Detection Of Banned/Restricted Azo Pigments In Food Based On Surface Enhanced Raman Scattering Spectroscopy

Most of the industrial used pigments are synthetic dyes,which usually are linked to cancer,teratogen and other diseases.Therefore,the use of these pigments in food is strictly prohibited/restricted.With the concerns about food safety issues increasing,many countries have set strict regulations on the types,dosages,and application scopes of pigments that could be added in food.However,the problem of unreasonable/illegal use of pigments in food still exists and has become one of the major food safety issues that needs to be addressed promptly.There is an urgent need for strengthening the government supervision over pigments in foods and establishing fast and accurate detection methods for pigments.Surface-enhanced Raman scattering/spectroscopy?SERS?has been shown great potential as a fast and sensitive detection technology in many fields such as material science,chemistry,biology,and medicine,and has attracted great interest from researcher in these fields.Compared with traditional analysis methods,SERS has some unique advantages,such as small sample size requirement,less interference from moisture,high sensitivity,providing molecular specific/fingerprint spectrum for a target compound and so on.SERS has the potential to be widely used as one of the most important tools for detecting trace substances.Therefore,the application of SERS in addressing food safety issues has caught enoumous attention from rsearchers in the field over the past decade.In this project,the banned/restricted azo pigments in food?Acid Orange II,Sudan Red I-IV,and Allura Red?were analyzed with SERS combined with chemometric methods.Different nano-substrates were employed as SERS substrate for rapid detecton of the pigments.The main contents and results were as follows:?1?Four types of SERS substrates with different morphological characteristics were fabricated using different synthetic methods and conditions.The four colloidal substrates were gold nanoparticles?Au NPs?,Au-Ag core-shell nanoparticles?Au-Ag NPs?,gold nanorods?Au NRs?as well as Au and/or Ag core-shell nanostars?Au/Ag NSs?.These substrates were characterized by UV-VIS spectroscopy,transmission electron microscope,and molecular probe.Four different sizes of Au NPs?diameter:23±2 nm,49±5 nm,64±6 nm,102±9 nm?were synthesized and the Au NPs had relatively uniform spherical shapes or close to spherical shapes.The preparation method for Au NPs was rather simple,controllable,and repeatable.Three different sizes of Au-Ag NPs?diameter:64±4 nm,73±5 nm,90±5 nm?were synthesized,which all had the same size of Au seeds?diameter:20±2 nm?but different thickness of Ag shells.Au-Ag NPs had relatively uniform spherical shapes.The Au NRs fabricated had three different aspect ratios,which were 1.8,2.4 and 3.5.The synthetic process for Au NRs with relatively uniform spherical shapes was somewhat simple and controllable.Three different methods including Au-seed growth,Ag-seed growth,and one step method were employed to fabricate Au/Ag NSs,but unable to get Au/Ag NSs with relatively uniform and reproducible shapes.The stability of the Au/Ag NSs and enhancement effects for the molecular probe also needed to be improved.Therefore,Au/Ag NSs was not used as a SERS substrate for the subsequent analysis of pigments.?2?The SERS enhancement effects for Acid Orange II as influenced by the wavelength of laser sources?633 nm and 780 nm?and the type of nano substrates(including a commercial product Klarite^TM,Au NPs,and Au NRs)were evaluated.The results showed that the excitation laser source with a wavelength of 633 nm was more effective for the SERS detection of Acid Orange II.The enhancement effects of Au NPs for Acid Orange II were influenced by the size of nanoparticles,and the smaller Au NPs?18.0±2.0 nm?was better.The lowest detection concentrations of Acid Orange II were0.2 mg/L,0.1 mg/L and 0.1 mg/L,respectively,correspoinding to the use of Klarite^TM,18.0±2.0 nm Au NPs and Au NRs with aspect ratio of 1.8.The Au NPs was further used to analyze the standard solutions of Acid Orange II and apple juice extraction spiked with different levels of Acid Orange II.The linear correlation between the intensity for each of the nine major SERS characteristic peaks(466,598,987,1180,1231,1337,1385,1499,and 1597cm^-1)and the concentration of Acid Orange II in either the standard solution system or apple juice extractions was established and used as simple linear model system for predicting the amount of Acid Orange II.For the standard solution system?0.05–10.0 mg/L,n=24?,the determination coefficient?R²?of the actural values versus predicted values was 0.950–0.985,and the root mean square error?RMSE?was 0.42–0.79 mg/L.The ratio of sample standard deviation to standard error of prediction?RPD?was 4.51–8.49.The best simple linear regression model was based on the characteristic peak of 987 cm^-1.For the apple juice extraction system?0.5–20.0 mg/L,n=24?,the corresponding R² was in the range of 0.950–0.982,RMSE was 0.91–1.57 mg/L,and RPD was 4.79–8.27.The simple linear regression model based on the characteristic peak at 1336 cm^-1 provided the best results.Partial least squares regression?PLS?was also employed for the quantitative analysis of Acid Orange II in the standard solution system and apple juice extractions.The PLS models could provide better predictability than the simple linear regression models?standard solutions:R²=0.993,RMSE=0.273 mg/L,RPD=13.06;apple juice extractions:R²=0.988,RMSE=0.746 mg/L,RPD=9.48?,indicating good predictability for quantitative analysis.?3?A rapid SERS method for detecting Sudan Red I-IV was developed using Au NPs and Au NRs as SERS substrates,and chemometric models were used for quantitative analysis of the pigments.The results showed that using Au NPs and Au NRs for SERS analysis resulted in similar sensitivity for Sudan Red I and II,and the lowest detectable concentrations for these two pigments were 0.5 mg/L and 1 mg/L,respectively.The lowest detectable concentrations for Sudan Red III were 0.5 mg/L and 1 mg/L,and that for Sudan Red IV were 0.1 mg/L and 0.5 mg/L,depending on the substrate used.The use of Au NRs provided better enhancement effects for Sudan Red III and IV,and thus higher sensitivity for the SERS analysis.Quantitative analysis models for Sudan Red I-IV were established using PLS.Based upon the SERS spectra collected via Au NPs,the indicators for the model predictabiliy for the four pigments were as follows:R²=0.900–0.998,RMSE=0.115–3.03 mg/L,RPD=2.563–67.52;while for that collected via Au NRs,the corresponding indicators were:R²=0.976–0.990,RMSE=0.241–1.48,RPD=5.098–8.160.The use of Au NPs outperformed Au NRs for the quantitative analysis of Sudan Red III,but the latter was better for quantative analysis of the other three Sudan dyes.?4?The SERS enhancement effects for Allura Red as influenced by the ratio of colloidal Au NPs to the pigement?1:3–3:1,v/v?and the concentration?0.1–0.2 mol/L?of salts?Mg SO₄ Mg Cl₂,used as aggregator?were investigated.The results showed that the best enhancement was achieved when the ratio of colloidal Au NPs to Allura Red used prior to spectral collection was 1:3.The optimal concentration of Mg SO₄ and Mg CL₂were 0.15 mo L/L and 0.1 mo L/L,respectively,which led to 4 and 5 times increase in the intensity of the characteristic peak at 1493 cm^-1,respectively.The lowest detectable concentration was 1 mg/L based on either one of the substrates under the optimal condition.PLS models for Allura Red were developed for quantitative analysis based on the Allura Red concentrations and SERS spectra of Allura Red solutions?1.00–20.0mg/L,n=25?collected by each of the two substrates.The R² of the actual values versus the predicted values of Allura Red by the models based on Au NRs and Au NPs were0.982 and 0.983,respectively,while the corresponding RMSE values were 0.913 and0.815 mg/L,respectively,and the RPD values were 8.505 and 9.528,respectively.SERS analysis methods for trace analysis of pigments?Acid Orange II,Sudan Red I-IV,and Allura Red?in food were developed.The results from this study could provide a new tool for food safety supervision and theoretical support for the application of SERS in trace analysis of various food products,which is of significance for promoting the application of SERS technology in food safety analysis.