SPE Journal
Volume 18, Number 2, April 2013, pp. 331-344

SPE-135563-PA

CO2 Injection in Vertical and Horizontal Cores: Measurements and Numerical Simulation

View full textPDF ( 1,298 KB )

DOI  More information 10.2118/135563-PA http://dx.doi.org/10.2118/135563-PA

Citation

  • Moortgat, J., Firoozabadi, A., Li, Z. et al. 2013. CO2 Injection in Vertical and Horizontal Cores: Measurements and Numerical Simulation. SPE J.  18 (2): 331-344. SPE-135563-PA. http://dx.doi.org/10.2118/135563-PA.

Summary

In certain high-pressure and low-temperature reservoirs, the density of CO2 may be substantially higher than the oil density. Upon mixing of CO2 and oil, a gas phase with a high content of methane (C1) may also appear. When the C1 content is high, this gas phase may have a lower density than the oil. In relation to this phenomenon, we have conducted three comprehensive experiments studying CO2 injection from the top and bottom of a vertical core and injection in a horizontal core. The injection rate is 1 PV/day. This low rate allows the study of diffusion. The core diameter used in this work is 3.8 cm and the length is 27.3 cm. The tests are conducted at a pressure of 441 bar and a temperature of 60°C. At 2.6 hydrocarbon pore volume injection (PVI), the coreflood results give a recovery of 98% for bottom injection, 84% for top injection, and 92% for horizontal injection. We have also conducted an extensive set of measurements to determine swelling, viscosity, and density for the calibration of an equation of state. We simulate the experiments using a state-of-the-art higher-order finite-element three-phase compositional model. The simulations suggest that the endpoint relative permeability of the CO2-rich phase may be lower than the oil phase. The results also show that Fickian diffusion should be taken into account, but that the diffusion coefficients are reduced, because the CO2/oil mixtures are in the near-critical region for much of the injection. Even for a horizontal core there is a considerable gravity effect. One main conclusion is that there may be vast differences between CO2 injection in a 1D slim tube and in a core where there may be a 2D flow. A related conclusion is that analysis of CO2 coreflooding may provide important parameters for field-scale problems.

© 2013. Society of Petroleum Engineers

View full textPDF ( 1,298 KB )

History

  • Original manuscript received: 2 May 2011
  • Meeting paper published: 4 October 2009
  • Revised manuscript received: 1 November 2011
  • Manuscript approved: 28 March 2012
  • Published online: 31 January 2013
  • Version of record: 5 April 2013