Effect of CO2 on the interfacial tension and swelling of crude oil during carbonated water flooding

Since the influence of carbon dioxide (CO2) on the dynamic interfacial tension of crude oil and aqueous solutions at the elevated temperatures and pressures is an important issue, this investigation is aimed to measure the IFT of crude oil/water, CO2, and carbonated water. The measured values reveal that the IFTs of crude oil/CO2 are decreased through the first interval, while a gradual and continuous reduction in IFT is obtained at the second interval. Although both intercepts and slopes of the first interval lines are more than the second interval for all the studied temperatures, they show a similar trend as a function of temperature. The used vanishing IFT method measurement reveals that the minimum miscibility pressure (MMP), first contact miscibility pressure, and intersection point (where the slopes of the linear curves change) of two linear equations have a linear increasing trend as temperature changes. However, this trend was not observed for the aqueous solution saturated with CO2 (i.e., carbonated water) which can be related to the difference between liquid- and gas-phase behavior as well as the formation of carbonic acid in the carbonated water process. IFT values reduced from 18.6 mN/m (@ pH = 8) to 0.3 mN/m (@ pH = 14) with an increase in pH of the aqueous solution, while miscibility conditions can be obtained for CO2 cases at pressures higher than the MMP points. Considerably higher IFT values were measured in the presence of water and carbonated water in the range of 9.1–17.5 mN/m and 12.2–17.4 mN/m, respectively. The most important feature of the CW process can be the swelling factor of crude oil due to the fact that the oil swelling of up to 40% was achieved at elevated temperature (75 °C) and pressure (4500 psi) after only1600 s.


Introduction
Enhanced oil recovery (EOR) methods such as CO 2 injection have attracted a lot of attention in the oil industries. CO 2 injection is one of the most fundamental and reliable EOR methods, regardless of the oil type (Al-Jarba and Al-Anazi 2009; Bougre and Gamadi 2021;Lashkarbolooki et al. 2016b). Resecting the unique features of this method, several CO 2 -based flooding processes were proposed and examined (Lashkarbolooki and Ayatollahi 2018c;Mahinpey et al. 2007;Rostami et al. 2017;Sasaki et al. 2013;Zhao et al. 2011;Zhu et al. 2021;Zihao et al. 2011). It is possible 1 3 to inject the CO 2 in the miscible or immiscible conditions regarding thermodynamic conditions and crude oil properties. Besides, it is mentioned that CO 2 is mostly immiscible fluid with the crude oil at the first contact, while it can be miscible through the second contiguity with the crude oil which is called multi-contact miscibility (MMP) (Alomair and Garrouch 2016;Choubineh et al. 2019;Green and Willhite 1998;Wang et al. 2010).
At the MMP point, CO 2 can be dissolved in crude oil through a multi-contact conditions (Green and Willhite 1998;Holm and Josendal 1974;Lashkarbolooki and Ayatollahi 2018c). Therefore, it is essential to determine the MMP to propose the most efficient procedure for the CO 2 flooding process. Unfortunately, direct injection of CO 2 has several drawbacks including gravity segregation, high mobility, low sweep efficiency, gas fingering, and quick CO 2 breakthrough making it an untrustworthy EOR method in some cases (Gogoi and Gogoi 2019;Riazi 2011). As a way out, carbonated water (CW) injection was proposed not only for EOR purposes but also even for underground CO 2 reservations (Ahmadi et al. 2016;Bisweswar et al. 2020;De Nevers 1966;Duguid et al. 2011;Honarvar et al. 2017;Riazi 2011;Walsh et al. 2014). In this process, CO 2 will diffuse from the aqueous phase into the oleic phase consequently leading to a desired swelling and viscosity reduction (Alizadeh et al. 2014;Foroozesh et al. 2016). Furthermore, CW is considered an efficient method since it can capture CO 2 underground through the CO 2 sequestration phenomenon which can reduce greenhouse gas emissions which is a global concern (Mosavat and Torabi 2013;Wang et al. 2015).
It has been thoroughly reported that as CO 2 dissolves in water, the interfacial tension (IFT) of crude oil/water decreases (i.e., the IFT values of CW/crude oil were lower than that obtained for crude oil/water), which means IFT reduction can be considered as an important mechanism for CW flooding. Conversely, the results reported by  demonstrated that the type of crude oil has an undeniable impact on the IFT of crude oil/CW. Their observations supported the idea that the movement and partitioning of natural active agents in crude oil and their interfacial orientation can be disordered by CO 2 diffusion leading to higher IFT values, especially for the acidic crude oil.
Among the different effective thermodynamic parameters, the temperature has a considerable effect on both IFT of crude oil/carbonated water and MMP Lashkarbolooki et al. 2019a). Besides, natural surface-active components existing in crude oil can introduce different dissociations at different pH values since carbonated water pH significantly reduces even at low CO 2 content compared to water (Crawford et al. 1963;. So, knowing the IFT behavior at different pH values is an important parameter that must be carefully examined especially through carbonated water and alkaline flooding processes. Performing a comprehensive literature review enlightens that IFT of crude oil has a profound effect on the fluid properties and several contradicting results were reported for IFT variation such as (a) IFT decreases as a function of pH because of dissociation of acidic components (Cratin 1993;Keleşoğlu et al. 2011), (b) dissociation of basic compounds at low pH values leading IFT reduction (Peters 1931) and (c) IFT decreases at low and high pH values due to ionization of both basic and acidic components and the highest IFT value was observed for neutral pH (Bai et al. 2010;Buckley 1996).
Because several contradicting results regarding IFT of crude oil and CO 2 , water and CW systems as functions of pressure and temperature have been reported Jaeger and Pietsch 2009;Jaeger and Eggers 2012;Wang et al. 2010); therefore, the current work is concentrated on the evaluation of DIFT of crude oil in the presence of different solutions. In this way, the IFT of crude oil/CO 2 was investigated and MMP values at three different temperatures including 25 °C, 50 °C, and 75 °C were determined. After the evaluation of the impact of pH on the IFT of crude oil, DIFT and equilibrium IFTs (EIFT) between the crude oil phase and aqueous solutions (water and carbonated phases) at two different isobars of 500 psi and 4500 psi were investigated to reveal the effect of pH and CO 2 on the IFT of crude oil/carbonated water. In the last stage, the adsorption time of studied cases was evaluated by an empirical adsorption decay model to clarify the effect of CO 2 on the interfacial properties of crude oil.

Materials
The required CO 2 with a purity of better than 99% was used to prepare the carbonated water. Besides, the crude oil sample with 32 API° was kindly supplied by National Iranian South Oil Co. (NISOC Co.) from one of the southern Iranian oil reservoirs. Moreover, different aqueous solutions with different pH values were prepared (using 1 M HCl and 1 M NaOH solutions) to investigate the effects of pH and the activity of basic and acidic components of crude oil.

IFT apparatus
Among the different possible methods for measuring the MMP point, vanishing interfacial tension (VIT) is one of the recent techniques proposed for measuring the MMP (Wang et al. 2010). In this method, IFT between two different phases is measured as a function of pressure and then extrapolated to the zero IFT value which is corresponded to the MMP where there is no differentiation between two liquids (Ahmad et al. 2016;Lashkarbolooki and Ayatollahi 2018c).
In the current investigation, a high pressure-high temperature pendant drop IFT measurement equipment was used (Fanavri Atiyeh Pouyandegan Exir Co., Arak, Iran). This device (see Fig. 1) consists of a mechanical section and an image processing section that provides the capability of IFT measurement for the operator. In detail, the equipment is comprised of a visual cell equipped with two HP sapphire sight glasses rated for maximum pressure and temperature of 600 bar and 150 °C. These sight glasses capable the operator to monitor the internal contents of the measuring chamber and dispatch the images of the formed drop at the tip of the injection needle to the drop shape analysis software. The required drop was formed using two manual hydraulic pumps that can inject the drop and bulk fluids into the main chamber at the desired pressure. After forming the drop, a CCD camera equipped with a macrolens (Computar, Japan) records the required images every second and delivers them to the online software where the IFT values can be calculated using the drop shape analysis approach (Lashkarbolooki et al. 2019b). It should be noted that the used software is rapid enough to capture the required images as the drop was formed. On the other hand, the IFT variation in the current investigation was slow enough to be measured with enough reliability. In other words, since no chemical surfactants existed in the current work and the possible effect of swelling on IFT variation is a time-consuming phenomenon, there was no limitation to utilize the software for IFT and swelling calculations.
Besides, the required pressure in the main measuring chamber and drop and bulk accumulators was provided using two separate manual hydraulic pumps. On the other side, the temperature of the main measuring chamber was controlled using a PID controller coupled with PT-100 with an accuracy of 0.1 K using heating elements implanted in the main chamber body.
For the preparation of carbonated water solutions, about 80% of the visual cell was filled with water, and then, the pressure of the visual cell and the CO 2 tank was elevated to the desired value using the manual pump. As the pressure reached a plateau (which is an indication of equilibration of water and CO 2 ), the oil drop was injected into the bulk phase to form a drop at the tip of the nozzle. Since the excess amount of CO 2 was injected at the top of the bulk phase, it is completely obvious that the solution was always in equilibrium and the aqueous phase was completely saturated with CO 2 . The noteworthy point is that the IFT values were measured three times for at least 1 h to ensure reaching equilibrium.

Decay model
To correlate DIFT values of crude oil and studied aqueous solutions, an exponential decay model was used: where 0 , t , e are the initial IFT, IFT at time t and EIFT, respectively, and is an adjustable parameter called adsorption or relaxation time. Despite the simplicity of this model (a model with one adjustable parameter), it has an acceptable level of accuracy to predict the DIFT behavior of crude oil/ water and crude oil/carbonated water systems (Hamidian et al. 2019;Lashkarbolooki and Ayatollahi 2018b;).

Evaluation of EIFT of crude oil/CO 2
The EIFT values of crude oil/CO 2 were measured as a function of pressure at different operating temperatures (see Fig. 2a). According to the measured IFT values, no dynamic behavior was obtained for crude oil/CO 2 since the EIFT was obtained at the first contact of crude oil and CO 2 . From a closer look at Fig. 2a, one can conclude that both temperature and pressure influenced the EIFT of the crude oil/CO 2 system. At each isotherm of 25 °C, 50 °C, and 75 °C, EIFT values were reduced as the pressure was enhanced due to higher CO 2 solubility in crude oil as the pressure was increased. On the other hand, the CO 2 solubility reduces as the temperatures increases which directly leads to higher IFT values. Commonly, the EIFT of crude oil experiences a reducing pattern as a function of pressure (Cao and Gu 2013;Wang and Gu 2011;Wang et al. 2010;Zolghadr et al. 2013). Respecting this fact, two pressures of 500 and 4500 psi were examined in a wide range of temperatures. The results illustrated that the EIFT values decrease sharply at the first examined pressure, while it faced a gradual change for the second examined pressure. The lines of the first interval correlated with the contiguity of CO 2 and light components, whereas the second interval lines demonstrate the adjacency of CO 2 with heavier components (Nobakht et al. 2008;Zolghadr et al. 2013). According to the results depicted in Fig. 2a, it is obvious that the lower oil extraction at lower temperatures can lead to CO 2 contiguity with lighter components leading to easier dissolution. In contrast, an increase in the temperature leading to higher oil extraction which exposes CO 2 to heavy components causes IFT reduction due to a reduction in CO 2 dissolution in oil (Zolghadr et al. 2013).
(1) = + 0 − 10 − As aforementioned, the VIT method is one of the reliable and fast methods for MMP calculation in which the EIFT values of first and second intervals at each temperature extrapolate to zero and the MMP and P max (first contact miscibility pressure) values are, respectively, determined ). According to Fig. 2a and Table 1, the MMP values of 960 psi, 1506 psi, and 2200 psi were obtained for temperatures of 25 °C, 50 °C, and 75 °C, respectively. A glance into the MMP values calculated in this stage revealed that temperature enhancement leads to higher required pressures for complete miscibility of CO 2 into the crude oil probably due to the lower CO 2 solubility at the higher temperatures (Enick and Klara 1990;Hutin et al. 2014).
The P max values also increased from 1038 to 2555 psi and 3677 psi as temperature increased from 25 to 50 and 75 °C. To identify the intersection point where the slope of the linear curve changes, the equations of lines for two intervals in each temperature were calculated. The tabulated equations along with their correlation coefficient (R 2 ) revealed the successfulness of the utilized equations to correlate the EITF value with an acceptable level of accuracy (see Table 1). Furthermore, CO 2 density at each temperature and pressure and its relationship with intersection points at different temperatures is depicted in Fig. 2b. This figure revealed that the density of crude oil has an inflection point as temperature changes. Moreover, the density value which is pressure-dependent at each isotherm was depicted and the differences between intersection, inflection and supercritical points are shown in Fig. 2c. For temperature of 25 °C which inflection point and CO 2 phase change from the gas phase to the liquid phase are the same, and the intersection and inflection points are similar to each other. (Their difference is 0 in Fig. 2c) However, for temperatures of 50 and 75 °C which inflection points are not coincided to change of CO 2 phase from the gas phase to supercritical phase, intersection points were lower compared with the inflection point (200 psi and 150 psi lower than inflection points, respectively) and higher than the point in which the CO 2 phase changes (220 psi and 710 higher than a supercritical point). In other words, an increase in temperature from 50 to 75 °C increases the difference between supercritical and intersection points which moves the system toward higher pressures to reach MMP pressure. Also, the results revealed a linear correlation between the intersection point and CO 2 density.
The depicted slopes and intercepts of VIT lines of the first and second intervals (Figs. 3a and b) show that the slopes and intercepts of EIFT versus temperature curves almost decrease with increasing temperature in the same pattern. Figure 4 shows that although the slopes of the first interval lines are more than the second interval at all the studied temperatures, the same variation for the slope was observed which depends on temperature.

3
A closer examination of the results depicted in Fig. 5 demonstrated that increasing the temperature reduces the slope of EIFT lines consequently enhancing MMP, P max, and intersection point values although there is almost a linear relationship between intersection point, MMP, and P max values and temperature. Among the examined parameters, the enhancement rate of P max versus T is the highest among these three parameters which are attributed to the higher Fig. 2 a EIFT between crude oil and CO 2 at three constant temperature versus pressure, b density of CO 2 obtained from NIST (Lemmon et al. 2010) and its relation with intersection point and c relation between intersection point, inflection and supercritical point slope of the second interval. So, higher pressure is required to achieve the first contact miscibility at elevated temperatures. Since intersection points are calculated by equality of first and second intervals lines, it will also increase as a function of temperature. For this type of crude oil, it can be concluded that the performance of CO 2 injection can be higher for a reservoir with a lower temperature since a lower pressure is required to reach miscibility conditions and extraction of trapped oil. According to these findings, the CO 2 injection for this type of crude oil is more practical if the reservoir temperature is low.

pH effect on the IFT
Another important parameter is the pH of the solution since it affects the ionization of natural surface-active agents (Hutin et al. 2014). The evaluation of the pH effect on the DIFT and EIFT (Fig. 6 and Fig. 7a) shows that in the presence of acidic solutions (pH = 2 and 4) and weak alkaline solutions (pH = 8 and 10), the differences between the initial and EIFT values were insignificant which means that surface-active materials are not capable to correctly pack at the interface as a function of time (see Fig. 7). But, the results illustrated that a lower initial IFT value and a reduction trend for DIFT were observed as the solutions become more basic (pH = 11 to 14). Generally, when crude oil comes into contact with the aqueous phase, the active materials that exist in crude oil can transfer from the oil phase to the aqueous phase leading to an IFT reduction time dependently.  Ionization of both basic and acidic constituents accelerates this partition leading to more IFT reduction. The point that must be considered is that the effectiveness of pH variation on IFT highly depends on basic and acidic constituents of oil and their ionization (Hutin et al. 2014). It shows that in the presence of active alkaline, the acidic components of crude oil have more influences which can cause a reduction in IFT values. Since the ionization of acidic components is more influential at high pH values, the lowest IFT value of 0.3 mN/m was observed at a pH of 14. In detail, IFT values sharply decreased at higher pH values since the acidic species that existed on the surface of crude oil are saponified (Hutin et al. 2016(Hutin et al. , 2014. Moreover, the slight change of EIFT at pH between 2 and 10 shows that the basic constituents could not be ionized properly at this range of pH. So, basic species are not active enough at the interface to decrease the IFT values in this pH interval. Furthermore, due to the low ionization of basic components that existed at the interface at low pH values, no significant effect on the IFT values was observed as the pH of solutions was decreased. Furthermore, the adsorption time values calculated by the MED model (Fig. 7b) demonstrated that the basic solutions have a more effective role than the acidic ones in the ionization and packing of acidic species present in the oil phase. This phenomenon causes a lower relaxation time of surface-active materials in contact with alkaline solutions compared to those obtained for acidic ones.

Evaluation of IFTs of crude oil/carbonated water and crude oil/ water
DIFT values of crude oil/water system were measured at two different pressures of 500 psi and 4500 psi and different temperatures of 25 °C, 50 °C, and 75 °C (see Figs. 8a and b). In all the studied cases, since natural surfactants of crude oil such as asphaltene and resin fractions move into the interface as time passes, IFT decreases as a function of time till reaching an equilibrium state and this is the equality of adsorption/ desorption rate of natural surfactants indicates the achievement of EIFT. As shown in Fig. 8, IFT decreases when the temperature increases from 25 to 50 °C for both examined pressures, while the IFT value experiences an increase if the temperature increases from 50 to 70 °C. This temperature in which the IFT variation behavior shows different patterns is known as phase inversion temperature (PIT).
In the next stage, the results of the empirical decay model against log ((γ 0 −γ t )/(γ t −γ e )) were also evaluated to ensure reaching to an equilibrium state (see Fig. 9). As it is clear, the DIFT values can be divided into four parts: (1) induction time, (2) rapid fall time, (3) meso-equilibrium time, and (4) equilibrium time (Hua and Rosen 1988;Lashkarbolooki et al. 2016a). The required time to reach the meso-equilibrium can be estimated by calculating of adsorption time from Eq. 1.
The effect of temperature on the DIFT between crude oil and carbonated water at two different pressures of 500 and 4500 psi as shown in Figs. 10a and b, respectively. As shown in Fig. 10, increasing the temperature from 25 °C to 50 °C for both examined pressures led to an increase in the EIFT, while EIFTs were decreased as the temperature was enhanced from 50 to 75 °C.
For more clarification, the EIFT values of water/crude oil and carbonated water/crude oil systems at all the studied operating conditions were compared (see Fig. 11a). Under the constant temperatures, pressure almost has no considerable effect on the EIFT of both solutions including water and CW. It should be noted that as CO 2 was dissolved in the water phase, the aqueous pH solution was decreased due to the carbonic acid formation. As aforementioned, pH reduction had no considerable effect on the IFT of the studied crude oil. Despite this, the adsorption time of the crude oil/ aqueous phase was considerably affected by CO 2 content and lower times were required to achieve the equilibrium at high temperatures and pressures. This observation confirmed that the interfacial behavior of crude oil is completely different if the CO 2 existed in the solution or not (both gas and supercritical phases). In detail, the IFT of crude oil regardless of the CO 2 status (being gaseous or supercritical) experienced a sharp reduction as the pressure was changed. However, the crude oil/CW IFT almost remained unaffected by the CO 2 content of the aqueous phase similar to the trend observed for the water phase.
It is also observed that temperature has a considerable influence on the IFT values compared with pressure since it can influence both the entropy and packing of natural surfactants at the interface (Duan et al. 2006;Lashkarbolooki and Ayatollahi 2018a;Lashkarbolooki et al. 2016a). In detail, an increase in the temperature leads to an increase in the molecular movement and even reduces the IFT. For the water/crude oil system, as the temperature increases from 25 to 50 °C, natural surfactants can easily move to the interface  DIFT of crude oil/water versus log ((γ 0 −γ t )/(γ t −γ e )) at 500 psi and 50 °C due to the molecules movement enhancement, while as the temperature rises to 75 °C, the higher motion of molecules impedes their proper packing. According to these facts, it seems that the best packing of natural surfactants can occur at a temperature of 50 °C so that increasing entropy is the dominant parameter in the first interval, while the reduction of packing due to reduction of surface excess concentration of active agents can be considered as a dominant factor in the second interval (Lashkarbolooki et al. 2016b). Besides, the measurements revealed that EIFT increases if temperature enhances from 25 to 50 °C, but it again decreases as temperature further increases to a value of 75 °C which indicates the temperature of 50 °C as the phase inversion temperature (PIT). Totally, according to Fig. 11a, no considerable positive effect was observed as CO 2 dissolved in the water phase except for a little more IFT reduction in the temperature of 25 °C, while at the elevated temperatures of 50 and 75 C, IFTs of crude oil/carbonated water were higher than IFT of crude oil/deionized water due to a reduction in the surface activity of natural materials and their orientation at the interface comes from the higher entropy and CO 2 partitioning.
Measuring the oil swelling at the presence of carbonated water for two different pressures of 500 and 4500 psi during 400 s measurements revealed no considerable change in this parameter for the pressure of 500 psi as a function of temperature, while this factor experienced an enhancement as the pressure was raised to 4500 psi (see Fig. 12). It was expected to observe a reduction in the oil swelling rate as the temperature was increased (at constant pressure) since the amount of dissolved CO 2 in the aqueous phase decreases with temperature, while a contradicting trend was observed for high pressure. These observed trends can be correlated with the higher CO 2 content as pressure increases, while  Oil swelling values for carbonated system a at two pressures of 500 psi and 4500 psi after 400 s, b swelling factor after 400 s and 1600 s for pressure of 4500 psi enhancement of swelling factor at elevated temperature can be attributed to higher partitioning of CO 2 from the aqueous phase toward the crude oil phase because of hydrogen bond reduction. Figure 12b also shows that temperature introduces a higher impact on the partitioning of CO 2 and oil swelling factor than pressure.

Conclusion
The current study is concentrated on the influence of CO 2 on the surface properties of crude oil/carbonated water solutions as functions of pressure and temperature. The obtained results can be classified as below: 1. Considering the crude oil/CO 2 IFT values, two distinguished pressure intervals were observed for all the studied temperatures. That is, the EIFT values decreased sharply and gradually in the first and second intervals, respectively, with no discontinuities in the supercritical CO 2 conditions. 2. Similar trends were obtained for the intercepts and slopes of two-interval lines as a function of temperature. 3. There was rather a linear relationship between the intersection point, MMP, and P max values with temperature. 4. Although the same intersection and inflection points (equal to CO 2 phase changes from the gas phase to liquid phase) were observed at 25 °C, the inflection points (unequal to change of CO 2 phase point from the gas phase to supercritical phase) introduce no coincide with intersection points at 50 and 75 °C. The difference between supercritical and intersection points increases as temperature increases from 50 to 75 °C leading to the higher required pressure for MMP. 5. Since changing pressure is ineffective on the IFT of crude oil-water and crude oil/CW in contrast to the crude oil/CO 2 IFT (significant reduction as pressure enhances), it is possible to conclude that CO 2 content has no considerable influence on the IFT variation of crude oil/aqueous solution. 6. Since IFT remained rather intact as the aqueous phase pH was decreased, carbonic acid formation by the dissolution of CO 2 in the solution is not a proper way for IFT reduction. 7. Two different trends were observed for phase inversion temperature of crude oil/water and crude oil/carbonated water systems. In the case of crude oil/water, increasing temperature from 25 to 50 °C led to a reduction in IFT values while further increase in temperature to 75 °C led to an increase in IFT. However, a reverse trend was obtained for the crude oil/carbonated water system which was probably related to different effects such as entropy, micro-emulsion formation, swelling of crude oil, and packing and orientation of natural surfactants and CO 2 at the fluid/fluid interface. 8. Higher oil swelling factor at higher pressure (4500 psi) can be attributed to the upper CO 2 content, while higher swelling factor at elevated temperature (75 °C) can be attributed to the greater partitioning of CO 2 from the aqueous phase toward the crude oil phase.