Abstract

Starting in 1985, Mobil conducted a series of pilots and commercial-size steamfoam operations in South Belridge, California, which demonstrated significantly higher recoveries of heavy oil in combination with attractive process economics. To evaluate these successes and guide future steamfoam field applications, a steamfoam model was incorporated into an in-house developed thermal simulator and was used in simulations to (1) match field performance, (2) confirm the extent and nature (i.e., accelerated vs. incremental) of additional oil recovered by steamfoam, and (3) determine optimum time for stem foam injection termination.

The model is shown to be robust, with affordable complexity in the description of the relevant physics of the steamfoam process, and faithful in reproducing field performance.

Introduction

In 1985 and 1987, Mobil conducted two steamfoam pilots in South Belridge, California, the first in patterns 506 and 507, the second in patterns 517 and 518 of the NWB project area. With the second pilot, 182,000 bbl of additional (i.e., accelerated + incremental) oil, under favorable economics, were produced. In November 1989, a commercial-size steamfoam operation in a four, 5-acre, inverted 9-spot, pattern area, which was previously optimized with regard to steam flooding, was initiated. The steamfoam operation produced a significant amount of additional oil, with reduced surfactant requirements and more favorable economics. These successes, which needed to be evaluated in terms of the amounts of accelerated and incremental oil and the optimum time for foam injection termination, necessitated the development of a steamfoam model to be used in guiding future field applications.

A model for steamfoam is required to reproduce the mechanics of increased oil recovery which is a result of sweep efficiency improvements. Steamfoam improves sweep efficiency by (1) diverting the steam to alleviate gravity override and channeling into high-permeability zones, (2) pressurizing the steam-swept regions, and (3) reducing the mobility of steam. These mechanics are controlled by reservoir thickness, interwell thermal communication, permeability contrast of neighboring regions and layer vertical communication. The increased oil recoveries attained by steamfoam depend on its ability to propagate into the reservoir.

Steamfoam, a dispersion of stem made discontinuous by free liquid thin films or lamellae and stabilized by surfactant, is continuously reshaped as it travels away from the injector into the reservoir. Gas mobility changes with changing foam texture. An appropriate model needs to provide, in addition to surfactant transport, for in-situ foam generation and destruction and modifications to gas mobility dependent on foam type. Two types of foam are mainly distinguished (1) continuous-gas foam, with part of the gas trapped in a portion of channels blocked by foam and another part flowing as a continuous phase, and (2) discontinuous-gas foam in the form of discrete, moving bubbles. A somewhat parallel distinction into weak and strong foams, is based on foam stability. In general, weak foams propagate by "break-and-reform" of the pore-throat occupying lamellae and strong foams propagate as trains of discrete bubbles which experience increased resistance to flow because of the presence and shape (i.e., constricted) pore walls and the constant altering of viscous and capillary forces at the liquid/gas interface of the bubble. There are limiting pressure gradients for generation and flow of each type of foam, not to be confused with each other. The first is related to the particular type of foam generation, i.e., snap-off, leave-behind and lamella division, the second, which acts as a yield stress, is related to foam mobilization. Capillary pressure, i.e, the difference between pressure in the gaseous and aqueous phases, is also important to foam generation and coalescence. A "limiting capillary pressure" exists which depends on flow rate, permeability and surfactant formulation, at which foam collapses abruptly.

P. 283

This content is only available via PDF.
You can access this article if you purchase or spend a download.