A Literature Review of Attempts to Increase the Viscosity of Dense Carbon Dioxide

Two important processes in the oil and gas industry that use dense carbon dioxide are fracture stimulation and enhanced oil recovery. For enhanced oil recovery, the problems with using carbon dioxide have been well studied and documented in laboratory and field studies. The low viscosity of carbon dioxide causes it to ‘finger’ towards the production wells and bypass large amounts of oil. Significant research has been conducted over the past 15 years searching for ways to increase the viscosity of (thicken) carbon dioxide. Because of the low viscosity of dense carbon dioxide, its effectiveness as a fracturing fluid has been questioned. Specifically, the effect of low fluid viscosity on proppant settling, placement of high sand concentration, and fluid leakoff is poorly understood. Nonetheless, Canadian Fracmaster has fracture stimulated thousands of wells in Canada via CO 2 sand fracturing (100% carbon dioxide and proppant) with great success. The U.S. Department of Energy has several initiatives to introduce CO -sand fracturing in the U.S. and demonstate its applicability as a non-damaging fracture 2 stimulation technique. Although most of the fracture stimulations have led to increased production/deliverability, the above questions with respect to the low viscosity still remain. This may in part be due to the trend over the last 7 years or more in the oil and gas industry to use high viscosity fluids, e.g. linear gels and cross-linked gels, for better fluid leakoff control and placement of higher sand concentrations. With the huge difference in viscosity between dense carbon dioxide and gelled fluids, an increase in the viscosity of carbon dioxide could improve the placement of more and larger sand particles and improve fluid leakoff. Hence the objective of this study was to review all previous and current research in the area of carbon dioxide viscosity enhancement and provide assessments of future research directions with respect to increasing the viscosity of carbon dioxide for CO -sand fracturing. Ideally, the viscosity of carbon dioxide 2 could be increased by a factor of 2-100 and the compound(s) used would be non-damaging to the formation (i.e., they would either be produced back with the gaseous carbon dioxide or be of such small quantity that porosity and permeability of the formation and proppant pack would not be reduced). The goal of most research efforts was to increase the viscosity of dense carbon dioxide via the dissolution of dilute concentrations (less than 1 wt. %) of ‘thickeners’. This review has indicated that three basic strategies have been followed. In the first, extremely high molecular weight polymers were considered. These compounds were either insoluble or sparingly soluble in carbon dioxide unless prohibitive amounts of co-solvent were added. In the second strategy, relatively low molecular weight compounds capable of forming viscosity-enhancing pseudo-networks of polymers via associations, hydrogen-bonding, or micelle formation were evaluated. These compounds contain polar groups that diminish carbon dioxide solubility. Hence, large amounts of co-solvent were required to enhance their CO 2 solubility in order to increase viscosity. The third group of studies included recent efforts to design novel viscosity enhancing molecules that exhibit very high carbon dioxide solubility. One high-molecular weight CO -thickening 2 polymer, a poly fluoroacrylate, and one low MW associative thickener, a fluoroether disulfate telechelic ionomer, have been identified. Both can increase the viscosity of carbon dioxide in concentrations of several weight percent without the need for a cosolvent, but they are synthesized using expensive, highly fluorinated precursors. Efforts to reduce the apparent viscosity of carbon dioxide mobility via non-aqueous emulsions (droplets of liquid carbon dioxide separated by films of an immiscible aqueous or oleic liquid) were also reviewed. For example, attempts have been made to develop emulsions that are about 95-98 vol% or more liquid carbon dioxide and 2-5 vol% non-aqueous films (unlike miscible displacement, water cannot be used in fracturing applications because the CO is injected at 2 temperatures well below the freezing point of water). These emulsions are relatively difficult to stabilize due to the difficulty in identifying effective surfactants for mixture of non-polar liquids. A limited amount of promising results were published, however. Several research groups continue to search for an effective carbon dioxide thickening agent. Their efforts continue to focus on (1) inexpensive thickening agents that require large volumes of co-solvents (e.g. 2 vol% thickener, 15% toluene, 83%carbon dioxide) or (2) expensive thickeners based on silicone-based or fluorine-based compounds designed to exhibit high carbon dioxide-solubility without any co-solvent. In the near-term, neither of these efforts is likely to provide an economically viable method for enhancing the performance of CO . In the long term, however, these 2 efforts may yield economic thickeners that require little or no co-solvent. The use of non-aqueous emulsions may still be a viable option because very small amounts of the non-aqueous phase are required (e.g. 1/10th to 1/50th the amount of liquid carbon dioxide). Further, small amounts of surfactants are required to stabilize the emulsion (e.g. 1/100 of the non-aqueous phase). Much progress has been made in the 1990's in surfactant development for chemical engineering processes employing liquid CO . Although these surfactants are 2 expensive because they are based on silicones or fluorinated functionalities, they would be required in very small amounts. A substantial portion of the surfactant and non-aqueous fluid would probably remain within the fracture and formation upon depressurization because of their low solubility in natural gas. Therefore, the benefits of the increased viscosity of the carbon dioxide would possibly be mitigated by the potential formation damage caused by these residual fluids. TABLE OF CONTENTS I. THE NEED FOR ENHANCED CARBON DIOXIDE VISCOSITY 1 1.0 Technical Issues and Objectives 1 1.1. Carbon Dioxide Miscible Displacement 1 1.2. Fracturing Formations with Liquid Carbon Dioxide 4 1.3. Previous Attempts to Decrease the Mobility of Carbon Dioxide 7 1.3.a. Reduction in Relative Permeability via Water Injection for Mobility Control 7 1.3.b. CO -Foams for Profile Modification, Mobility Control, and Fracturing 7 2 1.3.c. Identification of a Carbon Dioxide Thickening Agent for Mobility Control 9 NMIMT 10 University of Wyoming 11 NIPER 11 Canadian Fracmaster 13 University of Pittsburgh 13 University of North Carolina at Chapel Hill 18 Chevron 20 1.4 Barriers to Development of a Carbon Dioxide Thickening Agent 22 1.5 Summary 25 1.6 Guidelines for Design of Proposed Thickening Agents Based on the Literature Review 26 APPENDIX A Current CO -Thickening Research at the University of Pittsburgh 32 2