Identification of peach and apricot kernels for traditional Chinese medicines using near-infrared spectroscopy

Abstract

Peach kernels (persicae semen) and apricot kernels (armeniacae semen) are used in traditional Chinese medicines as an anti-blood stasis agent and an antitussive/expectorant, respectively. They are sometimes mixed with each other, either mistakenly or deliberately to counterfeit products, because they are similar in appearance. However, they contain greatly different amounts of the active ingredient. Their medicinal effects are also different. Therefore, it is necessary to distinguish them correctly. The traditional identification method, which is based on human senses, requires experience and skill and is time-consuming. Its accuracy thus varies by inspector. In this study, near-infrared (NIR) spectroscopy and multivariate analyses were employed to identify peach and apricot kernels. The analyses were performed either at 12500−4000 cm−1 (entire NIR region) or at 12500−9000 cm−1 (high-wavenumber region). The spectra in the latter region can be measured with an inexpensive and simple device suitable for practical use. An analysis of the spectra obtained at 12500−4000 cm−1 indicates that the two types of kernel can be distinguished using the principal component score plot. In the high-wavenumber region (12500−9000 cm−1), the analysis was performed using partial least squares discriminant analysis. All the test samples were correctly identified, showing that the proposed method is a powerful tool for accurately identifying peach and apricot kernels.