07122-36

Goal
The goals of the project are to: understand the impact of fracturing fluid additives on gas flowback in tight gas sandstones; investigate the effectiveness of gas and solvent injection as treatment options to improve gas flowback; and determine the most optimum process for liquid block removal.

Performers
University of Tulsa (TU), Tulsa, OK 74104-3189

Background
Fracture stimulation treatments are frequently performed with polymeric fluids to improve the ability of the fracturing treatment to create a propped fracture, but invasion of aqueous fracturing fluids during stimulation operations can reduce the relative permeability to gas, resulting in a “block.” In low-permeability reservoirs, little or no production is possible when capillary pressures are comparable to reservoir pressures.

In this project, the research team will support the development of novel fluids for remediation of this problem by enhancing our understanding of the impact of fracturing fluid properties on post-treatment gas flowback in tight gas wells. The results of this study will enable better selection of post-stimulation treatment fluids for remediating poor-performing tight gas wells and also of fracture treatment fluid additives for future applications.

The research team will use laboratory experiments and modeling to formulate remedial fluid composition and treatment processes for field application. The team will then test the process on existing non-performing wells to assess its ability to enhance productivity.

Project deliverables will include monthly status reports and a final report on the results of the defined effort. The research team will build and maintain a web site with information about the project and updates as appropriate.

Potential Impacts
Many tight gas sand wells are not producing p to their potential. For example, wells in the Green River field are producing at levels equivalent to that in an unfractured well; 100 Mcfd. With sufficient cleanup to increase the effective fracture length, these well rates should improve to as high as 900 Mcfd, and that could increase the reserves to nine times current levels. The Green River field is a prime target for further research to identify the problems and apply remedial solutions. Such solutions can significantly increase the reserves of the fields that produce marginally in spite of extensive stimulation treatments.

Accomplishments
Since the start of the project in September 2008 the purchase and assembly of the gas coreflooding systems has been completed to the extent of conducting the first set of experiments.

In the first step of this study, the applicability of dry gas injection on polymer block removal from damaged cores was investigated. Samples of fracturing fluids similar to those used in the Merna field were obtained from service companies and were injected into Berea sandstone cores to simulate the polymer blocking. The experiments were carried out as per the experimental procedure developed and described in the proposal. The polymer was injected into the rock at pressures ranging from 100 to 500 psi until breakthrough was achieved. Subsequently, wet gas was flowed through the rock sample to displace the polymer. The results showed that the return (wet) gas permeability is significantly reduced although the polymer invasion was not necessarily complete. It was observed that the polymer is strained and concentrated in the first few millimeters of the injected portion while the water traveled through to the outlet end. The polymer concentration at the outlet end was observed to be lower than the injected polymer concentration.

Subsequent to the wet gas flow, dry gas was injected through the rock sample to facilitate drying of the polymer. Results from the experiments show that the removal of polymer from the rock takes a very long time and the ultimate gas flow rates are not close to the undamaged gas flow rates. The results from these experiments were presented at the RPSEA Unconventional Gas Conference held April 6-7, 2010, in Golden, CO.

Polymer induced damage (gel damage) may be more pronounced on the fracture conductivity itself rather than the rock matrix conductivity. The research team is simultaneously investigating whether dry gas injection will have any effect on gel drying and removal in fractures by simulating the flow in sandpacks in fractured rocks. In order to prevent polymer straining, observed in the Berea samples, a sandpack using 20/40 mesh size sand was constructed. The results from this experiment were also presented at the April 2010 conference.

In the interpretation of experimental results with model solutions, it is possible to remove gel damage in sand packs by dry gas injection. Fractures that are choked with unbroken gel could be treated with a dry gas injection procedure to remove the damage. This will be theme of the conference paper that has been submitted for SPE formation damage symposium.

The geological modeling is complete. Based on the uncertainty in geological spatial relationships, we have developed several alternate models and upscaled these models to various levels using a newly developed technology. The researchers will compare the results of fine scale vs. coarse scale models to understand how well the results compare before starting the history matching process.

The model predictions for rate of removal of gel damage from sandpacks and the fractured rock compare well with the experimental observations. Therefore, the model may be used to predict field scale behavior when dry gas is injected into a fractured gas well. A common observation in all these drying experiments is the reduction in the end point gas relative permeability, subsequent to complete drying. The gas flow rates do not recover 100% due to the accumulation of polymer at the pore throats formed by the sand grains. This could cause the reduction observed in the drying experiments.

A website containing the details of the project plan and some results is now published. The website is part of the academic website providing information on other projects handled by the principal investigator. http://www.personal.utulsa.edu/~jmahadevan/novel_fluids.html [external site]

Current Status
A Project Management Plan with a work breakdown structure that concisely addresses the objectives and approach for each task with all major milestones and decision points, and a Technology Status Assessment describing the state-of-the-art of the proposed technology were completed.

Laboratory studies on the effect of gas injection. Two different proppant packs, a sandpack and fracture pack, were prepared to test the effectiveness of dry gas injection to remove residual trapped gel. The sandpack and the fracture pack were subjected to invasion by typical aqueous fracturing liquids followed by a brief period of flowback, then a dry gas injection cycle followed by gas flowback. Nitrogen, saturated with water to 100% relative humidity, was used for conducting the gas flowback experiments. Experimental data collected includes flow rate of gas and the saturation of liquid in the core with time. The results show that the recovery of gas flow rates is greater when the proppant packs were subjected to dry gas treatment as opposed to a simple gas displacement process. The dry gas treatment completely removes the water content from the gel thereby leading to a recovery of gas flow rate up to 30% of the initial undamaged rate. In comparison, a simple gas displacement process allows only a maximum of 5% recovery of the undamaged flow rate of gas.

Laboratory studies on the effect of solvent injection. Additives such as methanol and surfactant can result in an improvement in the production of gas from reservoirs by enhancing the rate of evaporation and also by changing the wettability. Proppant packs were subjected to invasion by typical aqueous fracturing liquids, followed by isopropyl alcohol injection cycle and then gas flow back. Nitrogen was again be used for conducting the gas flowback experiments. Experiments that combine both cycles of dry gas injection and solvent injection followed by gas flowback were performed in sequence. Two different cycles of alcohol treatment were followed. The first treatment procedure involved alcohol injection as soon as the gas flowback and displacement was completed. Gas flow back was resumed subsequently to monitor the change in gas flow rate recovery. It was observed that the simple alcohol treatment resulted in a doubling of the gas flow rate which is most likely due to the higher volatility of alcohol combined with the greater level of dissolution of residual gel by the injected alcohol.

In the second set of treatment procedures, the proppant pack, with residual gel, was subjected to both dry gas and alcohol treatments. The proppant pack, with residual gel, was treated with alcohol and left to soak for 3 hours followed by a dry gas treatment. This procedure resulted in gel removal and gas flow rate recovery of almost 43% of the original undamaged gas flow rate of the proppant pack. The recovery of gas flow rates was also much faster with only 1 hour needed to recover 15% of the undamaged flow rate when the proppant pack was treated with alcohol.

When the proppant pack was treated with alcohol “after” a dry gas treatment, the improvement was not as significant. The most likely reason for the low improvement was that the gel becomes very concentrated and therefore insoluble in alcohol after the drying process.

Modeling of dry gas and solvent injection. A modeling study has been carried out to ascertain the role of both dry gas injection and solvent injection on gas flowback rates. Injection of dry gas will produce drying fronts that propagates through the rock core. When a solvent is injected, miscible displacement takes place with water partitioning between the gel and alcohol phases. Two regimes were considered for modeling during the gas flowback: displacement and evaporation. Models for both regimes, including relative permeability and capillary pressure effects, were developed from conservation equations for components of gas and liquid phases.

Field test of solvent/gas injection. The laboratory experiments are expected to provide quantitative information on the rate of cleanup of the fracturing liquids under different treatment procedures. Based on the laboratory injection studies and subsequent modeling, field requirements will be computed. Once the field treatment volumes of chemicals are identified, a field experiment can be performed with the coordination of an operating company partner. A build up test is planned for the summer of 2010 which will form a basis for characterizing the reservoir and the near wellbore. The primary objective of this well test will be to understand the near wellbore fracture characteristics in a much greater detail than previously possible.

Project Start: September 2, 2008
Project End: September 1, 2011

DOE Contribution: $ 219,920
Performer Contribution: $ 219,590

Contact Information:
RPSEA – Kent Perry (kent.perry@gastechnology.org or 847-768-0961)
NETL – Virginia Weyland (Virginia.Weyland@netl.doe.gov or 281-494-2517)
University of Tulsa – Jagannathan Mahadevan (jmahadevan@utulsa.edu or 918-631-3906)