| Crayfish and its products with tasty flesh and high nutrition,are popular with the majority of consumers.However,in recent years,the continuous expansion of industrial production scale and the massive discharge of wastes,which lead to an increase of heavy metal content in water,and heavy metal pollution for crayfish is also becoming more and more serious.China has a large producer and consumer of crayfish,detection and control of excessive intake of heavy metals is the key to ensure the green and sustainable development of the crayfish industry.Traditional heavy metal detection methods with large-sacle instruments that are expensive price,high maintenance costs,complicated process,and operation with professional personnel,are not suitable for heavy metal detection with large batches of crayfish.Due to easy operation,inexpensive price,low maintenance cost,and high efficiency,etc.,it has a very important practical significance if near-infrared spectroscopy will be used to determine the heavy metals with large batches of crayfish.In this study,macroporous adsorption resins were used to adsorb arsenic,lead,and cadmium in crayfish and the NIR spectra of adsorption resin with arsenic,lead,and cadmium was collected.A regression model was established by using the PLS method and the predictive performance of the model was also investigated to achieve the application of NIR spectra to predict the content of arsenic,lead and cadmium in crayfish.The work may provide technical support for the detection of heavy metals with large batches of crayfish.The main research contents and conclusions are as follows.1.Modeling for quantitative prediction of arsenic in crayfish.The arsenic selective resin SQ-407 was placed in 100 m L of arsenic standard solution(100μg/L),and adjusted to p H 4.0～9.0 and adsorbed at 35～50°C for 60～150 min,the adsorption conditions were optimized.Furthermore,fresh crayfish were freeze-dried and microwave-digested,then the digested solution was subjected to SQ-407 resin adsorption with optimized adsorption conditions.The NIR spectra of the adsorbed resin were collected and the spectra were optimized by using different pretreatment methods(First derivative spectroscopy,Wavelet transform,Standard normal variate,and Multiplicative scatter correction,respectively),and the regression model between arsenic content and NIR spectra was established by the partial least square method,and the model was validated to explore the prediction accuracy of the model.The results showed that the high adsorption rate of arsenic(99.0%)was presented at p H7.0 and 4°C for 90 min.After pretreatment with first-order derivatives,the established regression model had an accurate prediction for the content of arsenic,with the RMSECV of 0.04 and R_cof 0.99 for the correction set,the RMSEP of 0.03and R_pof 0.99 for the prediction set,and the average recovery of 102%for the predicted arsenic content compared to the reference value.These results indicated that the regression model based on resin adsorption combined with NIR spectroscopy may accurately predict the arsenic in crayfish.2.Modeling for quantitative prediction of simultaneous lead and cadmium in crayfish.Heavy metals with lead and cadmium are divalent cations with similar chemical properties.In order to shorten the detection time and improve the detection efficiency,the simultaneous detection of lead and cadmium in crayfish using resin adsorption combined with NIR spectroscopy was further investigated on the basis of the detection of arsenic.The macroporous adsorption resin D405 was placed in 100m L of standard solutions containing the same concentration(100μg/L)of lead and cadmium,and the p H value of the solution was adjusted to 4.0～9.0.Then the resin was adsorbed at 35～50°C for 60～150 min,and the adsorption conditions of the resin were optimized.The microwave digest of the crayfish was subjected to resin adsorption at optimized conditions,while the NIR spectra of the resins adsorbed with lead and cadmium were collected,and the spectra were optimized using different pretreatment methods(First derivative spectroscopy,Wavelet transform,Standard normal variate,and Multiplicative scatter correction,respectively),then the band was selected using competitive adaptive reweighted sampling.The regression model between lead and cadmium concentrations and NIR spectra was established separately by using the partial least square method,and the model was validated to explore the prediction accuracy.The results showed that the high adsorption rates of lead and cadmium were 99.5%and 99.7%at the optimized conditions(p H 7.0,40°C,and 90min),respectively.After pretreatment with WT combined with band selection,the established regression model had an accurate prediction for the contents of lead and cadmium,with the RMSECV of 0.07 and 0.09 and R_cof 0.98 and 0.98 for the correction set,the RMSEP of 0.06 and 0.10 and R_pof 0.98 and 0.97 for the prediction set,and the average recoveries of 103.7%and 103.5%for the predicted lead and cadmium contents compared to the reference values,respectively.These results indicated that resin adsorption combined with NIR spectroscopy can be used for the simultaneous detection of lead and cadmium in crayfish,which may provide a guideline for the rapid detection of lead and cadmium with large batches of crayfish.3.Application and method validation of NIR spectral prediction model.Crayfish samples from different regions in the Yangtze River basin(Hefei,Chengdu,Wuhan,Yancheng Qianjiang,and Changsha)were collected,and the NIR spectroscopy model established was used to predict the arsenic,lead,and cadmium contents of whole shrimp and different parts(meat,shell,and offal),to analyze the contamination status of heavy metals in crayfish in different regions;the predicted results were also linearly fitted with the detection results of ICP-MS method(GB5009.268-2016)to examine the accuracy of the proposed model in the application of the actual system.Furthermore,the accuracy and stability of the NIR spectrometry method were analyzed through determining the precision and reproducibility,and compared with the ICP-MS method,to clarify the feasibility for the determination of heavy metals in crayfish.The results showed that the arsenic content was exceeded the range of Chinese national standards(GB2762-2017)in crayfish samples from different regions,but lead and cadmium with low contents were within the Chinese national standards.The enrichment pattern of arsenic,lead,and cadmium in crayfish from different regions had the same trend:offal>shell>meat.In addition,the contents of three heavy metals with meat were within the range of Chinese national standards,while lead and cadmium contents presented a low level.The correlation coefficients(R~2)all reached above 0.95 of the fitted curves,between the model predicted results and the ICP-MS method detected results.In precision test of NIR spectrometry,compared to the relative deviations(1.20%,1.88%,and 2.38%)of ICP-MS method detection,the relative deviations(3.71%,3.70%,and 4.87%)of the six replicate tests were higher,but less than 5%,and had good precision within the existence of permissible errors.In the repeatability test of NIR spectrometry,due to the influence of other interfering components in the spectral information,the relative deviations(2.29%,4.08%,and 2.70%)of the six parallel tests were higher than for ICP-MS method(1.45%,2.04%,and 1.05%),but all of which were less than 5%,suggesting that NIR spectrometry has a good repeatability.These results indicated that resin adsorption combined with NIR spectroscopy can be applied to the determination of heavy metals in crayfish in practical production. |
