Experimental and Modeling Studies on Foam in Porous Media: A Review

Abstract

Abstract Foams have been used for several decades to decrease the mobility of drive gas or steam, thereby enhancing the reservoir sweep efficiency. More recently foams found application in near-wellbore production enhancement operations such as acid diversion for matrix stimulation, gas shut-off, water shut-off and hydraulic fracturing. The optimization of these operations requires a thorough knowledge of the physical aspects involved in foam flow through porous media. In spite of the copious amount of experimental and theoretical investigations found in the literature, this knowledge is lacking. The present paper is aimed at reviewing experimental and modeling studies on foams in porous media, to stress the latest achievements and highlight the areas that are the least understood. The focus of the analysis is twofold. First, we review the most representative core flow experiments that were undertaken to investigate the behavior of relative permeabilities and mobilities with emphasis on the key microscopic mechanisms proposed. Second, the different classes of models of foam flow through porous media are presented, outlining their advantages and weaknesses. A comparative analysis of the models is presented. 1. Introduction Foam, a coarse or fine dispersion of a gas in a liquid, has long been an important subject for scientific study because of its applicability or impairing effects in a variety of petroleum industrial processes. As reviewed by Schramm [1], occurrence of foams in petroleum processes such as oil-water separation, oil flotation or distillation and fractionation cause problems by reducing process efficiency and operating controllability. In contrast to these impairing influences, foams also possess many desirable properties, which make them highly applicable in gas-mobility control in improved oil recovery, acid diversion in well stimulation, water shut-off and gas-blocking in water coning prevention and gas storage. The rationale for foam application in such processes is derived from basic properties of a stable-foam dispersion propagating in porous media: significant increase of gas viscosity in foam during gas-drive oil recovery processes; and the very low or even negligible gas and liquid permeabilities offered by a strong foam. Experience has shown that actual oil recovery by steam or CO2 processes are much lower than expected due primarily to poor "sweep efficiency": a substantial portion of the oil reservoir cannot be swept by a poorly contacted gas [2–12]. This undesired situation is most likely to be ascribed to three main causes: gravity overdrive; channeling; and viscous fingering. Foaming the displacing phases has dramatically improved the sweep efficiency and oil recovery. Gas flow in form of bubbles separated by thin films called lamellae exhibits more resistance produced not only by viscous shear stresses in thin films between the pore walls and the gas-liquid interface, but also by the forces required to push lamellae through constricted pore throats. A maximum pressure gradient at which all lamellae can be mobilized has not been found in all cases in foam core floods. This implies that there is always a portion of bubbles trapped inside porous media during foam flow. This portion is referred to as "trapped gas" or "stationary lamellae", and is thought of as a temporary blocking agent for gas flow. The reduction of both gas and liquid mobility are the result of the same physical phenomena but derive from different kinetic mechanisms. Increased viscosity and fraction of trapped gas decrease gas mobility while low liquid saturation due to foam induces decreased liquid mobility without any change in its viscosity. Unfortunately it is difficult to control foam behavior to achieve the desired purpose. Full understanding of the mechanisms governing foam behavior remains a challenge, pursued in many experimental and modeling studies. Some start by visualizing at the microscopic level the mechanisms of foam generation, transport and destruction. However, such micro-models, are not adequate for investigating real field conditions, which need to be simulated by core-flood experiments. The major aim of these macroscopic experimental studies has been to delineate causal relationships between global governing factors and microscopic foam mechanisms.