T E C H N O L O G Y

T O D A Y

S E R I E S

FRACTURE-STIMULATION TECHNOLOGY FOR GAS-STORAGE WELLS

S.R. Reeves, SPE, Advanced Resources Intl. Inc.
INTRODUCTION AND BACKGROUND

Since introduction of Federal Energy Regulatory Commission (FERC) Order 636,1 the fundamental role of the interstate gastransportation business has changed from that of intermediary between producer and consumer to that of a service provider. Further, because these services are now unbundled and cannot be lumped into the supplier rate base, pipeline companies must now provide each of their services individually and at increasingly competitive prices. Some of these services include gas transportation, market-center hub services (e.g., balancing, compression, loaning, parking, peaking, trading), and gas storage. Gas storage is used for two primary purposes: to meet seasonal winter demands for natural gas (base-load storage) and to meet short-term peaks in demand (peaking storage), which can range from a few hours to a few days. For purely peak demand, cavern storage has become increasingly popular because of the extremely high deliverability that can be achieved, albeit only for short periods of time. Base-load storage facilities, for which depleted oil/gas fields are frequently used, can store larger volumes of gas but usually have smaller overall deliverability capacities. Nevertheless, minimum deliverability requirements exist for base-load storage facilities. One of the keys to providing gas-storage services on a commercially competitive basis is to minimize the cost of constructing and/or maintaining minimum deliverability requirements. In a study performed in 1993, the Gas Research Inst. (GRI) determined that, on average, the storage industry loses more than 5% of its deliverability capacity annually from base-load fields.2 The reasons behind deliverability loss are currently the subject of ongoing R&D by GRI and the U.S. Dept. of Energy (DOE) even though they traditionally have been attributed to the introduction of foreign materials during gas-injection periods (e.g., compressor oils, rust, scale) and to the deposition of scales and/or fines mobilization during periods of high-rate withdrawal, both of which can contribute to near-well damage. To counteract deliverability decline, the industry spends more than U.S. $100 million annually, primarily to drill new infill wells to supplement older wells that have experienced substantial deliverability decline. While new wells can be used to increase overall storage capacity of a field if sited appropriately, the stimulation (or restimulation) of existing wells is a considerably more cost-effective approach to maintaining deliverability. Gas-storage operators realize this. However, the traditional approaches to deliverability enhancement (blowing, washing, or mechanically cleaning the wellbore; reperforating; and acidizing, which combined account for over 80% of all storage-well enhancement activity2) generally provide only limited, short-term results,

not substantial, long-term well stimulation. Additionally, poor candidate-well selection, frequently based only on limited or inaccurate deliverability data, may also contribute to this outcome.The GRI study2 identified fracturing as a promising deliverabilityenhancement technique for gas-storage applications; however, storage operators have historically avoided such methods because, in part, of concerns about reservoir seal integrity. Nevertheless, as a result of competitive pressures, fracturing of gas-storage wells is of increasing interest to storage operators. In response, the DOE (and more recently GRI) is sponsoring a joint program with industry to demonstrate application of new and novel fracture-stimulation technologies in gas-storage fields across the U.S. The ultimate objective of this program is to advance deliverability-enhancement technology to provide more efficient utilization of the countrys gas-storage assets in response to the growing demand for a widespread, reliable natural gas supply. This article describes some of these fracture-stimulation technologies and advances made as part of that project.
SPECIAL CONSIDERATIONS FOR FRACTURING GA S-STORAGE WELLS

Copyright 1998 Society of Petroleum Engineers This paper is SPE 39417. Technology Today Series articles provide useful summary information on both classic and emerging concepts in petroleum engineering. P u r p o s e : To provide the general reader with a basic understanding of a significant concept, technique, or development within a specific area of technology.

Gas-storage wells have a number of unique aspects that require attention when considering fracture stimulation. While, taken individually, none of these aspects is necessarily new to the oil and gas producing community, together they represent an interesting blend of special requirements and associated challenges. First and foremost is fracture-height growth. Frequently, only limited barriers exist between the storage formation and potential problem zones above and below it. Here, the concern is that the reservoir seal might be compromised by upward or downward fracture-height growth, resulting in gas loss. The potential to penetrate water-bearing zones also exists; significant amounts of water influx can reduce or eliminate the potential of a well to contribute to field deliverability and can increase well-operating costs. Hence, application of fracturing to gas-storage wells should be limited to those approaches, geologic settings, or reservoir conditions where fracture-height growth can be controlled or where the practical extent of height growth is not of concern. In most cases, this means limiting the size of the treatment and performing treatments when the reservoirs are depleted in the summer months (i.e., when a greater stress contrast exists between the storage horizon and the bounding layers to help contain height growth). Another important consideration for fracturing gas-storage wells is the final fracture properties needed for effective stimulation. Because of high deliverability, one of the principal requirements for gas-storage fields, storage-reservoir permeabilities are frequently quite high compared with those of the normal domain of (hydraulic) fracturing. A 1993 survey2 of storage operators suggested that storage-field permeabilities ranged from 20 to >800 md. The benefit in massively fracturing formations with permeabilities this high is debatable; instead, use of short, high61

FEBRUARY 1998

TABLE 1TSO-FRACTURING RESULTS, HUNTSMAN FIELD Well HS-23 Permeability, md Prefracture Total skin AOF (1,182 psia), MMcf/D Post-fracture Total skin AOF (1,182 psia), MMcf/D Deliverability improvement AOF, MMcf/D Percent Fig. 1Sites of new and novel fracture-stimulation tests.
*Estimated. AOF=absolute open flow.

HS-45 695 +40 147 To be tested 221* 74 50

46 +3.6 35 +0.1 51 16 46

conductivity fractures to bypass near-well formation damage is more appropriate. This is not unlike the stimulation requirements in the Gulf of Mexico, for example; therefore, techniques applicable to that environment should also be suitable for many gasstorage fields. A final consideration for storage reservoirs is their susceptibility to formation damage from stimulation (or other) fluids. The process of cycling dry, pipeline-quality gas into and out of storage wells can reduce the near-well area to a very low residual water saturation. Reintroduction of fluids into the system can in and of itself be damaging because a fluid saturation is re-established, reducing the relative permeability to gas. This situation can be exacerbated if the injected fluids contain potentially damaging chemicals and/or particulate matter; these might mix and react with already-in-place compressor oils, rust, scale, fines, and other such materials, significantly compromising any stimulation benefit of a fracture treatment. Frequently, storage-well deliverabilities actually decline following a traditional hydraulic-fracture treatment and can take several years or more to clean up to the point of meeting or exceeding prefracturing deliverability levels, not to mention expectations from an undamaged hydraulic fracture. This all indicates that gas-storage reservoirs can be extremely sensitive to aqueous stimulation fluids; therefore, fracturing treatments must be designed to account for and minimize the impact of this sensitivity.
RECENT APPLIC ATIONS OF FRACTURING FOR GA S-STORAGE-WELL STIMUL ATION

density slurry. Because damage bypass is the objective for most gasstorage fields, TSO fracturing is an obvious technology to consider. It also has two other potential advantages. First, it reduces nearwell pressure drops during peak withdrawal periods, which may mitigate mobilization of fines and deposition of scale (some of the primary causes for deliverability decline in gas-storage wells) as well as non-Darcy flow effects. Second, small treatments are generally the norm, and these minimize the risk of unacceptable fracture-height growth. The Huntsman gas-storage field in Nebraska (Fig. 1) is an example of use of this technology.3 The field exhibited the previously cited characteristics of concern to gas-storage operators, most notably a limited barrier to upward fracture-height growth between the storage reservoir and a potential gas thief zone and a downdip water leg in close proximity. Through careful treatment design and fracture diagnostics, two treatments were performed successfully without breaching the critical fracture-height constraints. Table 1 summarizes the deliverability-enhancement results that immediately followed the treatments; these results indicate that TSO fracturing can provide immediate and significant deliverability benefits (approximately 50% in this case). Further cleanup of the fracturing fluids is expected to improve these short-term results. Hence, TSO fracturing appears to represent a viable, exciting new technique to enhance deliverability of gas-storage reservoirs, even in reservoirs where fracture-height growth is of concern. Hydraulic Fracturing With Liquid CO2 and Proppant. As mentioned earlier, aqueous-based hydraulic-fracturing fluids can have a damaging effect on gas-storage wells, to the degree of more than offsetting the stimulation induced by the created fracture in some cases. As an alternative to aqueous-based stimulation fluids, fracturing with pure liquid CO2 and proppant is a legitimate consideration in gas-storage wells. This technology, first introduced in the early 1980s in Canada, has since been used to stimulate hundreds of gas wells there successfully. Its popularity has not yet extended into the U.S. because, in part, of the limited availability of specialized blending equipment required to perform liquid-CO2 treatments, In recent years, however, the U.S. DOE has been instrumental in making the equipment available and demonstrating successful application of the technique to water-sensitive gas formations in the lower 48 U.S. states. The pumping of ungelled, pure liquid CO2 as a fracturing fluid involves the use of self-contained pressurized blending equipment, which limits the proppant volumes that can be pumped from a single blender. Currently available blenders (in the U.S.)
FEBRUARY 1998

With the special requirements of gas-storage reservoirs in mind, several fracturing approaches, some newer and more novel than others, have been tested in various fields in the U.S. (Fig. 1). The most promising methods include tip-screenout fracturing (TSO), hydraulic fracturing with liquid CO2 and proppant, and extremeoverbalance (EOB) technology. Use of these methods and some advances in the technologies are described next. TSO Fracturing. TSO fracturing, originally conceived as a specific design objective in the late 1970s and early 1980s, has now become commonplace in such areas as the Gulf of Mexico, Alaska, and the North Sea. With the primary advantage of being able to create fracture conductivities that exceed those possible by manipulation of traditional treatment parameters (such as proppant type, fluid viscosity, and injection rate), TSO fracturing is ideally suited to stimulation of higher-permeability formations where the primary objective is bypassing near-well damage. The high fracture conductivities are achieved by effecting a screenout at the fracture tip to arrest further extensional growth, widening the fracture with continued pumping, and finally packing it with proppant in a high62

Fig. 2CO2 content of flowback gas in Well 4886.

are capable of holding up to 48,000 lbm of proppant, which is sufficient for most gas-storage applications. Also, the low viscosity of the fluid means that injection rate is the primary control over fluid leakoff, fracture width, and proppant transport. Rates as high as 70 bbl/min can be achieved, however, and have been shown to be sufficient for most applications. Primarily because of the cost of the CO2, treatment costs can be higher than those for more conventional fracture-stimulation treatments; nevertheless, the positive aspects of nonaqueous fracturing fluids for gas-storage applications are substantial. A three-well test of this technology at the Galbraith field in Pennsylvania (Fig. 1) demonstrated the potential of liquid-CO2 fracturing for gas-storage applications.3 The test showed that the treatment did not extensively contaminate the gas with CO2; the CO2 content was reduced to less than 5% within 50 hours of controlled flowback and to zero after 59 hours (Fig. 2). Table 2 shows the deliverability results of these treatments compared with those of conventional gelled-water treatments in the field. The CO2 treatments provided substantially superior short-term deliverability enhancement, enough so that the unit cost of added deliverability was similar. Considering the fact that these particular liquid-CO2 treatments screened out early (for operational reasons) and therefore placed only a small amount of proppant compared with the gelled-water treatments (<2,000 lbm each vs. >20,000 lbm each, respectively), this technology appears highly promising for gasstorage applications. EOB Technology. EOB technology for well stimulation can be traced back to the DOEs multiwell experiment site in the early 1980s where it was tested as a nonaqueous, low-damage completion method for the tight, naturally fractured Mesa Verde formation. Originally used with perforating operations, the technique involves overpressuring the wellbore with a gas (and frequently a liquid at the bottom of the well) to a level that exceeds the formation fracture

Fig. 3Fracture-initiation profiles; sh, min = minimum horizontal stress.

gradient by a factor of 1.5 to 2.0 and then quickly exposing that pressure to the formation as the perforator is fired. If pressure levels and rates are sufficient, multiple fractures extending radially outward may be created (Fig. 3). Because of the elevated pressures used, initial fracture orientation is governed primarily by wellbore and perforation orientation, not by in-situ stresses. This attribute can be useful when attempting to stimulate horizontal wells or in vertical wells where hydraulic fractures tend to grow horizontally. The radial nature of the created fracture pattern, as well as the limited treatment volumes involved, makes the technique particularly well suited for overcoming near-well damage. More recent developments in the technology include stimulation of already perforated completions by use of a rupture disk on the end of a tubing string, introduction of small proppant volumes into the process, and use of propellants to boost the downhole pressures created. Several reasons may make EOB technology applicable in gasstorage wells. First, the treatment volumes are limited, which limits the potential for undesirable fracture-height growth. Second, if created, the multiple-radial-fracture pattern may provide near-welldamage bypass characteristics superior to those of a normal twowing hydraulic fracture. Third, small fluid volumes reduce the possibility of fluid-induced damage. Finally, in locations of high in-situ stresses where horizontal fracturing is common (such as in many gas-storage fields in the Appalachian basin), the complete vertical extent of the storage horizon can be stimulated.

TABLE 2COMPARISON OF GELLED-WATER vs. LIQUID-CO2 FRACTURE RESULTS GALBRAITH FIELD Gelled Water (Three-Well Average) Immediate Prefracture AOF, MMscf/D Post-fracture AOF, MMscf/D AOF improvement, MMscf/D Folds-of-increase Treatment cost, U.S. $ Cost of added AOF, U.S. $/Mcf-D
AOF=absolute open flow.

Liquid CO2 (Three-Well Average) Immediate 2.93 7.40 4.47 2.5 47,300 10.58 Annual 2.93 8.50 5.57 2.9 47,300 8.49

Annual 3.00 5.00 2.00 1.7 16,200 8.10

3.00 4.87 1.87 1.6 16,200 8.66

64

FEBRUARY 1998

Fig. 4Vertical coverage of hydraulic-fracture and EOB treatments.

EOB technology has been tested in one gas-storage field, the Donegal field in Pennsylvania (Fig. 1).4 The deliverability response to previous (hydraulic) fracture-stimulation attempts at the field was disappointing; from rock-property, in-situ-stress, and fracturegeometry studies, this was attributed to the creation of horizontal fractures. EOB treatments were implemented to effect full vertical stimulation of the storage horizon. Fig. 4 illustrates the results of radioactive-tracer surveys for a hydraulically fractured well and one stimulated with the EOB technique. These results suggest that the EOB approach appears to stimulate the entire storage horizon. However, further work is needed to optimize the overall stimulation achieved with the process. To address this, advances have been made recently in numerical modeling of the EOB process under DOE R&D sponsorship.
FUTURE TECHNOLOGY NEEDS

which the damage cleans up, and over how long a period of time. Further, the properties of liquid-CO2 under fracturing conditions are not well-defined; therefore, the engineering tools to design and estimate the ultimate fracture geometry are essentially nonexistent at this time. Similar shortcomings in treatment design and analysis exist for EOB methods. However, with the assistance of DOE under this and other R&D programs, this technology need is now being addressed. A preliminary dynamic fracturing model has

Many technology needs still exist with respect to effective fracture stimulation of gas-storage wells. For example, the ability to compare the net long-term stimulation effect of a TSO-fracture treatment (with exceptional length/conductivity characteristics that are somewhat offset by fluid damage) with that of a nondamaging liquid-CO2 treatment (with presumably less favorable length/conductivity dimensions but without any offsetting damage) is required. The technical hurdles preventing such a comparison include the ability to quantify the damaging effects of aqueousbased fracturing fluids in gas-storage reservoirs, the degree to
66

Fig. 5Sample output for EOB modeling: Pennsylvania Well 5; 4, 1.75-in. CT vs. 4C, 2.375-in. tubing.

FEBRUARY 1998

been developed to predict the outcomes of EOB treatments in terms of pressures, number of fractures created, fracture length and width over time during the process, and other such parameters. Fig. 5, which compares the impact of different equipment arrangements on treatment results, illustrates how the model can be used. While further work is required to calibrate and validate the model, ultimately this should prove to be an exciting new development in EOB technology and should help improve the effectiveness of EOB treatments throughout the oil and gas industry.
ACKNOWLEDGMENTS

GENERAL REFERENCES
Application of New and Novel Fracture Stimulation Technologies to Enhance the Deliverability of Gas Storage Wells, topical report Contract No. DE-AC21-94MC31112, U.S. DOE, Washington, DC (April 1995). Natural Gas Storage: Historical Development and Expected Evolution, final report, Contract No. GRI-95/0214, GRI, Chicago, Illinois (June 1995).

SI METRIC CONVERSION FACTORS

I thank the U.S. DOE for funding this work under Contract No. DE-AC21-94MC31112 under the management of Jim Ammer and David Hill of GRI for organizing the preparation and publication of this paper.
REFERENCES
1. Order 636, final rule, Docket RM91-11-000, FERC, Washington, DC (8 April 1992). 2. State-of-Technology Assessment and Evaluation of Gas Storage Well Productivity Enhancement Techniques, topical report, Contract No. GRI93/0001, GRI, Chicago, Illinois (December 1993). 3. Reeves, S.R. et al.: Liquid-CO2 and Tip-Screenout Fracturing as Techniques for Restimulating Gas-Storage Wells, paper SPE 37343 presented at the 1996 SPE Eastern Regional Meeting, Columbus, Ohio, 2325 October. 4. Reeves, S.R., Pekot, L.J., and Koperna, G.J.: Fracture Stimulation Field Demonstration Projects, paper 9.1 presented at the 1997 U.S. DOE Natural Gas Conference, Houston, 2427 March.

bbl ft ft3 in. lbm md psi

1.589 873 3.048* 2.831 685 2.54* 4.535 924 9.869 233 6.894 757

E01 =m3 E01 =m E02 =m3 E+00 =cm E01 =kg E04 =m2 E+00 =kPa

*Conversion factor is exact.

Scott R. Reeves is Vice President of Advanced Resources Intl. Inc. in Arlington, Virginia, where he designs and manages R&D projects for the U.S. DOE, GRI, and others in the areas of production enhancement and well-stimulation technology. He also is an international consultant to the United Nations and private industry. Reeves holds a BS degree in petroleum engineering from Texas A&M U. and an MBA degree from Duke U.

FEBRUARY 1998

67