SOME PROBLEMS IN PETROLEUM GEOCHEMISTRY

Limited geological events are occasionally observable, but petroleum geochemistry is inevitably a wholly inferential science. It can only be simulated in the laboratory, with occasional help from industrial‐scale work, by using the Arrhenius equation to convert observations in human time to geotime by means of temperature change. There are difficulties here. First, the Arrhenius equation is purely empirical. The analogy with the theoretically rigorous van't Hoff isochore cannot give the Arrhenius equation any more than empirical status. A departure from linearity unobservable in human time could well make the extrapolation to geotime subject to significant errors. Secondly, the inability to define petroleum generation in terms of specific molecules prevents the assignment of specific activation energies to specific reactions. Energies of activation can vary significantly between even similar reactions. Thirdly, catalysis can accelerate a reaction by ten or twenty powers of ten, and even variations in natural clays can alter reaction rates almost a thousandfold. Each of these three factors could easily introduce errors of two powers of ten in translating laboratory simulation to geotime. From equilibrium considerations, petroleum generation and degeneration belong to a continuous spectrum of reactions in which very large molecules are broken down into smaller ones, and ultimately methane. Pressure inhibits the equilibrium of such reactions to such an extent, especially as the molecules become smaller, that there can be no validity to simulation experiments which are not carried out at geopressures. Altogether. it seems unlikely that the age of oil can be limited by thermodynamic considerations. Finally, an inorganic origin for some petroleum ought not to be completely ignored.