437%. For the multiple-adulterants samples, PLS2 models were established to determine concentrations of the different counterfeits simultaneously. The prediction results for Ziziphi Spinosae Semen were the best with R2p≥0.7115, while the results for Hovenia dulcis Thunb. were the worst with R2p≥0.2007. The results showed that the near infrared spectroscopy could be used to inspect different counterfeits adulterated in Ziziphi Spinosae Semen. The established method provided a theoretical basis for the subsequent development of portable detection equipment for authenticity of Ziziphi Spinosae Semen. It was also as a reference for other studies on quality inspection for seed Chinese medicinal materials. The method was of an important social significance for ensuring quality and safety of Chinese medicinal materials.

Keywords

• ziziphi spinosae semen
• adulteration detection
• near infrared spectroscopy
• partial least squares regression