CO2 huff-n-puff in shale cores: Live-oil vs. dead-oil

Abstract

Oil recovery in shale oil reservoirs by depressurization produces less than 10 % of the original oil in place (OOIP). Cyclic gas injection or huff-n-puff (HnP) is a promising EOR process for unconventional reservoirs. Most of the existing experimental works on huff-n-puff have been performed with dead-oil instead of live-oil. The dissolved gas in the live-oil can play a critical role in the huff-n-puff process. The goal of this paper is to compare the gas huff-n-puff process in shale cores using both live-oil and dead-oil. In our experiments, the live-oil was generated in situ by diffusing methane into dead-oil saturated cores. After live-oil generation, primary recovery and CO₂ huff-n-puff experiments were conducted sequentially. Oil recovery was obtained through core mass measurement, produced liquid analysis, and T₂ NMR inspection. Produced oil and gas compositions were analyzed by gas chromatography. A new analytical solution was proposed to model the pressure decline during the methane diffusion process to estimate the gas diffusion coefficient in oil-saturated porous media. Results indicate that a live-oil with a GOR of 402.6 SCF/STB was generated within the core. The diffusion coefficient of methane in an oil-saturated core was measured to be 4.52 × 10⁻¹⁰ m²/s. In the live-oil test, 8 % OOIP was recovered in primary recovery, the one cycle CO₂ HnP recovered an additional 32 % OOIP. In contrast, the dead-oil test results in 5 % OOIP in primary recovery and 19 % OOIP in the CO₂ huff-n-puff. The live-oil experiment shows higher oil recovery because the compressibility of live-oil allowing gas to enter the matrix through both convection and diffusion. However, in the dead-oil experiment, the lack of compressibility limits gas penetration to only diffusion, resulting in lower recovery. In field-scale, gas flow driven by convection would be more significant compared to core-scale.