Lawrence Berkeley National Laboratory

Carbon dioxide (CO2) extraction from deep reservoirs is currently important in CO2 enhanced oil recovery (EOR) and may become more important in the future if interim CO2 storage becomes common. In late 2014, we were involved in a production test of liquid CO2 from the Middle Duperow dolostone at Kevin Dome, Montana. The test resulted in lowering the temperature at the well bottom to ∼2 °C, and showed that the well and reservoir had very low CO2 productivity. We have used the CO2 modeling capabilities of the TOUGH codes to simulate the test and to show that liquid CO2 in the reservoir changes to gas phase as the pressure is lowered in the well during production testing. The associated phase change and decompression combine to drastically lower the bottom-hole temperature, creating the potential for water ice or CO2 hydrate to form. By hypothesizing a relatively high-permeability damage zone near the well surrounded by lower permeability reservoir rock, we can match the observed pressure, temperature, and production rate. Moving from the Kevin Dome test to the question of CO2 extraction from deep reservoirs in general, we carried out a parametric study to investigate the effects of reservoir depth and transmissivity on CO2 production rate for a prototypical reservoir. Simulations show that depth and high transmissivity favor productivity. Complex phase changes within the ranges of P-T encountered in typical CO2 production wells affect production rates. The results of our parametric study may be useful for the preliminary feasibility assessment of CO2 extraction from deep reservoirs. © 2019 Society of Chemical Industry and John Wiley & Sons, Ltd.