Unconventional Resources
Numerical and Laboratory Investigations for Maximization of Production from Tight/Shale Oil Reservoirs: From Fundamental Studies to Technology Development and Evaluation Last Reviewed
July 2018

ESD14089

Goal
The main project goal is to conduct multi-scale laboratory investigations and numerical simulations to (1) identify and quantify the various mechanisms involved in hydrocarbon production from tight systems; (2) describe the thermodynamic state and overall behavior of the various fluids in the nanometer-scale pores of tight media; (3) propose new methods for producing low-viscosity liquids from tight/shale reservoirs; and (4) investigate a wide range of such possible strategies, identify the promising ones, and quantitatively evaluate their performance. Laboratory research will include nano- to core-scale studies, and numerical simulations will examine molecular to field-scale conditions.  By covering the spectrum from fundamental studies to technology development and evaluation, the project team proposes to gain a deeper understanding of the dominant processes that control production from tight reservoirs and to develop a compendium and document the effectiveness of appropriate production strategies. 

Performer
Lawrence Berkeley National Laboratory (LBNL)

Background
Gas production from tight gas/shale gas reservoirs over the last decade has been met with spectacular success with the advent of advanced reservoir stimulation techniques (mainly hydraulic fracturing), to the extent that shale gas is now among the main contributors to US hydrocarbon production. This remarkable success has not been matched by similar progress in the production of (relatively) low-viscosity liquid hydrocarbons (including condensates) because of the significant challenges to liquid flow posed by the ultra-low permeability (and the correspondingly high capillary pressures and irreducible liquids saturations) of such reservoirs. These difficulties have limited liquids production to a very low fraction (usually <5%) of the resources-in-place. Increasing the recovery of liquids from these ultra-low permeability systems even by 50-100% over its current very low levels (to a level that is still low in absolute terms, but very significant in relative, hence economic, terms) will not only increase production and earnings, but will also have considerable wider economic implications, as the enhanced recovery will affect reserves and the valuation of companies.

In this multi-phase research effort, LBNL will conduct multi-scale laboratory investigations (nano- to core-scale) and numerical simulations (from molecular to field-scale) to: (1) identify and quantitatively describe mechanisms that control fluid flow and the various system interactions in oil shales; (2) quantitatively describe the behavior of the fluids involved in the production process in the extremely small pore space of shales, leading to promising strategies for enhanced liquid hydrocarbon recovery; (3) analyze the transport of proppants through realistic fractures (including inclined and sharply-angled ones) and evaluate the proppant long-term fate (embedment or pulverization); (4) describe the PVT behavior of fluids in shales, and propose novel approaches as new methods for enhanced production of low-viscosity fluids from tight/shale oil reservoirs after confirmation by laboratory (core-scale) experiments; (5) remove from further consideration potential production strategies that hold limited (if any) promise; and (6) identify and focus study on strategies that appear to have potential for significant enhancement of environmentally-conscious hydrocarbon production (in terms of maximization of both production and recovery), and numerically evaluate their large-scale and long-term performance.

Impact
Successful identification of the processes that control production within tight reservoirs may result in processes and methods that could increase production by 50 to 100 percent over the current low recovery rates of approximately 5 percent.  The impact to the industry will be significant and potentially dramatic because of an increase in the amount of hydrocarbon produced  and increases in reserve estimates. The result will be reflected in increased economic benefits to companies and consumers.

Accomplishments
In Phase I of this project, LBNL identified the parameters, objectives, and metrics of this study. Numerical simulations were conducted to evaluate production from unfractured/naturally fractured reservoirs and hydraulically fractured reservoirs (Figure 1). Initial simulations serve as reference cases for future research. The “success” of enhanced shale oil recovery will be achieved by an increase in recovery of at least 50% over the life of a shale oil well (3 to 5 years) when compared to the hydraulically fractured reservoir reference case.

  Results of the reference (base) case numerical simulations over a range of permeabilities
  Figure 1: Results of the reference (base) case numerical simulations over a range of permeabilities where solid lines represent the unfractured system and dashed lines represent the hydraulicaly fractured reservoir.

Phase I work also included field scale numerical simulations to assess the recovery enhancement associated with displacement and viscosity reduction in parallel horizontal wells (Figure 2). Simulations have been completed to evaluate enhanced liquid recovery by means of nitrogen (N2), methane (CH4), and carbon dioxide (CO2) displacement and viscosity reduction methods (gas dissolution and thermal stimulation) over a range of permeability. Displacement results indicate very little difference in recovery between N2 and CH4 gasses despite the affinity for CH4 dissolution in oil and its corresponding density and viscosity reductions that are beneficial to recovery (Figure 3). The lack of recovery contrast between the gasses is thought to be attributed to the difficulty of CH4 diffusion into the oil during displacement. Therefore, additional displacement simulations have been completed to investigate the production potential of oil with significant amounts of dissolved CH4 (Figure 3). Results indicate superior recovery of “gassy oil” compared “dead oil” and a much faster recovery over the range of permeability. Results from the CO2 displacement studies indicate greater recovery enhancement with CO2 when compared to N2 and CH4.

Viscosity reduction results associated with thermal stimulation indicate production enhancement after significant lead time and further recovery enhancement when thermal stimulation in initiated prior to production (Figure 4). However, this must be further evaluated against the energy requirements to raise the temperature of the shale system. A new semi-analytical solution (Transformational Decomposition Method [TMD]) has been developed to address the problem of 3D flow through hydraulically fractured media. The TMD solution was validated using published data and can be used to analyze well tests and determine flow properties of producing reservoirs at any desired simulation time without the computational expense of forward time integration. In Phase 1, molecular dynamics (MD) simulations were modified to include chemical reactivity and flow effects in order to understand pore-scale interactions between hydrocarbon molecules and clay surfaces. Exploratory MD runs were completed to determine reactivity in the clay pore molecular model system prior to the introduction of flow.

Detailed schematic of the shale reservoir investigated in the numberical simulations of enhanced oil recovery by means of gas displacement

Figure 2: Detailed schematic of the shale reservoir investigated in the numberical simulations of enhanced oil recovery by means of gas displacement.


Gas displacement and dissolution results from numerical simulations
Figure 3: Gas displacement and dissolution results from numerical simulations showing (a) CH4 and N2 displacement simulations with no discernible difference in production between the two gasses and (b) the effect of dissolved CH4 on enhanced oil recovery for various matrix permeabilities in fractured and unfractured media with superior recovery of "gassy" vs. "dead" oil.

Numerical simulation results of enhanced oil recovery from shale by means of thermal stimulation (viscosity reduction).
Figure 4: Numerical simulation results of enhanced oil recovery from shale by means of thermal stimulation (viscosity reduction).

In Phase I, the laboratory systems for the core-scale enhanced recovery experiments were designed, and initial experiments were completed on Niobrara shale and a well-characterized ceramic. Initial supercritical CO2 (SC-CO2) displacement experiments with the Niobrara shale produced a very small quantity of oil. In order to quantify the process at the laboratory scale, it was determined that a large, well-characterized sample (~1 m3) would be required with an excessive experimental run time. In response, the experimental system was redesigned to ensure lab-scale test durations and sufficient recovery from a well-characterized ceramic medium with known pore space, mineral phase wettability, hydrocarbon content, and starting conditions (Figure 5). Gas displacement results from the redesigned system agree with model results and indicate enhanced recovery with CO2 compared to CH4 and N2 and enhanced recovery with CH4 compared to N2.

Schematic of the redesigned core-scale laboratory system
Figure 5: Schematic of the redesigned core-scale laboratory system

The Advanced Light Source facility at LBNL was used for a series of nano-scale characterization and visualization studies on high quality Niobrara shale samples in Phase I. A comprehensive characterization study was completed on the Niobrara shale via electron microscopy, x-ray diffraction, and x-ray computed tomography (CT) to provide the mineralogy, chemical composition, and texture/microstructure of the samples. Characterization results indicate the Niobrara samples are carbonate rich (55.3 weight %) with a texture highly influenced by the carbonate distribution. While clay content in the samples is typical of many shales (24.1 weight %), chlorite and kaolinite are absent. Organic-rich particles are scattered throughout the samples but do not follow bedding planes. Following sample characterization, a series of imaging experiments were completed to understand micro-scale processes related to oil production techniques from tight shales. Fracture imaging experiments were completed to understand the relationship between textural features and fracture generation. Results from the fracture imaging experiments indicate that generated fractures are irregular and largely controlled by the stress state and bedding planes; however, secondary fractures appear to preferentially form in clay-rich layers (Figure 6). Additional microCT experiments were conducted to understand (1) fracture evolution during the flow of carbonated water and (2) the effect of sweeping a propped fracture with SC-CO2 (Figure 7). Results from the carbonated water flow test indicate increased porosity and permeability associated with worm-holing and preferential dissolution of carbonate-rich structures. Reacted water flood experiments in proppant filled fracture sample demonstrate the dissolution of carbonate along the fracture face and a lack of dissolution at proppant-grain boundaries. In the SC-CO2 sweeping test, results suggest that water in the sample cannot be easily displaced by SC-CO2 and that the effectiveness of SC-CO2 is strongly limited by the presence of trapped water.

Results from the micro-scale fracture imaging experiments
Figure 6: Results from the micro-scale fracture imaging experiments showing (a) fractures appear to be irregular and are primarily controlled by the applied stress state with no apparent interaction with the microstructure and (b) fractures appear irregular and orient with the stress state but the main crack was generated along a bedding plane between a carbonate-rich layer (left) and clay-rich layer (right). The clay-rich layer also contains a number of secondary fractures, while the carbonate-rich layer remains intact.

Local thickness aperature maps of the fracture at different stages of fracture evolution during the flow of carbonate water
Figure 7: (a) Local thickness aperture maps of the fracture at different stages of fracture evolution during the flow of carbonate water. Each step covers the whole sample (~3/8” x 1”) with the inlet at the bottom of the sample. (b) Cross-sections showing the fracture region before (A) and after (B) SC-CO2 sweeping of the propped fracture. Panel (C) shows the baseline image with color highlights in the narrow near-fracture regions that show minor modifications.

The Phase II research effort began on October 1, 2016. In Phase II, field scale simulations continue to be used to investigate a range of enhanced oil recovery techniques. Phase II simulation efforts have focused on the identification of production methods that hold limited promise for removal from futher consideration. Results of these simulations indicate poor performance of water injection/drive, steam injection/drive, and water-alternating-gas drive. These behaviors appear to be consistant across a wide range of injection rates, injection pressures, injection schedules/intervals, and reservoir properties. As such, these methods should be abandoned during future considerations of enhanced production techniques in tight/shale oil reservoirs. In order to quantify the most practical and least practical production enhancement methods, simulations have been expanded to explore additional displacement processes, methods of viscosity reduction, and combinations of these processes and their effect on production. Recent simulations have been expanded to evaluate the effects of various gases and injection strategies with heavier/more complex oil phases. While interpretation is still underway, results appear to agree with earlier investigations that suggested the superiority CO2 and positive effects of CH4.

In Phase II, molecular dynamics simulations have been recalibrated to a larger scale system frame, having a pore length of 15 nm in a single montmorillonite crystal. The larger scale model system has more than 60,000 atoms in the unit frame and will allow for flow simulations with reactive potentials that allow chemical reactions to occur between the fluid and pore walls, as well as within the fluids (Figure 8). Once the system is stable, evaluation of a larger range of chemistry will be sought with attention to edge-plane reactivity. One modification of the original plan will be to add a larger concentration of hydrocarbons to examine the details of clustering and associations with surfaces. So far the work has featured relatively low concentrations of reactant molecules, in aqueous solution. The other end of the possible fluid space (i.e. oil-based), requires exploration. It may also be possible to add CO2 to the system to explore its reaction on the molecular level. Ultimately, LBNL will attempt to leverage electron microscopy work performed outside this project as available, to connect MFD results to real-world visualizations of the 2-5 nm scale. These results will be compared to earlier Phase I results, and will also examine the behavior of less soluble alkanes, and the molecular behavior of high-organic-content fluids.

esd14089-figure8.png
Figure 8: 6x6x6 nm clay cell model with 3x3x2 nm pore used in past simulations with 10,000 atoms. New model will increase size to more than 60,000 atoms with a 3x15x2 nm pore.

In Phase II, the core-scale laboratory system has been updated to include temperature control and the ability to collect fluids from the top or bottom of the apparatus (Figure 9). These modifications to the experimental system allow for the evaluation of enhanced oil recovery techniques that include temperature and gravitational effects. LBNL performed and repeated Light Tight Oil (LTO) production tests using a number of injected fluids including supercritical CO2, water, methane, nitrogen, and helium, at both room temperature and elevated temperatures. Oil production was very high for the denser injected fluids (water, SC-CO2), and extremely good for injected water upon depressurization. The physics of this process are under investigation, as the magnitude of the depressurization was not expected to have an effect in this case, yet the effect was observed. The physics for the LTO production for SC-CO2 and water injection will require additional testing and model evaluation to identify processes and whether the process is real, or an artifact of the setup. A total of 62 tests have been performed to date assessing gas dissolution, depressurization, and oil imbibition.

esd14089-figure9.png
Figure 9: Modified core-scale laboratory system.

Work was also initiated in Phase II to analyze micro-scale proppant transport and fate through laboratory experimentation and numerical simulation. Recent progress on the micro-scale laboratory experiments has focused on proppant transport and fate in Eagle Ford, Niobrara, and Marcellus Shales under progressively increasing uniaxial stress conditions using an in-situ synchrotron X-ray microCT and mini-triaxial cell (Figure 10). Preliminary results from the experiments indicate that the (1) Eagle Ford shale becomes highly fractured, with proppant both embedding in rock and breaking; (2) Niobrara shale tends to break the proppant with limited fracturing in the shale; and (3) Marcellus shale demonstrates an intermediate behavior, with both fracturing of the rock and proppant. At the end of the experiment, a significant aperture of the propped fracture remains present in the Niobrara and Marcellus shales, while the fracture aperture in the Eagle Ford was greatly reduced. Results suggest that the abundance of calcite plays a significant role in proppant pulverization, while proppants tend to embed in clay-dominated rocks. Further analysis of the experimental results has led to the quatification of fracture aperture evolution as a function of the differential pressure, and modeling the evolution of permeability with fracture closure. Thus far, experiments have focused on the role of shale type during the closure of propped fractures. Future laboratory experiments will also address the role of bedding orientation and proppant type on the evolution of fractures under increasing stress conditions.

esd-14089-figure-10.png
Figure 10: X-ray microCT experimental system and imaging example from the Eagle Ford Shale during compression

Considerable work has been conducted to develop the numerical methods necessary to model proppant transport. A 2D/3D numerical model of fluid flow and accompanying proppant transport has been developed in Phase II. Preliminary simulations have demonstrated the ability to represent the moving fluid front and transport of proppants in vertical and horizontal fractures (Figure 11). This modeling work focuses on three major challenges (1) fluid lag behind the fracture tip during the fracturing process, (2) two-way coupled proppant transport inside of the fracturing fluid, and (3) flow of fluid and proppants through intersections of fractures.

esd-14089-figure-11.png
Figure 11: Example of a microscale simulation of flow through a segment of a branch with proppants. Proppants are represented by void spaces in this figure.

A laboratory-scale proppant transport visualization system has been designed, fabricated, and tested for laboratory experiments (Figure 12). Recent modifications have addressed several issues identified with the original system, and preliminary experiments are underway.

schematic-photo-proppant-visualization-system.png
Figure 12: Schematic and photo of the current proppant visualization system. (click to enlarge)

Current Status (July 2018)
The Phase II research effort expanded upon research conducted in Phase I to further investigate mechanisms that control fluid flow in oil shales (including proppant transport and long-term fate) and evaluate improved production strategies that have significant potential to enhance environmentally-conscious hydrocarbon production. The laboratory systems and numerical models have been developed and tested and experiments are currently underway. Field scale simulations have been expanded to evaluate the effects of various gases and injection strategies with heavier/more complex oil phases. Core-scale laboratory experiments have been completed to assess LTO production tests using a number of injected fluids, including supercritical CO2, water, methane, nitrogen, and helium—at both room temperature and elevated temperatures. Significant progress has been made on the proppant transport and fate experiments, and additional experiments and simulations are currently underway.

Project Start: October 1, 2014
Project End: September 30, 2018

DOE Contribution: $899,000
Performer Contribution: $0

Contact Information:
NETL – Stephen Henry (stephen.henry@netl.doe.gov or 304-285-2083)
Lawrence Berkeley National Laboratory – George Moridis, (gjmoridis@lbl.gov or 510-486-4746)

Additional Information:

Numerical and Laboratory Investigations for Maximization of Production from Tight/Shale Oil Reservoirs: From Fundamental Studies to Technology Development and Evaluation (Aug 2017) [PDF]
Presented by Matt Reagan, Lawrence Berkeley National Laboratory, 2017 Carbon Storage and Oil and Natural Gas Technologies Review Meeting, Pittsburgh, PA

Phase I Final Report (July 2016) [PDF] 

2015 Year End Progress Report  [PDF]

Topical Report - Definition of Metrics and Methodology for Screening Production Strategies [PDF]