Study On Quantitative Analysis Methods In The Enrichment Process Of Chelating Resin

Chelating resin,as a new type of material for filtering and purifying heavy metals in water,is widely used in wastewater treatment plants to test whether wastewater treatment meets standards.By detecting the quality of heavy metals enriched by chelating resins,one can understand their working status,so detecting the quality of heavy metals enriched by chelating resins is becoming increasingly important.However,traditional detection methods are slow in speed,low in accuracy,and cannot detect quickly.They can only detect the chelating resin after enrichment is completed,and cannot determine whether the chelating resin has reached saturation during the adsorption process and is replaced,resulting in unsatisfactory wastewater treatment results.Laser induced breakdown spectroscopy(LIBS)can detect the quality of enriched heavy metals during the chelating resin enrichment process,and LIBS technology is fast and can achieve rapid and in-situ determination.However,during the enrichment process,the detection of enriched metals results in changes in the mass of heavy metals enriched by chelating resins per unit time,resulting in low accuracy and poor repeatability in LIBS quantitative analysis.In response to the above issues,this article studies the trend of enrichment quality changes per unit time during the enrichment process,summarizes the impact of enrichment quality changes on quantitative analysis,and proposes a segmented analysis method.The Ada Boost integration algorithm is used to improve the accuracy of quantitative analysis of chelating resin enrichment quality.The specific research content is as follows:(1)The trend of mass change over time during the enrichment process:Taking Cu and Mn elements as research objects,the influence of experimental parameters on LIBS spectra was analyzed.By studying the changes in spectral intensity during the enrichment process,the variation law of mass change with contact time for amino phosphate chelating resin enrichment was summarized.This article analyzes the influence of experimental parameters on LIBS spectra through signal-to-noise ratio and spectral intensity analysis.Under the conditions of laser energy of 100 m J,3D mobile platform speed of 12 s/mm,and solution flow rate of 2 ml/min through chelating resin,LIBS spectral data is the best.As the contact time increases,the specific trend of change shows three stages.Phase 1:During the initial1-10 minutes of reaction,the chelating resin adsorbs Cu and Mn slowly,and the adsorption mass gradually increases per unit time;The second stage:10 minutes-saturation,with a saturation time of 16 minutes for Cu and 14 minutes for Mn.The adsorption mass tends to stabilize per unit time;Stage 3:After the chelating resin is saturated,the mass of Cu and Mn elements does not change.The mass of adsorbed Cu element is 42.34 mg,and the mass of adsorbed Mn element is 40.22 mg.(2)The impact of changes in enrichment quality on quantitative analysis:A univariate model was used for quantitative analysis of spectral data.The results of the quantitative model showed that when the concentrations of Cu and Mn elements were higher than 10wt.%,the fitting coefficients of the univariate model reached above 0.99.When the concentration is below 10 wt.%,the fitting coefficients of the univariate models for Cu I324.75 nm and Mn I 403.07 nm are 0.9174 and 0.9327,respectively;The RMSEC is 0.5751wt.%and 0.6296 wt.%,respectively;RMSEP is 0.6235 wt.%and 0.7785 wt.%respectively;ARE were 6.56%and 7.29%,respectively.At high concentrations,changes in enrichment quality have little impact on quantitative analysis,while at low concentrations,changes in enrichment quality have a greater impact on quantitative analysis.Therefore,quantitative analysis is divided into two stages for consideration.(3)Improve the precision of quantitative analysis of chelate resin enrichment quality:for nonlinear low concentration data,support vector machine regression(SVR)was used to establish a calibration model.Compared with the single variable model,R~2 was significantly improved.R~2 of Cu and Mn elements was 0.9873 and 0.9664,but it could not meet the requirements of experimental precision and overfitting occurred.On the basis of SVR,the Ada Boost algorithm is integrated with SVR to form a strong learner to improve the accuracy of quantitative analysis,with R~2 being 0.9913 and 0.9907,and RMSEC being 0.1064 wt.%and 0.1786 wt.%,respectively;RMSEP is 0.1795 wt.%and 0.22515 wt.%,respectively;ARE were 3.59%and 3.96%,respectively.The results indicate that the Ada Boost algorithm can effectively solve the problem of inaccurate quantitative analysis at low concentrations caused by changes in the enrichment quality of chelating resins.Finally,using a univariate model and segmented analysis method,the Ada Boost integrated SVR model was applied for quantitative analysis at low concentrations,and a relatively simple internal standard method was used for analysis at high concentrations to obtain a high-precision quantitative analysis model for chelating resin enrichment quality.This article focuses on the impact of changes in enrichment quality during the chelating resin enrichment process on the quantitative analysis of LIBS.Analyzing the trend of enrichment quality over time during the enrichment process and using different calibration models for segmented analysis to improve the accuracy of LIBS technology in quantitative analysis of enrichment quality is of great significance for wastewater treatment.