Near-infrared Spectroscopy Technology In Active Ingredients Of Medicinal Fungi Analysis

Near infrared spectroscopy technology (NIRS) is a novel technology applying to rapid determination of single component or multicomponents in materials depend on their optical characteristic in near infrared spectroscopy region. partial least squares (PLS) is wildly used in near-infrared spectral analysis and the effect is good. Using PLS method, information can be extract effectively from the original spectra, noise can be removed, the sensitivity of analysis model can beimproved.In this paper, Application of NIRS-PLS for analysis active contents in Medicinal fungi is studied. The influence of spectra pretreatment methods on quantitative analysis model have been considered. During the process of model establishing, wavelength range and factors were also studied.Various pretreating spectra methods including Savitzky-Golay smoothing, Fast Fourier transform, First derivative, and Second derivative were used for conversion of NIRS. The NIRS-PLS models for quantitative analysis of protein and polysaccharide in Coriolus versicolor were established. FFT spectra were used to establish the optimal model for quantitative analysis of protein content. The optimal wavelength range is 1330~1725 nm and the best factor is 6. The RMSECV value is 0.00946. Savitzky-Golay smoothing spectra were used to establish the optimal model for quantitative analysis of polysaccharide content. The optimal wavelength range is 800~2500 nm and the best factor is 5. The RMSECV value is0.01391. Using the optimal models to predict the active contents in samples of forecast set, the RMSEP of model for analysis protein is 0.00864 and the RSD is 0.138%. The RMSEP of model for analysis polysaccharide is 0.011 and the RSD is 0.207%.Various pretreating spectra methods including Savitzky-Golay smoothing, Fast Fourier transform, First derivative, and Second derivative were used for conversion of NIRS. The NIRS-PLS models for quantitative analysis of protein and adenosine in Cordyceps militaris were established. FFT spectra were used to establish the optimal model for quantitative analysis of adenosine content. The optimal wavelength range is 800~2500 nm and the best factor is 9. The RMSECV value is 0.25487. Savitzky-Golay smoothing spectra were used to establish the optimal model for quantitative analysis of protein content. The optimal wavelength range is 800~2500 nm and the best factor is 10. The RMSECV value is 0.01225. Using the optimal models to predict the active contents in samples of forecast set, the RMSEP of model for analysis adenosine is 0.11962 and the RSD is 0.127%. The RMSEP of model for analysis protein is 0.0362 and the RSD is 0.144%.In summary, this paper focuses on the study of pre-treatment methods of near-infrared spectroscopy. NIRS-PLS is used to quantitative analysis the active ingredient in fungi. The results indicate that it is feasible to apply NIRS combine with PLS to non-destruction determination of the active constituent of Coriolus versicolor and Cordyceps sinensis. It can be generalized to on-line and real-time quality control in pharmacy. It also has actual application value on quality control of quantity production.