【作者】 李军；

【导师】 熊远钦； 毛友安；

【作者基本信息】 湖南大学 ， 化学工程， 2013， 硕士

【摘要】 卷烟是最为广泛使用的烟草制品,它是通过吸食来达到使用效果的,感官品质是其重要的商品属性。随着社会各界对吸烟与健康问题的日益重视,我国烟草行业持续推行“减害降焦”工作,逐年降低了国产卷烟焦油、有害成分的释放量。在提高卷烟安全性的同时也带来了卷烟香气、吸食品质的下降,因此如何对损失的感官品质进行补偿成为我国烟草行业近期的研究重点。当前,我国烟草工业企业在烟草及其制品的品质控制、优质烟叶原料替代、中式卷烟风格特征剖析等方面的研究开发工作,仍然处于相对落后的状态,直接影响着烟草工业的发展后劲,这些方面还有大量的研究工作要做。以上几个方面的工作都离不开对卷烟烟气的分析研究,特别是烟气对感官品质贡献研究。为了改变这种技术落后的现状,充分利用有限的烟叶原料资源,维护和不断提升烟草制品的产品质量,不断研究开发符合消费者需求的新产品,使我国烟草制品在国际市场竞争中占有一席之地,很有必要进一步深化卷烟烟气分析技术的研究开发。本论文设计了一种模拟人体感觉器官吸收卷烟主流烟气的装置,并应用于主流烟气中的香味成分测定以及卷烟感官质量指标预测模型的构建。具体研究工作如下：(1)根据吸烟过程中主流烟气流经人体呼吸系统的生物物理顺序、路径、接触过程及接触方式等特点,分析出如何模拟人体感觉器官(口腔和鼻腔)对烟气进行捕集,确定仿生吸收装置关键参数及仿生吸收液制备方案。(2)模拟抽吸过程,完成全套仿生吸收装置的设计和制造,装置运行稳定而且具有较好仿真效果。通过试验确定模拟口腔、鼻腔吸收液的成分、使用量,以及仿生吸收装置其他配套参数。(3)利用仿生吸收装置对有代表性的39种卷烟样品主流烟气进行仿生吸收；根据烟气仿生吸收物质的组成和含量特点,使用合适的预处理方法,将仿生吸收样分为中性、碱性、酸性和脂溶性部分,采用优化后的分离分析条件使用GC、GC/MS技术分别对上述4部分进行分析和研究,鉴定出大部分对卷烟感官品质有重要作用的香气和香味物质,并对其定量。另外,对口腔仿生吸收液进行了pH测定。(4)利用化学计量学方法,深入探讨了基于卷烟烟气仿生吸收液化学成分组学数据预测卷烟感官品质的可行性,采用粒子群优化算法结合支持向量机构建了具有很高的拟合精度与很强推广能力的卷烟香型特征判别模型和感官指标回归模型。更多还原

【Abstract】 Cigarette is the most widely used tobacco products, it is by taking the smoke to achieve the effect, the sensory quality is an important product attributes. Along with the social from all walks of life to pay more attention to the problem of smoking and health, China Tobacco Industry continued implementation of the "tar and harm reducing" work, year by year to reduce the emission of cigarette tar and harmful components, to improve the security of cigarette.But the cigarette aroma and taste quality declined at the same time. So how to compensate the loss of sensory quality has become the focus of China Tobacco Industry’s recent research. At present, research and development in tobacco and tobacco products quality control, quality tobacco raw material substitution, Chinese style cigarette characteristic analysis and other aspects of the work of China Tobacco Industry, is still in a relatively backward state, directly affects the development potential of the tobacco industry, these areas have lots of work to do. The above aspects of work are inseparable from analysis of cigarette smoke, especially the research of smoke contribution to the sensory quality. In order to change the backward status, make full use of the limited tobacco raw resources, maintain and continuously improve the quality of tobacco products, continuous research and development of new products to meet consumer’s demand, make China tobacco products more competitive in the international market, it is necessary to go further to study and develop the cigarette smoke analysis technology.This thesis designed a device simulation of the human sensory organs to trapping mainstream cigarette smoke, and application the device for determination of aroma components in mainstream smoke. Specific research works are as follows:(1) According to the physical order, mainstream smoke in the smoking process flows through the human respiratory system path, contact process and contact way, analyze how to simulate the human sensory organs (oral and nasal) to trapping of the smoke. Determine the key parameters of bionic absorption device and bionic absorption liquid preparation scheme.(2) Simulation suction process, complete the design and manufacture of the full set of bionic absorption device. Device is stable and has good simulation effect. Determine the simulated oral, nasal absorption of liquid ingredients, through the test of usage, as well as bionic absorbing device supporting parameters. (3)Use bionic absorption device trapping the mainstream smoke of39brands of cigarettes. According to the characteristics of smoke bionic absorbing material’s composition and content, apply appropriate pretreatment method, bionic absorbing sample divided into neutral, alkaline, acid and fat-soluble part. Using GC, GC/MS technology analysis and studies the4part by optimized separating conditions. Identified and quantified the majority aroma and flavor components in mainstream smoke that had an important contribution to the cigarette sensory quality. In addition, determined the pH values of the bionic oral absorption liquid.(4) Use chemometric methods, based on bionic cigarette smoke absorption liquid chemical omics data, to study the feasibility of predict the cigarette sensory quality. By comparison, we selected particle swarm optimization algorithm with support vector machine to built prediction model.The model shows good accuracy and capacity when predicting cigarette flavor characteristics and sensory quality.更多还原