(617g) An Equilibrium Adsorption Model to Estimate Gas-in-Place and CO2 Storage Capacity of Shale Reservoirs

Advances in extraction technologies have propelled shale gas production; however, these processes can be quite inefficient, with about 90% of the original Gas-in-Place (GIP) remaining in the reservoir [1]. Fluid behaviour in these formations is not well understood, and the effect of confinement in the nanometric pores of shale presents additional complexity. The deployment of enhanced recovery using CO₂ injection can potentially improve production rates as well as lead to the permanent sequestration of CO₂ [2].

In shale, natural gas (composed mainly of methane and ethane) can be adsorbed in the pores of the organic matter and clay minerals and be held freely in larger pores and fractures [3]. The quantification of gas adsorption in these reservoirs is essential, as it dictates both GIP estimates and CO₂ storage potential of shale. Classical approaches for calculating the GIP are flawed, because they do not account for the reduction of the available pore space upon gas adsorption. In this work, we propose a method for estimating the GIP that utilises the so-called excess amount adsorbed, the sole quantity that can be directly accessed experimentally. As such, the method does not necessitate any assumptions of density or volume of the adsorbed phase. We further extend this formalism to evaluate the effect of CO₂ injection on recovery and storage in shale reservoirs.

The material balance equations using excess adsorption are presented for a typical shale reservoir and are applied to the Marcellus reservoir in the USA, which is considered as the base case. We first investigate primary recovery and obtain the GIP as a function of reservoir pressure (or P/Z, the ratio of reservoir pressure and gas compressibility). We also examine the effect of adsorption behaviour on the GIP and reservoir production by considering four different adsorption isotherms: Langmuir, Henry’s Law, BET and Anti-Langmuir. We then use a cyclic production/injection process to investigate the effect of CO₂ injection on recovery. Our model consists of a Production stage, whereby the reservoir is depressurised to a certain production pressure, followed by an Injection stage, where a specific volume of CO₂ is injected into the reservoir. As mass transfer limitations are inherent in shale reservoirs, both the Production and Injection stages are not at equilibrium, i.e. there is no change in the adsorbed phase, and gas is added or removed only from the free gas phase. At the Soak stage, the reservoir is finally allowed to equilibrate, and gas is exchanged between the bulk phase and adsorbed phase. The cycle then continues, as the reservoir produces again. Competitive adsorption is incorporated in this model through the use of the binary Langmuir isotherm.

We observe that the unique features of the supercritical adsorption isotherms are also clearly evident in the GIP curves and that adsorption plays a dominant role in gas storage in shales. We also show that the conventional approach that does not use the excess amount adsorbed can lead to a significant overestimation of the GIP. Injection of CO₂ yields up to 70% more CH₄ as compared to primary recovery, and about 50% of the CO₂ remains permanently sequestered in the reservoir. We also compare this to a N₂ sweeping scenario, where the same volume of N₂ is injected instead of CO₂. We observe that although CH₄ recovery is similar to the CO₂ injection scenario, higher reservoir pressures are required to sustain this recovery. Although gas injection in general leads to more CH₄ from the reservoir, the main benefit of using CO₂ is its storage. We also test the sensitivity of the model to various factors such as reservoir pressure, porosity and the adsorption model parameters. The equilibrium adsorption model is a useful tool as it can provide a quick means of testing injection scenarios, evaluate production methodologies and produce accurate estimates of the GIP and CO₂ storage potential of shale reservoirs.

References

1. Jarvie, R. Hill, T. Ruble, R. Pollastro, AAPG Bulletin, 91 (2007) p.475-499
2. R Edwards, M. Celia, K. Bandilla, F. Doster, C. Kanno, Environmental Science and Technology, 49 (2015) p.9222-9229
3. Yu, K. Sepehrnoori, T. Patzek, SPE Journal, 21 (2016) p.589-600