Evaluation Of Infant Formula Quality Using Multispectral Techniques

Infant formula was an important substitute for human milk,which could provide comprehensive nutrition for infants.Due to lack of mother's milk,physical discomfort,lactation time and other constraints,breast feed cannot thoroughly satisfy infant's growth,thus infant formula was widely consumed instead.As the production and consumption of infant formula rising,people have paid more attention to their quality.The simulation of human milk represents valuation standard for infant formula,which was highly concerned by the whole society.Herein,development of detection method for infant formula quality was urgently needed.Chromatography and mass spectrometry provide main methods for milk powder quality detection,but their applications were inhibited by some disadvantages,such as complex pretreatment/ limited target scope/long analysis time and high cost.In the study,whey protein and aroma components were selected as key research objects,following principles of selecting typical factors and highlighting main contradiction.Then new multispectral methods were developed,these methods were conducive to grasp the key nutritional quality and sensory quality of milk powder thoroughly.To encounter determination need of whey protein content,data driven raman spectroscopy method was developed to analyze infant formula.Based on chromatography and mass spectrometry,multispectral techniques were developed to realize automatic and intelligent evaluation of whey protein quality in infant formula.Taking advantage of high throughput and orthogonal separation for solid phase micro-extraction and two dimension gas chromatography,multispectral techniques were developed to detect volatile aroma components using mass spectrometry.These methods were efficient to identify the characteristics of aroma components,and also provided important technical support for simulation of breast milk fragrance.The main contents and contributions of this dissertation were listed as follows:1.Quantitative method of whey protein was developed using liquid chromatography tandem mass spectrometry.In view of the stability of whey protein peptide,we converted the protein detection into peptide detection through enzymatic hydrolysis.Thus,we solved difficulties encountered in the protein quantitative caused by the complicated molecular structures.To improve accuracy of the quantitative results,the matrix effect was evaluated systematically,then an effective and simple strategy was proposed to reduce the matrix effect.The strategy reduced the mutual interference effect between peptides and stable isotope labeled peptides efficiently,which combined plastic vial injection and dilution injection.To improve the stability of the quantitative results,data driven chromatography tandem mass spectrometry was developed,which optimized the shape of chromatographic peak and improved the signal-tonoise ratio efficiently.The accurately quantitative analysis of whey protein was realized successfully.2.Data driven raman spectroscopy for non-destructive detection of whey protein was established.We innovatively introduced high density wavelet transform(HDWT)to the raman spectrum analysis of whey protein.Firstly,HDWT improved the spectral resolution based on oversampling of double time/frequency domain,and then reduced the mutual interference of overlapping spectra sufficiently.On this basis,template orient frog algorithm(TOFA)was developed to extract characteristic variables of raman signal accurately.Based on HDWT and TOFA,high quality multi-scale prediction model was developed.Results showed that data driven raman spectroscopy could analyze signal accurately and finally realized nondestructive detection of whey protein in infant formula.3.Identification method of characteristic aroma components in infant formula was established based on solid-phase micro-extraction(SPME)and two dimensional gas chromatography tandem time of flight mass spectrometry.Firstly,we got the best condition of SPME by single factor optimization.Using gas enrichment sampling,volatile components in infant formula were captured without solid matrix effect,and the pretreatment could be completed quickly.Relying on different polarity and boiling point of volatile components,twodimensional gas chromatography was applied to improve detect capacity with orthogonal separation.As a result,peak overlap and signal interference were also solved.The fingerprint spectrum analysis of aroma components was carried out using time of flight mass spectrometer.Based on chromatographic separation and mass detection,a high throughput method was established to detect volatile components in one needle injection.And then,aroma characteristics of infant formula were analyzed using principal component analysis(PCA),and the differences between infant formula and human milk were identified by PCA.Results showed that it was useful to simulate breast milk for infant formula.In conclusion,the study innovatively developed multispectral techniques for milk powder quality detection,and realized effective detection of whey protein and aroma components.The present work provided novel approaches for quantitative evaluation of infant formula.