CHEMICAL AND MATHEMATICAL MODELING OF ASPHALTENE REACTION PATHWAYS

Precipitated asphaltene was subjected to pyrolysis and hydropyrolysis, both neat and in solvents, and catalytic hydroprocessing. A solvent extraction procedure defined gas, maltene, asphaltene, and coke product fractions. The apparent first order rate constant for asphaltene conversion at 400(DEGREES)C was relatively insensitive to the particular reaction scheme, and its value was always 0.013 (+OR-) 0.008 min('-1). The yield of gases likewise showed little variation and was always less than 10%. On the other hand, the maltene and coke yields were about 20% and 60%, respectively, from neat pyrolysis, and about 60% and less than 5%, respectively, from catalytic reactions.; The temporal variations of the product fractions allowed discernment of asphaltene reaction pathways. The primary reaction of asphaltene was to residual asphaltene, maltenes, and gases. The residual asphaltene reacted thermally to coke and catalytically to maltenes at the expense of coke. Secondary degradation of these primary products led to lighter compounds.; Pyrolysis of the asphaltene model compounds pentadecylbenzene, phenyldodecane, butylbenzene, 2-ethylnaphthalene, tridecylcyclohexane, and 2-ethyltetralin demonstrated that aliphatic C-C bond cleavage occurred predominantly near or at the ring. Aromatic and hydroaromatic rings were resistant to ring-opening. Saturated rings dehydrogenated to aromatics, and single-ring naphthenics underwent ring-opening in addition to dehydrogenation. The Arrhenius parameters log(,10)A (s('-1)), E* (kcal/mol) for disappearance of pentadecylbenzene, tridecylcyclohexane, and 2-ethyltetralin were 14.1, 55.4 , 14.9, 59.4 , and 12.7, 53.5 , respectively.; Pyrolyses of alkylaromatics in tetralin-d(,12) verified that only free-radical reaction steps, and not molecular mechanisms, were kinetically significant. Furthermore, organizing the elementary reaction steps as three parallel chain reactions, with chain transfer by radicals derived from the substrate by both (beta)-scission and hydrogen abstraction, admitted convenient derivation of analytical expressions for product selectivities and the apparent reaction rate.; The model compound kinetics results were combined with a stochastic description of asphaltene structure in a mathematical model of asphaltene pyrolysis. Individual molecular products were assigned to either the gas, maltene, asphaltene, or coke product fractions, and summation of the weights of each constituted the model's predictions. The temporal variation of the product fractions from simulated asphaltene pyrolysis compared favorably with experimental results.