Predicting petroleum gravity with basin modeling: New kinetic models

Abstract

Oil API gravity predictions using published basin modeling source rock (SR) reaction kinetics have displayed poor matches between modeled output and field observations because these kinetic models do not predict increasing API gravities with increasing maturity. Ideally, an SR kinetic model should use at least two liquid components of different densities, which are generated and expelled from the SR such that the API gravities are a consequence of relative mixing. Very few available kinetic models predict APIs with reasonable trends, but those are either not adjustable to calibrate to field observations or do not consider sorption, which is a necessary process when evaluating unconventional resources. Five new kinetics data sets are presented in this paper, each representing a standard SR type, which provide geologically reasonable API gravity trends and ranges. Each kinetic model uses two liquid pseudocomponents and two vapor pseudocomponents. The relative ratios between the pseudocomponents at full kerogen transformation are average ratios available from public and proprietary kinetic data sets. The primary generation follows published activation energies, including minor shifts, which allow peak generation to occur at lower activation energies for the heavier liquid pseudocomponent and at higher energies for the lighter one. This systematic shift of activation energies thus results in a constant change in API gravity as primary generation progresses. Additional in-SR sorption and secondary cracking schemes support the primary generated API gravity trends. The default ranges of API gravity for the new five kinetic models represent observed averages but can be adjusted easily.