Engineering Design Issues

The most critical issue is to determine the source of the water and the production mechanism. Numerous technologies are available for water shut-off, but the nature of the water production must be known in order to design an effective treatment. The most common cause of water shut-off treatment failure is the misdiagnosis of the problem. Elphick and Seright (1997) give details on classifying problem water types. Keng Sen Chan (1995) also published some excellent guides to using plots to help diagnose water problems.


Water Production Mechanism

The Tensleep reservoir at Teapot Dome has produced approximately 1.5 million barrels of oil, or 30% of the estimated 5 million STBO originally-in-place. The Tensleep reservoir is naturally fractured, multi-layered sandstone with interbedded dolomites. The matrix rock has oil saturations ranging from 35% to 61%, and is thought to be oil-wet. Water channels through the natural fracture system from the underlying Madison aquifer. It prevents adequate draw down pressure from being exerted on the oil-bearing matrix.

The Tensleep producers have similar behavioral characteristics. They are pumped with high-volume electrical submersible pumps and have very low oil cuts, ranging from 0.05% to 0.8%. Recoverable reserves are best predicted by plotting the log of oil cut versus the cumulative oil production. A typical plot is shown in Figure 1. All of the wells in the reservoir show similar straight-line performance on this plot, indicating that the water production is a reservoir-wide phenomenon, not an individual well problem.

Production tests have shown that the oil cut improves at higher withdrawal rates as shown in Figure 2. This demonstrates that a greater draw down pressure is required to access the oil production than is required to produce water. This indicates that the oil and water production have different sources. At a very high draw down pressure, withdrawal rates reach a limit and oil cut increases exponentially, demonstrating that water production is limited by the fracture conductivity.

The plots and some production experiments have revealed that the problem is a reservoir one, that the water is sourced in the fracture system, and that the oil must be in the rock matrix. This situation can be treated by plugging up the fracture system.


Treatment System Requirements

As a result of understanding the water production mechanism, the water shut-off treatment strategy is to place a rigid, high molecular weight, cross-linked polymer gel into the fracture system and reduce fracture conductivity significantly. This must be done without allowing the polymer to infiltrate the rock matrix. The point at which water production becomes limited moves to a much lower draw down pressure and enables the maximum draw down to be placed on oil-bearing matrix porosity. The gel must withstand the high pressure differential and remain in place over the life of the producing well. Polymer infiltration into the rock matrix is controlled by molecular weight. Therefore, cross-linking should occur prior to perforation entry. The base gel should be of a very high molecular weight to lower injection pressure and demand for polymer and chromium cross linker (Seright, 2000) It should be of a robust chemistry to withstand free radicals and temperatures up to 220° F.


Treatment System Selection

There are several treatment systems available to control water production and these have been described by Lane (1997). Low molecular weight polymers can penetrate the matrix and must be selectively placed. High molecular weight polymers do not penetrate the matrix and are generally used for plugging fractures. The production mechanism in the Tensleep dictates the use of a high molecular weight cross-linked polymer. The high molecular weight polymer is critical to achieving high gel pump rates and placing the gel with a minimum time for it to dehydrate.

The most commonly used system for plugging fractures in the Rockies is Marathon’s MARCIT CT™. Phillip’s Petroleum’s subsidiary, Drilling Specialties, also has a similar technology. These gel systems are low concentration polyacrylamide solutions that are cross-linked with chromium III carboxylate complexes (Sydansk and Southwell, 1998). They have been used at temperatures exceeding 200°F. An alternative system developed by Unocal called Unogelä utilizes an amide functionality on the polymer to attach to one of a variety of organic cross-linking materials. Some of the organic cross-linked systems are stable to temperatures exceeding 350°F (Dovan and Hutchens, 1996).


Treatment Volume

The treatment volume design procedure has not been well established in industry practice. Treatment volumes should account for channel volume, geometry, and distribution. These descriptions cannot be obtained with present technology. The history of treatment volumes and degrees of success in a field is usually considered the best way to size a treatment. Unfortunately, this does not help with the volume design for the first treatment in a field. Under-designed treatment volumes may fail due to water reestablishing a flow path around the treatment. Over designed treatment volumes may place gel into offset wellbores or risk economic failure due to high cost.

It is intuitive that the volume of the treatment should be related to the height, width, and length of the fracture system. It should also be related to the fracture spacing and distribution. It is difficult to determine the volume of the fracture system, especially in a producing well connected to a charged aquifer. The volume of the treatment should be large enough to approach offset wells, but still be economic to place. There are some wells that have been logged with both sonic and density tools, which may give an estimate of fracture porosity. Pressure buildup tests performed in Tensleep wells may contain a storativity ratio in dual porosity behavior.

A guideline offered in the literature by Portwood (1998) is to pump the volume of total fluid that a well is capable of producing in a pumped off condition in 24 hours. Other guidelines are 25 to 100 bbls per foot of perforations. However, these guidelines have little technical foundation and are based on experience.

Kinzele (1998) recommends basing the treatment volume on experience in the field. His treatments ranged from 12 to 76 bbls per foot of perforated interval. According to a phone conversation with Marathon Engineer Falinda Hall, a typical volume in the Spring Creek Field is now near 1,200 bbls. Since there is no previous experience documented in the Powder River Basin, the experience documented by Marathon in the Big Horn Basin is the best analogy.

Once the treatment is being pumped, a real time Hall plot should be useful to estimate the fracture system’s ability to take the treatment, and adjustments can be made during the pumping operation.


Polymer Concentration

Polymer concentration is an important variable in the treatment design. Kinzele (1998) reports on treatment results from three types of polymer concentration schedules in a Madison field in Wyoming’s Big Horn Basin. It was concluded that polymer concentration has a significant effect on gel placement. In early high concentration treatments, near-total shut-off was achieved. While this may initially seem to have met the objectives, oil production was no longer economic. Acid treatments were required to restore productivity and economic production. Subsequent treatments used lower volumes and polymer concentrations and had better economic success. The reason was that lower polymer concentrations penetrated deeper into the larger fractures with lower pressure gradients and were not forced into the smaller micro fracture systems that contributed mainly to oil flow. The optimized treatments were done with the first 90% of the volume using 5,000 ppm polymer and the last 10% using 8,000 ppm treatments.


Stimulation

In the process of optimizing treatment designs in a Big Horn Basin field, Kintzele (1998) found that the Extreme Over Balanced Perforating Technique was helpful in establishing good communication with the fracture systems near the wellbore, and minimizing the pressure drop during treatment. This technique enabled a deep treatment without damaging the micro fractures near the wellbore that are critical to oil flow after the treatment. Whisonant and Hall (1997) used gel treatments in the Oregon Basin Field in Wyoming and found that combining this technology with propellant stimulation and acid fracs optimized treatment economics.


Quality Control and Data Collection Program

Gould and Pender (1996) and Lane (1997) offer pointers on quality control for gelled polymer treatments. It is important for the operator to be actively involved in the quality control to ensure a good job. The essential elements of quality control include:

  • Verification that the field materials can make a gel with the desired properties for the job to be successful.
  • Verification that the equipment on site is mixing and delivering the expected product.
  • Elimination or minimization of sources of contamination or interfering chemical reactions.

Prior to the job, the job design should be formulated in the laboratory to verify the desired results. Gel time and gel quality should be tested at field conditions. The viscosity of the gelant must be measured to predict friction pressures. The actual mix water and chemicals to be used on the job must be tested at field conditions prior to pumping the job. A clean and consistent source of water must be verified.

The following items should be monitored continuously during the job and recorded as a function of time:

  1. Injection rate
  2. Cumulative Injection Volume
  3. Injection pressure
  4. Polymer Composition
  5. Polymer Concentration
  6. Gelant Viscosity
  7. Cross linker Composition
  8. Cross linker Concentration
  9. Feed water Temperature
  10. Sample gelant at wellhead using two-valve sampler to prevent shearing.