A parametric study on reservoir cooling for enhanced oil recovery from CO2 injection

Open Access
Wang, Zhenzhen
Graduate Program:
Petroleum and Natural Gas Engineering
Master of Science
Document Type:
Master Thesis
Date of Defense:
April 15, 2013
Committee Members:
  • Russell Taylor Johns, Thesis Advisor
  • Zuleima T Karpyn, Thesis Advisor
  • Li Li, Thesis Advisor
  • reservoir cooling
  • CO2 flooding
  • MMP reduction
Whorton et al. (1952) received a patent for their development of an oil recovery method by CO2 injection. Since then, CO2 flooding for secondary and tertiary oil recovery has developed greatly leading to more incremental oil recovery worldwide due to high displacement efficiency and low cost. Thus far, gas flooding is the most widely used EOR methods even over thermal methods, primarily because CO2 is able to develop multiple contact miscibility with light crude oil at relatively low reservoir pressures. CO2 flooding also has the potential for the reduction of greenhouse gas emissions by subsurface sequestration of CO2. Many papers have reported that minimum miscible pressure (MMP) decreases with a reduction in temperature. The objective of this thesis is to examine the feasibility of reservoir cooling by injection of a coolant (water) followed by CO2 at temperatures lower than reservoir temperature. Cooling the reservoir may result in incremental oil production due to MMP reduction. We have conducted a parametric study of the CO2 injection process, and examined how injection temperature, initial reservoir pressure, formation heterogeneity, heat transfer with over/underburden formations, and WAG ratio affect the incremental oil recovery. We have performed waterflood and CO2 injection at two different temperatures, one below the initial reservoir temperature and the other one equals to the reservoir temperature. The results show that performing a waterflood at cooler temperatures rather than performing the waterflood at reservoir temperature can lead to substantial cooling and improved incremental recovery (about 15 to 23% OOIP with cooling versus 4 to 22% without cooling for homogeneous reservoirs with no heat gain from the surroundings). When heterogeneity and heat transfer with surrounding formations are considered, the incremental oil recoveries are substantially less, although still significant. Our results show that it is beneficial to conduct a long waterflood at a lower temperature before CO2 injection, especially for reservoirs with initial pressures close but just below the CO2 MMP. Also, neglecting temperature changes during history matching may result in inaccurate results.