Appendix B

Supporting Materials for Oil Recovery ProjectionsFrom Application of Enhanced Recovery

ProcessesPage

TECHNOLOGICAL PROJECTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147Surfactant/Polymer Flooding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

State of the ArtTechnological Assessment . . . . . . . . . . . . . . . . . . . . . . . . . 147Oil Recovery Projections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150Composition and Costs of Injected Materials . . . . . . . . . . . . . . . . . . . . . . . . IS3Sensitivity Analyses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

Polymer Flooding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154State of the ArtTechnological Assessment . . . . . . . . . . . . . . . . . . . . . . . . . 154Oil Recovery Projections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156Sensitivity Analyses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156Effect of Polymer Flooding on Subsequent Application

of Surfactant/Polymer or Carbon Dioxide Miscible Processes . . . . . . . . . 157Steam Displacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

State of the ArtTechnological Assessment . . . . . . . . . . . . . . . . . . . . . . . . . 157Oil Recovery Projections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........158Steam Requirements and Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160Sensitivity Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

In Situ Combustion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162State of the ArtTechnological Assessment . . . . . . . . . . . . . . . . . . . . . . . . . 162Oil Recovery Projections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163Operating Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164Sensitivity Analyses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164

Carbon Dioxide Miscible . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165State of the ArtTechnological Assessment . . . . . . . . . . . . . . . . . . . . . . . . . 165Oil Recovery Projections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166Carbon Dioxide Costs.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168Results of Carbon Dioxide Cost Calculations . . . . . . . . . . . . . . . . . . . . . . . . 169Calculation Method and Details-Carbon Dioxide Costs . . . . . . . . . . . . . . 171Sensitivity Analyses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

ECONOMIC MODEL.... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175Structure of the Model.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175

Specific Economic Assumptions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176Economic Data-General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178Offshore Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190

Costs That Do Not Vary With Water Depth. . . . . . . . . . . . . . . . . . . . . . . . . 190Cost That Vary With Water Depth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191

143

144 . Appendix B

LIST OF TABLES

TableNumber Page

B-1 .

B-2.

B-3.B-4.B-5.B-6.

B-7.B-8.B-9,B-10.B-1 1.B-1 2.

B-1 3,

B-1 4.B-1 5.

B-1 6.B-1 7.B-1 8.B-1 9.B-20.B-21 .B-22.B-23.

B-24,

B-25.

B-26.

B-27.

B-28.

B-29.

B-30.

B-31 .

B-32.

B-33.

B-34.

B-35.

B-36.

B-37.

ERDA Cooperative Field-Demonstration Tests of EOR Using theSurfactant/Polymer Process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Summary of Surfactant Field Tests Being Conducted by Industry WithoutERDA Assistance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Development of a Five-Spot PatternSurfactant/Polymer Process . . . . . . . .Chemical Costs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Component Costs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Surfactant/Polymer Process-Ultimate Recovery, Summary of Computed

Results-Process and Economic Variations . . . . . . . . . . . . . . . . . . . . . . . . .Production Schedule for Polymer-Augmented Waterflood. . . . . . . . . . . . . . .Polymer-Augmented Waterflooding Ultimate Recovery, . . . . . . . . . . . .production Schedule for Steam Displacement Process . . . . . . . . . . . . . . . . . .Recovery Uncertainties Effecting Steam Displacement Results . . . . . . . .Effect of Uncertainties in Overall Recovery on Ultimate Production . . . . . . .Effect of Well Spacing on Ultimate Recovery of Oil Using the Steam

Displacement Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Advancing Technology Cases-Oil Displacement ModelWet

Combustion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .production ScheduleWet Combustion . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Effect of Changes in Compressor Operating Costs and Displacement

Efficiency in Ultimate Oil Recovery Using the In Situ Combustion ProcessCarbon Dioxide Injection Schedule. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .production Rate Schedule for Carbon Dioxide Miscible . . . . . . . . . . . . . . . . .Gas Injection Schedule-Offshore Carbon Dioxide Miscible . . . . . . . . . . . . .Oil Production Schedule-Offshore Carbon Dioxide Miscible. . . . . . . . . . . .Pipeline Capacity Versus Investment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Lateral Lines Associated With Pipeline Capacity . . . . . . . . . . . . . . . . . . . . . . .Total Costs per Mcf of C02. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . o Pipeline Capacity as a Function of Field Size . . . . . . . . . . . . . . . . . . . . . . . . . . .Estimated Recoveries for Advancing Technology--High-Process

Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Sensitivity of Ultimate Recovery to Carbon Dioxide Cost. . . . . . . . . . . . . . . .Sensitivity of Ultimate Recovery to Carbon Dioxide Cost. . . . . . . . . . . . . . . .Production Unit Size. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Schedule of Starting Dates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Timing of Reservoir Development. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Drilling and Completion Costs for production and Injection Wells . . . . . . . .Well, Lease, and Field Production Equipment Costs-Production Wells. . . .Costs of New Injection Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Well Workover and Conversion Costs for Production and Injection Wells,

Parts A and B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Basic Operating and Maintenance Costs for Production and Injection WellsIncremental Injection Operating and Maintenance Costs. . . . . . . . . . . . . . . .State and Local production Taxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .State Income Taxes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

148

149152153153

154156157160161161

162

163164

164167168168168172172173173

174175175177177178179180182

184186188190190

Appendix B . 145

B-38. Offshore Costs That Do Not Vary by Water Depth . . . . . . . . . . . . . . . . . . . . . 191B-39. Offshore Costs That Vary by Water Depth. . . . . . . . . . . . . . . . . . . . . . . . . . . . 191B-40. Drilling and Completion Cost Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192

LIST OF FIGURES

FigureNumber Page

B-1 . Historical Incremental production Therm Recovery-California ... , . . . . . . . . 159B-2. Pipeline Cost Versus Capacity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171B-3. Variable C02 Transportation Costs Versus Pipeline Capacity. . . . . . . . . . . . . 172B-4, Transportation of CO2 Break-Even Analysis . . . . . . . . . . . . . . . . . . . . . . . 173

Th i s append ix p resent s supp lementarymaterials which were used to prepare oil recov-ery projections and to compute the costs to pro-duce enhanced oil. It is organized into two sec-tions, the first describing the technologicalassumptions for each enhanced oil recovery(EOR) process. For each process the state of theart of the technology is assessed. Models used

to compute recoveries and production rates are

Appendix B . 147

presented in detail. Cost data which are specificto a process are documented. Results of calcula-tions not presented in the body of the report aregiven.

The second section describes the economicmodel used in the OTA study. Cost data whichare independent of the process are documentedin this section.

Technological ProjectionsSurfactant/Polymer Flooding

State of the ArtTechnological AssessmentThe surfactant/polymer process involves two

technologies. The first is the art of formulating achemical slug which can displace oil effectivelyover a wide range of crude oil compositions, for-mation water characteristics, and reservoir rockproperties. As used in this section the termchemical slug refers to all injected fluids whichcontain a surfactant mixed with hydrocarbons,alcohols, and other chemicals. Excluded from thisdefinition is alkaline flooding,1 a process in whichsurfactants are generated in situ by reaction ofcertain crude oils with caustic soda.

The second technology is the displacement ofthe injected chemical slug through the reservoir.This technology is governed by economic andgeologic constraints. The cost of the chemicalslug dictates use of small volumes in order tomake the process economically feasible. Thetechnology for displacement of the chemical slugthrough a reservoir relies on controlling the rela-tive rate of movement of the drive water to thechemical slug. Effective control (termed mobilitycontrol) through process design prevents ex-cessive dilution of the chemical slug. If mixedwith displaced oil or drive water, the chemicalslug would become ineffective as an oil-displac-ing agent. Control of the mobility of the chemicalslug or drive water is accomplished by alteringthe viscosities or resistance to flow of these fluidswhen they are formulated. z

NOTE: All references to footnotes in this appendix appearon page 193.

Research to find chemicals which displace oilfrom reservoir rocks has been conducted inGovernment, industry, and university laboratoriesfor the past 25 years. Research activity in theperiod from 1952 to about 1959 was based onthe injection of dilute solutions of surfactantwithout mobility control .-3 Activity peaked withthe advent of each new chemical formulation inthe laboratory and declined following disap-pointing field results. In some tests, surfactantswere injected into reservoirs with no observableresponse. in other tests, the response was sosmal l that the amount of incremental o i lrecovered was almost unmeasurable. The cost ofwhatever incremental oil was produced wasclearly uneconomic.

The period beginning in the late 1950s andextending into the present is characterized bymajor advances in formulation of the chemicalslug and control of slug movement through areservoir. Several laboratories developed for-mulations based on petroleum sulfonates whichmay displace as much as 95 percent of the oil insome portions of the reservoir which are sweptby the chemical slug.4,5 Addition of water-solublepolymer to drive water has led to mobility con-trol between the drive water and chemical slug.6

Field tests of the different processes have pro-duced mixed results. About 400,000 barrels of oilhave been produced from reservoirs which havebeen previously waterflooded to their economicI imit . 7,8,9 Oil from one test was considered~economic. All other oil was produced under con-ditions where operations were uneconomic.Offsetting these technically successful tests10 areseveral field tests which yielded considerably less

148 l Appendix B

incremental oil than anticipated. 11,12,13 The stateof technology is such that honest differences ofopinion exist concerning the reasons for disap-pointing field test results.14,15

The current ERDA program includes six large-scale, cooperative, field-demonstration tests. Thefields and locations are summarized in table B-1.The first five projects are in fields which havebeen intensively waterflooded. In these tests, theprincipal objectives are to demonstrate the effi-ciency and economics of recovery from a suc-cessfully depleted waterflood using the surfac-tant/polymer process. The Wilmington reservoircontains a viscous oil. An objective of this proj-ect is the development of a surfactant/polymers y s t e m w h i c h w i l l d i s p l a c e v i s c o u s o i leconomically.

Table B-1ERDA Cooperative Field-Demonstration Tests of EOR

Using the Surfactant/Polymer Process

Field Location

El Dorado. . . . . . . . . . . . . . . . . . . . . . . . . . KansasNorth Burbank. . . . . . . . . . . . . . . . . . . . . OklahomaBradford . . . . . . . . . . . . . . . . . . . . . . . . . . . PennsylvaniaBell Creek. . . . . . . . . . . . . . . . . . . . . . . . . . MontanaRobinson . . . . . . . . . . . . . . . . . . . . . . . . . . IllinoisWilmington . . . . . . . . . . . . . . . . . . . . . . . . California

Screening Criteria. The screening criteria intable 7 of the main text reflect estimates of tech-nological advances in the next 20 years as well ascurrent technology inferred from past and ongo-ing field tests. For example, technological ad-vances in temperature tolerance are projected sothat reservoirs which have a temperature of200 F can have a technical field test in 1985.

The OTA screening criteria coincide withthose used by the National Petroleum Council(NPC)16 with one exception. The OTA data basedid not contain adequate water-quality data forall reservoirs. Consequently, reservoirs were notscreened with respect to water quality.

The screening criteria were reviewed prior toacceptance. The review process included infor-mal contacts with personnel who did not partici-pate in the NPC study and an examination of thetechnical literature. The principal variables arediscussed in the following sections.

The screening criteria are judged to be repre-sentative of the present and future technologicallimits. As discussed later, it is recognized thatpermeability and viscosity criteria have economiccounterparts. However, the number of reservoirseliminated as candidates for the surfac -tant/polymer process by either of these screeningcriteria was insignificant.

Current Technology (1976).--Current limits oftechnology are reflected by field tests whichhave been conducted or are in an advanced stageof testing. These are summarized in table B-2.17

Field tests are generally conducted in reservoirswhere variation in rock properties is not largeenough to obscure the results of the displace-ment test due to reservoir heterogeneities. Thesereservoirs tend to be relatively clean sandstonewith moderate clay content. A crude oil viscosityless than 10 centipoise is characteristic of mostsurfactant/polymer field tests. Reservoir tem-peratures range from 55 F to 169 F.

Reservoir Temperature.Surfactants andpolymers are available which tolerate tem-peratures up to about 170 F. Research onsystems which will be stable at 200 F is under-way in several laboratories. The rate of tech-nological advance in this area will probably berelated to the success of field tests of the surfac-tant/polymer process in lower-temperature reser-voirs . Successful f ie ld tests wi l l s t imulatedevelopment of fluids for higher-temperaturedeeper reservoirs as potential applications inthose reservoirs become a reality. The assumedtiming of technological advances in temperaturelimitations appears attainable.

Permeability and Crude Oil Viscosity. -Per-meability of the reservoir rock is both a tech-nological and an economic factor. The surfac-tant/polymer process will displace oil from lowpermeability reservoir rock.18 A minimum per-meability based on technical performance of theprocess has not been established. Low per-meability may correlate with high-clay content ofthe reservoir rock and corresponding high-surfac-tant losses through adsorption. The surfactantslug must be designed so that its integrity can bemaintained in the presence of large adsorption

Table B-2Summary of Surfactant Field Tests Being Conducted by

Industry Without ERDA AssistanceProcess Area Porosity Perm. Depth Reservoir Oil Temp. Salinity

Field State County Operator Type* (Acres) Start Pay (%) (Md) (ft) o APl) (Cp) (F) (ppm) Comment0.75-40

4.30.75-45

10

1-160

0.65

8.231

0.8

209

1.25

2.02.52.52.5S.o

5.810.0

2.51.7

200

82

69

103

50052

75

53

*400

2,50090

457394

8793

950450

1,000 35-36 7 72 HPW 18,150 ppm 6 testsTDS (1 19-R)

11 /62

5/7012/68

5/71

11/67

9/70

Robinson 20

18

19

21

22.919.2

20

22

21

331821.724.2

14.817.13126

Robinson Ill. Crawford Marathon MSF

MSFMSF

MSF

Aqueous

AqueoussolutionSOFSF

SF

LTWF

AqueoussolutionSFMSFSFSFSF

LTWFSFLTWFSF

MarathonPennzoii

Aux VasesBradford

3,0001,860

Ill.Bingham Pa. McKean 5

4.5

4

68 2,800 Cl 2 tests

600 40 55Goodwill Pa. WaxyenHill

Benton Ill. Franklin

Quaker St. First Venongo

86 77,000 ppm TDS 2 testsShell Tar Springs 2,100

Est.9595114

64,000 Cl 104,000 TDS54,000 cl7,700 TDS,20 ppm fractured,CA+ Mg

Exxon ChesterCypressBluff CreekSecondWall CreekUpperCypressGunsight

1,460 4Loudon Ill.

8/698/73

11 /73

7/73

4.35.6

1,8703,100

3734

Higgs Unit Tex. JonesBig Muddy Wyo. Converse

UnionConoco

37Griffin Ind. GibsonConsol.Wichita Tex. WichitaCo. RegularBorregos Tex. Kleberg

2,400Conoco

2.2 89 160,000 TDS1,750 42Mobil

33,000 TDSmid60s

Frio 5,000 42 0.4 165Exxon

20,000 TDSJacksonKirkwoodFlappenCisco

2,2701,5001,9001,200

3638:39

3827

1.65.5

12272

Guerra Tex. StarBridgeport Ill. LawrenceSayles Tex. JonesMontague Tex. MontagueLoma NoviaTex. Duval

SunMarathonConocoConocoMobil

9/69/63/63

mid60s4/741 /756J741/75

150,000 TDS4% kaolinitc

5.5% montmorillonite40,000cl2,457 TDS60,000Cl 1,017 Ca++ and Mg++

3.60.80.717

0.8516516960

TexacoAmocoMobilTexaco

U. BenoistMuddy J.41AAlmy

1,7506,2505,700

700

38343226

Salem Ill. MarionSloss Nebr. KimballWest RanchTex. JacksonLa Barge Wyo. Sublette

* Process Type normally refers to specific surfactant floods used, but is not intended to characterize actual differences: Aqueous-dispersion of sulfonate in water with very littleoil in slug; MSFmicellar surfactant flood; SOFnormally considered oil external chemical slug; SF and LTWFsurfactint flood and low-tension waterflood normally similar toaqueous systems.

Source: Enhanced 0// Recovery, National Petroleum Council, December 1976, p. 97.

150 . Appendix B

losses. As a result, larger slugs or higher con-centrations may be needed with correspondingincreases in costs.

Permeability, fluid viscosities, well spacing,and maximum injection pressure affect the rate atwhich a chemical slug can displace oil from areservoir. Low permeability translates to low dis-placement rates or increased well density tomaintain a specific rate. Both lead to higher proc-ess costs.

The same reasoning applies to crude oilviscosity. As viscosity increases, displacementrates decrease or well density increases. Mobilitycontrol in the surfactant/polymer process is at-tained by increasing the viscosities of the chemi-cal slug and the drive water. Both of thesechanges require addition of expensive constit-uents to these fluids. Therefore both permeabilityand viscosity are constrained by economics.

It is known from laboratory tests that oil recov-ery by the surfactant/polymer process is a func-tion of displacement rate. For example, more oilis recovered at an average displacement rate of 5ft per day than at the rate of 1 ft per day19 whichexists in a typical reservoir. Rate effects in fieldsize patterns may be revealed in the Marathon-ERDA commercial demonstration test.20

Water Quality .-Composition of the formationwater i s a cr i t ical var iable in the surfac-tant/polymer process. Fluids under field tests cantolerate salinities of 10,000 to 20,000 ppm withmoderate concentrations of calcium and mag-nesium, although reservoirs containing low-salinity flu ids are preferred. Some field tests are inprogress in which preflushes are used to reducesalinity to levels which can be tolerated by theinjected chemicals. 21,22 However, in one largefield test23 the inability to attain a satisfactorypreflush was considered to be a major contribu-tor to poor flood performance. Potential short-ages of fresh water for preflushing and uncertain-ty in effectiveness of preflushes have stimulatedresearch to improve salinity tolerance.

Technological advances were projected in theNPC study which would increase the salinitytolerance from 20,000 ppm in 1976 to 150,000ppm in 1980 and 200,000 ppm in 1995. The OTAtechnical screen does not contain a similar

scenario because salinity data were not availablefor all the reservoirs in the OTA data base. It doesnot appear that results would have been affectedappreciably if the data were available in the database to schedule technological advances insalinity tolerance.

Rock Type. The surfactant/polymer processis considered to be applicable to sandstone reser-voirs. Carbonate reservoirs are less attractive can-didates because 1 ) the formulation of compatiblefluids is more difficult due to interaction withcalcium and magnesium in the rocks; 2) carbon-ate reservoirs frequently produce through small-and large-fracture systems in which maintenanceof an effective surfactant slug would be difficult;and 3) there is a consensus among technical per-sonnel that the C02 miscible displacement proc-ess is a superior process for carbonate reservoirs.

Reservoir Constraints. Reservoirs with largegas caps which could not be waterflooded eitherby natural water drive or water injection are likelyto be unacceptable. Also, reservoirs which pro-duce primarily through a fracture system fall inthe same category. However, there is thepossibil ity of technological developments 24

which would restrict flow in the fracture systemand perm t displacement of the surfactant slugthrough the porous matrix.

Oil Recovery ProjectionsThe surfactant/polymer process is applied in a

r e s e r v o i r w h i c h h a s b e e n p r e v i o u s l ywaterflooded, There are different opinionsamong technical personnel concerning thevolume of the reservoir which may be swept bythe process. Some consider that the sweptvolume will be less than the volume swept by thewaterflood, while others envision more volumeswept by the surfactant/polymer process. Thereasoning behind these viewpoints is summa-rized in the following subsections.

Swept Volume Less Than Water flood Sweep.Residual oil saturations and volumetric sweepefficiencies attributed to waterflooding are fre-quently the result of displacing many porevolumes of water through the pore space. In con-trast, the surfactant/polymer process can be ap-proximated as a 1- to 2-pore volume process

which may lead to a smaller fraction of the reser-voir being contacted by the surfactant/polymerprocess.

Many reservoirs are heterogeneous. It can bedemonstrated that heterogeneities in the verticaldirection of a reservoir which have relativelysmall effect on the sweep efficiency of awaterflood may have large effects on the sweepefficiency of the surfactant/polymer process.25

For instance, in a layered reservoir it may not bepossible to inject enough surfactant into all layersto effectively contact the regions which werepreviously waterflooded.

Surfactant/Polynmer Swept Volume Outside ofWaterflood Region. The region outside of thevolume swept by the waterflood contains a highoil saturation. in many surfactant processes, theviscosity of the injected fluids is much higherthan water used in the previous waterflood. Thiscould lead to increased volumetric sweep effi-ciency for the surfactant/polymer process.D a v i s 26 has presented data from a Maraflood T M

oil recovery process test in the Bradford ThirdSand of Pennsylvania. An increase of 7 to 10 per-cent in the volumetric sweep efficiency for thesurfactant process over the previous waterfloodwas indicated in his interpretation of the data.

OTA Model.-The OTA model is based on theassumption that the region contacted by the sur-factant/polymer process in most reservoirs is theregion swept by the previous waterflood. Thesurfactant/polymer process displaces oil from thepreviously water-swept region by reducing theoil saturation following the waterflood (Sorw) to alower saturation, termed SOf, which representsthe residual oil saturation after a region is sweptby the surfactant/polymter process. The oil recov-ery using this representation of the process wascomputed using equation 1 B for each patternarea,

B.

where

Npc

= oil displaced bystock-tank barrels

A p = area of the pattern

r-N (s

the chemical flood,

Appendix B . 151

h = net thickness of pay

9 = porosity, the fraction of the rock volumewhich is pore space

E vm = fraction of the reservoir volume whichwas contacted by water and surfac-tant/polymer process determined bymaterial balance calculations

BO = ratio of the volume of oil at reservoirtemperature and pressure to the volumeof the oil recovered at stock-tank condi-tions (60 F, atmospheric pressure)

Residual oil saturations left by the chemicalflood (SOf ranging from 0.05 to 0.15 have beenreported in laboratory 27,28 and field tests. 29 Avalue of 0.08 was selected for the OTA computa-tions.

T h e r e s i d u a l o i I s a t u r a t i o n f o l l o w i n gwaterflood (SOrw for the high-process perform-ance case was the oil saturation corresponding tothe particular geographic region in table A-1modified by the material balance calculation asdescribed in appendix A, in the section on Dis-tribution of (the Remaining Oil Resource on page139. In the low-process performance model, theresidual oil saturations following waterflood(SORW) were reduced by 5 saturation percent fromthe values in table A-1. This caused a decrease inrecoverable oil which averaged 28.6 percent forall surfactant/polymer reservoirs. Due to themethod of analysis, the process performance of asmall number of reservoirs was not affected bythis saturation change. Some reservoirs whichhad 90-percent volumetric sweep imposed bythe material balance discussed on page 139 forthe high-process performance case also had 90-percent volumetric sweep efficiency under low-process performance.

Pattern Area and Injection Rate.Each reser-voir was developed by subdividing the reservoirarea into five-spot patterns with equal areas. Thesize of a pattern was determined using the pro-cedure developed in the NPC study.30 A patternlife of 7 years was selected. Then, the patternarea and injection rates were chosen so that 1.Sswept-pore volumes of fluids could be injectedinto the pattern over the period of 7 years. Therelationship between pattern area and the injec-

tion rate is defined by equation 2B.

.

152 . Appendix B

injection rate, barrels per dayporositythickness, feetpattern area, acres

Maximum pattern area was limited to 40 acres.

Injection rates were constrained by two condi-tions. In Texas, California, and Louisiana, it wasassumed that maximum rates were limited bywell-bore hydraulics to 1,000 barrels per day,1,500 barrels per day, and 2,000 barrels per day,respectively. Rate constraints in the reservoirwere also computed from the steady-state equa-tion for single-phase flow in a five-spot patterngiven in equation 3B. The viscosity of the surfac-tant/polymer slug was assumed to be 20 timesthe viscosity of water at formation temperature.The lowest injection rate was selected. otherparameters are identified after the definition ofthe equation.

completion of wells and installation of surfacefacilities were done in the first 2 years. The sur-factant slug was injected during the third yearwith the polymer injected as a tapered slug fromyears 4 through 6. The oil displaced by the sur-factant/polymer process as computed from equa-tion IB was produced in years 5 through 9according to the schedule in table B-3.

Table B-3Development of a Five-Spot Pattern

Surfactant/Polymer Process

Year of Annual oil productionpattern % of

development Activity incremental recovery

1

2

3

456

3B

where

i = injection rate, barrels per dayk = average permeability, millidarciesh = average thickness, feetAP = pressure drop from injection to produc-

ing well, taken to equal depth/2

eff = effective viscosity of surfactant/polymerslug, or 20 times viscosity of water atreservoir temperature

In = natural logarithmd = distance between the injection and pro-

duction well, feet, or 147.58 ~~A P = pattern area, acres

RW = radius of the well bore

Development of Pattern.-Development ofeach five-spot pattern took place according tothe schedule shown in table B-3. Drilling and

789

Drill and completeinjection wells. Re-work productionwell.

I n s t a l l s u r f a c eequipment.

Inject surfactantslug.

Inject polymer slugwith average con-centration of 600ppm. Polymer con-c e n t r a t i o ntapered.

Injection of brine.

o

0

0

01026

322012

Total . . . . . . . 100

Volumes of Injected Materials.

Current technology

Surfactant Slug, . . 0.1 swept pore volume*Polymer Bank. . . . 1.0 swept pore volume

Advancing technology case

Surfactant Slug. . . 0.1 swept pore volumePolymer Bank, . . . 0.5 swept pore volume

l The swept pore volume of a pattern is defined by equa-tion 4B.

V p = Ev~ AP h o (7,758) 4B

= volume of pattern swept bythe surfactant/polymer proc-ess, barrels

The volumes of surfactant and polymer approxi-mate quantities which are being used in fieldtests. Volume of the surfactant slug needed tosweep the pattern is affected by adsorption ofsurfactant on the reservoir rock. The slug of 0.1swept-pore volume contains about 36 percentmore sulfonate than needed to compensate forloss of surfactant that would occur in a reservoirrock with porosity of 25 percent and a surfactantretention of 0.4 mg per gm rock. The OTA database contained insufficient information to con-sider differences in adsorption in individual reser-voirs. The effect of higher retention (and thushigher chemical costs) than assumed in the ad-vanced technology cases is examined in the high-chemical cost sensitivity runs.

Composition and Costs of Injected MaterialsThe surfactant slug for all cases except the cur-

rent technology case contained 5-wt percentpetroleum sulfonate (100-percent active), 1-wtpercent alcohol, and 10-volume percent leasecrude oil. In the current technology case, the sur-factant slug contained 20 percent lease crude oil.The concentration of the polymer solution was600 ppm for reservoir oils with viscosities lessthan or equal to 10 centipoise. Concentration ofpolymer was increased with viscosity for oilsabove 10 centipoise according to the multipliergiven in equation 5B.

Concentration Multiplier =(1 +32- API

) 5B10

Equation 5B is valid for API gravities greater than10. A polysaccharide polymer was used.

Table B-4 summarizes surfactant slug andpolymer costs as a function of oil price. Costs ofsurfactant and alcohol based on data from theNPC study are presented in table B-5.

Net Oil, -Projected oil recovery from the sur-factant/polymer process was reported as net bar-rels. The oil used in the surfactant slug and an

Appendix B l 153

estimate of the oil equivalent to the surfactantwas deducted from the gross oilproduction.

Table B-4Chemical Coats

Surfactant

slug cost -10-percent

Oil price lease crude$/bbl $/bbl

10 . . . . . . . . 7.6915 . . . . . . . . 9.7320 . . . . . . . . 11.7425 . . . . . . . . 13.78

to determine net

Surfactant

slug cost -20-percentlease crude

$/bbl

8.6911.2313.7416.28

Polymer cost*polysaccharide

$/lb

2.302.402.492.58

l Source: Enhanced Oil Recovery, National Petroleum Council,December 1976, p. 100.

Table B-5Component Costs*

Surfactant costOil price 100-percent active Alcohol cost

$/bbl $/lb $/lb

5 . . . . . . . . 0.29 0.1310 ....., . . 0.35 0.1615 . . . . . . . . 0.43 0.2020 . . . . . . . . 0.51 0.2325 . . . . . . . . 0.59 0.27

*Includlng tax and transportation.Source: Enhanced 011 Recovery,

December 1976, p. 99.National Petroleum Council,

Sensitivity AnalysesAdditional computations were made using the

low- and high-process performance models todetermine sensitivity to changes in chemicalcosts. Cost sensitivity analysis was accomplishedby altering the volumes of surfactant andpolymer used in the displacement process. Thelow-chemical cost case assumes a 40 percentreduction in the volume of the surfactant slugwhile the high-chemical cost case assumes that40 percent more surfactant and 50 percent more

polymer would be required than used in thebase-chemical cost case.

154 . Appendix B

Ultimate recoveries of oil using the surfac-tant/poIymer process with high- and low-chemi-cal cost assumptions are summarized in table B-6for the advancing technology cases. With highchemical costs, there would be a negligiblevolume of oil produced at world oil price. Thecombination of both high-process performanceand oil prices approaching the alternate fuelsprice would be needed to offset high chemicalcosts if the surfactant/polymer process is to con-tribute substantial volumes of oil to the Nationsreserves.

Low chemical costs have the largest impact onthe low-process performance case where sub-stantial increases in ultimate recovery could oc-cur at both upper tier and world oil price. Theeffect of lower chemical costs on the high-proc-ess performance case is to reduce the oil price re-quired to call forth a fairly constant level of pro-duction. For example, if chemical costs are low,the ultimate recovery projected at alternate fuelsprice is about the same as ultimate recovery atupper tier price. However, low chemical costshave a low probability of occurring unless a ma-jor technological breakthrough occurs.

The sensitivity analyses in this study weredesigned to bracket the extremes which might beexpected assuming technology develops aspostulated in the advancing technology cases.There are other process and economic variables

which would beindividual field

considered in the analysis of anproject which could not be

analyzed in a study of this magnitude.

Polymer FloodingState of the ArtTechnological Assessment

The concept of mobility control and its rela-tionship to the sweep efficiency of a waterfloodevolved in the early to mid-1950s.31,32 It wasfound that the sweep efficiency could be im-proved if the viscosity of the injected watercould be increased. Thickening agents were ac-tively sought. Numerous chemicals were evalu-ated but none which had economic potentialwere found until the early 1960s.

During this period, development in the fieldof polymer chemistry provided new moleculeswhich had unique properties. High-molecularweight polymers were developed which in-creased the apparent viscosity of water by factorsof 10 to 100 when as little as 0.1 percent (byweight) was d i s so lved in the water . The f i r s t

polymers investigated were partially hydrolyzed

polyacrylamides with average molecular weight

ranging from 3 million to 10 million.

The discovery of a potential low-cost methodto slow down the flow of water and improvesweep efficiency of the waterflood led to manyfield tests in the 1960s. Nearly all field tests used

Tabie B-6Surfactant/Polymer Process-Uitimate Recovery

Summary of Computed Results-Process and Economic Variations(billions of barrels)

Advancing technology casesOil price $/bbl

Case Low-process performance High-process performance

11.62 13.75 22.00 11.62 13.75 22.00

High chemical costs . . . . . . . . . 0.1 0.1 1.0 0.2 0.2 9,0

Base chemical costs . . . . . . . . . 1.0 2.3 7.1 7.2 10.0 12.2

Low chemical costs . . . . . . . . . 5,8 7.5 8.8 12.0 12.4 14.5

Appendix B 155

partially hydrolyzed polyacrylamides. By 1970 at

least 61 f ield tests had been init iated 33 a n d b y

1975 the number of polymer field tests exceeded100. Although most f ield tests were relativelysmall, two were substantial. These were the Pem-bina test in the Pembina Field in Alberta and the

Wilmington test in the Ranger V interval of the

Wilmington Field in California.

Results of field tests have been mixed. Suc-cessful use of polymers has been reported inseveral projects 3536 where incremental oilabove that expected from waterflooding hasbeen produced. At least 2 million barrels of oilhave been attributed to polymer flooding fromsuccessful projects. 37 Continuation of some proj-ects and expansion of others indicate commercialoperation is possible. However, polymer floodinghas not been widely adopted. Many field testsyielded marginal volumes of oil. Response topolymer flooding was not significant in either thePembina Flood or the Wilmington Flood.

Reasons for mixed field performance are notcompletely understood. polymer floods initiatedearly in the life of a waterflood are more likely tobe successful than those initiated toward the endof a project. Reservoirs which have beenwaterflooded to their economic limit have notresponded to polymer flooding as a tertiary proc-ess. Recent research 38 has demonstrated that par-tially hydrolyzed polyacrylamides degrade whensheared under conditions which may be presentin injection well bores. Thus, it is not certain inprevious field tests that a reservoir flooded withpolymer solution was contacted with the samefluid used in laboratory tests.

Further research and development produced apolysaccharide biopolymer 39 which has im-proved properties. Polysaccharides are relativelyinsensitive to mechanical shear and have hightolerance to salt, calcium, and magnesium ions.Solutions containing polysaccharides must befiltered prior to injection to remove bacterialdebris which may plug the injection wells. Sincethe polysaccharide is a product of a biologicalprocess, it is susceptible to further biological at-tack in the reservoir unless adequate biocide isincluded in the injected solution. Few field testshave been conducted using polysaccharidepolymers.

polymer flooding has economic potentialbecause it uses materials which are relatively lowcost. Field application is similar to waterfloodingwith minor changes to permit mixing and properhandling of the polymer solutions. Widespreaduse by most operators would be possible withoutextensive technical support. Performance ofpolymer floods cannot be predicted accurately,and well-documented demonstration projectssuch as those being conducted in the N. BurbankStanley Stringer40 and the Coalinga41 fields are es-sential to the widespread use of polymer flood-ing.

Screening Criteria. --Polymer flooding is not apotential process for all reservoirs which can bewaterflooded. Geologic constraints, properties ofthe reservoir rock and oil, and stage of thewaterflood are all critical parameters. Reservoirswhich produce primarily through large fracturesystems and reservoirs with large gas caps whichcould not be waterflooded were excluded. Inthese reservoirs, the polymer slug is likely tobypass much of the reservoir rock. A permeabilityconstraint of 20 millidarcies was selected. Whilethe lower limit of permeability is not known pre-cisely, there is a range of permeabilities wherethe polymer molecules are filtered out of the in-jected solution and cannot be propagatedthrough a reservoir. Selection of the correctmolecular weight distribution of the polymerreduces the minimum permeability.

Field experience indicates that polymer floodshave not been successful when applied after thewaterflood has been completed. Reservoirs underwaterflood which have volumetric sweep effi-ciency greater than 80 percent and low residualoil saturations are not good polymer candidates.Consequently, reservoirs with no ongoingwaterflood and reservoirs with high volumetricsweep efficiency and low oil saturation werescreened from the polymer flooding candidates.

Water quality was not used to screen reser-voirs because salinity and divalent ion content donot determine whether a reservoir can be floodedwith polymer solutions. These parameters do in-dicate the type of polymer which may be used.For example, par t ia l l y hydro lyzed po ly -acrylamides are frequently preferred in low-salinity systems. Polysaccharides are relatively in-

-

.

156 . Appendix B

sensitive to salinity and may be required in orderto flood successfully a reservoir which containshigh-salinity fluids.

The use of polymers is limited by temperaturestability. Proven temperature stability is about200 F. This limit is expected to be 250 F by1995. The same temperature limits used in thesurfactant/polymer process screen apply topolymer flooding.

Crude oil viscosity was the final screeningparameter. Field tests suggest an upper limit ofabout 200 centipoise. However, there is littlepublished literature which shows that polymerso lu t ions w i l l not d i sp lace o i l a t h igherviscosities. Other factors enter in the determina-tion of the upper viscosity limit. Steam displace-ment and in situ combustion are consideredsuperior processes because both can potentiallyrecover more oil. As crude oil viscosity increases,higher polymer concentrations are required tomaintain mobility control. Oil-displacement ratesdecline for a fixed pattern size. Both of these fac-tors operate in the direction of reducing the rateof return at fixed oil price or requiring a higher oilprice to produce a fixed rate of return. Then thecrude oil viscosity becomes an economic factorrather than a technical factor.

Most reservoirs which were polymer candi-dates yielded more oil when developed as C02,surfactant/polymer, steam, or in situ combustioncandidates. Thus, the OTA method of processselection, i.e., maximum oil if profitable at 10percent rate of return and world oil price, led toassignment of the poorest reservoirs to polymerflooding.

Oil Recovery ProjectionsEstimates of oil recovery from the application

of polymer-augmented waterflooding to reser-voirs which satisfied the technical screen weremade using an empirical model. Incrementalrecovery for the low-process performance casewas assumed to be 2.5 percent of the original oilin place. The incremental recovery for the high-process performance case was assumed to be 3percent of the original oil in place. These esti-mates closely approximate recent projections forthe N. Burbank Stanley Stringer and Coalinga field

demonstration tests. They also approximate theaverage performance of published field tests.42

Each reservoir was developed on 40-acre spac-ing with a ratio of 0.5 injection well per produc-tion well. Injection of polymer was continuedover the first 4 years of the project at a rate of0.05 pore volumes per year. Average polymerconcentration was 250 ppm. The polymer usedwas polysaccharide. Costs of polymer at variousoil prices were identical to those used for the sur-factant/polymer process (table B-4).

The recoverable oil was produced over an 11-year period according to the schedule in tableB-7.

Table B-7Production Schedule

for Polymer-Augmented Waterflood

Incremental oilYear percent of total

1-2 . . . . . . . . . . . . . . . . . . . . . . .3 . . . . . . . . . . . . . . . . . . . . . . . .4 . . . . . . . . . . . . . . . . . . . . . . . .5 . . . . . . . . . . . . . . . . . . . . . . . .

o5

102020151010

55

Total. . . . . . . . . ., . . . . . . . 100

Sensitivity AnalysesThe effects of changes in polymer costs and/or

volumes were examined for low- and high-polymer costs for both low- and high-processperformance cases. Bases for cost variation were+/- 25 percent change in polymer cost. Results ofthe economic evaluations are presented in tableB-8.

There is essentially no effect of chemical costson oil production from polymer flooding at theupper tier, world oil, and alternate fuels prices.The sensitivity analyses show that uncertainty inprocess performance is larger than uncertaintiesintroduced by chemical costs.

Appendix B . 757

Table B-8Polymer-Augmented Waterflooding

Ultimate Recovery(billions of barrels)

Case

High polymer cost(+25/0 over base) . . . . . . . . . . . . .

Base polymer cost. . . . . . . . . . . . .

Low chemical cost(-25/0 from base). . . . . . . . . . . . .

Advancing technology casesoil price $/bbl

Low-process performance

11.62

0.2

0.2

0.3

13.75

0.2

0.3

0.3

22.00

0.3

0.3

0.3

High-process performance

11.62

0.4

0.4

0.4

13.75

0.4

0.4

0.4

22.00

0.4

0.4

0.4

Effect of Polymer Flooding on SubsequentApplication of Surfactant/Polymer or CarbonDioxide Miscible Processes

The OTA analysis assumes a single processwould be applied to a reservoir. The possibilityof sequential application of two processes wasnot analyzed. Some reservoirs assigned to thesurfactant/polymer process or the C02 miscibleprocess would also be economic (rate of returngreater than 10 percent at world oil price) aspolymer floods. However, the decision rules forprocess assignment placed these reservoirs in theprocess which yielded the largest ultimate recov-ery.

One concern caused by this assignment pro-cedure was whether or not the low costs and low fi-nancial risk from the polymer projections wouldcause operators to use polymerflooding as the finalrecovery process for a reservoir, precluding use ofmethods which potentially recover more oil.

The principal displacement mechanism inpolymer flooding is an increase in the volume ofthe reservoir which is swept by the injected fluid.No reduction in residual oil saturation over thatexpected from waterflooding is anticipatedbecause the viscosities of the oils in these reser-voirs are low enough to make the residual oilsaturations relatively insensitive to the viscosityof the displacing fluid.

A successful polymer flood in the OTA high-process performance would recover 3 percent ofthe original oil in place. This corresponds roughlyto improved volumetric sweep efficiencies of 2to 7 percent. Both OTA models for surfac-tant/polymer and CO2 miscible processes arebased on recovery of the residual oil from somepercentage of the volume displaced by the pre-ceding waterflood. Polymer flooding increasesthis contacted volume. Slightly more oil wouldbe recovered from reservoirs which had beenpolymer flooded prior to surfactant flooding orC 02 flooding if the OTA models of these dis-placement processes are substantially correct.Therefore, the application of polymer floodingwill not prevent subsequent surfactant/polymeror C02 floods under the conditions postulated inthe OTA study.

Finally, polymer flooding prior to surfac-tant/polymer flooding has been proposed as amethod to improve volumetric sweep efficiencyby increasing the flow resistance in more permea-ble paths in the reservoir.43

Steam Displacement

State of the ArtTechnological Assessment

Steam displacement is a process which has pri-marily evolved in the last 10 to 15 years.

158 . Appendix B

Development of the process was motivated bypoor recovery efficiency of waterfloods in reser-voirs containing viscous oil and by low producingrates in fields which were producing by primaryenergy sources. Most of the development oc-curred in California and Venezuela, where largevolumes of heavy oil are located. Steam displace-ment has potential application in heavy oil reser-voirs in other oil-producing States.

Large-scale field tests of steam injection beganin the late 1950's 44,45 with field testing of hotwater injection underway at the same time46,47,48

in an attempt to improve the recovery efficiencyof the conventional waterflood. Early steam andhot water injection tests were not successful. in-jected fluids quickly broke through into the pro-ducing wells, resulting in low producing rates andcirculation of large volumes of heated fluids.

The process of cyclic steam injection was dis-covered accidentally in Venezuela in 19s9 andwas developed in California. 49 Cyclic injection ofsmall volumes of steam into producing wellsresulted in dramatic increases in oil production,particularly in California where incremental oildue to cyclic steam injection was about 130,000barrels per day in 1968.50 By 1971 about 53 per-cent of all wells in California had been steamedat least once.

Cyclic steam injection demonstrated that sig-nificant increases in production rate could be ob-tained by heating the reservoirs in the vicinity ofa producing well. However, the process is pri-marily a stimulation process because naturalreservoir energy sources like solution-gas drive orgravity drainage cause the oil to move from thereservoir to the producing well, Depletion of thisnatural reservoir energy with repeated applica-tion of cyclic steam injection will diminish thenumber of cyclic steam projects. Many of theseprojects will be converted to steam displace-ment.

The success of the steam displacement proc-ess is due to the high displacement efficiency ofsteam and the evolution of methods to heat areservoir using steam. Development of the steamdisplacement process in the United States can betraced to large-scale projects which began in theYorba Linda Field in 196051 and the Kern RiverField in 1964.52 Estimates of ultimate recoveries

(primary, secondary, cyclic steam, and steam dis-placement) from 30 to 55 percent of the originaloil in place have been reported for several fields.

A comparison 53 of trends in incremental oilproduction from cyclic steam and steam injectionfor California is shown in figure B-1. Cyclic steaminjection is expected to decline in importance asnatural reservoir energy is depleted. Productionfrom steam displacement could increase as cyclicprojects are converted to continuous steam injec-tion. The rate of conversion will be controlled byenvironmental constraints imposed on exhaustemissions from steam generators. Incremental oilfrom steam displacement will be limited to110,000 barrels per day in California, the levelwhich currently exists, unless technological ad-vances occur to reduce emissions.

Commercial steam-displacement projects arealso in operation in Wyoming,54 Arkansas, 55 andTexas.56 A large portion of the incremental oilnow produced by application of EOR processes isproduced by the steam displacement process.

Screen/rig Criteria. Steam displacement wasconsidered applicable in reservoirs which werelocated at depths between 500 and 5,000 feet.The upper depth limitation was imposed in orderto maintain sufficient steam injection pressure.The lower depth of 5,000 feet is determined bywell-bore heat losses in the injection wells. Atdepths approaching 5,000 feet, heat losses canbecome excessive even with insulated injectionstrings. In addition, as depth increases the injec-tion pressure increases, but the fraction of the in-jected fluid which is condensable decreases.Reduction in displacement efficiencies is ex-pected to occur under these conditions.

T h e s e c o n d s c r e e n i n g c r i t e r i o n w a st ransmi s s ib i I i t y . The t ransmi s s ib i l i t y (pe r -meability x thickness/oil viscosity) is a measureof the rate that the oil moves through a reservoirrock. A transmissibility of about 100 millidarcyfeet/centipoise is required for steam and hot-water injection processes in order to keep heatlosses from the reservoir to overlying and un-derlying formations from becoming excessive. ST

Oil Recovery ProjectionsRecovery Models.Although steam displace-

ment is the most advanced EOR process, it was

Appendix B . 159

Figure B-1. Historical Incremental Production Thermal Recovery-California

1968 1970

difficult to develop recovery models which ap-plied to an entire reservoir. The OTA data base aswell as the Lewin data bases used in the NPC andERDA reports contained little information onreservoir variability. Review of the technicalliterature and personal contacts with companiesoperating in fields with major steam displace-ment projects revealed considerable variability inthickness and oil saturation. It became apparentthat most steam displacement projects werebeing conducted in the best zones of a reservoir,where oil saturations were higher than theaverage values in the data base. Thus, OTA con-cluded that empirical recovery models based onthe results of these displacement tests could notbe extrapolated to poorer sections of larger reser-voirs with the available information. Subdivisionof several large reservoirs into smaller segmentsof different properties as done in the NPC studywas considered, but could not be done with theavailable computer program.

Recovery models were developed by OTA toestimate the recovery based on development ofthe entire reservoir. In taking this approach, it is

--

1972 1974 1976

Acknowledged that recovery from the better sec--tions of a reservoir will be understated and therecovery from poorer sections will be overstated.However, this approach was preferable to over-statement of recovery caused by applying empiri-cal recovery models from the better zones58 t oother intervals and areas of a reservoir, or ap-plication of recovery adjustment factors to ex-trapolate single-pattern performance to total-project performance. 59

Each reservoi r with mult ip le zones wasdeveloped zone by zone. The technology neces-sary to complete each zone selectively wasassumed to evolve through research anddevelopment. The average thickness per zonewas determined by dividing the net thickness bythe number of zones. Two displacement modelswere used based on the thickness of the zone.Single zone reservoirs were handled in the sameway-according to thickness of the zone.

High-Process Performance Case.Zone Thick-ness Less Than or Equal to 75 Feet. -Gross oilrecoverable by primary and secondary produc-tion followed by steam was considered to be 50

160 . Appendix B

percent of the original oil in place. Thus in each Zone Thickness Greater Than 75 Feet. Oilzone, displacement in thick reservoirs is based on the

Steam Displacement Oil =Original Oil (Primary + following model of the displacement process.

2 Secondary) Steam displacement patterns were developed

on 2.5-acre spacing with one injection well per

producing well.

Maximum verticalAreal sweep thickness of Residual oil

Region efficiency swept zone, feet saturation

Steam Zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.75 25 0.10

Hot Water Zone . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.90 35 0.25

Low-Process Performance Case. Well spacingwas increased to 5 acres. Gross oil displaced bysteam was 80 percent of the amount estimatedfor the high-process performance case.

Timing of Production. The incremental oilfrom the steam-displacement process was pro-duced according to the production schedule intable B-9.

Table B-9Production Schedule for

Steam Displacement Process

Annualincremental oil

Year percentage total

1-2 . . . . . . . . . . . . . . . . . ~ ~ ~ ~ . 03 12. . . . . . . . . . . . . . . . . . . . . . . .4 . . . . . . . . . . . . . . . . 225 : : : : : : : : . . . . . . . . . . . . . . . . 226 . . . . . . . . . . . . . . . . . . . . . . . . 207 . . . . . . . . . . . . . . . . . . . . . . . . 148 . . . . . . . . . . . . . . . . . . . . . . . . 10

Total . . . . . . . . . . . . . . . . . . 100

The same schedule was used for low- andhigh-process performance models.

Steam Requirements and CostsSteam requirement was 1 pore volume based

on net heated thickness. That is the volume oc-cupied by the combined steam and hot waterzones considering the areal sweep efficiency tobe 100 percent. Zones with thicknesses less thanor equal to 75 feet were assumed to be heated inthe entire vertical cross section. Steam was in-

jected over a 5-year period beginning in the thirdyear of field development at the rate of 0.2 porevolume per year.

Lease crude was used as fuel for the steamgenerators. Twelve barrels of steam were pro-duced per barrel of lease crude consumed. Thefull cost of the lease crude was charged as anoperating cost to the project. Oil consumed asfuel was deducted from the gross production toobtain the net production. Cost of steam genera-tion in addition to the fuel charge was $0.08 perbarrel of steam generated to cover incremental

operating and maintenance costs for the genera-

tor and water treatment.

Other Costs. The costs of installed steamgeneration equipment were scaled from a 50million Btu per hour steam generator costing$300,000 .60 A 1 million Btu per hour unit wasassumed to generate 20,000 barrels of steam(water equivalent) per year. The number (possi-bly fractional) of generators required per patternwas determined from the pore volume of the pat-tern. Since the steam generator life was longerthan pattern life, it was possible to use the samegenerator on two patterns in the field. The cost ofmoving a generator was assumed to be 30 per-cent of the initial cost. Thus the effective cost forthe steam generator per pattern was 65 percentof the initial generator cost.

Reservoi rs with mult ip le zones requiredworkovers in production and, injection wells toclose the zone just steamed and open the nextzone. These costs are discussed in the section onEconomic DataGeneral on page 178 of this ap-pendix.

Case

Low recovery . . . . . . . . ... . . .High-process performance ...High recovery. . . . . . . . . . . . . . . .

Appendix Be 161

Table B-10Recovery Uncertainties Effecting Steam Displacement Results

Production wellspacing, acres

2.52.52.5

Recovery Modela

Zone thickness

Maximumsteam zone

thickness

252530

75 ft.

Maximumhot water zone

thickness

303535

aAll other ~Odel parameters were the same as in the high-process Performance case

These extremes in recovery performance areSensitivity AnalysesProjections of oil recovery by steam displace-

ment contain uncertainties which are primari ly

related to the recovery efficiency of the process.Additional analyses were made to determine therange of variation in oil recovery due to uncer-tainties in process performance (table B-10).

One set of projections was based on variationsof recovery for a well spacing of 2.5 acres p e rproduction well. Projections for low recovery (45percent) and high recovery (55 percent) are com-pared with the high-process performance case(50 percent recovery) in table B-11. Results fromthe low recovery case are essentially the same asthe low-process per formance case. The pro-jections from the high recovery case are apprecia-bly higher than the high-process performance

case.

Table B-nEffect of Uncertainties in OverallRecovery on Ultimate Production

Steam Displacement Process(billions of barrels)

Upper World Alternatetier oil fuels

Case price price price($1 1.62/ ($1 3.75/ ($22.00/

bbl) bbl) bbl)

Low recovery . . . . . . . 2.1 2.5 3.4High-processp e r f o r m a n c e 2.8 3.3 6.0

High recovery. . . . . . . 3.9 5.9 8.8

also measures of energy efficiency. Crude oil isburned to produce steam. The amount of crudeconsumed is proportional to the volume of steamrequired to heat the reservoir. Nearly the samevolume of steam and consequently the sameamount of lease crude is consumed for each ofthe three cases. Slight variations occur for zoneswith thicknesses greater than 75 feet, Most of theadditional oil projected in the high recovery caseis produced with little additional lease crude re-quired for steam generation. In contrast, a largerfraction of the produced oil is consumed in thelow recovery case because about the sameamount of crude is consumed to produce steamwhile a smaller amount of oil is produced by thedisplacement process.

Pattern size is the second variable which wasinvestigated in sensitivity calculations. Oil recov-ery was estimated for two additional well spac-ings using the high-process performance model.Results are summarized in table B-12. If recoveryi s u n a f f e c t e d b y w e l l s p a c i n g , t h e r e i s a n

economic incentive to increase well spacing over

the 2.5-acre spacing used in the OTA study.Results are sensitive to spacing primarily becausethe costs to work over both injection and pro-duction wells in order to move from zone to zoneare significant.

Increasing well spacing reduces these costs inproducing wells by a margin which permitsseveral large reservoirs to meet the 10-percent

CJb-sn 4 (3 - 7 H - 12

162 . Appendix B

Table B-12Effect of Well Spacing on Ultimate Recovery of

Oil Using the Steam Displacement ProcessIncremental 011

(billions of barrels)

Production Upper tier World oil Alternate fuelswell spacing price price price

Case acres ($11.62/bbl) ($1 3.75/bbl) ($22.00/bbl)

High-process performance . . . . . . . . . . . . 2.5 2.8 3.3 6.0High-process performance ., . . . . . . ., 3.3 3.5 5.3 6.8H i g h - p r o c e s s p e r f o r m a n c e . . . . . 5.0 5.6 6.4 7.0

rate-of-return criteria at lower prices. This is a po-tential area for technological advances beyondthose which were assumed in this study.

In Situ Combustion

State of the ArtTechnological AssessmentIn s itu combustion has been investigated in

t h e u n i t e d S t a t e s s i n c e 1 9 4 8 .6 1 B y t h emid-1950s, two pilot tests had been conducted.One test was done in a reservoir containing alight oil (35 API) with a low viscosity (6 cp).62

The second reservoir tested contained 18.4 APIoil which had a viscosity of 5,000 cp.63 These ini-tial pilot tests demonstrated that a combustionfront could be initiated and propagated in oilreservoirs over a wide range of crude oil proper-ties.

The initial demonstrations of the technicalfeasibil ity of in situ combustion stimulatedresearch and development of the process both inthe laboratory and in the field. Over 100 fieldtests of in situ combustion have been conductedin the United States.64

Field testing developed considerable tech-nology. Methods were developed to initiatecombustion, control production from hot wells,and treat the emulsions produced in the process.Improved process efficiency evolved withresearch and field testing of methods to inject airand water simultaneously.65,66 The wet combus-tion process was found to have the potential ofreducing the air requirements by as much as 30 toso percent over dry combustion.

Many field tests have been conducted but fewhave resulted in projects which are commercially

successful. Economic information was not availa-ble on current in situ combustion projects. Con-tinued operation over a several-year period withf ie ldw ide expans ion imp l ie s sa t i s facto ryeconomics. California fields include the MocoUnit in the Midway Sunset.67 West Newport, 68

San Ardo, South Belridge, Lost Hills, and Brea-Olinda.69 Successful operations have also beenreported in the Glen Hummel, Gloriana, and TrixLiz Fields in Texas,70 and the Bellevue Field inLouisiana. 71 The number of commercial opera-tions in the United States is estimated to be 10.72

In situ combustion has not been appliedwidely because of marginal economics at existingoil prices, poor volumetric sweep efficiency insome reservoirs, and competition with steam dis-placement processes. Some field tests showed anet operating gain but could not generateenough income to return the large investment re-quired for an air compressor. The phrase a tech-nical success but an economic failure bestdescribes many projects.

The movement of the in situ combustion zonethrough a reservoir is controlled in part by varia-tions in reservoir properties. Directional move-ment has been observed in most in situ combus-tion projects. There has been limited success incontrolling the volume of the reservoir which isswept by the process. This is a major area forresearch and development.

Reservoirs which are candidates for steam dis-placement are also candidates for in situ combus-tion. Experience indicates that steam displace-ment is generally a superior process from theviewpoint of oil recovery, simplicity of opera-tion, and economics. Thus, applications of in situ

combustion have been limited by the develop-ment of the steam displacement process.

In situ combustion has one unique charac-teristic. It is the only process which may be ap-plicable over a wide range of crude gravities andviscosities.

Screening Criteria.--In situ combustion is ap-plicable to a wide range of oil gravities andviscosities. No constraints were placed on oilviscosity. The maximum permissible API gravityis determined by the capability of a particularreservoir rock/crude oil combination to depositenough coke to sustain combustion. Low-gravityoils which are composed of relatively large frac-tions of asphaltic-type components meet this re-quirement. It is also known that some mineralscatalyze in situ combustion, allowing high gravityoils to become candidates for in situ combus-tion. 73 The maximum oiI gravity which might be acandidate with catalytic effects was estimated tobe 45 API.

Minimum reservoir depth was set at 500 feet.74

Adequate reservoir transmissibility, i.e.,

Permeabilitv x thicknessoil viscosity

is necessary to prevent excessive heat losses tooverlying and underlying formations. Theminimum acceptable transmissibility for in situc o m b u s t i o n is about 20 miIIidarcyfeet/centipoise. 75 Carbonate reservoirs were notconsidered to be candidates for in situ combus-tion.

Oil Recovery ProjectionsThe wet combustion process was used for the

OTA study. All projects were developed as 20-

Appendix B . 163

acre patterns. In the wet combustion process,three distinct displacement zones are formed: aburned zone, a steam zone, and a hot waterzone. Gross oiI recovered from each pattern wascomputed from the sum of the volumes dis-placed from each zone. Areal sweep efficiency,maximum zone thickness, and residual oil satura-tion for each zone are included in table B-13 forthe advancing technology cases.

Fuel consumption was 200 barrels per acrefoot. 76 The equivalent oil saturation consumed inthe burned zone is Sob, where Sob = 200/7,758 X0); @ is the porosity of the rock, and 7,758 is bar-rels per acre foot.

The initial oil saturation was S,,,, the materialbalance average oil saturation computed fromequation 1. The volume of oil displaced wasdetermined in the following manner. The actualthickness of each zone was determined byallocating the net pay between the three zones inthe order shown in table B-13. A reservoir 20 feetthick would have a burned zone and a steamzone while a reservoir 100 feet thick would ex-perience the effects of three zones in a 50-footinterval. The volume of oil displaced from eachzone was computed from the product of the pat-tern area, areal sweep efficiency, zone thickness,porosity, and displaceable oil in the swept inter-val. All oil displaced from the swept zones wasconsidered captured by the producing well.

Timing of Production.The life of each pat-tern was 8 years. Drilling, completion, and otherdevelopment was completed in the first 2 years.Air and water injection began in year 3 and con-tinued through year 8 for a total productive life of6 years. The displaced oil was produced accord-ing to the schedule in table B-14.

Table B-13Advancing Technology Cases

Oil Displacement ModelWet Combustion

Areal sweepRegion efficiency

Burned zone ... . . 0.55S t e a m z o n e . . , 0.60H o t w a t e r z o n e 0 8 0

I Residualoil saturation

Max. vertical Low-process High-processthickness, ft. performance performance

10 0 010 (),20 0.1530 0.30 0.25

164 . Appendix B

Table B-14Production Schedule

Wet Combustion

Annual production ofincremental oil

Year Percentage of total

1 - 2, . . . . . . . . . . . . . . . . . . . . o3 . . . . . . . . . . . . . . . . . 104: : ::::. . . . . . . . . . . . . . . . . 165 . . . . . . . . , . . . . . . . . . . . . . , 226... . . . . . . . . . . . . . . . . . . . . 207 188 : : : : : : : : : : : : : : : : : : : : : : : 14

Total . . . . . . . . . . . . . . . . 100

Operating CostsAir required was computed on the basis of

110-acre feet burned per 20-acre pattern (if thereservoir is at least 10 feet thick) and a fuel con-sumption of 200 barrels per acre foot. If theair/oil ratio was less than 7,500 standard cubicfeet (Scf) per stock-tank barrel (STB), air require-ments were increased to yield 7,500. Air re-quirements were then used to size compressorsand to determine the equivalent amount of oilwhich would be consumed as compressor fuel.

The amount of oil used to fuel the com-pressors was computed as a Btu equivalent basedon 10,000 Btu per horsepower hour. Energy con-tent of the oil was 6,3 million Btu per barrel. Thisoil was deducted from the gross production.

The corresponding equations for the price ofair as the price per thousand standard cubic feet($/MScf) were derived from data used in the NPCstudy. 77

Depth Cost Equationfeet $/MScf

O - 2,500 0.08 + 0.01108 P2 ,500- 5 ,000 0.08 + 0.01299 P5,000-10,000 0.08 + 0.01863 P

10,000-15,000 0.08 + 0.02051 P

where

P = oil price in $/bbl and the multiplier of P isthe barrels of oil consumed to compress 1MScf of air to the pressure needed to in-ject into a reservoir at the specifieddepth.

Compressed air was supplied by a six-stagebank of compressors with 1 horsepower provid-ing 2.0 MScf per day.78 Compressor costs werecomputed on the basis of $40()/installed horse-power.

Sensitivity AnalysesThe effect of uncertainties in operating costs

was examined using the high-process perform-ance model. A low-cost case was analyzed byreducing the compressor maintenance cost from$0.08/MScf to $0.07/MScf. A high-cost case in-creased the compressor maintenance to$0.10/MScf. Results of these cases are comparedin table B-1 5. Cost reduction had little effect onthe projected results while the 25-percent in-crease in maintenance cost reduced the ultimaterecovery by 19 percent at upper tier price and 8percent at world oil price for the high-processperformance case,

A case was also simulated in which the dis-placement efficiency in the steam and hot waterzones was increased by changing the residual oilsaturation in the steam zone to 0.10 and in thehot water zone to 0.20, Results of this case are in-dicated as high-displacement efficiency in tableB-1 S. The effect of assumed improvement in dis-placement efficiency resulted in a 17- to 20-per-cent increase in ultimate recovery but littlechange in price elasticity.

Table B-15Effect of Changes in Compressor Operating Costs

and Displacement Efficiency in Ultimate OilRecovery Using the In Situ Combustion Process

Case

I Incremental oil(billions of barrels)Upper

tierprice

($1 1.62/bbl)

Worldoil

price($1 3.75/

bbl)

Alternatefuelsprice

($22.00/bbl)

High cost. . . . . . . . . . . .High-process

performance . . . . . . . .Low cost . . . . . . . . . . . .High-displacement

efficiency . . . . . . . .

1.4

1.71.7

2.1

1.7

1.91.9

2.2

1.9

1.91.9

2.3

Carbon Dioxide Miscible

State of the ArtTechnological AssessmentIt has been known for many years that oil can

be displaced from a reservoir by injection of asolvent that is miscible with the oil. Because suchsolvents are generally expensive, it is necessaryto use a slug of the solvent to displace the oiland then to drive the slug through the reservoirwith a cheaper fluid, This process was shown tobe feasible at least 20 years ago.79 An overviewof the various kinds of miscible displacements isgiven by Clark, et al.80

Hydrocarbon miscible processes have beendeveloped and studied fairly extensively. A num-ber of field tests have been conducted.81 While ithas been established that hydrocarbon miscibleprocesses are technically feasible, the high costof hydrocarbons used in a slug often makes theeconomics unattractive. Recently, attention hasfocused on carbon d iox ide (C02) as themiscibility agent.82

In the OTA study it was assumed that, ingeneral, economics and solvent availabilitywould favor the use of C02. The C02 processwas therefore used exclusively as the miscibledisplacement process in the study.

Carbon dioxide has several properties whichcan be used to promote the recovery of crude oilwhen it is brought into contact with the oil.These properties include: 1 ) volubility in oil withresultant swelling of oil volume; 2) reduction ofoil viscosity; 3) acidic effect on rock; and 4)ability to vaporize and extract portions of thecrude oil under certain conditions of composi-tion, pressure, and temperature.

Because of these properties, C02 can be usedin different ways to increase oil recovery, i.e.,different displacement mechanisms can be ex-ploited. The three primary mechanisms are solu-tion gas drive, immiscible displacement, anddynamic miscible displacement.

Solution-gas-drive recovery results from thefact that C0 2 is highly soluble in oil. When C02is brought into contact with oil under pressure,the C02 goes into solution. When the pressure islowered, part of the C02 will evolve and serve asan energy source to drive oil to producing wells.

Appendix B . 165

The mechanism is similar to the solution-gas-drive primary recovery mechanism and can beoperative in either immiscible or miscible dis-placement processes.

Helm and Josendalal83 have shown that C02 canbe used to displace oil immiscible. In experi-ments conducted with liquid C02 below thecritical temperature, residual oil saturations weresignificantly lower after flooding with C02 thanafter a waterflood. The improved recovery wasattributed primarily to viscosity reduction and oilswelling with resultant improvement in the rela-tive permeability. It was noted that the C02 dis-placement was not as ef f ic ient when awaterflood preceded the C02.

Carbon dioxide, at reservoir conditions, is notdirectly miscible with crude oil. However,because C02 dissolves in the oil phase and alsoextracts hydrocarbons from the crude, it is possi-ble to create a displacing phase composition inthe reservoir that is miscible with the crude oil.

Menzie and Nielson, in an early paper,84 pre-sented data indicating that when C02 is broughtinto contact with crude oil, part of the oil vapor-izes into the gaseous phase. Under certain condi-tions of pressure and temperature, the extractionof the hydrocarbons is significant, especially ex-traction of the intermediate molecular weight hy-drocarbons (C5 to C30). Helm and Josendahl

85 alsoshowed that C02 injected into an oil-saturatedcore extracts intermediate hydrocarbons from theoil phase and establishes a slug mixture which ismiscible with the original crude oil. Thus, whiledirect contact miscibility between crude oil andC02 does not occur, a miscible displacement canbe created in situ. The displacement process,termed dynamic miscibility, results in recoveriesfrom linear laboratory cores which are compara-ble to direct contact miscible displacement.

HoIm 86 has pointed out that the C02 miscibledisplacement process is similar to a dynamicmiscible displacement using high-pressure drygas. However, important differences are that C02extracts heavier hydrocarbons from the crude oiland does not depend upon the existence of lighthydrocarbons, such as propane and butanes, inthe oil. Miscible displacements can thus beachieved with COZ at much lower pressures than

166 . Appendix B

with a dry gas. Methods of estimating miscibilitypressure have been presented.87,88

The CO2 miscible process is being examined ina number of field pilot tests.89,90 The largest ofthese is the SACROC unit in the Kelly-SnyderField. 91 Different variations of the process arebeing tested. In one, a slug of CO2 is injectedfollowed by water injection. In another, CO2 andwater are injected alternatively in an attempt toimprove mobility control. 92

The preliminary indication from laboratory ex-periments and these field tests is that the C02process has significant potential. However, thefield experience is quite limited to date and somedifficulties have arisen. Early CO2 breakthroughhas occurred in some cases and the amount ofCO, required to be circulated through the reser-voir- has been greater than previously thought .93Operating problems such as corrosion and scal-ing can be more severe than with normalwaterflooding. Greater attention must be givento reservoir flow problems such as the effects ofreservoir heterogeneities and the potential forgravity override.

in general, the operating efficiency of the proc-ess or the economics have not been firmlyestablished. In the OTA study, the reportedlaboratory investigations and preliminary fieldresults were used as the basis for the recoverymodels and the economic calculations.

Screening Criteria.Technical screening cri-teria were set in accordance with the following:

Oil viscosity 30 API 1,200 psi

Temperature correction to miscibility pressureO psi if T < 120 F.

200 psi if T = 120- 150 F.350 psi if T = 150- 200 F.500 psi if T > 200 F.

This leads to depth criteria as follows (not tem-perature corrected):

< 27 API 7,200 ft27 - 30 API 5,500 ft> 30 API 2,500 ft

This was the same correlation as used in the NPCstudy.94 It is noted that the general validity of thiscorrelation has not been established. Crude oilsin particular reservoirs may or may not establishmiscibility with CO2 at the pressures and tem-peratures indicated. Other correlations havebeen presented in the literature, but they arebased on a knowledge of the crude oil composi-tion. Data on composition were not available inthe data base used in the OTA study, and ageneralized correlation of the type indicatedabove was therefore required.

Oil Recovery ProjectionsOnshore Reservoirs .The recovery model

used was as follows:

where

R = recovery by CO2 process, stock-tank barrels

s =orm residual oi l saturat ion in zoneswept by CO2. Set at 0.08, Nodistinction was made betweensandstone and carbonate reser-voirs.

E m = sweep efficiency of C02 misci-ble displacement. (Em/Evm) wasset at 0.70.

E =v m volumetric sweep efficiency of

the waterflood computed fromprocedure described in appendixA.

The sweep efficiency for CO2 miscible (Emwas determined by making example calculationson CO2 field tests. Field tests used were thefollowing:

SlaughterWassonLevel landKelly-Snyder (SACROC)

Appendix B . 167

Cowden-NorthCrosset

All projects except the Wasson test werereported in the SPE Field Reports. 95 Data onWasson were obtained from a private com-munication from Lewin and Associates, Inc.Based upon reported data and reported estimatesof the tertiary recovery for each field test, sweepefficiency values were calculated. The ratioEm/Evm averaged 0.87. Discarding the high andlow, the average was 0.80. It was judged that thenational average recovery would be less,therefore a value of EmE vm of 0.70 was used forall reservoirs in the OTA calculations.

The high-process performance model assumesthe waterflood residual (SOrw for each reservoir isdetermined f rom table A-1 according togeographic region. This value was used unlessthe volumetr ic sweep eff ic iency for thewaterflood (EVJ fell outside the limits describedin appendix A. The low-process performance wassimulated by reducing the SOrW values in tableA-1 by 5 saturation percent. The same limits onthe calculated values of Evm were used in thelow-process performance model. The recoverymodel (equation 6B) was unchanged except forEvm and SO:W.

The low-process performance model reducedthe EOR for those reservoirs in which the calcu-lated Evm fell within the prescribed limits. WhereE vm was outside the limits, SO ,W was recalculatedusing the l imit ing E v m value. Therefore, for theselatter reservoirs the recovery results were thesame in both the high- and low-process perform-ance models. For C02 miscible, this was the casefor about one-third of the total reservoirs. Theaverage recovery for all reservoirs was 20 percentless in the low-process performance case than inthe high-process performance case.

Volurnes of Injected Materials.The CO Z r e -quirement was established as follows:

Sandstone Reservoirs26 percent of pore volumeCarbonate Reservoirs22 percent of pore volume

Conversion of CO 2 from surface conditions toreservoir conditions was assumed to be:

2 Mcf C02 (std. cond.) per 1.0 reservoir bbl(A constant value was used.)

Twenty-five percent of the total CO2 require-ment was assumed to be from recovered, com-pressed, and reinfected gas. Seventy-five percentwas purchased.

The C02 injection schedule was as shown intable B-1 6. The water alternating gas process wasused. The ratios were:

S a n d s t o n e s I: 2 C O2H 2O

C a r b o n a t e s 1 : 1 C O2: H2O

Table B-16Carbon Dioxide Injection Schedule

I Purchased COZ I Recycled C02Year percent of total* percent of total*1-2 . . . . . . . . . . . 0 03 . . . . . . . . . . . . . 20 04 . . . . . . . . . . . . . 20 05 16 46 : : : : : : : : : : : : : 13 77 . . . . . . . . . . . . . 6 14

l Total refers to total volume of C02 Injected over Ilfc t)t pattern,

Fluid injection occurred over a 5-year period;reinfected C02 was used beginning in the thirdyear of the period, along with purchased COZ.

Timing of Production. The production profilewas set at a fixed percentage of the total recov-ery (as computed by the recovery model above).The schedule is shown in table B-17. All reser-voirs were developed on 40-acre spacing.

Offshore Reservoirs.--Offshore CO2 miscibledisplacement was calculated using a differentmodel than the onshore model. The reservoirs ofthe gulf offshore are steeply dipping becausethey are nearly universally associated with saltdome formations. This has limited effect on theother processes but great impact on CO2 misci-ble. Due to the dip, the CO), with small quan-tities of CH4 can be injected at the top of the dipand gravity stabilized. No production is noteduntil the oil bank ahead of the miscible slugreaches the first producers down dip. The bank isproduced until the slug breaks through, at whichtime the producer is shut in and the slug pro-ceeds further down dip, creating a new bankwhich is produced in like manner at the next pro-ducer further down. The process continues until

168 . Appendix B

Table B-17Production Rate Schedule

for Carbon Dioxide Miscible

Table B-18Gas Injection Schedule

Offshore Carbon Dioxide Miscible

PercentYear of EOR

Carbonates1-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

lo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11 . . . . . . . . . . . . . . . . . . . . . .fi. ......,12,. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

059

1317191410

6421

Total . . . . . . . . . . . . . . . . . . . . . . . . . . .

Sandstones

100

06

19262113

96

Total . . . . . . . . . . . . . . . . . . . . . . . . . 100

the final bank has been produced at the bottomof the formation. Because the integrity of themiscible slug must be maintained, no water injec-tion is contemplated. However, air is compressedand used to push the CO2-CH4 mixture after arelatively large volume of the mixture has beeninjected. Residual oil saturation after miscibledisplacement, S

Orm, was set at 0.08. Sweep effi-c iency , Em, was set at O.80 (i.e (Em/Evm) x Ev m

=

0.80). This is a significantly higher sweep efficien-cy than used, on the average, for onshore reser-voirs.

The fluid injection schedule for offshore reser-voirs is shown in table B-18 and the oil produc-tion schedule is given in table B-19.

Carbon Dioxide CostsWell Drilling and Completion Costs.-Because

of special requirements created by C02flooding,

Year I C02 -CH, I Air

1 . . . . . . . . . . . . . . 0 02 . . . . . . . . . . . . . . 0.25PV o3 . . . . . . . . . . . . . . 0.25PV o4 . . . . . . . . . . . . . . 0 0.15PV5 . . . . . . . . . . . . . . 0 0.15PV

Table B-19Oil Production Schedule

Offshore Carbon Dioxide Miscible

ProductionYear percent of total

1 . . . . . . . . . . . . . . . . . . . . . . . o2 . . . . . . . . . . . . . . . . . . . . . . . o3 . . . . . . . . . . . . . . . . . . . . . . . o4 . . . . . . . . . . . . . . . . . . . . . . . 505 . . . . . . . . . . . . . . . . . . . . . . . 50

Total . . . . . . . . . . . . . . . . 100

the base drilling and completion cost was in-creased by a factor of 1.25 for injection wells.

Compression Costs. Twenty-five percent ofthe CO2 requirement was met from recycledC O2. Compression equipment was purchasedand fuel costs were charged to this recompres-sion.

Carbon Dioxide Pricing Method.The cost ofC 02 is a variable of major importance. Costs ofCO 2 can vary widely depending on whether thesource is natural or manufactured gas and de-pending on the transportation method and dis-tance. In fact, this EOR technique probably hasthe greatest potential for economies of scalebecause of the variability of these costs.

The cost algorithm used in the OTA study was

developed by Lewin and Associates, Inc., and asummary of this analysis follows. Reservoirs wereplaced into one of four categories. These catego-ries are: