Incremental oil recovery

Oil-field chemistry has undergone major changes since the publication of earlier books on this subject Enhanced oil recovery research has shifted from processes in which surfactants and polymers are the primary promoters of increased oil production to processes in which surfactants are additives to improve the incremental oil recovery provided by steam and miscible gas injection fluids. Improved and more cost-effective cross-linked polymer systems have resulted from a better understanding of chemical cross-links in polysaccharides and of the rheological behavior of cross-linked fluids. The thrust of completion and hydraulic fracturing chemical research has shifted somewhat from systems designed for ever deeper, hotter formations to chemicals, particularly polymers, that exhibit improved cost effectiveness at more moderate reservoir conditions. 

Table 11.2 Illustration of the effect of changing the composition of injected solution on incremental oil recovery in an enhanced oil-recovery process. ... 

Field Application. Field trials of classical alkaline flooding have been disappointing. Mayer et al. (60) indicated that only 2 of 12 projects had significant incremental oil recovery North Ward Estes and Whittier with 6-8 and 5-7% pore volume, respectively. Estimated recovery from the Wilmington field was 14% with a classical alkaline flooding method (61). However, post-project evaluation of that field indicated no improvement over water-flooding (62). 

A common measure of the success of an EOR process is the incremental oil recovery factor. Figure 1.3 shows the schematic of incremental oil recovery from an EOR process. The oil production rates from B to C are extrapolated rates, and the cumulative oil at D is the predicted ultimate oil recovery had the EOR process not been initiated at B. The time from B to C is required to... 

Another measure of the success of chemical EOR is the amount of chemical injected in pounds per barrel of incremental oil produced (Ib/bbl), or tons of oil produced per ton of chemical injected, a figure often used in China to represent polymer flooding efficiency. Chang et al. (2006) reported that incremental oil recovery factors of up to 14% of the OOIP have been obtained in polymer flooding good-quaUty reservoirs, and incremental oil recovery factors of up to 25% of OOIP have been reported in ASP pilot areas. 

BP laboratory results (Lager et al., 2006) showed an average benefit of 14% with low-salinity brine, and a large scatter of results from -e4 to -t40% was observed. Such a wide spread of results was also observed by Morrow and coworkers. In some core floods, no incremental oil recovery was observed. 

Martin (1959) and Bernard (1967) observed that clay swelling and/or dispersion accompanied by increased pressure drop resulted in incremental oil recovery. Tang and Morrow (1999) concluded that line mobilization (mainly kaolinite) increased recovery based on their observations (1) fired/acidized Berea core showed insensitivity of salinity on oil recovery, whereas unlired Berea core did show sensitivity and (2) for clean sandstones, the increase in oil recovery with the decrease in salinity was less than that for the clay sands. Figure 3.4 shows some of their results. In the tests, the reservoir CS core was used. The reservoir brine, CS RB, was used as connate brine for the entire CS core tests. 

Table 5.13 shows more results based on the economic analysis of simulation data. This table shows that more polymer injection corresponds to higher incremental oil recovery but lower economic parameters (tons of incremental oil per ton of polymer injection Qi and Feng, 1998). 

For the same amount of polymer fixed, the question of optimization remains. Which option is better a higher concentration with a smaller injection pore volume or a lower concentration with a larger injection pore volume Generally, the ultimate incremental oil recovery mainly depends on the total amount of polymer injected. A higher concentration could result in more initial water-cut reduction due to polymer injection. However, a high concentration may be limited by the allowable injection pressure. From a mobility control point of view, a higher concentration should be injected at the front to counteract dilution. A commonly used concentration in China is around 1200 mg/L. 

Similar to (but not the same as) the concept to inject different MW polymers, different polymer concentrations can be injected for profile control. Yang et al. (2006) presented laboratory and pilot test results showing that the recovery increased by injecting high-concentration polymer solution in the early slugs. In this case, the high concentration used was 1500 to 2500 ppm, and the incremental oil recovery over waterflooding was about 20%. 

Laboratory and Numerical Simulation Studies Both 2D and 3D physical models were built to study the effectiveness of the profile control. In the 2D model, the incremental oil recovery factor was 8.19% over aquifer drive. In the 3D model, the incremental oil recovery factor was 6.2% (Li et al., 2005c). In the 3D model, 0.08 PV of 3000 mg/L polymer was injected. When crosslinked polymer was injected, high permeability channels were immediately blocked, the injection pressure rose, and the water cut fell. However, because of strong edge water, water bypassed the blocked zone, the injection pressure fell, and the water cut quickly rose again. A numerical simulation was carried out to study the feasibility of polymer injection and optimize the program (Yao et al., 2005). The optimum concentrations from the laboratory results were 0.3 to 0.5% polymer, 0.2% crosslinker concentration, pH 5,... 

Positive results similar to the pilot test were observed from these treatments. The total incremental oil recovery factor was 5.8%. In this case, the following measures were taken in the implementation ... 

The polymer injection was started in May and ended in December 2002. A 0.248 PV polymer solution with 1248 mg/L concentration (total 310 mg/L PV) was injected the injection solution viscosity was 54.8 mPa s. In November 2002, the injection pressure was 9.5 MPa compared with 5.4 MPa before injection. The increase was 0.6 MPa lower than that in the compared area where a polymer solution mixed with fresh water was injected. The water cut decreased by 38.6%, 9.4% more than that in the compared area. During the polymer injection period, the incremental oil recovery factor was 6%. The produced water was injected in other areas of the Lamadian field. Thus no produced water was disposed. 

The produced water had some polymer. The produced water may be injected before or after the main polymer slug as preflush or postflush slugs. A calculation showed that 2.5% incremental oil recovery could be obtained if the produced water was injected as preflush and 0.9% as postflush. A laboratory test showed that if the produced water had 400 mg/L polymer (with solution viscosity of 2.5 mPa s), the incremental oil recovery factor was about 3% (Zhang, 1998). 

The first pilot test in Daqing was started on August 30, 1972, and ended on September 24 in the same year, a total of 26 days. This test was conducted in the Sall7+8 layer. One inverted four-spot pattern was used with the injector, Well 501, in the center. Thus, it was called the Well 501 pattern. The distance between the injector and a producer was 75 m. The formation thickness was 5.2 m, and the permeability was 631 md. The reservoir temperature was 45°C. A 0.163 PV of polymer solution was injected having a concentration from 1000 to 1800 mg/L. The three producers started to respond after 12 days of polymer injection. The water cut at one producer (Well 503) was reduced from 99 to 60.4%, and the well pattern incremental oil recovery was about 5%. The well injection pressure increased, and the liquid production rate was reduced significantly. In this first pilot test, low molecular weight polymer (3 to 5 million Daltons) was used (Liu, 1995 Yang et al., 1996). 

On February 10, 1988, another pilot test was started in Bei-3-Qu-Xi PIl-3 in the Saertu field, Daqing, and ended on September 4, 1990. There were 4 injectors and 9 producers, with the distance between injector and producer being 200 m. The incremental oil recovery was 3% (Yang et al., 1996). 

Before PF, 0.1153 PV water had been injected from October 1996 to August 1997. A total polymer injection of 597.64 mg/L PV was started on August 28, 1997, and ended in October 2002. The average injected polymer concentration was 892 mg/L, and the injection PV was 0.67. The viscosity at the injection wellhead was 20 mPa s. In this case, the incremental oil recovery factor was 8.55%. 

The profile control agent was aluminum citrate. The penetration radius was 16.3 to 36.7 m, which is about 1/6 to 1/4 of the injector-producer distance. The AT-430 polymer was used at a concentration of 1000 mg/L. Injection of the profile control agent was started on March 28, 1986. After the control agent was injected, polymer was injected starting from December 4, 1986. A total of 124 mg/L PV was injected. The incremental oil recovery factor was 11.5%. For one ton of polymer injected, 514 tons of oil were recovered. However, it was realized that the amount of polymer injected in this project was not large enough. The produced polymer showed that the polymer MW was reduced... 

Daqing polymer flooding performance showed that incremental oil recovery factor peaked when the amount of polymer injected was 180 to 210 mg/L PV. 

In this section, simulation results are compared with the information from the literature for different polymer and surfactant-polymer injection schemes. We expect that UTCHEM simulation of a core-scale chemical process is the best simulation approach to study mechanisms. In this study, we use a ID core flood model with 100 blocks to represent a 1-foot-long core. The permeability is 2000 md, and the water and oil viscosities are 1 and 2 mPa s, respectively. To optimize injection schemes, we compare the incremental oil recovery factors over waterflooding and chemical costs. Chemical costs are evaluated using the amounts of chemicals injected per barrel of incremental oil (Ib/bbl oil). 

To investigate the effect of the amounts of polymer and surfactant injected, we use different concentrations and slug sizes and compare the incremental oil recovery factors and chemical costs (chemical Ib/bbl incremental oil) for different amounts of polymer and surfactant injection. The amount of surfactant injected is commonly presented in concentration (%) PV(%), and the amount of polymer is commonly presented in mg/L PV(fraction). Note that the unit PV is in fraction of pore volume. This section presents both polymer and surfactant in concentration (%) PV(%). The relationship between these two units is mg/L PV(fraction) = 100 concentration (%) PV(%). Figures 9.5 and 9.6 show the results. From these two figures, we can see that the more chemicals injected, the more incremental oil is recovered. However, in general, the chemical cost per barrel of incremental oil is also increased. Apparently, when a low load of chemicals is injected, the chemical cost per barrel of incremental oil is not sensitive to the amount of chemical injected. This observation is seen for both polymer and surfactant. 

Almost all chemical flood projects are started after some waterflood history. We want to know whether early chemical injection could be a better option. To do that, we change the water injection PV before chemical injection so that average oil saturations (So) before SP are different. The results are shown in Table 9.2. We can see that different total injection PVs are required to achieve about the same incremental recovery factor. The incremental oil recovery factor (RF) is defined as the RF from an SP case minus the RF from the 1.5 PV waterflooding case. The later SP is started, the higher the total injection PV is reqnired. Therefore, it is better to start snrfactant-polymer flood earlier to accelerate prodnction, and thns, less water will be injected. Such results have been confirmed by the ASP corefloods in Daqing (Li, 2007). 

To optimize the flooding processes, we first have to select which optimization criterion to use. Generally, we choose incremental oil recovery factor as a criterion. Alternatively, we may choose maximum NPV as a criterion with economic analysis. The latter choice is more proper because it takes into account discounted cash flow. However, performing economic analysis requires more economic data that are generally not available. The criterion to be used depends on the objective. 

This section discusses both the incremental oil recovery factor and chemical cost per barrel of incremental oil recovered. We have seen that the two criteria sometimes give different answers regarding the optimum process. 

Huang and Yu (2002) observed that emulsification was not completely reversible. When the dynamic IFT reached ultralow, emulsification occurred. Even when dynamic IFT went up, emulsified oil droplets did not easily coalesce. In alkaline flooding, emulsification is instant, and emulsions are very stable. From this emulsification point of view, the dynamic minimum IFT plays an important role in enhanced oil recovery. From the low IFT point of view, we may think we should use equilibrium IFT because reservoir flow is a slow process. However, the coreflood results in the Daqing laboratory showed that when the minimum dynamic IFT reached 10 mN/m level and the equilibrium IFT was at 10 mN/m the ASP incremental oil recovery factors were similar to those when the equilibrium IFT was 10 mN/m (Li, 2007). One explanation is that once the residual oil droplets become mobile owing to the instantaneous minimum IF F, they coalesce to form a continuous oil bank. This continuous oil bank can be move even when the IFT becomes high later. Then for this mechanism to work, the oil droplets must be able to coalesce before the IFT becomes high. It can be seen that it will be more difficult for such a mechanism to function in field conditions rather than in laboratory corefloods. This mecha-... 

From the previous discussions, we can see that ultralow IFT cannot be reached in alkaline flooding. The incremental oil recovery is not correlated with the IFT or crude acid number. The low IFT mechanism may not be the dominant mechanism. However, a reasonably low IFT is required for emulsification to occur, which is another proposed mechanism and summarized in the previous section. 

The oil recovery factor is shown in Figure 10.23. From this figure, we can see that the incremental oil recovery factor of alkaline flooding over waterflooding is about 4%. Table 10.13 also serves as an example explaining how to input salinity data into a performance prediction model. 

Figure 10.33 shows the cumulative percentage versus the total injected alkali, which is presented by the product of concentration (%) and slug size (% PV). At the 50% cumulative percentage, the product is 17. For most of the field projects, the incremental oil recovery factors were 1 to 2%. 

---