Two laboratory experiments were performed to investigate the molecular weight selectivity of crude oil hydrocarbons mobilized into CO2 at the reservoir temperature (42 °C) and pressures ranging from below the minimum miscibility pressure (MMP) of 9.69 MPa (1406 psi) to pressures well above the MMP (8.27–27.58 MPa, 1200–4000 psi). The hydrocarbon composition at equilibrium in the CO2-rich upper "miscible" phase was measured, as was the ability of four pore volumes of CO2 to recover crude oil hydrocarbons from a sand bed. Both experiments showed significant selectivity against producing higher molecular weight hydrocarbons at lower pressures. In addition, the bias against higher molecular weight hydrocarbons continued even at pressures well above the 9.69 MPa MMP. For example, both the total hydrocarbon concentration in the CO2-rich "miscible" phase and the fraction of C21–C36 increased dramatically from 10.34 to 21.13 MPa (from 1500 to 3500 psi), even though both pressures were above the MMP. In addition, when crude oil was first equilibrated at 20.68 MPa (3000 psi) and then lowered before sampling the "miscible" phase, substantial deposition of the higher molecular weight hydrocarbons occurred with the result that the hydrocarbon composition and concentrations were similar to those when the crude oil was exposed to only the lower CO2 pressures without first being exposed to any higher pressure. The sand flush experiments showed similar results to the "miscible" phase samplings in that both the total hydrocarbons and the fraction of C21–C36 recovered increased dramatically with pressure, regardless of whether the pressures were below or more than double the MMP. The crude oil samples remaining after CO2 exposures also showed large increases in molecular weight as well as higher viscosity and lower American Petroleum Institute (API) gravity than the original crude oil. As predicted by both lab experiments, produced crude oil samples collected from two wells before and after CO2 breakthrough showed significant bias against high-molecular-weight hydrocarbons, with the fraction of C21–C36 produced dropping by as much as 60–80% after CO2 breakthrough compared to crude oil samples collected from the same wells before CO2 breakthrough. These results verify that crude oil during a CO2 flood does not move as a homogeneous "miscible" phase and that hydrocarbon dissolution (vaporization into the "miscible" phase) of lower molecular weight hydrocarbons dominates oil recovery in both the field and lab experiments.
Horizontal well drilling and multistage hydraulic fracturing have been demonstrated as effective approaches for stimulating oil production in the Bakken tight oil reservoir. However, after multiple years of production, primary oil recovery in the Bakken is generally less than 10% of the estimated original oil in place. Gas huff 'n' puff (HnP) has been tested in the Bakken Formation as an enhanced oil recovery (EOR) method; however, most field pilot test results showed no significant incremental oil production. One of the factors affecting HnP EOR performance is premature gas breakthrough, which is one of the most critical issues observed in the field because of the presence of interwell fractures. Consequently, injected gas rapidly reaches adjacent production wells without contacting reservoir rock and increasing oil recovery. Proper conformance control is therefore needed to avoid early gas breakthrough and improve EOR performance. In this study, a rich gas EOR pilot in the Bakken was carefully analyzed to collect the essential reservoir and operational data. A simulation model with 16 wells was then developed to reproduce the production history and predict the EOR performance with and without conformance control. EOR operational strategies, including single- and multiple-well HnP, with different gas injection constraints were investigated. The simulation results of single-well HnP without conformance control showed that a rich gas injection rate of at least 10 MMscfd was needed to yield meaningful incremental oil production. The strategy of conformance control via water injection could significantly improve oil production in the HnP well, but injecting an excessive amount of water also leads to water breakthrough and loss of oil production in the offset wells. By analyzing the production performance of the wells individually, the arrangement of wells was optimized for multiple-well HnP EOR. The multiwell results showed that rich gas EOR could improve oil production up to 7.4% by employing conformance control strategies. Furthermore, replacing rich gas with propane as the injection gas could result in 14% of incremental oil production.
Aimed at advancing gas injection enhanced oil recovery (EOR) technologies in unconventional reservoirs, this study comprised a series of activities to bridge the gap between the theoretical study and actual field applications. Twenty-four EOR pilot tests were collected from the major unconventional plays in North America to evaluate the performance of different EOR technologies. Fit-for-purpose experiments and simulations were performed to investigate the effects of injection rate and pressure on EOR performance, as well as to reveal the effectiveness of huff "n" puff (HnP) cycles in actual field operations. The selection of injection rate and pressure as key parameters for investigation was based on field observations and communications with oil and gas operators because these two parameters play critical roles in both facility design and overall cost for an EOR project. Results showed that miscible EOR with a high gas injection rate and pressure is required for field operations because the injected gas needs to penetrate and extract oil from the tight matrix. Experimental results indicated that there is a correlation between oil recovery and the logarithm of core volume for miscible EOR. Immiscible gas EOR could not yield a satisfactory EOR response in actual reservoirs because the injected gas tends to flow through fractures instead of penetrating the matrix to interact with oil. Results also showed that reaching minimum miscibility pressure (MMP) does not guarantee an optimum EOR operation in unconventional reservoirs. Pressure higher than MMP is preferred in field operations. When designed properly, up to a tenfold oil production rate boost is achievable in field applications within a short period. However, such a high-performance operation is only effective in the first several HnP cycles due to the limited gas penetration depth into the rock matrix.
This study presents the results of a detailed life cycle analysis of greenhouse gas (GHG) emissions associated with carbon dioxide-enhanced oil recovery (CO2-EOR) where the CO2 is sourced from a coal-fired power plant. This work builds upon previous investigations and integrates new information to provide more plausible ranges for CO2 storage in the reservoir during CO2-EOR. The system model includes three segments: upstream, gate-to-gate, and downstream processes. Our base case model using Ryan–Holmes gas separation technology for the CO2-EOR site determined the emissions from upstream, gate-to-gate, and downstream processes to be 117, 98, and 470 kg CO2e/bbl (CO2 equivalents per barrel of incremental oil produced), respectively, for total emissions of 685 kg CO2e/bbl. However, these emissions are offset by CO2 storage in the reservoir and the resulting displacement credit of U.S. grid electricity, which results in a net life cycle emission factor of 438 kg CO2e/bbl. Therefore, CO2-EOR produces oil with a lower emission factor than conventional oil (∼500 kg CO2e/bbl). Optimization scenarios are presented that define a performance envelope based on the CO2 capture rate and net CO2 utilization and suggest that lower emission factors below 300 kg CO2e/bbl are achievable. Based on these results, CO2-EOR where the CO2 is sourced from a coal-fired power plant provides one potential means for addressing the energy demand–climate change conundrum, by simultaneously producing electricity and oil to meet growing energy demand and reducing GHG emissions to abate global warming.
Carbon capture and storage (CCS) is one approach being studied by the U.S. Department of Energy to help mitigate global warming. The process involves capturing CO 2 emissions from industrial sources and permanently storing them in deep geologic formations (storage reservoirs). However, CCS projects generally target “green field sites,” where there is often little characterization data and therefore large uncertainty about the petrophysical properties and other geologic attributes of the storage reservoir. Consequently, ensemble-based approaches are often used to forecast multiple realizations prior to CO 2 injection to visualize a range of potential outcomes. In addition, monitoring data during injection operations are used to update the pre-injection forecasts and thereby improve agreement between forecasted and observed behavior. Thus, a system for generating accurate, timely forecasts of pressure buildup and CO 2 movement and distribution within the storage reservoir and for updating those forecasts via monitoring measurements becomes crucial. This study proposes a learning-based prediction method that can accurately and rapidly forecast spatial distribution of CO 2 concentration and pressure with uncertainty quantification without relying on traditional inverse modeling. The machine learning techniques include dimension reduction, multivariate data analysis, and Bayesian learning. The outcome is expected to provide CO 2 storage site operators with an effective tool for timely and informative decision making based on limited simulation and monitoring data.