Analysis Of Complex Components In Mixed Tobacco Flavors By Moving Window Characteristic Peak Orthogonal Projections Resolution Algorithm And Flavor Database

Flavors and flavor material are indispensable material and products in modern social production, the recipe of flavors and fragrances is one of the core technologies in many enterprises.Tobacco flavors are often added into tobacco products to increase the flavor in order to make the cigarette has its own unique characteristics of aroma and achieve the purpose of coordination,it can also make the different grades and types of tobacco aroma to coordinate together.In current production of cigarettes, flavor recipe is the most important factor for quality and character of cigarette,while tobacco flavor is mixed by many natural and synthetic perfumes. So the research on the chemical components of natural perfumes and tobacco flavor may contribute to the analysis of flavor compositions and quality control of tobacco flavor.In this paper, the raw material flavor database was established, and two-dimensional data from gas chromatography/mass spectrometry was used to analyze the main components of the complex flavors and fragrances by Moving Window Characteristic Peak Orthogonal Projections Resolution Algorithm.In this paper, the method which can be used to analyze the monomer substances and other mixed compounds from complex cigarette mixtures was proposed,and the parameters related to the algorithm and the factors influencing the qualitative results were discussed in detail. After analyzing the complex mixed tobacco flavors from the point of view of chemistry, we turned into "odor space", the note attributes related to components analyzed already were listed by consulting related literature,approximate formula can finally be roughly obtained.The paper aimed at analyzing complex flavor by combing chemical analysis and odor analysis. This could help flavorists to improve the ability to control flavor quality, and also give them some technical supports.In chapter one, the classification of the natural perfumes and tobacco flavor was reviewed, and then, the comparison of the methods to extract tobacco flavor and chemometric methods commonly used in the analyses of flavors and fragrances were summarized.Finally,the odor indicators of tobacco flavor,such as the main aroma,note,and the flavor formula were discussed in detail. In chapter two,different types of flavors, such as monomer substances, oil,tincture,extract,were selected to research.The appropriate extraction method was optimized through the comparison and improvement of the commonly used methods.The overlapped chromatographic peaks were resolved by Heuristic evolving latent projections (HELP).The information of retention times and mass spectra in each flavor were extracted.In chapter three, the principle of a chemometric method named "Moving Window Characteristic Peak Orthogonal Projections" was introduced firstly,which can be used to analysis the recipe of tobacco flavor. And then the raw flavor material database which contained five hundred kinds of flavors and fragrants was proposed,it was essential to choose the right characteristic peaks of the tobacco flavor in database.In addition, how the database and Moving Window Characteristic Peak Orthogonal Projections Resolution Algorithm combined with each other to analyze the mixed tobacco flavor was systematic analyzed,the parameters related to the algorithm and the factors influencing the qualitative results are discussed in detail.In chapter four, one flavor which was mixed by monomers and essential oils and the other flavor which was mixed by tinctures and essential oils in complex base were analyzed by Moving Window Characteristic Peak Orthogonal Projections Resolution Algorithm,Gas Chromatography-Mass Spectrometry and flavor database, the results showed that this method could be applied to the formula analysis of actual flavor.In chapter five, after analyzing the complex mixed tobacco flavors from the point of view of chemistry, we turned into "odor space", the note attributes related to components analyzed already were listed by consulting related literature, approximate formula could finally be roughly obtained. In this chapter, we aimed at analyzing complex flavor by combing chemical analysis and odor analysis. It could help flavorists to improve the ability to control flavor quality, and also give them some technical support.