Waterflooding observations

An alternative to this process is low (<10 N/m (10 dynes /cm)) tension polymer flooding where lower concentrations of surfactant are used compared to micellar polymer flooding. Chemical adsorption is reduced compared to micellar polymer flooding. Increases in oil production compared to waterflooding have been observed in laboratory tests. The physical chemistry of this process has been reviewed (247). Among the surfactants used in this process are alcohol propoxyethoxy sulfonates, the stmcture of which can be adjusted to the salinity of the injection water (248). 

B. licheniformis JF-2 and Clostridium acetogutylicum were investigated under simulated reservoir conditions. Sandstone cores were equilibrated to the desired simulated reservoir conditions, saturated with oil and brine, and flooded to residual oil saturation. The waterflood brine was displaced with a nutrient solution. The MEOR efficiency was directly related to the dissolved gas/oil ratio. The principal MEOR mechanism observed in this work was solution gas drive [505]. 

The observed increase in waterflood recovery indicated that rock surface wettability might be changed as a result of cyclic waterflooding—from Flood 1 (TlFl) to Flood 3 (T1F3). 

It was also observed that the compaction did not stop in the waterflooded areas, even though the reservoir was repressurized to the initial condition. Thus, seawater appeared to have a special interaction with chaUc at high temperatures, which has an impact on oil recovery and rock mechanics (Austad et ah, 2008). Austad and his coworkers started to work on the issues related to seawater flooding in carbonate reservoirs in 1990s. In the next section, the salinity effect on oil recovery is briefly summarized. 

Figure 6.19 shows a set of relative permeability curves for waterflooding and polymer flooding. The following observations can be made ... 

The pilot test included 10 injectors, 16 producers, 2 observation wells, and 2 wells for coring. Before the test, the water cut was 97.4%, and the recovery factor was 34.3% with the expected waterflooding recovery factor of 36.3%. The residual oil saturation was 0.33, and the average sweep efficiency was 0.54. The injection pressure was about 11.9 MPa (1725.8 psi). 

Figure 10.17 illustrates the pressure and saturation changes that occur during an alkaline waterflood. Shown are typical pressure gradients and oil saturations during a flood of a sand-packed column previously waterflooded to residual oil saturation. Oil and water flow simultaneously ahead of the alkaline water front. Note the sharp gradient in oil saturation that occurs at the front (Figure 10.17c), and the rise and fall in pressure gradient behind the alkaline water front (Figure 10.17b). The schematic (Figure 10.17d) illustrates what is observed microscopically.

Castor et al. (1981b) observed that the IFT in the alkaline flooding was on the order of 0.1 mN/m. Their capillary numbers of alkaline floods are presented in Figure 10.18. The capillary number of alkaline floods was about 100 times higher than the capillary numbers in waterfloods. The alkaline flooding results from Castor et al. show that the recovery efficiencies could be better correlated with the stability of emulsions and wettability alteration than with IFT of the systems. 

In summary, P/A was better than A/P, and A-PP was better than A/P or P/A. These observations were also made by Chen et al. (1999b). Figure 11.6 shows the residual oil recovery factor after waterflooding. The system was as follows oil viscosity, 180 mPa s at room temperature polymer, 5000 mg/L HPAM and alkali, Na2Si04. This figure shows that A-pP was better than any sequential or single injection process. This result was also observed when a biopolymer or less-viscous oil (62 mPa s) was used. Although in this example A/P was much better than P/A, P/A was better than A/P in the case of 62 mPa s (Krumrine and Falcone, 1983). 

In these three schemes, the amount of alkali or polymer was the same. Only the injected surfactant was gradually reduced. In these schemes, it was observed that the incremental oil recovery factors over waterflooding were almost the same. Because less surfactant was injected in Schemes 2 and 3, Schemes 2 and 3 were economically more attractive than Scheme 1. 

September 23, 1994, followed by 0.327 PV ASP flood, and 0.273 PV polymer drive and water drive. The ASP solution viscosity was 16 mPa s. During water preflush, the oil recovery before ASP was 31.63% from the SII1.3 layer. The response to ASP injection was observed in November 1994 (after 0.0693 PV of injection). The average water cut in the entire pilot area decreased from 82.7% to a low of 59.7%, and the daily oil production increased from 37 mVd toapeakof91.5 mVd. The water injectivity decreased from 1.75 mV(md-MPa), stabilized at about 1.42 mV(m-d-MPa), and then dropped to 1.19 mV(m d-MPa). In general, after an ASP slug is injected, flow resistance increases, and water injectivity decreases. The simulation prediction showed about 20% incremental oil recovery factor over waterflood. The early performance matched very well with the simulation prediction. In this pilot test, the simulator used was GCOMP. 

Charles et al. (1985) observed that in oilfield waterflood systems, some attempts to use metallic salts and oxides as scavengers have resulted in the formation of undesirable solid, metallic sulfides. But both formaldehyde and acrolein are aldehydes and are the most frequently nsed chemical scavengers in oilfield waterflood operations. However, they noted that the best performance of these sulfide scavengers can be seen in surface cleaning operations such as the treatment of oilfield water-flood systems (Charles et al. 1985). 

The distribution of the oil, gas and water in the porous medium was better understood when Botset and Wyckoff (9) carried out the first experiments on relative permeability. They showed that either oil or gas would flow only if a specific minimum saturation of the phase in question existed in the flow region of the porous material. Some of the early workers also recognized that either the oil or gas droplets could be discontinuous, and in this condition, would be hard to displace by flowing water because of the Jamin effect. In 1927, Uren and Fahmy (10) investigated a number of "factors which affect the recovery of petroleum from unconsolidated sands by waterflooding. Table 1 lists these factors and the general results observed by Uren and Fahmy. With one exception (rate), the results observed by Uren and Fahmy are similar to generalizations which most experts in this field claim today after work of more than 50 years. 

Run DY-15 shown in Figure 4, In this experiment, the flow rate decreased to zero just before breakthrough of the driving fluid (at about T = 0.6 PV in a nominal waterflood). Substantial differences in the upstream pressure of the flow apparatus (400 psi at T = 1.2 PV) and the differential pressure across the core (75 psi at T = 1.2 PV) indicate that most of the emulsification and entrapment occurred in the entrance region of the sandpack. This mechanism was repeatedly observed in high pH, non-saline floods of moderate acid number (> 2.0) oils. 

For an ordinary water flood under water-wet conditions, is usually in the range of 10 to 10The critical capillary number may be in the range of 10 to 10 whereas complete desaturation of the nonwetting phase (oil) may occur at a capillary number in the range of 10 to 10 [JO]. The waterflooded residual oil saturation may be in the range of 30 to 40%. It must be noticed that these data are mainly based on model cores (Berea and other outcrop sandstone cores) which have never been in contact with reservoir crude oil. Much lower values are, however, observed under mixed-wet conditions [JJ]. This implies that it is about... 

In conventional oil recovery activities via waterflooding, low yields are normally observed, basically as a result of high oil viscosity and high interfacial tensions developed when water is injected. When the viscosity of the injected fluid is lower than that of the fluid to be displaced, the previous one flows more swiftly than the latter across the porous medium, often finding preferred paths. In view of high-interfacial tensions, the capacity of the injected fluid to displace the oil is rather impaired, resulting in high contents of residual oil in the wells. 

This same water-cut flat or dip was also observed in the Niagara pilot flood. In the cases where considerable mobile water is present at the start, the floods started on the waterflood curve at an elevated water-oil ratio value. 

---