Microscopic Observations of Solution Gas-Drive Behavior

Microscopic mechanisms are critical in the behavior of reservoir systems and our knowledge of them uncovers pertinent factors made apparent in no other way. Mechanistic problems associated with evolution of solution gas have been considered by earlier investigators. Sage, Kennedy, Berry, and their co-workers have studied the number of bubbles formed in the evolution process. Van Meurs, has visually examined the patterns formed by the evolved gas in displacements in ideal packs. With enlargogram techniques Kimbler and Caudle have examined location of the points of genesis of free gas in unconsolidated sand packs. In the present work 116 laboratory experiments were performed in a visual study of the solution gas-drive mechanism. Thin glass-bead packings and thin sections of natural sandstone and limestone were used as porous matrices in the experiments. As gas evolved the system was examined visually at different magnifications. Certain differences in fluid behavior were observed for the different types of matrices. However, the findings of most interest were generally observed in all tests. These findings help clarify the roles of the physical parameters governing solution gas-drive behavior. They also provide a basis for further experimentation to help define this behavior in reservoir formations. The experimental approach and techniques employed were developed by American Petroleum institute Research Project 47-B. Details of these are published in the literature. Observation flow cells were saturated with butane-saturated mineral oil at atmospheric pressure and drawn down to sub-atmospheric pressures to effect gas evolution. Results of Gas Evolution Experiments Gas Nucleation In all the gas evolution experiments pressure had to be reduced to several psi less than atmospheric before the first point of gas nucleation appeared. This indicates that some degree of supersaturation must be reached before gas evolution will start. However, supersaturation behavior is variable and not precisely reproducible, even in identical systems.