Abstract:

Shale gas and liquid-rich shales have become important energy sources in the US and other parts of the world. Unlike conventional oil and gas reservoirs, unconventional shale resources contain a very heterogeneous pore system. The pore size varies between micro-, meso- and macroscales (50 nm). The mineral composition of shale rocks varies widely as well from clay-rich to calcite-rich. The nanoscale nature of the pores, coupled with rock mineral heterogeneity, makes the \conventional" understanding of uid transport in conventional reservoirs no longer suitable to explain and predict accurately the ow behavior in unconventional resources. The research work aimed to bridge the gap in the understanding of the uid ow behavior of unconventional resources by applying various experimental and molecular simulation tools. Specically, this research work studied how the rock (i.e. permeability), the uid (i.e. composition and phase behavior) and the uid-rock interactions (i.e. adsorption) all behaved with depletion in nanoporous rock formations. Several laboratory experiments and molecular simulation techniques were applied in this research work. Laboratory experiments included a gas-condensate core- ooding experiment, permeability measurements and adsorption measurements. In the core- ooding experiment, a real gas-condensate mixture obtained from the Marcellus shale play was injected into a Marcellus shale core at in-situ conditions and the composition of gas samples collected along the core was monitored during ow. To investigate the e
ect of rock mineralogy and pore structure on the transport mechanisms in nanoporous shale reservoirs, the permeability of Utica, Permian and Eagle Ford shale samples were measured using argon as a nonadsorbing gas and CO2 as an adsorbing gas. In addition, CO2 adsorption experiments were conducted on di
erent v shale samples in order to investigate the role of shale mineral constituents in adsorption. Moreover, molecular simulation techniques were applied to model the selective adsorption of binary hydrocarbon mixtures in carbon-based slit-pores and to estimate the shift in the critical properties of hydrocarbons due to connement in nanometersize pores. The molecular simulation techniques included the grand canonical Monte Carlo (GCMC) and the Gibbs ensemble Monte Carlo (GEMC). This research work revealed that clay content in shale reservoirs played a significant role in the stress-dependent permeability. For clay-rich samples, higher pore throat compressibility was observed which in turn led to higher permeability reduction with increasing e
ective stress compared to calcite-rich samples. Numerical simulation results showed that failing to account for stress-dependent permeability in clay-rich shale reservoirs may lead to overestimating the cumulative gas recovery by a factor of two after ten years of production. Permeability measurements with CO2 indicated that CO2 permeability decreased in comparison with the nonadsorbing gases by as high as an order of magnitude due to a combination of CO2 adsorption, sorption-induced swelling and molecular sieving e
ects. CO2 adsorption measurements indicated that adsorption was controlled mainly by the clay content. Clay-rich shale samples showed higher adsorption capacity compared to clay-poor shale samples. The predominant clay mineral in those shale samples was illite. The platy shape of illite provided the surface area for enhanced adsorption capacity. This study concluded that in gas-condensate systems of liquid-rich shales, the produced gas becomes leaner during production and signicant volumes of condensates, which contain predominantly heavy components, are left behind in the reservoir. The gas-condensate core- ooding experiment showed that composition of the owing mixture below the dew-point pressure contained less heavy components along the direction of ow. Molecular simulations revealed that the change in gas composition was not only due to condensate dropout and relative permeability e
ects, but also due to the preferential adsorption of heavy hydrocarbons over methane. This means that initial production from shale reservoirs contain both methane and other heavy components from the free phase. However, as reservoir pressure decreases, methane from the adsorbed phase starts to desorb preferentially and the adsorption sites where vi methane molecules used to reside start to accept heavier components. In addition, molecular simulations conducted at subcritical conditions to estimate the vapor and liquid densities of pure hydrocarbons inside 5 and 10-nm pores revealed that rock- uid interactions in the form of adsorption caused the critical pressure and temperature of the conned molecules to decrease. This was observed clearly for methane and ethane. The decrease in the critical properties was a
ected by the size of the pores. For example, the estimated critical pressure and temperature of methane in 5-nm pore were lower than the critical pressure and temperature in 10-nm pore.

Press the Back button in your browser.

Copyright 2016, Maytham I. Al Ismail: Please note that the reports and theses are copyright to their original authors. Authors have given written permission for their work to be made available here. Readers who download reports from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the author, Maytham I. Al Ismail.