| Kimchi is one of the most popular vegetable products in East Asian countries such as China and South Korea.However,the continuous fermentation of lactic acid bacteria(LAB)during storage and eating will cause the acidification of kimchi,which seriously affects the taste and quality.In this paper,through screening molecular blockers of LAB,the selected water-soluble substance sodium lactate was prepared into a sustained-release preparation which can be released in the kimchi environment system by adopting slow release technology.What's more,for the first time,the prescription and process of its sustained-release preparation were optimized by BP neural network technology.The specific research content and results are as follows:Part I.Screening of LAB molecular blockersThe molecular blockers of LAB were screened by investigating the growth inhibitory effect of the pollution-free,water-soluble and weakly alkaline substances sodium lactate,sodium glutamate and sodium linoleate on the growth of Leuconostoc mesenteroides and Lactobacillus plantarum and the changes of pH during kimchi fermentation.The results showed that sodium lactate showed double effects of inhibiting LAB and delaying the pH change rate of kimchi.Therefore,sodium lactate was selected as the molecular blocker of LAB.Part II.Development of molecular blocking preparations of LABSodium lactate microcapsules were prepared by emulsification-solvent volatilization method.Based on the results of single factor tests,response surface methodology was used to investigate the effects of drug-to-excipients ratio,emulsifier concentration,volume ratio of solution phase and curing time on medicine loading,encapsulation efficiency,yield and comprehensive score of microcapsules.What's more,it carries out verification tests,and the test results are as follows.(1)Taking the drug loading as the evaluation index,the obtained process conditions are as follows: drug-to-excipients ratio 1.82: 1,Span80 dosage 2.0%,volume ratio of solution phase 0.95: 1,curing time 5.14 h.The drug loading was 29.51%,the encapsulation efficiency was 70.95%,the yield was 80.57%,and the comprehensive score was 65.548.(2)Taking the encapsulation efficiency as the evaluation index,the obtained process conditions are as follows: drug-to-excipients ratio 1.33: 1,Span80 dosage 1.0%,volume ratio of solution phase 1.07: 1,curing time 5.10 h.The drug loading was 28.81%,the encapsulation efficiency was 73.07%,the yield was 81.28%,and the comprehensive score was 66.681.(3)Taking the yield as the evaluation index,the obtained process conditions are as follows: drug-to-excipients ratio 1.27: 1,Span80 dosage 1.0%,volume ratio of solution phase 1.12: 1,curing time 5.18 h.The drug loading was 28.85%,the encapsulation efficiency was 72.49%,the yield was 83.38%,and the comprehensive score was 67.029.(4)Taking the comprehensive score as the evaluation index,the obtained process conditions are as follows: drug-to-excipients ratio 1.17: 1,Span80 dosage 1.0%,volume ratio of solution phase 1.16: 1,curing time 5.19 h.The drug loading was 29.34%,the encapsulation efficiency was 72.84%,the yield was 83.07%,and the comprehensive score was 67.209.There are many shortcomings of subjectivity and local optimization when using response surface methodology to optimize multiple response values,which leads to poor predictability of the optimal prescription and preparation process of preparations.Part III.The application of BP neural network to the preparation of sodium lactate microcapsulesUsing Matlab 2017 a software,BP neural network model was constructed by selecting algorithm,number of hidden layer neurons,learning rate and momentum factor.What's more,the established neural network was simulated and tested with the preparation data of molecular blocking preparations of LAB.The results show that the prediction and optimization results of BP neural network are closer to the actual values.Finally,the experimental results obtained by neural network optimization are as follows: drug-to-excipients ratio 1.12: 1,Span80 dosage 1.36%,volume ratio of solution phase 1.15: 1,and curing time 5.05 h.In the meantime,the drug loadingwas 30.11%,encapsulation efficiency was 73.22%,yield was 85.61%,and comprehensive score was 68.315 point.The adaptive learning system of BP neural network makes it have higher fitting and prediction accuracy for multi-response value models.The release law under different conditions conforms to Higuchi equation,and the pH change caused by it conforms to Logistic equation.Part IV.The effects of molecular blocking preparations of LAB on the fermentation process of kimchi.The effect of sodium lactate sustained-release microcapsules on the shelf-life of kimchi was investigated.The results showed that it was recommended to use 2.0% sodium lactate microcapsule to treat kimchi,it could delay the change rate of pH and titratable acidity of kimchi,inhibit the growth of LAB and total microorganisms,and improve sensory score.Under the storage condition of 25?,the shelf-life of kimchi can be prolonged by 1 time.Under the storage condition of 4?,the shelf-life of kimchi can be prolonged to 1.3 times of the original shelf-life.In conclusion,this thesis prepared a slow-release formulation of a small molecule water-soluble drug,sodium lactate,and its release rule under different conditions all complied with the Higuchi equation.BP neural network was used to optimize the formulation and preparation process of sustained-release preparations for the first time,which overcoming the drawbacks of multi-index optimization of response surface methodology.The prepared microcapsules have stable technology and can prolong the shelf-life of kimchi at different fermentation temperatures.It provides detailed and reliable theoretical basis for the industrial production of kimchi and the preparation of small molecule water-soluble medicine sustained-release preparations. |
PDF Full Text Request