John W. Ely
Stephen A. Holditch
S.A. Holditch & Associates Inc.
College Station, Tex.
Varying formation properties in coal seams creates the need for variable completion practices for optimum methane production.
One primary concept in successful coal seam fracturing requires interconnecting the cleat system to the well bore. This interconnection is similar to interconnecting natural fractures to the well bore in conventional reservoirs.
The authors believe that interconnecting the cleat system to the well bore is all that is required of a fracture treatment in moderate to high permeability coals (k 20 md).
In low permeability coal reservoirs (k
COAL SEAM FRACTURING
Fracture treatment technology in coal seams is just now starting to evolve. Because permeable coal seams are highly cleated, complex hydraulic fracture systems are commonly created. These complex fractures usually result in very high pressures during fracture treatments in thick coal sections.
When multiple, thin seams in close proximity to one another are fracture treated, a single, planar vertical fracture is usually created. When this type of fracture occurs, the fracture treatment pressures will normally be less than 1.0 psi/ft.
The majority of completions and fracture treatments in coal seams are currently conducted in the San Juan and Black Warrior basins. Some wells have been drilled in the Piceance, Raton, and the Northern Appalachian basins. Also, some activity is evident in several European countries, including France, Poland, and Germany.
One of the major differences between fracturing in coal seams and sandstone is the stiffness of the two types of formations. Typically, Young's modulus in coal will range between 100 and 500 thousand psi. Young's modulus for typical sandstones and limestones will range from 2 to 13 million psi. Because of the low modulus and the highly cleated nature of a permeable thick (h 20 ft) coal seam, we often create multiple, complex, and wide hydraulic fracture systems.
FRACTURING SCENARIOS
Over the past few years, while investigating the types of fracture treatments conducted in coal, we have categorized five separate stimulation scenarios. These scenarios can be described as follows:
- A shallow coal seam where a horizontal fracture will be created.
- A series of thin coal seams in a depth range where single, planar vertical fractures will be created.
- A single thick coal seam where the hydraulic fracture will be confined entirely in the coal and a complex fracture system (multivertical or T shaped fractures) created.
- An hydraulic fracture where the fracture is initially contained within a single coal seam, but during the latter portion of the treatment the fracture begins to propagate vertically into the bounding layers.
- A high-permeability (highly cleated-fractured) coal seam that does not require stimulation.
SCENARIO 1
Under the first scenario (shallow coal seams with horizontal fractures), we feel that individual horizontal fractures will have to be created using either limited entry methods or mechanical diversion. An example of horizontal fractures is shown in Fig. 1. When horizontal fractures are expected, we recommend using linear fracturing fluids with a moderate-size pad volume.
For a horizontal fracture, the bottomhole treating pressure will be in excess of 1 psi/ft, because the fracture must lift the overburden. Complex multiple fracture systems may be created if bottom hole pressures increase substantially during the treatment.
SCENARIO 2
The second scenario, where a single vertical fracture through a series of coal seams is created, has the following characteristics.
- The excess pressure in the fracture caused by the fluid moving down the fracture will be minimal, i.e., only 200-500 psi. This is far less than the excess pressure of 1,000-2,000 psi experienced when one observes fracture containment within thicker coals.
- Fracture propagation pressures will be less than 1 psi and, even more typically, between 0.6 and 0.8 psi/ft.
- When a single vertical fracture is created cutting through several coal seams, a conventional three dimensional fracture design model can be used to estimate fracture dimensions. A simplified drawing of this scenario is shown in Fig. 2.
SCENARIO 3
We have the following comments about the third scenario, where complex hydraulic fracture is contained entirely in a single thick coal seam:
- High injection rates must be used during the fracture treatment to combat high leakoff.
- In addition to high injection rates, we recommend using a high viscosity, shear stable fracturing fluid such, as a 35 lb/1,000 gal cross linked gel using borate. In some cases, bridging fluid loss additives are needed to offset high leakoff.
- Because of a low Young's modulus and high formation compressibility in the presence of complex fracture systems, the hydraulic fracture will seldom penetrate more than 200-300 ft from the well bore. Fig. 3 presents a drawing of a complex fracture system.
SCENARIO 4
The fourth scenario is similar to either Scenario 1 or 3, with the fracture contained within the coal seam. But at some point during the treatment, the fracture breaks out of the coal seam and vertical growth occurs.
To be prepared for this particular scenario, one must have extra fluid on location to reinitiate the pad if the fracture does grow out of zone. Also, if the pad is reinitiated, one should pump a large enough volume to develop a fracture wide enough to accept proppant for restarting the proppant.
This particular scenario is very unpredictable and involves a lot of prejob planning. Experienced personnel must be on location to be able successfully to adjust the pumping schedule during pumping operations.
SCENARIO 5
The fifth scenario is a high permeability, high-pressure coal seam that does not require hydraulic fracturing. The most important factor in a highly cleated, highly permeable coal seam is to minimize near well bore damage caused by mud filtrate and cement filtrate invasion.
One small portion of the Fruitland coal in the San Juan basin can be produced satisfactorily without hydraulic fracturing. However, well bore enlargement by jetting and well bore stabilization using slotted liners are needed to produce these high-permeability, high-pressure coals.
One major problem with producing high-permeability coal is the persistent production of coal fines. On occasion, these wells have to be cleaned out to remove coal from the well bore. Fig. 4 presents an example of the stable cavity technique used to complete high-permeability coals in the San Juan basin.
PREFRACTURE WELL TESTING
The single most common error in developing coal seam properties is not determining permeability estimates early in the project's life. Fairly inexpensive techniques are available to determine the ability of the reservoir to produce.
An easy way to measure coal seam permeability is to run an injection-falloff test, at low rates, using water. Additionally, slug tests can be used to estimate the permeability.
One must know the permeability to design the fracture treatment and to predict long term gas recovery.
A second critical item is to estimate the gas in place. One must obtain cores to measure accurately the gas content.
Some operators will set pipe and complete wells whenever a coal seam is encountered. Because coal seams desorb gas, and the gas flow rate tends to increase with time, the operators believe their wells will become economic someday as gas desorption continues.
However, if the coal seam permeability is very low or the gas content is very low, any specific well may never be economic to drill or produce. Like any other gas reservoir, coal seams must be evaluated and the cutoff values of permeability, thickness, and gas content determined.
FRACTURE DESIGN
The procedures for optimizing a fracture treatment in a coal seam reservoir are similar to procedures used in designing stimulation treatment for sandstone or carbonate reservoirs. Basically, there are two different kinds of parameters in fracture design. One group of parameters can be controlled by the engineer, while the second group can only be estimated.
The design parameters that can be controlled and measured or estimated are listed in Table 1.
To understand the fracture design process, we recommend that an engineer conduct a sensitivity study. For a coal seam, the engineer must determine the most likely fracture scenario and fracture orientation for a given coal seam or series of coal seams. The engineer must then estimate the most probable value of each design parameter and run a fracture treatment design model for the range of injection rates and volumes.
After these base runs are completed, the engineer should determine the formation properties most likely to be in error and establish a range for each parameter. Finally, the fracture design model can be run by changing one parameter at a time and observing how each affects the created and/or propped fracture dimensions.
FLUID SELECTION
Four basic fluids are commonly used in the coal: fresh water, linear gel, foam fluids, and crosslink gels. Fresh water was used several years ago to stimulate the thin seams in the Black Warrior basin. The advantages of using fresh water include low cost and minimal environmental implications.
The distinct disadvantages of water are its lack of friction reduction plus the inability of the fluid to create wide fractures and transport high proppant concentrations.
Linear-gel systems have been widely used for multiple seams where horizontal fractures are propagated. Linear gels can also be used to create vertical fractures in multiple seams.
The advantages of linear gels are adequate proppant transport, minimal cost, and well-defined rheology. Some disadvantages of using linear gel include lack of perfect proppant suspension; however, this can be overcome in some cases by using forced closure fracture techniques.
Although foamed fluids have been tried, foam is not widely used to stimulate coal seams at this time. The only place where energized fluids might be recommended is where fluid removal is costly, and the gas in the fluid helps to clean up the well. However, most coal seam wells have to be pumped, so the use of an energized fluid does not normally make sense.
The advantages of using foam include good proppant transport, minimizing the volume of water needed, and, of course, energy assist in flowback. If one chooses to use foam, field quality control is essential to be assured the treatment is pumped correctly.
The fourth system is a crosslinked polymer gel, normally a borate system. Many treatments are pumped using 30-35 lb/1,000 gal of polymer. A crosslinked gel has the advantages of excellent proppant transport and high viscosity yielding wide fractures. And the fluid can be used to transport high proppant concentrations.
Disadvantages of these gels include higher cost than linear gels, complicated rheology, and the requirement of intense quality control.
For ease of removal, one must use every possible measure to be assured that the gel does indeed break back to water. However, the fluid must also maintain the required minimum viscosity to transport proppant during the job.
PAD VOLUMES
For horizontal fractures, the pad volume should typically be in the range of 25%. For linear and crosslinked fluids where a single, planar vertical fracture intersects multiple coal seams, we have been successful using pad volumes of 25% or less. Where the fracture is contained within a single coal seam and complex fractures are created, pad volumes of 40-60% are needed.
PUMP RATE AND FLUID LOSS
High injection rates should be used when fluid loss is a problem. High leakoff will normally occur in thick, high permeability coals. In low permeability coals where a single planar fracture through multiple seams is created, we recommend using lower injection rates.
Most operators, when they encounter high leakoff, will compensate by using a higher rate rather than bridging type fluid-loss additives. We have found substantial benefit in using both injection rates and granular additives to combat high leakoff.
Hundred-mesh sand and 40/70-mesh sand can help to minimize leakoff and keep the fracture cleats open after a treatment. The data in Tables 2-6 have been included to illustrate typical injection schedules for coal seam fracture treatments.
FRACTURE GRADIENTS
We have seen variations in bottom hole treating pressures of as much as 0.5 psi/ft from the same zones in offset wells. We believe these variations are a result of multiple fracture systems and near well bore pressure drop and that minimizing the number of perforations and initiating the fracture in a competent zone optimizes the chance for creating a single, planar vertical fracture.
By creating a clean flow path from the well bore to the fracture, the injection pressure will be minimized.
PROPPANT CONCENTRATION
Whether stimulating a coal seam or a conventional reservoir, the authors recommend the use of high proppant concentrations. The cleat systems can be propped by smaller mesh sand (100 or 40/70 mesh). This sand can also restrict fluid entry during the treatment which should help to divert flow into a single, main fracture.
Numerous high sand-concentration treatments have had initial fracture gradients exceeding 1.2 psi/ft, but by the end of the treatment, normal gradients (0.6-0.8 psi/ft) are observed. Using high sand concentrations is beneficial because the purpose of a fracture treatment is to create a conductive flow path that connects the cleat system with the well bore.
High sand concentrations will lead to high slurry viscosities, which in turn will create wide fractures and improve proppant transport. The purpose of a fracture treatment is to pump proppant.
GEL CONCENTRATION
Many crosslinked fracturing treatments have been conducted with a 30 lb/1,000 gal gel system crosslinked using borate. Borate cross linkers are preferred because these gels are not shear degradable and are ideal for low-temperature (less than 170 F.) reservoirs.
These fracturing fluids will create excellent downhole viscosity at typical coal seam temperatures. The 30 lb/1,000 gal gel systems, however, may not work well because the polymer concentration may be too low to crosslink.
Because the polymer concentration required for a successful crosslink is so critical, we recommend the use of 35 lb/1,000 gal of gel (as a minimum) to be sure the fluid crosslinks and remains stable.
The authors recommend that enzyme breakers should never be used when the gel is crosslinked with borate. Catalyzed oxidizer systems available from the service companies should be used.
QUALITY CONTROL
Serious quality control problems have been encountered during some coal seam fracture treatments. These problems usually have the potential to reduce postfracture gas production. On some treatments, 60-70% had problems.
Low-temperature and low pressure environments create a situation where the stimulation fluids can become permanent plugging agents. We believe intense quality control is a necessity on every coal seam fracture treatment.
Intense quality control includes pilot testing of the fracturing fluids on location using a heat cup and the Fann 35 viscometer with a B 2 bob, as well as other procedures to measure the fluid and proppant quality.
Two common problems encountered are that the breaker system does not function as designed or the crosslinker is not working properly. Testing the gel at temperature assures that the correct viscosity is achieved at bottom hole temperature and the gel will break when desired.
We have often observed that the types and concentration of the fluid additives must be changed in the field to obtain the desired performance of the fracture fluid.
FIELD OBSERVATIONS
Supervision of coal seam fracture treatments in the San Juan, Piceance, Raton, San Wash, Black Warrior, and Northern Appalachian basins and Europe has taught us several important lessons.
One basic lesson is to do everything possible to keep the coal and surrounding formations competent.
If the coal rubbelizes near the well bore, high injection pressures will result and it will be more difficult to pump high sand concentrations.
One procedure for improving well bore competence is to not perforate within the coal. Perforating between the coal seams can stop or reduce coal fines production.
Another basic lesson is that the purpose of the fracture treatment is to interconnect the cleat system with the well bore. Following this premise, one does not necessarily have to utilize extremely large treatments or large mesh proppants.
In areas of low-permeability coal seams, there have been indications that the coal seam permeability decreases as the flowing bottom hole pressure is reduced. This fact causes real problems for the operator because the drawdown pressure (p - pwf) must be maximized for optimum gas production.
Some low-permeability coals have ceased to produce water or gas when pumped off, but will continue to produce gas if a water level is left above the coal seam. Therefore, some intermediate value of pressure drawdown could be best for certain coal seams.
In areas containing multiple coal seams, excessive height growth during a treatment can sometimes be a problem.
Field data indicate that vertical height growth from 200 to 600 ft is possible.
Data from mechanical properties logs and fracture height logs also indicate that few or no barriers to vertical fracture growth exist in many areas. We are conducting further tests to confirm this height growth and are initiating studies with less viscous fracture fluids, i.e., linear-gel systems and lower pump rates, to try to control height growth.
Some engineers have recommended using large-mesh proppants to create high fracture conductivity. The authors do not agree with that concept, but recommend using 100-mesh, 40/70-mesh, and 20/40-mesh proppants because these proppants:
- Seal the cleats and small fractures near the well bore during a treatment, thus helping to funnel most of the fluid down a single main fracture.
- Erode the corners and smooth the fracture path near the well bore, thus helping to reduce excess pressure.
- Are easily transported by linear gel or a 35 lb/1,000 gal crosslinked gel.
- Can prop open the cleats near the well bore during production (100-mesh and 40/70-mesh sand).
Typically, we design a treatment with all three proppant sizes. Tables 2-6 illustrate such designs.
In one field example, several wells in a five-spot pattern were stimulated using the 100-mesh stage while the remaining wells only used the 40/70 and 20/40 stages.
Production data indicate the near well bore permeability was improved in the wells where 100-mesh sand had been used.
Additionally, we believe that by using the smaller mesh proppants, the production of coal fines seems to have been reduced.
The 100-mesh and 40/70-mesh sand tends to keep the fines in place.
In fracture treatments where large (10/20 or 8/12) mesh proppants are used, coal fines are usually more of a problem.
The authors believe that it is more efficient to stop the coal fines production in the reservoir rather than allow the fines to enter the proppant pack.
Our observations show fewer coal fines are produced when small-mesh proppants are used.
RECOMMENDATIONS
Based upon several years of designing and supervising fracture treatments in coal seams, the authors offer the following recommendations.
- Careful pre and postfracture analysis should be conducted to determine the optimum fracture treatment design.
- Due to the low-temperature, low-pressure environment, quality control in the field is needed to be sure the optimum fluid is pumped during a treatment.
- Coal seam reservoirs vary in permeability, hydrocarbon content, and reservoir pressure similar to conventional reservoirs. Variable formation properties in conventional reservoirs create the need for variable completion practices. These same factors create the need for varying completion procedures in coal.
- The operator needs to be sure that the coal seam contains sufficient permeability and gas content to be a commercial producer after a fracture treatment.
- Most coal seams can be successfully stimulated using either a linear gel or a 35 lb/1,000 gal gel crosslinked with borate.
- The treatment should be designed using the correct pad volume, then the proppant should be pumped, starting first with 100 mesh. That should be followed by 40/70 mesh and, finally, 20/40 mesh.
- High injection rates are required to combat high leakoff in permeable, highly cleated coal seams.
Copyright 1990 Oil & Gas Journal. All Rights Reserved.