Displacing fluid mobility

The existing concept of mobility control is that the displacing fluid mobility should be equal to or less than the (minimum) total mobility of displaced multiphase fluids. This chapter first uses a simulation approach to demonstrate that the existing concept is invalid the simulation results suggest that the displacing fluid mobility should be related to the displaced oil phase mobility, rather than the total mobility of the displaced fluids. From a stability point of view, a new criterion regarding the mobility control requirement is derived when one fluid displaces two mobile oil and water fluids. The chapter presents numerical verification and analyzes some published experimental data to justify the proposed criterion. 

When discussing viscous fingering, generaUy we deal with the case of displacing one mobile fluid (e.g., oil) by another fluid (e.g., water). The concept is that the displacing fluid mobility in the upstream (A,) should be equal to or less than the displaced fluid mobility in the downstream (A,[Pg.80]

The conventional mobility ratio in multiphase flow is defined as the displacing fluid mobility divided by the total mobility of displaced water and oil phases. From the previous section, we can see that the unit mobility ratio based on the conventional definition is not a valid criterion to distinguish favorable and unfavorable mobility control conditions. We have found that a better criterion should be the unit mobility ratio, which is defined as the displacing fluid mobility divided by the oil mobility multiplied by the oil saturation (Eq. 4.9). In this section, we attempt to justify the proposed idea from the stability of displacement front. 

Figures 4.21 and 4.22 show the recovery factors versus M c and for the initial water saturation of 0.7. These figures show that the observations in the homogeneous model are still valid in the heterogeneous model. In other words, if we define the mobility ratio as the ratio of injection (displacing) fluid mobility to oil mobility multiplied by the normalized oil saturation, the unit mobility ratio is a much better criterion than the conventional one using the total mobility.

As discussed in Chapter 4, the mobihty control reqnirement is closely related to the ratio of displacing flnid mobility to displaced flnid mobility. Because changing displaced oil mobihty (relative permeability and/or viscosity) often is not feasible without the injection of heat, most often we inject chemicals to change displacing fluid mobility. Primarily, the injected chemicals are polymers whose obvious function is to increase the displacing polymer solution viscosity, although other mechanisms are involved, as discnssed in Chapter 6. 

As described in Chapter 4, the higher the displacing fluid viscosity, the higher the sweep efficiency (recovery factor). Chapter 4 proposed that the displacing fluid mobility should be equal to the displaced oil mobility corrected by the oil saturation. In Daqing, however, the polymer solution viscosity of three to five times the oil viscosity was used so that a high oil recovery factor can be obtained in heterogeneous reservoirs. 

The macroscopic displacement efficiency is a measure of how well the displacing fluid has come in contact with the oil-bearing parts of the reservoir. The microscopic displacement efficiency is a measure of how well the displacing fluid mobilizes the residual oil once the fluid has come in contact with the oil. 

Polymer flooding alms at reducing the amount of by-passed oil by increasing the viscosity of the displacing fluid, say water, and thereby improving the mobility ratio (M). 

Miscible processes are aimed at recovering oil which would normally be left behind as residual oil, by using a displacing fluid which actually mixes with the oil. Because the miscible drive fluid is usually more mobile than oil, it tends to bypass the oil giving rise to a low macroscopic sweep efficiency. The method is therefore best suited to high dip reservoirs. Typical miscible drive fluids include hydrocarbon solvents, hydrocarbon gases, carbon dioxide and nitrogen. 

Chemical processes work either to change the mobility of a displacing fluid like water, or to reduce the capillary trapping of oil in the rock matrix pores. Reducing the mobility of water, for example by adding polymers, helps to prevent fingering, in which the less viscous water bypasses the oil and... 

Oil recovery can also be affected by extreme variations in rock permeability, such as when high-permeability thief zones between injectors and producers allow most of the injected drive fluid to channel quickly to producers, leaving oil in other zones relatively unrecovered. A need exists for a low-cost fluid that can be injected into such thief zones (from either injectors or producers) to reduce fluid mobility, thus diverting pressure energy into displacing oil from adjacent lower-permeability zones. 

Alteration is always a cause for concern in geochemical investigations and the best approach will always be to avoid samples with visual or chemical evidence for alteration. The differential fluid mobility of U, Th, Pa and Ra undoubtedly provides the potential for weathering or hydrothermal circulation to disturb the U-series signatures of arc lavas. In a study of lavas from Mt. Pelee on Martinique, Villemant et al. (1996) found that domeforming lavas were in U-Th equilibrium whereas plinian deposits from the same eruptions had small U-excesses which they interpreted to reflect hydrothermal alteration. However, whilst the addition of U could be due to hydrothermal alteration, the plinian deposits were also displaced to lower °Th/ Th ratios which cannot. Instead, the two rock types may just be from separate magma batches. 

The displacing fluid may be steam, supercritical carbon dioxide, hydrocarbon miscible gases, nitrogen or solutions of surfactants or polymers instead of water. The VSE increases with lower mobility ratio values (253). A mobility ratio of 1.0 is considered optimum. The mobility of water is usually high relative to that of oil. Steam and oil-miscible gases such as supercritical carbon dioxide also exhibit even higher mobility ratios and consequent low volumetric sweep efficiencies. 

Polymers increase the viscosity of the soil washing fiuids. Increased viscosity provides mobility control, which reduces the fingering of the displacing fiuid past the displaced fluid. It also helps ensure that the contaminated area is efficiently contacted by the soil washing solution. 

When the mobility ratio is greater than one, the front between the displaced and displacing fluids is unstable if the porous medium is sufficiently wide (> 10 cm) to allow the formation and propagation of viscous fingers. The lower viscosity fluid channels or "fingers" through the displaced fluid, leaving much of it uncontacted. (See... 

Figure 1.) The channeling can occur with either miscible or immiscible floods and results in much lower production of the displaced fluid for any given throughput of the injection fluid once the latter reaches the production well (10,12,15). The problem, which is common to water flooding and to all EOR processes, is most severe for gas flooding simply because it is in gas flooding that the injected fluids have the lowest viscosities (most unfavorable mobility ratios).

The inherently unstable nature of miscible displacements with unfavorable mobility ratios (the viscosity of the displacing fluid is less than the viscosity of the displaced fluid) and unfavorable density ratios (for a downward vertical displacement, the density of the displacing fluid is greater than the density of the displaced fluid) has been well documented (6-18). For a downward vertical displacement with a favorable mobility ratio and an unfavorable density ratio. Hill ( ) proposed an approximate theory that... 

Achievement of low mobility ratios at the fronts between displacing and displaced fluids is of even greater concern in enhanced oil recovery than in waterflooding owing to the high costs and/or low viscosities of the injected fluids. One response to this concern has been the continuing effort to develop a fundamental understanding of so-called foam flow, which employs aqueous solutions of properly chosen surfactants at relatively low capillary numbers to reduce the effective mobility of low viscosity fluids (see 5,6 and papers on foam flow in this volume). 

This developed miscibility process results in a miscible fluid, that is capable of displacing all the oil which it contacts in the reservoir... The efficiency of this displacement is controlled by the mobility (ratio of relative permeability to viscosity) of each fluid. If the displacing fluid (i.e. carbon dioxide) is more mobile than that being displaced (i.e. crude oil) then the displacement will be relatively inefficient. Some of the residual oil saturation will never come into contact with carbon dioxide. Both laboratory and field tests have indicated, that even under favourable condition, injection of 0.15-0.6 10 m of carbon dioxide is required for recovery of an additional barrel (0.16 m ) of oil". Here our goal is to obtain a mass ratio of CO2 to incremental oil of 1 to 4, on the basis of the Bonder s data. 

Mobility control is one of the most important concepts in any enhanced oil recovery process. It can be achieved throngh injection of chemicals to change displacing fluid viscosity or to preferentially rednce specific flnid relative permeability through injection of foams, or even through injection of chemicals, to modify wettability. This chapter does not address a specific mobility control process. Instead, it discusses the general concept of the mobility control requirement in enhanced oil recovery (EOR). 

In enhanced oil recovery processes, such as polymer flooding, one fluid (or even several fluids) displaces several mobile fluids (e.g., water and oil). According to the conventional concept, when one or several fluids displace several mobile fluids ahead, the total mobility of displacing fluids should be equal to or less than the total mobility of the several displaced fluids (Dyes et al., 1954 Lake, 1989) ... 

Our first task is to evaluate the validity of the conventional concept about the mobility control requirement using a simulation approach. This model uses the UTCHEM-9.0 simulator (2000). The dimensions of the two-dimensional XZ cross-section model are 300 ft x 1 ft x 10 ft. One injection well and one production well are at the two extreme ends in the X direction, and they are fully penetrated. The injection velocity is 1 ft/day the initial water saturation and oil saturation are 0.5. The displacing fluid is a polymer solution. The purpose of using the polymer solutuion in the model is to change the viscosity of the displacing fluid. Therefore, polymer adsorption, shear dilution effect, and so on are not included in the model. To simplify the problem, it is assumed that the oil and water densities are the same that the capillary pressure is not included that the relative permeabilities of water and oil are straight lines with the connate water saturation and residual oil saturation equal to 0 and that the water and oil viscosity is 1 mPa s. Under these assumptions and conditions, we can know the fluid mobilities at any saturation. The model uses an isotropic permeability of 10 mD. 

The grid blocks tested are listed in Table 4.1. The recovery factors (RF) from each grid shown in the table are all greater than 99.48% at one pore volume (PV) of injection. That means, at least from the recovery factor point of view, all these models provide reasonably accurate results (close to theoretical RF of 100% for the built base model with the mobilities of displacing and displaced fluids being equal). 

In Case visc03, even though the oil viscosity is 100 mPa-s, the same as that in visc02, when the injection fluid mobility is adjusted to be the same as the oil mobility only (not the total mobility) based on Eq. 4.9, the recovery factor is 98.34%, almost the same as that in Case viscOl (only 1% different). Based on these results, we can see that to satisfy the mobiUty control requirement for a high oil recovery factor (favorable displacement condition), the injection mobility should be equal to or less than the oil mobility corrected by the initial oil saturation by Eq. 4.9, not the total mobility of fluids ahead of the displacing front. 

The water saturation distributions for the previous cases can further explain what would happen at different mobility ratios. The water saturation profile for Case viscOl at 0.5 PV injection is shown in Figure 4.6. Because the mobility ratio between the displacing fluid and displaced fluid is 1, the displacing front is stable. The finger is not further developed, and the displacing front is sharp. 

The physical meaning of M c dehned by Eq. 4.14 is the mobility ratio of the displacing fluid to the displaced oil phase in the assumed oil channel. This mobility ratio in the assumed oil channel is the mobility ratio of the displacing fluid to the displaced oil phase multiplied by the normalized movable oil saturation, So. 

One obvious mechanism in polymer flooding is the reduced mobility ratio of displacing fluid to the displaced fluid so that viscous Angering is reduced. When viscous Angering is reduced, the sweep efficiency is improved, as shown in Figure 1.2. This mechanism is discussed extensively in the waterflooding literature it is also discussed in Chapter 4. When polymer is injected in vertical heterogeneous layers, crossflow between layers improves polymer allocation in the vertical layers so that vertical sweep efficiency is improved. This mechanism is detailed in Sorbie (1991). 

---