Étude de la mouillabilité des roches réservoir à l'échelle du pore par cryomicroscopie électronique à balayage Wettability of Reservoir Rock At the Pore Scale: Contribution of Cryo-Scanning Electron Microscopy

Abstract

Le but de cette étude est de caractériser, à l'échelle du pore, la mouillabilité des roches réservoir, en relation avec leur géométrie et/ou leur minéralogie. Cette caractérisation se fait, après congélation des échantillons, par l'observation de la distribution des fluides au sein du milieu poreux (saumure et huile brute), en microscopie électronique à balayage. Les expériences ont d'abord été effectuées sur des roches modèles parfaitement mouillables à l'eau, verre fritté et grès naturels. Certains de ces minéraux ont été rendus hydrophobes par greffage de silane. L'étude de ces systèmes a mis en évidence une corrélation entre la mouillabilité et la distribution des fluides. Puis, une roche réservoir (grès argileux de la formation de Brent, de mer du Nord) connue comme étant de mouillabilité intermédiaire a été étudiée. Un travail précédent (étude de déplacements eau/huile par tomographie X) avait abouti à la conclusion que si ces roches présentaient des hétérogénéités de mouillabilité, l'échelle de ces hétérogénéités devait être inférieure au millimètre. Les études de cryomicroscopie ont montré le caractère hydrophobe de la kaolinite, tandis que les illites, le quartz et les feldspaths sont préférentiellement mouillables à l'eau. L'imbibition spontanée d'huile pourrait ainsi être attribuée à l'existence au sein de la roche d'un réseau de kaolinite, tandis que l'imbibition spontanée de saumure serait due à l'existence d'un second réseau plus ou moins imbriqué avec le premier et constitué des autres minéraux. Un autre cas de roche réservoir a été étudié, à savoir un carbonate du Moyen-Orient. Les mésopores intergranulaires y ont été observés comme étant mouillables à l'huile tandis que les micropores restaient mouillables à l'eau. Dans ce cas, la mouillabilité intermédiaire de ces échantillons s'explique par la géométrie plutôt que par la minéralogie qui est plutôt homogène. Wettability is generally considered to be one of the principal parameters influencing the distribution, saturation and flow of fluids in porous media. Reservoir rock wettability has long been approached by overall or indirect methods [1] (capillary pressure or relative permeability curves, contact angle, fluid displacement, etc. ). Few studies until now have led to a detailed description of porous media with intermediate wettability. - Is there any evidence of an intermediate behavior of fluids in contact with minerals distributed homogeneously throughout the medium, or is there a heterogeneous distribution of water- and oil-wettabilities within the porous medium?- What influence does the local heterogeneity of the minerals (size, geometry, surface chemistry, etc. ) have on fluid distribution [2 to 7]?The answer to these questions requires a microscopic-scale description of saturated porous media [8 to 11]. By using the imaging and analytical capabilities of a scanning electron microscope coupled with a cold stage unit, fluids can be visualized and identified by detection of their natural tracer element (sulfur for oil and chlorine for brine), and their relative distribution within the pore space can be analyzed in terms of wettability. Results presented here illustrate both the interest of the method and its applicability to actual reservoir rocks. Small cores of the chosen porous media were first saturated with brine, flooded to irreducible water saturation by centrifuging in oil, aged in oil for one month and finally flooded to residual oil saturation by centrifuging in brine. Samples were then frozen in nitrogen slush, freeze fractured and coated before being transferred to the cold stage of the microscope for observation. Experiments were first conducted on porous media with controlled wettability : model sintered glass media, natural clean sandstone (Fontainebleau) and clayey sandstones (Vosges, Velaines). All these porous media are naturally water wet. Some of them were treated with alkyltrichlorosilane so as to become oil wet. - The analysis of the relative distribution of fluids within the pore space enables conclusions to be drawn about the wettability of the pore walls. The residual non wetting fluid appears as globules trapped in the center of the pores, while the irreducible wetting fluid appears as films surrounding some grains. Films observed were rather thick (1 to 5 microns) and rare, but this does not exclude the systematic presence of a film less than 0. 1 micron thick, this being the limit resolution in the operating conditions. - The porous media made of spherical glass beads, eroded quartz grains or silica overgrowth with no or low content of clays, have a comparable behavior. - Quartz and feldspars are naturally water-wet; the wettability of the quartz grains is efficiently reversed by silanation. - Illite has a marked affinity for brine even after silanation. - The presence of small size minerals (weathered feldspars and clays) enhances oil entrapment by reducing the pore throats. Experiments were then conducted on actual field cores with intermediate wettability (i. e. spontaneously imbibing both brine and oil), a sandstone and a limestone. Both led to interesting conclusions concerning the origin of their behavior. The main results concerning a sandstone from the North Sea (Brent formation) composed of quartz and feldspar grains, and a high content of clay minerals (mainly kaolinite and some illite) are as follows :- residual oil is systematically associated with kaolinite,- illite and weathered feldspars are always observed associated with brine,- quartz and feldspar grains are preferentially water-wet. In some cases, detrital feldspar grains were observed partially covered by oil. Dissolution roughness then seems to play a role in oil entrapment. The hydrophobic character of this sandstone can then be attributed mainly to the presence of kaolinite and its affinity for oil. Spontaneous imbibition in both crude oil and brine can be interpreted as a consequence of two coexisting networks that are oil- or water-wet : kaolinite on one hand (due to high content, and homogeneous distribution), and essentially quartz, feldspar and illite on the other. A quantitative analysis of phase distribution could confirm this hypothesis. - The difference between the Wettability Index measured for oil and water zone samples could be due to differences in clay distribution (depending on the diagenetic history). For an actual field limestone, cryo-SEM observations of fluid distribution lead to an interpretation of sample behavior during displacement tests. They show that intergranular mesopores are preferentially oil-wet, whilst cement micropores remain water-wet. This points out a wettability heterogeneity at the pore scale, leading to an intermediate wettability on a macroscopic scale and thus demonstrates the importance of pore size and geometry. Wettability alteration could be related to geometric parameters during oil invasion. When oil invaded the initially waterwet pore space, its distribution was controlled by both pore size and prevailing capillary pressure : the largest pores were invaded by oil while the smaller ones remained oil free. Aging then caused adsorption of polar oil compounds on the exposed surface. Spontaneous imbibition of oil could therefore be due to a continuous pore network within the oil-wet intergranular mesopores, whilst spontaneous imbibition of brine could be related to brine circulation in water-wet micropores of the calcitic cement. Cryo-SEM has a resolution below the size of minerals constituting natural porous media. It makes it possible to study in situ the influence of various parameters (pore mineralogy, geometry, surface chemistry, etc. ) on wettability. Microscopic studies of oil-brine-rock systems (associated with other imaging techniques such as the X-ray computed tomography) contribute to a better understanding of intermediate wettability causes and can thus explain the macroscopic behavior of some reservoir rocks.