节点文献
激光诱导击穿光谱的单标样定量检测研究
Research on Quantitative Analysis of Single Sample Calibration Laser-Induced Breakdown Spectroscopy
【作者】 邓凡；
【作者基本信息】 华中科技大学 ， 软件工程（专业学位）， 2022， 硕士
【摘要】 激光诱导击穿光谱（简称LIBS）是一种新型的元素检测技术,因其具有快速、微损与原位检测等优势,已初步应用于冶金、环境监测、生物医疗和深海探测等领域。然而,LIBS传统多标样法通常需要5个以上标准样品建立定标曲线来完成定量分析,难以适用于地质勘探、考古以及太空探测等标样较难获取的领域;无需标样的免定标法定量误差通常大于20%,准确度较低。而单标样法融合了传统多标样法和免定标法的优点,仅需单个标样即可实现较高准确度的定量分析,但其定量准确度易受自吸收效应影响。因此,本文通过研究单标样法的选线方法,并创新性地提出适用于单标样的自吸收校正算法,进一步提高了单标样LIBS的定量准确度,取得的主要研究成果和创新点如下:（1）由于不同分析线受自吸收效应影响程度不同,本文研究了单标样法的选线方法。首先采用遗传算法优化出最佳分析线,与常用分析线的定量结果进行对比发现,铝青铜样品中元素（Al、Fe、Mn和Cu）和矿石样品中氧化物（Si O2、Al2O3、Ca O、K2O和Na2O）定量的平均相对误差分别从17.28%和36.99%降低至11.20%和21.16%。在此基础上进一步研究发现,当分析线的原子参数具备跃迁几率较低、下能级能量较高、下能级简并度较小和波长较短等特征时,该分析线受自吸收效应影响较小。因此直接选取具备以上特征的谱线作为分析线对于单标样LIBS是一种有效的选线方法。（2）采用选线方法仅能被动筛选出自吸收相对较弱的谱线。为进一步消除自吸收对谱线的影响,本文在选线的基础上提出了一种适用于单标样法的自吸收校正算法。该算法通过引入Hα参考谱线计算相对自吸收系数即可实现对待测元素谱线强度的精准校正。使用铝青铜样品和矿石样品验证该方法的有效性。实验结果显示,该算法使得两种样品定量的平均相对误差分别从11.20%和21.16%进一步降低至5.70%和12.65%。综上所述,本文所总结的选线优化方法可筛选出自吸收效应较弱的谱线,在此基础上通过自吸收校正算法可进一步消除自吸收效应对谱线的影响,大幅提高单标样法的定量准确度,该研究对LIBS技术的推广应用具有重要意义。
【Abstract】 Laser induced breakdown spectroscopy（LIBS）is a new element detection technology.Because of the advantages of fast,micro-loss and in-situ detection,LIBS has been preliminarily applied in metallurgy,environmental monitoring,biomedicine and deep-sea exploration.However,the traditional calibration strategies of LIBS usually require more than five standard samples to establish calibration curve for quantitative analysis,which is challenging to apply to areas where the standard samples are difficult to obtain such as geological exploration,archaeology and space exploration.The calibration-free LIBS（CF-LIBS）requires no standard samples,but the relative error of quantitative analysis usually exceeds 20%,which does not meet the actual demand.Single sample calibration LIBS（SSC-LIBS）combines the advantages of traditional calibration strategies and CF-LIBS.It can achieve high quantitative accuracy only by single standard sample.However,the quantitative accuracy of SSC-LIBS is susceptible to self-absorption effect.Therefore,the line selection strategy of SSC-LIBS was studied in this thesis,and the self-absorption correction algorithm suitable for SSC-LIBS was innovatively proposed,which can further improve the quantitative accuracy of SSC-LIBS.The main achievement and innovations are as follows:（1）Due to different analytical lines are affected by self-absorption effect in different degrees,the line selection strategy of SSC-LIBS was studied.Firstly,the optimal analytical line was optimized by genetic algorithm.Comparing the quantitative results based on optimal analytical line with those based on the common analytical lines,it is found that the average relative errors（ARE）of elements（Al,Fe,Mn and Cu）in aluminum bronze samples and oxides（Si O2,Al2O3,Ca O,K2O and Na2O）in ore samples decreased from 17.28%and 36.99%to 11.20%and 21.16%,respectively.On this basis,it is further found that when the atomic parameters of the optimal analytical line have the characteristics of low transition probability,high energy of lower level,small degeneracy of lower level and short central wavelength,they are less affected by the self-absorption effect.Therefore,directly selecting the analytical lines with weak self-absorption by atomic parameters can be used as an effective strategy for line selection of SSC-LIBS.（2）The line selection can only passively screen out the spectral lines with weak self-absorption effect.In order to further eliminate the influence of self-absorption on the spectral lines,a self-absorption correction algorithm suitable for the SSC-LIBS was proposed on the basis of line selection.The algorithm can correct the spectral line intensity of the test samples and standard samples by introducing the reference line（Hα）to calculate the relative self-absorption coefficient.Aluminum bronze samples and pressed ore samples were used to verify the effectiveness of this method.The experimental results show that the ARE of the two samples are further reduced from 11.20%and 21.16%to 5.70%and 12.65%,respectively.In summary,the line selection strategy summarized in this paper can filter out the spectral lines with weak self-absorption effect,and the self-absorption correction algorithm proposed on this basis can further eliminate the influence of self-absorption on the spectral lines and greatly improve the quantitative accuracy of SSC-LIBS,which is of great significance to the popularization and application of LIBS technology.