Degradation kinetics of tobacco—survey by fast thermogravimetry analysis

Abstract

Kinetic parameters of pyrolytic and thermal oxidative degradation for cellulose material as well as for tobacco are discussed in the literature extensively. Most of the experiments are conducted at low or moderate heating rates, 0.5–10 K/min typically. The experiments show two steps in the degradation curve when conducted in air. Different activation energies and pre-exponential factors can be derived. The first step of the degradation is related to pyrolysis of the cellulose matrix. The other one reflects oxidation processes. The behavior of tobacco material at fast heating rates up to 500 K/s (as in a burning cigarette) has been subject to various theories. Most of them result in a superposition of the degradation steps, being characterized by global kinetic parameters. One calculation predicts that pyrolysis processes should be the predominant effect at high heating rates. In a standard thermogravimetry analysis either a high heating rate can be achieved at a low sample weight, or a low heating rate must be used at a high sample weight. Therefore, there is a lack of experimental data. In our study thermal gravimetric analysis was carried out at heating rates between 100 and 300 K/s in air. A newly developed high power, radiation heated pyrolysis system was used. The light of a 1 kW high pressure arc lamp, chopped by pneumatic driven mechanics, was focused by an elliptical mirror on to the sample. Temperature measurement was done by an infrared technique. The sample mass was recorded by using a fast electric scale. Data logging and time control of the experiment was done by a personal computer. The kinetic parameters for burley and oriental tobacco species were derived from the thermogravimetry curves. These parameters are in good agreement with those values extracted from the first step of a low heating rate experiment, thus supporting the published hypothesis of a predominant pyrolysis effect (T. Kashawagi and H. Nambo, Combust. Flame, 88 (1992) 345–368.