Experimental Three-Phase Flow in Porous Media: Development of Saturated Structures Dominated by Viscous Flow, Gravity, and Capillarity
- Alexandru Barbu (Pennsylvania State U.) | P.J. Hicks Jr. (Pennsylvania State U.) | A.S. Grader (Pennsylvania State U.)
- Document ID
- Society of Petroleum Engineers
- SPE Journal
- Publication Date
- December 1999
- Document Type
- Journal Paper
- 368 - 379
- 1999. Society of Petroleum Engineers
- 5.3 Reservoir Fluid Dynamics, 1.6.9 Coring, Fishing, 5.2 Reservoir Fluid Dynamics, 5.4.1 Waterflooding, 4.6 Natural Gas, 6.5.2 Water use, produced water discharge and disposal, 1.10 Drilling Equipment, 4.1.2 Separation and Treating, 4.1.5 Processing Equipment, 5.3.1 Flow in Porous Media, 5.1.1 Exploration, Development, Structural Geology, 5.5.2 Core Analysis, 4.3.4 Scale, 5.3.2 Multiphase Flow
- 3 in the last 30 days
- 273 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 10.00|
|SPE Non-Member Price:||USD 30.00|
The motivation for this work is the need to understand the formation of oil banks during waterflooding operations as well counter-current gravity driven oil recovery mechanisms. The success of these operations depends on a combination of viscous, gravitational, and capillary forces. This paper focuses on the development of three-phase saturation structures driven by viscous flow, gravity, and capillarity. Three liquids: water, benzyl alcohol, and decane are flowed through glass bead packs. Dynamic water floods of cores saturated with mixtures of benzyl alcohol and decane are described. Saturation distributions in the core are determined using x-ray computer tomography (CT). The injection of water generates a benzyl alcohol bank that displaces most of the decane ahead of it. Benzyl alcohol and decane, driven by gravity, exchange places in the core. This gravity driven counter-current flow is rapid in the presence of low water saturation ahead of the benzyl alcohol bank and within the benzyl alcohol bank. The gravity exchange is slow in the presence of mobile water behind the benzyl alcohol bank. The water saturation structure is stable and is not significantly affected by the counter-current flow of benzyl alcohol and decane. The accumulation of decane at the top of the core triggers a backward migration of a decane tongue. The decane tongue disappears when water injection is resumed and the benzyl alcohol bank re-forms where the decane tongue existed. When mobile decane is not present, significant capillary driven longitudinal fluid transport is observed while the high gradient saturation structure in the transverse direction remains stable. Even high saturation gradients in the longitudinal direction are quite stable. Fluid flow patterns during the shut in periods and displacement mechanisms are proposed.
Three-phase flow in porous media occurs in oil and gas production, as well as in other fields such as ground water hydrology and in situ mining. Understanding of three-phase fluid flow mechanisms is essential in the process of forecasting and optimizing human interaction with the upper crust of the earth. The interactions between viscous, gravity, and interfacial tension forces create complicated fluid flow phenomena that need to be modeled for predictive purposes. This paper furthers our ability to predict the outcome of injection and production schemes so that more effective procedures can be implemented.
There are several papers in the literature concerning modeling three-phase flow in porous media, with a focus on relative permeabilities.1-5 Some papers in the three-phase field concern both experimental work and theoretical work.6-8 The mathematical description of three-phase flow in porous media is difficult and many more experimental sets are needed to develop an acceptable theory for practical usage. There is no attempt here to summarize the current status of experimental and theoretical three-phase flow in the literature. This paper focuses on presenting experimental evidence to three-phase flow saturation structures and some mechanistic puzzles.
This paper describes the flow of three phases: water, benzyl alcohol, and decane through glass bead packs. The mechanisms controlling the motion of three phases in a porous material are not fully understood. Relative permeabilities and capillary pressure functions are traditionally used to describe fluid flow in a continuous manner. There are very few papers concerning three-phase flow experimental work in the literature. The main objective of this paper is to further our understanding of three-phase flow in porous media. The paper focuses on three-phase flow observations and suggests possible explanations to these observations.
The experiments examine the flow of three immiscible liquids in glass bead packs. The liquids are water, benzyl alcohol, and decane. Water and decane are tagged with iodine (sodium iodide and iodododecane, respectively) to yield the same high CT registration, while benzyl alcohol has a low CT registration. The matching of the CT response of water and decane allows the determination of benzyl alcohol saturation with a single high energy scan. The water and the decane saturations cannot be separated using only the high energy CT values. The densities of pre-equilibrated tagged water, benzyl alcohol, and decane are 1.03, 1.03, and 0.77 g/cc at room conditions, respectively. The density of benzyl alcohol is slightly less than the density of water, allowing gravitational separation in collection vessels. The viscosities of tagged water, benzyl alcohol, and decane at 20\(deC) are 1.19, 5.44, and 1.14 cp, respectively.9 The fluid-fluid interfacial tensions for the tagged fluids are: 3.04 dyne/cm (water-benzyl alcohol), 1.83 dyne/cm (benzyl alcohol-decane), and 6.15 dyne/cm (decane-water).10 Three types of scanning sequences are used, designated as coarse (3-in. spacing), intermediate (1-in. spacing), and detailed (1/3-in. spacing). The glass bead core is 36 in. long with a diameter of 2 in. In situ saturations are determined using CT. X-ray CT technology is described by many papers in the literature.11-14 Complete details of the CT application in this work are available.15
Two experimental sequences are summarized in Fig. 1. Each sequence has two main stages. In the first stage, a steady-state saturation of benzyl alcohol and decane at low water saturation is established by evacuating the core (1), saturating the core with water (2), injecting benzyl alcohol (3) and finally, injecting a steady mixture of benzyl alcohol and decane (4). All steps in the first stage are performed while the core is in a vertical position.
In the second stage, a dynamic water flood is performed. In Experiment A, 0.3 pore volumes of water are injected while the core is in a vertical position (5). The core is then rotated to a horizontal position and placed in the CT unit for scanning (6). Following several sequences of scans, the core is rotated by 180° around its long axis (7) and several sequences of scans are performed (8). Water injection is resumed and another 0.1 pore volumes of water are injected (9) and the core is scanned again (10). Finally, the water flood is completed (11) and the core is scanned for the last time (12). The fluids produced during the water injection periods are collected and recovery curves are constructed.
|File Size||1 MB||Number of Pages||12|