| Milk is a highly nutritious and widely consumed food.As people’s awareness of food safety has generally increased,how to effectively monitor and control possible harmful substances in milk has become an important issue closely watched by management,consumers,and dairy companies.Milk is a complex mixture,and it is challenging to achieve sensitive detection of low concentrations of hazardous substances in milk.Surface-enhanced Raman spectroscopy is an ultrasensitive detection technique with a high enhancement factor of 106-108 that enables the signal-to-noise ratio of the target molecule to be significantly increased in complex environments.In this paper,based on Surface-enhanced Raman spectroscopy,the amplification of the characteristic signals of the target is realized by developing the SERS substrates with high enhancement factor and selectivity to reduce the influence of Mie scattering and fluorescence effects.The Raman spectral properties of milk were experimentally investigated.In view of the physicochemical properties and molecular structure of the hazardous substances,the Raman enhancement mechanism in milk was explored.Based on the enhancement mechanism,SERS substrates with excellent performance were developed.The spectral detection was combined with intelligent algorithms to establish detection methods for antibiotics,non-protein nitrogen and preservatives in milk.In order to meet the needs of rapid detection in the field,a portable milk Raman detector was designed and applied for detection.The main research content of this paper is as follows:1.Research on the Raman spectral characteristics of the main monomers in milk and the Raman enhancement mechanism of hazardous substances.The Raman spectral properties of different brands of milk with different fat contents and the main monomers were investigated,and the main monomers constituting the Raman spectra of milk were identified.The DFT calculations have been carried out on the SERS substrate synthesized on the basis of covalent organic framework（COF）,and the Raman enhanment mechanism of harmful molecules in milk has been investigated in terms of their physicochemical properties and molecular structures,laying a foundation for the research of the detection methods.2.Development of methods for the detection of antibiotic residues in milk based on Surface-enhanced Raman spectroscopy.The charge distribution of antibiotics in milk is uneven,Raman signals are weak and there are fluorescence interferences.The similar structures of antibiotics lead to their similar Raman spectra.By developing SERS substrates with high enhancement factor,combining with ligand modification technology and machine learning algorithms,rapid detection methods of three antibiotics have been established.Based on the 4-amino 1-butanol-modified PDMS cavity,a rapid detection method for tetracycline without pretreatment was constructed,and the limit of detection was 0.28μg/L.The magnetic Fe3O4@COF@Ag substrate was developed,and combined with the principal component analysis and classification algorithm,the spectra of the quinolone antibiotics with similar structures were classified,and the limits of detection ciprofloxacin,levofloxacin,norfloxacin were 5.61×10-9 M,1.44×10-8 M,1.56×10-8 M.Based on the secondary modification of magnetic Fe3O4@COF@Ag substrate by MPBA,a detection method with excellent selectivity for amoxicillin was established,and the limit of detection was 3.64 n M.3.Development of methods for the detection of non-protein nitrogen adulterants in milk based on Surface-enhanced Raman spectroscopy.Non-protein nitrogen molecules have remarkable Raman features due to the special structure of triazine ring and cyano group,but,they are easily interfered by protein components and have low detection signal-to-noise ratios.By developing SERS substrates with high enhancement factors and strong interactions with non-protein nitrogen substances,three methods for the detection of non-protein nitrogen substances have been established.A sensitive method for the detection of melamine in milk was established by developing a carboxyl-functionalized substrate,which was combined with melamine through hydrogen bonding,and the limit of detection was 3.34×10-10 M.The hydrogen bonding transition mechanism between L-hydroxyproline and the substrate was investigated,and based on the significant effect of the mechanism on the charge transfer of L-hydroxyproline,a sensitive label-free method for the determination of L-hydroxyproline in milk was established,and the limit of detection was 0.13μg/m L.Based on the characteristic peaks of Raman spectra of dicyandiamide molecules located in the silent region of biomolecules,a hydroxyl-functionalized substrate was developed,and a sensitive method for the determination of dicyandiamide in milk was established with a limit of detection of 0.029μg/m L.4.Development of methods for the detection of preservatives in milk based on Surface-enhanced Raman spectroscopy.The Raman signal of preservative molecules is weak and the spectral signal-to-noise ratio is low in milk.By developing SERS substrates with good enrichment effect and strong anti-interference ability,the sensitivity of the detection methods was improved,and the detection methods of three preservatives were established.An amino-functionalized SERS substrate was developed,and based on the enrichment of benzoic acid molecules by the sub-micrometer microporous structure and the interaction between the amino group and the carboxylic acid group,the sensitive detection of benzoic acid in milk was realized,and the limit of detection was 0.13μg/m L.The amino-functionalized substrate was modified with cysteamine,which solved the problem of the difficulty of Raman enhancement of salicylic acid due to the intermolecular hydrogen bonding.A sensitive method for the detection of salicylic acid in milk with a limit of detection of 0.065μg/m L has been established.The hydroxyl-functionalized substrate was developed for the detection of sodium thiocyanate through the dual mechanism of hydroxyl group and cyano group,and sulfur atoms and Ag NPs.The binding sites between sodium thiocyanate and the substrate were increased,thus a sensitive method for the detection of sodium thiocyanate in milk was established,and the limit of detection was 0.057μg/m L.5.Development of a portable milk Raman detector.For the purpose of on-site detection,a portable milk Raman detector has been developed with the studied detection method.Melamine,benzoic acid and ciprofloxacin,which are among the three types of harmful substances,were used as application examples for rapid detection.The results showed that the limits of detection of melamine,benzoic acid and ciprofloxacin were 7.13 n M,3.92μM and12.89 n M,respectively,which met the requirements of national standards.In addition,the Raman spectra of pure milk,fresh milk,yogurt and prepared milk were detected by this instrument,and the rapid identification of different kinds of milk was realized based on the t-SNE method and the classification algorithms.This work enriches the mechanism of Surface Enhanced Raman spectroscopy and expands its application in the detection of complex systems.Meanwhile,this research can help the development of high-sensitivity detection technology for complex systems and support the development of food safety detection technology. |
PDF Full Text Request