Study On Gemstone Component Analysis Methods And Applications Using Laser Induced Breakdown Spectroscopy

In recent years,Laser-Induced Breakdown Spectroscopy?LIBS?,a new rising atomic emission spectroscopy,has been studied as an element analysis technique and has been showing huge potentials in loads of fields due to its unique advantages of remote distance,multiple elements,quickly in-situ and real-time on-line analytical capability.However,so far there have been few studies and applications of laser induced breakdown spectroscopy in gemology,and the analysis methods of LIBS is still can't satisfy actual applied demand at present,especial its measurement accuracy and stability are not as good as that of traditional spectroscopy technique.The main aim of this dissertation is to improve measurement accuracy and stability to promote LIBS practical applications in Gemology.The appropriate analysis methods and actual applications in gemological field were studied as follow.?1?The influence of LIBS system parameters?including delay time,laser energy and accumulated frequency?on the characteristics of jadeite plasma was studied and discussed to get the optimized parameters.The optimized delay time was 1.1us.The optimized laser energy was about 70mJ,and the optimized accumulated frequency was20 times.Additionally,the pretreatment methods of the LIBS spectral data were studied and compared.The results indicated that the normalization and calibration of the spectra had significant influence on the accuracy and stability of the subsequent quantitative analysis.?2?LIBS coupled with traditional quantitative analysis methods and multivariate statistical analysis methods were used to measure the multiple elements?including Na,Ca,Mg,Al,Fe and Cr?in jadeite simultaneously.Calibration curves of Na,Ca,Mg,Al,Fe and Cr were fitted based on the normalized intensity,integrated intensity and intensity ratio?Si element internal method?,and the linear relationship between the normalized intensity,integrated intensity,intensity ratio and the element contents was compared.The calibration curves were verified by using validation samples.It indicated that the optimized linear relationship was between the normalized integrated intensity and the element contents.In addition,the method of principal component analysis?PCA?was used for the similarity analysis of calibration samples and verification samples.The PCA model of the relationship between the spectral data of the jadeite samples and the content of the elements was established,and the calibration samples which were the most similar to the chemical composition of the validation samples were found as the standard sample.Then,the validation samples were quantitatively analyzed by using the normalization external reference method.Studies have shown that the normalization external reference method coupled with PCA basically met the requirements of quantitative analysis of jadeite,but the quantitative analysis of trace elements still needed further improvement.?3?The partial least squares analysis model of six elements in jadeite?including Na,Ca,Mg,Al,Fe and Cr?was established by using LIBS technique and partial least-squares method.The correlation coefficients of the PLS model of calibration set were 0.969,0.978,0.99,0.991,0.989 and 0.996 for Na,Ca,Mg,Al,Fe and Cr respectively,and the root mean square errors of the correction set were 0.121,0.101,0.089,0.054,0.031 and 0.001.By using the established PLS model,the composition of6 jadeite samples in the verification set were predicted.The absolute errors of predicted content of Na element were less than 0.3%,and the relative errors were less than 3%.The absolute errors of predicted content of Al element were less than 0.6%,and the relative errors were less than 6%.The experimental value was consistent with the actual value.The other elements'relative error was also small.It indicated that the content of element in the jadeite sample could be determined by LIBS technique combined with PLS and the PLS model had good applicability and validity.PLS method basically met the requirement of the quantitative analysis of the jadeite samples.?4?Omphacite samples were measured by LIBS,and the optimized external standard method was used to analyze the content of samples.Crystal chemical formula of the sample was gotten from measure results.The ratio of the total number of different crystal cations coefficientwas?0.913 to 1.075?:?0.902 to 1.182?:?1.909 to 2.087?,which was basically consistent with the jadeite cation crystal-bit coefficient 1:1:2 theoretical value.The result verified the feasibility of LIBS quantitative distinction between jadeite and omphacite.It indicated that the percentage of SiO₂ in jadeite and omphacite was basically stable at around 59%,and the percentage of Na₂O and Al₂O₃ in jadeite was about 13%and 23%respectively.The total percentage content of CaO,Fe O and MgO and was less than 4%.However,the percentage content of Na₂O and Al₂O₃ in omphacite was less than jadeite,and the content of CaO,FeO and MgO could reach to 27%.In addition,in omphacite the percentage of Na₂O and CaO showed negative linear correlation?correlation coefficient r1=-0.9901?,and the percentage of Al₂O₃ and MgO showed negative linear correlation?correlation coefficient r1=-0.966?.?5?In this work,the univariate and the partial least square?PLS?techniques were used to analyze the LIBS spectra of Myanmar jade samples,and to build calibration models were established to predict the content of Ca and Al.The result indicated that PLS was able to significantly improve the analytical results in comparison with the univariate technique.The normalized root mean square error?NRMSE?and r² of the univariate models were 1.14 and 0.93 in calibration and 1.43 and 0.92 in prediction for Ca and 3.04 and 0.63 in calibration and 3.59 and 0.58 in prediction for Al.For the PLS models using the spectral range 197.33 to 762.8nm,the NRMSE and r² were 0.81 and0.97 for Ca,1.08 and 0.95 for Al in calibration respectively,and 1.36 and 0.92 for Ca and 1.79 and 0.89 for Al in prediction respectively.The results indicated that PLS was able to improve the quantitative analytical capability of LIBS for Myanmar jade samples.?6?Using laser-induced breakdown spectroscopy,quantitative analysis of iron?Fe?in Myanmar jades was studied in this work.Ion's LIBS spectrum at 275.57nm was chosen for quantitative spectral analysis,and Silicon's LIBS spectrum at 288.17nm was selected as internal standard line.The experimental results indicated that the relative standard deviation of the intensity of iron was between 1.4%and 8.3%,and the fitting correlation?R²?of was 0.979,and the maximum relative error of test samples was 10.6%when traditional method was used.While the internal standard method was used the relative standard deviation of the intensity ratio of iron and silicon is between0.9%and 5.7%,and the fitting correlation?R²?of is 0.989,and the relative error of test samples reduced to less than 7%.The analyzed result indicated that using LIBS coupled with internal standard method was much better to quantitatively analyze the content of iron in Myanmar jades compared with applying traditional method,as internal standard method could reduce the relative error of measurement.?7?Identifying the origins of nephrite from three different places was studied in this work,using laser-induced breakdown spectroscopy?LIBS?coupled with partial least squares discriminant analysis?PLS-DA?approach.146 nephrite specimens from Xinjiang?Hotan,Yutian,Qiemo?,Qinghai?Golmud?and Russia?Baikal?were selected as research subjects.111 of the specimens were chosen as calibration samples to build the PLS-DA model,and the rest 35 specimens were used as prediction samples to test the PLS-DA model.In this study,LIBS was used to test and analyze the element composition of nephrite specimens.Na,K,Al,Li,Be,Mn,Sr,Zr,Ba,Y and Ce were chosen as analysis object elements,and their spectral lines at 589.995nm?766.490nm?396.152nm?670.793nm?313.042nm?257.610nm?407.771nm?389.138nm?455.403nm?437.493nm?401.239nm were selected as analysis spectral lines.The LIBS spectrum of Si at 288.158nm was chosen as internal spectral line.Ratio?Rx?between the intensity of analysis spectral lines and the intensity of analysis internal spectral line was calculated.The matrix of independent variables composed by Rx was applied to calibrate and test the PLS-DA mode.The correlations between the predicted category variables of calibration or validation and the measured category variables were all remarkable with a correlation coefficient?r?over 0.90,and low root mean squared error of cross calibration and root mean square error of prediction?<0.29?.The discrimination accuracy for the nephrite from three different origins was 92%by PLS-DA model based on the validation set of samples.The results indicate that using LIBS coupled with PLS-AD approach can achieve a well recognition of the origins of nephrite specimens.