|Fig. 1: Illustrative cases for stable and unstable displacement.|
Oil remains the world's leading choice of energy, accounting for 33% of global energy consumption, despite losing market share 13 years in a row. In 2012, global oil consumption increased by 890,000 barrels per day, bringing total daily consumption to approximately 87 million barrels.  To satisfy demand and increase reserves, companies have been relying on enhanced oil recovery (EOR) methods for decades. 
The life of a hydrocarbon reservoir is commonly divided into three phases. During the initial phase, or primary recovery, oil is produced by natural drive mechanisms like supporting aquifers, solution gas drive, gas cap drive, or gravity drainage. Once natural drive mechanisms are depleted, secondary recovery techniques are put into action to maintain the reservoir pressure. This includes techniques like water or gas injection. The lifetime of a reservoir is further increased by special techniques during tertiary recovery. The third stage is commonly referred to as enhanced oil recovery; though, it should be noted that enhanced oil recovery is not limited to a certain stage of recovery, and that several definitions for EOR exist in the literature. [3-5] Improved oil recovery (IOR) has sometimes been used interchangeably with EOR, and while there is no formal definition for IOR, it is generally understood to summarize any approach improving oil recovery. IOR, therefore, encompasses EOR. 
Generally, natural depletion of a reservoir allows for a very limited recovery of the oil in place. Depending on the type of reservoir, values are around 20%. During secondary recovery, another 15 to 20% might be added.  Values for recovery at the tertiary stage can vary significantly depending on many factors. For example, original production forecasts under primary depletion for the Suplacu de Barcau oil reservoir, a heavy oil reservoir located in Romania, were estimated to be around 9% for an 80 years time frame. Through the introduction of a combination of thermal enhanced oil recovery methods, the cumulative extracted crude in 35 years corresponds to a current recovery factor of 44.6%.  We need to understand the mechanisms limiting primary and secondary recovery to understand the significance of enhanced oil recovery. There are two main reasons oil cannot be swept completely from a reservoir when displacing it with another fluid. First, we have trapping of oil on the pore scale. The mechanism is described by a snap-off model and can be elegantly summarized by the capillary number.  The capillary number is a dimensionless number and defined as
|Nca||=||ν × μ
σ × cos(θ)
where v is the velocity, μ the viscosity, σ the surface tension, and θ the wetting angle. [8,9] In essence, the capillary number is the ratio of viscous versus capillary forces. A small capillary number suggests that the motion of the fluid is dominated by capillary forces. Conversely, a large capillary number indicates a viscous dominated regime. From a practical point of view, we wish to increase the capillary number thereby reducing trapping. As pointed out by S. Thomas, we would need an increase by three orders of magnitude to halve the residual oil saturation. 
Looking at the capillary number one would assume any technique raising the product of velocity and viscosity does suffice. The product, however, is directly proportional to the pressure drop, consequently limiting an injection well to fracture pressure. For example, chemical enhanced oil recovery methods focus especially on increasing the capillary number by reducing interfacial tension rather than increasing the product of velocity and viscosity. 
The second phenomena affecting recovery happens on the reservoir scale and is described by another dimensionless number, the mobility ratio.  Mobility itself is defined as the ratio of permeability κ and viscosity μ
The mobility ratio is then given as the ratio of the displacing phase mobility to the displaced phase mobility:
From a sweep efficiency perspective, mobility ratios larger than 1 are unfavorable because it would result in an unstable displacement. Displacing viscous oil with water is an example for a mobility ratio larger than 1. The water will tend to finger through the reservoir yielding poor sweep efficiency. Mobility ratios smaller than 1 are preferred because the injected fluid is then able to displace the oil in a more piston like manner.  Both cartoons in Fig. 1 give an aerial view of a reservoir with a simple injector-producer configuration. The oil is colored in green, water injected is colored in blue. The figure on the left illustrates the viscous fingering problem for mobility ratios larger than 1. Once a finger has reached the producer it will become harder and harder to sweep the remaining oil. The water injected will prefer the flowpath established by the finger. The figure on the right illustrates the case for a mobility ratio smaller than 1. The water front is growing in a stable, radial manner with no fingers trying to get ahead of the front. Ultimately, this will yield a better sweep of the reservoir.
Enhanced Oil Recovery methods are divided into three main groups: chemical, miscible or solvent injection, and thermal.  All of them are subject to extensive research at Stanford.
As mentioned above, chemical methods focus on injecting interfacial active components such as surfactants thereby increasing the capillary number.  According to Lake, one of the most common methods is micellar-polymer flooding.  Newer methods like alkaline-surfactant-polymer and surfactant-polymer target to change both dimensionless numbers at the same time. In the case for alkaline-surfactant-polymer, the polymer targets the mobility control issue. The surfactant and the alkali combine forces to lower interfacial tension. 
Miscible or solvent injection relies on the injectants miscibility with the oil phase. This type of enhanced oil recovery method aims to decrease the viscosity of the oil and/or cause it to swell.  With respect to the second dimensionless number, the mobility ratio, we can already foresee the challenges that go along with miscible or solvent injection. The fluid injected usually has a lower viscosity and viscous fingering is not unlikely. In addition, depending on the density contrast between the oil and the injectant, gravity over- or underride occurs reducing sweep efficiency. 
Thermal methods are especially geared towards reservoirs with high viscosity oils. Adding thermal energy to the reservoir by injecting steam results in a decrease of the oil viscosity making it more mobile. In addition, residual oil is reduced and the presence of a gas phase in the reservoir causes light components in the oil to be distilled.  Kern River in Bakersfield is a prominent example for steamflooding. 
|Fig. 2: Example of a successful combustion tube experiment.|
Another possibility is to create the thermal energy in place by injecting air, combusting some of the oil.  This process is referred to as in-situ combustion and is part of the development strategy of the aforementioned Suplacu de Barcau oil reservoir. Interestingly, not every crude oil has the properties lending itself to this process. To determine whether or not a crude oil would be possible candidate for in-situ combustion, extensive laboratory experiments have to be carried out. Combustion tube experiments are of special interest at Stanford. A metal tube is filled with the reservoir matrix and oil mixture under investigation. One end of the tube is then heated in an inert atmosphere. Once ignition temperatures are reached, air is injected to initialize the combustion process. For a successful combustion run, the combustion front will slowly travel through the tube pushing the oil out.  Fig. 2 shows the result of a successful combustion tube run. Air was injected from right to left and we can clearly distinguish between the part swept by the combustion front, a transition zone, and the untouched part.
The demand for hydrocarbons is still increasing. To keep up with demand, the hydrocarbon industry is constantly challenged to improve current recovery methods and find new means to produce oil. In this paper the mechanisms limiting primary and secondary recovery were discussed to explain the significance of enhanced oil recovery methods. In essence, the limiting mechanisms can be summarized with two dimensionless numbers, the capillary number and the mobility ratio. Enhanced oil recovery methods aim to overcome these limitations attacking the challenges from different angles.
© Guenther Glatz. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.
 "BP Statistical Review of World Energy," British Petroleum, June 2013.
 E.C. Hammershaimb et al., "Recovery Efficiency of Enhanced Oil Recovery Methods: A Review of Significant Field Tests," One Petro 12114-MS, 5 Oct 83.
 J. Sheng, Modern Chemical Enhanced Oil Recovery: Theory and Practice (Gulf Professional Publishing, 2010).
 A. Carcoana, Applied Enhanced Oil Recovery (Prentice Hall, 1992).
 S. Thomas, "Enhanced Oil Recovery - An Overview," Oil & Gas Sci. Technol. - Rev. IFP 63, 9 (2008).
 A. Panait-Patică, D. Şerban and N. Ille, "Suplacu de Barcau Field - a Case History of a Successful In-Situ Combustion Exploitation," One Petro 100346-MS, 12 Jun 06.
 L. W. Lake, Enhanced Oil Recovery (Prentice Hall, 1996).
 J. J. Taber, "Research on enhanced oil recovery: Past, present and future," in Surface Phenomenon in Enhanced Oil Recovery Plenum, ed. by D. O. Shah (Springer, 1981).
 V. Alvarado and E. Manrique, Enhanced Oil Recovery: Field Planning and Development Strategies (Gulf Professional Publishing, 2010).
 R. A. Fulcher, T. Ertkin and C. D. Stahl, "Effect of Capillary Number and Its Constituents on Two-Phase Relative Permeability Curves," J. Petrol. Technol. 37, 249 (1985).
 J. S. Aronofsky, "Mobility Ratio - Its Influence on Flood Patterns During Water Encroachment," J. Petrol. Technol. 4, 15 (1952).
 M. Prats, Thermal Recovery (Soc. Petrol. Eng, 1982).
 G. R. Greaser and R. A. Shore, "Steamflood Performance in the Kern River Field," One Petro 8834-MS, 20 Apr 80.
 H. J. Ramey, Jr., "In Situ Combustion," One Petro WPC-14229, 13 Jun 71.
 G. Glatz et al., "Kinetic Cell and Combustion Tube Results for a Central European Crude," One Petro 146089-MS, 30 Oct 11.