07122-33

Goal
The goal of this project is to develop new methods for creating extensive, conductive hydraulic fractures in low permeability gas reservoirs. The plan is to evaluate fractured well productivity achieved using conventional fracturing practices; to dynamically measure conductivity created with high rate proppant fracturing in the laboratory; and to develop design models to implement the optimal fracture treatments determined from the laboratory and field data

Performers
Texas Engineering Experiment Station, College Station, TX 77843
Crisman Research Institute, Texas A&M University, College Station, TX 77843
Carbo Ceramics, Inc., Houston, TX 77024
BJ Services, Houston, TX 77092
Halliburton, Duncan, OK 73536
Schlumberger, Sugar Land, TX 77478

Background
The key to producing gas from tight gas reservoirs is to create long, highly conductive hydraulic fractures to stimulate gas flow from the reservoir to the wellbore. To maintain conductivity in the fracture, it is also important to pump sufficient quantities of an effective propping agent into the fracture.

Current hydraulic fracturing methods in tight gas reservoirs have been developed largely through an ad-hoc application of low-cost water fracs, without much effort to optimize the hydraulic fracturing fluid and overall fracturing process. The widespread use of water fracs instead of gel fracs is in part due to the complex problem of gel damage – a reduction in permeability that occurs due to plugging of the proppant pack inside the fracture, filter cake deposition on the fracture walls, and fracture fluid invasion into the formation adjacent to the fracture walls.

This project aims to develop modern techniques to optimize design of the fracturing treatment and fluid type in low permeability formations. The first part of the project will focus on the analysis of production data from thousands of wells to evaluate the effect of the volume of proppant and type of fluid used on the long-term production behavior of tight gas wells. The second part of the project will involve laboratory studies of the mechanisms of proppant interaction with the fracture walls and creation of fracture conductivity, using dynamic fracture conductivity testing procedures. Results of the field and laboratory studies will be used to recommend new procedures for fracturing low permeability formations.

The project will be executed by a team of researchers from the Crisman Institute in the Department of Petroleum Engineering at Texas A&M, with cost sharing and technical support from four industry partners: Carbo Ceramics (a leading proppant producer), BJ Services (a primary provider of hydraulic fracture services), and Halliburton and Schlumberger (both leading service companies with extensive field experience in hydraulic fracturing projects).

Deliverables for this project will include: 1) a database of fractured well performance; 2) recommendations on methods to minimize gel damage; 3) guidelines for optimizing fracture conductivity under dynamic conditions; and 4) optimized tight gas fracturing treatment designs. All of these deliverables will be included in the final report.

Potential Impacts
If successful, this project could lead to improved hydraulic fracturing practices in tight gas basins throughout the United States. Advances in hydraulic fracturing and mitigating gel damage have the potential to add substantial unconventional gas reserves to the nation’s gas supply. Improved hydraulic fracturing practices resulting from this project could bring a significant portion of the nation’s 293 Tcf of technically recoverable tight gas to market. The cumulative benefit would be determined by the total number of fields where the technology is ultimately applied. An incremental increase in domestic gas production would result in increased tax revenues, royalties, and regional economic benefits.

Background Work and Accomplishments

Assessment of Field Treatment Results
Over the past three years an advisory system has been built for stimulation in tight gas reservoirs. The program was built in response to the lack of experience of many stimulation design engineers in these types of reservoirs. The goal of the project was to design a program that would suggest good completion and stimulation designs to help guide young and/or inexperienced engineers to better solutions. We do not assert that the recommendations made by the advisory program are the optimal solution, but they are a very good reference point. The program uses information from reviewed literature and interviews with experts to build its decision trees. The decision trees are then used to give the best suggestions possible based on information provided to the program by the user, such as reservoir properties.

The program allows the user to run a study on a single or multiple producible formations. Based on the reservoir properties input by the user, the program will suggest whether to drill a vertical or horizontal well. Next, the program will compute the best completion package for the well. The program assumes that the well will be hydraulically fractured, or otherwise would not be producible; therefore the default completion design recommended by the program is cased and perforated.

After the completion design phase, the user will be able to make inputs regarding the design of the stimulation treatment. The stimulation design section of the program is the most involved portion of the program as regards to user interaction, as well as sophistication of the program. The program includes suggestions on injection methods, fracture fluid design, fracture fluid additives, proppant types, proppant concentrations, and optimal fracture half-lengths using several models.

Upon completion of the Tight Gas Advisory System, a study of the suggestions made by the system was compared to leading expert’s designs with promising results. The current status / future work portion of this document will detail how we intend to move forward with this project.

Dynamic Conductivity Testing
The Project Management Plan and the Technology Status Assessment were both completed in April 2009. The system and equipment was moved to another laboratory with more space and set up to satisfy the standard safety conditions. The system was redesigned to better simulate the conditions that exist during an actual hydraulic fracturing operation. The pump was replaced with a diaphragm pump to ensure we could maintain control over the slurry injection rate. Different conditions of flow (rate and temperature) and characteristics of fracturing fluids (polymer concentrations, effect of cross-linker, and effect of breaker) are being tested. Marpaung (2008) already established that the polymer concentration, gas flux and the presence of breaker affect dynamic fracture conductivity. Also, static tests would give optimistic clean-up efficiencies when compared to dynamic conductivity tests.

Table 1 shows the parameters we have determined might have influence on dynamic fracture conductivity. We decided to set up an experimental schedule (based on design of experiment methodology) that would enable us to determine which variables are the most important. We will rank these variables once the experiments specified in our schedule are completed. The experimental schedule itself will be discussed in the ‘Current / Future Work’ section.

Parameters that may have an influence on dynamic fracture conductivity

Modeling of Gel Cleanup
Gel damage is a very complex problem and this phenomenon includes proppant pack damage inside fracture, filter cake deposition at the fracture face, and fracture fluid invasion in the formation adjacent to the fracture. We did a literature review and we present a summary below. May et al. (1997) investigated the effect of yield stress on fracture-fluid cleanup by using a numerical stimulation model. Balhoff and Miller (2002) derived an analytical model for non-Newtonian fracture fluid to predict fracture fluid clean-up by incorporating geometry and reservoir properties. Yi (2004) established a model and studied the effects of rheological parameters and injection rate. Ayoub et al (2006) presented some of the results of an investigation of fracture cleanup mechanisms and demonstrated strategies that mitigate the effects of excessive filter cake. Wang (2008) used a reservoir simulator to model how polymers in the fracture affect fracture fluid cleanup.

In our work, we investigated the process of displacement of gel by gas and modeled two-phase flow at the pore scale to obtain a better understanding of fracture fluid clean up mechanisms. Fracture clean up efficiency is believed to be a function of basic parameters such as proppant size, fluid rheology and imposed pressure gradients. Equations for stratified two-phase laminar flow in an idealized fracture were developed. In this model, the fracture is represented by a pipe coated with a layer of gel residue on its internal perimeter with a gas core flowing in the middle of the pipe. These equations describe when the gel starts to move and the velocity profile of both the gel and the gas once the yield point of the gel is exceeded. For this idealized case, the equations of motion and momentum in porous media are solved analytically with appropriate boundary conditions. Three flow patterns were identified for this system under imposed pressure gradients (Figure 1). They include:

1. Case 1: Gas core is flowing but the gel layer on the inside perimeter of the pipe is stationary.
2. Case 2: Gas core is flowing and the gel layer on the inside perimeter of the pipe is also flowing. However, the velocity profile of the gel is flat near the interface between the gel and the gas.
3. Case 3: Gas Core is flowing and the gel layer on the inside perimeter of the pipe is also flowing.

Figure 1: Velocity Profiles for stratified gas-gel laminar flow in pipe

Measurement of Yield Stress
To experimentally study the relationship between yield stress and polymer concentration, we developed and setup a yield stress testing apparatus (Figure 2b). Our apparatus is an adaptation of the apparatus developed by Zhu et al. (2001) - Figure 2a. Detailed experimental procedures and results can be seen in Wang (2009). A series of experiments were conducted to determine the relationship between yield stress, guar concentration and breaker concentration. The major conclusions were as follows:

1. Guar concentration has a significant impact on yield stress of fracturing fluid. The yield stress is very low when the guar concentration is 40lb/Mgal or less. Increasing guar concentration will increase yield stress rapidly.
2. Yield stress decreases dramatically when breaker is added to fracturing fluid. The yield stress becomes zero when the breaker concentration reaches some critical value. The relationship between guar concentration and optimal breaker concentration is almost linear.

Figure 3: Static Yield Stress of Polymers

Current Status
The key tasks to being undertaken are outlined below.

Assessment of Field Treatment Results
We are currently in the process of beginning a more involved field and application study of the software. We have been contacted recently by Newfield Exploration Company in regards to supporting the field study of the TGS Advisory System. The field study will consist of comparing current hydraulic fracturing designs in the field to ones recommended by the system. Production data will be computed with the software prediction. During this study we will not only be able to validate the program as making successful recommendations, but also learn how we can use the output from the software to reach the goal of optimal fracture treatment design. Initial meetings will be set up with Newfield Exploration Company in order to go over the system in detail and begin compiling data for the study.

Dynamic Conductivity Testing
The entire fracture conductivity laboratory was moved to a different location and was assembled. A new load frame capable of dynamically loading the conductivity cell was installed. Using this new frame, we will be able to measure the core displacement as a function of increasing load. This will lead to a more accurate measurement of the fracture width after closure. Figures 4a and b show schematics of the current setup while Figures 5 to 7 show actual pictures of the laboratory setup. A special software system is also used to record data (load, differential pressure and nitrogen flow rates) in real time (Figure 8).

We also installed 60 mesh screens in the high pressure vessels downstream of the conductivity cells. This was to prevent the plugging of the back pressure valves in the return line during the fracture conductivity measurement stage of the experiment. A systematic procedure for equipment use was developed and initial tests were ran to ensure the proper functioning of all the equipment in the test setup. We are currently running tests to confirm that we can control all the test variables (e.g. proppant concentration, slurry flow rates) accurately and also to ensure that the results we get under all anticipated test conditions are repeatable and consistent.

We developed an experimental schedule based on design of experiment methodology – Table 2. This schedule of experiments will serve as a screening tool for us to identify the most important parameters affecting dynamic fracture conductivity. In the future, we would use two-factor or if required multiple-factor experiments to refine the information derived from the screening experiments and develop a statistical model. The end point of this process is optimization, that is, the determination of the levels of the critical variables that will result in the best system performance.

Modeling of Gel Damage
For the analytical equation development part, the equations for stratified two-phase stratified flow will be derived not just for laminar flow but also for turbulent flow. An analytical expression for determining the minimum required velocity required to initiate flow of gel in the fracture would be derived. The model will be extended to more complex flow geometries that are more representative of proppant packs.

Advanced Treatment Design
The project team will develop a fracture treatment design program based on the findings of the literature and laboratory studies in the previous tasks, as well as on feedback from the project’s industry partners. The fracture treatment design program will include guidelines on the fracture treatment size, fluid type, proppant type, and pumping schedule for an optimal end-job proppant concentration. Guidelines will be tailored to specific reservoir conditions, including permeability, depth of target zone, pressure, and temperature.

Table 2: Experimental Schedule for screening out the least important factors

Technology Transfer
The following posters/presentations are as a result of work from this project.

Conference Papers:
Marpaung, F., Chen, F., Pongthunya, P., Zhu, D., and Hill, A.D. : ”Measurement of Gel Cleanup in a Propped Fracture With Dynamic Fracture Conductivity Experiments,” SPE paper 115653 presented at the 2008 SPE Annual Technical Conference and Exhibition, Denver, Colorado, 21-24 September.

Wei, Y.N., and Holditch, S.A.: “Computing Estimated Values of Optimal Fracture Half Length in the Tight Gas Sand Advisor Program,” SPE paper 119374 presented at the SPE Hydraulic Fracturing Technology Conference, 19-21 January 2009, The Woodlands, Texas.

Wang, Y., Holditch, S.A., and McVay, D.A.: “Modeling Fracture Fluid Cleanup in Tight Gas Wells,” SPE paper 119624 presented at the SPE Hydraulic Fracturing Technology Conference, Texas, 19-21 January 2009, The Woodlands, Texas.

Wei, Y.N., and Holditch, S.A.: “Multicomponent Advisory System Can Expedite Evaluation of Unconventional Gas Reservoirs,” SPE paper 124323 presented at the SPE Annual Technical Conference and Exhibition, 4-7 October 2009, New Orleans, Louisiana.

Posters:
Hydraulic Fracturing for Tight Gas Poster, presented at the SPE ATCE 2009, New Orleans.

Presentations:
RPSEA Presentation Texas A&M University, 2009.

RPSEA Unconventional Gas Conference, April 2009, Golden, CO. “Hydraulic Fracturing in Tight Gas Formations,” Crisman Institute for Petroleum Research, May 19, 2008.

“Technology Transfer Meeting on Unconventional Gas and Hydraulic Fracturing,” Crisman Institute for Petroleum Research, October 16, 2009.

“Advanced Hydraulic Fracturing”, Crisman Institute for Petroleum Research, February 18, 2009.

Thesis/Dissertations:
Wang, Y.: “Simulation of Fracture Fluid Cleanup and Its Effect on Long Term Recovery in Tight Gas Reservoirs” PhD Dissertation, Texas A&M University., College Station, Texas, 2008.

Wei, Yunan, “An Advisory System for the Development of Unconventional Gas Reservoirs” PhD Dissertation, Texas A&M University, College Station, Texas, 2008.

Marpaung, Fivman, “Investigation of the Effective of Gel Residue on Hydraulic Fracture Conductivity Using Dynamic Fracture Conductivity Test” M.S. Thesis, Texas A&M University, College Station, Texas, 2007.

Project Start: September 3, 2008
Project End: September 2, 2011

DOE Contribution: $1,045,551
Performer Contribution: $261,388

Contact Information:
RPSEA – Charlotte Schroeder (cschroeder@rpsea.org or 281-690-5506)
NETL – Virginia Weyland (Virginia.Weyland@netl.doe.gov or 918-699-2041)
TEES – Ding Zhu (dingzhu@tamu.edu or (979) 458-4522

References:
Ayoub, J.A., Hutchins, R.D., F. Van Der Bas: “New Findings in Fracture Cleanup Change Common Industry Perceptions,” SPE paper 98746, presented at SPE International Symposium and Exhibition on Formation Damage Control, Lafayette, LA, February 15-17, 2006.

Balhoff, M.T. and Miller, M.J. 2005. An Analytical Model for Cleanup of Yield-Stress Fluids in Hydraulic Fractures. SPEJ 10(1): 5-12. SPE-77596-PA.

Marpaung, F., Chen, F., Pongthunya, P., Zhu, D., and Hill, A.D. : ”Measurement of Gel Cleanup in a Propped Fracture With Dynamic Fracture Conductivity Experiments,” SPE paper 115653 presented at the 2008 SPE Annual Technical Conference and Exhibition, Denver, Colorado, 21-24 September.

May, E.A., Britt, L.K., and Nolte, K.G.: ”The Effect of Yield Stress on Fracture Fluid Cleanup,” SPE paper 38619 presented at the 1997 SPE Annual Technical Conference and Exhibition, San Antonio, Texas, October 5-8.

Wang, Lei: “Experimental Research on Yield Stress of Fracturing Fluid”. Internal Report, Texas A&M U., College Station, Texas, (2009).

Wang, Y.: “Simulation of Fracture Fluid Cleanup and Its Effect on Long Term Recovery in Tight Gas Reservoirs” PhD Dissertation, Texas A&M U., College Station, Texas, (2008).

Yi, X.: “Model for Displacement of Herschel-Bulkley Non-Newtonian Fluid by Newtonian Fluid in Porous Media and Its Application in Fracturing Fluid Cleanup.,” SPE paper 86491, presented at SPE International Symposium and Exhibition on Formation Damage Control, Lafayette, LA, February 18-20, 2004.

Zhu, L., Sun, N. Papadopoulos, K., and Kee, D.D.: “A Slotted Plate Device for Measuring Static Yield Stress,” J. Rheol, 45(5), September/October 2001, p. 1105-1122.