
==== Front
ACS Omega
ACS Omega
ao
acsodf
ACS Omega
2470-1343
American Chemical Society

10.1021/acsomega.4c04137
Article
CO2 Geostorage and Enhanced Oil Recovery: Challenges in Controlling Foam/Emulsion Generation and Propagation
https://orcid.org/0000-0003-3810-385X
Klimenko Alexandra *†‡
https://orcid.org/0000-0003-2164-7586
Cui Leyu †‡⊥
https://orcid.org/0000-0002-0097-3000
Ding Lei ‡∇
Bourrel Maurice ‡
† Pôle d’Etudes et de Recherches de Lacq, TotalEnergies S.E., BP 47, 64170 Lacq, France
‡ Physico-Chimie des Interfaces Complexes, Laboratoire Commun TotalEnergies/ESPCI, Bâtiment CHEMSTARTUP, RD 817, 64170 Lacq, France
* Email: alexandra.klimenko@totalenergies.com.
20 08 2024
03 09 2024
9 35 3709437104
30 04 2024
08 08 2024
06 08 2024
© 2024 The Authors. Published by American Chemical Society
2024
The Authors
https://creativecommons.org/licenses/by-nc-nd/4.0/ Permits non-commercial access and re-use, provided that author attribution and integrity are maintained; but does not permit creation of adaptations or other derivative works (https://creativecommons.org/licenses/by-nc-nd/4.0/).

CO2 injection in subterranean reservoirs for storage, oil recovery, or both is challenging because of its very high mobility. Using a CO2 foam or emulsion is a way to remedy this problem by increasing CO2’s apparent viscosity. However, the generation of the foam and its propagation in porous media present several issues that have to be overcome for this process to be economically realistic in practice. For example, it may take time, i.e., a number of pore volumes to be injected, before the foam is created. It is the objective of this Article to investigate these issues thoroughly and to identify the mechanisms underlying them by looking at the effects of various parameters. It is found that surfactant adsorption on the surface of the rock is an important factor involved in the delay of foam formation, but this may not explain all of the results. The nature and morphology of the porous medium may be, in some cases, the dominant factors for foam generation and propagation. From an understanding of the origin of the encountered problem, relevant mitigation strategies are envisioned and evaluated. It is found, for example, that when appropriately formulated and injected with the proper process, foam or emulsion generation is strongly accelerated, which very significantly shortens the delay for achieving CO2 storage.

document-id-old-9ao4c04137
document-id-new-14ao4c04137
ccc-price
==== Body
pmc1 Introduction

CO2 injection in subterranean reservoirs was initially developed for enhanced oil recovery (EOR) in depleted oil-bearing reservoirs.1 Since then, with the awareness of climate breakdown, CO2 EOR has appeared as an opportunity to sequestrate CO2 underground while producing oil with a lower CO2 emission footprint.2−6 Besides, it has been recognized that saline aquifers offer also high storage potential,7−10 possibly combined with geothermal heat extraction.11

In all of these situations, CO2 has to displace residing water and, in some cases, some oil. Since CO2 has a low viscosity, displacement is not piston-like, and in certain cases unstable viscous fingering may develop, yielding early breakthrough. Thus, reducing CO2 mobility is highly attractive to stabilize the displacement front and increase the CO2 storage capacity for a given reservoir. The potential advantages of foam for CO2 storage and related challenges are discussed at length in the work by Rossen et al.12

A number of methods have been proposed for that purpose, including using additives in CO2 to increase its viscosity or surfactants to create foams.13 Depending on the relative permeability curves, the presence of oil, and the viscosity of the residing and injected fluids, it may be necessary to drastically reduce the mobility of CO2 to displace the residing fluid. As an example, in experiences without oil or at residual oil saturation on a limestone core (Figure 1) without surfactant, we have estimated that the maximum apparent viscosity that needs to be overcome in the CO2/water displacement front may vary from 0.6 to ∼5 cP. Since in this example the CO2 viscosity is only 0.0273 cP, the mobility of CO2 must be reduced by a factor of 22–183 to achieve piston-like displacement.

Figure 1 Apparent viscosity during brine/CO2 coinjection on Estaillades at 4 ft/d in the absence (pink points) and in the presence (blue points) of the residual oil. Conditions: 110 °C, 150 barg.

Such a reduction is possible if the CO2 is injected in the form of a CO2 foam, which forms between the brine and CO2 thanks to a CO2-philic surfactant. Clearly, the foam must be as CO2-rich as possible while maintaining a high apparent viscosity to ensure the low mobility of the CO2-rich phase. Recently, Li et al.14 showed that, with CO2 foam injection, the gas saturation increased from ∼21% (100% CO2 injection) to ∼84% (85% CO2-rich foam injection) after ∼10 PV of injection, therefore improving the CO2 storage capacity. The best-performing foam also had the highest apparent viscosity, as derived from the measured pressure gradient, underlining the importance of good mobility control for CO2 storage.

In this Article, we first record, from an application standpoint, several issues often encountered in the formation and propagation of the foam in porous media. We then present results obtained on quartz sandpacks of high permeability on the one side and on limestone cores of low permeability on the other side when varying various parameters. Afterward, we discuss the possible mechanisms underlying the observed phenomena. Finally, we propose some mitigation strategies depending on the identified issue.

2 Experimental Conditions

Details are given in the Supporting Information. All experiments have been conducted at 150 barg and 110 °C unless specified. Under these conditions, CO2 is in a supercritical state, and its density is 0.3031 g/cm3, i.e., in between gas and liquid. We thus term the fluid either emulsion, foam, or emulsion/foam. All the indicated CO2 fractions are volumetric and corrected for CO2 solubility in brine (see Supporting Information).

The proprietary15 surfactant R-CADA (reduced cocoalkyl dimethylpropane diamine) was used with a purity higher than 90%.16 The brine salinity is 257.55 g/L total dissolved solids. Relative to other surfactants, this surfactant was shown to be capable of generating strong CO2 foam with supercritical CO2 at high temperatures with high-salinity brine. More details can be found in the work by Cui et al.,16 and the molecular structure is presented in Figure 2.

Figure 2 Molecular structure of R-CADA.

At these conditions, this surfactant is soluble in both CO2 and saline brine (at pH ≤ 8). The measured partition coefficient is 1.82 ± 0.23 mole fraction in CO2/mole fraction in brine (0.21 ± 0.07 g/L CO2 per g/L brine) at 110 °C and 0.37 ± 0.13 mole fraction in CO2 per mole fraction in brine (0.12 ± 0.04 g/L CO2 per g/L brine) at 28 °C in 220 g/L NaCl brine and 150 barg (see the Supporting Information for more details). These values give minimum solubilities of ∼0.16 and 0.62 mg surfactant/g CO2 at 28 and 110 °C, respectively.

The sandpack column was packed with quartz sand, giving 14 D permeability; limestone cores were from Indiana and Estaillades, with permeabilities around 330 and 130 mD, respectively.

To describe the mobility and strength of the CO2 emulsion in different porous media, the measured pressure gradient ΔP is converted into apparent viscosity AV with the following equation based on the single-phase Darcy law:

where k is the absolute permeability, A is the core cross-section, q is the total volumetric injection rate, and L is the core length. All experiments are carried out by co-injecting CO2 and brine containing R-CADA at a constant injection flow rate.

3 Potential Challenges

A typical evolution of the apparent viscosity as a function of injected pore volumes is shown in Figure 3 together with the identification of its main features, which helps to display the requirements for successful foam/emulsion transport in the porous medium.

Figure 3 Definition of different typical characteristics of transient foam/emulsion transport in porous media initially saturated with brine. Example is based on an experiment on quartz sandpack, performed at 150 barg, 110 °C, and 0.95 CO2 volume fraction (i.e., 95%); 0.2 wt % surfactant R-CADA in injected brine.

When co-injecting CO2 and brine, with the surfactant injected into one of the phases (in this study, in the brine), it is common to observe a delay in emulsion/foam generation and development, the so-called minimum pore volume (MPV), which needs to be minimized to rapidly ensure the required mobility control.

At the minimum pore volume, the pressure gradient rises gradually until it sudden increases, indicating the formation of a strong foam/emulsion.17 The MPV is generally attributed to the time required for the pressure to reach a minimum pressure gradient (MPG) that must be exceeded to start the formation of a strong foam/emulsion. Among the other parameters listed below, MPV is a strong function of surfactant adsorption.

MPG increases with the foam quality, i.e., the CO2 volumetric fraction,17−19 with decreasing surfactant concentration,19 with increasing surface tension,17,18 and with decreasing pore throat radius.18 Due to the latter, the relationship of MPG with permeability k is complex for consolidated porous media but simpler for unconsolidated media where MPG scales with 1/k.17 As the interfacial tension with CO2 is lower than that with N2, it has been shown20 that MPG is much lower for CO2 emulsion/foam (<2 psi/ft) than for nitrogen foam.

Once the foam/emulsion starts to form, it propagates through the porous medium, which takes a certain amount of time, defined here as the pore volume required for emulsion propagation (PVP). In constant-velocity experiments, when conditions are favorable for stable strong foam, this time is influenced by a number of parameters: the surfactant concentration,21 which is subjected to adsorption and CO2/brine partitioning; the dispersivity of the porous media; the injected foam quality; the core orientation;22 the nature of the porous media; etc. During this time, the apparent viscosity continues to increase until it reaches a steady-state plateau.

The sum of MPV and PVP is the number of pore volumes required for the foam/emulsion to reach equilibrium (i.e., the plateau) and is defined here as EPV. The apparent viscosity at the plateau (PAV) is the steady-state emulsion/foam force for a given injected CO2 volume fraction (FCO2). It is commonly observed that PAV exhibits a maximum upon variation of FCO2.

Besides, an important feature of foam/emulsion transport is the possible arrival of the foam/emulsion before EPV, which implies that the foam texture continues to evolve in the porous medium. This is the sign that foam forms at the outlet due to the capillary end effect and that, once formed, it generates backward against flow toward the inlet, resulting in an increase in apparent viscosity.23−26

Therefore, for successful injection, MPV, PVP, and MPG must be minimized while ensuring a high apparent viscosity PAV at the highest possible FCO2. The CO2 volume fraction at the highest apparent viscosity is defined here as the transition quality FCO2*. Finally, the effect of the presence of oil in the porous medium has to be taken into account.

4 Results and Discussion

The investigation of the effect of various parameters on the features of the curve of apparent viscosity vs injected pore volumes (Figure 3) has been carried out on quartz sandpack (see the Supporting Information). The effect of the type and morphology of the porous medium is presented later.

4.1 Results

The apparent viscosity behavior at different CO2 volume fractions FCO2 at 28 and 110 °C is reported in Figure 4.

Figure 4 CO2 emulsion generation at different CO2 volume fractions coinjected at a Darcy velocity of 60 ft/d and 0.2 wt % R-CADA in brine in quartz sandpack (∼14 D, 33% porosity). The sandpack is saturated initially with brine. Violet points define the emulsion breakthrough time.

The steady-state apparent viscosity PAV increases with FCO2 and is the highest at a foam quality of about 95%, thus satisfying one of the criteria for mobility control during CO2 injection; additionally, it is not sensitive to temperature with the surfactant at hand, as shown in Figure 5.

Figure 5 Steady-state apparent viscosity of the CO2 emulsion generated in quartz sandpack (∼14 D) at 60 ft/d and 0.2 wt % R-CADA in brine, 150 barg, and 110 °C.

The effect of the CO2 volume fraction on MPV, PVP, and MPG is reported in Figure 6.

Figure 6 Characteristics of transient emulsion behavior during CO2 emulsion injection in quartz sandpack (∼14 D) at 60 ft/d and 0.2 wt % R-CADA in brine, 150 barg, and 110 °C.

As shown in Figure 6 (left side) and Figure 5, the steady-state emulsion strength increases with FCO2 but is accompanied by an increase in MPV and PVP: more than 25 PV of injection is required to reach the steady-state condition at FCO2 95%.

Interestingly, MPG appears to be almost constant (1–2 psi/ft) with FCO2 (Figure 6, right side), in contrast to the studies from Gauglitz at al.17 and Yu et al.,19 which predicted and showed an increase in MPG with FCO2. It is very likely that the interstitial gas velocity of 26–55 m/day in our experiments on sandpack was higher than the minimum velocity required for foam generation.18 Therefore, the reported MPG is probably apparent, meaning that the observed delay (MPV) is related to other reasons and that the actual MPG is <1–2 psi/ft

As mentioned above, if oil is present in a target reservoir, even a small amount may be enough to increase the EPV and deteriorate the emulsion/foam strength: on quartz sandpacks (Figure 7, left side), the presence of residual oil considerably delayed CO2 emulsion generation and, at 28 °C, decreased emulsion strength. This was confirmed by other experiments conducted on Estaillades limestone (Figure 7, right side), where 6% of oil was sufficient to hinder emulsion generation even after ∼25 PV of emulsion injection. The negative impact of oil on CO2 foam/emulsion generation and propagation is known under both immiscible and miscible conditions.27−32 Therefore, the impact of oil on emulsion/foam generation, propagation, and stability must also be considered if the CO2 emulsion/foam is to be injected into an oil-bearing reservoir.

Figure 7 Oil impact during CO2 emulsion injection (0.2 wt % R-CADA in the injected brine) under immiscible conditions. On the left: quartz sandpack, CO2 fraction 48%, Darcy velocity 60 ft/d, 28 and 110 °C, 150 barg. On the right: quality scan on Estaillades limestone, 110 °C, 150 barg, 2 ft/d. Violet points define the emulsion breakthrough time.

In the following section, possible reasons for a delay in foam/emulsion generation and propagation are examined. Only oil-free conditions are discussed.

4.2 Discussion

The reasons for the delay of strong emulsion/foam generation and propagation must be elucidated in order to develop a strategy for accelerating foam/emulsion propagation for effective use of CO2 injection. Hereafter, various possible reasons are examined. Only the strong foam/emulsion case of interest is discussed, as the propagation and generation of weak foams are out of the scope of this work.

4.2.1 Surfactant Adsorption

A necessary condition for foam and emulsion propagation is surfactant propagation, which can be compromised by adsorption.

The adsorption of a surfactant on a mineral is largely dependent on temperature, wettability, surface charge, pH, brine compositions, and salinity.33,34 Presumably, the adsorption of the cationic surfactant (as amine surfactants at low pH) on carbonate minerals should be low at low pH since they are both positively charged.35,36

Static adsorption tests have generally been used to measure the adsorption of switchable amine surfactants on carbonate, and the pH was adjusted by pressurizing under 1–2 barg of CO2.37−39 However, the CO2 solubility in brine at 2 barg is significantly lower than that at target high pressure (150 barg in the examples of this study). The brine composition and the extent of ion exchange may be largely different at different CO2 pressures, which can impact the mineral surface charge.40 More generally, the reliability of the static adsorption test is questionable since, first, normalization by the surface area must be performed to convert the data into the usual units of mg/g of rock and, second, the rock must be ground prior to the test, which can change the nature and even the charge of the surface41 exposed to the surfactant solution.

Ideally, surfactant adsorption is more relevantly determined from dynamic measurements, i.e., by measuring the delay between the production profiles of the surfactant and a nonadsorbing tracer in a flooding experiment in a core of the porous medium under investigation. In the case of CO2 injection at high pressure, as stated above, care must be taken with regard to the CO2 solubility in water, and measurements at atmospheric pressure might not be relevant. A method has been developed in our laboratory to overcome this difficulty (patent pending,42 see the Supporting Information). Applied on Estaillades limestone, it yields an adsorption of 0.19 ± 0.02 mg/g of rock or 0.42 ± 0.04 mg/m2 of rock surface, the specific area having been determined by BET measurements.

However, this method is inconclusive in the case of the sandpack owing to its small pore volume and solid surface area. In the experiments on the sandpack, since the pressure gradient is assumed to be sufficient to overcome the MPG, the most obvious reason for the observed delay for foam/emulsion generation remains surfactant retention, likely essentially adsorption. This can be evaluated by calculating the mass of surfactant consumed before emulsion breakthrough, which corresponds to EPV, as follows:

where PVbt is the number of pore volumes at which the emulsion arrives, PV is the pore volume and Cs is the surfactant concentration.

At 110 °C, the injected mass before steady state is established as almost constant (5.9 ± 0.8 mg, see Figure 8). At 28 °C and higher CO2 fractions, the mass is also constant (4.3 ± 0.5 mg); the results at FCO2 48% and 69% are close to zero, and the experiments are worth repeating. In terms of adsorption, these values provide 0.12 ± 0.01 and 0.17 ± 0.02 mg/g of rock at 28 and 110 °C, respectively. The increase in estimated adsorption with temperature can probably be related to weaker interactions of the surfactant with brine at higher temperature.

Figure 8 Injected surfactant mass “consumed” before emulsion arrival in the experiments on quartz sandpack.

The measured specific surface of the sand is 0.22 m2/g, yielding adsorption values of 0.56 and 0.78 mg/m2 at 28 and 110 °C, correspondingly. These values are higher than that measured on Estaillades limestone, which can be related to the difference in surface charges between quartz and calcite. A comparison with the adsorption of surfactants of comparable structure reported in the literature is provided in Table S3 of the Supporting Information.

To further establish the importance of surfactant adsorption in the transport process, an experiment was carried out on the same sandpack without restoration, i.e., co-injecting CO2/brine with stepwise increasing fractions of CO2. If adsorption is one cause of the delay, emulsion generation and propagation are expected to be rapid once adsorption is satisfied, since, as discussed before, the interstitial CO2 velocity is sufficiently high to provide a pressure gradient higher than MPG. Indeed, as shown in Figure 9, less than 1 PV is required to generate the new emulsion, and only a few PV are required to reach steady-state conditions. Additionally, quality scans performed with and without restoration are in excellent accord (see Figure S4 in the Supporting Information).

Figure 9 Flooding history of the coinjection of CO2 and R-CADA solution at 0.2 wt % in brine in quartz sandpack at a superficial velocity of 60 ft/d, 110 °C, and 150 bar. % indicates the CO2 volume fraction. The violet point defines the emulsion breakthrough time.

Surfactant partitioning into CO2, which significantly depletes the brine, especially at high CO2 fractions, is a phenomenon that also possibly affects its transport in the porous medium. Interestingly, in our experiments, it does not lead to a decrease in emulsion strength at steady state, nor does it prevent emulsion/foam generation. At FCO2 95%, the estimated concentration in brine, obtained from the measured the partitioning coefficient KCO2/w = 1.82 ± 0.23 mole fraction in CO2 per mole fraction in brine at 110 °C, is 0.042%, close to the critical micellar concentration (CMC) determined from surface tension measurements, i.e., 0.046 wt %.43 Such a low concentration should be far below the critical surfactant concentration where the emulsion/foam strength decreases with decreasing concentration. In addition, Mannhardt and Svorstøl44 showed that the lowest surfactant concentration at which foam is generated overlaps with the CMC region. Therefore, it may be hypothesized that the surfactant present in both phases plays a role in foam/emulsion flow, not only in the aqueous or CO2 phase.

4.2.2 Effect of the Nature of the Porous Media

Besides the propagation of the surfactant, the foam/emulsion itself has to be generated and propagated.

Generation, through MPG, is inversely related to the pore throat radius and the distance between two moving lamellae.18,17 Therefore, the morphology of the porous media is a paramount factor for foam generation.

Its impact on propagation, i.e., on PVP, is less clear. In different population-balance models, which try to reproduce foam generation and the transient regime of foam propagation, lamella creation and destruction rates as well as foam texture are tuned to provide gas resistance to the flow. Depending on the model, rates of generation and coalescence are the function of foam texture, gas velocity, surfactant concentration, capillary pressure, pore geometry, and matching coefficients.45,46 Dependence on the gas velocity reflects the idea that limiting capillary pressure decreases with increasing gas velocity,47 whereas in a recent experimental study the opposite trend was shown.48 In any case, capillary pressure and limiting capillary pressure are related to the two characteristics of the porous media, i.e., wettability and pore throat diameter, which are expected to impact the transient behavior, i.e., the PVP value.

The effect of the rock type was investigated on limestone cores from two different origins: Estaillades and Indiana. To avoid being hampered by surfactant adsorption, quality scans in the coreflood experiments have been carried out without any restoration in between two successive injections, similar to Figure 9.

The results at steady-state conditions obtained on Estaillades (Figure 10, left side) and Indiana limestones (Figure 10, right side) are very different: FCO2* is near ∼20% and ∼90%, respectively, while in the quartz sandpack it is ≥95% (Figure 5). The low FCO2* value observed for Estaillades has already been mentioned by Ding et al.,49 who reported FCO2* < 50% with nitrogen foam in the presence of oil. As yet, no explanation has been found for this unusual behavior, but it is probably related to the complex porous structure of this limestone.50

Figure 10 Quality scan and MPG in Estaillades (on the right) and Indiana (on the left) limestones. Conditions: 110 °C, 150 barg, 0.2 wt % R-CADA in brine. MPG on both limestones is measured at 4 ft/d.

Both limestones have similar pore throat diameter distributions51 and thus the notable difference in quality scans is surprising. This possibly underlines the importance of other petrophysical characteristics: Indiana limestone is more permeable and less porous than Estaillades.

Emulsion generation and propagation were studied for two FCO2 values after a preflush with a surfactant solution to satisfy the adsorption at 15% and 48% on Estaillades and at 15% and 90% on Indiana limestone. The results are presented in Figure 11, and MPG is reported in Figure 10.

Figure 11 CO2 emulsion generation after preflushing with a R-CADA solution at different CO2 volume fractions on Estaillades (on the right) and Indiana limestone (on the left) at a Darcy velocity of 4 ft/d and 0.2 wt % R-CADA in brine. Violet points define the emulsion breakthrough time. Magenta points indicate CO2 arrival.

On Indiana limestone, MPVs are nonzero and close for both FCO2 (∼0.2 PV); as for MPG, it increases slightly with FCO2 from 0.7 to 0.9 psi/ft (Figure 10, left). Propagation is much faster for a stronger emulsion, i.e., for a higher CO2 fraction.

On Estaillades, the behavior is very different: despite satisfied adsorption, MPV is more important at a higher CO2 fraction (∼0.8 PV versus almost zero at FCO2 15%) and accompanied by a notable MPG increase from 0.4 to ∼7.5 psi/ft (Figure 10, right). The experiment at the low CO2 fraction was interrupted before steady state was reached, but this continuous generation and propagation appears to be faster for a stronger emulsion as in Indiana limestone but at low CO2 fraction. More importantly, emulsion breakthrough appears near 1 PV for both fractions tested and pressure continues to increase, indicating bubble refinement and emulsion texture evolution in the porous media, probably due to the presence of end effect on Estaillades: a strong emulsion is generated near the outlet and generates backward, progressively filling the core, as previously reported by Apaydin and Kovscek in 2001,24 Nguyen at al. in 2003,25 and Almajid et al. in 2019.26 According to the studies cited, the weak, coarse emulsion first broke through, and then the emulsion strengthened as the front receded.

In light of this, the difference in MPG (Figure 10) between the two limestones may be related to different emulsion generation conditions.

In Indiana limestone, due to the low MPG, generation most likely occurs in the inlet, since the emulsion arrival coincides with the EPV. The observed MPV is therefore related to the time required to reach the MPG.

Generation on Estaillades appears to occur at the core outlet due to the end effect, meaning that reported MPGs are apparent since the pressure gradient in gas phase would be higher at the outlet due to higher capillary pressure gradient and water saturation gradient.26 Therefore, the actual MPGs for strong emulsion generation should be even higher. At low CO2 fractions, the apparent MPG is lower as well as the actual MPG because the end effect and capillary pressure gradient are less pronounced for a wetter foam, as experimentally confirmed by Almajid et al. in 2019.26

Analysis of the atypical quality scan with low FCO2* and the difficulties in generating strong emulsions at high CO2 fractions despite the satisfied adsorption observed on Estaillades underline the important impact of the porous media on CO2 emulsion behavior. It is therefore essential to perform experiments on a reservoir rock as soon as possible, since the choice of an analogue is very often based on similar porosity, permeability and, in the best case, pore throat distribution. However, this analogy may prove insufficient.

Indeed, particular attention must be paid to laboratory artifacts such as foam/emulsion generation due to the end effect. It must be pointed out, however, that this phenomenon can possibly also be produced in reservoirs by heterogeneities yielding discontinuities of capillary forces, as observed by Almajid et al. in 2019,26 resulting in that case in a positive effect. However, taking this effect into account when designing the implementation of a CO2 emulsion injection is challenging.

4.2.3 Surfactant Concentration Effect

For a given porous medium, there exists a surfactant concentration, so-called “critical”, above which the emulsion/foam strength no longer depends on the concentration. Knowledge of this critical surfactant concentration is important for economic considerations.

It is actually different for the two carbonates studied in this work: for Estaillades it is ∼0.2 wt %, while for Indiana limestone it is ∼0.1 wt % R-CADA (Figure 12). These values are higher than the critical micelle concentration (CMC) determined for this surfactant (0.046 wt % by Chen et al. in 202343), in agreement with Mannhardt and Svorstøl44 conclusions according to which concentrations higher than the CMC are necessary to generate and propagate the foam. It should be noted that Mannhardt and Svorstøl also showed that, once formed and propagated, the foam could be maintained even at sub-CMC concentrations, which can help minimize surfactant requirements during field application.

Figure 12 Quality scan in Estaillades (on the right) and Indiana limestone (on the left) for R-CADA concentrations in brine.

Besides, it is interesting to note that the transition emulsion quality FCO2* decreases when the surfactant concentration is reduced from 0.1 to 0.01 wt % on Indiana limestone, going from 90% to 58%. This is related probably to less resistant water films at low surfactant concentration causing film breakage, eventual bubble coalescence, and therefore a coarser foam/emulsion. This phenomenon is consistent with those reported in literature52 and implies that an injection at high CO2 fraction may be a compromise when performed at a surfactant concentration lower than the critical one.

The nature and morphology of the porous medium thus play a major role not only in the steady-state foam/emulsion force but also in foam/emulsion generation and propagation. While surfactant retention can be more or less easily overcome to improve foam/emulsion generation and propagation, the impact of porous media is more difficult to handle. Several mitigation strategies are proposed depending on the main cause of the issue.

4.2.4 Mitigation Strategies

Indeed, for CO2 emulsion injection for EOR or CO2 storage applications, the higher the CO2 fraction is, the better. It is required that at this fraction the emulsion/foam has a high apparent viscosity to efficiently displace water and/or oil. As was shown, for a given surfactant, this requirement can be met (Figures 5 and 10, right) or not (Figure 10, left) depending on the porous media. Of course, efficient injection requires also rapid emulsion/foam generation and propagation, which depends on adsorption, emulsion quality, force, and the porous nature, as seen above.

If the adsorption is a key factor in the delay of foam/emulsion generation, then the injection of a slug with a higher surfactant concentration is expected to reduce this delay. This strategy was tested on Indiana and Estaillades limestones.

Preflush injection of 0.2 wt % R-CADA solution for several pore volumes did accelerate the emulsion generation when injecting FCO2 90% on Indiana limestone (Figure 13): MPV decreased from ∼5 PV to ∼0.2 PV, and PVP decreased from ∼6.3 to ∼0.6 PV with no impact on MPG. Therefore, adsorption has an impact on both MPV and PVP, as previously hypothesized. If the difference in EPV of ∼10.6 PV is attributed to the adsorption, this would give ∼0.17 mg/g of rock, a value similar to that measured on Estaillades.

Figure 13 Impact of the preflush of R-CADA on transient emulsion behavior on Indiana limestone (on the left, FCO2 90%) and on Estaillades (on the right, FCO2 48%) at a Darcy velocity of 4 ft/d and 0.2 wt % R-CADA in brine. Violet points define the emulsion breakthrough time.

Using the adsorption value measured on Estaillades (§ 4.2.1 above), ∼1.5 PV of emulsion injection at FCO2 48% would be required to satisfy adsorption. However, preflush on Estaillades in the experiments at FCO2 48% only impacts MPV and emulsion breakthrough time without impacting PVP: MPV decreases slightly from ∼1 to ∼0.8 PV, and emulsion breakthrough time decreases from ∼1.6 to ∼1.0 PV while remaining well before the steady-state plateau, again indicating the presence of an end effect and backward strong emulsion generation into the core. Consequently, the observed MPV is only apparent and should not be considered as relevant. As emulsion generation occurs at the outlet, attaining the steady state (i.e., PVP) reflects just backward emulsion generation toward the inlet, which explains the comparable values of EPV for both experiments.

Therefore, the generation of strong emulsion is much more difficult on Estaillades, especially at high CO2 fractions, and adsorption, i.e., the surfactant dilution effect, may not be as important as the MPG effect.

At low CO2 fractions, as discussed before, the MPG is likely lower, and the adsorption may play a role in that case in delaying emulsion generation. Indeed, when injected at FCO2 15%, the same strategy of preflush is more effective: after preflush, an emulsion is generated almost instantaneously (Figure 14), but this emulsion is coarse since pressure continues to increase after emulsion breakthrough, as noted before.

Figure 14 Zoomed-in view of the first 1.5 PV of Figure 11 (right). Violet points define the emulsion breakthrough time.

For cases such as Estaillades requiring high MPG for strong emulsion generation, a possible strategy may consist of a primary injection of a CO2 emulsion requiring lower MPG (i.e., at lower CO2 fraction) at sufficiently high flow rates to generate strong foam, and then the CO2 fraction can be gradually increased to the desired value.

Alternatively, a combination of both strategies could be envisioned. This approach has been tested on quartz sandpack: the first injected slug was at FCO2 48%, for which fast emulsion generation was observed before. Moreover, to accelerate surfactant adsorption, this first slug contains 0.5 wt % R-CADA. Then, the slug of interest at FCO2 95% and 0.2 wt % R-CADA is injected, for which an important delay was previously observed with direct injection (Figure 6). The results are presented in Figure 15: for both temperatures tested, the use of a concentrated preflush, capable of rapidly generating an emulsion at low FCO2, considerably accelerates emulsion generation and propagation at higher CO2 fractions.

Figure 15 Impact of gradient injection (FCO2 48% to 95%) on emulsion generation on quartz sandpack. The slug at FCO2 48% contains 0.5 wt % R-CADA in brine, and the slug at FCO2 95% contains 0.2 wt % R-CADA in brine. Violet points define the emulsion breakthrough time.

Again, from a practical standpoint, given the importance of the nature of the reservoir rock as demonstrated above, it is advisable to use a reservoir core as early as possible starting from the first stage of the study, as the criteria for choosing the appropriate rock analogue are unclear. The strategy of injecting 2 slugs can be tested: either a preflush at high surfactant concentration or at low FCO2, if adsorption is found to be the reason for delayed foam/emulsion generation (as recommended by Mannhardt and Svorstøl in 200144) or, if high MPG is the issue, the first slug at a CO2 fraction for which fast generation of foam/emulsion was observed (generally, at low FCO2), followed by the target FCO2.

Regarding CO2 storage implementation, ideally, the CO2-containing surfactant should be injected into the target geological storage site. It is therefore important to assess whether an emulsion can be generated during the process to improve the sweep efficiency. Due to technical limitations in the laboratory, it was not possible to inject R-CADA dissolved in the CO2 phase. To simulate this, CO2 was injected into a surfactant-filled core of Indiana limestone. In doing so, we assume an instantaneous partition of the surfactant between the two phases.

A 3 PV slug at FCO2 5%, 0.2% R-CADA was first injected, followed by continuous injection of CO2. Figure 16 illustrates the apparent viscosity and in situ CO2 saturation during the R-CADA preflush and subsequent continuous CO2 injection. It is observed that upon the injection of pure CO2 into the core, the emulsion is generated almost instantaneously. The peak of the apparent viscosity is around 27 cP, and the apparent viscosity is almost constant at 5 cP after 3 PV of CO2 injection. CO2 saturation was near 63% after the injection of ∼3.5 PV of CO2, while in the experiment without surfactant (not shown here) the injection of 9 PV of CO2 was necessary to achieve this value.

Figure 16 Emulsion apparent viscosity and CO2 in situ saturation during R-CADA preflush and continuous CO2 injection at 4 ft/d in Indiana limestone. Conditions: 110 °C, 150 barg, 0.2 wt % R-CADA in brine.

5 Conclusions

It has been shown that the nature of the porous media and adsorption are key factors for the success of CO2 emulsion/foam injection, controlling fast generation and propagation, as well as the high apparent viscosity at a high CO2 fraction. As demonstrated, the mitigation strategy will vary according to which factor is most relevant to the desired objective.

The effects of various parameters (CO2 volume fraction, temperature, and presence of oil) on transient and steady-state characteristics of foam/emulsion transport have been investigated on quartz sandpack of high permeability under conditions such that the pressure gradient was above the minimum pressure gradient. The steady-state emulsion strength increases with FCO2, but it is accompanied by an increase in the minimum pore volume and pore volume for emulsion propagation, which is a drawback to overcome.

Surfactant adsorption and the nature of porous media have been studied on two limestone materials coming from different sources. Despite having the same type of mineralogy, they yield very different behaviors: in one case, the foam/emulsion is formed at the inlet of the core with a delay of foam generation and propagation due to surfactant adsorption, while in the other case it is generated at the outlet of the core by the end effect, probably due to elevated minimum pressure gradient.

Two different mitigation strategies were thus proposed and evaluated to accelerate the achievement of the steady-state regime:If the consumption of surfactants by adsorption is the main reason for the delay, the injection of a concentrated preflush containing a sacrificial amount of surfactant can be a solution. It may be pertinent to inject this preflush at low FCO2 to accelerate the satisfaction of adsorption.

If the limitation is related to an insufficient pressure gradient to overcome the MPG, preinjecting a short slug at a CO2 fraction for which MPG is lower enables the pressure gradient to be increased sufficiently to facilitate the foam/emulsion generation of the subsequent target FCO2 slug.

In some situations, a slug at low FCO2 may be the best choice for both strategies: a higher absolute amount of surfactant will satisfy adsorption faster than the target slug. Once adsorption is satisfied, it will provide some mobility control; finally, since generally the MPG of the wetter foam is lower, this simplifies the generation of the emulsion/foam of the target slug.

For applications in CO2 storage, the CO2 emulsion can substantially improve mobility control at the injected CO2 front, provided it contains the surfactant capable of creating a strong emulsion.

This study also highlights the importance of porous media not only for steady-state conditions but also for the transition behavior of emulsion/foam (generation and propagation).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.4c04137.Materials, experimental procedures, porous media properties, procedure for dynamic adsorption measurements, corresponding results, and interpretation (PDF)

Supplementary Material

ao4c04137_si_001.pdf

Author Present Address

⊥ Sinopec Shanghai Research Institute of Petrochemical Technology, 1658 Pudong Beilu, Shanghai 201208, China

Author Present Address

∇ Aramco Asia, F 43, China World Tower, No. 1 Jian Guo Men Wai Avenue, Chaoyang District, Beijing 100004, P. R. China

The authors declare no competing financial interest.

Acknowledgments

TotalEnergies is gratefully acknowledged for allowing the publication of this work. We thank Mss. Géraldine Salabert and Michèle Joly for performing part of the experimental work.
==== Refs
References

Holm L. W. Carbon dioxide solvent flooding for increased oil recovery. Trans. AIME 1959, 216 , 225–231. 10.2118/1250-G.
Godec M. L. ; Kuuskraa V. A. ; Dipietro P. Opportunities for using anthropogenic CO2 for Enhanced Oil Recovery and CO2 storage. Energy Fuels 2013, 27 (8 ), 4183–4189. 10.1021/ef302040u.
Ampomah W. ; Balch R. ; Cather M. ; Rose-Coss D. ; Dai Z. ; Heath J. ; Dewers T. ; Mozley P. Evaluation of CO2 storage mechanisms in CO2 enhanced oil recovery sites: application to Morrow sandstone reservoir. Energy Fuels 2016, 30 (10 ), 8545–8555. 10.1021/acs.energyfuels.6b01888.
Kolster C. ; Masnadi M. S. ; Krevor S. ; Mac Dowell N. ; Brandt A. R. CO2 -Enhanced Oil Recovery: a catalyst for gigatonne-scale carbon capture and storage deployment?. Energy Envir. Sci. 2017, 10 (12 ), 2594–2608. 10.1039/C7EE02102J.
Iglauer S. ; Paluszny A. ; Rahman T. ; Zhang Y. ; Wülling W. ; Lebedev M. Residual trapping of CO2 in an oil-filled, oil-wet sandstone core: results of three-phase pore-scale imaging. Geophys. Res. Lett. 2019, 46 (20 ), 11146–11154. 10.1029/2019GL083401.
Sun Q. ; Ampomah W. ; Kutsienyo E. ; Appold M. ; Adu-Gyamfi B. ; Dai Z. ; Soltanian M. R. Assess-ment of CO2 trapping mechanisms in partially depleted oil-bearing sands. Fuel 2020, 278 , 118356 10.1016/j.fuel.2020.118356.
Garcia S. ; Kaminska A. ; Maroto-Valer M. M. Underground carbon dioxide storage in saline formations. Proc. Inst.Civil Eng. 2010, 163 (2 ), 77–88. 10.1680/warm.2010.163.2.77.
Krevor S. ; Pini R. ; Zuo L. ; Benson S. M. Relative permeabilities and trapping of CO2 and water in sandstone rocks at reservoir conditions. Water Resour. Res. 2012, 10.1029/2011WR010859.
Iglauer S. CO2-water-rock wettability: variability, influencing factors, and implications for CO2 storage. Acc. Chem. Res. 2017, 50 (5 ), 1134–1142. 10.1021/acs.accounts.6b00602.28406029
Ringrose P. S. ; Furre A.-K. ; Gilfillan S. M. V. ; Krevor S. ; Landro M. ; Leslie R. ; Meckel T. ; Nazarian B. ; Zahid A. Storage of carbon dioxide in saline aquifers: Physico-chemical, key constraints, and scale-up potentials. Ann. Rev. Chem. Biomol. Eng. 2021, 12 , 471–494. 10.1146/annurev-chembioeng-093020-091447.33872518
CO2-dissolved: successfully combining CO2 storage and geothermal extraction. BRGM, 2021. https://www.brgm.fr/en/reference-completed-project/co2-dissolved-successfully-combining-co2-storage-geothermal-heat
Rossen W. R. ; Farajzadeh R. ; Hirasaki G. J. ; Amirmoshiri M. Potential and Challenges of Foam-Assisted CO2 Sequestration. Geoenergy Science and Engineering 2024, 239 , 212929 10.1016/j.geoen.2024.212929.
Massarweh O. ; Abushaikha A. S. A review of recent developments in CO2 mobility control in enhanced oil recovery. Petroleum 2022, 8 (3 ), 291–317. 10.1016/j.petlm.2021.05.002.
Li S. ; Wang P. ; Wang Z. ; Cheng H. ; Zhang K. Strategy to enhance geological CO2 storage capacity in saline aquifer. Geophys. Res. Lett. 2023, 50 (3 ), e2022GL101431 10.1029/2022GL101431.
Cui L. ; Bourrel M. ; Dubos F. ; Klimenko A. Surfactant for enhanced oil recovery. US 11254854 B2, 2022.
Cui L. ; Dubos F. ; Bourrel M. Novel alkyl-amine surfactants for CO2 emulsion assisted enhanced oil recovery. Energy Fuels 2018, 32 (8 ), 8220–8229. 10.1021/acs.energyfuels.8b01555.
Gauglitz P. A. ; Friedmann F. ; Kam S. I. ; Rossen W. R. Foam generation in porous media. In Proceedings of the SPE/DOE Improved Oil Recovery Conference, Tulsa, OK, April 13–17, 2002; OnePetro, 2002; SPE-75177-MS. 10.2118/75177-MS
Rossen W. R. ; Gauglitz P. A. Percolation theory of creation and mobilization of foams in porous media. AIChE J. 1990, 36 (8 ), 1176–1188. 10.1002/aic.690360807.
Yu G. ; Vincent-Bonnieu S. ; Rossen W. R. Foam propagation at low superficial velocity: implications for long-distance foam propagation. SPE Journal 2020, 25 (06 ), 3457–3471. 10.2118/201251-PA.
Rossen W. R. Theory of mobilization pressure gradient of flowing foams in porous media: I. Incompressible foam. Journal of Colloid and Interface Science 1990, 136 (1 ), 1–16. 10.1016/0021-9797(90)90074-X.
Chang S. H. ; Owusu L. A. ; French S. B. ; Kovarik F. S. The effect of microscopic heterogeneity on CO2-foam mobility: Part 2-mechanistic foam simulation. In Proceedings of the SPE/DOE Improved Oil Recovery Conference, Tulsa, OK, April 22–25, 1990; OnePetro, 1990; SPE-20191-MS. 10.2118/20191-MS
M’barki O. ; Ma K. ; Mateen K. ; Ren G. ; Luo H. ; Neillo V. ; Bourdarot G. ; Morel D. ; Nguyen Q. Experimental Investigation of Multiple Pore Vololumes Needed to Reach Steady-state Foam Flow in a Porous Media. In Proceedings of IOR 2019–20th European Symposium on Improved Oil Recovery, April 2019; European Association of Geoscientists & Engineers; Vol. 2019 , pp 1–17. 10.3997/2214-4609.201900148
Myers T. J. ; Radke C. J. Transient foam displacement in the presence of residual oil: experiment and simulation using a population-balance model. Industrial & engineering chemistry research 2000, 39 (8 ), 2725–2741. 10.1021/ie990909u.
Apaydin O. G. ; Kovscek A. R. Surfactant concentration and end effects on foam flow in porous media. Transport in porous media 2001, 43 , 511–536. 10.1023/A:1010740811277.
Nguyen Q. P. ; Currie P. K. ; Zitha P. L. Determination of foam induced fluid partitioning in porous media using X-ray computed tomography. In Proceedings of the International Conference on Oilfield Chemistry, Houston, TX, February 5–7, 2003; OnePetro, 2003; SPE-80245-MS. 10.2118/80245-MS
Almajid M. M. ; Nazari N. ; Kovscek A. R. Modeling steady-state foam flow: hysteresis and backward front movement. Energy Fuels 2019, 33 (11 ), 11353–11363. 10.1021/acs.energyfuels.9b01842.
AlYousef Z. ; Gizzatov A. ; AlMatouq H. ; Jian G. Effect of Crude Oil on CO2–Foam Stability. In Proceedings of the Offshore Technology Conference Asia, Kuala Lumpur, Malaysia, November 2–6, 2020; OnePetro, 2020; OTC-30210-MS. 10.4043/30210-MS
Beunat V. ; Pannacci N. ; Batot G. ; Gland N. ; Chevallier E. ; Cuenca A. Study on the Impact of Core Wettability and Oil Saturation on the Rheological Behavior of CO2-Foams. In Proceedings of the SPE Middle East Oil and Gas Show and Conference, Manama, Bahrain, March 18–21, 2019; OnePetro, 2019; SPE-194963-MS. 10.2118/194963-MS
Chen H. ; Elhag A. S. ; Chen Y. ; Noguera J. A. ; AlSumaiti A. M. ; Hirasaki G. J. ; Nguyen Q. P. ; Biswal S. L. ; Yang S. ; Johnston K. P. Oil effect on CO2 foam stabilized by a switchable amine surfactant at high temperature and high salinity. Fuel 2018, 227 , 247–255. 10.1016/j.fuel.2018.04.020.
Jian G. ; Zhang L. ; Da C. ; Puerto M. ; Johnston K. P. ; Biswal S. L. ; Hirasaki G. J. Evaluating the transport behavior of CO2 foam in the presence of crude oil under high-temperature and high-salinity conditions for carbonate reservoirs. Energy Fuels 2019, 33 (7 ), 6038–6047. 10.1021/acs.energyfuels.9b00667.
Hosseini-Nasab S. M. ; Zitha P. L. J. Chemical-foam Design as a Novel Approach Towards Immiscible Foam Flooding for Enhanced Oil Recovery. In Proceedings of IOR 2017–19th European Symposium on Improved Oil Recovery, April 2017; European Association of Geoscientists & Engineers, 2017; Vol. 2017 , pp 1–16. 10.3997/2214-4609.201702311
Jian G. ; Fernandez C. A. ; Puerto M. ; Sarathi R. ; Bonneville A. ; Biswal S. L. Advances and challenges in CO2 foam technologies for enhanced oil recovery in carbonate reservoirs. J. Pet. Sci. Eng. 2021, 202 , 108447 10.1016/j.petrol.2021.108447.
Belhaj A. F. ; Elraies K. A. ; Mahmood S. M. ; Zulkifli N. N. ; Akbari S. ; Hussien O. S. The effect of surfactant concentration, salinity, temperature, and pH on surfactant adsorption for chemical enhanced oil recovery: a review. Journal of Petroleum Exploration and Production Technology 2020, 10 , 125–137. 10.1007/s13202-019-0685-y.
Amirmoshiri M. ; Zhang L. ; Puerto M. C. ; Tewari R. D. ; Bahrim R. Z. B. K. ; Farajzadeh R. ; Hirasaki G. J. ; Biswal S. L. Role of wettability on the adsorption of an anionic surfactant on sandstone cores. Langmuir 2020, 36 (36 ), 10725–10738. 10.1021/acs.langmuir.0c01521.32870010
Heberling F. ; Trainor T. P. ; Lützenkirchen J. ; Eng P. ; Denecke M. A. ; Bosbach D. Structure and reactivity of the calcite–water interface. J. Colloid Interface Sci. 2011, 354 (2 ), 843–857. 10.1016/j.jcis.2010.10.047.21087772
Song J. ; Zeng Y. ; Wang L. ; Duan X. ; Puerto M. ; Chapman W. G. ; Biswal S. L. ; Hirasaki G. J. Surface complexation modeling of calcite zeta potential measurements in brines with mixed potential determining ions (Ca2+, CO32–, Mg2+, SO42−) for characterizing carbonate wettability. J. Colloid Interface Sci. 2017, 506 , 169–179. 10.1016/j.jcis.2017.06.096.28735190
Chen Y. ; Elhag A. S. ; Poon B. M. ; Cui L. ; Ma K. ; Liao S. Y. ; Reddy P. P. ; Worthen A. J. ; Hirasaki G. J. ; Nguyen Q. P. ; Biswal S. L. ; Johnston K. P. Switchable nonionic to cationic ethoxylated amine surfactants for CO2 enhanced oil recovery in high-temperature, high-salinity carbonate reservoirs. SPE J. 2014, 19 (02 ), 249–259. 10.2118/154222-PA.
Cui L. ; Ma K. ; Abdala A. A. ; Lu L. J. ; Tanakov I. ; Biswal S. L. ; Hirasaki G. J. Adsorption of a switchable cationic surfactant on natural carbonate minerals. SPE Journal 2015, 20 (01 ), 70–78. 10.2118/169040-PA.
Zhang L. ; Jian G. ; Puerto M. ; Wang X. ; Chen Z. ; Da C. ; Johnston K. ; Hirasaki G. ; Biswal S. L. Crude Oil Recovery with Duomeen CTM-Stabilized Supercritical CO2 Foams for HPHT and Ultrahigh-Salinity Carbonate Reservoirs. Energy Fuels 2020, 34 (12 ), 15727–15735. 10.1021/acs.energyfuels.0c02048.
Al Mahrouqi D. ; Vinogradov J. ; Jackson M. D. Zeta potential of artificial and natural calcite in aqueous solution. Advances in colloid and interface science 2017, 240 , 60–76. 10.1016/j.cis.2016.12.006.28063520
Li S. ; Jackson M. D. ; Agenet N. Role of the calcite-water interface in wettability alteration during low salinity waterflooding. Fuel 2020, 276 , 118097 10.1016/j.fuel.2020.118097.
Deposition no. PCT/FR2023/051334.

Chen X. ; Da C. ; Hatchell D. C. ; Daigle H. ; Ordonez-Varela J. R. ; Blondeau C. ; Johnston K. P. Ultra-stable CO2-in-water foam by generating switchable Janus nanoparticles in-situ. J. Colloid Interface Sci. 2023, 630 , 828–843. 10.1016/j.jcis.2022.10.102.36356449
Mannhardt K. ; Svorstøl I. Surfactant concentration for foam formation and propagation in Snorre reservoir core. J. Pet. Sci. Eng. 2001, 30 (2 ), 105–119. 10.1016/S0920-4105(01)00107-3.
Lotfollahi M. ; Farajzadeh R. ; Delshad M. ; Varavei A. ; Rossen W. R. Comparison of implicit-texture and population-balance foam models. In Proceedings of the SPE EOR Conference at Oil and Gas West Asia, Muscat, Oman, March 21–23, 2016; OnePetro, 2016; SPE-179808-MS. 10.2118/179808-MS
Ma K. ; Ren G. ; Mateen K. ; Morel D. ; Cordelier P. Modeling techniques for foam flow in porous media. SPE Journal 2015, 20 (03 ), 453–470. 10.2118/169104-PA.
Farajzadeh R. ; Lotfollahi M. ; Eftekhari A. A. ; Rossen W. R. ; Hirasaki G. J. H. Effect of permeability on implicit-texture foam model parameters and the limiting capillary pressure. Energy Fuels 2015, 29 , 3011–3018. 10.1021/acs.energyfuels.5b00248.
Vavra E. ; Bai C. ; Puerto M. ; Ma K. ; Mateen K. ; Hirasaki G. J. ; Biswal S. L. Effects of velocity on N2 and CO2 foam flow with in-situ capillary pressure measurements in a high-permeability homogeneous sandpack. Sci. Rep. 2023, 13 , 10029 10.1038/s41598-023-36345-4.37340000
Ding L. ; Jouenne S. ; Gharbi O. ; Pal M. ; Bertin H. ; Rahman M. A. ; Economou I. G. ; Romero C. ; Guérillot D. An experimental investigation of the foam enhanced oil recovery process for a dual porosity and heterogeneous carbonate reservoir under strongly oil-wet condition. Fuel 2022, 313 , 122684 10.1016/j.fuel.2021.122684.
Blunt M. J. ; Bijeljic B. ; Dong H. ; Gharbi O. ; Iglauer S. ; Mostaghimi P. ; Paluszny A. ; Pentland C. Pore-scale imaging and modelling. Adv. Water Res. 2013, 51 , 197–216. 10.1016/j.advwatres.2012.03.003.
Bijeljic B. ; Mostaghimi P. ; Blunt M. J. Insights into non-Fickian solute transport in carbonates. Water Resour. Res. 2013, 49 (5 ), 2714–2728. 10.1002/wrcr.20238.24223444
Jones S. A. ; Kahrobaei S. ; Van Wageningen N. ; Farajzadeh R. CO2 Foam Behavior in Carbonate Rock: Effect of Surfactant Type and Concentration. Ind. Eng. Chem. Res. 2022, 61 (32 ), 11977–11987. 10.1021/acs.iecr.2c01186.
