Multiscale and Multicomponent Flow and Storage Capacity Investigation of Unconventional Resources

Advisor

Anthony R. Kovscek

Abstract

The goal of this research was to understand the complex interplay of gas transport and storativity in nanoporous shale formations. The potential of carbon storage and enhanced gas recovery using pure and multicomponent preferential adsorption studies was examined through original custom designed static and dynamic experimental setups. Results were validated with a molecular simulator and also confirmed using a novel multiscale imaging workflow to characterize storage capacity. This work improves our fundamental knowledge of the physical and flow characteristics of shale reservoirs and is essential for accurate reserve estimation, future production forecasts, and ultimately better field development and planning.
A novel high precision volumetric gas adsorption apparatus was first constructed in house to investigate multicomponent gas sorption at the core level. The thermodynamically closed system allowed for simultaneous measurements of three fundamental rock properties at the core scale including porosity, permeability, and excess adsorption. Experiments were carried out to measure the preferential adsorption of different components in a gas mixture on intact core samples from the Haynesville and Barnett shale plays. Based on the pore size distribution (PSD) of the sample, the experimental observations were confirmed using a grand canonical Monte Carlo (GCMC) simulation employing a slit pore model composed of parallel planar graphitic surfaces. A multiscale approach was then adopted where experimental characterization at the core level was linked with a digital laboratory through molecular modeling to simulate and predict pure and multicomponent adsorption conditions that were hard to achieve in the lab. Experimental results revealed multilayer adsorption coverage with CO2 in comparison to a monolayer coverage with N2 and CH4. Mixed gas sorption measurements showed preferential adsorption of CO2 over CH4 as indicated by a selectivity coefficient greater than 1. GCMC simulations on pore sizes ranging from 1 nm to 50 nm were consistent with the core-scale experimental results and revealed an increasing selectivity for CO2 over CH4 in model shale pores. N2 injection, however, proved to be an unsuccessful enhanced gas recovery technique due to a preferential adsorption of CH4 over N2.
Dynamic Column Breakthrough (DCB) measurements were also carried out for the first time on idealized shale samples based on a custom-designed system. To better understand the contribution of different shale minerals on flow and storativity, measurements were carried out on composition-controlled shales having known weight percentages of total organic carbon (TOC) and illite. Experimental results reveal an increase in permeability and CO2 adsorption with either increasing TOC or illite content. This is attributed to the complex porous structure of kerogen as well as the interlayering characteristics of clay minerals, resulting in large surface area and pore volume ratios. In fact, DCB experiments revealed the potential for CO2 storage in shale formations with adsorption capacities exceeding that of CH4 by 4 to 12 times depending on the content of TOC and illite. Through a series of low-pressure gas adsorption experiments, it was found that each weight percent increase in TOC has a larger influence on the pore volume and surface area compared to each weight percent increase in illite content. The coupled results clearly establish the comparative role of the organic versus inorganic adsorbing components of gas shales while overcoming the material heterogeneity through the investigation of `idealized` compositions.
Finally, we introduced a novel multiscale (cm to nm) workflow for gas flow and storage capacity imaging of whole shale cores. Gas storage (free and sorbed gas) capacity was investigated at the core-scale with carbon dioxide (CO2) and krypton (Kr) using X-ray computed tomography (CT). Also, two-dimensional tiled images were acquired using a scanning electron microscope (SEM) and stitched together to form one-inch diameter mosaics. Multiscale image registration was then carried out to align the CT data with the SEM mosaics. Energy dispersive spectroscopy (EDS) generated elemental spectra maps and subsequent component maps for regions with either substantial or minimal gas storage to assess the interplay of structural features (e.g., fractures) and matrix composition with respect to gas accessibility and storage. Results reveal the relationship of enhanced storage zones with open fractures and reduced storage regions with secondary mineralization (such as nodules) in the carbonaceous samples. These gas sorption experiments prove the feasibility of dynamic core- to nm-scale CT/SEM/EDS image registration to improve sample characterization. To our knowledge, this is the first investigation of core-scale CO2 gas storage employing multiscale imaging. This work also supports the potential of carbon storage in shale formations as well as guides engineers toward optimal CO2 injection zones for enhanced gas recovery.