In-situ combustion is the oldest thermal recovery technique. It has been used for more than nine decades with many economically successful projects. In-situ combustion is regarded as a high-risk process by many, primarily because of the many failures of early field tests. Most of those failures came from the application of a good process to the wrong reservoirs or the poorest prospects. The objective of this page is to describe the potential of in-situ combustion as an economically viable oil recovery technique for a variety of reservoirs.
Brooks, David (Shell Intl. E&P Co.) | De Zwart, Albert Hendrik (Shell Intl. E&P Co.) | Bychkov, Andrey (Shell) | Azri, Nasser (Shell International EP) | Hern, Carolinne (Shell) | Al Ajmi, Widad (Petroleum Development Oman) | Mukmin, Mukmin (Petroleum Development Oman)
The primary recovery of a medium-heavy oil reservoir with a strong bottom aquifer is generally poor. The introduction of horizontal wells that are drilled at the top of the oil column has improved the oil recovery. However, even horizontal wells suffer from fast water breakthrough that leads to oil production at a high water cut. Given the low primary recoveries, such fields are attractive EOR targets.
In situ combustion (ISC) is a displacement process generally applied to medium-heavy oil reservoirs in order to increase oil production by reducing the oil viscosity. In thick reservoirs (thicker than 10 meters), oil recovery could be severely challenged by gravity override of the injected gasses. In reservoirs without active aquifers, a significant part of the incremental oil is produced by gravity drainage after breakthrough.
We propose an ISC strategy where an infill producer is drilled close to the oil-water contact so that a significant amount of heat can be rapidly deployed in the middle and upper sections of the reservoir. Subsequently, the aquifer is used to sweep the warm oil through the heated zone towards the producers. The ISC process is compared with steam injection that also employs an additional infill producer. ISC and steam injection are used to deploy heat in the reservoir.
Numerical simulations show that the oil is produced at much lower water cut compared to the cold case (50-60 % versus 95%). Simulated oil recoveries increase significantly for both ISC and steam injection. A detailed comparison of these two processes is presented in this paper.
Mass grading of a hillside residential project within the Brea-Olinda oil field in southern California provided an opportunity to mitigate the naturally occurring oil and gas seeps at the site during the grading process. Multiple strategies, such as subsurface collection and venting systems, were used to reduce the potential for oil seeps to form in graded lots, and to prevent hazardous accumulations of methane gas in homes. Monitoring of the systems will be conducted by an environmental consultant in accordance with maintenance manuals prepared for the project.
This paper addresses the reservoir engineering aspects of air injection as an enhanced oil recovery technique for light-oil reservoirs. In its most successful form, the process has been applied in deep, carbonate reservoirs. The development of this process in conjunction with an application of the in-situ combustion (ISC) process to light-oil reservoirs, as well as the main mechanisms pertaining to ISC and to gas miscible flooding, are analyzed. It is seen that various air-injection processes (AIP's) can be classified, depending on their spontaneous ignition potential and gas miscibility at reservoir conditions, into four different processes. Based on an in-depth literature review, the best reservoir conditions for application of each of these four processes are derived. The main differences in operational aspects (pollution, corrosion, safety) for these processes are also discussed. Design considerations for pilot testing of the technique are presented. The crucial point is location of the pilot on the structure, which is also a key element in its proper evaluation, and its subsequent development to a commercial-size operation. Finally, recommendations on laboratory work in support of design and evaluation of a field pilot are also presented.
AIP's comprise those oil recovery processes that occur naturally when air is injected in an oil reservoir. The ISC process is one variation of air injection. Although the focus of this paper is not on ISC, the experience gained from ISC is used whenever relevant.
ISC is an AIP, but the reverse is not true; some AIP's cannot be considered as ISC processes at all. Usually, the application of the ISC process is associated with the existence of a high peak temperature (350 to 600°C) or an ISC front that travels from injection to production wells. On the other hand, the application of an AIP does not necessarily assume the existence of a high peak temperature. In other words, application of ISC sometimes requires an ignition operation to initiate it (create the heat wave), while the application of an AIP does not.
The ISC does not appear feasible in low-porosity matrix reservoirs; the porosity requirement is directly related to heat losses within the matrix. However, if the intent of air injection is merely pressure maintenance, the air injection should still be feasible, either as a miscible or an immiscible gas displacement process. Because the composition of air/flue gases, or a mixture of nitrogen with hydrocarbons in the vapor phase, is closer (from the miscibility point of view) to that of nitrogen, the miscibility of nitrogen can be a starting point in analyzing the feasibility of AIP's; this is illustrated in this paper. Generally, if the miscibility with nitrogen cannot be achieved, only an immiscible gas displacement needs to be evaluated.
This paper analyzes the main reservoir engineering aspects of air-injection application through a new classification based on the main mechanisms of ISC and miscible flooding, as well as in light of limited experience gained from airflooding light and very light-oil reservoirs.
Air-Injection-Based Oil-Recovery Processes
Air-injection-based oil-recovery processes were evaluated based on the screening criteria for improved-oil-recovery (IOR) processes used in the software program PRIze™, a package that evaluates the IOR potential of oil reservoirs.1 Basically, the screening criteria for application of ISC, gas miscible flooding, and immiscible gasflooding were used.
When air is injected into an oil reservoir, two simultaneous phenomena occur: displacement of oil and oxidation of oil. According to the efficiency of displacement and the intensity of oxidation, four main types of processes can occur.
Immiscible Airflooding (IAF) with High Temperature Oxidation (HTO)
IAF with Low Temperature Oxidation (LTO)
Miscible Airflooding (MAF) with HTO
MAF with LTO
The last two processes are commonly known as high pressure air injection (HPAI) processes.
Depending on the intensity of oxidation, either the LTO or the HTO reactions can dominate development of the process. Actually, when HTO takes place in immiscible airflooding, the classic ISC process is obtained, while if only LTO takes place, the process is called LTO-IAF (LTO combined with IAF).
The LTO-IAF was unintentionally obtained while attempting ISC, either when the ignition operation was not successful or when it was successful, but the ISC front did not sustain itself. Therefore, this kind of process has been applied only for relatively viscous oils. So far, the LTO-IAF process has not proved to be an effective IOR process (as compared to the ISC process). As a matter of fact, it seems to be the least efficient one among the four possible combinations.
Stoichiometrically, the volume of gases produced during an HTO process is roughly the same as that of air injected; hence, the oxidation reactions do not significantly impact pressure maintenance. For an LTO process, a part of oxygen is consumed without releasing carbon oxides, leading to a shrinkage of the injected gas volume. Consequently, benefits of pressurization are somewhat less for this process, and some overinjection may be considered.
Air injection can be used in both horizontal and vertical flooding. In a vertical flood, air is injected at the top of the structure (which may be a reef), and oil is produced from lower intervals, taking full advantage of gravity. This way, the volumetric sweep efficiency and displacement efficiency are aided by natural forces and are usually extremely efficient. In the hydrocarbon miscible flood, field experience has indicated incremental oil recovery using a vertical flood to be of the order of 30% original oil in place (OOIP), whereas for horizontal floods, the incremental oil recovery is typically 10% OOIP. A similar difference in the magnitude of oil recovery is expected for the application of air injection in these two modes. A typical horizontal immiscible gas injection can increase the ultimate oil recovery by up to 5 to 6% OOIP. For a vertical immiscible flood, this increment is expected to be much higher. In general, the IAF is expected to increase the ultimate oil recovery by at least as much as that obtained during immiscible flooding with such gases as nitrogen, flue gas, or hydrocarbon gas.
For cases where a combination of extensive fracturing and unfavorable mobility ratio between air and oil causes severe channeling, horizontal gasflooding may not be a viable option. However, gas injection at the top of the reservoir with velocities of displacement lower than the critical velocity may still be feasible.
It is expected that the oxygen contained in the injected air will not appear at the production well; rather, it will be consumed by reacting with the oil (oxygen uptake). Following gas breakthrough, the produced gas will consist mainly of nitrogen and hydrocarbon gases. This is true for most oil reservoirs, even in cases where air injection is accompanied by LTO, so long as heterogeneity is not too high.
A comprehensive geologic and engineering study of the A-10 North pool of theSansinena oil field was conducted to improve recovery of the pool of theSansinena oil field was conducted to improve recovery of the remainingreserves. Two reservoir facies were identified that exhibited distinctivelydifferent reservoir properties and oil production histories. The Channel Coresands have produced 31.7% of original stock-tank oil in place (OSTOIP), arecoarse-grained and locally pebbly, contain very little place (OSTOIP), arecoarse-grained and locally pebbly, contain very little associated clayminerals, and have high permeability and good lateral continuity. The ChannelFlank sands, on the other hand, have produced only 12.3% of OSTOIP, arefine-grained and silty, contain abundant dispersed and laminar clay minerals,and have low permeability and poor lateral continuity. A closer look at theinternal arrangement of reservoir facies, coupled with historical productionperformance, explains the large differences in oil recovery from the ChannelCore and Channel Flank sands and provides a better understanding of theproducing mechanisms within these different facies. The sands with goodpotential for enhanced recovery are identified along with the remaininginfill-drilling and recompletion potentials to enhance primary production fromthis geologically complex reservoir.
The Sansinena oil field is located in Los Angeles County along the Whittierfault trend. The oil seepages along the Whittier fault zone, a major geologicfeature in this area, resulted in the discovery of the oldest oil fields insouthern California. The Puente Hills field, discovered in 1880, is the secondoldest oil field in California. This field is located approximately 3 miles [5km] east of the original Sansinena discovery of 1898. Full-scale development ofthe Sansinena field began in 1944, with the major producing zones being theupper Miocene Divs. A, C, and D sands. The Div. A (Zone A-10) reservoirs arethe second most productive in the Sansinena field and are geologically complex,with traps formed by structural and stratigraphic features. Cumulativeproduction from Zone A-10 pools to Nov. 1988 exceeded 13.3 million bbl [2.1 X10(6) m3], a 22.8% recovery of the OSTOIP. Of the 52 wells completed in theA-10 North pool, only seven are currently producing. Only minor gas- andwater-injection projects have been implemented to date, and the pool is in anadvanced state of primary depletion. The original geologic interpretationstreated the entire A-10 North pool as a single reservoir unit with minor faultcomplications. Facies variations were not considered. This current study,however, integrates modem geological concepts with historical production datato describe the remaining production data to describe the remainingrecompletion and infill-drilling potentials and to evaluate the feasibility ofenhanced recovery projects.
The Sansinena oil field is one of several major oil accumulations on thenortheast margin of the Los Angeles basin. Fig. 1 shows the location of theSansinena field, which is on trend with the Whittier and giant Brea-Olindafields. The Sansinena field consists of numerous stratigraphically andstructurally controlled sandstone reservoirs within the Upper Miocene Puenteand the Lower Pliocene Repetto formations (Fig. 2). Coarse-grainedsiliciclastic sediments form a sequence of coalesced submarine fans and feederchannel deposits that are interbedded and encased within thick, deepwatershales. The sands were derived from the highlands of the San Gabriel mountainsto the northeast and were transported across alluvial and shallow marineterrains to the shelf/slope break controlled by the Whittier fault uplift. Thestratigraphic thickness of the Upper Miocene/Lower Pliocene compositestratigraphic section exceeds 12,000 ft [3658 m] in this portion of the basin.The section is complicated by regional and local unconformities. These complexstratigraphic relationships appear to be controlled by syndepositional wrenchfault tectonism along the Whittier fault system. Continued fault deformationthrough the Pleistocene Age and even recent times has further obscured thesestratigraphic relationships. The entire section dips steeply toward the basincenter on the southwest flank of the Whittier fault trend. Thispostdepositional tilting was caused by rapid basin subsidence coupled withuplift along the Whittier fault. Oil accumulations in the Puente/Repetto sandsare trapped along the Whittier fault by a combination of fault truncation andupdip pinchout of submarine fan sands and by local anticlinal flexuring of thearea as a result of Whittier fault movement. Fig. 2 shows the stratigraphy andproducing zones in the eastern portion of the producing zones in the easternportion of the Sansinena field. The Miocene section of the Los Angeles basin issubdivided into a number of biostratigraphic divisions based on foraminiferalwork. Oil production from the Upper Miocene section in the Sansinena fieldcomes predominantly from Divs. A, C, and D sands. The A-10 North pool is thelargest and most productive pool in the eastern portion of the field.
The A-10 North Pool
The A-10 North pool is a northeast/ southwest-trending submarine channelcomplex with a variable internal fill of sands, silts, and shales. The channelis truncated updip (Fig. 3) by the frontal reverse fault of the Whittier faultsystem.
This paper is a sequel to the state-of-the art review of fireflood field projects recently made by the same author, and discusses largely topics an current combustion technology which were not covered in the previous review. In this paper, examples are given to show why combustion was chosen as a primary, secondary or tertiary recovery process. primary, secondary or tertiary recovery process. Laboratory experimentation and numerical modeling in support of field projects are discussed. Expanded discussions are given an comparison between dry and wet combustion, and pattern selections. A review of monitoring and coring programs is followed by discussions on new frontier areas which include: in-situ combustion of tar sands and coal, utilization of flue gas, and a sand control technique.
In-situ combustion, one of the most important enhanced oil recovery methods, has received increased attention in recent years. A state-of-the-art review of combustion field projects, recently made by Chu, covered the following topics: screening guides, reservoir performance predictions, project design, well completions, ignition methods project design, well completions, ignition methods and operational problems. This paper complements the previous review and discusses the following new topics: why combustion was chosen, laboratory experimentation and numerical modeling, monitoring and coring programs, and new frontier areas. Expanded discussions are also given on dry vs. wet combustion and pattern selection, which topics have already been touched upon in the previous paper. paper. WHY IN-SITU COMBUSTION
Most of the in-situ combustion field projects have been undertaken for secondary recovery purposes. However, in some cases, in-situ combustion was also used as a tertiary recovery process, and as a primary recovery process. In the following, examples primary recovery process. In the following, examples will be given to illustrate why in-situ combustion was used for recovering oil in various depletion stages of a reservoir.
1. In-Situ Combustion as a Primary Recovery Process
When Mobil started the combustion project in the Moco Zone Reservoir, Midway Sunset Field, California, the cumulative oil production by primary recovery means amounted to only 0.4% of primary recovery means amounted to only 0.4% of the original oil in place. The reservoir pressure was only slightly below the initial pressure of the virgin reservoir. It was expected that, without fluid injection of sane kind, there would be a rapid decline in the oil production rates. Reservoir properties appeared to be favorable for in-situ combustion. Predictions of performance under cambustion were made, based on the results of the South Belridge thermal recovery experiment and on prior supporting work. Evaluation of several modes of operation indicated that the most desirable economics would be obtained if the combustion process was applied immediately. Accordingly, process was applied immediately. Accordingly, combustion operations were undertaken as soon as practicable after the reservoir was returned to practicable after the reservoir was returned to production. production. Although examples are few in which in-situ combustion served as a primary recovery process, such a possibility should be examined and its economics evaluated. In commenting on the timing of initiating a combustion project for secondary recovery purposes, Poettmann stated that, the sooner the application of the combustion process to a reservoir, the better. Fireflooding a virgin reservoir is just one step further in the right direction.
This paper embodies a state-of-the-art review of fireflood field projects. On the basis of reservoir data on 25 selected successful fireflood projects and nine aborted projects, a new screening guide has been developed for firefloods. A new regression equation also was developed that will allow prediction of air requirements with known reservoir characteristics. Methods of predicting other performance variables such as fuel content, sweep efficiency, oil recovery and air/oil ratio are discussed. Industry experience on project design is reviewed in regard to the choice between dry and wet combustion and determination of pattern type and size, air injection rate, water/oil ratio and completion intervals. Special well completions needed for fireflood are reviewed as well as the various ignition methods. Operational problems plaguing the fireflood projects, which include poor injectivity or productivity, corrosion, erosion, emulsion, and explosion hazards, are discussed along with their remedies.
The in-situ combustion process, ever since its inception in the mid-1930's, has proved a significant method for recovering oil, especially heavy oil.
Comprehensive reviews of fireflood field projects have been given by Farouq Ali1 and Chu,2among others. This review covers the following topics: screening guides, reservoir performance predictions, project design, well completions for injectors and producers, ignition methods, and operational problems and their remedies.
In dealing with oil prospects, the first step is to find out whether the field in question can be produced by certain recovery methods. Screening guides are useful for this purpose.
Screening guides for in-situ combustion processes have been proposed by various authors, including Poettman,3 Geffen,4 Lewin and Assocs.,5 Chu,2 and Iyoho.6 The screening guides proposed by these authors are listed in Table 1. In recent years, the price structure of the crude oil has changed tremendously. A fireflood project that previously was considered uneconomical could become economically feasible if conducted today. Moreover, more information recently has become available for various new or continuing fireflood field projects. In view of this, a new set of screening guides has been developed and is listed on the last entry in Table 1.
The approach for developing this new screening guide differs from the approaches used by Chu in 1977 in two respects. First, the capability of producing oil rather than the economical feasibility of the projects is used as the criterion for acceptance or rejection of a prospect. Second, both the confidence-limit and regression-analysis approaches used previously presuppose that the frequency distributions of the various reservoir characteristics conform to a normal distribution. In the present approach, the frequency distributions are used as they appear, regardless of whether they are normal, log-normal or otherwise.
As a part of the data base for the development of the new screening guide, 25 successful fireflood projects were selected. All these projects have exhibited oil production rates of at least 100 B/D for an extended period of time, mostly for more than a year. These are considered positive projects. As another part of the data base, nine aborted projects have been included as negative projects. The lists of the positive and negative projects are not meant to be exhaustive. However, these 34 were selected from all the projects that appeared in the literature available to me, with sufficiently detailed description of reservoir characteristics, process variables, and fireflood performance. The reservoir characteristics of these two project types are listed in Table 2.
More than 40 domestic and foreign firefloods were reviewed. Regression equations were developed for predicting production performance of firefloods. Two screening guides were established for selecting reservoirs for the application of the fireflood process.
Enhanced recovery methods that show most promise for commercial application include hydrocarbon miscible flood, carbon dioxide miscible flood, micellar flood, and thermal recovery methods. Since the first three processes are applicable only to low-viscosity crudes (10 cp or less), the recovery of medium- to high-viscosity crudes depends solely on thermal methods - steam stimulation, steamflood, and fireflood.
A recent domestic survey showed that, of the three thermal methods, steam stimulation has been practiced exclusively in California and steamflood has been used predominantly there. Fireflood projects, on the other hand, cover a much wider area including California, Texas, Louisiana, Oklahoma, Arkansas, Illinois, Mississippi, Nebraska, and Wyoming. This wide geographical distribution of fields amenable to firefloods clearly indicates that firefloods are applicable to a broader range of reservoir and crude properties than steam stimulation and steamflood.
More than 70 firefloods have been completed or are in progress in the U.S. Firefloods also have been practiced in other countries such as Venezuela, U.S.S.R., Romania, and Japan. Varying information is available on more than 40 domestic and foreign firefloods. The purpose of this paper is to review these firefloods and thereby develop criteria for screening prospects for the application of the fireflood process.
This paper first gives a general description of fireflood and its variations and discusses variables characterizing the performance of firefloods. Screening guides for fireflood prospects are developed by two different statistical approaches - the confidence-limits approach and the regression-analysis approach. The choice between fireflood and steamflood is also discussed.
Fireflood and Its Variations
The most commonly used form of the fireflood process is dry forward combustion. The process is called dry combustion because no water is injected along with air. The combustion is forward because ignition occurs near the injection well and the burning front moves forward from the injection well to the production well. The advantage of this process is that an undesirable fraction of the crude is burned in the form of coke, leaving clean sand in the region behind the burning front. However, it has two limitations. First, the produced oil has to pass through a cold region of the reservoir. If the oil is highly viscous, liquid blocking will occur, which may terminate the process. Second, heat stored in the burned-out region is not utilized efficiently because injected air is not sufficiently effective to carry the heat forward.
Reverse combustion is a variation of the fireflood that remedies the first limitation of the conventional fireflood. Ignition occurs near the production well and the burning front moves countercurrent to the flow of the injected air. Since oil moves through a hot region toward the produced well, there is no upper limit to the viscosity of the reservoir crude. Reverse combustion is not as efficient as forward combustion because a desirable fraction of the oil is burned as the fuel, and the undesirable fraction remains in the region behind the combustion front.
This paper was prepared for the Improved Oil Recovery Symposium of the Society of Petroleum Engineers of AIME, to be held in Tulsa, Okla., April 22-24, 1974. Permission to copy is restricted to an abstract of not more than 300 words. Illustrations may not be copied. The abstract should contain conspicuous acknowledgment of where and by whom the paper is presented. Publication elsewhere after publication in the JOURNAL paper is presented. Publication elsewhere after publication in the JOURNAL OF PETROLEUM TECHNOLOGY or the SOCIETY OF PETROLEUM ENGINEERS JOURNAL is usually granted upon requested to the Editor of the appropriate journal, provided agreement to give proper credit is made. provided agreement to give proper credit is made. Discussion of this paper is invited. Three copies of any discussion should be sent to the Society of Petroleum Engineers office. Such discussions may be presented at the above meeting and, with the paper, may be considered for publication in one of the two SPE magazines.
A combustion drive project has been underway for 1-1/2 years at Brea-Olinda, Orange County, California. Air injection was commenced in one fault block early in 1972, and in a second adjacent block early in 1973. One compressor supplies 5 MM scf/d of air to the two injection wells at a maximum pressure of 1200 psig.
About 22 production wells are in communication with the two injection wells as evidenced by increased hitrogen in their produced gas. So far, the large increases in oil production have occurred in wells downstructure from the injection well in the first fault block. The original injection well for the second block has been replaced by another well nearer the top of the structure in an attempt to improve oil recovery from that block.
The reservoir is First Miocene sand dipping 45 degrees at depths of 3000-4000 feet. Oil gravity is 22 degrees API. Oxygen utilization is 100%. Spontaneous ignition occurred in about one week.
Principal Conclusion: Principal Conclusion: We believe that burning from the top down in a steeply dipping reservoir maximizes oil recovery.
Fireflooding has been extensively pilot tested by the industry during the past two decades. Sufficient information is now available to make it possible to select reservoirs from which fireflooding will recover economic amounts of additional oil. As with other secondary and tertiary recovery methods, technological success (stimulated oil recovery) may occur with or without economic success (profit).
The desire to recover more of the oil in place from fields before they are abandoned will favor the use of fireflooding where applicable. Increases in the price of crude oil, supporting larger production costs, will make fireflooding appear promising in some of the fields for which it has been rejected in the past as insufficiently rewarding. The information in this paper is presented in the spirit that these increased expectations for fireflooding are real and enduring.
HISTORY OF DEVELOPMENT OF BREA OLINDA
The discovery well at Brea-olinda was completed about 1884. The field is now credited with 2,300 proved acres (FIGURE 1). Oil production has been about 350,000,000 barrels from more than 1,250 wells.