Enhanced recovery

Huff ’n’ Puff Gas-Injection Pilot Improves Oil Recovery in the Eagle Ford

The Eagle Ford formation has produced approximately 2 billion bbl of oil during the last 7 years, yet its potential may be even greater. Using improved oil-recovery (IOR) methods could result in billions of additional barrels of production.


The Eagle Ford formation has produced approximately 2 billion bbl of oil during the last 7 years, yet its potential may be even greater. Using improved oil-recovery (IOR) methods could result in billions of additional barrels of production. While a number of companies have field-tested an IOR method called huff ’n’ puff gas injection, most of the published results are from laboratory and modeling studies. This paper evaluates the results of these field tests and discusses the potential of the huff ’n’ puff method in the formation.


The Eagle Ford formation, formed during the late Cretaceous, is a laminated calcareous organic rich shale. The shale was deposited in a low-energy anoxic marine shelf environment, which allowed for rapid deposition and burial of abundant organic material. The Eagle Ford is divided typically into two sections: the Upper, which was deposited during a regional marine regression, and the Lower, which was deposited during a transgressional period and tends to have more organic-rich black shale. The Eagle Ford is laterally continuous and spatially extensive throughout much of southern Texas.

The matrix permeability of the reservoir is low enough that typical hydrocarbon migration is restricted, causing the oil-rich rocks to be stratigraphically higher than the gas-filled ones.

Results of Previous Huff ’n’ Puff Pilot Projects

The huff ’n’ puff process involves injecting a miscible gas into a well and then, after some time has passed, producing back from that same well. Data have been collected on seven pilot projects in the Eagle Ford that have been completed during the last 5 years. Data from four of these seven are discussed in this synopsis. All of the pilots used hydrocarbon gas, but the composition of the gas varied across the locations. In addition, all of the field trials used a huff ’n’ puff injection scheme.

Pilot A. The first pilot was a single isolated well where the nearest offset producers were more than 2 miles away. This provided some assurance that the injected gas would not migrate to other wells. The oil rates for the well indicated a positive response to the process. Each cycle increased the production rate to approximately half of the well’s initial rate. Once the oil rate began to drop, another injection cycle was started. This is encouraging because not only the first cycle but also subsequent cycles were able to increase production.

Pilots B and C. These are both multiwell pilots in which approximately half of the wells in the lease are cyclic-­injection wells and the remaining wells produce without huff ’n’ puff. The offset producers, however, showed production increases that might be the result of the huff ’n’ puff wells. Pilot B involved four injection wells and four noninjection wells. Although the patterns are unknown, the assumption is that the four injection wells are in the middle with two buffer wells on each side of the injection wells to ensure isolation from offset patterns. For Pilot C, eight horizontal wells run northwest to southeast while the other wells run perpendicular at the toe and heel of the original eight wells. The injection pattern is not known, but the assumption is that the six injection wells are in middle of the eight, and the surrounding wells are used to monitor for gas leaving the piloted area.

The average lease oil-production rate since injection started in early 2015 is 370 STB/D for Pilot B and 1,065 STB/D for Pilot C. If this project had not been implemented, the average oil rates would have been 170 and 420 STB/D for leases associated with Pilots B and C, respectively. This calculation is based on the preinjection decline curves. The leases are producing at more than double the rate that they would have if only half of the wells received injection fluids.

Pilot B injected for approximately 1½ years and then produced without injection for another year. This lease shows a 17% increase in cumulative production in little more than a 2½-year span. Pilot C has continued huff ’n’ puff injection for 2½ years, and the increase in production is more than 550,000 STB, which amounts to a 20% increase in cumulative production.

Pilot D. This pilot is more than 100 miles to the southwest of the other three pilots. All four wells in the lease are injected at the same time, so interpretation is clearer. Because all four wells are cyclic injectors, the lease rates were divided by the number of wells to derive an average well rate.

The Pilot D huff ’n’ puff project has been ongoing for 3 years. The four wells have produced 300,000 STB during that time. Without the injection process, the estimate is that the lease would have produced only 130,000 STB during that time. During the last 3 years, therefore, the gas-injection process has increased production by 170,000 STB, more than doubling the incremental production during that time. When considering all 6 years of the lease’s production history, the injection case already has produced 1.3 times more than the predicted preinjection cumulative production.  

Because data quality is better for this pilot, an attempt is made to predict future production for these wells in this lease. The extended huff ’n’ puff process would be expected to recover an additional 370,000 STB, or 50% more, than the preinjection decline-curve-analysis model would predict.

Pilot Summary. Fig. 1 shows a standardized version of a cumulative oil plot of lease oil production for the three pilots where data can be analyzed. The lowest purple line represents a standardized lease production on primary production (no-injection case) on the basis of decline-curve extrapolation. The red shaded area indicates improved production as the result of gas injection of 30 to 70% over primary production, which is based on reported estimates.

Fig. 1—Comparison of measured production from three leases with reported estimated recovery.

Considerations for Future IOR Pilot Projects in Unconventionals

Location and Infrastructure. In conventional reservoirs, the adage is that the best primary wells make the best waterflood/IOR wells, and this also likely applies to unconventional oil reservoirs. For a pilot project, however, it is probably best to test on an average area of the reservoir. Locating the pilot where the operator has 100% working interest is preferable, but that is often impractical or does not fit with the other criteria for selecting a location. In those cases, selecting a location that can be easily unitized or where partners are committed and willing to work together on the project is important.

The injectant is likely to be the most expensive part of the operation, particularly if an IOR method is being used. Therefore, during the planning stages for the pilot, the source of the injectant needs to be resolved for the fieldwide development.

Isolating wells from offsets by 2 or more miles is best, but, with many plays already drilled up, finding such isolated wells or pads may be difficult. If wells are in pressure communication (or are even more connected) with each other, then all of the wells need to be considered as part of the pilot, regardless of whether fluid is being injected into all of them. Vertical containment also needs to be considered; if injected fluid is being lost to the overburden, the economics of the project will be reduced significantly.

Drilling and Completion Strategies. The wells need to be in zone as much as possible, and the hydraulic fractures need to be designed to create the greatest possible surface area. Matrix permeability is so low that the injectant cannot push oil through the matrix like a conventional flood, so the injected fluid has to access the remaining oil through the fracture networks. The more fracture complexity, the better the opportunity for success with IOR.

Data Acquisition. Data-collection strategies are needed for three distinct phases in the unconventional IOR development: the preinjection study phase, the pilot phase, and fieldwide development.

Improved Reservoir Characterization. Pilot projects are only one tool that can be used to engineer successful IOR projects. Laboratory studies and numerical reservoir modeling (building fit-for-purpose models and testing various ideas) provide a complementary data set. These tools will describe better how IOR processes will work and will help extract more value from large-scale field development.


  • The huff ’n’ puff gas-injection pilots have shown clearly that this process will recover additional oil in the black-oil window of the Eagle Ford.
  • Economics during the pilots appear to be marginally successful; however, as more pilots are completed and fieldwide development occurs, improved efficiencies and better economics should be realized.
  • As the industry moves to the next generation of pilot projects, setting clear objectives for the field trials is important; otherwise, capital is likely to be wasted, either by not obtaining the data from the pilots needed to learn how to improve the process or by overspending on what is truly needed from the pilot.
  • Important considerations when designing a pilot include the pilot location and the proximity to existing infrastructure. Additionally, completion design will have a significant effect on the success of injection projects in unconventional oil reservoirs.

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 189816, “Huff ’n’ Puff Gas-Injection Pilot Projects in the Eagle Ford,” by B. Todd Hoffman, SPE, Montana Tech, prepared for the 2018 SPE Canada Unconventional Resources Conference, Calgary, 13–14 March. The paper has not been peer reviewed.