The Rapid Determination And Evaluation Of Food Quality Using Fluorescence Spectroscopy

With the continuous improvement of people’s living standard,the requirements for food quality are also becoming higher and higher.But at present,food quality and safety incidents take place occasionally in China.Therefore,it’s especially important to have a quick,real-time and accurate detection method.Through using the detection method of fluorescence technique,this paper analyzes the properties and qualities of some common foods including milk,beef and edible oil.The details of research contents and results are as follows:1.As the fluorescence properties of substances are closely related to environmental factors,the influence of environmental factors such as solution concentration,pH value,temperature,solvent,sample pretreatment,ordered media and fluorescence quenchers on fluorescence spectra is studied respectively.The results indicate that:when the solution concentration is too high,inner filter effect may occur,making fluorescence intensity decrease,and only when the solution concentration is low enough,will the measured fluorescence intensity be in direct proportion to the concentration of fluorescent substances;there may be differences in the number of fluorescence characteristic peaks,the location of excitation wavelength and emission wavelength as well as the fluorescence intensity for the same substance in different pH conditions,different temperature conditions,different solvents and different sample pretreatments;ordered media may react with non-fluorescing substances or the substances with low fluorescence quantum yield to produce a derivative with strong fluorescence,and on the contrary,some fluorescence quenchers such as metal ions may make the fluorescence intensity of a substance decrease.2.The difference of fluorescence characteristics among three kinds of milk with different degrees of heating is studied.Compared with that in raw milk,in both pasteurized milk and UHT milk which have been heated,the fluorescence intensity of tryptophan significantly decreases,while the fluorescence intensity of Maillard reaction products and riboflavin slightly increases.And the change of fluorescence intensity of substances in milk can serve as a determination index for the degree of heating.The accuracy of qualitative discriminant analysis for milk with different degrees of heating by both EEMs combined with PARAFAC and SFS combined with PLS-DA can reach up to 100%.In the detection of the fresh milk and the pasteurized milk mixed with reconstituted milk by using 2nd derivative synchronous fluorescence spectroscopy,the accuracy of fluorescence intensity ratio method is 86.7%and 90%,respectively,and the accuracy of PLS-DA method is 100%and 96.9%,respectively.3.The quality change of beef during storage is studied by using fluorescence spectroscopy.There are two kinds of new fluorescence substances generated in the process of beef storage,locating at Ex/Em:320 nm/380 nm and Ex/Em:380 nm/470 nm,respectively.The two characteristic peaks are related with Maillard reaction and oxidation reaction in beef,and the two kinds of new fluorescence substances can be used to characterize shelf life and freshness of beef.The accuracy of qualitative model for beefs freshness based on surface synchronous fluorescence spectroscopy(with the fixed wavelength difference Δλ of 75 nm)combined with PLS-DA to distinguish the samples from calibration set and validation set is 96.61%and 94.44%,respectively,and the difference of beefs origin has no effect on modeling results.4.A single spectrum analysis approach may have some limitations in the application,so we try to combine fluorescence spectra with ultraviolet spectra,infrared spectra,near infrared spectra and Raman spectra to make qualitative discriminant analysis for the types of edible oil and conduct quantitative analysis for the acid value and peroxide value of frying oil,and then compare the results of multisource spectra with that of single spectrum.After the fusion of the five spectra,the accuracy of qualitative analysis for the types of edible oil can reach up to 100%;the Rc2 and Rp2 in the quantitative analysis model for the acid value are 0.945 and 0.927,respectively,and the RMSEC and RMSEP are 0.037 and 0.046,respectively;the Rc2 and Rp2 in the quantitative analysis model for the peroxide value are 0.878 and 0.788,respectively,and the RMSEC and RMSEP are 0.105 and 0.142,respectively;the above modeling results are all obviously better than that of single spectrum.