Horizontal well shale oil development in the U.S. began in earnest around 2006, first in the Williston basin of the northern U.S. Development expanded to the Eagle Ford in 2007 and has since expanded to several more basins, such as the Permian, DJ, Powder River, Uinta, San Juan and SCOOP/STACK. Today, there are more than 94,000 horizontal shale oil producing wells contributing 7 MMbpd to our nation’s supply.
About half of these wells are located in the Permian and Eagle Ford shale plays of Texas. Production from all these horizontal shale oil wells is characterized by high initial rates and steep declines over the first few years, followed by shallower declines to economic limit. Economic production may continue from eight to more than 15 years, depending upon the basin and well completion characteristics and commodity prices.
An industry data analytics firm has calculated the horizontal shale oil well daily production by year of first output and the number of wells by state, Fig. 1. In the DJ basin, shale oil development began with vertical wells focused on the Ft. Hays-Codell shaley limestone and sandstone sequence in the 1990s and transitioned to horizontal wells completed in the Codell and Niobrara shale in 2009.
Fig. 1. Horizontal shale oil daily oil production by year of first production and number of horizontal shale oil wells by state. (Source Novi Labs)
Today, there are approximately 9,025 Niobrara and Codell shale wells producing, with about 8,625 in the Wattenberg field area of northeastern Colorado and about 400 in southeastern Wyoming near Cheyenne. Figure 2 shows the Niobrara and Codell horizontal shale oil well daily production by year of first production and the number of wells by state.
Fig. 2. Horizontal Niobrara and Codell shale oil well daily oil production by year of first production and number of horizontal shale oil wells by state.
The Codell is a sandy shale situated above the J sand, with a thickness ranging from 5 ft to more than 50 ft. The Niobrara is a thick interval of chalks and marlstones that is ubiquitous across the entire basin. The deepest portion of the basin is north of Denver, and development has been predominantly in the area of the large Wattenberg field, where natural gas in the deeper J-Sand was discovered in 1970. EIA lists the Wattenberg—which covers about 2,000 mi2, from northern Jefferson County to southeastern Weld County, Colorado—as the ninth-largest gas field in the U.S. A stratigraphic column of the productive intervals in the DJ basin is shown in Fig. 3.
Fig. 3. DJ basin stratigraphic column.
The Niobrara shale is a stacked sequence of chalkstones and marlstones, with a gross thickness ranging from 300 ft to more than 900 ft. The Niobrara is the primary target of shale oil development in the DJ basin, with horizontal wells completed individually in one of the chalky intervals.
Initially, horizontal wells were drilled and completed with lateral lengths of 4,000 to 5,000 ft, but lateral lengths have since increased to more than 9,000 ft. Drilling conditions in the DJ basin are simple and low-cost, enabling monobore completions. Fracture stimulations have varied over time, as operators experimented with varying fluid types and proppants, but the most common fracture stimulation treatment method is slickwater with proppant loading of 0.5 to 1.0 ppg. Lateral spacing, likewise, varies but is generally 300 to 600 ft. The thick, multi-chalk intervals of the Niobrara require well stacking, with varying lateral spacing and varying vertical offsets, as dictated by rock thickness and reservoir rock properties.
Political and pandemic setbacks. Development of the Niobrara shale slowed after 2018, when the Colorado legislature enacted Senate Bill 181, which changed the Colorado Oil and Gas Conservation Commission’s (COGCC) focus and makeup. Following its passage, the Colorado governor replaced the COGCC board of directors, and the new board began enacting a series of new rules that greatly restricted development.
The drop in crude prices in March 2020, due to Covid-19, also slowed development. Higher crude prices in 2021 and 2022 have improved economics for shale oil development; however, continued development is hampered by long well permitting times and excessive well setback restrictions. Niobrara shale wells have an average life of about 10 to 12 years at a $60/bbl crude price. Therefore, it is anticipated that more than half of the wells now on production will likely reach their economically limited production level within the next five years.
EOR LIMITATIONS
Enhanced oil recovery has never provided a significant amount of oil production worldwide. According to an International Energy Agency 2018 study, Whatever happened to enhanced oil recovery?, “EOR’s share of global crude oil production seems to have remained broadly stable over time at around 2%.” Reasons for lack of EOR development include lack of concerns over oil scarcity and preference for upstream industry projects that can generate fast returns. Also, reasons include governmental policy incentives, lack of skills and technologies/expertise, in addition to high costs and risks of EOR.
Shale oil development in some U.S. basins, such as the DJ basin, is reaching maturity and as mentioned earlier, many horizontal shale oil wells drilled since 2009 are reaching or have reached depletion. These wells will have to be plugged or abandoned at significant cost, or some action must be taken to increase oil production. Refracs are a consideration; however, the costs of conducting refracs can be prohibitive, i.e., equal to or greater than the fracture stimulation treatment conducted upon completion. And they may not yield sufficient production to recover costs and generate an attractive ROI.
Shale oil enhanced oil recovery (EOR) is in its infancy and has been implemented in a few hundred wells in the Eagle Ford shale. According to Railroad Commission of Texas (RRC) data, as of mid-2022, there are 36 permitted shale oil EOR projects in the Eagle Ford, all of which, except for one, involve the cyclic injection of natural gas. Thirteen shale oil EOR tests have been conducted, or are planned, in other shale basins, such as the Permian, the Williston basin and the SCOOP, with limited success.
Nearly all of these EOR projections have involved the cyclic injection of natural gas, often referred to as huff-and-puff (HnP). In several of these EOR projects, interwell communication, due to natural fractures or fracture stimulation treatments, has limited maximum natural gas injection pressures and resulted in reduced oil recoveries. Enriched natural gas HnP in the DJ basin of the Niobrara shale was evaluated, and an incremental oil recovery of about 13% in 15 years of EOR operation was predicted.
Shale EOR, using natural gas or CO2, requires the use of large, specially designed high-pressure compressor sets that can inject very large volumes of gas or CO2 into the shale reservoir to reach pressures exceeding the minimum miscibility pressure (MMP). These compressor sets are hig- cost, and they require long delivery lead times and, in most cases, cannot be rented.
Natural gas is usually purchased from a midstream company and delivered to the wellsite. Carbon dioxide is generally in short supply and requires a dedicated delivery pipeline and higher costs for transmission and materials. The cyclic nature of injection and production at high rates, and for relatively short periods of time, can be difficult to manage in the field. Injection and production cycles for natural gas and CO2 HnP may be many days or even months, depending upon the compression capacity and wellbore configurations.
Compression fuel consumption is about 7% to 8% of the injected volumes. All of these costs—equipment, injectant, fuel, maintenance, and resulting production downtimes—during the injection cycles make many of these projects uneconomic or only marginally economic. To date, these shale oil EOR projects have been developed by 16 companies, mostly majors and large independent producers, who have the financial capacity and technical expertise required to design and implement them.
SHALE EOR METHODS
Shale Ingenuity has invented and developed two innovative shale oil EOR methods, named SuperEOR and UltraEOR. They involve the injection of a liquid solvent of a specified composition into the shale oil reservoir, at a high rate and for a short injection period (a few days), followed by flowback to a specific minimum wellbore pressure. The solvent is recovered in equipment at the surface and reinjected, thereby reducing costs. Gas produced with the crude oil continues to flow to the midstream facility for NGL recovery. Solvent recovery is highly efficient and low-cost. UltraEOR adds a step at the start of the EOR process that generates a complex fracture network around the wellbore lateral, thereby increasing the contact of the solvent with the shale rock matrix.
Specific composition, or “tuned” liquid solvents recover oil in a manner significantly different from natural gas or CO2. The solvent has a very low solubility or miscibility pressure; it immediately reduces the interfacial tension between the crude oil and rock matrix; it reduces the oil viscosity; it increases the relative permeability to oil; it adds solution gas pressure to the crude oil and becomes a gas when the pressure is reduced, expanding and driving the oil through the shale pores; and also benefits from other forces, such as advective and molecular-diffusion flux. As a result, it is very effective at recovering oil from the shale oil matrix, and high-pressure injection is not required.
The SuperEOR and UltraEOR shale oil EOR processes, like any other EOR process, require a sound understanding of the shale oil reservoir and the wells completed into it. A compositional reservoir simulation model of the reservoir was developed to obtain a production and pressure history match of the reservoir. As with any reservoir simulation model, the more detailed and accurate data available, the more accurate the model forecast results under SuperEOR and UltraEOR performance. Compositional simulation modeling of shale oil reservoirs shows that oil recovery is optimized when the solvent composition is tuned, and the solvent composition may require periodic adjustment for optimum oil recovery as the process continues.
Shale Ingenuity owns patents related to the optimization of oil recovery under cyclic injection of natural gas and has additional patents in application relating to the optimization of oil recovery under cyclic injection of solvents.
SuperEOR compositional reservoir simulation process. To illustrate the shale oil EOR potential of the SuperEOR and UltraEOR processes in the Niobrara formation of the DJ basin, a representative Niobrara horizontal shale oil producing well was selected in the central-northeastern area of Wattenberg field.
The selected well was completed in the Niobrara A interval, with a 4,000-ft lateral and a 25-stage slickwater fracture stimulation treatment. At the time of the simulation, the well had a 5½-year production history. Well completion, open-hole logs, PVT, production and surface pressure data were used to develop a compositional simulation model. The PVT data were used to develop a tuned equation of state used by the model to track the individual hydrocarbon components in the reservoir and the injection and production of the SuperEOR and UltraEOR liquid solvent. The production history of the well is shown in Fig. 4.
Fig. 4. Production history of case study well.
A nine-component EOS model was developed with the PVT data from a well in the immediate vicinity of the modeled well. The nine pseudo components were N2, CO2, CH4, C2H6, C3H8, C4, C5, C6, and C7+. Table 1 shows the composition and selected EOS parameters. Figures 5a and 5b show experimental data versus simulation data match for constant mass expansion experiment. Figures 6a and 6b show experimental data versus simulation data match for differential liberation experiment.
Table 1. Pseudo components and selected EOS parameters.
Fig. 5a and 5b. Experimental data versus simulation data match for constant mass experiment.
Fig. 6a and 6b. Experimental versus simulation data match for differential liberation experiment.
A compositional, single-porosity model was utilized to match the historical pressure and production of the selected well. An element of symmetry approach was used, with a model comprised of five fracture stages, or 20% of the total number of fracture stimulation treatment stages. Results were then scaled up to 25 fracture stages for the entire well. Reducing the number of modeled frac stages reduces model complexity and simulation computation time.
The modeled Niobrara reservoir has a top at 6,700-ft depth, with a pay thickness of 270 ft. The model has 10 layers of constant grid thickness. Reservoir temperature is 190°F. Initial reservoir pressure is 4,000 psig. Absolute shale matrix permeability was estimated to be .0001 md. Hydraulic fractures are simple planar fracs, one per frac stage, having a half-length of 225 ft and frac height of 81 ft.
Logarithmically spaced, locally refined grid blocks were used in the model to accommodate the large transient response from the low-permeability shale matrix to the high-permeability propped fractures. The history match of primary oil, gas and water production, and surface production pressure, was completed, using an automated history matching tool. The tuning parameters included matrix permeability, relative permeability, porosity, hydraulic fracture geometry, hydraulic fracture conductivity, and formation compressibility.
Bayesian probabilistic history matching required approximately 500 simulation runs to achieve an acceptable history match with less than 1% history match error. Figures 7a through 7d show the history match error plot and matches on oil production rate, gas production rate, water production rate and BHP. The history-matched model was utilized to evaluate and optimize the forecast oil recovery via the SuperEOR solvent EOR process. Several scenarios were evaluated, varying injection rate, injection pressure, soak time, flowback rate, flowback pressure, and solvent injectant composition. The SuperEOR forecast was run for a period of approximately 13 years. Optimization of the NPV for a SuperEOR project was not conducted, as this would require several other assumptions for such a project, such as the number of wells in SuperEOR operation, the cost of solvent and the solvent recovery system.
Fig. 7a-7d. Historical production and pressure versus history match case.
However, an unoptimized SuperEOR development scenario is offered, assuming an eight-well project consisting of wells completed identical to the modeled well. Maximum injection BHP was assumed to be 3,900 psi. The single-well (five stages) forecast oil production for the selected constant composition solvent solution, with a maximum injection rate of 330 bpd, injection period of 26 days, maximum production rate of 1,500 Mcfd, production period of 30 days and minimum production BHP of 750 psi, Fig. 8. This projection assumes no soak time, as other simulation runs show that soak time has no significant impact on oil recovery.
Fig. 8. Historical oil production and improved crude output under SuperEOR.
Cumulative oil production, for the five modeled frac stages, under primary (pressure depletion) production with no assumed economic limit, and after 13 years of SuperEOR operation, is shown as the black line, Fig. 9. After five years of SuperEOR operation, cumulative oil is forecast to be 27,600 bbl or 146% of the cumulative oil recovered after five years of continued primary oil production. After 13 years of SuperEOR operation, cumulative oil from the well is forecast to be 47,658 bbl or 172% of the cumulative oil recovered after 13 years of continued primary oil production.
Fig. 9. Cumulative primary oil production and increase output with SuperEOR.
Primary EUR for the entire well was forecast to be 78,600 bbl of oil, assuming an economic limit rate of production of 10 bopd. Therefore, the well production at the start of SuperEOR operation was within one year of reaching the calculated economic limit, Fig. 9. Thus, implementation of SuperEOR could extend the life of the well by more than 10 years and increase the EUR from 78,600 to 238,290 bbl of oil, a 300% increase.
Fig. 10. Cumulative primary oil production and with SuperEOR at varying maximum injection BHP.
A sensitivity analysis to maximum bottomhole injection pressure was run, to assess the impact on oil recovery, if the SuperEOR process were to be conducted at lower maximum injection BHP. This is an especially important consideration, as several natural gas HnP EOR projects in other shale oil basins have shown inter-well communication via natural or hydraulic stimulation treatment fractures once the BHP reaches a certain high level during injection. Figure 10 shows the cumulative oil production under primary and SuperEOR at maximum injection BHP of 3,500, 3,000, 2,500 and 2,000 psi. The oil recovery after 13 years of SuperEOR operation is reduced by 8.3%, 13.3%, 18.9% and 24.2%, respectively. The simulation model also shows the areas of the reservoir where oil is being recovered by the process. Figure 11 shows a top view of the oil saturation of the middle layer of the reservoir after 13 years of SuperEOR process operation, indicating that oil recovery is localized almost entirely to those grid blocks between the hydraulic fractures.
Fig. 11. Top view of middle block sim grid showing oil saturation after 13 years SuperEOR.
Value added. The process has multiple advantages over natural gas or CO2 HnP EOR. Most notably, it recovers far more oil in less time. The cost of recovery is also much lower, as the solvent injectant is lower-cost than natural gas or CO2 and is purchased only once; recovery of the solvent is almost 100%, and the cost of recovery is low. Cycle times are short, and as a result, wells experience little downtime and higher effective daily oil production rates. A multi-well process can be designed to enable staggered injection and production schedules to afford near constant flowback and injection rates. Artificial lift equipment is not needed, as the wells flow at rates sufficient to efficiently lift oil out of the wellbores.
An unoptimized SuperEOR project economic analysis for an eight-well project was conducted, assuming 9,000-ft completed lateral lengths in each well. The key project assumptions and results and the forecast oil production for the project are shown in Fig. 12. As shown at a net $83/bbl price, and providing no economic benefit for any additional natural gas that may be produced via SuperEOR, the unrisked, unoptimized process should generate an attractive ROI on capital expenditure.
Fig. 12. SuperEOR project assumptions and economic results.
Modeling of the SuperEOR process has been completed in most of the U.S. shale plays, and SPE 206186 and SPE 208312 describe similar modeling work in the Midland basin Wolfcamp D and Eagle Ford shales. Core tests of the SuperEOR process have been conducted on Eagle Ford and Midland basin Wolfcamp B shale, demonstrating about 90% oil recovery after six injection and production cycles. These core tests and results are presented in SPE 209348. The SuperEOR process has also been field-tested successfully in other shale oil basins.
Compositional reservoir simulation of the UltraEOR process. The UltraEOR process was modeled, using the same procedure, with the addition of a complex fracture network situated in the areas between the primary completion planar stage fractures. The CMG GEM compositional reservoir simulation model provides a feature enabling the delineation of a complex fracture network. In this case, we assumed the generation of a complex fracture network having fracture widths of .001 feet and an effective permeability of 17.5 millidarcies. A comparison of the five-stage model with planar (SuperEOR) and complex (UltraEOR) fractures are shown in Fig. 13.
Fig. 13. Top view of middle block SuperEOR and UltraEOR simulation grids.
Complex fracture generation in the areas between the primary completion planar fracture stages may be achieved with a proprietary process that recognizes the altered stress orientations of the shale. This fracture generation process has been proven and demonstrated successfully in many wells worldwide.
Fig. 14. Forecasted cumulative oil recovery under primary SuperEOR and UltraEOR.
This complex fracture network effectively doubles the fracture surface area in the region between the five primary planar fractures. As a result, the cumulative oil recovery is about double the oil recovery from SuperEOR. Figure 14 shows the cumulative oil versus time for the five fracture stages modeled, under primary recovery, SuperEOR recovery and UltraEOR recovery. As shown, after 13 years of EOR operation, SuperEOR may increase oil recovery by 1.8 times versus primary and UltraEOR may achieve 3.7 times as much oil as primary recovery.
Fig. 15. UltraEOR project assumptions and economic results.
An unoptimized UltraEOR project economic analysis for an eight-well project was conducted, assuming 9,000-ft completed lateral lengths in each well. The key project assumptions and results and the forecast oil production for the project are shown in Fig. 15.
CONCLUSIONS
The two novel shale oil EOR processes, SuperEOR and UltraEOR, should enable oil recovery that is far superior to natural gas or CO2 HnP EOR, and at much lower cost. SuperEOR and UltraEOR have several important operational advantages of greater simplicity and lower capital and operating cost. These processes require rigorous compositional reservoir simulation modeling to design, and its use during SuperEOR and UltraEOR operation for optimized for maximum oil recovery as the process is placed into field operation.
ACKNOWLEDGEMENTS
Shale Ingenuity is grateful for the collaboration with the Computer Modeling Group and the application of its advanced compositional reservoir simulation software in the development of the SuperEOR and UltraEOR shale oil EOR processes. Also, thanks to Novi Labs for access to their well database , enabling the generation of the Niobrara production and well graphs. This article contains excerpts from SPE paper 213082-MS, “Advanced, superior shale oil EOR methods for the DJ basin,” presented at the SPE Oklahoma City Oil and Gas Symposium, held at the Oklahoma City Convention Center, April 17-19, 2023.
About the Authors
Robert Downey
Shale Ingenuity
Robert Downey has over 30 years of experience in upstream oil and gas and energy technology development. Prior to Shale Ingenuity, he held numerous executive and technical professional positions at Amoco, Encana, Synthetic Genomics, Ciris Energy and Gunnison Energy. He has been awarded several patents in coal bioconversion, reservoir simulation modeling, reservoir engineering and well drilling and completions. Mr. Downey earned a BS degree in petroleum engineering from the Colorado School of Mines.
Dr. Jim Erdle
The Computer Modeling Group
Dr. Jim Erdle has 49 years of industry experience primarily in reservoir and production engineering-related positions in the upstream services and software segments. Early in his career, he was involved with some of the industry’s leading advances in well testing design, monitoring and interpretation technology including closed chamber and surface pressure readout (SPRO), drill stem testing and production enhancement via NODAL analysis. Dr. Erdle also worked on stimulation treatment design, monitoring techniques and production surveillance software. He recently retired from the Computer Modeling Group after 25 years, most recently as the company’s V.P. of software sales and support for the U.S. and Latin America. He holds a BS degree (1971) and a PhD (1974) in Petroleum Engineering from Penn State University
Kiran Venepalli
The Computer Modeling Group
Kiran Venepalli is a product manager at the Computer Modeling Group specializing in reservoir modelling and simulation. In this role, he plays a pivotal role in shaping the direction and development of CMG's solutions in line with the goals of energy transition. He is also responsible for supervising unconventional solutions. This entails exploring and developing unconventional oil and gas resources, such as shale gas and tight oil, with an emphasis on leveraging technological advancements to improve extraction techniques and minimize environmental impact. Mr. Venepalli has 12 years of industry experience and holds a master’s degree in petroleum engineering from the University of Alaska Fairbanks and a bachelor’s degree from JNTU, India.
