A Generalized Partial Molar Algorithm Provides Fast Estimates of CO2 Storage Capacity in Depleted Oil and Gas Reservoirs

Tuesday, April 3, 2012: 4:00 PM
Grand Ballroom G (Hilton of the Americas)
Maria Barrufet and Ernesto Valbuena, Petroleum Engineering, Texas A&M University, College Station, TX

Understanding hydrocarbon fluid behavior is critical for the success of any gas injection project into a reservoir whether this is to recover additional oil or to store and sequester greenhouse gases including CO2. Fluid phase behavior of mixtures resulting from the injected streams and reservoir fluids determine the design of production and injection schemes and facilities.

This manuscript presents an analytical method to estimate the ultimate CO2 storage capacity in depleted oil and gas reservoirs by implementing a volume-constrained thermodynamic equation of state (EOS) using reservoir average pressure and fluid composition. This method can handle the impurities contained in the injection stream using a generalized partial molar volume definition.

The algorithm developed provides fast and thermodynamically consistent estimates of storage capacity enabling the selection of target storage reservoirs, schedule injection strategies and design surface facilities.

Results from this analytical method are in excellent agreement with those from a commercial reservoir simulator. A total of 24 numerical runs were conducted to evaluate scenarios with large pressure and compositional gradients while injecting. The reservoir used was heterogeneous, had a five-spot injection pattern and local grid refinement in the neighborhood of wells.  CO2 storage capacity was predicted with an average difference of 1.3% (on a molar basis) between analytical and numerical methods; the average oil, gas, and water saturations were also matched. Additionally, the analytical algorithm performed several orders of magnitude faster than numerical simulation, with an average of 5 seconds per run.

Extended Abstract: File Uploaded