Gas Transport, Sorption and Fracture in Shale

Open Access
Li, Xiang
Graduate Program:
Energy and Mineral Engineering
Doctor of Philosophy
Document Type:
Date of Defense:
February 16, 2016
Committee Members:
  • Derek Elsworth, Dissertation Advisor
  • Derek Elsworth, Committee Chair
  • Chirs Marone, Committee Member
  • Zuleima T Karpyn, Committee Member
  • Li Li, Committee Member
  • enhanced gas recovery
  • hydraulic fracturing
  • breakdown pressure
  • fracture complexity
  • adsorption
  • swelling
  • stress redistribution
Shale gas has become an increasingly important source of natural gas (CH4) in the United States over the last decade. As an unconventional resource, various stimulation techniques including hydraulic fracturing and enhanced gas recovery have been proposed to maximize production. This study examines some of these techniques combining experiments and models. Part I of this dissertation (Chapter 1) examines the use of CO2 for enhanced shale gas recovery (CO2-ESGR) using a dual porosity dual permeability model to better understand its feasibility and effectiveness. Part II (Chapter 2) explores the use of gas stimulants for hydraulic fracturing to assess the form and behavior of fractures in shale driven by different gas compositions and states. Part III (Chapter 3) examines the evolution of permeability in artificially propped fractures in Green River Shale for native CH4 contrasted against sorbing CO2, slightly sorbing N2 and non-sorbing He, specifically to examine the deleterious influence of proppant embedment. Together, the findings of these experiments and analyses aid in the understanding of proppant embedment and fracture diagenesis in shales. Finally, part IV (Chapter 4) examines the role of stress reorienation in aiding productivity increases in the refracturing of previoslu fractured wells in shales to determine the optimal timing of refracturing as well as in quantifying its potential improvement. Chapter 1 explores the roles of important coupled phenomena activated during gas substitution especially vigorous feedbacks between sorptive behavior and permeability evolution. Permeability and porosity evolution models developed for sorptive fractured coal are adapted to the component characteristics of gas shales. These adapted models are used to probe the optimization of CO2-ESGR for injection of CO2 at overpressures of 0MPa, 4MPa and 8MPa to investigate magnitudes of elevated CH4 production, CO2 storage rate and capacity, and of CO2 early-breakthrough and permeability evolution in the reservoir. For the injection pressures selected, CH4 production was enhanced by 2.3%, 14.3%, 28.5%, respectively, over the case where CO2 is not injected. Distinctly different evolutions are noted for permeability in both fractures and matrix due to different dominating mechanisms. Fracture permeability increased by ~1/3 for the injection scenarios due to the dominant influence of CH4 de-sorption over CO2 sorption. CO2 sequestration capacity was only of the order of when supercritical for a net recovery of CH4 of m3. We investigated the potential of optimal CO2-pulsed injection to enhance CH4 production (absolute mass recovered)-without the undesirable effects of CO2 early-breakthrough and also minimum cost on CO2 injection. This utilizes the competitive sorptive behavior between CH4 and CO2, can also reduce the potential for induced seismicity hence the entire system can be near net neutrality in terms of its carbon and seismic footprint. In Chapter 2, fracturing is completed on cylindrical samples containing a single blind axial borehole under simple triaxial conditions with confining pressure ranging from 10~25MPa and axial stress ranging from 0-35MPa ( ). Results show that: 1) under the same stress conditions, CO2 returns the highest breakdown pressure, followed by N2, and with H2O exhibiting the lowest breakdown pressure; 2) CO2 fracturing, compared to other fracturing fluids, creates nominally the most complex fracturing patterns as well as the roughest fracture surface and with the greatest apparent local damage followed by H2O and then N2; 3) under conditions of constant injection rate, the CO2 pressure build-up record exhibits condensation between ~5-7MPa and transits from gas to liquid through a mixed-phase region rather than directly to liquid as for H2O and N2 which do not; 4) there is a positive correlation between minimum principal stress and breakdown pressure for failure both by transverse fracturing ( ) and by longitudinal fracturing ( ) for each fracturing fluid with CO2 having the highest correlation coefficient/slope and lowest for H2O. We explain these results in terms of a mechanistic understanding of breakdown, and through correlations with the specific properties of the stimulating fluids. In Chapter 3, experiments are conducted on 1inch diameter, 2 inch long split cylindrical samples sandwiched with proppant at a constant confining stress of 20 MPa and with varied pore pressure — increases in pore pressure represent concomitant decreases in effective stress. Permeability and sorption characteristics are measured by pulse transient methods. To explore the effect of swelling and embedment on fracture surface geometry, we measure the evolution of transport characteristics for different proppant geometries (single layer vs. multi-layer), gas saturation, and sample variance. In order to simulate both production and enhanced gas recovery processes, both injection and depletion cases are investigated. Embedment morphology is measured by scanning white light interferometry and characterized by roughness through a variety of surface roughness parameters including Sa, Sq and Sz as well as complexity through fractal dimension. For both strongly- (CO2, CH4) and slightly-adsorptive gases (N2) the permeability first decreases when gas pressure increases because of swelling. It then increases beyond the Langmuir threshold due to the over-riding influence of effective stresses. Due to its highest adsorptive affinity, CO2 returns the lowest permeability among these three gas permeants. Compared to the case of a mono-layer propped sample, the sample with four layers exhibits less swelling as implied by its elevated k/k0 ratio and reduced embedment surface roughness and complexity. Interestingly the duration of gas exposure and saturation tested here which is up to ~20hrs does not have a significant influence on permeability for either adsorptive or non-adsorptive gases. Permeabilities recovered from both injection and depletion cycles generally overlap each other and are repeatable with little hysteresis. This suggests the dominant role of reversible swelling over irreversible embedment. Permeability variance between different samples is of the order of ~1.5 - 2 times but with repeatable trends and order of magnitude parity. Gas permeant composition and related swelling effects exert important influences on the permeability evolution of shales under nominally in situ conditions. In Chapter 4, key factors include the time dependency of the stress reorientation, the threshold of σhmax/σhmin for the presence of the stress reversal region, the influences of permeability anisotropy/heterogeneity, pressure drawdown and rock-fluid properties - these factors are investigated. The results show that stress reorientation develops as soon as the reservoir begins to produce and the stress reversal region extends with time to a maximum extent following which it retracts until the direction of the maximum principal stress gradually returns to the initial state. The optimal refrac timing and the size of the stress reversal region are either positively or negatively correlated to the factors examined here. σhmax/σhmin ratio and Poisson’s ratio are negatively correlated to the size of stress reversal region as well as the timing of optimal refracturing; permeability magnitude and porosity have no influence on the size but are negatively and positively correlated to the timing, respectively; permeability anisotropy is positively correlated to the size and negatively correlated to the timing if the permeability is greater in the direction perpendicular the initial fracture, however it is negatively correlated to the size and positively correlated to the timing if the permeability is greater in the direction parallel to the initial fracture; permeability heterogeneity has no influence on the size nor the timing; pressure drawdown and Biot coefficient are positively correlated to the size as well as the timing.