Water Imbibition and Characterization of North Sea Chalk Reservoir Surfaces

Abstract

Summary Water imbibition of a reservoir core was dramatically affected by the removal of small amounts of tightly bound material. Both water-wet and oil-wet chalk cores contained sufficient extractable material to form multimolecular coatings over their entire surfaces. Analysis of the material removed from these reservoir cores provided some clues regarding the fluid dynamics. Introduction Over the past half-century, research into the interaction between mobile fluids and the immobile reservoir surface has produced a variety of interpretations. Some early workers considered the reservoir surface to be the inorganic rock that was completely wet by interstitial water and never contacted by either the gas or oil phases.1,2 Others speculated that oil-wet surfaces existed in some reservoirs and could be generated in the laboratory by the addition of certain oils or other chemicals to clean sands.3–5 More recent studies have shown that some cores from oil-producing reservoirs are predominantly oil-wet or at least not completely water-wet.6–11 These findings corroborate Nutting's12,13 early evidence of insoluble black hydrocarbon coatings on several oil-bearing sandstones and limestones and Rudakov's14 conclusions that highly condensed organics are not extracted from sediments during the migration of petroleum. The results obtained by these last two authors strongly suggest that the inorganic substrate is covered with organic material, at least for the petroleum reservoirs they studied. Several researchers have attempted to analyze material removal from reservoir surfaces.15,16 Some authors17,18 have proposed that reservoir wettability is largely, if not entirely, determined by the components in the oil phase; however, others19–21 claim that the surface is equally important in determining wettability. The magnitude of the fluid-surface interaction, as measured by wettability, affects the movement and ultimate production of oil. For example, relative permeabilities were found to vary with the surface wettability,22,23 and the relative oil permeability decreased as the strength of oil interaction with the immobile surface, oil wettability, increased. It appears to be generally assumed by reservoir engineers that a water-wet core has a clean, inorganic surface, while an oil-wet one is covered with some component(s) of the crude oil; however, one must emphasize that this is an assumption and not a proven fact. Knowledge of the chemical nature of the immobile reservoir surface that contacts the mobile fluids is important for both economic and technical reasons. This is especially true of waterflooding in fractured chalk reservoirs where the production mechanism is the expulsion of oil by spontaneous imbibition of water. If the major functional groups on the reservoir surface were known, one could predict the approximate amount of water that would imbibe and displace oil. In addition to predicting performance of a waterflood, this knowledge could lead, in favorable cases, to the design of a flood fluid that would maximize oil displacement. Experimental Characterization of the reservoir surfaces was made both directly on the surfaces [thermal gravimetric analysis (TGA), X-ray photoelectron spectroscopy (XPS), and argon ion etching] and after the material was removed and concentrated by sequential extraction (elemental analysis, infrared, fluorescence, and weight). the sequential extraction process is summarized in the block diagram of Fig. 1. The organic components in the original oil are assumed to have reached a state of equilibrium with regard to adsorption during geologic times in the reservoir. With the proper choice of solvents, the most weakly held materials are removed first, without disturbing the more strongly bound components. The extracted materials are thus fractionated according to the strength of their interactions with the surface. The weakly held material was similar to the produced crude, while the most strongly bound material would not have been produced during normal production. These latter materials thus form the immobile surface of the reservoir. The sequential extraction of core material was generally performed in a Soxhlet extractor with approximately 50 g of core material and 200 mL of solvent used. Each solvent was cycled until no evidence (generally color) of additional extraction was observed for 24 hours. Extraction, which lasted 3 to 4 weeks, was run under an argon gas blanket to reduce the chance of reaction. The extracted materials were isolated from the solvent by vacuum evaporation at ambient temperature. The solvents, n-heptane, toluene, and tetrahydrofuran (THF), were selected for their increasing solvent strength and relatively low boiling points. All the extracted materials were soluble in any of the solvents once removed from the core. This was necessary to eliminate the possibility that the solvents, particularly the least polar, would cause a fraction of the oil to precipitate. As much of the original core surface as possible was preserved to keep selective adsorption on freshly exposed matrix surfaces to a minimum. In sample preparation, we exposed 1% new surface. This contrasts to others,24 who crushed the rock to grains of <0.15 mm [<0.006 in.] to generate new surfaces and thus form a chromatographic material from the original rock. TGA was performed with a Dupon 990™ thermanalytical system and a 951 thermogravimetric analyzer. The sample to be analyzed was placed on the pan of a microbalance and either heated at a constant rate or held at a single temperature. The weight changes are automatically determined by an electronic balancing system. The baseline was a single crystal of pure calcite. XPS, which provides qualitative and semiquantitative analyses of surface films, was made with a PHI Model 548™ spectrometer. This technique consists of irradiating the surface with narrowband X-rays and measuring the energy of the ejected photoelectrons. The specific energy of photoelectrons from individual elements allows a qualitative analysis of the surface elements. With proper ratioing of elemental intensities, it is possible to obtain a semiquantitative analysis. Further information about surface film thickness and composition as a function of depth can be obtained by bombarding the surface with high-energy argon ions between XPS analyses. This process mechanically removes surface elements, and thus exposes underlying features. Reservoir core material was obtained from standard coring of North Sea chalk reservoirs. Core samples were preserved at the drilling site by being wrapped in Saran Wrap™ and aluminum foil, then dipped in molten plastic. Cores, after the coating was removed and plugs were cut, were stored and extracted under argon to minimize the opportunity for oxidation. Similar experiments on unpreserved core indicated that core preservation did not significantly affect imbibition or the character of the more tightly bound material. Lack of preservation resulted in loss of light ends and slightly more oxidation in the heptane extract.