A Pilot Carbon Dioxide Test, Hall Gurney Field, Kansas

A pilot carbon dioxide miscible flood was initiated in the Lansing Kansas City C formation in the Hall Gurney Field, Russell County, Kansas. The reservoir zone is an oomoldic carbonate located at a depth of about 2900 feet. The pilot consists of one carbon dioxide injection well and three production wells. Continuous carbon dioxide injection began on December 2, 2003. By the end of June 2005, 16.19 MM lb of carbon dioxide was injected into the pilot area. Injection was converted to water on June 21, 2005 to reduce operating costs to a breakeven level with the expectation that sufficient carbon dioxide was injected to displace the oil bank to the production wells by water injection. By March 7,2010, 8,736 bbl of oil were produced from the pilot. Production from wells to the northwest of the pilot region indicates that oil displaced from carbon dioxide injection was produced from Colliver A7, Colliver A3, Colliver A14 and Graham A4 located on adjacent leases. About 19,166 bbl of incremental oil were estimated to have been produced from these wells as of March 7, 2010. There is evidence of a directional permeability trend toward the NW through the pilot region. The majority of the injected carbon dioxide remains in the pilot region, which has been maintained at a pressure at or above the minimum miscibility pressure. Estimated oil recovery attributed to the CO2 flood is 27,902 bbl which is equivalent to a gross CO2 utilization of 4.8 MCF/bbl. The pilot project is not economic.

A pilot carbon dioxide miscible flood was initiated in the Lansing Kansas City C formation in the Hall Gurney Field, Russell County, Kansas. The reservoir zone is an oomoldic carbonate located at a depth of about 2900 feet. The pilot consists of one carbon dioxide injection well and three production wells. Continuous carbon dioxide injection began on December 2, 2003. By the end of June 2005, 16.19 MM lb of carbon dioxide were injected into the pilot area. Injection was converted to water on June 21, 2005 to reduce operating costs to a breakeven level with the expectation that sufficient carbon dioxide has been injected to displace the oil bank to the production wells by water injection. By December 31, 2006, 79,072 bbls of water were injected into CO2 I-1 and 3,923 bbl of oil were produced from the pilot. Water injection rates into CO2 I-1, CO2 No.10 and CO2 No.18 were stabilized during this period. Oil production rates increased from 4.7 B/D to 5.5 to 6 B/D confirming the arrival of an oil bank at CO2 No.12. Production from wells to the northwest of the pilot region indicates that oil displaced from carbon dioxide injection was produced from Colliver No.7, Colliver No.3 and possibly Graham A4 located on an adjacent property. There is evidence of a directional permeability trend toward the NW through the pilot region. The majority of the injected carbon dioxide remains in the pilot region, which has been maintained at a pressure at or above the minimum miscibility pressure. Our management plan is to continue water injection maintaining oil displacement by displacing the carbon dioxide remaining in the C zone,. If the decline rate of production from the Colliver Lease remains as estimated and the oil rate from the pilot region remains constant, we estimate that the oil production attributed to carbon dioxide injection will be about 12,000 bbl by December 31, 2007. Oil recovery would be equivalent to 12 MCF/bbl, which is consistent with field experience in established West Texas carbon dioxide floods. The project is not economic.

Summary A pilot carbon dioxide (CO2) -miscible flood was initiated in the Lansing-Kansas City C formation in the Hall-Gurney Field, Russell County, Kansas. The reservoir zone is an oomoldic limestone located at a depth of approximately 2,900 ft. The pilot consisted of one CO2 injection well and three production wells. Continuous CO2 injection began in December 2003 and continued through June 2005, at which point 16.19 million lbm of CO2 had been injected into the pilot area. Injection was converted to water in June 2005 to reduce operating costs to a break-even level with the expectation that sufficient CO2 was injected to displace the oil bank to the production wells by water injection. By March 2010, 8,736 bbl of oil had been produced from the pilot. Production from wells to the northwest of the pilot region indicated that oil displaced by CO2 injection was produced from five wells outside of the pilot area, to the northwest. Approximately 19,166 bbl of incremental oil was estimated to have been produced from these wells as of March 2010. There was evidence of a directional permeability trend toward the northwest through the pilot region. The majority of the injected CO2 remained in the pilot region, which was maintained at or above the minimum miscibility pressure (MMP). Although the four-well pilot was uneconomical, the estimated oil recovery attributed to the CO2 flood is 27,902 bbl, which is equivalent to a gross CO2 usage of 4.8 Mcf/bbl.

A pilot carbon dioxide miscible flood was initiated in the Lansing Kansas City C formation in the Hall Gurney Field, Russell County, Kansas. The reservoir zone is an oomoldic carbonate located at a depth of about 2900 feet. The pilot consists of one carbon dioxide injection well and two production wells on about 10 acre spacing. Continuous carbon dioxide injection began on December 2, 2003. By the end of June 2005, 16.19 MM lb of carbon dioxide were injected into the pilot area. Injection was converted to water on June 21, 2005 to reduce operating costs to a breakeven level with the expectation that sufficient carbon dioxide has been injected to displace the oil bank to the production wells by water injection. Wells in the pilot area produced 100% water at the beginning of the flood. Oil production began in February 2004, increasing to an average of about 3.78 B/D for the six month period between January 1 and June 30, 2005 before declining. By June 30, 2006, 41,566 bbls of water were injected into CO2I-1 and 2,726 bbl of oil were produced from the pilot. Injection rates into CO2I-1 declined with time, dropping to an unacceptable level for the project. The injection pressure was increased to reach a stable water injection rate of 100 B/D. However, the injection rate continued to decline with time, suggesting that water was being injected into a region with limited leakoff and production. Oil production rates remained in the range of 3-3.5 B/D following conversion to water injection. Oil rates increased from about 3.3 B/D for the period from January through March to about 4.7 B/D for the period from April through June. If the oil rate is sustained, this may be the first indication of the arrival of the oil bank mobilized by carbon dioxide injection. A sustained fluid withdrawal rate of about 200 B/D from CO2 No.12 and CO2 No.13 appears to be necessary to obtain higher oil rates. There is no evidence that the oil bank generated by injection of carbon dioxide has reached either production well. Water injection will continue to displace oil mobilized by carbon dioxide to the production wells and to maintain the pressure in the PPV region at a level that supports continued miscible displacement as the carbon dioxide is displaced by the injected water.

A pilot carbon dioxide miscible flood was initiated in the Lansing Kansas City C formation in the Hall Gurney Field, Russell County, Kansas. Continuous carbon dioxide injection began on December 2, 2003. By the end of December 2004, 11.39 MM lb of carbon dioxide were injected into the pilot area. Carbon dioxide injection rates averaged about 242 MCFD. Vent losses were excessive during June as ambient temperatures increased. Installation of smaller plungers in the carbon dioxide injection pump reduced the recycle and vent loss substantially. Carbon dioxide was detected in one production well near the end of May and in the second production well in August. No channeling of carbon dioxide was observed. The GOR has remained within the range of 3000-4000 for most the last six months. Wells in the pilot area produced 100% water at the beginning of the flood. Oil production began in February, increasing to an average of about 2.35 B/D for the six month period between July 1 and December 31. Cumulative oil production was 814 bbls. Neither well has experienced increased oil production rates expected from the arrival of the oil bank generated by carbon dioxide injection.

A pilot carbon dioxide miscible flood was initiated in the Lansing Kansas City C formation in the Hall Gurney Field, Russell County, Kansas. Continuous carbon dioxide injection began on December 2, 2003. By the end of June 2004, 6.26 MM lb of carbon dioxide were injected into the pilot area. Carbon dioxide injection rates averaged about 250 MCFD. Carbon dioxide was detected in one production well near the end of May. The amount of carbon dioxide produced was small during this period. Wells in the pilot area produced 100% water at the beginning of the flood. Oil production began in February, increasing to an average of about 2.5 B/D in May and June. Operational problems encountered during the initial stages of the flood were identified and resolved.

The Rock Creek Field, discovered in 1906, is the site of a pilot carbon dioxide flood designed and installed by Pennzoil Company with financial support from ERDA. Favorable reservoir characteristics and conditions, the inability to successfully stimulate production by conventional waterflooding, and the production by conventional waterflooding, and the fact that nearly 150,000,000 barrels of oil will remain in the reservoir after completion of the low pressure gas recycles now in operation were the main pressure gas recycles now in operation were the main factors responsible for the initial study of miscible fluid displacement by carbon dioxide injection. After laboratory tests indicated the reservoir oil and carbon dioxide to be miscible at a reasonable pressure, the design and installation work for the pilot pressure, the design and installation work for the pilot was initiated. Currently in the water injection or pressure build-up phase, the project has progressed pressure build-up phase, the project has progressed satisfactorily with injection of carbon dioxide scheduled for early 1978. Introduction The Rock Creek Field, which has produced from the Big Injun Sand since 1906, is located in Roane County, West Virginia. During the 71 year life of the field, seven separate projects utilizing three different secondary recovery methods, have been attempted. Low pressure gas recycling, which was implemented in 1935 and continues today, has proven to be the only effective extra recovery mechanism. Waterflooding efforts failed to bank significant quantities of oil with failure attributed to the high connate water saturation, and steamflooding efforts failed due to the low injectivity encountered in the project. Primary recovery by solution-gas drive and project. Primary recovery by solution-gas drive and the secondary recovery by gas recycle account for a total recovery of approximately 20% of the original stock tank oil-in-place (OSTOIP). Therefore, 80% of the OSTOIP will remain after gas recycling. The amount of oil remaining in place coupled with a favorable reservoir temperature, a permeability profile with a high degree of homogeneity, and profile with a high degree of homogeneity, and successful laboratory miscibility tests provided the incentive for a pilot test of a miscible carbon dioxide flood. Pennzoil Company designed, developed, and installed this pilot project with financial assistance from the Energy Research and Development Administration of the United States Government (ERDA). The pilot consists of two contiguous five- spot patterns encompassing approximately ten acres each. These dual five spots are surrounded by thirteen back-up water injection wells drilled as or converted to water injection wells for the sole purpose of containing the injected carbon dioxide. purpose of containing the injected carbon dioxide. Prior to carbon dioxide injection, the reservoir Prior to carbon dioxide injection, the reservoir pressure of 91 psia must be increased to near 1000 pressure of 91 psia must be increased to near 1000 psia. This pressure increase will be accomplished psia. This pressure increase will be accomplished by water injection and monitored by pressure fall-off tests. The project is currently in the water injection or pressure build-up phase. The objectives of this pilot are to determine the oil recovery efficiency of a multiple contact carbon dioxide miscible flood and to define and evaluate operational problems associated with this particular recovery method. The quantitative data particular recovery method. The quantitative data gathered as a result of the installation of the project and the injection performance experienced project and the injection performance experienced thus far conform to expectations. GENERAL FIELD INFORMATION The Rock Creek Field is located in the southeastern portion of Roane County, West Virginia, approximately twenty miles northeast of Charleston, the capitol of West Virginia. The acreage included in the Rock Creek Field makes up the larger portion of the Rock Creek Trend which also includes the Hammack Field to the northeast. The trend itself encompasses 11,200 acres, laterally extends 12.5 miles and ranges between 1.5 and 5 miles in width. (See Figure No. 1) Field History - The discovery well in the Rock Creek Trend was drilled and completed in 1906.

In this study, a rising bubble apparatus (RBA) was used to determine the CO2 MMP for various oils. RBA tests permit direct observation of changes in bubble behaviour and thus offered insight into the phase behaviour for the CO2-reservoir fluid system. The CO2 MMPs were estimated for two Steelman reservoir fluids with a high gas-oil ratio, the partially flashed reservoir fluids, and the dead oils. The composition of solution gas of each partially flashed reservoir fluid was determined and the effects of different gas components were analysed. The MMP was also determined and discussed for Weyburn reservoir fluids which had a low gas-oil ratio with pure and impure CO2. The results of this study demonstrated that the effect of solution gas in oil on CO2 MP could be significant. Furthermore, achieving a miscible CO2 flood (in a reservoir with a Steelman-like reservoir fluid) could be possible at a lower operating pressure than the measured CO2 MMP, by partially depleting the reservoir. This may be the only option for some reservoirs which cannot sustain the relatively high pressure required for achieving miscibility. Introduction Carbon dioxide flooding is a proven oil recovery process(1). Over the last decade, carbon dioxide injection has become the leading enhanced oil recovery (EOR) process for light oils(2). CO2 injection can prolong, by 15 to 20 years, the production life of light oil fields nearing depletion under waterflood, and may recover 15 to 25% of the original oil in place. More than 20 years of field experience of CO2 injection has advanced the CO2 technology. CO2 injection can be introduced gradually and use some of the same equipment currently in place for waterflooding. Saskatchewan's light and medium oil resource, representing nearly 45% of proven reserves, has been on waterflood for over 30 years and is fast approaching its economic limit of production(3). For the light and medium oil reservoirs in Saskatchewan, carbon dioxide or hydrocarbon injection is considered to be the most effective EOR process(4–6). These gases can be injected into the reservoir to develop miscible or immiscible conditions with the oil depending upon the operating pressure. Carbon dioxide is preferred over hydrocarbon gases (e.g., ethane, propane) because it is cheaper, has higher density, and offers environmental benefits by providing storage for CO2 in the reservoir. The Saskatchewan Research Council (SRC) is conducting a comprehensive research program to assess the suitability of miscible CO2 displacement for reservoirs in southeast Saskatchewan and to optimize the field operating procedures. The first step in determining if a field is a viable CO2 flood candidate is to conduct a screening study to provide a reasonable estimate of CO2 injection performance. The minimum miscibility pressure (MMP) is a key parameter used in the assessment. It has been recognized that the CO2 MMP for a reservoir oil depends on the reservoir temperature, oil composition, and the purity of injected CO2. The minimum miscibility pressure increases with increasing reservoir temperature.

Huff-n-puff gas injection enhanced oil recovery has received increased attention especially in the unconventional plays like the Eagle Ford, where oil recovery is as low as 5 - 10%. An increase in 1% of recovery could realize a potential of 2.3 billion barrels of oil, which has an enormous economic value. Through a laboratory investigation of huff-n-puff conducted on preserved Eagle Ford samples; we evaluate different factors that can affect the recovery performance such as minimum miscibility pressure (MMP), surface area, soaking time, injection pressure, composition of injection gas and injection gas rate. In addition, different recovery mechanisms such as a vaporization (concentration gradient) and a miscible flowback (pressure gradient) were also investigated. Two sets of experiments were conducted utilizing a high-pressure chamber: one with Eagle Ford oil, providing MMP values using a VIT technique and vaporization test with different soaking times (2 days, 4 days, and 6 days). Another set of experiment were performed with preserved Eagle Ford samples. Different types of gas: carbon dioxide (CO2,) methane (C1), ethane (C2), C1:C2 (72:28) mixture, C1:C2 (95:5) mixture, and field gas were injected at various pressures from 1000 psi below MMP, MMP to 1000 psi above MMP with various soaking time of (1 hr, 3 hr and 6 hr). Nuclear Magnetic Resonance (NMR), HAWK pyrolysis, isothermal nitrogen adsorption tests (BET), Mercury Injection Capillary Pressure (MICP) and Gas Chromatography (GC), were performed to qualitatively and quantitively monitor the changes in Eagle Ford hydrocarbons recovered from shale samples. The experimental results demonstrated that: 1) as methane concentration in gas is increased, MMP also increased, 2) residence time (soaking time + production time) controls the recovery, 3) injection pressure determines the fraction of hydrocarbons mobilized, 4) surface area variation studies showed that the samples with higher surface area have greater recoveries, 5) ethane showed the best performance of all the gases tested (40% recovery). CO2 performed the second best (32%). C1:C2(72:28) mixture and field gas exhibit the similar efficacy in recovery (24% and 21%). C1:C2(95:5) mixture showed the worst recovery (11%). 6) high injection rate yielded better recovery (37%) than low injection rate (24%), 6) Increase in pore surface area by factor of 2.5 was observed from the opening of small pores and pore- throat on post huff-n-puff sample. In addition, recovery mechanism study shows that miscible flow back mobilized hydrocarbon up to C30, vaporization at 1000psi above MMP mobilized hydrocarbon up to C23 and vaporization 1000psi below MMP mobilized hydrocarbon up to C15. The results also indicated that longer soaking times increased diffused oil concentration in vapor phase.

Publication Rights Reserved This paper is to be presented at the 39th Annual Fall Meeting to be held in Houston, Tex., on Oct. 11–14, 1964, and is considered property of the Society of Petroleum Engineers. Permission to publish is hereby restricted to an abstract of not more than 300 words, with no illustrations, unless the paper is specifically released to the press by the Editor of the Journal of Petroleum Engineers or the Executive Secretary. Such abstract should contain conspicuous acknowledgment of where and by whom the paper is presented. Publication elsewhere after publication in JOURNAL OF PETROLEUM TECHNOLOGY or SOCIETY OF PETROLEUM ENGINEERS JOURNAL is granted on request, providing proper credit is given that publication and the original presentation of the paper. Discussion of this paper is invited. Three copies of any discussion should be sent to the Society of Petroleum Engineers office. Such discussion may be presented at the above meeting and, with the paper, may be considered for publication in one of the two SPE magazines. Abstract The problem confronted in the initiation of the Adell Field Multi-zone Lansing-Kansas City Waterflood was isolation of injection because of the behind-the-pipe communication between the various producing intervals. This problem is common to producing operations in Western Kansas because of past drilling and completion procedures. In general, past waterflooding practices for the Lansing-Kansas City reservoir make no particular effort to control injection, disregarding reserves left in the intervals receiving inadequate water input. The potential secondary reserves in the Adell Lansing-Kansas City Field were sufficient to justify something better than uncontrolled water injection. To gain maximum secondary recovery, nine new input wells were drilled around the periphery of the field. By the use of a specific type of drilling mud, casing cementing procedure, and completion program, injection of water was controlled into the zones proposed for waterflooding. Introduction The Adell Field, Sheridan County, Kansas, is a multi-zone Lansing-Kansas City lime reservoir which has produced over 5 million barrels of primary oil since discovery in 1945., Fig. 1. Pilot waterflood operations proved the feasibility of secondary recovery and the seriousness of the communication problem. Approximately 3 million barrels of secondary oil were predicted, providing each zone of the Lansing-Kansas City reservoir could be effectively waterflooded. To assure maximum effective waterflood operations, the isolation of injection appeared necessary. The use of existing producing wells for injection was deemed unsound, because of the cost and difficulty involved with behind-the-pipe repair. Severe communications existed in the producing wells causing the foreseeable problem of isolation of water injection into the desired intervals. At this point, further investigation was made of the exact nature of the communication problem, beginning with core samples of the Lansing-Kansas City shales. Laboratory analyses of these shale samples showed a high percentage of calcareous material. The samples disintegrated in fresh water, salt water and acid, but were stable in oil and oil-water emulsions. This knowledge gave thought that communication existed because of the hydration and sloughing of the shale sections separating each Lansing-Kansas City zone, when the wells were drilled and cemented with conventional water-based fluids. This was confirmed by the character of caliper logs run through the Lansing-Kansas City formation. The decision was then made to proceed with the planned waterflood operations by drilling nine injection wells, Fig. 2. The drilling program was designed to minimize shale hydration by use of an inverted emulsion mud and use of salt cement with a fluid loss additive during casing setting operations. Nine injection wells were drilled in the above manner, with final completion made by perforating one-foot intervals in six prospective waterflood zones.

One of the essential parameters in carbon dioxide (CO2) miscible flooding is the minimum miscibility pressure (MMP). Minimum miscibility pressure (MMP) is defined as the lowest pressure at which recovery of oil is (90–92%) at injection (1.2 PV) of carbon dioxide (CO2). The injected gas and oil become a multi-contact miscible at a fixed temperature. Before any field trial, minimum miscibility pressure (MMP) must be determined. This parameter must be determined before any field trial because any engineer needs a suitable plan to develop an injection and surface facilities environment. Estimation of reliable (MMP) maybe by traditional laboratory techniques, but it is very costly and time-consuming. Also, it can rely on various literature (MMP) empirical correlations, but this is not a good strategy because each minimum miscibility pressure (MMP) correlation relates to a unique formation condition.

This numerical study investigates the potential of mutually fulfilling EOR and sequestration objectives of carbon dioxide (CO2) injection into a hypothetical light-oil, carbonate reservoir. Formation and fluid properties were selected considering the ranges published in the literature for light-oil, carbonate reservoirs. The potential sources of CO2 were assumed to be the upstream and downstream activities in the reservoir geographic area, such as atmospheric emissions from refineries and steam boilers operating in nearby fields. Therefore, the example case is intended to explore sequestration and recovery potential of enhanced oil recovery (EOR) by CO2 injection (CO2-EOR) — not in isolation but as part of regular oil-field operations. The light oil carbonate reservoir meets the EOR screening criteria for miscible CO2 flood. An equation of state was used to characterize the reservoir hydrocarbon system to generate phase envelope and minimum miscibility pressure (MMP) via a ‘mixing cell method’ and benchmarked against published correlations. Compositional numerical simulations of waterflooding, CO2-flooding, water-alternating-gas with CO2 (WAG-CO2), and methane (CH4) were performed under the assumptions of anticipated reservoir management and operational conditions. The EOR and sequestration potentials of CO2 flooding and WAG-CO2 were quantified by CO2 gross and net utilization ratios (GUR and NUR, respectively) and retention capacity (RC). Simulation results yielded the highest incremental oil recovery for the WAG-CO2 flooding surpassing that of the continuous CO2 flooding by a factor of 1.75. Typical GURs and NURs for continuous CO2 flooding were 29,000 SCF/STB and 9,600 SCF/STB, respectively. For WAG-CO2, the gross CO2 utilization ratio was 8,300 SCF/STB and the net CO2 utilization ratio was 3,100 SCF/STB, which indicated considerably lower CO2 requirements per barrel of incremental oil than those for the continuous CO2 injection case. Although the CO2 utilization ratios indicated a higher EOR efficiency of WAG-CO2, the RCs of the two methods were similar: 33% and 36% for the continuous CO2 and WAG-CO2 injection cases, respectively. This amounted to 83% more CO2 being sequestered for the continuous CO2 injection compared to WAG-CO2. Adding 2,000 ppm H2S impurity and 10 – 20 % CH4 to the injected CO2 mixture caused little effect on the efficiency of CO2-EOR and sequestration. Therefore, recycling produced CO2, contaminated by the reservoir gases, does not impair the EOR and sequestration objectives of CO2 injection into the light-oil, carbonate reservoir considered in this study. The results of this work corroborate the conflicting effectiveness of continuous-CO2 and WAG-CO2 methods in meeting the EOR and CO2 sequestration objectives. Operational restrictions, such as to accommodate continuous injection of CO2, the WAG-CO2 method requires at least two parallel patterns with alternating CO2 time periods, further exacerbate the optimization considerations.

The mechanisms of carbon dioxide flooding at pressures below the minimum miscibility pressure (MMP) were studied using a numerical model of a slim tube to determine a means of increasing the efficiency of such floods. Results of these studies indicate that, in multiple contact flooding (MCF), the gas phase at the liquid-gas front approaches a constant composition denoting a bank of solvent approaching conditions of miscibility, but not achieving it because of the quantity of methane, nitrogen, and other light gases that overwhelms it. The ethane plus components (C2+) composed approximately five percent of the reservoir gas phase. This constant compositional gas phase formed early in the flood and persisted throughout the flood until eventual gas breakthrough. A simulated low-temperature flash of the reservoir gas phase produced a solvent that contained more than 75 percent ethane and propane. Slugs of this solvent were used to produce miscible displacements with CO2 gas at pressures 40 percent below the MMP. These findings were confirmed in further studies using fluids from several other reservoirs.

Low primary recovery percentages (usually 10% or less) in unconventional reservoirs, such as the Three Forks Formation in the Williston Basin, mean that potentially billions of barrels of oil are left behind in the reservoirs. Gas enhanced oil recovery (EOR) pilot studies that were performed in the Three Forks Formation suggest that cyclic gas injection is a viable approach for enhancing oil recovery in unconventional reservoirs. Due to limited knowledge of the cyclic gas injection mechanisms and the interaction between the injected gases and Three Forks reservoir fluids, additional investigations are mandatory despite of the pilot studies and the published findings in the Three Forks Formation. This paper conducts a series of experiments on core samples from the Upper Three Forks (UTF) and Middle Three Forks (MTF) under different constraints. The parameters that affect the performance of the cyclic gas injection during the Huff and Puff (HnP) technique are analyzed, including soaking time in miscible and immiscible conditions, injection pressure, vapor-supercritical states, density change, gas type (CO2), Ethane (C2H6), Propane (C3H8), within these formations (Upper Three Forks, and Middle Three Forks), and the microscopic pore distribution effect on hydrocarbon transport for the production using the Nuclear Magnetic Resonance (NMR) approach. The findings indicated that the soaking time improved recovery factors significantly below and close to the CO2's minimum miscibility pressure (MMP), but its effectiveness decreased for the injection pressures over the MMP. At low injection pressures, propane (C3H8) was shown to be the most effective gas, followed by ethane (C2H6) and then carbon dioxide (CO2). Due to a significant shift in density, ethane (C2H6) and carbon dioxide (CO2) both performed better during increasing pressure. In contrast, increasing the injection pressure of propane (C3H8) above 6.89 MPa did not affect the performance since the density of propane at 6.89 MPa caused a negligible density change. Additionally, the NMR measurements conducted at the UTF indicated that in the first cycle of HnP, oil extraction was dominated by micropores, mesopores, and less by small pores. Increasing the HnP cycles, micropores will be the dominant source of oil production. However, in MTF, the mesopores were the dominant pore distributions, followed by small pores, which affect the micro oil mobilization during the HnP process. This work provides a new insight into understanding the mechanisms of gas HnP enhanced recovery in the Three Forks formation, which is of great significance for the efficient hydrocarbon exploitation and greenhouse gas utilization in the Three Forks.

Carbon dioxide in enhanced oil recovery