Discrimination Of The Origin And Quality Of Lycium Ruthenicum Murr And Real-time Detection Of The Liquid Based On Spectroscopy

Lycium ruthenicum Murr is a kind of medicinal and edible fruit.It is widespread cultivated in the northwestern of China.It is an edible fruit as rural or tea by pouring hot water.The total flavonoids,anthocyanins and other active ingredients accumulated in Lycium ruthenicum Murr from different habitats are different,which directly affects the quality of wolfberry fruit.Therefore,in order to distinguish Lycium ruthenicum Murr from different origin areas,chemical values and infrared spectra were collected from the dried fruits of the Lycium ruthenicum Murr in Qinghai,Xinjiang,Ningxia and Gansu provinces by near-infrared（NIR）and mid-infrared（MIR）spectroscopy combined with chemometric.The quality model,origin model and real-time monitoring model of Lycium ruthenicum Murr were established,which provide theoretical bases for rational brewing of Lycium ruthenicum Murr.These provide the basis for the reasonable brewing of black fruit Lycium ruthenicum Murr.The results of the study are as follows:（1）Collection and routine physical and chemical analysis of the samples of Chinese wolfberry.70 samples picked in 6⁷ months were collected from four main producing areas of Lycium ruthenicum Murr in Qinghai,Xinjiang,Ningxia and Gansu provinces.The spectral information of Lycium ruthenicum Murr and its tea was collected on the basis of optimizing the parameters of near infrared spectroscopy and mid infrared spectrometer.Then the acquisition parameters of NIR/MIR spectroscopy and the spectroscopy of Lycium ruthenicum Murr and tea were determined.Then the total flavonoids（TFC）and anthocyanins（TAC）contents in Lycium ruthenicum Murr were determined by conventional method.The results showed that the average TFC in the Lycium ruthenicum Murr were the highest from Gansu province,which is 46.152mg/g.The average TAC in the Lycium ruthenicum Murr was the highest from Qinghai province,which is 4.098 mg/g.The TFC and TAC from different origins were significantly different in Lycium ruthenicum Murr.The best soaking temperature of Lycium ruthenicum Murr is 100°C,and the TFC is in the range of 0.752-7.833 mg/g,and the TAC is in the range of 0.108-2.729mg/g in Lycium ruthenicum Murr tea.All above provide the following qualitative and quantitative model analysis,which were conducive to distinguish the origins and quality with model.（2）Lycium ruthenicum Murr were distinguished from different origins.Chemometrics including principal component analysis（PCA）,K-nearest neighbor analysis（KNN）,inverse-artificial neural network analysis（BP-ANN）and least squares-support vector machine（LS-SVM）were used to establish the discriminant model for four different origins of Lycium ruthenicum Murr with NIR and MIR spectroscopy.The PCA results showed that the contribution of variance of the first three principal components is 95.72%.The recognition rates of the calibration set were 100.00%for NIR and MIR spectroscope,BP-ANN prediction set recognition rates were 100.00%,97.50%,respectively.The high recognition rate and optimized model indicating that NIR/MIR spectroscopy could quickly and accurately identify the origins of Lycium ruthenicum Murr.（3）Rapid determination of total flavonoids and anthocyanin content in Lycium ruthenicum Murr with NIR and MIR spectroscopy.Two spectroscopy combined with synergy interval partial least squares（Si-PLS）were to predict the total flavonoids and anthocyanin content in Lycium ruthenicum Murr.The results showed that the quantitative model of the TAC based on the NIR spectroscopy subinterval(4462.72⁴921.59 cm^-1,4925.45⁵384.32 cm^-1,6776.35-7235.22 cm^-1,9086.12⁹541.13 cm^-1)was good.The r_c and r_p for the calibration set and prediction set were0.867 and 0.7928,RMSECV and RMSEP were 0.888 mg/g and 1.44 mg/g,respectively;the prediction model of the TFC based on the MIR spectroscopy subinterval(734.23⁸50.84 cm^-1,2251.24²367.36 cm^-1,3534.03³650.15 cm^-1,3883.88⁴000.00 cm^-1)was best,the r_c and RMSECV were 0.9265 and 6.130 mg/g in the calibration model,r_p and RMSEP were 0.8936 and1.010 g/mg in the prediction model.The NIR and MIR data were fused after Min-Max Normalization.And the results showed that r_c and r_p for the calibration set and prediction set were 0.9099 and 0.8985,RMSECV and RMSEP were 0.776 mg/g and 1.080 mg/g,respectively.It is the best Si-PLS model.It is of great significance to establish the rapid prediction model for the main active ingredients in Lycium ruthenicum Murr by spectroscopic techniques.（4）The main active ingredients were real-time monitored in the Lycium ruthenicum Murr tea from different origins.Firstly,the Si-PLS real-time monitoring models were established to predict the TFC and TAC in Lycium ruthenicum Murr tea by single NIR or MIR spectroscope and the results showed the correlation coefficient were above 0.800 but the models were weak.The TFC and TAC model were best based on fusion spectroscope combined with characteristic wavelength.The r_c,RMSECV,r_p and RMSEP were 0.9035,0.706 mg/g,0.892 and 1.020 mg/g in TFC model.The r_c,r_p,RMSECV and RMSEP were 0.8617,0.8550,0.318 mg/g and 0.377mg/g in TAC model.Secondly,the TFC and TAC were predicted using the best fusion model in Lycium ruthenicum Murr tea.The results showed that the release law of TFC and TAC in the tea was consistent with its actual release pattern,which indicated the prediction results of the model were reliable.Finally,when the Lycium ruthenicum Murr was soaked in water for five times,the TFC and TAC were 3.70%and 7.15%of the total content.In this case,the Lycium ruthenicum Murr should be replaced.This research accomplished that NIR and MIR spectroscopy can be effectively achieved accurate determination for the Lycium ruthenicum Murr base on different origins and quality with chemometrics.At the same time,it also established the TFC and TAC in Lycium ruthenicum Murr tea,and explained the release law in water.During daily drinking,the real-time monitoring model of the TFC and TAC can guide people how to drink.