Identification Of Walnut From Different Regions Of China Based On The Determination Of Multi-elements And Statistical Methods

In this paper,a method of determination of ICP-MS elements in walnut samples was established,which provided a theoretical basis of the protection for walnut geographical indications.Thirty-five kinds of elements in walnut were determined by microwave digestion and microwave ashing method combining with inductively coupled plasma mass spectrometry(ICP-MS).One-way analysis of variance,principal component analysis(PCA)and linear discriminant analysis(LDA)were performed on the data of the multi-elements.A discriminant model of walnuts was established according to the content and composition of the multi-elements,and it successfully identified the walnut from different regions.The results showed that:(1)A method for the determination of 19 kinds trace elements in walnut was established by optimizing the conditions of microwave digestion and ICP-MS instrument.The detection limit was from 0.047?g kg-1 to 78.4 ?g kg-1,the recovery of standard addition was from 85.7% to 109.8%,the RSD was from 0.11% to 2.2%,and the measured values of beans standard material(GBW10021)verified by this method were within the recommended values.Therefore,this method for the determination of 19 elements in walnut was reliable and stable.(2)A method of the determination of 16 kinds of rare earth elements(REEs)in walnut was established by optimizing the conditions of Microwave Ashing and ICP-MS instrument.The detection limit is from 0.00059 ?g kg-1 to 0.12 ?g kg-1,the recovery standard addition was from 73.5% ~ 110.9%,the RSD was from 0.3% ~ 19.7%,and the measured values of soybean and beans standard materials verified by this method were within the recommended values.Therefore,the method is accurate and suitable for determination of trace REEs in walnut.The results of REEs in walnut show that the total content of REEs in walnut was from 2.7?g kg-1 to 11.9 ?g kg-1,furthermore,particularly the content of Sc,Y and La are differ greatly among the walnut groups.(3)The contents of 35 elements of 128 walnut samples from eight major producing provinces of China were determined by microwave digestion and microwave ashing method combining with inductively coupled plasma mass spectrometry.One-way analysis of variance,PCA and LDA were used to identify the walnut from different regions.In comparison,there were significant differences(p < 0.05)in the composition of multi-elements of walnut from different regions.PCA could represent most of the information about the original variables,but the classification performance of walnut was not satisfactory;LDA was an effective evaluation method to screen identification elements related to origin.Discrimination model of walnut from different origins and discrimination model of geographical indications(GI)and non-geographical indications(Non-GI)were established by LDA,the model achieved 99.2% and 95.7% discriminant accuracy,respectively.