Study On Laser Induced Breakdown Spectroscopy For Detecting Heavy Metals In Paint Coatings

Paint coatings are widely found on the surface of various daily necessities.The coatings may contain heavy metal elements that can migrate.They will enter the human body through oral and skin contact,which will damage human health.Paint coating will also destroy the ecological environment as domestic waste flows into soil and water.Therefore,it is of great significance to quickly and accurately measure heavy metal elements in product paint coatings in industrial production.It is often used to detect heavy metal detection in paint coatings.The detection efficiency is low,the damage to the sample is large,and it is difficult to meet the needs of on-site inspection.LIBS technology has the advantages of multi-element online detection and in-situ detection,and has been widely used in various fields in recent years.In this thesis,based on LIBS technology,a series of studies have been carried out to quantitatively detect trace heavy metals in paint coatings,starting with physical factors affecting LIBS spectra and spectral pretreatment and analysis methods that affect LIBS quantitative detection.Its main work and conclusions are as follows:(1)Introduced and analyzed the sources and hazards of heavy metals in paint coatings and the limitations of several existing paint coating heavy metal detection methods,and illustrated the application potential of LIBS technology in the detection of trace heavy metals in paint coatings.(2)Based on the in-depth study of the principle and method of quantitative analysis of LIBS,a LIBS experimental system for heavy metal detection of paint coatings was built.The LIBS spectra of yellow and red paint coating samples were measured.The results of spectral analysis showed that the two paint coatings contained Ca,Ba and Sr matrix elements.The spectral difference mainly came from the difference of pigments in the paint coating.(3)Based on the LIBS experimental system,the physical factors affecting the LIBS spectra of paint coatings(spectral acquisition delay,laser energy,focal plane position of lens,average number of pulses)were studied by taking the intensity and the ratio of signal to back of the characteristic spectra of heavy metals Pb and Cr as indicators.The results are as follows: the signal-to-back ratios of Pb and Cr elements reach the maximum at 1.8 and 2.2 us,respectively;when the laser energy is less than 86.5 mJ,intensity of the Pb spectral line is basically linear with the laser energy.When the laser energy is more than 86.5 mJ,the intensity of the spectral line changes little and tends to be saturated gradually;considering the intensity-to-signal-to-back ratios,it is considered that the focal plane has the best spectral quality at 1.5 mm below the sample surface.The RSD of the spectral intensity of Pb element is approximately exponentially attenuated with the average number of pulses.After the average number of pulses reaches 80,the RSD is basically stable at about 5%.(4)A univariate analysis model and a multiple linear regression model for Pb and Cr elements were established.Before univariate analysis,the analytical lines selected for Pb and Cr elements were 405.78 and 425.44 nm,respectively.The determined coefficients of the model were 0.951 and 0.953,respectively,and the RMSECV of the one-cross-check was 0.0069 and 0.00354,respectively.In the process of establishing multiple regression models of Pb and Cr elements,the effects of target elements and matrix elements were considered.The multivariate regression models with three,four and six variables were established respectively.The coefficient of the model was higher than that of the univariate analysis model.The more variables in the model,the higher the resolvable coefficient,the lower the prediction accuracy of the model,which indicates that the multiple linear regression model is prone to over-fitting.(5)Quantitative analysis of heavy metals Pb and Cr in paint coatings by competitive adaptive reweighted sampling(CARS)algorithm combined with partial least squares(PLS)method.Compared with the univariate analysis model and the multiple linear regression model,the CARS-PLS model of Pb and Cr elements is the best.The coefficient of the model can be as high as 0.991 and 0.998,and the RMSECV of the one-out cross-validation is 0.0039 and 0.0021 respectively.The above results indicate the combination of LIBS technology.The CARS-PLS model accurately quantifies heavy metal elements in paint coatings.