Effect of Conformance Control Patterns and Size of the Slug of In Situ Supercritical CO2 Emulsion on Tertiary Oil Recovery by Supercritical CO2 Miscible Injection for Carbonate Reservoirs.

The reservoir heterogeneity is the major cause of poor volumetric sweep efficiency in sandstone and carbonate reservoirs. Displacing fluids (water, chemical solution, gas, and supercritical CO2 (sc-CO2)) flow toward the high permeable zone. A significant fraction of oil remains in the low permeable zone due to the permeability contrast. This study used in situ sc-CO2 emulsion as a conformance control agent to plug the high permeable zone and improve the low permeable zone's volumetric sweep efficiency in carbonate formation. We investigated the effect of two types of conformance control patterns and the size of sc-CO2 emulsion on tertiary oil recovery performance by sc-CO2 miscible injection for carbonate reservoirs at reservoir conditions. The conformance control patterns are achieved using two different approaches. In the first approach, the low permeable zone was isolated, and the diverting gel system, a 0.4 pore volume slug, was injected into a high permeable zone. In the second approach, the simultaneous injection of the diverting gel system, a 0.2 pore volume slug, was done on both the low and high permeable zones. The first sc-CO2 injection was conducted as a tertiary oil recovery mode to recover the remaining oil after water flooding. The diverting gel system was injected after the first sc-CO2 flood for the conformance control. The second or post sc-CO2 injection was conducted after the diverting gel system injection. The diverting gel system used in this study consisted of a polymer and a surfactant. An in situ emulsion was generated when the injected diverting gel system interacts with the sc-CO2 in the core plug. Results obtained from dual-core core flooding experiments suggested that the in situ sc-CO2 emulsion was generated successfully in the formation based on the different pressure increases and observation of the dual-core core flooding experiments. The volumetric sweep efficiency and oil recovery in both conformance control patterns were improved. The production performances were also compared for both conformance control models before and after the diverting gel system injection. The conformance control model 2 (simultaneous injection of the diverting gel system into low and high permeability cores) has a better choice to be applied in field application due to high recovery with a small sc-CO2 emulsion easy operation in the field.

Introduction

The heterogeneity of the reservoir could result in the early production of injection fluid. The excess water production of injection fluids from a hydrocarbon reservoir is an extremely serious problem during secondary and tertiary oil recovery processes. High water cut causes many problems such as decreased oil production, shortened life of oil production, increased water treatment cost, and so on. Conformance control is one of the major technologies to improve the volumetric sweep efficiency and oil recovery for both sandstone and carbonate reservoirs. Researchers and reservoir engineers have been following the development and application of the conformance control practice in the core and field scales. Conformance control materials can be classified as follows: (1) polymer gel emulsion, (2) oil-in-water or water-in-oil emulsion, (3) surfactant-assisted micro-emulsion, (4) nanoparticle-stabilized emulsion, and (5) CO2 emulsion.

Polymer Gel Emulsion

Sydansk and Southwell, Kabir, and Bai et al. have comprehensively reviewed the conformance control and water shutoff for diversified polymer gel technologies in the laboratory and field scales.[1−3] When the water cut reaches above 90%, polymer gels can be applied as a water shutoff treatment or the permeability blockers for the production zone flooded by water injection[4−11] and relative permeability modifiers to reduce water permeability and increase oil permeability.[12−15] In recent years, the research focus has been shifted to developing new cross-linkers to improve the conformance control efficiency for extremely complex environmental conditions such as high temperature, high reservoir pressure, high concentration of injection fluid, and extremely heterogeneous reservoir. The cross-linkers such as the less-toxic cross-linker (PEI), re-cross-linkable preformed particle gel, and terpolymer–gel system were developed. Their gel strength, gelation time, and polymer gels’ thermal stability by using cross-linkers have been evaluated at different temperatures for such conditions.[16,17] Lower oil prices in the market can be a hindrance to applying this technology in the field due to higher costs associated with effluent treatment.

Oil-in-Water or Water-in-Oil Emulsion

Water-in-oil (W/O) or oil-in-water (O/W) emulsion is another conformance control technology. The basic mechanisms of improving displacement efficiency using emulsion or micro-emulsion rely on a pore throat blockage by a droplet known as the “Jamin” effect. When an emulsion droplet tries to pass through a smaller pore throat than the droplet’s diameter, a capillary resistance force generates, which is the opposite of the driving force and leads to the redistribution of fluids in the formation. To date, many researchers and engineers have proposed a variety of emulsion treatments for oilfield applications.[18,19,28,20−27] O/W or W/O emulsion’s main advantages include moderate injectivity due to the low bulk viscosity of O/W emulsion, less formation damage, and low cost.[21,29,30] This technique can be applied if an O/W or W/O is generated in the formation. Unfortunately, in most cases of field application, the crude oil produced from the reservoir is used to form the emulsion with water in the well site and then injected such emulsion into the reservoir again, which is a risky approach and may result in the additional residual oil from such emulsion left in the reservoir.

Surfactant-Assisted Micro-emulsion

To improve the stability and rheology of the emulsion or micro-emulsion for the conformance control in the high permeable zone and the oil displacement efficiency in the low permeable zone, surfactant-assisted micro/nano-emulsion technologies have been evaluated in the past decade.[31−36] The improvement of oil displacement efficiency by using surfactant-assisted micro-emulsion relies on interfacial tension (IFT) reduction. The droplets in such emulsion can plug the pore throat. The anionic surfactants, such as nonyl surfactants, sodium sulfonates, TRS 10–80 petroleum sulfonate, WITCO TRS-18 petroleum sulfonate, sodium lignosulfonate, sodium dodecylbenzene sulfonate, alkyl sulfate Gemini surfactant, Guerbet alkoxy sulfate, sulfonate surfactants, and so on, and nonionic surfactants such as ethoxylated alcohols, Triton X100, Span-80 and Tween 80, 1,3,5-triazine surfactants, Tergitol, and so on, were used to form emulsion for reducing IFT, improving solubilization capacity and compatibility with oil phase, and increasing stability in high-salinity brines. The surfactants can retain on rock surfaces that could increase the cost significantly. In addition, the interactions of surfactants with minerals result in produced effluent from the production well in which it is difficult to separate oil from the effluent.

Nanoparticle-Stabilized Emulsion

The application of nanoparticle-assisted micro-emulsion in reservoir engineering is another attractive choice.[37−43] The nanoparticles have a large surface area, and small size can be several orders of magnitude low compared with mineral fines. Nanoparticles could easily move into the pore and pore throat with a very small amount of adsorption and without severe retention and formation damage compared with surfactants. The physical properties of micro-emulsions have been greatly improved by using nanoparticles. With the continuous development of nanoparticles in the field of tertiary oil recovery, the synergistic effect of surfactant- and nanoparticle-assisted emulsion as a displacing agent or mobility/conformance control agent has been proposed in the past decade. This technique may embody the combined effect of physicochemical and mechanical mechanisms when it flows through a porous medium. The emulsion assisted by a surfactant with nanoparticles will reduce IFT and increase its bulk viscosity and strengthen the O/W interface of emulsions.[44−47] These techniques are being studied in the laboratory. The main research focuses on its stability, rheological properties, particle suspension, and aggregation with different rocks at harsh conditions. It is too early to make a comprehensive evaluation of the technique for field application.

CO2 Emulsion

So far, research on the conformance control by CO2 emulsion is rare. Al Otaibi et al. used two kinds of Aerosil fumed and Ludox colloidal silica nanoparticles to screen the possibility of generating both water-in-CO2 and CO2-in-water emulsions.[48] They used hydrophilic nanoparticles in the water/isooctane system to generate water-in-CO2 emulsion. The other emulsion, CO2-in-water emulsion, was formed using hydrophilic nanoparticles in the same water/isooctane system. To the characteristics of two types of emulsions, the emulsions’ stability and viscosity were studied. Stability and viscosity of the emulsions, water-in-CO2, and CO2-in-water emulsions were studied with different ratios of isooctane to water at a temperature of 25 and 100 °C, respectively. The results showed that the stable water/isooctane emulsion depends on nanoparticles’ concentration and the volume percentage of water/isooctane. For CO2/water emulsion and water/CO2 emulsion, both emulsions’ viscosities increase with an increase in nanoparticles’ concentration when the CO2/water volume ratio was constant and decreased in the CO2 volume percentage in the system when the nanoparticle concentration was constant.

In this study, we used in situ sc-CO2 emulsion consisted of sc-CO2 and the diverting gel system (water-based polysaccharide linear polymer and surfactant) as a conformance control agent to improve the volumetric sweep efficiency in the lower permeable zone using reservoir cores with different levels of permeability of rocks (permeability contracts). Two conformance control patterns were designed and carried out experimentally at reservoir conditions using a dual-core core flooding apparatus. The stability, compatibility, and rheological studies have been reported in our previous publication.[49] To improve the volumetric sweep efficiency in the lower permeable zone, we studied the effect of two types of conformance control patterns on oil recovery by sc-CO2 miscible injection for carbonate reservoirs.

Results and Discussion

In the core flooding experiments for experiments 1 and 2, the HPCP1 or HPCP3 core holder has been placed horizontally above the LPCP2 or LPCP4 core holder for both experiments. The main objective is to investigate the effect of different conformance control patterns using in situ sc-CO2 emulsion on oil recovery. The two physical models of the conformance control patterns were studied in this work. In experiment #1 represented by physical model 1, the diverting gel system was injected to HPCP1 alone in the presence of sc-CO2 in the core, as shown in Figure . In experiment #2 represented using physical model 2, the diverting gel system was injected into both HPCP3 and LPCP4 in the presence of sc-CO2 flooding in the core, as shown in Figure .

Figure 1

Schematic diagram of physical model 1.

Figure 2

Schematic diagram of physical model 2.

Oil Recovery by Seawater Flooding

The purpose of conducting seawater flooding in the dual-core core flooding as a secondary oil recovery mode is to obtain the water-flooded oil recovery factor, simulate the performance of seawater flooding physically, and establish remaining oil saturation, Sorw in the case of displacing oil by water for heterogeneous carbonate reservoirs. The process is represented by physical models consisting of high and low permeable core plugs. The remaining oil after seawater flooding is the target to be recovered by sc-CO2 miscible flooding.

Experiment #1

Experiment #1 includes two composite core plugs, HPCP1 (composite #1) consisted of carbonate core plugs A and #B, and LPCP2 (composite #2) consisted of carbonate core plugs C and D. The permeability ratio of HPCP1 to LPCP2 or permeability contrast for experiment #1 was 43:1, which is based on the brine permeability of the core plugs as listed in Table . Before the diverting gel system was injected into the HPCP1, seawater and initial sc-CO2 flooding have been conducted at reservoir conditions.

Table 1

Dynamic Data of Brine Permeability, Initial Water Saturation, and Original Oil in Composites

experiment ID	composite ID	length (cm)	diameter (cm)	PV (cc)	Kb (mD)	Swi (%)	Soi (% OOIC)
experiment 1 (physical model 1)	composite #1 (HPCP1)	6.39	3.8	20.57	966.7	24.64	75.36
	composite #2 (LPCP2)	6.02	3.8	14.06	22.3	17.56	82.44
experiment 2 (physical model 2)	composite #3 (HPCP3)	8.23	3.8	22.5	1400	9.87	90.13
	composite #4 (LPCP4)	8.55	3.8	18.23	23.47	18.65	81.35

Table provides the initial conditions of water and oil saturation (Swi and Soi) for both the HPCP1 and the LPCP2 composite cores at the beginning of the seawater injection. Simultaneous seawater injections into both the HPCP1 and the LPCP2 were conducted at injection flow rates of 0.5 cc/min for 2 PVs, 1.0 cc/min for 1 PV, and 2.0 cc/min for 1 PV. After injecting about 4.0 PV of total seawater injection and achieving a water cut of 99%, the remaining oil saturations were determined for both the HPCP1 and the LPCP2 using the material balance method. The remaining oil saturation after seawater flooding was about 49.2% in OOIC for HPCP1 and 58.6% in OOIC for LPCP2, shown under Sorw in Table .

Table 2

Summary of Oil Recovery and Residual Oil with Different Injection Fluids for Experiments #1 and #2a

					seawater flooding		initial sc-CO2 flooding		2nd sc-CO2 flooding
composite ID	L (cm)	D (cm)	PV (cc)	Swi (%)	RF (%)	Sorw (%)	RF (%)	Sorsc-CO2 (%)	RF (%)	Sor2nd-sc-CO2 (%)
composite #1 (HPCP1)	6.39	3.8	20.57	24.6	50.8	49.2	47.4	1.83	X	1.83
composite #2 (LPCP2)	6.02	3.8	14.06	17.56	41.4	58.6	21.8	36.8	19	17.75
composite #3 (HPCP3)	8.23	3.8	22.5	9.87	49.9	50.1	44.9	5.2	2.5	2.7
composite #4 (LPCP4)	8.55	3.8	18.23	18.7	46.4	53.6	7.8	45.8	21	24.8

L: length, D: diameter, RF: recovery factor, PV: pore volume of the composite core plug, Swi: initial water saturation, Sorw: remaining oil saturation after water flooding, SorCO2: remaining oil saturation at initial sc-CO2 flooding, Sor2nd-CO2: residual oil saturation at the second sc-CO2 flood.

Experiment #2

Experimental procedures and conditions of seawater flooding in experiment #1 were applied to experiment #2. The routine data of core plugs used for experiment #2 are listed in Table . Initial conditions of water and oil saturation (Swi and Soi) for both the HPCP3 and LPCP4 cores at the beginning of the seawater injection are listed in Table . The injection strategy and pore volume remained the same as in experiment #1. The permeability contrast is 60:1 based on the brine permeability. Simultaneous seawater injections into both the HPCP3 and LPCP4 were conducted at injection flow rates of 0.5 cc/min for 2.8 PVs, 1.0 cc/min for 1.2 PV, and 2.0 cc/min for 1 PV. After injecting about 5.0 PV of total seawater injection and achieving a water cut of 99%, the remaining oil saturations were determined for both the HPCP3 and the LPCP4 using the material balance method. The remaining oil saturation after seawater flooding was about 50.1% in OOIC for HPCP3 and 53.6% in OOIC for LPCP4, shown under Sorw in Table .

Table 3

Routine Data of Core Plugs and Assembly Composite Cores

test ID	composite ID	sample ID	length (cm)	diameter (cm)	PV (cc)	porosity (%)	air permeability (mD)
experiment 1 (physical model 1)	composite #1 (HPCP1)	A	3.03	3.8	9.5	28.1	917.3
		B	3.364	3.8	11.06	29.1	746
		A + B	6.394	3.8	20.56	28.6	831.7
	composite #2 (LPCP2)	C	2.878	3.8	6.25	19.4	51.5
		D	3.14	3.8	7.84	24.3	86.5
		C + D	6.018	3.8	14.09	21.9	69
experiment 2 (physical model 2)	composite #3 (HPCP3)	E	2.352	3.8	6.39	23.95	1729
		F	5.875	3.8	16.11	24.18	1559
		E + F	8.227	3.8	22.5	24.07	1644
	composite #4 (LPCP4)	G	2.374	3.8	5.3	19.7	53.8
		H	6.171	3.8	12.92	18.47	23.2
		G + H	8.545	3.8	18.23	19.08	38.5

Initial sc-CO2 Miscible Flooding after Seawater Injection

A significant fraction of the remaining oil was left after seawater injection. The tertiary oil recovery process was done using sc-SC2 miscible flooding. The experimental procedure and conditions remained the same as seawater flooding except for the injection flow rate of sc-CO2. The flow rate for both experiment #1 and experiment #2 was 0.2 cc/min.

Experiment #1

During the process of the sc-CO2 injection, a pore pressure of 3200 psi was set up to ensure that this process is miscible flooding. The minimum miscibility pressure between live crude oil and sc-CO2 for a particular carbonate reservoir is 2600 psi.[50] The sc-CO2 was initially injected into the two composites, HPCP1 and LPCP2, simultaneously at 0.2 cc/min for about a 0.75 total PV. The oil recovery performance of sc-CO2 injection is shown in Figure for both HPCP1 and LPCP2. Final oil recovery by initial sc-CO2 flooding was 47.4% in OOIC and 21.8% in OOIC for the HPCP1 and LPCP2. Residual oil saturation was less than 2% in OOIC for HPCP1 and 37% in OOIC for LPCP2 at the end of the initial sc-CO2 injection. These values are listed in Table . The results show that there is still a lot of crude oil left in the LPCP2 due to reservoir heterogeneity, and displacing fluids bypass low permeable zone despite seawater and initial sc-CO2 flooding.

Figure 3

Oil recovery by simultaneous sc-CO2 injection for both the HPCP1 and the LPCP2 at reservoir conditions (experiment #1).

Experiment #2

Table shows the remaining oil saturation after seawater injection for HPCP3 and LPCP4. After seawater injection, about 0.9 PV of sc-CO2 was injected simultaneously at a constant rate of 0.2 cc/min. When the injection pore volume reaches about 0.6, no more oil is produced. Figure shows recovery performance by initial sc-CO2 flooding for both HPCP3 and LPCP4. Final oil recoveries were 8 and 45% in OOIC for LPCP4 and HPCP3 at the end of the initial sc-CO2 flooding, respectively, as shown in Table . The reservoir heterogeneity in terms of permeability contrast resulted in a significant recovery difference in the LPCP4 and HPCP3. The poor displacement efficiency in LPCL4 can be attributed to the amount of sc-CO2 flow in the HPCP3 or bypass of the LPCP4.

Figure 4

Oil recovery by simultaneous injection of initial sc-CO2 for both HPCP3 and LPCP4 at reservoir conditions (experiment #2).

Diverting Gel System Injection and Oil Recovery Performance by Second sc-CO2 Miscible Flooding

The results after seawater and initial sc-CO2 flooding indicate that most of the oil was produced from the high permeable zone, and a significant amount of oil was left behind in the low permeable zone. To obtain such remaining oil and improve displacement efficiency from the low permeable zone, a slug of the diverting gel system (DGS) was injected into the high permeable zone after sc-CO2 flooding with different conformance patterns in experiments #1 and #2. Physical model 1 described in experiment #1 represents a field case with an isolation layer between high and low permeable zones. Physical model 2 in experiment #2 represents the heterogeneity in a single formation.

The process of injecting a diverting gel system into the HPCP1 to generate in situ sc-CO2 emulsion in experiment #1 was performed based on the design of physical model 1. In experiment #1, we injected 0.4 PV of a slug of a DGS into the HPCP1 at an injection rate of 0.5 cc/min when the LPCP2 composite was isolated using physical model 1, as shown in Figure . During the injection of the diverting gel system, the differential pressure across the HPCP1 was monitored. The maximum injection pressure recorded was more than 200 psi for injecting the diverting gel system at reservoir conditions. The differential pressure increase across HPCP1 indicates that a slug of the DGS was injected successfully into HPCP1, and then in situ sc-CO2 emulsion was generated at the presence of sc-CO2 in HPCP1. The main reason for the successful injection of DGS into the core is that the high permeability of the HPCP1, about a 1D core, makes the DGS easier to inject into the rock. Another implication for the successful injection of the DGS into the high permeable zone is that we found that the cumulative output of CO2 increases with DGS injection pore volume. The profile of differential pressure across the HPCP1 is shown in Figure . After 0.4 PV of a slug of the DGS was reached and more than 200 psi differential pressure was built up in HPCP1, second sc-CO2 was injected again into both the composites, HPCP1 and LPCP2, to determine the oil recovery performance. Oil production only from the LPCP2 was observed. Figure shows the oil recovery during the second sc-CO2 flooding cycle, indicating that 19% in OOIC of extra oil was recovered at the end of the second sc-CO2 injection after about 1 PV of the second sc-CO2 injection. The higher pressure observed for the second sc-CO2 injection cycle is caused by the plugging of the HPCP1 with the 0.4 PV slug of the DGS. Such diversion causes the subsequent sc-CO2 to go through the LPCP2 and produce a miscible displacement there. To further confirm that the DGS was injected into the HPCP1, we disintegrated the composite core (HPCP1). We did not find DGS deposition and aggregation at the injection end, thus eliminating the pressure rise caused by the blockage at the injection end.

Figure 5

Profile of differential pressure during injection of the diverting gel system into HPCP1 in experiment #1.

Figure 6

Oil recovery by the second sc-CO2 flooding for the LPCP2 composite after DGS injection at reservoir conditions.

After seawater and initial sc-CO2 flooding, the remaining oil saturations were 45.8% in OOIC for LPCP4 and 5.2% in OOIC for HPCP3, respectively. More than 50% of the remaining oil was left in LPCP4 and HPCP3 composites. For such remaining oil in both composites, a DGS injection strategy can be performed, such as physical model 2, as shown in Figure to conduct the oil recovery experiment. Then, 0.2 PV a slug of the DGS was injected into both HPCP3 and LPCP4 to contact sc-CO2 at an injection rate of 0.5 cc/min. During the DGS injection, we observed the change of differential pressure across both HPCP3 and LPCP4 with the increasing pore volume injected by DGS injection, as shown in Figure . The pressure buildup for LPCP4 was higher than that of HPCP3 at the beginning of DGS injection. The low permeability of LPCP4 is the main reason for this phenomenon. The values of brine permeability of HPCP3 and LPCP4 are listed in Table . With the increasing PV of DGS injection into the HPCP3 and LPCP4, differential pressure in HPCP3 increased and decreased in LPCL4, which indicates that a small amount of the DGS was injected into LPCP4, and a large amount of DGS was injected into HPCP3. Therefore, the in situ sc-CO2 emulsions were formed gradually in the HPCP3. After injecting about 0.17 PV of DGS, differential pressure in HPCP3 was reduced, and the differential in LPCP4 has increased again at the same time, as shown in Figure , which could be an indication of the in situ sc-CO2 emulsion breakdown in the HPCP3. On the contrary, we observed another phenomenon that differential pressure in the LPCP4 increased gradually from 0.4 to 2.1 psi, which may generate in situ sc-CO2 emulsion in LPCP4 and benefit second sc-CO2 miscible flooding. To further confirm that the DGS was injected into the HPCP1, we disintegrated the HPCP3 and LPCP4. We did not find DGS deposition and aggregation at the injection end, thus eliminating the pressure rise caused by the blockage during the injection.

Figure 7

Profile of injection pressure building up for HPCP3 and LPCP4 during a diverting gel system injection in experiment #2.

Figure 8

Oil recovery by the second sc-CO2 flooding for HPCP3 and LPCP4 after DGS injection at reservoir conditions.

As shown in Figure , the oil recovery from the HPCP3 was 44.9% in OOIC, and 7.8% in OOIC was recovered from the LPCP4. The low oil recovery in the LPCP4 is due to the high value of permeability contrast and sc-CO2 bypassing the LPCP4 through the higher permeable zone (HPCP3). To recover the remaining oil in the LPCP4, we conducted second sc-CO2 flooding at 0.2 cc/min by following DGS injection. The cumulative oil recovery from the second sc-CO2 flooding was 21% in OOIC from the LPCP4, which is an additional incremental oil recovery beyond the seawater and initial sc-CO2 flooding. The oil was recovered during the first 0.4 PV of injected sc-CO2. The improvement of oil recovery for LPCP4 is attributed to the successful formation of in situ sc-CO2 emulsion. Figure shows the oil recovery profile versus the sum of PV injected by the second sc-CO2 for HPCP3 and LPCP4 composites. Compared to the oil recovery by second sc-CO2 injection after injecting the base–gel system between two physical models, we found no significant difference between physical models 1 and 2, as shown in Table . However, about 5% in OOIC of oil recovery was recovered from physical model 1 based on the remaining oil saturation after initial sc-CO2 injection compared with physical model 2 (see Table ). From a field application and economic point of view, physical model 2 is more applicable because 0.2 PV of a small slug was injected into both models to recover 46% of remaining oil during the second sc-CO2 injection, and it is easier for the operation of field application.

Conclusions

Based on results and observations from core flooding experiments, the following conclusions can be drawn:

The results suggest that the low permeability zone’s poor sweep efficiency is associated with the bypassing of displacing fluid in both seawater and sc-CO2 flooding. Permeability contact has a significant impact on oil recovery by seawater and sc-CO2 injection.

Two proposed physical models with different permeability ratios of heterogeneous reservoir rocks can be used to simulate the conformance control using in situ CO2 emulsion, which was generated by injecting a diverting gel system into the formation in the presence of sc-CO2. The in situ CO2 emulsion was generated successfully in physical models to achieve the conformance control for heterogeneous reservoirs based on the pressure building analysis.

With an increase in the permeability ratio, the final residual oil saturation increased because the larger permeability ratio indicated that the formation is more heterogeneous.

In situ CO2 emulsion demonstrated an excellent blocking and displacing agent to improve the sweep efficiency in heterogeneous reservoirs. The generation of in situ sc-CO2 in the high permeable zone can be beneficial to improve the sweep efficiency and oil recovery in the low permeable zone.

Although more oil of about 5% in OOIC was recovered from LPCP2 in physical model 1 based on the remaining oil saturation after initial sc-CO2 injection compared with physical model 2, from a field application and economic point of view, physical model 2 is more convenient. Because 0.2 PV of a small slug was injected into both cores to recover 46% of the remaining oil during the second sc-CO2 injection, the design of physical model 2 is acceptable for field applications compared to physical model 1 wherein a 0.4 PV slug was injected into the high permeable zone.

Experimental Work

Materials

Two types of brines are used in this work for core flooding experiments. The seawater was used as an injection brine, while the formation water was used as a saturation fluid of the core to achieve the initial water saturation (Swi). The composition of both brines is listed in Table . The properties of different fluids are given in Tables and 6.

Table 4

Composition of Formation Water and Seawater

component	formation water (g/L)	seawater (g/L)
NaCl	150.446	41.041
CaCl2·2HO	69.841	2.384
MgCl2·6H2O	20.396	17.645
Na2SO4	0.518	6.343
NaHCO3	0.487	0.165
total dissolved solids	213,734 ppm	57,670 ppm

Table 5

Properties of Crude Oil and Gas

saturation pressure, psia @ 102 °C	1684
gas oil ratio, SCF/STB	524
stock tank oil gravity °API @ 60 °F	32
average gas gravity (air = 1.0)	1.22
formation volume factor	1.32

Table 6

Fluid Properties at Ambient Temperature and Reservoir Conditions

	ambient temperature at 25 °C		reservoir condition at 102 °C and 3200 psi
fluids	density (g/cc)	viscosity (cP)	density (g/cc)	viscosity (cP)
formation water	1.1462	1.45	1.0906	0.73
seawater	1.0385	0.97	1.0018	0.5
dead oil	0.881	20.51	0.823	2.5
live oil			0.755	0.73
sc-CO2			0.5337	0.04

A dead crude oil from a carbonate reservoir was used to restore the wettability and initial water saturation (Swi) of this study’s core plugs. Separator crude oil and gas were collected from the same reservoir for recombination in the live crude oil sample, which was then used as an oil phase for the water flooding and the sc-CO2 miscible flooding experiment. The saturation pressure, gas/oil ratio, and formation volume factor are listed in Table . The viscosity and density of the dead and live crude oils at reservoir temperature are also listed in Table . The molecular weight of the recombined live crude oil in this study was 121.

sc-CO2 was used as a displacing fluid in the secondary and tertiary stages. The minimum miscibility pressure (MMP) was 2600 psi for a specific system consisting of crude oil and sc-CO2. The MMP is a key factor in designing sc-CO2 injection and measured experimentally using the slim tube apparatus at reservoir conditions. The value of the MMP depends on oil composition and reservoir temperature and pressure. Al Otaibi et al. have reported laboratory studies on the MMP in detail.[50]

This work’s diverting gel system consisted of a gelling agent (polysaccharide-based gel) and a foaming agent (surfactant). A local service company provided these chemicals. The gelling system consists of an 8.9 gpt polymer and a 7.0 gpt surfactant. This gelling system was injected into the core plug with residual oil after initial sc-CO2 injection to generate the emulsion with sc-CO2 in the core plug. Al Otaibi et al. described the commercial CO2 base diverting gel system formula and reported the characteristics of such a system at reservoir conditions at a temperature of 102 °C and pore pressure of 3200 psi.[49]

Core Plugs and Initial Water Saturation Setup

Carbonate Core Plugs

Eight carbonate core plugs, core IDs A, B, C, D, E, F, G, and H, were used for core flooding experiments. The core plugs were cleaned by extracting with methanol/toluene and trying them at a temperature of 104 °C until the dry weight was constant before we used these core plugs to process any tests. Routine core analysis was first conducted to measure the dimensions, air permeability, porosity, and helium PV of the core plugs. Four composite cores #1, 2, 3, and 4 were used in this study. Composite #1 consisted of core plugs A and B, which was considered a high permeable core plug (HPCP1), and composite #2 consisted of core plugs C and D, which was considered a low permeable core plug (LPCP2). Composite #1 and composite #2 (HPCP1 and LPCP2) were used for conformance pattern #1 in experiment #1. For experiment #2 (conformance pattern #2), we used another combination of composite #3 and #4 to run the core flooding experiment. Composite #3 included core plugs E and F as a high permeable core plug (HPCP3), and both core plugs G and H constituted composite #4, which is referred to as a low permeable core plug (LPCP4). The routine core analysis data and composite details are given in Table .

Initial Water Saturation Setup

The core plugs were vacuumed at least 4 h or 1 day, which depends on the permeability of the rock, and then saturated with formation brine. The ionic equilibrium was achieved by immersing the saturated core plugs in the formation brine for 10 days of the aging period. The pore volume and porosity of the core plug was determined by weight change before and after saturation of the core. The brine permeability (Kb) was determined by injecting fresh formation brine using the core flooding apparatus, as shown in Table . After aging core plugs with formation water and determining rock permeability to formation water, the physical parameters of the composite cores stacked together based on the core plugs are listed in Table . We wrapped two core plugs together using a layer of Teflon tape and aluminum foil and put them into a Teflon heat shrink tube. Then, we heated the shrink tube using a heating gun to assemble a composite core.

The formation water in the composite core was then displaced by dead crude oil at a variable injection flow rate of 0.1, 0.2, 0.4, 0.8, 1.0, and 2.0 cc/min. At each flow rate during dead crude oil flooding, the amount of water produced and the differential pressure across the composite core were recorded, continuing until no more water was produced. During oil flooding, the direction of oil flow was reversed to alleviate possible end effects. At this stage, the initial water saturation (Swi) and original oil saturation (Soi) were calculated by material balance and are listed in Table , respectively.

Aging Core with Crude Oil

After determining initial water saturation, Swi, and original oil in the core, Soi, with dead crude oil flooding, the live crude oil was injected at reservoir conditions at a temperature of 102 °C and pore pressure of 3200 psi. The aging core with live crude oil was achieved by a continuous injection (1 cc/min) for one pore volume (PV) per day for three weeks. After the composite core was aged with dead and live oil, it was expected to be weak oil–wet or mixed wet behavior for such carbonate rocks.[51,52]

Core Flooding Experiment

Dual-Core Flooding Apparatus

Figure shows a schematic of the core flooding apparatus. The tests can be run at overburden pressures up to ∼10,000 psi, pore pressures up to 9500 psi, and temperatures up to 150 °C. The system is designed to be extremely versatile. All pore fluid-wetted parts are constructed from corrosion-resistant materials, including Hastelloy C-276, Viton, and Teflon, except for the pressure transducers constructed from stainless steel. Oil, brines, sc-CO2, and a diverting gel system are delivered by using external QX pumps wherein the Quizix Q5000-5 K pump has a minimum flow rate of 0.00007 cc/min and a maximum flow rate of 30 cc/min. Fluid injection is accomplished through a metering pump connected by a valve placed ahead of the core. The pore pressure of the core plugs is maintained by two back pressure regulators (BPRs) and controlled through a pressurized N2 accumulator or pump. The upstream (inlet), downstream (outlet), and differential pressures for each composite are recorded separately. The temperature, flow rate, and other parameters during the dual-core flooding test are measured and recorded through an elaborate data acquisition system. Graduated glass tubes are used to measure the produced oil individually from the HPCP and LPCP composites during seawater and initial and post sc-CO2 miscible flooding.

Figure 9

Schematic for the dual-core core flooding setup at reservoir conditions.

Experimental Procedure

Two core flooding experiments with different injection patterns were conducted for the conformance control at reservoir conditions using formation brine, seawater, living crude oil, and sc-CO2. The pore pressure was set up for 3200 psi using BPRs. The temperature and confining pressure in the experiment were set up at 102 °C and 4500 psi, respectively. We kept the same procedures for seawater, initial sc-CO2, and second sc-CO2 injection to compare conformance control effects.

We placed the dual-core core holders horizontally to load the different permeable carbonate composite cores. The core holder with the high permeable core plugs (HPCP1) was above the other core holder with the low permeable core plug (LPCP2). The brine permeability ratio of HPCP1 to LPCP2 was 43 to 1 for the first physical model of the conformance control pattern #1, which was used to isolate the low permeable zone and open the high permeable zone when a diverting system injection was injected into the core plug in the presence of sc-CO2. For the second physical model of the conformance control pattern #2, which was used to open both the low and high permeable zones when a diverting gel system injection was injected into the core plugs with sc-CO2 flooding, the ratio of permeability between HPCP3 and LPCP4 was 60 to 1, which is based on brine permeability.

Seawater Injection

The purpose of this step is to obtain the remaining oil saturation in the cores by seawater water flooding at reservoir conditions at a pore pressure of 3200 psi and a temperature of 102 °C. After aging with the crude oil, the seawater was injected into both high and low permeable core plugs at three different flow rates (0.5, 1.0, and 2.0 cc/min) until a 99% water cut was achieved during dual-core core flooding experiments. The recovered oil was plotted as a function of PVs of injected seawater. The oil production was measured at ambient conditions using graduated centrifuge tubes with 0.1 cc graduation and then corrected that to reservoir conditions by multiplying with the formation volume factor 1.32. The upstream, downstream, and differential pressures were also recorded across both composites.

sc-CO2 Miscible Injection

After the seawater flooding, there are remaining oil and water in the core. sc-CO2 was simultaneously injected into both composites, high and low permeable core plugs, at 0.2 cc/min at reservoir conditions for experiments #1 and #2 to recover the remaining oil after seawater flooding. The oil production was measured at ambient conditions using graduated centrifuge tubes with 0.1 cc graduation and then corrected that to reservoir conditions by multiplying with the formation volume factor 1.32. The upstream, downstream, and differential pressures were also recorded across both composites during sc-CO2 miscible flooding, respectively.

Diverting Gel System Injection

The effect of different conformance patterns on oil recovery was investigated by injecting diverting gel systems at reservoir conditions. For experiment #1 (physical model 1), including HPCP1 and LPCP2, a slug, 0.4 PV of the diverting gel system (DGS), was injected into the HPCP1 at 0.5 cc/min after initial sc-CO2 injection when LPCP2 was isolated. For experiment #2, a sample of 0.2 PV of the DGS was injected into both HPCP3 and LPCP4 simultaneously at an injection rate of 0.5 cc/min to simulate the reservoir’s application. During DGS injection, the upstream, downstream, and differential pressures were also recorded across both composites using differential pressure transducers, respectively.

Second (Post) sc-CO2 Miscible Injection

After the DGS was injected into composite #1 for experiment #1 and composites #3 and #4 for experiment #2, the sc-CO2 was simultaneously injected into both composites, composites #1 and #2 for experiment #1 and composites #3 and #4 for experiment #2, at a rate of 0.5 cc/min to displace the residual oil in the core left behind the first sc-CO2 flooding. The oil production from the core was measured at ambient conditions using graduated centrifuge tubes with 0.1 cc graduation and then corrected that to reservoir conditions by multiplying with the formation volume factor 1.32. The upstream, downstream, and differential pressures were also recorded across both composites for each experiment during sc-CO2 miscible flooding, respectively.

It should be noted that the difference between experiments #1 and #2 is the injection mode of the DGS. Such differences are shown in Figures and 2, respectively. The DGS was injected into the HPCP1 when the LPCP2 was closed, as shown in Figure for experiment #1. The DGS was injected into both the HPCP3 and the LPCP4, as shown in Figure for experiment #2.