No abstract

The viscoelastic behaviors of extra-heavy crude oils were investigated using a dynamic rheometer. The spectromechanical analyses of the fluids showed the presence of a relaxation process on the storage modulus curve. Both position and amplitude of this phenomenon were strongly dependent upon the value of the controlled stress. In particular, an increase of the applied stress induced a reduction of its average relaxation time. Similar effects could also be obtained with a temperature rise. The mechanical relaxation phenomenon was interpreted as being characteristic of the spatial rearrangement of interacting asphaltene aggregates. The global structure could be dislocated provided that the stress exceeded a critical value as predicted by theoretical approaches based on the potential gap concept. The original conformation of the asphaltene aggregates could be recovered under precise conditions of stress and time.

The study of multiphase flow pore level physic has scientific appeal as well as many applications, mainly in oil reservoir engineering. In this work, we show that micro-tomography is an effective tool to extract the structure of many solid systems in a non-destructive and classical manner. Recently, X-ray computed tomography tools have been extended to the ability to contrast fluids in the pore space of core samples. As time required for collecting a CT image is much longer than almost flow time scales, CT imaging must occur at static flow conditions to accurate geometrical information on fluid- fluid and solid - fluid interaction. Considering new decisive developments of the visualization cell we claim to have developed a new petrophysical tool which might permit to access experimentally to the visualization in quasi-static flow of the capillary phenomena and multiphase flows analysis in 3D form or in dynamic flow in 2D form. The in-situ measurements are realized in reservoir conditions of extra-heavy oils. The aim of the measurements is first to contribute to a predictive bubble population balance which will then be integrated in a network flow model computation. This step is required to develop physical models to obtain a more efficient reservoir simulation. More precisely we focus our attention on a quasi static model of bubble nucleation which appears decisive and we demonstrate how to simulate the whole bubble nucleation and the corresponding experimental results to corroborate the experimental observations. We focus here on the influence of the gas availability in the porous media (diffusivity, GOR or porous media morphology). Introduction More than a match for energy resources, the productivity of Extra-Heavy oil reservoir is one of the main challenges for oil companies. Numerous Enhanced Oil Recovery processes are available to product this very viscous and dense oil but primary production techniques are always used. The low environmental and economical cost of the primary production in comparison to steam injection for example is very decisive to select this kind of process. In primary production we focus our attention on a process called solution gas drive [2]. An Extra-Heavy Oil solution gas drive reservoir is one in which most of the production is due to the expansion of the originaly dissolved gas phase [3, 6]. Theoretically oil is undersaturated when reservoir pressure is above the bubble point pressure. Below the bubble point oil is saturated, the gas bubbles form or nucleate at the bubble point pressure and expand with continued decreasing pressure [5]. However the bubble point measured during the experiments, whatever in porous media or in PVT cells, and the bubble density which is the most critical parameter to obtain high OOIP, are mainly dependant on the decline rates. Thus, for higher pressure decline rates, greater numbers of bubbles are formed [4]. The number of bubbles formed was also observed to depend on the gas diffusion rate through the oil which tends to give rise to demonstrate bubble competitiveness and / or Ostwald ripening. Various works on a predictive modeling of bubble density report a need to fit the various parameters [2, 4, 6]. Our objective is to answer to this question that's why we first have been interested in developing a Harvey's nuclei (pre-existent bubble nuclei) model which permits to qualitatively predict or confirm the whole bubble expansion during the depletion [5]. In this way we have been interested in a pore scale approach to confirm this first step. Pore-scale physics governs fundamental behavior of the upscaled multiphase flows used in reservoir engineering. These systems are very complex and integrate various scientific fields as mechanistic modeling, physical - chemistry, thermodynamics, geology …

We examine the gas bubble nucleation phenomenon encountered in extra heavy oil during cold production. The nucleation model described in this work is based on so-called non-classical nucleation. Using this method, we show mesoporous cavities could be at the origin of the nanobubble trapping mechanism. The results obtained show this physical approach tends to demonstrate the pre-existence of gas bubbles in these crevices (surface roughness). The physics of capillarity used here is based on traditional Laplace's law and an original disjoining pressure expression. We test for several wettabilities in our mathematical model. The first configuration envisaged is for oil-wet rocks, although the cavity is assumed to be gas-wet. Water wettability is considered a second time, taking into account a precursor water film between the rock and the entrapped bubbles. The mix of these two configurations could represent nucleation in a global mixed-wet porous media. However, we show in the first part of this article that water presence does not affect the initial bubble radius. Nevertheless, a bubble growth model developed in the second configuration shows that bubble confinement could play an important role on gas bubble nucleation and the early first steps of its development. Introduction Macroscopic nucleation mechanisms have been the subject of many studies over the last decades because of both its fascinating underlying physics and its technological importance in many domains. The physics of gas bubble generation and growth in supersaturated solutions is a confusing subject in the literature. Jones et al.(1) proposed a classification system for the kinds of nucleation that occur experimentally. They define mechanistically four types of nucleation and place the specific forms of nucleation into a better defined context. Several authors concluded that the heterogeneous nucleation form is the most plausible mechanism in porous media. Nevertheless, homogeneous or heterogeneous nucleation on molecularly smooth surfaces requires a very high level of supersaturation. During a depletion experiment with heavy oil in porous media, the supersaturations reached are low, so another type of nucleation has to be envisaged. The opposite of the classical bubble nucleation theory which requires very high levels of supersaturation, is one where the activation of the pre-existing gas cavities needs very low levels. Therefore, to study the nucleation phenomenon in heavy oil we assume the pre-existence of trapped gas bubbles in contact with the supersaturated liquid. Lubetkin(2) focused on this and claimed that theory predicts a need for a much higher supersaturation to cause bubble nucleation than is found experimentally. Moreover, the theory predicts that the nature of the liquid influences the conditions required for nucleation and experimental works show a strong influence with the chemical nature of the gas. Considering this, we try to take into account the influence of the physical and chemical properties of the liquid and gas in equilibrium in the crevice and develop a detailed net analysis of the various existing forces. Most previous work relates to spontaneous bubble formations which occur after a rapid decompression of a liquid solution initially saturated with dissolved gas molecules.