Viability of Gas-Injection EOR in Eagle Ford Shale Reservoirs

You have access to this full article to experience the outstanding content available to SPE members and JPT subscribers.

To ensure continued access to JPT's content, please Sign In, JOIN SPE, or Subscribe to JPT

Gas-injection huff ’n’ puff enhanced-oil-recovery (EOR) techniques have the potential to improve liquid hydrocarbon recovery in ultratight, unconventional reservoirs. This paper studies the technical and economic viability of this EOR technique in Eagle Ford shale reservoirs using natural gas injection, generally after some period of primary depletion, typically through long, hydraulically fractured horizontal-reach wells.

Reservoir-Analysis Work Flow

Model Description. A compositional, fine-scale, dual-porosity, dual-permeability, symmetry-element numerical model was used in this study to model the current primary depletion and the EOR huff ’n’ puff process. In this study, the element of symmetry, which represents the bottom half of a cluster within a fracturing stage, is being extended to include three wells to allow the investigation of the effect of interference and containment on pad-level cyclic-gas-injection deployment.

Production and injection of the entire well come from all active stages in each well, with equal weighting. The production and the injection for each cluster in every well are calculated proportionally by use of these assumptions.

Model Input. The complete paper describes the critical input for the numerical model, including fracture half-length and fracture height, shape factor, absolute and relative permeability, and pressure/volume/temperature (PVT).

History-Matching Work Flow. Because the history-matching solution is known to be highly nonunique, a comprehensive probabilistic approach was used to identify the realistic solution space. To facilitate this process, the authors used a three-step approach:

  1. Identify the solution space through parametric study.
  2. Perform combinatorial analysis of predetermined variables and understand available solutions within the solution space.
  3. Fine-tune history matching to identify the correct combination of parameters for probable solutions.

Characterizing the Performance of Cyclic Natural-Gas Injection

The compositional fluid-flow analysis of the combined sample of the separator liquid and gas at reservoir conditions indicated a volatile oil in the reservoir. To evaluate the phase behavior of the mixture, the original oil and the gas supplied for injection purposes were mixed at different ratios. For each mixture, the calibrated equation of state (EOS) is used to evaluate the resulting phase diagram and fluid properties by modeling a constant-composition-expansion experiment at reservoir conditions.

The 1979 Peng-Robinson EOS with nine components was used to model the PVT changes resulting from the composition changes related to each injection ratio. As the composition changes, the phase behavior and the fluid properties also change. The fluid system in the reservoir gradually changed from a volatile oil to a gas condensate.

Results show that when injected-gas makeup is 20% or greater, the phase behavior changes from volatile oil to gas condensate. When the mixture consists of more than 70% injected gas, at any pressure below 6,000 psi, there is a possibility that condensation can occur.

Operational Conditions. Injection and Production Periods. To find the duration of the injection cycle, it was necessary to monitor the injection rate decline vs. time in the simulation results, assuming a maximum injection rate of 10 MMscf/D and a maximum bottomhole injection pressure of 10,300 psi. Significant decline in the injection rate signals that the reservoir is being pressured up. Continuing injection after this point will be counterproductive because other wells in the pads are waiting for gas to be re-energized. The injection period of 9 weeks at an injection rate of 10 MMscf/D was determined as optimal for these cases.

The simulator was set at a bottomhole pressure (BHP) of 6,000 psi and the production rate was monitored until it started to decline. With this methodology, the optimal production duration was determined by using the point at which production is negligible, which was determined to be approximately 4–5 weeks.

Shut-In Period. Finding an optimal duration for the shut-in period is an exercise in compromise. Pressure distribution and stabilization are monitored immediately after injection has stopped. Once the well is shut in, the pressure at the points near the wellbore begin to decline while the pressure at points farther from the wellbore start to increase until equilibrium is reached.  The pressure profiles from the models indicate that 3–-4 weeks are enough, depending on the cycle number, to propagate the injected fluids and to stabilize the pressure in the stimulated reservoir volume (SRV) and the surrounding matrix.

Sensitivity Analysis. Because the optimization approach was designed on a pad rather than on a well basis, for the sensitivity analysis, the injection-production schedule for Well B only (out of six wells within a pad) for the case of a $50/STB oil price and a $1.50/Mscf gas price was selected to accomplish the sensitivity analysis. The Well B schedule in all sensitivity cases is represented as Well B in the simulation model. For all the sensitivity cases, the base case is selected from the cases with fracture half-length of 175 ft and fracture height of 50%. The sensitivity cases outlined in this section of the complete paper were represented by only one well within the pad.

Compaction Table. Compaction of the formation may cause induced fractures to close, resulting in a reduction of conductivity of the fracture networks, and can influence the fluid transformation from the fracture networks into the wellbore. Because there were no available data provided for the compaction effect in this area, similar studies were reviewed in the literature and three published compaction functions for the Eagle Ford were found. The most-conservative of these compaction functions was selected for the starting point. Further tuning was applied using the conservative function to achieve a reasonable history match.

During the injection-production cycles, two scenarios of compaction functions—when compaction is reversible and when it is irreversible—are evaluated. In the irreversible scenario, the permeability in the SRV is maintained to be the same as at the end of the depletion period (the end of history matching) and is not increased during the injection cycles. Because of low conductivity at the end of history matching, a lower injectivity rate, coupled with higher BHP, is expected. Moreover, low conductivity will also cause lower oil production, as compared with the reversible scenario. If reversible compaction is considered, the injection will create an increase of permeability in the SRV and also in the main conduit during the injection cycle and could increase gas leakage from the injected well toward the offset well.

Gas-Diffusion Effect. To simulate the effect of gas diffusion, the Langmuir isotherm option is borrowed. There were no data available for gas-diffusion parameters for this study; however, two different sets of values for Langmuir parameters in the literature were found for the Eagle Ford play. To cover the spectrum of the variables reported in the literature, both cases were simulated for comparison purposes. A slight increase was observed in the oil production because of diffusion effects during the historical depletion period. However, during the injection and production cycles of cyclic gas injection, this effect becomes minimal.

Injected-Gas Composition. The composition of the injected gas has a significant effect on phase behavior of the mixture. As mentioned previously, the fluid system in the reservoir gradually changed from a volatile oil to a gas condensate as composition changed in each injection cycle. Through the injection of 100% methane (dry gas) at higher ratios, the phase envelope expands, which increases the bubblepoint and dewpoint pressure of the mixture. Larger changes will be anticipated if the composition of the injected gas switches to lean gas. The miscibility process by methane gas injection is dominated mainly by the vaporization process.

Fracture Geometry. Two fracture half-lengths were used for history-matching purposes: a fracture half-length of 91 ft and a half-length of 175 ft. For the cyclic gas-injection simulations and all sensitivity scenarios, one of the most likely cases with a fracture half-length of 175 ft was selected as the base case. Results for cyclic gas injection and the sensitivities were based on this fracture half-length and height. The results of optimistic, pessimistic, and most-likely cases are compared with the base case with a half-length of 175 ft. The base case and optimistic case using a fracture half-length of 91 ft both yield similar uplift, 41 vs. 43%, respectively. Similarly, the uplift of 25 and 20% were expected from the most-likely and pessimistic cases using a half-length of 91 ft.

Summary. The sensitivity analysis is summarized in Fig. 1. On the basis of this sensitivity study, the top parameters are the SRV and quality and containment of the injected gas.

Fig. 1—Summary of the sensitivity analysis, showing the effect of individual parameters compared with the uplift of the base case.

Economic Analysis

Through the cyclic natural-gas-injection process, light components of the volatile oil vaporize into the injected gas and enrich the gas in each cycle. The efficiency of the vaporization process gradually declines with each cycle because the remaining moles stranded in the reservoirs are fewer and the difference between the liquid phase and vapor phase is less accentuated than in the earlier cycles. The recovery process then becomes less effective but still economical. To find this optimal economic point at which to switch the compressor to the next pad, an economic work flow has been developed as a function of commodity price and costs.


  • The results of this study suggest a likely incremental oil recovery of 41% with gas-injection EOR using produced gas.
  • On the basis of the sensitivity work, containment of the injected gas is the highest risk to the project.
  • Compressor use, rotating the compressors as a function of the pad performance, and economic variables could result in a highly economical project, even in a low-price environment.
  • Modeling efforts, although imperfect, provide a guide for designing a suitable surveillance plan.

For a limited time, the complete paper SPE 191673 is free to SPE members.

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 191673, “Viability of Gas-Injection EOR in Eagle Ford Shale Reservoirs,” by Safian Atan, SPE, Arashi Ajayi, Matt Honarpour, Edward Turek, SPE, Eric Dillenbeck, and Cheryl Mock, BHP, and Mahmood Ahmadi and Carlos Pereira, MI3 Petroleum Engineering, prepared for the 2018 SPE Annual Technical Conference and Exhibition, Dallas, 24–26 September. The paper has not been peer reviewed.

Viability of Gas-Injection EOR in Eagle Ford Shale Reservoirs

03 January 2019

Volume: 71 | Issue: 1


Don't miss out on the latest technology delivered to your email weekly.  Sign up for the JPT newsletter.  If you are not logged in, you will receive a confirmation email that you will need to click on to confirm you want to receive the newsletter.