Steam assisted gravity drainage

Steam assisted gravity drainage (SAGD) is an outstanding example of a steam injection process devised for exploitation of heavy oil or bitumen reservoirs utilizing horizontal wells. It is widely used in Alberta Canada, Russia, and China for recovery of heavy and extra-heavy oilsands resources.

Basic process

Several variations of the basic process have been developed, and are being tested. The original SAGD process, as developed by Butler, McNab, and Lo in 1979, utilizes two parallel horizontal wells in a vertical plane: the injector being the upper well and the producer the lower well (Fig. 1, taken from Butler). If the oil/bitumen mobility is initially very low, steam is circulated in both wells for conduction heating of the oil around the wells. The high-pressure steam is then injected into the upper well and mobilize heavy oil/bitumen. The mobilized heavy oil/bitumen flow downward due to gravity drainage to the lower well and is pumped to surface. As a result, steam rises forming a steam saturated zone called 'steam chamber' with heavy oil/bitumen flowing at the sides of the chamber and condensate flowing inside the chamber, as shown in Fig. 1. This is an idealized situation. Other flow regimes may occur depending on the oil and formation properties.

• 

  Fig. 1 – Conceptual diagram of the SAGD process (after Butler).

Major mechanisms

There are four main mechanisms involved in the SAGD process for enhancing heavy oil/bitumen recovery:

Heat conduction at chamber edges

In SAGD process, steam is injected into a horizontal injection well and is forced outward, losing its latent heat when it comes into contact with the cold bitumen at the edge of a depletion chamber.

Gravity drainage

After heating by conduction at the chamber edge, the viscosity of bitumen falls several orders of magnitude. The bitumen then flows under gravity forces toward a horizontal production well that is placed several meters below the injection well.

Thermal expansion

The volume of bitumen will expand due to heat conduction and rising temperature. This expansion can generate high pressure and dilate the sand, thereby increasing its permeability.

Solution gas drive

Sometimes a certain amount of solution gas can present in the heavy oil/bitumen reservoirs. Under this case, oil can be pushed or squeezed out of the reservoir due to gas exsolution and gas volume expansion.

Estimating production rates

Butler's SAGD theory

Butler’s SAGD theory shows a correlation related to the oil drainage flow rate:

, ....................(1)

where Φ is porosity (fraction), ΔS_o is recoverable oil saturation (fraction), k is permeability (md or μm²), α is thermal diffusivity (m² s^-1 or ft² s^-1), h is the height of the reservoir (m or ft), m is a constant used to reflect dependence of viscosity on temperature and γ_s is kinematic viscosity (m²/s or cS).

For the simplest case, the oil production rate q_oh, in B/D per ft well length, is given by (multiplier "2" indicates flow from two sides of the steam chamber)

....................(2)

where the kinematic viscosity of oil (in centistokes) at the steam temperature, T_s, is given by ν_s, and that at any other temperature, T, is given by

....................(3)

where m is derived from the viscosity-temperature relationship of the oil.

Eq. 3 predicts rates of 0.1 to 0.7 B/D per ft for a horizontal well for an oil viscosity of 100,000 cp. For example, a 2,000-ft long well may be expected to produce about 800 B/D at a steam temperature of 400°F. The theory has been verified by laboratory experiments. Field results to-date have been encouraging. One commercial project (EnCana’s Foster Creek Project), consisting of 22 well pairs, has been in operation since October 2001. In 2002, steam/oil ratios were averaging 2.5 bbl steam/bbl of oil. Earlier field tests of SAGD in Athabasca oilsands were successful at a depth of about 600 ft, which is too deep for surface mining and not deep enough for high-pressure steam injection.

Considerations

SAGD is a complex process because gravity flow strongly relies on a high vertical permeability. Typical vertical well spacing between injector and producer is 5 to 6 m [16 to 20 ft]. It is also important that the steam chamber be sealed with an impermeable shale caprock to ensure reservoir containment. There is no steam migration to offset vertical wells. In California, SAGD failed to achieve commercial success because of relatively high initial mobility of oil, as well as other reasons.

To have a successful SAGD treatment, the following three parameters should be considered.

Thickness

The reservoir should be thick enough to place two parallel horizontal wells, also ensure less energy loss. The recommended minimum reservoir thickness is 15 meters.

Vertical permeability

Since the steam need to travel inside the reservoir and the mobilized heavy oil need to flow downward to the well through gravity drainage, a high vertical permeability of the reservoir is essential.

Oil saturation

Higher heavy oil saturation will lead to higher efficiency of the steam chamber.

Variations

One variation of SAGD is known as single-well SAGD. Here, insulated tubing is used to inject steam into a single horizontal well, with production from the annulus. This process was successful in a few cases but generally failed. Another variation (Vapex) utilizes a suitable solvent (such as ethane, propane, etc.) instead of steam and is being field tested.

Negative environmental effects from SAGD operations

Although there is no clear evidence that SAGD could directly causes environmental damage, this approach usually has some impact on the formation. According to the Inside Climate News, in 2016, there were four runaway bitumen leaks that occurred in Alberta. Many geologists believed that this was caused by high pressure steam injection, which is part of SAGD treatment. Geologists also found that there were natural fractures within the bedrock layers due to salt dissolution and any high-pressure treatment would let oil flow up through these fractures and come to the surface.

Aside from that, water and air pollutions are also related to SAGD. The water used for steam injection will contain various contaminants afterwards. This would cause water waste and further pollution to the surface water if not handled properly.

Carbon dioxide emission along with other emissions like sulfur caused by SAGD treatment applied to heavy oil reservoirs is much higher than conventional oil development. Since natural gases are needed to produce steam, this would also lead to high consumption of natural gas.

Applications

Due to the high consumption of the conventional oil resource, the development of unconventional oil resources like heavy oil has been grown rapidly, and SAGD has also been widely used. Canada is one of the regions with the most applications of SAGD technique, since Chanda has the largest known reservoir of crude bitumen (Athabasca).

Cenovus foster creek

Cen Foster Creek is one of the oil sands projects in Alberta, Canada, started in 1996, with the first SAGD application in 2001. There are approximately 2607 MMbbls proved and probable reserves of bitumen (9° - 11° API gravity), the reservoir depth is about 450 meters and the thickness of the pay zone is 25 to 30 meters. With the application of SAGD, the average production per well is 544 barrels/d and the total production was about 166000 barrels/d in 2018.

Nomenclature

hh	=	fluid level in stimulated reservoir, ft [m]
k	=	reservoir permeability, md [μm2]
kro	=	relative permeability to oil
m*	=	exponent in Eqs. 2, and 3
qoh	=	hot oil production rate, B/D [m3/d]
So	=	oil saturation
Sors	=	residual oil saturation to steam fraction
T	=	average temperature in heated reservoir, °F
TR	=	unaffected reservoir temperature, °F
Ts	=	steam temperature, °F
Ф	=	porosity

Noteworthy papers in OnePetro

Mukhametshina, A., Hascakir, B., Bitumen extraction by expanding solvent-steam assisted gravity drainage (ES-SAGD) with asphaltene solvents and non-solvents, 2014 SPE Heavy Oil Conference, 10-12 June, 2014, Calgary, Alberta, Canada, SPE 170013-MS., https://www.onepetro.org/conference-paper/SPE-170013-MS

See also

Thermal recovery by steam injection