Isothermal and nonisothermal methods have been used to investigate the kinetics of oil generation during decomposition of 91.7 ml/kg (22 US. gal/short ton) Colorado oil shale. The result from the nonisothermal method gives an apparent activation energy of 219.4 kJ/mol and a frequency factor of 2.81 x 1 013 s-t. Furthermore, the process is found to be first-order to within experimental error. These results compare favourably with isothermal data reported here and in the literature. The results show the reliability and convenience of nonisothermal kinetic experiments in studying oil-shale decomposition reactions. The principal advantages are short-term experiments and the lack of initial heat-up periods. Moreover, nonisothermal experiments more accurately simulate actual conditions of above-ground and in situ oil-shale retorting. These kinetics are 'effective' values and can only properly be used to describe the macroscopic oil-production process rather than the complex microchemistry. Several investigations have been carried out on the kinetics of the decomposition of kerogen in oil shalelp4. Perhaps the best known of these have been articles by Hubbard and Robinson' and by Allred'. Both investigators attempted to treat the kinetics using a simple two-step mechanism for decomposition of kerogen to bitumen and subsequently bitumen to oil: Kerogen k bitumen k' oil, where k > k' at T < 3OO'C. Recently Braun and Rothman have more completely analysed the Hubbard and Robinson data, taking into account an initial 'thermal induction period', and they report activation energies and frequency factors for each step of the proposed two-step mechanism.
Good kinetic data are essential for accurate mathematical modelling of various oil-shale processes (e.g. in situ processing). Of particular importance are kinetics for the oilevolution step. Previous experiments on the rate of oil evolution have been carried out under isothermal conditions or under nonisothermal conditions by thermogravimetric analysis (TGA). Both methods have some experimental difficulties. Because of the initial heat-up period, it is often hard to carry out true isothermal experiments.
Nonisothermal TGA experiments, on the other hand, are complicated by the fact that total weight loss (i.e. of noncondensible gas, oil and water) is measured rather than just oil release.
The major goal of this study is to measure directly the rate of oil evolution during decomposition of 9 1.7 ml/kg (22 US gal/short ton) Colorado oil shale. For comparison, we took kinetic data by both isotherma! and nonisothermal techniques. The nonisothermal results show that the oilevolution process can be quite accurately represented as a