Miscible Displacement Oil Recovery

Abstract

Miscible gas injection (MGI) is an effective enhanced oil recovery (EOR) method used worldwide often for light oil recovery. In the petroleum industry, many MGI processes typically involve injection of an associated gas (AG) mixture or CO2, which have both been recognised as excellent candidates for such processes. The initial part of this chapter provides a broad introduction and background to the EOR techniques used worldwide as well as those implemented in Oman oil fields and briefly discusses their critical importance. Oman is one of the most active countries in terms of successful MGI processes in the Middle East, hence the emphasis given in this chapter to such projects in this country. The second part covers the technical details of the MGI process and the potential problems and challenges associated with it, while the third part focuses mainly on the common techniques used to control gas mobility during gas flooding including MGI. The impediments and challenges for wider application of the mobility control techniques are also covered. The last section presents a sample feasibility evaluation for a real oil field around the implementation of mobility control techniques for an MGI process.

1. Introduction

Over the past few decades, the rate of the new substantial oil discoveries has been on the decline. As a result nowadays, many oil companies are trying to maximise oil production from their existing reserves and maintain oil flow rates at or above the economic level through production optimisation and the use of enhanced oil recovery (EOR) techniques [1]. Enhanced oil recovery refers to the methods of increasing or maintaining the ability of oil to flow through interconnected pores towards the production wells by changing the physical and/or chemical properties of the in-situ fluid-rock system. Presently, the average recovery factor (RF) from mature oilfields under the primary and secondary recovery is only 20–40% [2]. Given the earlier mentioned lack of substantial new discoveries, increasing the RF from matures fields has become important to meet the growing energy demand in the years to come.

During the life cycle of an oil field, the oil extraction may occur typically in three recovery stages of primary, secondary and tertiary (i.e. EOR). Essentially, the petroleum product is produced from the reservoir initially by the natural reservoir energy such as the solution gas drive, gas cap drive and aquifer influx [3]. This is often termed as primary recovery, where the first wells drilled in the field are able to produce the oil from the reservoir without any intervention. In this stage primarily, the pressure gradient between the reservoir and surface controls the hydrocarbon flow into the well and then to surface. Over time, the reservoir pressure may decline reducing the pressure at the bottom hole which may then become closer to the hydrostatic head of the fluid column in a production well reducing the oil flow rate achievable from the well. Subsequently, secondary recovery methods may be applied, for example, by injecting water or gas via injection wells into the reservoir to maintain the reservoir pressure and eliminate or minimise the previously observed decline in oil flow. This type of recovery methods has its own technical and economic limitations as may be determined by the cost and availability of injection fluids and/or the issues that may arise during the development of the in-situ flooding. For instance, in both water and gas flooding, the difference in fluid properties between the displacing fluids and to be displaced in-situ oil can result in unstable displacement, leading to a large oil volume left behind due to poor displacement efficiency and early breakthrough. Therefore, the application of such techniques may typically add up to only 40–50% of eventual oil recovery.

When the oil in a reservoir can no longer be produced by natural reservoir pressure (i.e. primary recovery), or by water or immiscible gas injection (i.e. secondary, improved recovery methods (IOR) or pressure maintenance), EOR techniques may be considered. In general, as briefly referred to earlier, EOR techniques aim to stimulate oil flow by overcoming the physical, chemical and geologic factors that inhibit the production of the remaining hydrocarbons [4]. One of the most widely implemented EOR processes today is thermal recovery, which involves heating the oil bearing interval with steam or hot water to reduce the oil viscosity. Miscible gas injection (MGI) is another most widely used approach today, which is carried out by the injection of a high-pressure gas, such as carbon dioxide or hydrocarbon-associated gas, to sweep additional oil towards the wellbores by employing a number of in-situ mechanisms such as oil viscosity and IFT reductions. Chemical agents dissolved in water and injected into the reservoir can also improve the displacement properties during a water flood. Currently, various EOR projects executed around the world, as shown in Figure 1, account for only 3.5% (3 million barrels per day (MMbpd)) of the total world oil production (96 MMbpd) [5]. However, further application of these technologies has the potential to increase oil recovery from existing fields and new discoveries and alleviate oil supply shortage in the future [5, 6].

Figure 1.

Worldwide EOR projects contribute to global oil production [5].

The application of EOR techniques in Oman may be considered as a successful example of how such techniques may be used to boost oil production and achieve substantial enhancements in recovery. Over the past decade, a number of EOR projects in the Middle East (ME) have been executed. Among the ME countries, Oman leads the way mainly owing to its declining overall oil production rate [7] which has seen EOR to become a major strategy to meet target oil production from its existing fields [8]. In 2007, this country’s oil production declined to an average of 700,000 bpd. However, with the aid of EOR methods, the field operators have been able to increase the country’s overall oil production to its current level of nearly 1 million bpd. Miscible gas injection is one of the EOR techniques used in the country. The largest fields produced using EOR techniques in Oman and the indicative contributions made by such techniques in each field are depicted in Figure 2. The daily oil production rate from these fields with implanting EOR techniques varies between 40 and 80 thousand bpd (Mbpd). While without EOR, it was 3–45 Mbpd making such techniques the key driver of Oman’s oil production nowadays [9].

Figure 2.

The contribution of current EOR projects implemented in Oman oil fields [7].

2. Miscible gas injection (MGI)

The MGI is one of the most effective EOR methods used to enhance the production of light crude oil in the petroleum industry [10]. This method is PVT driven in which the injected gas (CO₂, associated gas (AG) or natural gas liquids (NGL)), in addition to helping with pressure maintenance, would mix with and alter the properties of the in-situ oil allowing the otherwise trapped oil to become mobile and easily displaced [11, 12]. During the miscible gas flooding, the injected gas would become miscible with the reservoir oil at or above the minimum miscibility pressure (MMP) of the reservoir oil (Figure 3). By definition, the MMP is the pressure at which the mass transfer and molecular interactions between the gas and oil intensify forcing the physical and chemical properties of the two phases to converge [14, 15]. In other words, upon reaching MMP, the gas acts as a solvent for the oil towards forming a single fluid phase (liquid) in the reservoir with the potential of effectively reducing the saturation of the remaining oil to near zero under ideal conditions [4]. During this process, the improved displacement efficiency of the flood is realised via three main mechanisms including substantial reduction in IFT (i.e. elimination of the interface between the two fluids and reduction of capillary pressure to zero), reduction of oil viscosity and oil swelling [16, 17]. The value of MMP depends on the reservoir temperature as well as the compositions of the injected gas and in-situ oil [18, 19].

Figure 3.

Development of miscibility of injected CO₂ in oil at miscible and immiscible pressures [13].

In general, the miscibility process of the crude oil-gas system may occur through two paths of multi-contact miscibility (MCM) and first-contact miscibility (FCM) [20]. The MCM would take place if the in-situ pressure is equal to MMP which, as discussed previously, is a critical property to be taken into account for designing an MGI process [21]. The MCM may develop gradually via a number of processes including vaporising gas drive, condensing gas drive and a combination of the two drives [15]. On the other hand, when the reservoir pressure is adequately high and well above MMP, FCM would take place in which the injected gas would develop miscibility with the in-situ oil at all proportions as soon as they are brought in contact. Since FCM would only occur at high enough pressures, depending on the type of injectant used, achieving this type of miscibility could be challenging.

The type of the injected gas used for MGI depends on the gas availability and reservoir conditions [2] with the common gases used around the world being CO₂, hydrocarbon gas mixture (AG, NGL), flue gas and N2 [20, 22]. Carbon dioxide, which has been most widely used in the United State, Canada and China [23], can achieve miscibility at relatively low pressures (when compared with other gases) and has a relatively high density (can be similar to oil). The latter can help to reduce the severity of gravity segregation and override which can negatively affect the sweep efficiency. The use of this gas for flooding can also help to reduce the global level of CO₂ emissions. However, some of the main challenges for a successful CO₂ flooding in general are the availability of CO₂ and corrosion in wells and surface facilities, which can result in considerable cost increases, in particular, for remotely located fields.

In the Middle East, the available CO₂ supply is limited to those associated with large industrial sources [24] which, when combined with the earlier mentioned issues associated with using this gas, has made its wide application limited. However, the hydrocarbon gas injection could be considered for MGI processes more widely for which the produced AG is usually readily available from the field itself or those close by. On the other hand, as mentioned earlier, unlike CO₂, conducting MGI using AG, depending on the gas composition, requires a relatively high pressure to achieve miscibility. To date, there have been three MGI projects (at either pilot- or field-scale) in the Middle East as reported in the literature [25]. Two of the projects involve miscible CO₂ injection and the other utilises AG injection. The Rumaitha Field in Abu Dhabi was the first pilot miscible CO₂ injection implemented in the region [25, 26, 27, 28]. The second pilot CO₂-EOR project has been implemented in Minagish Oolite Reservoir in west Kuwait [29]. The third project has been implemented in Field A located in the Harwell Cluster in southern Oman in which the field’s AG mixture (CH₄ enriched with light and heavy hydrocarbon fractions found in natural gas as well as considerable amounts of sour gases (3–5 mol % H₂S and 10–25 mol % CO₂)) is used for reinjection [30, 31]. As will be discussed in further details later with field case for improving the MGI process in this field.

Harweel Fields consist of a cluster of reservoirs deep within the tight carbonate oil-bearing rocks in the south of Oman in the Petroleum Development Oman (PDO) concession area, as shown in Figure 4. The figure also presents a geological cross section of the carbonate stringers, as encased in the Ara salt and the general geological setting of the area. The fields are expected to make a significant contribution to the Sultanate’s oil production over the coming 30 years. The reservoir rocks in these fields are more than half a billion years old (where the hydrocarbon deposits are among the oldest in the world) located at a depth of about 5 km, making them PDO’s deepest producing oil fields [33, 34]. As indicated earlier, the MGI in Field A, located in the cluster, has already begun in which the source of the injection gas is the Field’s AG [30]. The produced AG mixture is reinjected into the reservoir at high pressures of up to 45 MPa during which the injected gas develops miscibility with the in-situ oil under the reservoir’s high temperature (up to 377 K). The reservoir contains a light crude oil with a typical gravity of 42°API and a viscosity of 0.23 cP at reservoir conditions. It was initially estimated that up to 47% of the Field’s original oil in place (OOIP) could be recovered with the MGI process [33]. However, it has been realised since then that the presumed RF might not be eventually achievable due to the technical and operational challenges faced in this field, e.g. premature gas breakthrough and high degree of reservoir heterogeneity. As mentioned earlier, this chapter will be mainly focusing on addressing some of the technical challenges experienced during MGI in Field A and similar fields by proposing and testing a novel mobility control technique applicable to such a high-pressure and temperature environment.

Figure 4.

Geological cross section of the carbonate stringers (left) and an aerial overview of Harweel Fields in southern Oman (right) [32].

3. Challenges associated with MGI process

As with other EOR techniques, MGI can be economically expensive and technically challenging to implement [35]. For example, even when the injection gas is readily available on site (e.g. associated gas), gas processing, handling and compression as part of the expected gas recycling scheme can be costly [2, 36]. Full life cycle economics of a gas injection project, therefore, must be taken into account to justify its implementation. In addition, as an example, a technical challenge in achieving a profitable MGI is the instability of the oil displacement process in the reservoir mainly due to the expected unfavourable mobility ratio and possible gravity segregation whose effects may be intensified by the level of reservoir heterogeneity.

From a more general technical perspective, the efficiency of an MGI is controlled by the collective effects of several physical forces acting on the displacement front. These forces include the viscous forces that stem from viscosity contrast in the flood, gravity forces caused by fluid-fluid density differences, dispersive forces driven by the fluid concentration gradients and, finally, the capillary forces that have roots in the IFT between any immiscible fluids. The large differences in fluid viscosities can cause viscous fingers at the displacement front. If the vertical permeability in the reservoir is quite high, a pronounced density difference can cause gravity segregation. Both of the above have the potential to leave a large amount of oil unswept. The capillary and dispersive forces tend to enhance the fluid mixing but do not often overwhelm the viscous fingering [37, 38]. Therefore, the gravity and viscosity forces are the essential forces driving the instability of the oil displacement process during MGI [39]. Provided in the following two subsections are further details about the underlying mechanisms behind these two forces and how they may interfere with the performance of an MGI process. Possible mitigation strategies to lessen their effects will be outlined and adequately discussed in later sections of this chapter.

3.1 Viscous fingering

When a fluid is injected into a reservoir to displace another, there is almost never a collective perfect piston-like displacement across the entire reservoir interval. Especially in a gas flood, unstable displacement due to viscous fingering can lead to uneven or poor sweep, as depicted in Figure 5 [40]. Viscous fingering is generally defined as a hydrodynamic instability that occurs between two fluids of differing mobility/viscosity in the porous media that could lead to reduced sweep efficiency and early breakthrough [39, 41, 42]. The terms mobility, mobility ratio and that in a gas flood mobility ratio may be interchangeably used with viscosity ratio would be defined and discussed shortly. In MGI, there are several parameters that affect the viscous instability at the fluid-fluid interface including fluid viscosities, degree of miscibility, gas dissolution and exsolution and reservoir heterogeneity [39, 42, 43, 44]. However, the viscosity contrast and permeability heterogeneity are the two that mainly control the dynamics of the fingering phenomenon [37, 45]. The importance of mobility/viscosity ratio may be further realised after defining the mobility ratio ($\mathrm{M}$) as a widely used criterion to characterise and determine the occurrence and possible effects of viscous fingering.

Figure 5.

Effect of viscous fingering on the development of areal sweep efficiency against time (t) in a quarter of a five-spot flood pattern during gas flooding, (A) an unstable displacement with poor macroscopic sweep, (B) a stable displacement good with macroscopic sweep, () an injection well and (O) a production well [40].

As indicated by Eq. (1), the mobility of a fluid (${\mathrm{\lambda }}_{\mathrm{i}}$) in a porous medium may be defined as the ratio of effective permeability (${\mathrm{K}}_i$) and effective viscosity (${\mathrm{\mu }}_{\mathrm{i}}$) experienced by the fluid while flowing in the medium [40, 46, 47],

$${\mathrm{\lambda }}_{\mathrm{i}}=\frac{{\mathrm{K}}_{\mathrm{i}}}{{\mathrm{\mu }}_{\mathrm{i}}}$$E1

Subsequently, for any fluid-fluid displacement, such as an MGI, the mobility ratio ($\mathrm{M}$) can be simply defined as the mobility of the displacing fluid over that of the displaced fluid [40, 47]. For instance, Eq. (2) defines $\mathrm{M}$ for an MGI process where gas displaces the in-situ oil.

$$\mathrm{M}=\frac{{\mathrm{\lambda }}_{\mathrm{gas}}}{{\mathrm{\lambda }}_{\text{oil}}}=\frac{\raisebox{1ex}{${\mathrm{K}}_{\mathrm{gas}}$}\!\left/ \!\raisebox{-1ex}{${\mathrm{\mu }}_{\mathrm{gas}}$}\right.}{\raisebox{1ex}{${\mathrm{K}}_{\text{oil}}$}\!\left/ \!\raisebox{-1ex}{${\mathrm{\mu }}_{\text{oil}}$}\right.}$$E2

Where ${\mathrm{\mu }}_{\text{oil}}$ and ${\mathrm{\mu }}_{\mathrm{gas}}$ are the oil and injected gas viscosities, respectively. For a miscible displacement, where the gas solvent may displace the oil at irreducible water saturation and the effective permeability to both fluids may be considered to be the similar, Eq. (2) may be reduced to Eq. (3) [40]. Furthermore, during gas flooding, due to the large viscosity contrast between the gas and the in-situ oil, viscosity ratio may be considered adequate for qualitative evaluation of viscous instability in the flood [40, 45, 48]. Therefore, for the purpose of qualitatively characterising the effect of viscous fingering on the performance of an MGI process, the viscosity ratio may be used interchangeably with the mobility ratio [40].

$$\mathrm{M}=\frac{{\mathrm{\mu }}_{\text{oil}}}{{\mathrm{\mu }}_{\mathrm{gas}}}$$E3

During its development, the severity of viscous fingering increases with increase in the mobility/viscosity ratio of the fluid system. If $\mathrm{M}$ is larger than unity, the displacement becomes unstable resulting in the development of viscous fingers. Therefore, to achieve a stable displacement, where possible, the viscosity of the displacing fluid may be increased or its effective permeability reduced until the value of $\mathrm{M}$ approaches unity or less. For instance, if the injected gas viscosity is increased, the gas mobility may be suppressed. Hence, the severity of the viscous fingering and the chance of developing premature breakthrough can be reduced, resulting in improved displacment efficiency. Figure 6 demonstrates the effect of mobility ratio on the area sweep efficiency of an MGI process as reported by Habermann [40]. As can be seen from the figure, when $\mathrm{M}=1$, the ultimate areal sweep reaches as high as 99%, however, if $\mathrm{M}$ increases to 38.2, the areal sweep would decrease by more than 20%. The physical development of viscous fingers as the mobility ratio changes for the cases presented in Figure 6 is demonstrated by the diagrams included in Figure 7. As can be seen in this figure, the displacement is characterised as stable if the value of $\mathrm{M}$ is one or lower. The effect of $\mathrm{M}$ as demonstrated through the above sweep values and Figures 6 and 7 was for a homogeneous porous system. The presence of permeability heterogeneity would also make considerable contribution towards initiating and development of viscous fingering [41]. A high permeability layer would present a preferential flow path for the fingering of the injected gas causing early gas breakthrough and a low overall oil recovery factor [28, 29, 49].

Figure 6.

Areal sweep efficiency as a function of mobility ratio and pore volumes of displacing phase injected for an MGI process [40].

Figure 7.

Viscous fingering growth for different mobility ratio and injected pore volume [40].

3.2 Gravity segregation

As indicated earlier, another possible major technical challenge faced by an MGI process that influences the vertical sweep efficiency is the gravity segregation or gravity override. The injected gas (such as CO₂ or hydrocarbon gas) is usually less dense than the in-situ oil which may lead the injected gas to flow upwards, rather than lateral, forming a gravity tongue [49, 50]. Such a behaviour, similar to unfavourable mobility ratio, would result in early gas breakthrough and reduced vertical sweep efficiency in horizontal MGI processes as depicted in Figure 8. The effect of gravitational force on an MGI process has been studied by Moissis et al. [51] using numerical simulation. They found two dimensionless parameters of relevance, the dimensional density difference ($∆\mathrm{\rho }$):

Figure 8.

(A) Reservoir heterogeneity due to permeability variation versus depth in field a located in South of Oman, (B) example effect of possible gravity segregation on vertical sweep efficiency.

$$∆\mathrm{\rho }=\frac{{\mathrm{\rho }}_{\mathrm{o}}-{\mathrm{\rho }}_{\mathrm{g}}}{{\mathrm{\rho }}_{\mathrm{g}}}$$E4

and the dimensionless gravity number (${\mathrm{N}}_{\mathrm{g}}$):

$${\mathrm{N}}_{\mathrm{g}}=\frac{\left({\mathrm{\rho }}_{\mathrm{o}}-{\mathrm{\rho }}_{\mathrm{g}}\right)\mathrm{g}{\mathrm{K}}_{\mathrm{e}}}{\mathrm{q}{\mathrm{\mu }}_{\mathrm{o}}}$$E5

where ${\mathrm{N}}_{\mathrm{g}}$ represents the ratio of gravity forces to viscous forces, ${\mathrm{\rho }}_{\mathrm{o}}$ and ${\mathrm{\rho }}_{\mathrm{g}}$ are the oil and gas densities, respectively, ${\mathrm{K}}_{\mathrm{e}}$ is equivalent permeability, ${\mathrm{\mu }}_{\mathrm{o}}$ is the oil viscosity, q is the flow rate of the less viscous fluid in the porous medium of interest, and g is the gravitational acceleration. The simulation results obtained by Moissis et al. [51] show that the gravity force does not influence viscous fingering growth at small ${\mathrm{N}}_{\mathrm{g}}$ values indicating the dominance of the viscous forces under such a condition [51]. As ${\mathrm{N}}_{\mathrm{g}}$ increases to larger values, the gravity force begins to influence the growth rate of viscous fingering in the upper part of the porous medium. For sufficiently large ${\mathrm{N}}_{\mathrm{g}}$ values, gravity override completely dominates the displacement where, eventhough the viscous fingering can still occur near the gravity tongue, it is suppressed in the bottom part leaving this part of the porous medium completely unswept. Overall, as may be expected, with increase in ${\mathrm{N}}_{\mathrm{g}}$ the gas breakthrough occurs earlier reducing the overall oil recovery [51].

Further interplay between the gravity and viscous forces towards controlling the efficiency of a gas flood may be deduced by further scrutiny of Eq. (5). Controlled by the magnitude of ${\mathrm{N}}_{\mathrm{g}}$, the effect of the gravitational force is expected to be even larger at high flood viscous ratios because the gravity to viscous forces ratio is inversely proportional to the viscosity of the fluid available in the porous medium. At the beginning of the flood, as defined by Eq. (5), this ratio is equal to ${\mathrm{N}}_{\mathrm{g}}$. However, as the displacement proceeds and more of the less viscous gas enters the porous medium at constant flow rate, the gravity to viscous forces ratio begins to increase resulting in more sever gravity override. Such an effect would be more pronounced in the case of floods characterised by a high viscosity ratio [51].

Scott [50] has suggested to combat the gravity segregation by adjusting the density of the miscible gas injected as part of an MGI. For example, the pressure within the formation can be maintained high enough so that the density of the injected fluid approaches that of the reservoir oil. However, for measurable outcomes in general, the density of the miscible fluid should be maintained within about 10% of the density of to be displaced in-situ oil [50]. Furthermore, this technique may be proven difficult and impractical if large injection volumes are required to maintain the reservoir pressure. Scott [50] has also indicated that the density adjustment may be obtained by injecting carbon dioxide or intermediate natural gas fractions (C₂H₆, C₃H₈ and C₄H₁₀). Carbon dioxide in its supercritical state is capable of exhibiting a density greater than that of the reservoir oil [50]. However, hydrocarbon gases alone may not normally achieve a density equal or close to that of the resident crude oil under typical reservoir conditions; therefore, sever gravity override could still occur. Another technique to increase the density of injected gas is the use of chemical additives; however, to date, suitable and viable chemical additives to be used for this purpose are yet to be developed [52]. As suggested in the literature, the mitigation of the gravity segregation can be possibly achieved by mobility or conformance control [52].

4. Gas mobility control techniques

As mentioned above, the major challenge with the ongoing MGI flooding in the oil field is the unfavourable mobility ratio. This challenge can be addressed by the implementation of several approaches as proposed in the literature (although mainly for CO₂ flooding) including water alternating gas flooding (WAG) [53, 54, 55], foam flooding [56, 57, 58, 59, 60, 61, 62] and increasing the gas viscosity using the addition of polymers as thickening agents [52, 63, 64, 65, 66, 67, 68, 69, 70, 71]. The common main objective of these approaches would be to control the gas mobility effectively and, as a result, increase the sweep efficiency of the gas flooding [72]. Further technical details about each of the abovementioned techniques are provided in the upcoming subsections of this chapter.

4.1 WAG process

As an EOR method, the WAG process is defined as the injection of a gas (e.g. CO₂ or hydrocarbon gases) and water alternately into an oil-bearing formation (Figure 9). The WAG injection scheme was initially proposed by Claudle and Dyes in 1958 [55] to improve sweep efficiency during gas flooding. Their study showed that this injection scheme would result in the reduction of the relative permeability to the gas phase and suppress its mobility. In other words, the WAG would improve the sweep efficiency of the injected gas by using water to control the gas mobility and stabilise the displacement front. In general, depending on the MMP of the in-situ oil, this technique can be classified into two categories of miscible and immiscible WAG displacements [73]; however, as reported in the literature, the majority (79%) of the historical WAG field applications fall into the miscible category [74, 75]. In some recent field applications, in an injection scheme similar to WAG, the produced gas has been reinjected through water injection wells to improve the oil recovery and help to provide pressure maintenance [76]. The majority of the WAG injection projects are found onshore (88%), and few others are reported to have been implemented in an offshore environment (12%) [75].

Figure 9.

A typical WAG injection process as an EOR method that involves the injection of gas and water alternatively into an oil reservoir.

In general, there are a number of factors affecting the performance of the WAG process including the degree of reservoir heterogeneity, in-situ fluid properties, injection technique, miscibility conditions and other WAG parameters such as the individual gas and water slug sizes and their size ratio (WAG ratio), number of injection cycles and injection rates [77, 78, 79]. Similar to other EOR processes, the WAG flooding has a number of advantages and disadvantages that will be presented and discussed below.

4.1.1 The mechanisms and factors influencing WAG flooding

During WAG injection, the improved recovery is not often achieved through modifying the fluid properties of each of the injected phases, it rather tends to combine the advantages of each of the continuous gas or continuous water floods through creating a synergism between the in-situ flow properties of the two phases if injected on their own. Overall, when WAG injection is applied in an oil reservoir, it may yield favourable outcomes through several mechanisms [80]. Firstly, the injection process may help to maintain the reservoir pressure above the MMP of the oil resulting the achievement of the more desirable miscible flood. Secondly, the injected gas mobility is reduced by supressing the gas relative permeability in any existing preferential flow channels. This is achieved by the increase in water saturation in these zones and therefore reduction in gas saturation suppressing the possibility of gas channelling and viscous fingering [52, 81, 82]. Thirdly, in the case of a miscible flood, the excellent microscopic displacement efficiency of the miscible gas flooding is put into use across a larger portion of the reservoir by the mobility control and conformance control provided by the water phase, leading to higher oil recovery. Lastly, compared with a continuous gas injection process (e.g. continuous MGI), the WAG flooding decreases the amount of the gas needed for injection leading to possible improvement in the economics of the overall flooding process. Considering the collective advantages mentioned here, the WAG injection process may become a viable option for some fields around the world.

Laboratory experiments have been used to study the effect of various parameters such as WAG slug size, WAG ratio (tapering), number of WAG cycle and injection flow rate on the performance of WAG [53, 54, 73, 83, 84, 85, 86, 87, 88, 89]. In general, these parameters show strong effects on the oil recovery trends of a WAG injection. It has been found that, in general, decreasing slug size and WAG ratio and increasing the number of WAG cycles would lead to a higher oil recovery [53, 80]. However, the optimum WAG ratio often depends on the wettability of reservoir rock, in-situ fluid properties and the type of gas being used as well as economic evaluations [53]. The optimum WAG ratio is considered as a key parameter for the successful implementation of a WAG injection process. A high WAG ratio may lead to an excessive water injection into the reservoir giving rise to the water blocking effect where the water phase would surround the trapped oil at low permeable zones and reduce accessibility by the injected gas decreasing the overall oil recovery. On the other hand, if the ratio is too low, the conformance control of the WAG flood would be lost and the injected gas would penetrate through the reservoir very fast under the effect of unfavourable mobility ratio and lead to early breakthrough. Overall, the experimental results have demonstrated that the WAG process may help to suppress viscous fingering and lead to increased oil recovery in gas flooding [74, 75].

4.1.2 Challenges of WAG flooding

The WAG injection has been successfully applied in several oilfields worldwide demonstrating that it could result in considerable incremental oil recovery at the field scale (5–10% of oil initially in place (OIIP)) [90]. However, some published literature also indicates that some of the field-scale WAG processes have not reached their expected target recovery factors, especially in naturally fractured, highly permeable and highly heterogeneous reservoirs [75]. Furthermore, the field-scale implementation of this technique has also helped to identify a number of challenges that may be faced by the field operators. Such challenges are presented and discussed below by first dividing them into the two categories of operational challenges versus those of subsurface reservoir related.

4.1.2.1 Operational issues

A numbers of operational related issues have been reported in the literature including [74, 75, 90].

4.1.2.2 Reduced injectivity

The ability to inject the required amounts of gas and water through the injection wells is critical towards achieving the desirable WAG performance. Reduced injectivity can result in a pressure reduction in the reservoir, which may impact on, for example, miscibility, performance of the displacement and the eventual production yield. This issue may be caused by changes in the phase relative permeabilities and/or near wellbore formation damage. In general, the field trials have shown that the reduced injectivity may be experienced for the water injection rather than the gas injection stage during the alternating injection of the two phases [75, 90].

4.1.2.3 Corrosion

Corrosion problems have been reported in many projects that have involved WAG injection. Often such issues have been encountered because the pre-existing injection and production facilities were not initially designed to handle the WAG injection process. Six fields are reported to have experienced corrosion problems, mainly on the injection facilities. The existing case studies indicate that in most cases, such problems could be adequately addressed by using corrosion-resistant materials in the manufacture of equipment, coating the flow-lines and chemical treatments [75, 90].

4.1.2.4 Asphaltene, scale and hydrate formation

Asphaltene and scale precipitation and hydrate formation are among other problems that have been experienced in various WAG field trials. These problems would lead to production disturbance and even flowlines blockage which may increase the operating costs of a WAG process. Three fields (East Vacuum, Wertz Tensleep, Mitsue) have experienced asphaltene precipitation, and two fields (Ekofisk and Wasson Denver) have reported the formation of hydrate in the injection wells due to the low temperature in the injectors or cold weather at the wellhead. Some of these problems could be resolved by chemical treatments [75, 90].

4.1.2.5 Subsurface reservoir issues

Besides the operational problems discussed above, there are also a number of issues related specifically to the subsurface and fluid flow in the bulk of the reservoir presenting challenges for the WAG implementation:

4.1.2.6 Premature gas breakthrough

Unexpected early gas breakthrough has been reported in several WAG field applications despite the fact that WAG is often implemented to combat this issue in particular. The main cause for this problem has often been inadequate characterisation of the reservoir, poor design of the WAG process or limitations imposed by the existing versus required infrastructure (e.g. limited number of injection/production wells). Regardless of the cause, early gas breakthrough would often occur due to gas channelling through highly permeable layers or gravity override [91, 92]. The early gas breakthrough leads to loss of reservoir pressure and lost miscibility in a miscible WAG project [93, 94]. As reported in the literature, five oil fields (University Block 9, Juravlevsko-Stepanovskoye, Lick Creek, Caroline and Snorre) have experienced this problem because of gas channelling [93, 95, 96, 97, 98]. Unfortunately, this problem is hard to resolve as once occurred, its root causes (as mentioned at the beginning of this paragraph) are difficult to address. However, adequate reservoir characterisation before the implementation of this mobility control technique can be helpful in avoiding unexpected early gas breakthrough [75].

4.1.2.7 Oil trapping

Several studies have demonstrated the occurence of oil trapping by water in the WAG flooding [99, 100, 101, 102]. This phenomenon is also referred to as water blocking [102]. During the WAG injection, the injected mobile water traps/encases the residual oil which then becomes difficult for the gas phase to access and mobilise. Therefore, a high residual oil saturation may be left behind in the reservoir even after WAG flooding. It has been determined that rock wettability and WAG ratio can strongly affect the oil trapping with being more sever in the case of water-wet rock formations or high WAG ratios [80, 99, 100, 103].

4.1.2.8 High water production

The injection of large amounts of the water into the reservoir (i.e. high WAG ratio) can cause high water saturation [104] leading to excessive water production and, hence, reduced oil recovery [105]. In addition, the excessive water production would require additional water treatment capacity that brings about additional costs impacting on the project economics [103].

4.2 Gas foam flooding process

4.2.1 Gas-foam generation and foaming agents

Gas-foam injection is another approach to combat the conformance and mobility limitations encountered in an MGI process. Furthermore, this technique may also bring about some of the advantages of the chemical EOR due to the chemical additives required for foam stabilisation and generally better foam generation. The foam flooding was first introduced by Bond and Halbrook in 1958 to show that the foam generated by the injection of an aqueous surfactant solution and miscible/immiscible gas could increase sweep efficiency [106]. With the favourable results obtained from the above study in the subsequent years, it was proposed to use foam injection as a means of gas mobility control. However, the concept did not become widely known and immediately adopted due to the lack of understanding of mobility control mechanisms behind the foam flooding [107].

In the context of fluid flow in porous media, a foam is generally defined as a gas-liquid mixture where the liquid phase exists as a continuous wetting phase in the rock, whereas all or parts of the gas form the discontinuous phase surrounded by a thin liquid film or Lamellae [60]. According to the literature, the research conducted in the area of gas foam flooding mostly relates to CO₂-EOR because the required chemicals are much easier to dissolve in CO₂ towards the generation of a CO₂ foam at reservoir conditions [72]. A gas foam may be stabilised by the addition of effective surfactants, which contain a hydrophobic and hydrophilic segment [72]. Surfactants then can be either water-soluble or CO₂-soluble [60, 108, 109]. The selection of surfactant depends on the reservoir conditions. If the reservoir condition is suitable for a surfactant to be soluble in the injected gas, then injection of water with the surfactant can be eliminated [110]. Numerous CO₂ soluble surfactants have been experimentally identified [56, 109, 110]. For example, the hydrocarbon-based ethoxylates surfactant has been suggested by Scheievelbein et al. as a CO₂ foam agent instead of using a water-soluble surfactant [110]. The other reported surfactant products include Tergitol TMN-6, oligo (vinyl acetate), poly(ethylene glycol) 2,6,8-trimethyl-4-nonyl ethers, and ethoxylated amine surfactant [111, 112, 113, 114, 115]. For miscible hydrocarbon gas flooding, only water-soluble surfactants can be used as the foaming agent because no effective surfactant directly soluble in hydrocarbon gases for gas-foam generation has been reported in the literature [57]. Nine water-soluble surfactants have been identified for foam generation with hydrocarbon solvents, including alkanolamides, amine oxides, betaine derivatives, ethoxylated and propoxylated alcohols and alkylphenols, ethoxylated and propoxylated fatty acids, ethoxylated fatty amines, fatty acid esters, fluorocarbon-based surfactants and sulfate and sulfonate derivatives [57]. As the temperature increases, most of the water-soluble surfactants become less soluble in water. Therefore, it may be necessary to evaluate the surfactant solubility in either CO₂ or water for application in high-temperature reservoirs [72, 116].

The foam used for gas foam flooding may be generated in several ways as discussed in the literature. It may be formed within the target porous media by alternating injection or co-injection of a suitable surfactant and gas (CO₂ or hydrocarbon gas mixture). In the case of CO₂ foam flooding, the foam can be formed when a surfactant is dissolved into CO₂ (usually in supercritical state) and then injected into the porous media, without requiring the injection of a liquid slug [59]. The foam can also be generated at the wellhead by the simultaneous injection of the gas and surfactant solution. Then, as the foam leaves the wellbore, it could be re-formed and strengthened as it enters the micropores of the reservoir rock [72].

As a gas foam enters a rock formation, it would need to propagate through the entire formation suppressing the high gas mobility for the whole duration of the flood. However, the injected foam is not often thermodynamically stable under in-situ conditions, and therefore, the two-phase foam system may collapse with time. On the other hand, as mentioned earlier, the passage of the fluids through the porous rock formation could result in the regeneration of the foam due to shearing effects applied by the micron-sized tortuous pores and pore channels [117]. Therefore, in order to have an effective foam for mobility control, the rate of in-situ foam generation would need to be equal to or greater than the rate of its decay [72]. In general, the foam propagation at the large reservoir scale and the foam stability are the main challenges faced by the gas foam flooding technique.

4.2.2 Main mechanisms of gas-foam flooding

A gas foam may be used as part of an EOR scheme for two purposes [57]. Firstly, it can be designed to reduce the gas mobility to a level that is comparable to or even less than that of the displaced oil so that the gas viscous fingering and channelling can be effectively suppressed. Thereby the areal sweep efficiency could be improved considerably. However, it is worth noting that the reduction level in the foam mobility has to be optimised and controlled to avoid the prohibitive pressure drop in the reservoir caused by extremely low foam mobility. Therefore as a compromise, a weak and modest foam may be generated by varying the surfactant concentration in a gas-foam injection [118]. The second possible purpose of using a gas foam is for conformance control or blocking of a thief gas channel to divert the injection fluids away from it and into other unswept lower permeability oil-rich zones to mobilise the otherwise bypassed oil [72, 109]. Typically, this can be achieved by the alternating injection of an aqueous solution with a high concentration of a surfactant [57]. The high concentration of surfactant then generates a strong foam that would flow in the highly permeable or thief zone [118] resulting in the diversion of the gas flow into the lower permeability zones.

The enhanced recovery of a gas foam injection is usually achieved through a number of different mechanisms as summarised and briefly discussed below.

4.2.2.1 Stabilising the displacement front

The efficiency of a fluid-fluid displacement in a porous medium is in general controlled by the three gravity, viscous and capillary forces [60, 119]. Therefore, the manipulation of these forces can result in enhanced recovery. Concerning the application of a gas foam during a gas flooding process such as MGI, the mobility control and, therefore, stabilisation of the flood front may be achieved by the higher viscosity and reduced relative permeability of the gas foam both relative to the case of injecting the gas on its own. Typically, these effects may be achieved through two mechanisms [60]. The first mechanism is related to the movment and re-arrangement of bubbles due to the local gradient in the surfactant concentration and, therefore, the interfacial tension. The surfactant movement within the liquid film (Lamellae) lowers the surface tension between the two phases (liquid and gas) that slows down the bubble motion and causes an increase in the gas phase effective viscosity [120, 121, 122]. The second mechanism that reduces the gas-foam mobility is gas trapping [123, 124]. As the foam injected and/or formed in a porous medium, as also indicated earlier, it prefers to flow through highly permeable and porous zones, while the low permeability areas with small pores remain occupied by the wetting phase [125] (Figure 10). Thus, the gas bubbles may enter and become trapped in the intermediate size pores, where a large fraction of foam bubbles are immobilised due to the high enough capillary pressure [59]. Nguyen et al. [126] found that the amount of trapped gas in this form is governed by several factors, such as the foam texture, pore geometry and pressure gradients. The blocked intermediate size pores decrease the pore volume available for the gas foam to flow through, thus the reduced relative permeability and suppressed gas-foam mobility [60].

Figure 10.

A micro-pore illustration of foam flow and gas trapping in the porous media. The cross-hatched spaces represent the solid grains, and the dotted spaces indicate the wetting liquid [60, 117].

A gas foam can help to combat gravity segregation too [60]. Figure 11 demonstrates the effectiveness of a CO₂ foam towards stabilising the displacement front in the X-ray CT scanned core-flooding experiments conducted by Wellington and Vinegar [127]. As can be seen from the left-hand side images, the researchers found that CO₂ injection alone would lead to the formation of a gravity tongue, whereas the right-hand side images show that the CO₂-foam injection prevented the gravity and viscous instabilities towards the uniform displacement of the in-situ oil.

Figure 11.

X-ray CT scan images for (A) a CO₂ miscible flood (blue) in a core saturated with oil (red) and residual brine (yellow) and (B) CO₂-foam flooding (blue) in a core saturated with oil (red) and a surfactant solution (yellow) [72, 127].

Overall, based on the discussion presented so far, a gas foam would not change the gas phase density but exhibit its effectiveness by suppressing the gravity and viscous forces, leading to stabilisation of the displacement front.

4.2.2.2 Reducing the capillary force

Capillary pressure is usually held responsible for the bulk of the entrapped oil (often non-wetting phase) in rock formations. That is why Zhang et al. [128] point out that the removal of the trapped crude from a reservoir rock needs ultra-low interfacial tension through an emulsification mechanism. The capillary number as set out in Eq. (6) defines the ratio between viscous and capillary forces acting on a displacement. The lower the interfacial tension (low capillary forces), the higher the capillary number and, therefore, the more dominant would be the viscous forces resulting in higher recoveries.

$${\mathrm{N}}_{\mathrm{c}}=\frac{\mathrm{K}∆\mathrm{P}}{\text{σLcosθ}}$$E6

Where ${\mathrm{N}}_{\mathrm{c}}:$ capillary number, dimensionless, K: absolute permeability of the porous medium, $∆\mathrm{P}$: pressure drop along the porous medium, $\mathrm{\sigma }$: the interfacial tension between the two fluids, L: length of the porous medium, and $\mathrm{\theta }$: contact angle.

Once during foam injection, the surfactant in the injected slugs proceeds through the porous rock, different interactions occur at oil, foam and rock interface [129] leading to ultra-reduction of the interfacial tension between the oil and water resulting in the formation of an oil-in-water emulsion. Accordingly, the capillary force reduces to near zero allowing the emulsion to move through the pore throats (Figure 12) resulting in enhanced recovery [60].

Figure 12.

(A) A high interfacial tension results in large capillary force, which prevents an oil droplet from crossing through the downstream pore throat, (B) ultra-low interfacial tension leads to near zero capillary force, which allows the oil droplet to flow through the pore throat and be produced [59].

4.2.2.3 Altering the rock wettability

The wettability of a porous rock formation is an essential factor to be taken into account in its characterisation because of its impact on the bond between oil and rock, the multiphase flow behaviour and distribution of fluid saturations in the reservoir [108]. Wettability alteration may occur in the foam flooding process due to the interactions between the surfactants used and the rock surface [59]. According to Eq. (6), the capillary number can also be increased by changing the contact angle, which means altering the rock wettability. As mentioned before, increasing the capillary number can result in lower residually trapped oil [59]. The importance of wettability alteration is not often considered in both experimental and simulation work, because of the erroneous assumption that all rocks remain water-wet during foam injection, and it is difficult to quantify the reservoir wettability in a meaningful and repeatable manner [130]. Although Charanjit and Bernard [131] do not agree that wettability may change due to a foaming agent, in a number of other studies, wettability alteration due the surfactant adsorption has been reported to change porous rocks from oil-wet to water-wet [131, 132, 133].

Overall, the foam injection process can enhance the oil recovery by mobility control in combination with ultra-low IFT and possible alteration of the rock wettability due to the presence of surfactant in the foam.

4.2.3 Challenges and field application for gas-foam EOR

The application of the gas-foam process in oil fields for mobility control has shown to be technically and economically challenging. This is because the effectiveness of a gas foam flooding highly depends on several parameters such as oil type, oil and water saturation, brine salinity and pH, surfactant formulation and concentration, reservoir heterogeneity, capillary pressure and gas flow rate [134, 135]. For example, a high oil saturation and low water saturation in the presence of light oil may cause the foam to decay and collapse [136]. As a consequence, before applying a foam EOR process, it is extremely important to gain a comprehensive understanding of the physical aspects of the process and how the foam may flow and behave once injected through a porous rock formation. The two main broad technical and operational difficulties in applying foam EOR at the large field scale are described below.

4.2.3.1 Foam stability and propagation

According to the numerous studies conducted to date, it may be difficult to achieve a stable and reliable foam generation under the harsh reservoir condition (high temperature and high salinity) often encountered and also control the propagation of the foam over large distances in the reservoir scale. Under high salinity and high temperature, the gas foam cannot be stabilised with the surfactant, because under such conditions the surfactant solubility in water or CO₂ would be reduced resulting in its precipitation onto the rock surface [115, 137]. In addition, with the loss of the surfactant, the necessary ultra-low IFT may not be achievable [138, 139]. The levels of oil and water saturations are other parameters that affect foam stability. Mayberry and Kam [140] examined the foam strength at different oil and water saturations. Their experimental results indicate that the apparent foam viscosity is significantly reduced at oil saturations greater or lower than a critical oil saturation. The presence of the oil in the formation has a strong effect on the foam rupture and breakdown due to the interactions occurring between the foam lamellae and the oil phase [141]. Law et al. [142] also found that foam is degraded if the oil saturation exceeds critical foaming oil saturation of the surfactant. It is also shown that the light and less viscous oils are more destructive to foam stability than heavy oils [136]. Moreover, the reservoir water saturation is crucial for the foam stability. When a foam is injected at water saturations below a critical value, which corresponds to a limiting capillary pressure, the foam may begin to coalesce and dry out. It should be noted that below the critical water saturation and above the critical oil saturation, the foam is eliminated [56, 136].

4.2.3.2 Scale-up from pilot to full field application

There have been several CO₂-foam trials performed since 1990 mainly in the United States [143, 144, 145]. Some of these, such as that performed in Joffre Viking oil field, were unsuccessful, because of the foam propagation control failure [146]. On the other hand, a few of the pilot tests have been successful, including that conducted in the Rock Creek Field [147] and Northward-Estes Field. In Northward-Estes Field, it was observed that the foam injection led to reduced CO₂ injectivity by 40–85% [143]. Several other pilot studies were conducted using CO₂ foam in East Vacuum Grayberg/San Andreas Unit [148] and SACROC Field in West Texas [149, 150], all of which proved that CO₂ mobility could be reduced and oil production increased. However, a transition from pilot scale to a wider field application has not been implemented due to various challenges such as issues associated with chemical supply and transportation, processing and separation of the produced fluids, offshore supply and also safety concern [151, 152, 153, 154].

4.3 Direct gas thickeners

The use of direct gas thickeners is another method that brings together the combined possible advantages of using chemical additives and MGI. This technique has been recognised as a “game-changing technology” for mobility control, which was first reported in late 1960 [68, 69, 72, 155]. Since then, the interest in synthesising and designing affordable gas thickeners has been carrying on steadily. However, until now the term “gas thickener” has been used in laboratory investigations only, and its effectiveness has not yet been verified in any field-scale applications around the world. In general, this technique involves increasing the injected gas viscosity by directly adding chemicals that exhibit good solubility in common supercritical fluids (SCF) used for EOR such as CO₂ or hydrocarbon solvents. Chemicals that may increase the viscosity of an SCF include entrainers, conventional oligomers and polymers and small associating compounds [156]. In an ideal situation, chemical compounds need to be readily soluble in the dense CO₂ or hydrocarbons solvents and insoluble in both crude oil and brine at reservoir conditions [52]. It should be noted that the thickening level of the gas is not expected to affect its injectivity because this solution would exhibit a shear-thinning behaviour near the wellbore which facilitates the mobility of the thickened gas in this area but, the mobility ratio of the gas flood would be improved in the bulk of the rock formation leading to enhanced recovery (Figure 13). In addition, the thickened gas would uniformly flow into different zones, allowing the gas to also mobilise the trapped oil in the low permeable zones. In other words, this technique can be applied as a way of improving the flood conformance and mobility control as illustrated in Figure 13.

Figure 13.

Simplified illustration of a thickened gas flooding.

Two fundamental strategies have been introduced in the literature to increase the injected gas viscosity [157].

Direct dissolution of polymers: In this strategy, a gas thickener is typically a synthesised or identified polymer or oligomer that promotes attractive interactions and dissolution with gas molecules. However, it has been recognised that the use of polymers with extraordinary molecular weight for the above purpose would be quite challenging since most of the SCF fluids are very stable and weak solvents due to the very low dielectric constant, no dipole momentum and sometimes low density. The intermolecular attractions between the polymer molecules are typically strong enough at ambient temperature so that even stirring them would be insufficient to attain dissolution. Therefore, they may only dissolve in a gas solvent at elevated pressure and temperature because such conditions give rise to the intermolecular forces between the solvent-polymer segments or solvent-solvent or polymer segment-segment pairs in the solution given by difference on the free volume between the polymer and gas solvent and the free energy [158]. In addition, heat may be required to weaken intermolecular interactions between the polymer molecules (e.g. hydrogen bond) [158]. Another approach for obtaining high solubility of the polymer in solvents is to introduce associating or functional groups in the polymer’s molecular chains, for example to become CO₂ philic, and therefore assist the polymer dissolution in the solvent [159, 160]. Some examples of the associated polymers include polyvinyl acetate (PVAc), oligo (3-acetoxy oxetane), poly [(1-O-(vinyloxy) ethyl-2, 3, 4, 6-tetra-O-acetyl-β-D-glucopyranoside)] and amorphous polylactic acid [161, 162]. Once the molecules of the polymer are dissolved in the solvent, the intermolecular/intramolecular association may occur which would result in an increased solution viscosity. Some of the polymers can increase the solvent viscosity significantly by simply changing the thickener concentration or by twining their molecular structure like a hair between different polymer chains [163].

Dissolution of small molecules (self-assembling and associating compound): The second strategy is focusing on the design of small-molecules material that contains a self-assembling and associating compound to form a viscosity-enhancing supramolecular network structure in the solution. Such a material contains an associating group composed of a solvent philic segment that facilitates dissolution and one or more solvent-phobic segments that would induce the intramolecular association with neighbouring molecules, thereby molecular association establishing a viscosity enhancement for the solution, but its impact on viscosity could be minimal [72]. The small-molecules thickeners have shown little success to thicken CO₂ and light alkane solvents primarily because these are regarded as weak solvents for the ionic and polar associating compounds that are commonly composed into the small-molecules thickeners [71, 157].

Overall, a polymeric or small-molecules compound thickener capable of dissolving into CO₂ or light hydrocarbon solvents has to be identified to increase the solution viscosity under typical field conditions. The ideal chemical additives are those that can effectively increase the viscosity of the injected gas very close to that of the crude oil. Furthermore, a viscosified gas used for EOR has to be transparent and single phase rather than opaque viscous solution in order to be capable of flowing through micro-pore throats in rock formations [157]. A viscosified gas with the above-described desirable characteristics used for an MGI process can suppress the gas mobility in the reservoir reducing the severity of viscous fingering and the chance of developing premature gas breakthrough and high production gas oil ratio (GOR). As a result, the sweep efficiency would be improved for the gas flood. Various studies conducted over the past several decades have resulted in successful laboratory-scale progress in thickening of CO₂ and NGL (natural gas liquefied). The successful CO₂ thickeners include the fluoroacrylate-styrene copolymer polyFAST and poly(dimethylsiloxane)-toluene solutions [160]. These two thickeners have been found to be capable of increasing the CO₂ viscosity by approximately 10 and 4 fold, respectively, at dilute concentrations [160]. A drag-reducing agent (DRA) poly(α-olefin) was presented as the most significant thickener that can increase the viscosity of the NGL [164].

4.3.1 Challenges and opportunity for gas thickeners

The use of gas thickeners has the potential to eliminate many of the earlier mentioned challenges and difficulties associated with WAG and gas-foam injections. However, the discovery of inexpensive polymers or small-molecules materials soluble in CO₂ or alkane solvents has so far been a major challenge. Furthermore, the performance of none of the identified or synthesised thickeners has been verified in even a field pilot test yet.

In general, the following challenges have hindered the identification of effective thickeners that could be used for a gas flood:

Thickener solubility: The attainment of adequate solubility has been the primary obstacle in finding viable thickeners because most of the designed and identified polymers for the CO₂ and hydrocarbon gases exhibit extremely low solubility unless a large volume of a co-solvent (e.g., 10–15 wt% toluene) is added. The reason behind this problem is that CO₂ and alkane gases are poor solvents for extremely high molecular weight, polar and ionic-associated groups that are composed in small-molecules thickeners. The alkane gases (methane and ethane) do not have dipole or quadruple moments, so the dispersion interactions are dominant with these solvents. Thereby, alkane gases would not be suitable SCF solvents unless the density of these solvents is increased considerably by increasing the system pressure. Unlike alkane gases, CO₂ has a substantial quadrupole moment that induces quadruple interaction as the temperature is low [158]. In addition, CO₂ acts as a Lewis acid for the polymers containing oxygen [165]. In general, as mentioned before, a polymeric CO₂ thickener needs to contain a CO₂-philic function group that facilitates the polymer solubility and CO₂-phobic function group that promotes intermolecular associations to enhance the viscosity [166]. To date, solubility remains a key major challenge in the identification of an inexpensive thickener for CO₂ and hydrocarbon solvents.

Cost and environmental persistence: The high price and environmental issues are other challenges that impede the use of the identified or developed thickeners to date in field applications. In fact, most of such thickeners are unaffordable and/or unavailable in large enough quantities. The requirment of an organic co-solvent to obtain the necessary dissolution levels further adds to the cost. Moreover, some of the developed thickeners, such as fluoroacrylate-styrene copolymers (polyFAST) and semi-fluorinated trialkyltin fluorides, are fluorinated compounds that contain Fluorine. These thickeners have been identified as the best thickeners for CO₂ and NGL, respectively. However, the fluorine in these thickeners would bring about potential negative effects on the environment making them unsuitable for EOR applications [52, 72, 157, 160].

5. Mobility control: feasibility evaluation field A

Overall, from the discussions presented so far, it is clear that each of the proposed mobility control methods, as applicable to an MGI process, has its own challenges and deficiencies. The possible field-scale implementation of each method often depends primarily on the in-situ conditions and specific characteristics of the field of interest. The objective of this section of the chapter is to present an evaluation of the applicability of each of the techniques discussed earlier in Field A given its specific conditions and characteristics.

5.1 WAG technique

As mentioned earlier, in field applications, the WAG process has been applied successfully in a number of oil fields around the world [74, 75]. A total of 72 field-scale miscible and immiscible WAG projects were reviewed by Skauge et al. that have utilised hydrocarbon or non-hydrocarbon gases. Majority of these projects have been successful resulting in incremental oil recoveries in the range of 5–10% of OIIP. For successful projects, the WAG process consistently yielded better oil recovery than that could be achieved with continuous gas injection even though, often, a large amount of oil (35–65% of OIIP) would still be left behind [52]. Some of reviewed projects have also been unsuccessful due to operational and/or reservoir related difficulties such gas gravity segregation, extreme reservoir heterogeneity, excessive water production, corrosion, scale and/or hydrate formation, etc. [74] In the case of Field A, in-situ water saturation is very low (<10%) and, therefore, the field surface facilities and well completions are not designed to inject or handle large amounts of water. Therefore, the WAG strategy is not the best choice to implement in this field.

5.1.1 Gas foam technique

It was previously discussed that the gas foam injection process has been tried at the pilot scale in some fields in the United States and Canada. However, this technique has never been performed in any field in the Middle East due to the difficulties of finding a suitable surfactant (water soluble) or due to the harsh reservoir conditions encountered including high salinity and high temperature. Although, there has been a number laboratory-scale studies done to date evaluating the application the technique under conditions encountered in this region. For example, in a recent study conducted by Sumaiti et al. [56, 111], the foamability and mobility of CO₂-ethoxylated amine in carbonate cores were investigated at a salinity of 220,000 ppm and temperature of 393 K. The foamability of Ethomeen (C₁₂) and apparent foam viscosity increase were confirmed at these conditions. In addition, CO₂-foam core flooding obtained 8.89% of additional oil recovery. However, the availability of CO₂ is very limited in the Middle East. Concerning Field A, the reservoir presents a harsh environment with a formation brine salinity of 275,000 ppm and a reservoir temperature of 377 K with low in-situ water saturation and a very light oil (42${}^{°}$ API). It is extremely difficult to find a surfactant, especially water-soluble, which can work under these conditions. For the CO₂-foam process, there is a lack of adequate CO₂ availability in Oman. As a result, it is expected that achieving adequate foam stability would be a major challenge to implement a gas-foam process in Field A.

5.2 Direct thickened technique

As discussed earlier, several laboratory-scale studies have been conducted to date to find and/or develop direct thickeners for CO₂ and NGL. However, the cost and environmental issues associated with these thickeners have prevented their application beyond the laboratory scale [167]. As outlined earlier, this technique has several distinct advantages compared with the other two mobility/conformance control techniques of WAG and gas-foam injection. Firstly, a screened thickener additive would be thermodynamically stable and chemically inert (with no or minimal interaction with reservoir sediments), making it ideal for application in harsh reservoir conditions (i.e. high formation salinity and temperature). Secondly, the gas viscosity increase achievable by a thickener does not dependent on rock characteristics, properties and saturations of other fluids in the reservoir and injection flow rates. Thirdly, it eliminates the need for water co-injection which minimises the chance of excessive water production and treatment requirements substantially and eliminates the water blocking effect too. Lastly, it has been demonstrated at the laboratory scale that this technique can increase the sweep efficiency considerably because of delayed gas breakthrough and improved gas mobility. Hence, it is believed that CO₂ or AG mixture thickening may be the only viable technique for Field A to counteract unfavourable mobility conditions present in the Field and further enhance the oil recovery of the current ongoing MGI.

6. Conclusion and recommendations

This chapter presents the process of miscible gas injection (MGI) and the implementation of MGI in the petroleum industry especially for the recovery of light oil. It briefly discussed the challenges associated with the MGI flooding, and several solutions proposed in the literature to overcome these challenges include: water alternating gas flooding (WAG), foam flooding and the use of thickening agents. Despite many efforts made to date to identify a viable approach to counteract unfavourable mobility conditions and improve sweep efficiency. These approaches are not applicable in the fields as means of mobility control at field scale. Therefore, a further work requires that can improve the industry’s confidence in employing these approaches at the field scale using numerical simulation followed by economic analysis to investigate and verify the feasibility of these techniques for field applications.