SPE Member now with Western Atlas Software.
This paper discusses the outlook for Large Grid Block (LGB) models as predictive tools. The interactions between viscous fingering and heterogeneity are considered for combinations of secondary miscible, immiscible, and continuous and Water-Alternate-Gas injection tertiary miscible displacements. The approach uses detailed simulations in very fine-scale heterogeneous grids to calibrate simpler analytic and coarse grid models. It concludes that although much progress can be made in calibrating continuous solvent injection LGB models, these same models prove to be less reliable for more complicated situations, such as Water-Alternate-Gas injection. In particular, for continuous solvent injection: (1) The mixing parameter model can be calibrated with a potentially measurable indicator of heterogeneity to match secondary miscible recovery and injectivity. (2) Integrations between heterogeneity and effective relative permeability can be calibrated with this same indicator of heterogeneity. (3) These calibrations for secondary miscible and purely immiscible displacements can be combined and reasonably applied to tertiary miscible displacements. However, severe complications arise when considering Water-Alternate-Gas injection, particularly in heterogeneous systems which are not at their residual saturation. These complications result because: (1) The mixing parameter must vary with WAG ratio and cycle size. (2) The effective relative permeabilities depend on the nature of the phase being injected. In these situations, it is not possible to simultaneously match breakthrough times, recoveries and injectivies using a single set of LGB parameters.
Miscible flooding simulations often use models which do not adequately represent the physics of what is actually occurring on the scale of a simulation grid block. Explicit simulation of the effects of viscous fingering, reservoir heterogeneity and/or the development of miscibility generally require levels of detail which cannot currently be attained in field-scale reservoir simulation models. Acknowledging such limitations, Large Grid Block (LGB) models have commonly been used to simulate field-scale miscible and developed miscible displacements. With these models, the effects of sub-grid block scale phenomena are incorporated with effective property modifications and/or pseudo-physics type corrections. Examples of these are the mixing parameter (to account for viscous fingering) , pseudo-relative permeabilities (to account for multiphase flow and heterogeneity) , and miscible residual representations of phase behavior. LGB models can be very useful once they have been calibrated. However, they may be unreliable if used for situations where calibrations have not been established.
Viscous fingering has been one of the most widely analyzed aspects of LGB modeling. Many studies have been performed to understand the mechanisms behind initiation and propagation of fingering .
Most laboratory work in MEOR, both in screening and in development, is performed under in situ temperatures and in brines similar to those of the target reservoir. However, the effect of In situ pore pressures in such work is normally ignored. Much of our understanding of the effects of hydrostatic pressure on microorganisms comes from studies of microbial communities at the ocean floor. Since nearly 90% of the ocean floor has pressures ranging from 10 MPa to over 100 MPa, there is a paucity of data concerning effects on microorganisms at the lower pressures found in many oil reservoirs.
In our studies of the indigenous cells from injection brine at the North Burbank Unit, we have found that pressures ranging from 0.1 MPa to 8.8 MPa can have a significant effect on pH changes, substrate utilization, end product formation and permeability reduction. In general, hydrogen ion concentrations and carbohydrate utilization rates increased with increasing pressures. Methanol was found to be a significant end-product in the presence of hydrostatic pressure but found to be absent under atmospheric conditions. The fact that some of the observed effects could be demonstrated by simply reducing the gas headspace suggests that gas solubility could be an important factor. Gas formation in a core was estimated to account for as much as 50% of the permeability decrease observed in cores run under atmospheric pressure. Therefore the changes observed are of such magnitude as to alter the MEOR process in the reservoir from that indicated by laboratory studies not done under pressure.
Our results strongly suggest that all MEOR studies pertaining to a known reservoir should be evaluated under the in situ conditions of the reservoir, including pressure.
A number of mechanisms have been proposed for the recovery of residual oil through the application of microbially enhanced oil recovery (MEOR).
This paper describes the procedures of a simulation study to design an optimum CO2 flood process for the Texaco operated Roberts Unit located in the Wasson Field, Yoakum County, Texas. The designing process includes geological study to define flow units, CO2 target area selection, model area determination, fluid characterization, model calibration by history matching waterflood performance, compositional simulation forecasting for tertiary recovery, and economic analyses.
An equation-of-state (EOS) compositional simulator was used for both waterflood history matching and CO2 flood predictions.
Phase behavior of the CO2/Hydrocarbon system was calibrated by matching laboratory PVT test results. Tertiary forecasts were run for various continuous, water-alternating-gas (WAG) and hybrid modes of operation. Simulation results were scaled to represent the entire Roberts Unit CO2 target area. An economic indicator was then used to determine the optimum method of CO2 injection.
This paper summarizes the results Of field application of EOR techniques in the former Soviet Union. 175,000 barrels of incremental oil per day were produced in 1991 due to their implementation in different geological and environmental conditions. 55% of total production or 96,500 BOPD were produced with the application of chemical methods, 64,500 BPD or 37% were produced due to thermal and the rest - due to gas injection methods. The hydrodynamic improved oil recovery methods based on waterflooding technique and reservoir management brought another 600,000 BPD additional oil. The future of enhanced oil production depends on wide application of both conventional and nonconventional advanced methods tested in laboratories and in field sites and provided for their high efficiency.
The outlook of wide application of EOR and IOR techniques by the years 2000 and 2010 at different economical conditions is presented.
Oil production in the former Soviet Union after many years of sharp increase started to decline. In 1991 our country produced 10.4 MM BOPD, 17% less than in 1987 and 1988 when the maximum oil production level of 12.5 MM BPD was reached.
EOR and IOR techniques are considered to be an important tool to stabilize domestic oil production or reduce the rate of its decline.
All known conventional EOR methods widely apply in different geological and environmental conditions.
This paper describes an instrument to measure. the interfacial tension (IFI) of aqueous surfactant solutions and crude oil. The method involves injection of a drop of fluid (such as crude oil) into a second immiscible phase to determine the IFT between the two phases. The instrument is composed of an AT-class computer, optical cell, illumination, video camera and lens, video frame digitizer board, monitor, and software. The camera displays an image of the pendant drop on the monitor, which is then processed by the frame digitizer board and non-proprietary software to determine the IFT. Several binary and ternary phase systems were taken from the literature and used to measure the precision and accuracy of the instrument in determining IFTs.
The instrument has been used to determine the IFTs for several bacterial supernatants and unfractionated acid precipitates of microbial cultures containing biosurfactants against medium to heavy crude oils.
These experiments demonstrate that the use of automated video imaging of pendant drops is a simple and fast method to reliably determine interfacial tension between two immiscible liquid phases, or between a gas and a liquid phase.
It has been estimated that approximately 60% of all oil discovered will remain trapped after current oil recovery technologies have been employed. Waterflood recovery is the most economic and widely applied secondary recovery technique. However, waterflooding does not remove all the remaining oil in the reservoir due to incomplete reservoir flooding and to Viscous and capillary forces.
The unproduced oil remaining after waterflooding includes immobile oil left in place. The immobile oil is trapped in pore structures by viscous and capillary forces, and cannot be removed by waterflooding. Ultra-low IFTs (about 10 mN/m) allow the oil trapped by capillary forces to become mobile. Lowering the IFT by use of surfactants decreases the pressure required to force a nonwetting phase (oil) through small capillaries and pore constrictions. Micellar (or surfactant) flooding is of particular interest at the Idaho National Engineering Laboratory (INEL) due to research on microbial enhanced oil recovery (MEOR) being conducted there.
Measurement of IFTs has been accomplished by a variety of methods, including pendant drop, sessile drop, and spinning drop methods. These methods place a drop of oil into a surfactant solution, and then measure the shape of the drop formed. Current developments in personal computers coupled with the ability to digitize video images enables the researcher to automate the process, and provides an improved process to determine the IFT. Video digital methods have been applied to the measurement of the IFTs of bacterial biosurfactants. A simple, fast, automated system for measuring interfacial tensions using the pendant drop method has been developed at the INEL.
THEORY OF IFT DETERMINATION BY PENDANT DROPS
Capillary forces between crude oil and formation water determine the pressures needed to displace the oil from formation pores, which is given by the Young-Laplace equation.
Thermal oil recovery was first started in 1966, in Trintopec's operations, with a small cyclic pilot project in the Palo Seco field. Twenty-five (25) years of thermal recovery, comprising cyclic and flood operations, have witnessed vigorous growth and dynamic expansion to the extent that, by 1991, the thermal recovery statistics are as follows :
(a) Steamflood operations exist in all the Company's major land fields, viz., Palo Seco, Central Los Bajos, Guapo, North Fyzabad and Apex-Quarry/Coora/Quarry.
(b) Total production from thermal recovery averages 9100 bopd, representing 55% of Trintopec's current land production.
(c) A total of approximately 40000 bspd is being supplied by 23 steam generators to more than 150 steam injectors.
This paper presents highlights of Trintopec's experiences in the design, implementation and operation of thermal oil recovery schemes. New concepts, innovations, modelling and monitoring techniques over the past twenty-five (25) years are outlined, and, in addition, projections for the future are indicated.
Of the enhanced oil recovery techniques available to the industry, viz., thermal (steam and in-situ combustion), miscible, chemical and polymer floods, the single method that has attained most widespread acceptance has been the use of steam, both in cyclic and flood-type operations.
Not surprisingly, this method has manifested itself as the major thrust of EOR operations at Trintopec. The primary factors that motivate its use are as follows
(i) Cost effectiveness, with water being cheap and readily available,
(ii) Abundance of heavy oil reserves in Trintopec's leases,
(iii) Projected recovery of lighter oil, by distillation,
(iv) Proven success of method, both internationally and at Trintopec.
Significantly, and expectedly, the growth of associated technology in the Company's thermal operations has paralleled the increased production rates from thermal operations. Although the current performance of the Company's thermal schemes is such that 55% of land production is thereby accrued, and prospects for further enhancement of thermal production are encouraging, it would be erroneous to suggest that the path was not fraught with difficulties and operational problems. Nevertheless, through a vigorous and dynamic growth and expansion programme, the method employed in the thermal recovery schemes, although not fully state of the art, have been refined to the point of acceptable confidence levels. Trintopec then, continues to maintain its pioneering status in thermal recovery in Trinidad and Tobago.
The technical performances of horizontal and vertical wells were examined for a tertiary, carbon dioxide miscible flood in a 240-acre (97 ha) area of and west Texas field using a black-oil, pseudo-miscible simulator. Although and 4D acre (16 ha), inverted fivespot pattern was used initially for both vertical and horizontal wells, the spacing was increased to 60 acres (32 ha) for the horizontal wells to maintain the miscibility pressure in the reservoir. Horizontal injection and production wells, 1,320 feet (402 m) in length, completed at the bottom of the formation, recovered 14 to 22 percent more oil than vertical wells. Economic analyses for the horizontal injectors and producers were compared to economic analyses conducted for vertical injectors and producers. Project economics were significantly affected by the capital expense for drilling new wells. The vertical wells provided the better rate of return if no now drilling or only one now well for every six patterns was required. If horizontal wells could be drilled from existing vertical wells or from the surface at 1.5 times the cost of vertical wells, the rate of return was comparable or better than vertical wells requiring one new well for every three patterns. Horizontal wells drilled from the surface at twice the cost of vertical wells provided the lowest rate of return.
To assess the economic viability of a proposed polymer augmented waterflood and to design the treatment, it is necessary to determine the polymer retention level in the rock matrix.
Static adsorption experiments on rock samples are unsuitable, because the crushing process may expose new sites, and mechanical entrapment of molecules in pores is not included in the measurements. Dynamic core-flooding measurements, at deep reservoir flow rates, are therefore required. However, reservoir core plugs may be contaminated by drilling muds, particularly at the ends, increasing the adsorption capacity. In this case, detailed mass balance calculations on the polymer effluent profiles may significantly overestimate retention levels in the formation.
To overcome these problems, a pyrolysis/beta scintillation counting technique has been developed. The core plug is flooded with 14 C-labelled polymer. It is then disassembled and samples from the interior are crushed and pyrolysed, to convert the carbon to CO2. The 4C content of the gas is measured via beta-scintillation counting, giving the polymer retention level in the plug's interior, where face filtration and drilling mud contamination are minimised.
This paper describes the validation of the technique. The reproducibility and efficiency of the recovery are shown to be independent of the crushed particle size, over the range 50-1000 um, the polymer retention level, over the range 1-120 ug/g, and the presence of oil at residual saturation. Finally, the retention levels deduced from a detailed mass balance calculation and subsequent pyrolysis/beta scintillation counting of samples from an uncontaminated core are compared, and found to be consistent.
This technique is recommended for the determination of polymer retention levels in the rock matrix, under as near as possible reservoir conditions, assuming that a small amount of C-labelled polymer is available.
To design a polymer project and assess its economic viability, it is necessary to estimate the polymer retention level in the native rock at deep reservoir flow rates. The retention may include chemical adsorption onto clays or rock surface adsorption sites and mechanical or Hydrodynamic entrapment of the molecules in pore throat constrictions and dead-end spaces.
Static adsorption experiments on crushed rock samples are unsuitable for estimating the retention level for two reasons. Firstly, dynamic effects, such as mechanical and hydrodynamic entrapment in pore spaces, are not included in the measurement Secondly, new adsorption sites may be exposed and activated in the crushing process, resulting in an overestimate of the adsorption capacity of the consolidated rock.
The effluent profiles from core-flooding experiments at appropriate flow rates can be analysed to obtain dynamic retention levels for the core sample, via a detailed mass balance calculation. However, the core sample may not be representative of the native rock in the formation.
It has been well documented that tertiary carbon dioxide WAG cycle injectivity during displacements above the minimum pressure for dynamic miscibility cannot be reliably predicted on the basis of viscosity ratios and waterflood injectivity performance alone. Approximate analytical models based upon gravity-free displacement with nondispersive plug flow of constant composition slugs in noncommunicating radial layers have been used to interpret injectivity during reservoir tests as a function of relative permeability, flow geometry, effective wellbore radius, and layering. We examine compositional simulation as an alternative for incorporating the additional effects of phase behavior, dispersive mixing, gravity, capillary forces. viscous instability, crossflow, and more realistic representations of reservoir heterogeneity in injectivity calculations for field test interpretation. We show how simple modifications can be made to an existing finite-difference equation-of-state simulator to allow for detailed modeling with these mechanisms of reservoir-scale cross-sections that incorporate geostatistical representations of reservoir heterogeneity and radial flow near injection and production wells. The compositional model is applied to simulate and interpret an injectivity test conducted in the Mabee Field in the San Andres Formation, Martin County, Texas. Injectivity during the field test was significantly greater than initial waterflood injectivity during all WAG cycles, increased during each CO2 cycle, and was greater than would be anticipated from other field tests in the San Andres formation or from available laboratory data. Compositional simulation of the field test indicates that the reservoir response during the first cycles of CO2 and brine injection, including the initial increasing CO2 injectivity trend and higher brine injectivity after the first CO2 cycle, are consistent with measured relative permeability, phase behavior, and reservoir characterization data, but that the observed injectivity increase during the second cycle Of CO2 injection cannot be attributed to either the near-wellbore condition of the reservoir at the start of the test or to the presence of thief zones that are statistically consistent with measured core data and well logs. Injectivity is a key variable in determining the economic incentive associated with a proposed CO2 project, and an important implication of this study is the validation of compositional simulation as a means for interpreting field tests and developing improved predictions of reservoir injectivity performance. Another important conclusion is that geostadstical techniques can be used successfully to characterize high heterogeneity in carbonate reservoirs for injectivity calculations.
The J.E. Mabee Field in the San Andres Formation, Martin County, Texas. is a candidate for tertiary recovery by multiple-contact miscible (MCM) displacement with carbon dioxide.
Two thermally enhanced oil recovery methods, cyclic steaming and steamflooding, have been investigated by means of numerical modeling for a massive, dipping, Midway Sunset Field reservoir. Comparative economics have been evaluated for the development alternatives.
The results of the investigation indicate that ultimate recovery efficiency is not sensitive to the method of steam delivery to the reservoir. Cyclic steaming methods recovered the same percentage of original oil in place as did steamflooding. The cyclic operations required several times longer than steamflooding to reach the same ultimate recovery efficiency.
On a steam utilization efficiency basis (oil to steam ratio), cyclic steaming with small steam slug volumes proved to be the most efficient recovery method. As the rate of steam injection increased, the steam utilization efficiency decreased, regardless of the manner in which steam was delivered to the reservoir.
Despite having the poorest steam utilization efficiency, steamflooding provided far superior economics as compared to conventional low or moderate slug volume cyclic steaming. This is directly attributable to steamflooding having the fastest rate of increase in average reservoir temperature and therefore the greatest rate of net recovery as compared to the alternatives investigated. Steamflooding exhibited a much quicker payout of development capital and a greater present value return per dollar invested.
While thermally enhanced oil recovery methods are an attractive method of extracting heavy oil, due to the large development capital costs and the significant expense in steam generation, it is important to look critically at development alternatives and to choose the method that is most attractive to the investor.
A cyclic steaming project is operated significantly different from that of a steamflood. Consequently, it requires a different facilities configuration in order to allow effective optimization during its operation. It is costly to install a facilities design tailored to one method, and then to retrofit for an alternative operating strategy. It is therefore essential to choose the more attractive manner in which steam is to be delivered to the reservoir and to install and operate the project in this manner.
Thermal EOR textbooks indicate that, in general, greater recovery efficiencies are possible with steamflooding than with cyclic steaming. It is generally written that while steamflooding can be expected to recover 50-60% of the original oil in place, cyclic steaming typically recovers only 10-25%. Cyclic steaming is viewed as a stimulation method with rapid response while there is a lag in the production response associated with steamflooding.
The Potter Sand in the Midway Sunset Field has been developed and operated with cyclic steaming as well as a steamflood by various operators. Both methods are viewed by their respective operators as technical and economic successes.