Based On NIR-Master For The Conventional Chemical Composition Analysis Model Of Flue-cured Tobacco

In tobacco industry production process, Ts, Rs, TKN, Nic, K and Cl were essential part of cigarette formula designing and quality control. And convenient methods for tobacco manufacture process analysis and quality control has been gotten more and more attention. NIR has been widely applied in field of tobacco in the recent ten years. On the one hand, profit from the rapid development of spectrum instrument, chemometrics and computer science, on the other hand is due to the near infrared spectrometer in the analysis such as test speed, nondestructive measurement, and quantitative analysis, no pollution, no chemical change simultaneously, low investment. So the near infrared spectral analysis as a kind of ’sustainable development’ analysis technology, effectively applied to a large number of samples of rapid measurement, in recent years, further promote the use of the tobacco industry.This study of408kinds of flue-cured tobacco of total sugar, reducing sugar, total nitrogen, nicotine, potassium and chlorine in six of the chemical composition of the near infrared quantitative analysis were studied. The satisfactory results indicated that the model of quantitative analyzing main chemical elements in cured leaf and growing leaf meets the analysis requirements and proves to be practically valuable. Compared with cireulating analysis and artifieial analysis, near infrared analysis will reduce costs and enhance effieiene. Total sugar and reducing sugar were uesd SNV+PLS; total nitrogen was used MSC+PLS; nicotine, potassium and chlorine were usd PLS for instrumental methods. And theirs pectra were recorded between4000and10000cm-1with8cm-1of resolution, moreover, the average of spectral data of64times scan was obtained.Through the experiment, the establishment of Ts, Rs, TKN, Nic, K and Cl six main chemical composition of the near infrared quantitative analysis model. Checked by100kinds of flue-cured tobacco, and each component average relative error were:5.94%,6.01%.4.88%,7.97%,3.66%,4.11%, less than10%. The satisfactory results indieated that the model of quantitative analyzing main chemical elements in cured leaf and growing leaf meets the analysis requirements and proves to be practically valuable.