EXPERIMENTAL SIMULATION IN A CONFINED SYSTEM AND KINETIC MODELING OF KEROGEN AND OIL CRACKING

The purpose of this study is to experimentally simulate both kerogen and oil cracking in a closed pyrolysis system and then, to model the kinetic scheme. Kerogens were isolated from their mineral matrix in order to obtain a complete recovery of the insoluble residue. Experiments were conducted under anhydrous conditions in order to calculate atomic balances for carbon, hydrogen, oxygen and sulfur. The pyrolysis products were fractionated by molecular weight and relative thermal stability. The kinetic scheme comprises four kinds of reactions: depolymerization reactions for kerogen and heavy products such as resins and asphaltenes. This type of reaction involves the most labile compounds, which are likely to be generated by C-O or/and C-S bond cracking. It produces heavy soluble compounds but very small amounts of gases and pure liquid hydrocarbons. C-C bond cracking of C6+ saturated chains. These reactions occur at higher apparent activation energies than the previous ones. They generate shorter aliphatic chains but again, very low amounts of methane and ethane. demethylation reactions of aromatic structures such as C-9-C-13 compounds, polycondensed C-14+ nuclei and the insoluble residue. The generated products are mainly gaseous compounds and coke. -C-C bond cracking of C3-C5 aliphatic chains which produces mainly methane and ethane, ethane being degraded into methane. Our results show that a unique kinetic scheme can be used for secondary cracking reactions either when oil is pyrolysed alone or when bitumen is first generated during kerogen pyrolysis. The kinetic scheme involves 11 chemical classes fractionated by molecular weight and thermal stability; among them 3 are stable (methane, a mixture comprising benzene/toluene/xylenes/naphthalene and coke), 8 are unstable (ethane, C3-C5, C-9-C-13 aromatics, C6-C-13 saturates, C-14+ unstable aromatics, C-14 + condensed aromatics and precoke). Although the two kerogens selected in this study are degraded with the same apparent activation energy and preexponential factor, stoichiometric coefficients must be specifically determined for each equation of the kinetic scheme in which kerogen cracking is included.