Rheology of Gas/Water Foam in the Quality Range Relevant to Steam Foam

Abstract

Summary Experimental results on the rheology of gas/water foam essentially show that if the gas flow rate is increased at an imposed water rate, the pressure gradient increases, reaches a maximum (the break point), and then decreases. A model for foam flow was formulated that explains this behavior semiquantitatively. It consists of a highly simplified model for foam in a single capillary coupled with a description for foam flow in a bundle of identical parallel capillaries. Below the break point, capillaries filled with foam occur with water-filled capillaries; above the break point, the water-filled capillaries are replaced by gas-filled capillaries. A consequence of the model is that if one plots the results in a special way, a universal curve is obtained, dependent on only the porous medium and the fluid properties. The experimental data more or less conform to this curve, which can be used (in the same way as relative permeabilities) to predict pressure gradients for gas/water rates other than those experimentally measured. Consequences of the foam model for numerical simulation of foam flow are discussed. Introduction Because the mobility of gas flowing through a porous medium is reduced considerably when the gas is flowing as a foam, foams have been, proposed for many oil recovery processes involving gas injection. For practical applications, one would lie to know the basic relationships between gas/liquid fluxes, pressure gradient, and saturations. Also, the dependence of these relationships on several factors-such as surfactant type and concentration, the properties of the porous medium, and the presence of oil-warrants investigation but was ignored in this study. Some confusion exists in the literature about the basic relationships between fluxes and pressure gradient. Most investigators find a decrease in foam mobility when the foam quality is increased for "cold" gas/water foams, whereas others find an increase. For steam foam, only an increase in foam mobility with increasing quality has been reported. (Foam quality is the volume gas rate in the foam as a fraction of the total volumetric rate, whereas steam quality is the mass vapor rate in the steam as a fraction of the total mass rate.) Although in steam foam other processes than those in gas/water foam (condensation and evaporation) play a role, we thought that these additional processes should not effect such significant differences. Consequently, we studied the rheology of nitrogen/ water foam in porous media, concentrating on a quality range between gas/water foam (qualities normally less than 95 %) and steam foam (qualities normally >99 %). We show that steam and gas/water foams of comparable foam quality exhibit similar behavior. Experimental Setup and Results The experimental setup is basically a standard coreflow experiment in which gas and liquid flow rates are imposed and the resulting steady-state pressure gradients are measured with a number of pressure taps. Foam is generated outside the porous medium by means of a stainless-steel porous plug in front of the inlet. The foam's texture (the bubble size) has been shown to be an important variable in determining foam mobility. Varying the texture of our injected foam by varying the properties of the foam generator did not influence the results within experimental error. We conclude, therefore, that the foam texture is changed by the porous medium to a value independent of the injected texture in porous medium to a value independent of the injected texture in such a short distance that the effects are not noticeable in our experiments. The pressure gradients were measured over several sections of the porous medium, and the results were found to be comparable. The pressure gradient was measured over a 10-mm [0.4-in.] section approximately halfway along the porous medium. Because of the high pressure gradients measured and the compressibility of the gas phase, pressure differences should be small compared with the phase, pressure differences should be small compared with the pressure level. pressure level. We performed two sets of experiments (Sets 1 and 2) in slightly different equipment. The data are summarized in Table 1. Set 1 experiments were performed at ambient temperature on a Clemtex sandpack of 4.2-/im permeability.