Enhanced
gas recovery (EGR) is believed to be a promising technology
to improve the production of shale gas reservoirs and simultaneously
reduce the emissions of greenhouse gas via the injection (sequestration)
of carbon dioxide, to which great effort has been devoted by scholars.
However, traditional investigations are generally limited to the ideal
model of nanochannels and statistic characterization of competitive
adsorption, neglecting the nanoporous structure of the kerogen matrix
and the complex dynamic behavior during the EGR process. In this work,
we present a comprehensive study of the EGR process in a realistic
kerogen pore network (matrix) which is obtained from the artificial
pyrolysis of bulk kerogen through reactive force field molecular dynamics
(ReaxFF MD) simulations. The influence of pore properties (e.g., porosity)
of the kerogen matrix under different maturities, and the proportions
(i.e., methane and carbon dioxide) of injection gas with various injection
pressures are revealed and meticulously discussed. In addition, the
underlying mechanisms including diffusion and displacement effects
behind the EGR process are analyzed by combining them with with particle
trajectory capture technology. In particular, based on the MD simulation
results, an analytical model to depict the dynamic recovery process
in the kerogen matrix is proposed by coupling consideration of recovery
time and capacity, which are examined against the simulation and experimental
data. The hybrid recovery strategy is developed by utilizing the advantages
of depressurization and gas-injection recoveries to achieve the optimization
of both recovery time and capacity. The insights acquired from this
work would be helpful for efficient exploitation of shale gas reservoirs
and pave the way to capture the realistic EGR processes within the
kerogen matrix from molecular and theoretical perspectives.