Detection Of Residual Drugs In Fish Muscle Based On Surface Enhanced Raman Spectroscopy

China is a large fish producing and exporting country.The fish production of China has been continuously increased on a yearly basis over the past three decades.The use of fish drugs plays an important role to ensure such a large scale of fishery.However,the excessive or illegal use of fish drugs could cause a series of food safety problems,which negatively affects public health,the stability of our society and the export trade of aquatic products.To date,the detection methods for these drugs are mainly liquid chromatography and mass spectrometry.However,these methods are quite costly and time consuming,which generally cannot meet the requirement for rapid detection of a large quantity of samples.In recent years,surface enhanced Raman spectroscopy（SERS）has been widely studied for rapid detection of harmful chemicals in foods due to its potential as a fast,accurate and high sensitive analytical tool.The development of SERS technology has been inseparable with the development of nano-substrates.The morphology,size and dispersion of nano particles have important impact on the enhancement effect of SERS.In this study,gold nanorods（Au NRs）with three different aspect ratios were synthesized via a seed growth method.SERS technology combined with Au NRs was used to analyze residual of dye drugs in different four fish species.Quantitative models of PLS were established to evaluate the possibility for quantitative analysis of fish drugs and the related metabolites via SERS.During the synthesis of Au NRs via seed growth method,the aspect ratios（AR）of2.2（50.8±3.1 nm,22.8±2.1 nm）,3.5（46.7±2.9 nm,13.3±1.1 nm）and 2.0（35.5±2.9nm,16.6±1.1 nm）of Au NRs were controlled by adjusting the amount of silver nitrate（50,240,and 480μL）which was an auxiliary growth agent.Results showed that excessive Ag NO₃ in the growth solution would inhibit the growth of gold nanorods.The surface plasmon resonance and morphological characteristics of the three gold nanorods were characterized by UV-visible spectroscopy and transmission electron microscopy,respectively.For malachite green,the Au NRs with AR 3.5 showed the best SERS enhancement effect,providing about 1.5 to 2 times stronger enhancement effect for the major characteristic peak at 1615 cm^-1.SERS combined with 3.5 AR Au NRs was applied to detect malachite green and methylene blue in fish（including snakehead fish,yellow catfish,tilapia fillets and black carp）extracts spiked with different levels of one of the drugs.The results showed that simplified pre-treatment steps could allow the extraction of malachite green or methylene blue from fish muscle within 40 min,and achieve a relatively ideal removal of other compounds in the muscle.The lowest concentrations detected for the malachite green and methylene blue in standard solutions were 0.1 ng/m L and 0.5 ng/m L,respectively.The detectable concentrations for methylene blue in snakehead fish yellow catfish,tilapia fillets,and black carp were 5.0 ng/m L,1.0 ng/m L,1.0 ng/m L and 1.0ng/m L,respectively.And the lowest detectable concentration for malachite green in snakehead fish yellow catfish,tilapia fillets,and black carp were 1.0,0.5,0.3,and 0.5ng/m L,respectively.Partial least squares（PLS）model was established to predict malachite green（0-20 ng/g）and methylene blue（0-100 ng/g）in fish extracts.The range of R²for the amounts of malachite green in the four fish extracts versus the model predicted values was 0.883-0.972.The root mean square error（RMSE）was 0.807-2.780ng/g,and the ratio of sample standard deviation to standard error of prediction（RPD）was 2.388-4.776.The R² for the methylene blue in the fish extract was 0.947-0.993,the RMSE was 0.811-10.80 ng/g,and the RPD was 3.484-9.575.The result demonstrated the feasibility of applying SERS technique for quantitative analysis of malachite green and methylene blue in fish muscle.Azure A,Azure B,and Azure C are three metabolites of methylene blue.Gold nanorods were used as SERS substrates for detection of a series of the Azure A,Azure B,and Azure C in standard solutions（0-30 ng/m L）and in tilapia fish extracts（0-30 ng/g）.A quantitative PLS model correlating the SERS spectral and the concentrations of target drug was developed.For the standard solution,the lowest detection concentrations of the Azure A,B and C,were 2 ng/m L,1 ng/m L and 1 ng/m L,respectively.Among the best PLS models for the three drugs,the range of R² value was 0.910-0.996,the RMSE value was 0.459-2.093 ng/m L,and the RPD value was 3.715-16.981.The linear relationship between the intensity of each of the four major characteristic peaks and the concentration of each drug was also evaluated（R²:0.864-0.996;RMSE:0.41-3.04ng/m L;RPD:2.82-18.93）.The lowest detectable concentration for all three drugs in the tilapia extract was 2 ng/g.The R² values of the best PLS models for Azure A,B and C were 0.993,0.997,and 0.974,respectively,while the corresponding RMSE values were0.80,0.51,and 1.65 ng/g,and the RPD values were 14.40,22.82,and 6.98,respectively.Relatively ideal quantitative analysis results were achieved.In summary,gold nanorods were synthesized and used as SERS substrate for determination of malachite green,methylene blue and its metabolites in different fish species.SERS test conditions were optimized to achieve the ideal enhancement effects.A rapid analytical method for drugs residue in fish extract was established through this study,indicating that the SERS technology combined with chemometric models can be a reliable method for quantitative analysis of trace residues of fish drugs within a certain concentration range.This provides a great potential for rapid quantitative detection of drug residues in aquatic products.