Quantitative Monitoring Of Harmful Constituents In Mainstream Smoke By Near Infrared Spectroscopy Combined With Cambridge Filter

Cigarette smoke is a complex mixture composed of more than 4000 chemical specises detectable by current analytical techniques.Most of these chemicals are harmful to humans.Chromatography methods are generally used to analyze harmful chemical species in mainstream cigarette smoke.However,chromatography methods require tedious sample preparation and time-consuming instrumental procedures,and consume large amounts of chemicals(mobile phases)which may pollute the environment.They are not suitable for rapid detection of large quantities of samples.Recently,a new high-efficiency analytical technique—near infrared spectroscopy(NIR)has been receiving considerable attention in tobacco industry.NIR as a non-destructive analytical technique has many advantages over tradictional analytical methods such as high measuring speed,zero pollution,very low running costs and requirement of less or even no sample preparations which make it highly suitable for in-line process monitoring.Morover,NIR measurements of cigarette smoke contain spectral informations of multiple chemical specises.The combination of NIR technique with multivariate calibration methods can realize simultaneous determination of multi-components in large quantities of samples.The development of reliable multivariate calibration models for quantitative analysis of cigarette smoke using NIR instruments is generally difficult,time-consuming and costly.Therefore,it is preferable if the multivariate calibration models established can be widely shared among NIR instruments of different brands accross different laboratories within tobacco industry,and used for an extended period,without the need to replace them.The main aim of this thesis is to quantitative monitor harmful constituents in mainstream cigarette smoke by NIR technique combined with Cambridge filter,and to establish an intelligent operation and maintenance system for the wide sharing of NIR calibrtation models within tobacco industry.The main achievements of this thesis are as follows:1)A robust calibration model was established between NIR spectral measuremnts of the particle phase of mainstream cigarette smoke collected by Cambridge filter pads and the contents of BaP and TSNAs in cigarette mainstream smoke by partial least squares method(PLS).The experimental results showed that the established PLS calibration models realized rapid quantitative detection of harmful ingredients in cigarette mainstream smoke,and gave quite accurate predictions for BaP and TSNAs contents with average relative predictive error values in the range of 7%?10%with respect to the corresponding BaP and TSNAs contents determined by conventional standard methods.2)Crotonaldehyde is one of the main harmful species in cigarette mainstream smoke.Its quantification method currently adopted in tobacco industry is tedious and time-consuming.A PLS quantitative model was established between the NIR spectrum of the particle phase of mainstream cigarette smoke collected by Cambridge filter pads treated with 2,4-dinitrobenzene hydrazine solution and the content of crotonaldehyde in cigarette mainstream smoke.The experimental results demonstrated that the established quantitative model could provide quite accurate predictions of crotonaldehyde content.Its average realtive predictive error is about 8%,satisfying the requirement of qunantitative analysis of crotonaldehyde in cigarette mainstream smoke in tobacco industry.3)with a view to realize the long term usage of the NIR calibration models built for harmful species in cigarette mainstream smoke and their wide sharing accross different laboratories within tobacco industry,an intelligent operation and maintenance system was designed and established based on the calibraion model transfer methods developed in our laboratory.With the application of the intelligent operation and maintenance system,the NIR calibration models built by the central laboratoy can be shared with other laboratories within tobacco industry even when the brands of NIR instruments and measurement conditions differ across laboratories.The quantitative methods developed for BaP,TSNAs and crotonaldehyde in cigarette mainstream smoke have the advantages of high detecting speed and excellent prediction accuracy,and can realize simultaneous determination of multi-components in large quantities of samples.Therefore,it is reasonable to expect that they would becom promising and convenient entries for quantitative monitoring and control of tobacco products.