Highly Sensitive And Accurate Elemental Analysis Of Matters Based On New Plasma Spectroscopic Technologies

The quantitative elemental analysis is closely related to the production and life of human society.It is important to realize high-sensitive and high-accurate quantitative elemental analysis of matters.Laser-induced breakdown spectroscopy(LIBS),as a new type of atomic spectroscopy analysis technology,has been rapidly developed in recent years due to its advantages of fast,real-time,in-situ,non-destructive,remote measurement,simple operation,no sample pretreatment,and multi-element analysis.LIBS has been widely used in environmental monitoring,metallurgical process analysis,coal quality analysis,deep space exploration,biomedicine,material processing,geological survey and other many fields.However,the negative problems of LIBS in the quantitative elemental analysis include low accuracy,low sensitivity and requiring a series of standard samples,which block the development of LIBS.To save these problems,studies have been carried out based on the new plasma spectroscopic technologies of laser induced breakdown spectroscopy combined with laser induced fluorescence(LIBS-LIF)and laser ablation-spark induced breakdown spectroscopy(LA-SIBS).Univariate calibration curve method,multivariate calibration method and one-point calibration method have been adopted to improve the sensitivity and accuracy of quantitative elemental analysis of LIBS.This paper includes the following four aspects:1.LIBS-LIF experimental system with a home-made dye laser was built to detect trace copper in water and trace lead in wheat flour.To eliminate the matrix effect of water,a wood slice was used as a substrate to convert liquid samples into solid samples.The current experimental parameters of LIBS-LIF were optimized and a calibration curve of copper in water was established.The limit of detection was 3.6 ppb,which was three orders of magnitude better than that obtained by LIBS.The flour sample was pressed to pellet,the 266 nm laser with low pulse energy was used as ablation laser and the optimal time delay was explored to be 400 ns.Under the current optimal experimental conditions,the calibration curve of lead in wheat flour was established,the limit of detection was 73.8 ppb and the relative errors of unknow samples were less than 4.07%.The feasibility of using a multimode fiber to transmit laser pulses to build the LIBS-LIF experimental system with only one Nd:YAG laser was discussed.2.The effect of discharge channel on the analytical sensitivity and the resolution in LA-SIBS was studied.A gated pulsed high-voltage power supply was used in LA-SIBS,and the relationship between the time delay and the discharge channel was explored under external trigger mode.Under appropriate laser pulse energy and geometrical of the electrodes,the discharge channel could change V-shape to parallel when the time delay was 20?s-35?s.The intensity of plasma emission was significantly enhanced and no additional mass of samples were ablated with parallel discharge channel.Under the current experimental conditions,the detection limit of Mg,Cr,Pb,Al were 14.4 ppm,8.8 ppm,3.4 ppm and 5.1 ppm,respectively.It showed that LA-SIBS with parallel discharge channel can achieve high-sensitive and high-spatial resolution quantitative elemental analysis.3.The high repetition rate LA-SIBS technology combined with multivariable calibration method was studied to realize fast and high-accurate quantitative elemental analysis.The particle swarm optimization-extreme learning machine algorithm(PSO-ELM)was used to improve the accuracy of quantitative analysis and the 1 k Hz repetition rate laser was used as the ablation laser source to improve the analytical speed.The fiber spectrometer worked under non-gated signal recording mode was used to record the spectrum,and the median filter method was utilized for baseline subtraction.Multiple characteristic spectral lines of elements were selected as the input variables and the particle swarm optimization algorithm was applied to optimize the parameters of ELM model.The correlation coefficients of the prediction set were0.9996,0.9972 and 0.9999,and the root mean square errors of the prediction set were 0.0138%,0.0045%and 0.0095%,respectively.The relative error was in the range of 0.002%to 2.127%.The same experimental dataset was used to establish the univariate calibration curve and support vector machine calibration models for comparation,and the PSO-ELM model had the best prediction performance.It showed that the combination of PSO-ELM algorithm and high repetition rate LA-SIBS technology can realize fast,high accurate,and convenient quantitative analysis of multiple elements.4.The high repetition rate LA-SIBS coupled with one-point calibration(OPC)method was studied to realize rapid and high accurate quantitative elemental analysis in aluminum alloy samples.A matrix-matched standard sample was applied to synchronously correct the uncertainty of traditional calibration-free method,and the correction factors were substituted into the CF algorithm to analyze four unknown samples.The electron density was estimated from the Stark broadening of three appropriate ionic lines to be(1.26-1.54)×10¹⁷cm^-3.The Saha-Boltzmann plots were built and the average plasma temperature were estimated to be6,200±100 K.The Mc Whirter criterion and the Critoforetti additional criterion confirmed that the plasma was close to the local thermodynamic equilibrium state.Compared with the standard value,the average relative errors of Al,Mg,Cu,Cr,Mn and Zn in the four unknown samples are 0.065%,13.03%,11.97%,19.45%,9.77%and 7.81%,respectively.The results show that OPC method coupled with high repetition rate LA-SIBS can achieve rapid and high-accurate quantitative analysis of elements in aluminum alloys.