Experimental Study on Conformance Control Using Acidic Nanoparticles in a Heterogeneous Reservoir by Flue Gas Flooding

Abstract

Flue gas flooding has been applied in many oilfields for its accessibility and low cost. However, the problem of gas channeling during flue gas flooding is significantly more serious due to reservoir heterogeneity and gravity override, and the traditional profile control agent is inapplicable because of flue gas acidity. In order to solve this challenge, a novel acidic nanoparticle was presented first; then, the profile control performance of both water slugs and this novel nanoparticle for flue gas flooding in heterogeneous reservoirs was studied using core samples with different rhythms. The results show that the stability of the acidic nanoparticles is good, and the viscosity of the nanoparticle solution increases as the pH decreases, which is suitable for acidic flue gas flooding. The oil recovery of flue gas flooding in a positive rhythm core is 5–10% greater than that in a reverse rhythm core. The water slug can improve oil recovery by 5% in the reverse rhythm core, and oil recovery was less than 2% in the positive rhythm core. The effect of a nanoparticle slug is much better than the water slug. It improved the oil recovery by 10% in the positive rhythm core by continuing flue gas flooding after nanoparticle slug treatment, which was more than the 20% in the reverse rhythm core. The ultimate oil recovery of both positive and reverse-rhythm cores by acidic nanoparticle slug treatment was around 50%, which was 10% greater than the water slug treatment. The conformance control using acidic nanoparticles is more suitable for reverse rhythm formation due to its plugging capacity, deformation characteristic, and viscosity increment in an acidic environment. This research demonstrated that these novel acidic nanoparticles could be effectively applied to conformance control during flue gas flooding in heterogeneous reservoirs.

1. Introduction

Gas flooding is an essential method of enhanced oil recovery [1,2,3,4]. Flue gas flooding has drawn much attention due to the requirements of reducing greenhouse gas emissions [5,6]. Flue gas is mainly composed of N₂ and CO₂, with a small amount of CO, H₂S, SO₂, and nitrogen oxide [7,8,9]. Since the cost of CO₂ capture from flue gas is high, it is desirable to directly inject flue gas into the reservoir to enhance oil recovery and store CO₂ [10,11,12]. Flue gas flooding can avoid the injectivity problems of water flooding in low-permeability reservoirs and store greenhouse gases (such as CO₂, CO, and nitrogen oxide) in underground geological formations as well. Flue gas is a kind of non-condensable gas. It is much more difficult for the flue gas to be miscible with crude oil than CO₂, so the displacement efficiency of flue gas is between CO₂ and N₂. Thus, flue gas flooding is used for immiscible flooding and gravity drainage. During the immiscible gas flooding, gravity overrides and gas channeling tend to happen for the low density and high mobility of flue gas and heterogeneity, which leads to decline in the gas sweep efficiency, and adversely affecting the efficiency of gas flooding.

The periodic changes of the sedimentary environment caused by crustal movement, sea level, climate, and sediment change lead to different grain size ratios, resulting in reservoirs with different rhythms [13,14]. This kind of reservoir heterogeneity has an important influence on reservoir development [15,16]. The effect of different fluid injection qualities on the development of reservoirs with different rhythms is quite different. The gas or steam overburden reduces the swept volume of gas in the reservoir with reverse rhythms [17,18]. The water invades the lower part of the reservoir with higher permeability under the action of gravity, which affects the efficient development of the reservoir with a positive rhythm [19,20]. The difference in remaining oil distribution caused by different development methods for different rhythmic reservoirs determines the selection of profile control measures.

Therefore, conformance control in reservoirs with different rhythms during flue gas flooding is a key investigation area [21]. Because there is no barrier interface between the dominant channel and the reservoir, the channeling path coexists in different regions. If all of the dominant channels are not plugged, gas will flow around the remaining oil enrichment area and back to the dominant channel, which cannot maximize the remaining oil recovery [22]. Several continuous plugging agents, such as foam [23,24], polyacrylamide gel [25], and isopropylamine, have been proposed in the literature. However, the above conventional plugging agents have the following disadvantages: the plugging agents, without deep transport performance, plug only the middle and front parts of the gas channeling, and their acid resistance is poor. As a new nanometer-sized plugging agent, nanoparticles have the advantages of good suspension, injection, and stability properties. Through elastic deformation and adsorption [26], nanoparticles act in deep reservoirs [27] and are widely used in water plugging, profile control of high water-cut reservoirs, and gas channel plugging [28]. However, the stability of traditional nanoparticles in an acidic environment is unsatisfactory. Moreover, a few studies have been conducted on the flooding characteristics of nanoparticles in different rhythmic reservoirs.

Therefore, we first developed a novel kind of acid-thickening nanoparticle. Then, the properties were evaluated. After that, it was used to investigate the effectiveness of conformance control in heterogeneous reservoirs by flue gas flooding of core samples with different rhythms. The dynamic displacement rules of flue gas flooding treated by water slug and nanoparticle slug was discussed. Finally, the conformance control mechanisms of both water slugs and nanoparticles treated in different rhythm cores were investigated. The results demonstrated the effectiveness of nanoparticles on conformance control and enhanced oil recovery, which may benefit the pilot tests of conformance control in oil fields.

2. Experiments

2.1. Experimental Materials

The core samples were artificially made based on the reservoir permeability. The quartz sand was first cemented into semi-cylinders with different permeabilities, and then they were cemented into a whole core based on the predetermined permeability ratio and rhythm. Each core sample consists of two halves. One half is a low-permeability region, and the other half is a high-permeability region. The two halves arrangement was used to simulate the rhythmic characteristics. Four core samples were used in this experiment, as shown in Figure 1. Two core samples were made with positive rhythm, while the other two were made with reverse rhythm. The dimensions, porosity, and permeability of the core samples are shown in

Table 1. Properties of the core samples used in the experiments.

Number	Diameter, cm	Length, cm	Porosity, %	Soi	Kl, mD	Kh, mD
①	2.51	10.02	20.91	0.6827	9.56	53.82
②		10.21	20.72	0.6857
③		9.74	19.23	0.6989
④		9.91	20.12	0.6566

.

The oil sample used in the experiments is the crude oil from the Xinjiang Oilfield, with a viscosity of 15 mPa·s at 42 °C and atmospheric pressure. The aqueous solution used in the experiment is produced water with a salinity of 16,100 mg/L. As

Table 2. The composition of flue gas used in the experiments.

Components	Composition, mol%
N2	74.97
CO2	15.25
O2	3.38
H2S	0.06
C3H8	0.14
C2H6	0.29
CH4	5.92

shows, it mainly consists of N₂, CO₂, CH₄, and O₂. The minimum miscible pressure (MMP) of flue gas with crude oil is higher than 35 MPa at 42 °C.

In order to make the nanoparticles adapt to the acidic conditions of flue gas, the novel acid-adaptive nanoparticles were developed using silica nanoparticles by modifying the carboxylic group. The particle size of silica nanoparticles is shown in Figure 2. The zeta-potential of the nanoparticle system is about −17.5, measured by a Malvern Panako Zetasizer, as shown in Figure 3. Additionally, the dispersion solution had no obvious settlement after 6 h, and it can well recover good uniform dispersion after settlement, which indicates that the novel system has good dispersion performance and stability.

The nanoparticles were dissolved in brine water, and the viscosity of the nanoparticle system at different concentrations was measured using a Brookfield viscometer. The test result is shown in Figure 4. As the solution concentration gradually increases, the shear viscosity of the solution system also gradually increases.

The nanoparticle is suitable for conformance control during acidic gas flooding. The acidic flue gas can cause the brine formation to be acidic, which could affect the stability of the conventional conformance control agents. This research investigated the application of self-adapting nanoparticles in an acidic environment to enhance the viscosity of the aqueous phase, as nanoparticles can cross-link in the acidic environment, thus increasing the viscosity of the aqueous phase. Figure 5 shows the viscosity of the aqueous solution of nanoparticles in reservoir brine at different pHs with a nanoparticle concentration of 0.5 wt.%. The aqueous solution’s viscosity was measured at room temperature and atmospheric pressure. The viscosity of the aqueous solution increased with a decrease in pH. The value of viscosity doubled as pH was reduced by 1.

2.2. Experimental Design

Figure 6 shows the experimental set-up, which consists of a gas cylinder, a pump, valves, an accumulator, a core holder, a backpressure regulator (BPR), a separator, a measuring cylinder, and a data acquisition system. There were four sets of experiments. To ensure the accuracy of the experiment, each group of experiments was conducted three times. The first two sets of experiments investigated the effect of heterogeneity (different rhythmic characteristics) on flue gas flooding and conformance control by brine. Each core sample had positive and reverse rhythms, which were used in the experiments. The core sample was first flooded with flue gas, and then the brine-slug was injected into the core sample. After that, flue gas was injected into the core sample until there was no oil production. The last two experiments investigated the effect of nanoparticles on conformance control during flue gas flooding in heterogeneous reservoirs. During the experiments, flue gas was injected into the core first, and then an aqueous nanoparticle solution was injected. After that, flue gas was injected until there was no oil production. The last two sets of experiments were compared with the first two sets of experiments to investigate the effect of nanoparticles on conformance control.

2.3. Experimental Procedure

The experimental procedure is as follows:

• Preparation of the core samples: the diameter and length of the core samples were measured, and the bulk volume of the core samples was calculated;
• Measurement of the porosity and permeability: the core samples were saturated with brine at an injection rate of 0.5 mL/min for 90 min, then 1 mL/min for 45 min. The injected brine volume was about ten pore volumes (10 PV) to ensure that the core samples were fully saturated with the brine;
• Oil saturation and measurement of initial oil saturation: crude oil was injected into the core at 0.5 mL/min until the water cut was zero, then the initial oil saturation was calculated. After that, the core was aged for 2–3 days at 42 °C;
• Flue gas flooding: the flue gas flooding was performed in an oven at 42 °C. The overburden pressure was 20 MPa, and the back pressure was 14 MPa. The gas injection rate was 0.5 mL/min. The amount of oil produced and the inlet pressure were measured every 2 min. The flue gas was injected into the core sample until there was no oil production;
• Injection of brine-slug or nanoparticle solution slug: the brine-slug or nanoparticle solution slug was injected into the core sample at 0.5 mL/min for 0.2 PV. The amount of oil produced and the inlet pressure were recorded;
• Flue gas flooding: the flue gas was injected into the core samples at 0.5 mL/min until there was no oil production. The amount of oil produced and the inlet pressure were recorded every 2 min. The oil recovery was then calculated.

3. Results and Discussion

3.1. Effect of Heterogeneity on Flue Gas Flooding and Conformance Control

Figure 7 presents the results of the first two sets of experiments. It can be noted that the oil recovery during the first flue gas flooding for the sample with a positive rhythm was higher than that of the sample with a reverse rhythm by 5%. The reason is that the density of flue gas is lower than that of oil, so it tends to flow into the upper section of the core sample under gravity segregation. Since the permeability of the upper section in the reverse rhythm core is higher than that of the lower section, the gas channeling is more serious than that of the positive rhythm core; in other words, the heterogeneity in the positive rhythm core is favorable for flue gas flooding.

Concerning the injection pressure, in the positive rhythm core, the flue gas can flow in both the lower and upper sections, so the flow area is larger. However, in the reverse rhythm core, the gas flow is mainly in the upper section, and the flow area is smaller. Therefore, the pressure difference is lower in a positive rhythm core. At the continued gas flooding stage, a good water slug was formed under the action of both gravity and heterogeneity in reverse rhythm formation, so the injection pressure is higher and the displacement effect is better.

3.2. The Effect of Acidic Nanoparticles on Conformance Control during Flue Gas Flooding in Heterogeneous Reservoirs

Figure 8 presents the results of the last two sets of experiments that examined conformance control by an acidic nanoparticle slug. The injection pressure increased as the nanoparticles were injected, and the pressure difference in the reverse rhythm core is much higher than in the positive rhythm core. The enhanced oil recovery during the continued flue gas flooding for the core sample with positive and reverse rhythms was 10% and 22%, respectively. The ultimate oil recovery for the two cases was about 50%, which is 10% greater than that of conformance control by water slug.

Figure 9 shows the comparison of the development effects between the water slug and nanoparticle solution slug in both positive and reverse rhythm cores. Water slug injection showed limited conformance control for the sample with a positive rhythm, while nanoparticle solution injection effectively enhanced oil recovery. Water slug injection and nanoparticle solution slug injection showed good conformance control for the sample with a reverse rhythm. However, nanoparticle solution injection showed much better effective conformance control in contrast to water slug.

Figure 10a,b shows the movement of flue gas before and after water slug treatment in the positive rhythm core. When the flue gas was injected into a positive rhythm core, the gas was inclined to flow into both the upper low-permeability layer due to gravity differentiation and the lower high-permeability layer for least resistance. Therefore, the oil in the middle of the core will be displaced. For the reverse rhythm core, the gas was strongly inclined to flow into the upper high-permeability layer, so the gas channeling is serious, as shown in Figure 10c. When the water slug was injected into the positive rhythm core, it tended to plug the lower high-permeability layer for both gravity differentiation and least resistance where the remaining oil was enriched, so there was little effect, as shown in Figure 10b. However, when the water slug was injected into the reverse-rhythm core, it plugged the upper high-permeability layer for least resistance, so the continued gas would flow into the lower low-permeability layer, where the remaining oil was enriched. However, because the plugging capacity is limited, gas will channel back to the upper high-permeability layer, as shown in Figure 10d.

Figure 11a,b shows the movement of flue gas before and after nanoparticle slug treatment in the positive rhythm core. The flue gas flooding process for both positive and reverse rhythm cores was similar to those described above. When the nanoparticle slug was injected into the positive rhythm core, although it also tended to flow into the lower high-permeability layer for both gravity differentiation and least resistance, where the remaining oil was enriched, it had a higher viscosity, and the nanoparticles had good deformability. Therefore, it played the combined role of displacement and plugging at the same time, and part of the remaining oil in the lower high-permeability layer will be displaced out, as shown in Figure 11b. Generally, the plugging ability in the upper channel is limited. However, when the nanoparticle slug is injected into the reverse rhythm core, the nanoparticle will flow into the upper high-permeability layer and plug the channel, so the continued flue gas will displace the lower low-permeability layer, where the remaining oil is enriched. Moreover, the nanoparticles also play the role of displacement in the upper high-permeability layer because of their deformation characteristics and increased viscosity in an acidic environment. Therefore, the remaining oil in both layers will be recovered, as shown in Figure 11d.

4. Conclusions

• A novel acidic nanoparticle with good dispersion and stability performance was developed. The zeta potential of the nanoparticle system is about −17.5, and the viscosity of the nanoparticle solution doubles as the pH decreases by 1;
• The oil recovery of flue gas flooding from a positive rhythm core was 5–10% greater than that of a reverse rhythm core. The effect of water slug treatment on a reverse rhythm core is better than on a positive rhythm core;
• The oil recovery is enhanced by 10% for nanoparticle treatment in positive rhythm cores and by more than 20% in reverse rhythm cores. The ultimate oil recovery is around 50% by nanoparticle treatment of both positive and reverse rhythm cores, and it is 10% higher than that of water slug treatment. Therefore, the nanoparticle is more suitable than the aqueous solution for conformance control during flue gas flooding in heterogeneous reservoirs;
• The effect of flue gas flooding is better in a positive rhythm core under the action of gravity differentiation and least resistance. Water slugs can plug the upper high-permeability layer for least resistance to achieve a better effect for continued flue gas flooding in reverse rhythm cores. Nanoparticle treatment can achieve excellent effects due to its plugging ability, deformation characteristic, and viscosity increment in an acidic environment in reverse rhythm cores.