Doug Walser has extensive (31 years) Permian Basin, Mid-continent, Appalachia, Rockies and South Texas experience with Dowell Schlumberger; The Western Company of North America; BJ Services and Pinnacle, a Halliburton business line. He has specialized in the calibration of three-dimensional fracture modeling via a number of methods. Recently, he has specialized in the examination and comparison of the various emerging resource plays in North America, and more specifically, plays with liquid hydrocarbons. He has authored 14 papers, and holds three patents in his areas of interest. He can be reached at email@example.com
The New Albany Shale has been marginally attractive as a producer for many years, pre-dating most unconventional shale development efforts by a wide margin.
The first drilling attempts targeting New Albany Shale gas in Indiana occurred in 1885. A relatively large number of fields were drilled in the first half of the 20th century in all three Illinois Basin States.
Traditional completion and stim-ulation practice involved the use of gelatinated nitroglycerine charges and/or hydrochloric acid in open hole sections of vertical wellbores. The success of this process appeared to be related to the connection of the wellbore with naturally occurring open fractures.
Experimentation with water-based fracturing fluids began in the 1970s and led to a general consensus that techniques involving water were not fiscally viable.
This is possibly due to less-than-normal static pressure gradients and the propensity of water to adsorb on various clay-rich surfaces.
With the introduction of the nitrogen foam fracturing processes, it quickly became apparent that treatments with gaseous fluids were statistically out-performing near-wellbore treatments involving explosives and/or acid.
Vertical completions involving straight nitrogen (with no water or proppant) became popular, and slowly began to replace foam as a fluid of choice. Clearly, there are localized portions of NAS properties where 100 percent N2 treatments result in superior acceleration of reserve recovery, but there is not enough statistical data to indicate whether this type of stimulation fluid outperforms foam fracture treatments across wide swaths of the play.
The gradual acceptance of horizontal drilling and multistage completions led to the realization that these improvements generally resulted in higher reserve recovery efficiency per D&C dollar compared to vertical drill wells.
Though a number of operators still work with vertical completions, the preponderance of clearly commercial completions have been associated with horizontal wellbores within a wide range of gaseous commodity prices.
Verifiable success with slickwater in the Ft. Worth Basin Barnett and other horizontal plays was followed by renewed speculation and a short resurgence of experimentation with large volumes of slickwater in the New Albany Shale.
To date, no known successes involving slickwater completions in this formation have been reported in the literature or in public data.
Stimulation Fluid Viscosity
The sporadic success of 100 percent N2 treatments [as compared to foams and other completion methods] in the play has led to an industry consensus that stimulation fluid viscosity and its benign impact on the clays in the shale may be “drivers” of well performance. Nitrogen has approximately 80 to 100 times less viscosity than typical slickwater formulations (possibly promoting even more far-field fracture complexity), and it is essentially inert in terms of its physical and chemical interaction with typical New Albany shale mineralogy.
Though completions involving 100 percent nitrogen-based fluids statistically outperform near-wellbore stimulation and stimulation with large volumes of liquid, there is no evidence to demonstrate that they out-perform foam-based systems uniformly across the play or that they regularly result in commercial completions comparable to other domestic plays for a given commodity and D&C cost structure.
A major concern for industry participants with respect to 100 percent N2 treatments was that the lack of proppants could theoretically lead to rapid and premature closure of some large percentage of induced fractures, therefore limiting the total exposed induced fracture surface area. A study was undertaken by RPSEA and its subcontractors, under a contract with the U.S. Department of Energy’s National Energy Technology Laboratory, that examined methods to address these concerns. Microseismic fracture mapping was performed during the hydraulic fracture stimulation of one horizontal New Albany Shale gas well in Christian County, Ky., and two parallel horizontal completions in McLean County, Ky. These two mapping efforts revealed a number of points:
- Interaction of hydraulically induced fractures with naturally occurring pre-stressed geological features (such as faulting) may be more common than was previously assumed and may decrease stimulated reservoir volume (SRV) substantially. This results in less total induced fracture area that is exposed to potentially productive reservoir. This same interaction can result in significant out-of-zone penetration, exposing bounding layers to communication with the wellbore and the subsequent production of undesirable fluids and/or movement of undesirable fluids into the potentially productive reservoir.
- Induced fractures or fracture network azimuths were consistent with area expectations: N85°E in Christian County, Ky., and N60°E in McLean County, Ky.
- Though some degree of far-field fracture network complexity was implied on both projects, it is possible that the completion method (open hole packers and ports with a “low” port count) may have severely limited total fracture initiation point count. A further implication is that when fracture initiation point count is low, it is possible that total surface area exposed to the potential reservoir is not as high as it could be if additional fractures were artificially induced along the wellbore.
Christian County Project
In the Christian County, Ky., project, a horizontal wellbore was drilled transverse (perpendicular to the expected fracture azimuth) and completed as an uncemented lateral with a series of open hole packers and sliding sleeves. The well was stimulated for gas production with eight nitrogen hydraulic fracture treatment stages and a minimal quantity of nitrified HCl in front of each stage (no proppant).
The receiver array was first located in a temporarily abandoned adjacent vertical well. As the stages progressed to the south in the heelward direction, the locations of sliding sleeves and events exceeded the listening distance of the geophones.
After the fourth stage’s completion, the toolstring was rigged out and placed into the intermediate casing of a vertical (future horizontal) offset located on the same surface pad as the treatment well.
It was known from core processing activities that non-negligible quantities of calcium carbonate lined a large percentage of both open and closed natural fracture faces. The process was recommended with the thought that some limited etching would be better than no etching at all, and the damage caused by leaving small quantities of high-chloride saltwater (spent acid) would be overwhelmed by the stimulation resulting from the etching. Induced fractures that remained in-zone resulted in microseismic activity approximately 1,200 feet from the lateral with volumes of approximately 1.0 million standard cubic feet (MMSCF) N2 per stage.
Figure 1 is a side view of the microseismic response from the treatments. The toe-most stages appeared to be relatively confined by the Ft. Payne lime. The heel-most stages demonstrated interaction with pre-stressed geological “feature(s),” and may have been responsible for the subsequent (uneconomic) production of formation water. During the last stage, enough events were recorded to clearly define upward growth through the Ft. Payne, the Warsaw and into the Salem.
Figure 1. Christian County, Ky., fracture mapping of a single wellbore-side view
In McLean County Ky., two parallel wells were drilled transverse to the expected fracture azimuth and completed with open hole packers isolating individual portions of each wellbore. The pair of wells was stimulated with 18 hydraulic fracture treatment stages in a “simulfrac” process. During this process, nine adjacent stages were sequentially pumped in each wellbore with two separate fracturing fleets at approximately the same time.
A dedicated vertical observation well was drilled between the two laterals, and the geophone array was placed in such a manner as to straddle the elevation of the expected fractures. The fracturing fluid was a 92 – 99 downhole slurry quality (DHSQ) nitrogen/water mist. A 30/80-mesh, ultra-lightweight proppant (a thermoplastic bead with a specific gravity of ±1.05) was slurried in a KCl water solution and batch-pumped at several concentrations ranging from 0.05 ppa to 0.25 ppa. The process was employed with the assumption that damage caused by the small volumes of water would be overwhelmed by the extra stimulation that would result by placing some limited quantities of proppant into the induced fracture system network.
Induced fractures that remained in-zone resulted in microseismic activity approximately 800 feet from the lateral, with volumes of approximately 2.6 MMSCF N2 per stage. During most stages, propagation away from both wellbores was extremely rapid, and the full extent of fracture length was observed only a few minutes after individual stage initiation.
Figure 2 is a map view showing the typical induced fracture azimuths (dashed lines), along with substantial interaction with pre-stressed geological “features” oriented N37°E between and on either side of the two stimulated wellbores. Figure 3 is a typical treatment record for one of the stages.
Though the approaches above (with respect to fluid choice) may have had some technical merit, poor production results from both projects were such that any improvements that might have been attributable to the process changes were not verified. The public release of the microseismic data is valuable, however, in that it provides some off-the-cuff estimates of fracture geometry and azimuth for typical treatments.
Figure 2. McLean County, Ky., mapping of a simulfrac on two wellbores-map view.
Figure 3. McLean County, Ky., treatment record for one of the stages on one of the stimulated wellbores during the simulfrac completion.
Optimizing stimulation processes in the New Albany Shale will require that operators and service providers cooperate and share information with respect to the application of specific technologies. Clearly, these technologies will continue to involve pumping high volumes of nitrogen at high rates for the short-term future. Incremental advances in the safe storage, handling and pumping of nitrogen will dominate the industry’s attention.
Further geological, engineering and operational information on these projects can be found in the report http://www.rpsea.org/attachments/contentmanagers/2729/07122-16-FR-New_Al.... The funding was provided by RPSEA (www.rpsea.org) through the “Ultra-Deepwater and Unconventional Natural Gas and Other Petroleum Resources” program authorized by the U.S. Energy Policy Act of 2005.
RPSEA is a nonprofit with a mission to provide a stewardship role to ensure the focused research, development and deployment of safe, environmentally-responsible technology that can effectively deliver hydrocarbons from domestic resources to the citizens of the U.S. RPSEA, a consortium of premier U.S. energy research universities, industry and independent research organizations, manages its programs under a contract with the U.S. Department of Energy’s National Energy Technology Laboratory.
Upstream Pumping Systems, Winter 2012