Amidst global warming and greenhouse gas emissions, carbon dioxide-enhanced oil recovery, referred to as CO2-EOR, has become the potential method to enhance recovery from mature oilfields. CO2 is either used in alternated flooding with water or continuous flooding. Water stream flooding still leaves almost 50% of the oil in the reservoir that can be recovered through CO2-EOR. CO2 possesses multi-contact miscibility leading to significant oil extraction, but to attain miscibility, a complete understanding of minimum miscibility pressure is essential. Therefore, the important concepts surrounding CO2-EOR are explained in this article, along with the multiple methods developed for the estimation of gas minimum miscibility pressure. At present, the EOR based on CO2 flooding is quite promising but needs special legal and financial incentives for scaling up the capacity and market outreach.
In CO2-EOR, CO2 injected into the reservoir pulls the light-to-intermediate components from the reservoir even if it is not truly miscible with the oil. As the pressure becomes considerably high, the miscibility develops and the oil is displaced from the reservoir. The formation of CO2 miscibility with crude oil usually follows two stages .
Stage-1: Swelling of the reservoir oil via initial contact with the gas
Stage-2: Decreasing the velocity of the reservoir oil followed by the decrease in interfacial tension between injected CO2 – reservoir oil
Oil reservoirs are naturally available underground at a noticeable depth with oil/gas in porous rocks under extreme (high pressure and temperature) conditions. There are two conventional ways to extract oil from the reservoir: primary and secondary methods. Primary production of oil which is 5-10% of the total is commonly achieved under the pressure of the oilfield itself, whereas the rest 40-50% of the reservoir is extracted via secondary methods . In the secondary method, gas or water is injected into the underground reservoir bed to displace or force the oil into the production well. This method is relatively more successful than the primary one, but still leaves around 50-60% of the oil in the reservoir.
Here comes the need for an enhanced recovery method that can change the oil properties to make it more suitable for extraction. Gas injection-based oil recovery has gained considerable importance in this regard . At present, there are different gas-injected oil recovery (EOR) methods available and, among these, miscible or immiscible gas injection is a popular method for recovering light or medium oils from carbonate reservoirs.
2 Understanding EOR: Water Flooding & CO2 Injection
The term “Enhanced Oil Recovery” (EOR) describes the application of additional engineering methods to enhance and boost the quantity of recovered oil. Chemical injection, gas injection, steam injection, thermal injection, and other techniques are common EOR techniques. Water flooding & gas/CO2 injection process are considered the second stage of crude oil production during which an external fluid such as water or gas is injected into the reservoir. Secondary recovery processes work to keep the reservoir under pressure while moving hydrocarbons away from the wellbore. Commonly known secondary methods are water flooding and gas/CO2 injection.
2.1 Water flooding
Water Flooding is the process in which water is injected into the oil reservoir to maintain pressure as well as to displace oil from the reservoir up to the production well . While around 60% of EOR methods involve gas injection, production companies are having significant success with oil recovery using the water flooding process.
2.1.1 Basic Mechanism
When the water is injected into the reservoir, it increases the exhausting pressure, forms the homogeneous mixture with the oil and enables the oil to move up to the production well to complete the recovery. For onshore and offshore developments, water flooding is frequently utilized . The technique can be used to transfer oil and increase depleted pressure inside reservoirs, as shown in the following figure.
Figure 1: Oil Recovery Process via water flooding; Recaptured from 
2.1.2 Water Flooding – Objectives
Water flooding is utilized to accomplish the following set of objectives :
Maintaining the reservoir pressure.
Effective displacement of oil from the reservoir to the production well.
2.1.3 Water Flooding – Widespread Acceptance
Some of the explanations for the widespread acceptance of water flooding include the following :
Water being inert is proved to be effective in displacing oil ranging from light to medium gravity
It is relatively simple to inject water into oil-bearing formations
In most cases, water is readily available
Relatively lower capital investment and operating costs are required, which results in favorable economics
2.2 CO2 Injected Oil Recovery Process
2.2.1 The Basic Mechanism
Another method to enhance oil recovery from the reservoir is to inject CO2 gas (CO2-EOR). The injection of CO2 into the oil reservoir enables the expulsion of oil from the rock pores towards the production well. The miscibility behavior of the oil and gas accomplishes the noticeable enhancement in the oil recovery process (figure 2). It has the capacity to recover an additional 10-20% of the original oil comparing to the water flooding process .
2.2.2 CO2 – EOR: What Does It Accomplish?
CO2 – EOR is employed to accomplish the following objectives :
Decrease the density of the oil making its flow difficult into the pores or congested areas, where recovery becomes a challenge
Evaporation of some contaminants presents within the oil, making it relatively pure for extraction
Reduction in the surface tension between the CO2 and the reservoir oil and water-oil thus increases permeability
Achieve miscibility at a lower pressure than the other hydrocarbon-based injection gases.
2.2.3 CO2 – EOR: Different Injection Schemes
Following are the different schemes for the CO2 – EOR process depending upon the reservoir geological location, fluid type and the nature of the rock :
Continuous CO2 Injection with no other fluid.
Continuous CO2 – water injection.
Continuous water-alternating gas injection followed by an equal volume of water flooding.
WAG Injection is followed by a cheap gas injection to increase the inflow of CO2 for better oil recovery.
Figure 2. CO2 – EOR Step-by-Step Mechanism; recaptured from 
2.2.4 CO2-EOR: Miscible & Immiscible Processes
The miscibility of CO2 with oil depends on a number of factors. The common ones include the oil’s composition, reservoir’s pressure, and temperature. Consequently, the advanced CO2-EOR method relies on the two predominant mechanisms: Miscible CO2 driven EOR and Immiscible CO2 driven EOR. Abedini et al. studied that miscible injection processes extract lighter components more frequently than immiscible injection processes, producing higher-quality oil as a result . Considering these circumstances, miscible CO2 EOR is often preferred over immiscible injection scheme.
2.2.5 Miscible CO2 Injection
CO2 miscible flooding or injection is an effective method for recovering oil from spent reservoirs. CO2 when injected at sufficiently high-temperature forms a viscous mixture miscible in the crude oil, resulting in almost 40% recovery of the remaining oil in the reservoir. In the USA, UAE, and other countries, half of the water-flooded oilfields are exploited by CO2 injection . The CO2 also blends with the oil, thus improving its mobility and allowing it to flow more conveniently towards the production wells. When CO2 miscible flooding is used, the interfacial tension is almost negligible and as a result, the capillary forces are non-existent. This results in the reduction of oil saturation as compared to the residual oil saturation and hence the yield of recoverable oil increases. The CO2 gas made multiple contact miscibility (MCM) with the oil and achieves miscibility through spontaneous mass transfer between the two phases. Such a type of miscibility is further elaborated below.
2.2.6 Multiple-Contact Miscibility (MCM)
In most gas-oil interactions, the miscibility is not developed through first contact and is made through the transference of mass between the two phases. When CO2 gas moves through the porous medium of the reservoir, it interacts with the oil, causing a decrease in the surface tension and viscosity, thus resulting in the swelling of the oil. This phenomenon increases the fluidity of the oil and the resulting miscibility is called “multi-contact miscibility”, established at a pressure termed minimum miscibility pressure (MMP) and is an integral part of EOR.
MCM occurs through three types of displacement processes , :
Vaporizing gas drive
Condensing gas drive
These three processes are briefly discussed below:
Condensing gas drive: Such a gas drive is also known as an “enriched gas drive” because, during the process, reservoir oil is enriched by the intermediate components found in the injection gas. As soon as the gas comes into contact with the reservoir oil, its density decreases due to the condensing of gas-to-oil intermediates. This oil undergoes repeated interactions with the incoming fresh gas and becomes sufficiently enriched to develop miscibility with the gas at the trailing edge of displacement.
Vaporizing gas drive – In contrast to the condensing gas drive, this form of gas drive enriches the injected gas. Fresh gas is injected into the reservoir, where it interacts with fresh oil at the front of displacement, causing intermediates to vaporize into the gas phase. Through several interactions, the gas is sufficiently enriched to create miscibility with the heavier reservoir oil. Between the two phases, a miscible transition zone forms, which effectively displaces oil. Such a form of gas drive is usually observed with the injection of nitrogen.
Condensing/vaporizing drive – This drive is often termed a hybrid gas drive and combines the qualities of vaporizing and condensing gas drives. Here, none of the two phases (oil and gas) becomes sufficiently enriched to produce miscibility in this form of displacement. Instead, there is a significant transfer of intermediates, which causes the two phases to mix. The formation of miscibility occurs in this type of gas drive halfway between the leading and trailing edges of the displacement.
All three types are schematically explained below:
Figure 3. Mechanism of condensing drive responsible for CO2 – Reservoir Oil Miscibility – Recaptured from 
Figure 4. Schematic explaining vaporizing drive mechanism responsible for CO2 – Reservoir Oil Miscibility – Recaptured from 
Figure 5. Mechanism of vaporizing/condensing combined drive – Recaptured from 
3 MINIMUM MISCIBILITY PRESSURE
It is defined as the least pressure above which the miscibility is achieved between the injected gas and the oil through dynamic multi-contact process at the reservoir temperature, resulting in oil recoveries as high as 90% of the total oil reservoir .
3.2 Factors affecting MMP
There are multiple factors contributing to the decrease or increase in the magnitude of MMP, some are elaborated below :
3.2.1 Reservoir Temperature
According to research studies, the CO2 MMP is primarily changed by the temperature of the reservoir. Yellig and Metcalfe analyzed the impact of reservoir temperature on MMP, and concluded that with every 1oF increase, there was a 15 psi increase in MMP was observed .
3.2.2 Type of Oil
Experiments have been carried out in lab studies to estimate MMP utilizing both live and dead oils. Live oil was gathered from the field, and experiments were run under reservoir conditions. In contrast, the dead oil is devoid of its volatile components, is rather thick, and does not contain any dissolved gases. The outcomes of studies employing live and dead oils in slender tubes were compared in a research study. It was discovered that for both oils, the rate of recovery and the precision of the estimated MMPs were identical. However, compared to live oil samples, the MMP value for all dead samples was 10% lower .
3.2.3 Concentration of Intermediate Components/Impurities
The MMP of the system is significantly impacted by the presence of contaminants in the injected CO2 gas. Different gas contaminants affect the MMP differently. For instance, if pollutants like hydrogen sulfide or intermediate hydrocarbons (such as ethane, propane, etc.) are present, the system MMP drops. On the other hand, methane and nitrogen gas caused a significant rise in the system MMP. In most cases, methane is delivered into the system through the re-injected CO2 and flue gases .
3.3 Significance of MMP
Determining the MMP value is vital for understanding the dynamics of the EOR process because it provides operational parameters especially the initial injected gas pressure required to attain efficient oil recovery process. Injecting CO2 above the MMP into the oil reservoir results in a comparatively higher oil recovery ratio. So, basically knowing the value of MMP for any process, makes it easy to know the minimum gas injection pressure necessary for achieving higher oil recovery from the reservoir.
4 METHODS OF MMP DETERMINATION
As mentioned previously, precise determination of MMP plays a crucial role in implementing effective gas injection. Conventionally, both the experimental and mathematical methods are used to determine MMP and each method has its application limits. Some of the most common experimental techniques and theoretical/mathematical models available in the research literature are briefly discussed in this section.
4.1 Experimental Methods
It is crucial to directly assess MMP using experimental equipment to provide useful first-hand information and validate theoretical models. Researchers and engineers have worked over the past few decades to develop a variety of MMP experimental equipment and techniques, such as the slim tube experiment, rising bubble apparatus, vanishing interfacial test (VIT), and other techniques. The experimental methods range from tests on horizontal or vertical shale to sandstone cores through the utilization of different diameter packing. The results obtained from each of the techniques are different from the others and have their significance. The most common experimental methods are:
Slim Tube Experiment (STE)
Rising Bubble Apparatus (RBA)
Mixing cell experiment (MCE)
Vanishing Interfacial Test (VIT)
4.1.1 Slim Tube Experiments
The slim tube test method is widely used and is the acknowledged measuring technique in the oil industry . It is a straightforward and generally acknowledged for calculating MMP in the event of CO2 flooding. The test typically uses a slim tube with the following dimensions ranges and packing:
Tube length: 10ft to 120 ft
Diameter of tube: 0.12 inches to 0.63 inches (typical diameter is 0.25 inches).
Packing materials: unconsolidated sand or glass beads having sizes ranging from 50 to 270 mesh.
In the slim tube experiment, a gas, such as CO2, is injected into a slim tube that has been saturated with oil. The system’s conditions (temperature and pressure) are maintained constant, and gas is added at a pace that creates steady-state conditions. Typically, a slim tube’s displacement velocity ranges from 120 to 200 ft/D. Numerous tests are carried out to precisely determine MMP, and in each experiment, the amount of gas injected and the amount of recovered oil is recorded .
The gathered information is then utilized to calculate MMP using a pressure versus recovery plot and a widely used criterion known as brake-over pressure. Other requirements include an 80% recovery at gas breakthrough and a 95% final recovery rate. Figure 4 depicts a common configuration for slim tube equipment.
Figure 6. Slim Tube Apparatus; Reproduced from 
4.1.2 Rising Bubble Apparatus (RBA)
Another well-known MMP measurement method that has been around for more than 30 years and modified for many applications is the RBA test. This method was introduced in 1987 . Comparing this procedure to other experimental techniques, it performs quickly and with more simplicity, which attracts attention. The apparatus is shown in Figure 7. A glass tube mounted on a high-pressure gauge and housed in a bath with a controlled temperature. The experiment involves introducing a gas from the tube’s bottom and then observe the gas bubbles rise. Under various pressures, the speed and shape of the gas bubbles are seen. To estimate the MMP, the pressure dependence of rising gas bubbles is empirically translated.
Similar to the previously stated method, this method is much more dependable for estimating MMP under vaporizing gas drive, but its results are typically unreliable for condensing or vaporizing/condensing gas drives. This technology can only be used on a small scale because the majority of gas drives are vaporizing/condensing types.
Figure 7. Rising Bubble Apparatus; Reproduced from 
4.2 Mixing Cell Experiment (MCE)
This experiment’s main goal is to examine the phase behavior of oil and gas after their injection into the reservoir. A precise prediction of MMP values can be made by the mixing cell experiment in some circumstances . The primary prerequisite is that there must be a vaporizing or condensing gas drive and not a hybrid vaporizing/condensing gas drive. In a pressure-volume temperature cell, the oil and gas are repeatedly in contact, and equilibrium is then allowed to develop. These tests are carried out under a variety of pressures, and in each instance, they are finished when the composition of any phase does not change further. The main disadvantage of this method is that it cannot be used to estimate MMP in the event of a hybrid gas drive. Therefore, it is important to understand the miscibility process before using this method .
4.2.1 Vanishing Interfacial Test
The vanishing interfacial test (VIT), which was created in 1997 and has since evolved into various modified variants, is a third extensively utilized technology . Investigating the interfacial tension (IFT) behavior between gas and oil as a function of pressure in a PVT cell and extrapolating the observed curve of IFT vs pressure yields the MMP. However, in case of multi contact miscibility, it is not a reliable method for accurate estimation of the MMP. In addition, the composition of the mixture significantly effects the accuracy of the MMP as compared to the slim tube method.
4.3 Mathematical/Theoretical Models
According to a review of the literature on mathematical estimation, numerous correlations have been created by using regressions on experimental data. These mathematical models are helpful in forecasting CO2 MMP because they often only require a small number of input factors, very little knowledge about the fluids, and are simple to implement. However, their accuracy is typically lower compared to experimental values. They are nonetheless highly helpful for screening a wide range of reservoirs for potential CO2 displacements despite this drawback. In this part, a brief overview of the many correlations found in the literature and used for the prediction of pure CO2 MMP is provided.
4.3.1 The Cronquist Model
A mathematical relationship was presented by Cronquist in the year 1978 for estimating pure CO2 MMP . It was based on nearly 60 experimental measurements using different oils at different temperatures. The correlation is given as follows,
MC5+ = Molecular Weight of pentane fractions plus reservoir fluid fractions
C1 = Mole fraction of lighter components (methane and nitrogen) in the reservoir fluid.
The tested oils were evaluated in a reservoir with a temperature range of 75 to 54°F and a specific gravity of 23 to 44° API. The resulting MMP values ranged from 1075 to 5000 psi, with the experimental and projected values differing by an average of 310 psi. The highest inaccuracy, 1700 psi, was discovered, and this large mistake was attributable to MMP readings acquired from sources with conflicting definitions of MMP.
4.3.2 The Lee Model
This correlation was created in 1979 by a study including the injection of both miscible and immiscible CO2 . When the temperature of the reservoir oil was below the critical temperature for CO2, the MMP was calculated using the CO2 vapor pressure. In the opposite scenario, the developed correlation is used to predict MMP. Following is the developed mathematical model:
MMP = 7.3942 × 10b
b = 2.772 – (1519 / (492 + 1.8 TR)
TR = Temperature of the reservoir
The drawback of this correlation is that bubble point pressure can be taken as CO2 MMP when MMP is found smaller than bubble point pressure.
4.3.3 The Glasø Correlation
In 1985, Glasø conducted research on North Sea oil resources using experimental and correlational methods . This study used several injection fluids, including pure CO2, nitrogen, and hydrocarbon gases, to build a generalized correlation for the prediction of MMP. The correlation was derived using Benham et al graphical correlations . Additionally, it made predictions about MMP based on reservoir temperature, the molecular weight of the heavier fractions (C7+), and intermediate fractions (C2 to C6). The two forms of correlation depending upon the percentage of intermediates are given as follows,
MC5+ = Molecular Weight of heavier fluid fractions
FR = Mole percentage of intermediates
By using the above two equations it was found that the average deviation from experimental values was just -4.41% and a standard deviation of 11.65% was obtained in the oppression range of 900 to 4500 psi.
4.3.4 The Yellig & Metcalfe Model
Yellig and Metcalfe proposed a mathematical model in 1980 after researching both live and dead oils . Light fractions (methane, nitrogen, and CO2), intermediate fractions (hydrocarbons with molecular weights between C2 and C6), and heavier fractions (C7+) were considered to make up the three main fractions of oils. Following are the correlation’s details:
The range of temperatures for this correlation was from 95 to 190 °F. Two factors, such as reservoir temperature and oil content, were first taken into consideration in this investigation. The correlation was based solely on reservoir temperature, and it was discovered that the effect of temperature is much more significant than the effect of oil composition. Over the temperature range of 95 to 190 °F, it was discovered that an increase in temperature resulted in a rise of 15 psi per °F. For those oils with MMP lower than bubble point pressures, the postulated association was sufficiently accurate. While errors were seen in situations where the bubble point pressure exceeded the predicted MMP. Additionally, compositional effects were discovered to be a significant cause of inaccuracy and should be taken into account when calculating MMP .
4.3.5 The Orr & Jensen Correlation
The extrapolated vapor pressure (EVP) method, which is used for MMP assessment of both living and dead oils, constituted the foundation for this correlation, which first appeared in 1984 . The Yellig & Metcalfe and Holm & Josendal and EVP correlations were very similar. However, at lower reservoir temperatures (less than 121°F), it offered a straightforward physical explanation for the two connections. Following are the correlation’s details:
For dead oils, this correlation was remarkably accurate. Though, the estimated MMP had to be increased by an acceptable safety margin of 200 to 300 psi in the event of dissolved gases. Furthermore, it was oversimplified to examine temperature alone because the compositional impacts of the oil were not taken into account oversimplification.
4.3.6 The Yuan Model
This correlation was created to forecast the MMP of pure and impure CO2 gas displacement of multicomponent oil at relatively high reservoir temperatures . It used developments in the analytical theory of the equation of state for the formulation of the correlation involving three parameters such as reservoir temperature, percentage of the intermediate compounds (C2 to C6), and molecule weight of heavier compounds (C7+).
Seventy MMP measurements were made in total, and the data were then fitted with a quadratic equation. The reservoir has a temperature range of 71 °F to 300 °F. When compared to the correlation data, an absolute average inaccuracy of 11.9% was found for the 41 slim tube MMP readings.
4.3.7 The Emera & Sarma Model
Relationship between pure and impure CO2 injections into oil reservoirs was presented by Emera & Sarma in 2005 . This correlation applied to oils with bubble point pressures under 50 psi and was based on a novel genetic algorithm technique. The proposed model for pure CO2 is given as follows:
MC5+ = Molecular Weight of heavier fluid fractions
Xvol = Mole fraction of volatiles like nitrogen and methane
Xmed = Mole fraction of intermediates (C2-C6)
4.3.8 The Hao-Dali-Kai Model
Hao et al. recently presented this model in the year 2015 . This had its foundation in the earlier Emera-Sarma model. It used a modified conjugate gradient and a global optimization approach to build a prediction model. The researchers discovered that the accuracy of the correlation coefficient benefits from the molecular weight of bigger molecules like MC7+ rather than MC5+. This relationship is given as follows:
Xvol = Mole fraction of volatiles like nitrogen and methane
Xmed = Mole fraction of intermediates (C2-C6)
In the earlier literature, a comparison was made with other models and found that this model had a wide application range. For instance, it is applied in temperature limits of 36.80 °F to 377.54 °F and a pressure of 0 to 10,000 psi. The average absolute error was noticeably less than other models and hence it was more precise in predicting the MMP of CO2.
5 FUTURE PROSPECTS OF CO2 – EOR:
For the past four decades, CO2-EOR showed noticeable success, but there is still some great potential lying ahead for its increased growth. This potential can be further exploited by capturing CO2 from different natural sources like biomass plants, power plants, and industrial facilities. “Carbon capture” technology involves capturing CO2 from different natural or artificial sources to provide climate and green energy solutions.
Climate change or global warming is one of the leading global challenges to a safer living environment. CO2 –EOR holds great potential in addressing this challenge and is considered the only large-scale carbon sequestration industry. It uses a great deal of CO2 and buries a lot under the ground, contributing to a carbon-free environment. Experts are aware of the science underlying the process, and there are many CO2 emitters from various sources as well as several oil reservoirs to store the CO2 they produce. The economic choice, which is crucial, mostly depends on several variables. Although it may be challenging to foresee these variables, creating a platform to aid in decision-making could considerably speed up CO2-EOR.
CO2-EOR does need some more global attention because it can supply reduced carbon oil to support the green energy-based processes. It is indeed a fact that fossil fuels currently supply the energy basis upon which global societies function, and that an abrupt alteration in the composition of that basis could potentially disturb the balance of the global economy and integral components of today’s society. Greater financial and legal incentives are required to start growing CO2-EOR with storage globally to fulfill this pressing demand.
Beyond the potential of CO2-EOR to extract oil from the reservoir, different industries, governments, and environmental organizations are closely monitoring CO2-EOR because of its potential to permanently store CO2. Not only does this method is maximizing the oil recovery, but also offers a bridge to a decreased Carbon emissions future  , .
CO2 gas possesses great potential in enhancing the overall oil recovery process as it forms the complete miscible solution with the reservoir oil, making it easy for the oil to float up towards the production well. Almost 80% of the total reservoir oil is made possible to recover via CO2 – EOR, compared to water flooding (40-50%). The Hao-Dali-Kai model stands out amongst the rest, in accurately predicting MMP, one of the crucial factors in EOR process. However, none of the methods are universally acceptable and further studies are required for their commercial-scale applications. There is immense potential in CO2 – EOR process as it can supply reduced carbon oil to support the green processes, but greater financial and legal incentives are required to scale up this technique.