●
✎
SPESPE 23952
Horizontal Injection and Production Wells forEOR or WaterfhodingJ*J, Taber and FLS. Seright, NM Petroleum Recovery Research Center
SPEMembers
Copyrlgh! 1992, Sc41y of Petroleum En@neers Ino
Tfrls paper was prepared for p#ewrrlation at the 1992 SPE Permian Baaln 011and Gas Recovery Conference held In Midland, Texas, March 18-20, 1992.
Thie paper was selected for presentation by an SPE Prcgram Commltfee follow!ngreview of informationcontained in an abstract submifted by the author(a).Contentsof the paper,aa proeeoted, heve not been revlawod by the Society of Petroloum Engirreereand are aub@ctto correctionby the author(a).The materiel, se presented, does not nemssarlly reflectany positionof the Societyof Pewo!eumEngineer$, Its officers, or members, Pepera presentedat SPE meetlngaare ?>5jecl to publicationreview by Edllorial Committeesof the Bocletyof PetroleumEn$ineere,Permleeion!0 co+IyIerestrictedto an abstractof notmorethen SDOwords.Ilfuetratbnsmay no:be qlad. The abstractshouldccmtelnconepkuousacknowWgmmtof where and by whom the paper is fxeeented, Write Llbrarlan Manager, SPE, P,O. Box 6S2S.?.?,Richardson, TX 750S2WSS. Talex, 720SS9 SPEDAL.
ABSTRACT INTRODUCTION
This paper describes the potential benefits of using The use of horizontal wells has been increasing verycombinationsof horizontalinjectionand productionwells for rapidly throughout the oil industry as advances in drillingEORprocessesor waWiooding, techniquescontinue. In manyreservoirs,horizontalwellscan
help solvea numberof oil productionproblems, However,inOur results show that a very favorable configuration spite of a tremendousincrease in literaturers&mces, little
occurs when two opposed horizontalwells are drilled from information is available on horizontal-wellapplicationsfixinjectionand productionwells so that the opposedlateralsare EOR methods. A survey of the extensive horizontal-wellparallel in the patterns, and extended until the horizontal literatureis beyond the scope of this paper, but reviews,l”Ssegments almost meet midway between UIc like wells. books,6-10SPEReprintSeries,ll}’2andpublishedarticles(overComparedto five-spotpatterns with vertied wells, opposed one thousand,even excludingnewspaperarticles) show thathorizontal wells can increase injectivity (injection rate per horizontalwellsare still usedprimarilyin problemmaervoirsapplied pressure drop) by as much as a factor of ten, or to solvespecificproductionproblems. Theseinclude !ow-dependingon well spacingand formationthickness. Areal permeabilityformations especiallyfracturedformationssuchsweep efficiencycan be increasedby 25% to 40%, The as the Austin Chalk,13-k0low-permeability as reservoirs,2*horizontal-welladvantagesarcgreatestfor thinformationswith &unusual as sourcessuchas coal-bedmethane ‘u or Devonianwide spacing, and decline signifiesrrtlyfor thick formations
‘5shale,24-6 as or waterconing,27’2sthin formations,29”31and
and/orclosespacing, Also, for a giveninjectiorrpressure,the viscousoil, 2’33prcssum gradient in the bulk of the reservoir CM averageseveraltimesgreaterwhenusingopposedhorizontalwellsthan Mostof the EORactivityhas bec;Iin the w of thermalwhen using verticatwells. This could significantlyimprove
d~ves~“sg where both horizontal injection wells andrecove , primarily for steam stimulation and steam
microscopicdisplacementefficienciesfor EORprocesses,suchas miceHsr/polymerflooding,that are sensitiveto interracial produ&ionwells have been tried. A potentialfor gas EORtensions, projectswith horizontalwellsis indkatd by recentsimulation
and model studies for inert gss’Oand C02 injection,” TheBecauseof the better sweepefficiencies,faster flooding interest in horizontal-wellwaterfloodingis very recent with
rates, and/or lower injectionpressuresthat are possiblewith mostreportsor publicationsappearingin 1991and 1992.424horizontalwells,SHBORmethodsshouldbenefitby theiruse,For example,polymerfloodscan be improvedby the higher Althoughthereis relativelylittlepublishedinformationoninjectivityand lower rates of shearat the injectionsandface, the use of horizontal_ wells, other than for thermalThe advantagesof horizontalwells for CO floodinginclude:
/(a) delayed C02 breakthroughbecauseo the &tter sweeprecovery,the needforpatternsof bothhorizontali@ction andproductionwells,or opposedverticalfmctures,to increaseth
efficiency,(b) the potentialfor maintainingthe MMP in more 8rate of flooding in EOR processeshas been mentionedt”-sof the reservoir with no increase in the ir@tion pressure, Thispaperdescribesthe improvementsin sweepefficiencyand(c) better injectivity at the same pressures, and (d) the floodingrstes that are ~ssible if horizontalwellsare usedforopportunity to convert more pumped producers to flowing waterfloodingor for any EOR methodwhichrequiresthe usewells, ThermatEORwasnot investigatedin thiswork,but the of both htjection and production wells. In addition, thecited references show that horizontal wells have been potentialfor increasedmicroscopicdisplwxmentefficiencyatsucesssfulin aeveratfieldapplicationsand moreprojectsare the faster rates (with no increase in well-headpressure) Isbeingplanned, examinedfor someof the EOR methods.
87
INCREASEDINJECTION AND PRODUCTIONRATESFOR HORIZONTAL-WELLPATI’ERNS
A relationshipbetweenthe rateof floodingandtheamountof oil displacedby surfactantsolutions was observed manyyearsago.sl Ingeneral,fasterdispimment ratesincreasedtheoil recoveryfor almostaIl Moratory experimentswithsurfaceactive agents. Althoughlater work showed that there is aspecific minimum rate for the displacement of oil with
!8,52 Themfom diff~~t possibilitiesforswfactants, floodn at rates above this criticalvaluealwaysproducedmom oil.increasing the rate of flooding ior the average pressuregradient)for surfactantfloodsin the fieldwereexamined. Amajor obstacle to efficient oil recovery was the inefficientradial flowbetweeninjectionwellsand productionwells, Asearly as 1933, studies showed that the available pressuregradientf~rabout75% of a waterfloodpatternwasonly Moftheaveragepressuregradient.53 To avoidhighpressurelossesfrom radial flow near wells, other geometric arrangementswere considered. The two best possibilitiesappearedto beopposed lateral drainhole (now usually termed “horizontalwells”) and opposed vertical fracture& Because !akraldrainhole hadnot beendrilledmorethan MOfeetin the fifties(althoughEastman claimed that they could be extended togreater distances),54and the costs were much greater thanverticalwelIs, the paralleland opposedvertical fractures(tocreatea linear flood)weregiven moreattention,47
Before discussing the benefits of hoti?id injectionwells, it is instructiveto examinethe differemccbetweentheflcodingrates in a normalfive-spotpatternand the theoreticalrate thatcouldbe achievedby a fill linearflood(possiblywithverticalfractures)with the samespacing.
The steady-sta~~rate of water injection in a five-spotpattern is givenby:33
% =LS4.khAp
(M%.: -0.420w )
where:
% = rate of water injection,bblahiayk = permeability,darciesh m ggndthicknegg,ft
Ap = pressuredifferencebetweeniqjectionand producingwells, psi
P = viscosity,cp~ M dis~~ between~, ftrw = rsdhs of injectionwell, ft
The linear tate in the same ..+ts is givenby
In this equation,A is the rectangularcross-sectionareaofthe sandfats betweentwo like wells, In Fig. 1, if the entire
fkce could be opened to the water (for example if preciseverticalfi’acturesreplacedthe lateralholes),A willbe equaltohW, Onthe otherhand, if the formationthicknessis close tothe diameterof the lateralholesshownin Fig. 1, E+. 2 shouldbe a goodapproximationfor thealmost-linearflowbetweenthehorizontalinjectionand producingwells in the pattern.
In Fig. 1, L is %W,and again, A is equalto AWiftheentiresandfaceis open to flow. Sincethe totalrate in the five-spot includesflowin both directionsfmm the horizontalwells(or verticalfractures),the totallinearrate in tkrrnsof h and Wis
Therefore, the linear rate for the five-spotbecomes
(3)
4.52khAqL($ = ~ ‘. (4)
Notethat the rate dependsmlv on the thicknessof the sandfora given k, P, and Ap. Ms, the linear rate wiil becompletely independent of the well spacing in five-spotpatterns.
Eqs. 1and4 canbe combinedto expressthe linearrate interms of the five-spotrote, The result is
( :-0”4+q’($)= 2.93 q5 @ (5)
Eq. 5 was used to prepare Table 1, which shows theamountthat the linearmte exceedsthe five-spotmteat variousspacings, Theoretically,the mtesare equalwhenthe injectionwells are 1.9 ft apart. However,the five-spotmte dr6paoffso rapidly with increasingdistance betwmn wells, that thelinearmte is 8.4 timesas fastas the five-spotrateat a spacingof 10 acres and can be more than 10 times as fast at widerspacings.
The horizontal-weilrate differs only slightly from thelinear mte in very thin formations. To determinethe amountthat the nte differs from the Ml linear rote, M approximatesolution can be obtained by solving the flow problem in amanner similar to determiningthe rate of flow for differentpermeabilitiesin series.
In Fig. 2, flowis radialbetweenthewelland the mdius,r From G, to q , the flow will be linear, and fkom~a to/w; the flowagain)s radial. ‘Ihe samepermeabilityexists inail regions,soq canbe expmsaedin termsof thepressuredropacross each section. In addkion, the total pressure drop0, - PJ till O@ the sum of the Pressuredifferentialsin
I theothti three s&ions or
(Pt-P4) -63- Pa)+ @- P9)+(PS-P4). (6)
w
●
●
SPE 23952 J.J. TABERANDR.S. SERIGHT 3
&“ = x kL .
4r@ +uLrw
(9)
s61utioI~usingour symbolsis
RearrangingEq, 10 to the same form gives
2r#wApqm m ~,
jcJ +21n325 rw
The flowin the entirematrix can be expressedas
wherek’ is tileaverageor effectivepermtillity if the flowisassumedto be linearwith the end sectionsshowinga reducedpermeabilitycausedby the radial flow patterns. qm is therate for the combinationof radialand lincaxflowbetweenthetwo lateralholes.
In Fig. 2, A is actuallyW(2r,)(neglectingwellradius),so~, 7 is rewrittenas
fi” Wr,(pl - P4)9HW= (8)
pL “
The flow for each of the other sections is written insimilar units and each equation is solved for the pressuredifferential. The resultsalongwith Eq. 8 are substitutedinto~. 6 and solvedfor k“. The result is
Substitutionof this valuecf k’ into Eq. 8 and multiplyingby 2 to give the total flowin both directionsyields
Eq. 10is an approximationof the flowcapacityof a five-spotpatternwhereinthedistancebetweenlikewellsis spannedby lateral holes, After obtaining the above equation, J,E.Warrennotedthat theproblemis comparableto Muskat’sline-drive flood56if thedistancebetweeninjectionwellsis replacedby the formation thickness iii Fig, 2, Muskat’sline-drive
(11)
(12)
Eqs. 11and 12are similar, as long as L is muchgreaterthan the formationtldcknew To determinehow closelythe
flow betweenthe lateral holesapproachesa linear flood,qr~~ be eXpltSXdill tCM’lSOf q’ (* substituting2f, tid%h for W and rI in I@ 12). ~e result, after changingtologlo, is
‘m=q’(o[%L+4;:*(?)} ’13)
The approximaterelationshipin Eq. 13 shows thatq~and qL@am almost equalfor very wide spacingand/or thinformations. Thus, if five-spotpatternscan be drilledwith theopposedand parallelhorizontalwellsas shownin Fig. 1, thelinear rates in Table 1 may be taken as the theoreticalmaximumrates for very thin formations. For thickpay zmesand/or close spacing, the horizontal-well rate departsappreciablyfrom the linear-flowmaximumas more of theinjectedfluid followsthe radial-flowregime.
Fig, 3 was prepared to provide a direct comparisonbetweenthe flowratesof horizontalwellsandverticalwellsinfive-spotpatterns, Horizontal-wellrateswerecalculatedwithEq. 11 (from Muskat) and compared to the five-spotrates(Eq. 1) for differentspacingsand as a functionof formationthickness. The flow-roteadvantageof horizontalwellsfor thinformationsand widespacingis clear in Fii. 3. For example,for ten-ft formations,the horizontal-wellflowrates sre eightto ten times the mtes for the usual five-spot, vertical-wellpatterns. The advantage of horizontal wells is reia!ivedyinsensitiveto increasingformationthicknessuntilthepayzonesexceed100feet. Evenwith KN10-footformations,horisontal-WW flow ra&a are double those of the usual five-spotwaterfloodingpatterns unless the spacing is ckxwr than 20acres ‘fliehorizontalnet mte couldbe even Wter than thatindicatedby Fig. 3 if the water entersthe formationt%omtheverticalportionof the well along with the water enteringthelateralhole. For the simplegeometryin Figs. 1 and 2, it isassumedthe vertical portion of the well is cased and waterenters the lateral hole only, The potential for improvedoilrecoveryandbetteroperatingconditionswiththesefastermtes(atno increasesin injectionpressure)willbe examinedin latersections.
SWEEP EFFICIENCY WITH OPKMID HORIZONTALINJECTION ANDPRODUCTIONVb~.LLS
In thissection,wecomparesweepefficienciesexpectedinopposedhorizontalwells withthose in verticalwells, Craig57has extensivelyreviewedareal‘sweepefficienciesfor variouspatterns of unfractured vextical wells. In a developed,homogeneous five-spot pattern with a unit-mobilitydisplacement,the areal sweep et%cieneyat breakthroughisabout70%, accordhg to mostauthors. For example,Muskatpredcted a sweepefflci~cy of 71.5% for verticalwalls in ahomogeneousk-spot pattern.%
I Muskat also developed an quation to predict sweepefficiency for a unit-moMlitvdisplacementh a direct-linedrive. I; Eq. 14, we have- Muskat’sdirect-linedriveequation to estimate weep efficiency (Zf~ for opposed
80
.●
HORIZONTALINJECTIONAND PRODUCTIONWELLSFOR EOR OR WATERPLOODING SPE 23952..-. ——-. .-. .— -..-——. . . .
horizontalwells as a fimctionof formationtldcknesa(h) and Wick~Oi~ withbSrriCSSto VCltiCd flOW. WIS will be anpatternsize (L). importantarea for futureresearch.
E*= I-* (14)USE OF HORIZONTAL WELIJi 70 IMPROVEDISPLACEMENTEFFICIENCY OF EOR PROCESSES
Eq. 14 was used to generate Fig. 4, which comparesanticipated sweep eff’cicncies (at breakthrough for unit- AllEORor waterfloodingprojectsshouldbenefitfromthe
mobility displacement in homogeneous formations) for faster injection-productionrates and better sweepefficiencies
opposedhorizontalwellswith thoseforverticalwellsin a five- thatare possiblewithhorizontalwellpatternssuchas shownin
spotpattern. In this idealizedmmparison, sweepefficiencies Fig. 1. However, for those processeswhich exhMt higher
for the vertical weils are independentof well spacing and recoveries because of improved microscopicdisplacement
formation thickness. In contrast, for the horizontal wells, efficiencies at higher rates, the horizontal injection wells
sweepefficiencydecreaseswith increasedformationthickness shouldprovidea significantadditionalbenefit.
and with decreasedwell spacing. Using dpposedhorizontalwells, sweep efficiencies can exceed 90% for formation In Fig. 5, the EOR methodsare arrangedin order of the
thicknessesup to 100 fl. However, as formationthicknessAPI gravitk of the oils to be dispkwcd.w The figurecanbe
increasesabove300 ft., sweepefficienciesquicklyfalIbelow usedas a verygeneralscreeningguideto matchtheoil gravity
thosefor theverticalwells,especiallyfor tighterwellspacings.(or correspondingoil viscosity) to the moat effectiveEORmethod. The rangeof APIgravitiesfor mostof theEOR field
Much has been published on the effects of vertical projectslistedin recentsumeys’sare givenby the limitsof the
fractureson areal sweepefficiency.S8-71In ~n~pt, PSdkl solid lines whichenclosethe nameof each method. A rough
verticalfracturesin both injectionand productionwellscouldindication of the importance of the method (in terms of
provide areal sweep efficienciesapproaching 100% if theincrementaloil production)is shownby therelativesi= of the
fractureshave the proper orientation, length, and height, so letters.
that flow is linear between opposed vertical fractures.Unfortunately,fractureorientation,len th, andheightareoften
#Pig. 5 indicatesthatadditionalbenefitsshouldbe observed
difficultto controlduring fracturing. With recentadvances foreachof the threegeneraltypesof EORmethodx miscible,
in the precise placementof horizontalwells,m’” the use of water-basedchemical, and thermal. l’he masons for the
opposed horizontal wells may now offer advantagesover improvementsthat accompanythe faster rates and/or lower
vertical fractures, with respect to improved mwd sweep pressuresthat are possiblewith horizontalwellsare examined
eftlciency.in the next sectionfor someof the EORpmcewea.
Previous analysesof the impact of fractureson sweepefficiencycan be of valuein assessingthe meritsof horizontal The use of }oriamtsl injection wells can improveversus vertical wells. When comparing fractured versusunfractured vertical wells, some importam published injectivityand reduce polymer degradationduring polymer
conclusionsare:5s*61 floods and other chemicai floods. Becauseof the viscousnatureofpolymersolutions,irijectivhyinunfiactumdwellscan
1. Fractureorientationcan stronglyaffixt sweepefficiency. be substantiallyleas during a polymer flood than during awaterflood.’c ‘l%ehigher injectivitiesallowedby horizontal
2. The impactof fractureson sweepefficiencyincreasesas injectionwellscan help to alleviatethis problem,
the mobility ratio increases. In particular, for a givenadverse mobility ratio, the presence of an opposed For a given injectionpressure, the fluid velocityat the
vertical-fracturewell pattern would improve the areal wellboresandfacecan be significantlyless in a horizontalwell
sweepefficiencycomparedto that for unfracturedvertical thanin a verticalwell. Pig. 6 demonstratesthat tMsdifference
wells. is most pronouncedin thin formationsand with large wellspacings.The reducedsandfacevelocityallowelby horizontal
3, The greatest impact on areal sweep efficiencyoccurs injectorscould significantlyreducemechanicaldegradationof
whenbothproducersand injectorsare fractured,whileth~ polymersolutions.
smallest effect occurs when only the producers arefractured. The higherinjectionratesassociatedwithhorizontalwells
couldalso help to mitigatethe effectsof chemicaland thermsd
4. The ratio o! fracture len&#hto interwelldistance, x~L, degradationof polymersand other chemicalsduring floodsin
mustbe greaterthan0.1 (andsometimesgraster than0.4) high-temperatureresmoirs, Sincehigherinjectionrates lead
in walerto significantlyaffectsweepefficiency. to lower maidencetimesfor injectedfluids, the requirementsfor long-termstabilityfor the chemicalscan be relaxed.
By analogy, we anticipate that similar conclusionscan bedemonstratedfor horizontalwells.
Im -
In stratifiedreservoirs, the performanceof a horizontal In this section, we discuss the potential for increased
well will be affectedby its verticalplacement,especiallyin micro=pic displacementefficiencywhen the faster rates of
w
.,
SPR 23M2 1.1. TAFIER AND R.S. SEIUGHT s-. - ----- ---- . ..-—. ---
horizontalwellsare used for those ~ that produceoilby virtue of low oil-watexinterfacialtensions(IIT).
Theaforementionedrelationshipbetweenthe floodhg rateand oil recoveryby surfactantsolutionsis often correlatedbya dimcnsionleasgrouptermedthecapillarynumtm, designatedNCA,Nvc, or Nc. The numberis usefid for expressingthemagnitudeof the ratio betweenviscous and capillary forcesduring the displacemat of one phase by another in porousmedia, such as water, polymer, or surfactantfloods. Fig, 7showsexamplesof experimentalandcalculated(byStegemeier)capillarydesaturationcurves.48,52,77,78 The oil Sahmjtions~.normalizedso that the displacementsthat started at high oilsaturations (continuous oil) can be compared to thedisplacementsof residual oil. Many capillary desaturationcurves for water or surfactantsolutionshave appearedin thelitmture (for examples,see Refs. 48, 52, 77-83), All of thecurves for the displacementof residualoil, by increasingthecapillary number, have characteristics similar to theexperimentalresults shownin Fig. 7. For adstones similarto Bsrea, residualoil displacementstarts at a “critical”NCAvalue of 1V5-10A,and about 50% of the oil is usuallyrecoveredby an order of magnitudeincreasein the capillarynumber. At least95% of the oil is normallyrecoveredif theNCAvaluereaches103 to 10-2.The highNCAvalue? requiredfor tMs improvedoil recoveryare achievedin the field byadding surfactantsto reduce the oil-brineinterfaciaitension,sincefieldfloodhg rates (andpressuregradients)are normallysignificantly below that required to be in the criticaldesaturation region in Fig. 7. Maintaining low IFI’sthroughout the resmoir is difficu$tbecause of adsorptionandlordilutionof componentsin the surfactantslug. Tlds &tpartly explainswhy oil recoveryhas beamvery disappointingfor many micellar/polymer field projects. However, ifhorizontalwells could be used to increasethe floodingrate,and thus the capillarynumber, some increasein oil recoverywould be expected. For example, a tenfold increase infloodingrate should double the oil recovery for a dynamicsystemthat has an NCAvalueof 3 x 1~ beforethe increaseinrate (see Fig. 7).
Laboratorydata for the displacementof residualoil from15sandstoneswithpermeabilitiesrangingfrom40 to 2,190 mdare presented in Table 2.*1 Although the dimensionless“critical”capillarynumbers(theminimumNCAvalueat whichresidualoil is displaced)are similarfor the differentracks, the“critical” pressure gradient per unit of interracial tension(AP/La) was much higher for the low-pernmkdlityrocks.Sincehorizontalwellscamraise the avemgepressuregradientsin the reservoir, horizontal wells could extend surfactantfloodingto tighterSandstones.
The greaterareal sweepand fmter floodingnatesthat arepossible with horizontal wells can provide a number ofeconomic and operational improvementsfor two sodled“miscible” EOR method~ i.e., C% flooding andhydrocarbon-misciblemethods. The hydrocarbonpmcesseainclude high-pressure(vaporizing)gas driva and enriched(condensing)gas drives. The horizontal-wellbemefitaapplyprimafilyto thoseprocesea whichgeneratemiscibilityin the
.—
i-.
resemir by one of he multiple-contactpromwes. CQfloodingis the fasteatgrowingEOR methodtmla~7s*U andmuch of the following discussion of the horizontal-weilimprovementsappliesprimarilyto COpiscible flooding.Thehorizontal-well advantages and operational improvemcatsinclude:
- Betterareal sweepand delayedbreakthroughof C%.- Increasedinjectionratea(withno increasein pressure).- Mhdmummiscibilitypressure(MMP)maintainedinwider
area of reservoir.- More productionwellsconvertedto flowingwells,- Better oil recovery in near-miscibleams (just below
MMP)becauseof higherNCAvaluesfromfasterinjectionrates.
Importantaspects of these five projectedimprovementswith horizontal wells for C02-miscible projects will bediscussedbriefly.
. .and ~ .
Althoughmuch of the flush oil productionin C% floodingcomesafter C02 breakthrough,the early gas breakthroughatproductionwells adds to the projectcwts and eauseavariousoperational problems. Increaxcl areal sweep efficiency,illustratedin FVg.4, Smu!ddelay the C~ breakkmgh fornew C% projects that use horizontalinjectionwells. If theoperatormwtdecidebetweenhorimntd h@etoraorproducecs,horizontalinjectionwellsshouldprovidethemostimprovementof are4 sweep. TIds suggestionMOWSfrom the work ofBargas and Yanosik?s who inveatigetedthe u of verticalfractures to improve sweep efficiencyduring dispkemenbwith unfavorablemobilityratioe.
.Fig. 3 shows that injection rates for C% floods can beincmsed several fold with no increasein wellheadpreaaurewhenhorizontalwells are used insteadof verthd-well, five-spotpatterns. Injectivityis oftena seriousproblemin tertiaryC% floods. Althoughoperatorshaveexperiencedinjectivitiesthat are both higherand lowerthan predkted from waterandC02 viscosities(or mobilities),w-ma loss of injectivityforborhC~ and water, especiallyafter switchingWAGcycles,is common. The potentialfor faster injectionrata, with noincrease in pressure, could mitigate various injectivityproblems,especiallyfor thoseC02 floodswhichare operatedwitha pressureconstraint,i.e., at pressuresnearthe formationpartingpressure.
.~. operatorsshouldbe able to maintainmoreuniformpressuredistributionsin C% floods that utilize horizontal Wells. Because lesspressureis lost nearthewellbore,hi@erpreasutea(ator aboveMMP) can be achieved in a much broader area of thereservoir,perhapswithevena loweriqjecdonpressure, MostC~-miscible floods are operatedai i~ection pmasum wellabovethe MMP to try to includees muchof the redemir aspossiblein the multi-contact-miscibleflowregime, Horizontalinjectionand productionwells shouldreducesignificantlytheamountof over-prasure required, Again,if it ia necemry todecidebetweeniqjectorsw producers,the switchto horizontal
81
HORIZONTALINJECTIONAND PRODU~
injectionwells shouldb ernoreusefulfimmaintain”g MMPuniformlyat lower injectionpmaures.
possibilityfor much hater floodingrates, with no incr&e inwdk!ad pmsaure, should pCOVideqaators with moresh@&~~~&m@ producers to flowing wells for
.
of ~A~. Shyeh-Yunghas shownthatthenear-miscible
pressureregionsmaybe very importantin C% displacementsof oil.m In laboratory studies from outcrop samples, sheshowedthattheC02 mobilityis lessat pressuresjust belowtheMMP, and that oil recoveryis quite good, i,e., only a deless than the high recoveriesabovethe MMP, She attributessomeof these high oil recoveriesto the very low IFT valuesand attendanthigh NCAVSIUCS at pressures~ the MMP.Therefore, one would expect even higher recoverb at thehigher rates possible with horizontalwells, if the IFh andC02 displacementrates are already in the sensitivepart of acapillarydesaturationcuwe (asshownin Fig. 7). Shyeh-YunghascomparedC02 displacementwithswfactant-fkoddisphce-mcnt (see Fig. 4 of Ref. 90) and finds that more oil isdisplacedby C% thanby surfactantsat theequivalentc@larynumbers shown. However, it appearsthat she did not vwyNCAvalues by floodhg at differentrates, but obtahwdbertiues by calculatingthe NCAVSIuefor the IFI’s ~td ti
different floocthgpressuresnear the MMP. To makea directcomparison,experimentsat different rates at each pressurewouldbe hdpfil!.
Although C% displacementshave been performed atvarious rates above the MMP, we arc not aware of C%experimentswhere the displacementrates havebeenvariedatpressuresbelow, and at, the MMP. However,oil-recovery-versus-rateexperimentshave been conductedwith all-liquidmiscible displacements(alcoholdisplacingoil with connatewater resent),and oii recoveryimprovedmarkedlyat higher
7rates.9 The higheroil recoverieswereattributedto reducedtrappingand/or increased~ displacementof the oilphaseas miscibilitywas approachedwith the oil-waterIFT atvery low valu~, tmdthereforeat even higherNCAwIues, ssthe displacementrates were increased
Wepostulatethat similarincreasedoil recoverywouldbeexperiencedwith faster mtes for C~miscible displacementsnear the MMP. Stem92did observea smalldecreasein SO(an increasein oil recovery)when going from a rate of O%ft/day to 1.0 R/day, but this was followedby an increaseinS- whenthe rate was raisedto 5 ft/day. Watkins93showedan evengreateradverseeffectof hi@er mteson oil recovery.However,both Stem and WatkinsperformedtheirC02 floodsat pressures much higher than the MMP, where thecontribution from high-capillary-number, immiscible-displacementmechanismswould be reladvely less, and thecontributionfkomextractionwould be greater.m Thus, thecompetingefftxXs-improvedrecovery from high NCA,anddecreasedrecoveryffom poorer extractionat high mtes-canexplain the optimumrmvery obsemd by Stern at 1 R/day(based on only three different rates) even though his
..
NWELLSFOR EOROR WATERPUXIDINQ SPB239S2
experimentswerewcuabove MMP, bcuuaetherate waaatillslowenoughfor efhctive extracb,
Therefore,untilmoreco2displacementsareccmductedatmanydiffbremtrates (qeeially at pressuresnear the MMP),adefinitive conclusioncannotbereachsd ontheeffectofmteon the microscopicdisplacemente!llciencyof C% floods. Inany event, if the optimumconditionsof pressureand floodhgrate are ever determined more precisely fm C% floods,horizontal wells would be very useful to establish thoseoptimumconditionsin the field,
In a recentarticle in the SPE TechnologyToday series,Joshi pointed out thatmost of the applicationsof horizontalwells for EOR to datehavekm for steamprojects.M I&eatreviews and other publicationsexplaknhat horizontalwellsshouldprovidethe sametypesof befits for thermalrecovmyas for waterfloodingand other EOR methods, i.e. bettersweepefficiencyand faster injection/productionrate3.~3”39Inaddition,horizontal-wellsystemscan be devisedto permit theinjectionof the requiredheat into moreparts of the reservoir,and in a shorter time than with the usual vertical wells.However, reservoirs with viscous oils present differentchallengesand opportunitiesfor horizontalwell systems. Forexample,thermalprojectshaveusedmorehorizontalproducingwells than injectors, whereas we anticipate that horissmtalinjectors should be nwm beneficialthan produem @r mostotherEORprocases. Oneproblemwithlonghorizontalsteaminjectorsis the condensationof steam in the oold part of thedlwti~a~ti titiatimmlyatimofthe injector’sfull length. Preheatingthe longwellboreshouldreduceor eliminatethis problem.x
Eecause of the need for early heat to reduce the oilviscositybefore it can flow at a ussfbl rate, many differentpatternsand arrangementsof hozixmtalwells havebees trialor suggested,eqccially configumtionsdesignedto place thehorizontal injectors and producers close together. Foradditionalinformation,the reader is referred to some of theliteratureon the subject,32-39~WM
HORIZONTALWELU3 lWR WATERFLOODING
Althoughmost of this papsr deals with the potentialofhorizontal wells for EOR methods, the benefits of usinghorizontalwells for waterfloodingshouldnot be overlooked.Manyimprovementsin waterfloochgopemtionsshouldcomefmm the fasterinjectionmtesandbettersweepefficicociesthatarepossiblewithhorizontalwellsas illustratedh Figs. 3 and4.
The faster mtes, shown in Fig. 3 for horizontal-wellpatterns, should provide significantoperatingand economicadvantagesfor waterflood, especiallyif the horizontalwellscan be drilled in thinner formations at wide spacing fixreasonablecosts. (A recent survey showed that horizontalwells cost only 17% more per foot even thoughthe averagetotalmeasured‘depth’ of the horbntal wellwasalmosttwicethat for vertical wells,97) We emphasii that the increasedfloodingmtes shownin Fig. 3 are achievedwith no increase
U2
orxa WZae ● .a. B m- ruvu W.U, u—v.. a
in pesaure, i.e. the rates are comjxucdto standwdfive-spot horizontalwells fromanewinfill welltoapprmch thepattemrates atthesame pressureandspacing. typeshownby Fig.1. Inthiscsse, thelateral segmentswotdd
notneed toextend as fartowsrds theexisting h@tionwellsSinccrhe flowrate isdh’ectly propordodb presaum torecavermost of thepattern-bypassedoil, Carefideconomic
differencesbetweenthe bottom-holepresums of the iqjection studies are needal @ detedne if any of these retrofitandpKXhKtiOnWCh, Fi. 3 couldSdSObe USed to estimate the possibilitiescould recoveroil profitablyfrom existingfloods.lower pressures that would result for a horixmtal-well At present, it appears that wme type of horizontal-wellwaterfloodcompared to an equivalmt five-spotflood at the application should compare very favorably with the infiUasme rate. In general, the pressure requiredfor horizontal- drillingof verticalwells, eitherin newor maturewatcsfloods.well floodiig at the same rate as an equivalentvertical five-spotpatternshouldbe the reciprocalof the ratentio shownonthe ordinateof Fig. 3. For example,in a 10-ftformationat
CONCLUSIONS
80-acre spacing, the prmure required to maintainthe sameinjectionrate with horizontalwellswould be about one-tenth
1. The fiwterfloodingratesand improvedsweepefficicmcies
the presswerequiredfor normalfive-spotfloods. Maintainingthat are possiblewithcombinationsof horizontalinjection
equivalent mtes at lower pressures would have many atld production wells should be very beneficial for
advantages for waterflood operators. For floods wherewaterfloodngand the followingEORpmcessew
injection rates are limited by !he parting pressure of therecovery, C02 flooding,hydrocarbon-misciblefloodhg,
formation,much higher rates couldIMachievedat acceptablemicdkidpolymer (low IFT) floodhg, and polymer
pressures. In practice, some combinationof lowerpressuresflooding. An espeddy favorable configuration ofparallel and oppscd horizontalinjectionand production
and faster rates would probablybe selectedas the optimumcondhionto maximizeoil recoveryand profits.
wells is ahownin Fig. 1.
2. Injection/productionratescanbe increasedby as muchasten times (with no increase in pressure) by usingcombinationsof horiimntslWction and wduction wells
Figures 3, 4, and 6 show that the Wan- ofin tldn formationsand at wi& spacing. *advantage of
combinationsof horizontalinjectionand productionwells forfaster mte3 with hori20nadwells (comparedto vertical”
Waterfloodingare greatest for patterns with wide spacing.wellpattems) deGreUa fixthUerfbrmWna andkr
Therefore, equivalent (or better) water injection and oilcloser well spacing.
pfOdUCtiOi mk can b achievedwith ti h horizontalwellsthanwithverticalwells. Althoughindividualhorizontal
3. Compared to vertical wells, fewer horizontaliqjectiord
wellscostmore, the totaldrillingcostscouldbe muchless thanproductionwellsare neededto maintainthe fast flooding
for vertical-wellmtterns becausefewerwellsare drilledat therates possiblein patternswith wide spacing.
wider spacing. From Table 1 and Figures3 and4, it appearstialymbbti~ltid~~inp~oftit10 verticalwells to achievethe sameoverall floodingrate inthin formationsat wide spacing. However, a ratio of onehorizontalwellforeachthreeto eightverticalwellsis probablyabetter estimateto achievetheequivalentproductionratesandbetter sweerIefficienciesin thickezformations. Oneopemtor,
4. Significantincrease-sinarealsweepefkiency arepossiblewithhorizontalinjectionandproductionwells. l%esweepadvantageis greatest (up to 99% areal sweep)* tldnformationsand wide spacing, For very tidck formationsand closer spacing, it appears that there will be noadvantageover vertical-wellpatterns,
who has &ady drilled at least one horizontalinj=-tor and 5, The fMer rated possible with’horizontal wells shouldproducer in SoutheastTexas, estimates that “two horizontalwellscouldtakethe placeof six vemicaiwells,Wowin higher
hincrease the microscopicdisplacementefficienciesforEOR methods such as micelhdpolymer (surfhctant)
injection/productionrates and boostingoil recovery.’ flooding, that show increased recoveries at higher
‘lTie sweep-efficiency advantage of horizontal-well,floodingpatternswouldbe observedbest for newwaterfloodor an expandeddevelopmentof existing floods. ‘l’hehighersweep efficiencies shown in Fig. 4 (especially for thinformations)should delay water breakthroughfor new floodsand providemuchhighernet oil recoveryin a shortertime,
Forexistingwaterflood, infilldrillingis showingpromiseas a goodmethodfor recoveringthe oil that is bypassedby thepoor areal sweep of normal, vertical-well waterhodingpattems.9*”’~ Horizontal-well systems should comparefavorably with new infill wells, even in mature floods,especiallyif thecostaof horizontalwellscontinueto fhllwhilethe techniquesbecome more versatile. Mathods are now
availablefor drillinga horizontalwell froman existingwell sothat a pattern geometry similar to Fig, 1 cwld be achievedevenin olderwaterflood. Analternativewouldbe to drill two
6.
7,
8,
capill~” numbers.
Polymer floods should be improved by the higherinjectivity and lower rates of shear at the injectionsandface (see Fig. 6) that are possible with horizontalwells.
Advantagesof horizontalweM for C% floodinginclude:(a) delayedC02 breakthroughbecauseof thebettersweepefficiency,(b) the potentialfor maintainingthe MMP inmore of the reservoir with no increase in the injectionpressure, (c) better i@ctivity at the samepressures,and(d) the opportunityto convertmorepumpedproducerstoflowingwells.
Further research,development,and economicsmdiesareneeded to determine the most beneficial ways for theapplicationof horizontal-welltduwlogy to all i@ntionmethodsand EORpmcesaes.
●
●
8 HORIZONTALINJECTIONAND PRODUCTINOMENCLATURE
A.
Eh =
k=
k“ =
h=
L=
NCA sc
P=
q =
%r =
qL =
qL($ =
(?s =
r, =
rW =
s- =
w ~~
Xf =
v=
a=
CrOSS-SdOld a Of the sandface,ft
areal sweepefficiency
pernwibility,darcies
averagepermeability,darcies
formationthickness,ft
dMancebetweenopposedhorizontalwells infive-spotpattern(Fig. 1), ft
dimensionlesscapillarynumber(alsoNc orhkmAp
Nvc), definedas — or ~ (severalotherULU a
forms listedin Ref.“49)
pressure,psi
flOW rate, bbhkiay
flowrate betweenopposedparallelhorizontalwells, bbhdday
linear flowrate betweenparallelfacesof a reservoirblock, bbhhy
linear flowrate in five-spotpattern, bblsklay
water injectionrate for five-spotpattern,bbls/day
radius to externalboundary,ft
wellfkceradius, ft
misciblefloodresidualoil saturation
distancebetweenlike wells, tl
il; cture length, ft
visawy, cp
interracialtension(F17 betweenoil and wateror betweenoil and’C02-richphase,dynekrn
REFERENCES
I
1,
2.
3.
Jurgens, R., Bitto, R,, Henderson,B., White, C,, andMullins, G.: “Honz.ontslDrilling and Completions:AReview of AvailableTechnology,Part 1 - Short- andMedium-RadiusHorizontalDrilling,” PEI (Feb. 1991)63, No, 2, 14-21.
Jurgcns, R., Bitto, R,, Henderson,B., White, C., andMullins, G,: ‘HorizontalDrilling and Completions:AReviewof AvailabicTechnology,Part 2- Medium-andLong-RadiusHorizontalDrilling,”P&f(March1991)63,No. 3,32-37.
Leaf, R.C. and PM@ F,J.: ‘A Reviewof HorizontalMethods and Drilling Technology,”paper SPE 21862presented at the 1991 SPE Rocky MountainRegional/Low Permeability Reservoirs Sympsium,Denver, April 15-17.
i WELLSFOR EOROR WATBRFLOODINQ SPE 239S
4,
5.
6.
7.
8.
9.
10,
11.
12.
13,
14,
15.
16,
17*
Nods, S.0., Hunt, J.L., and Solhnan, M,Y.:“PredictingHorizontalWell Mkmancw A Reviewofcurrent Technolo~,” ~ SPE 21793presentedat the1991 SPE Western Re@owdMeeting, LOW Beach,March2s22.
Nevile, G., Smith, P.J., Terdor, D.G.V., and Banks,D.: “AReviewof theSuccessfulUseof Horizontalwellsin the Developmentof the UKCS Cyrus Field,” paperSPE 23742 presentedat the 1992SPE htin AmericanPetroleumEngineeringConf~cc, Caracus,Vcmezucla,March8-11.
Joshi, S,D.: ‘HorizontalWell Technology,”PennWellPublishing,Tulsa (1991).
French Oil & Gas Industry Association TechnicalCommittee:“DirectionalDrillingand Deviition ControlTechnology,”GulfPublishingCompany,Houston(1990).
Crouse,P.C.: ‘EconomicAnalysisof HorizontalIXillingInvestment,”GulfPublishingCompany,Houston(1991).
Aguilcra, R,: “Volume 9: Horizontal Drilling,Completion,andProduction,”GulfPublisMngCompany,Houston(March 1991).
“World oil’S mbook of Horizontal Drilling andCompletion Technology,” Gulf Publishing Company,Houston(1951).
Aldrich,C.S. et d.: ‘Dkecdonal Drilling,gSPE RepsintSeries#30 (1991).
“HorizontalDrilling,”SPEReprintSeriesNo. 33(1991).
Stang,C,W,: ‘AlternativeElectronicbgging TechniqueLocatesFmcturesin AustinChalkHorizontalWell,” Oil& GusJ. (NOV. 6, 1989)87, No. 45,42-45,
Kuich, N.: “Seismicand Horizontal Drilling UnlockAustinChalk,” Wor&iOil (Sept. 1990)211, No. 3, 47-54.
Dees, J.M,, Freet, T,G, and Hollabaugh, G.S.:“HorizontalWellStimulationResultsin theAustinChalkFormation, Pearsall Field, Texas,” Pm,, 65th SPJ3AnnualTech. Conf, & Exhib,, New Orleans(Sept, 23-26, 1990), Production Operations and EnginAngVolume,665-76;SPE 20683,
Pope, C,D, and Handren,P.J,: “CompletionTechniquesfor Horizontal Wells in the Pcarsall Austin Chalk,”Pmt., 65th SPE AnnualTech. Conf, & Exhib,, New~kans (Sept. 23-26, 1990),hXhlCtiOll opSt’&tiOflS andEngineeringVolume,657-64;SPE 20682,
Stayton, R,J.: “Underbalanced Horizontal DrillingEnhances Production of Austin Chalk Well,” pap&IADC!ISPE19984 presented at the 1990 IADC/SPEDrillingConference,Houston,Feb. 27-March2,
94
●✎
. - -“.”- #.d. a41Dnn nnw she. 0L3nAun 1 7
18. ~y, R.: “HorizontalMWD Tool Developmentand 30. Fa@, R.A., Tmty, J.E., Bmmet, L.R., and Mayo,Applicationin the Chalkand Texas,”pl’SW#ltCdat 1991 M.K.: ‘MWD ReeistivityTool ChddesBit ~tdlyWorld Oil 2nd Annual Horizontal Well ‘L’ethnology in Thin Bed,” Oil& ti if. (Dec. 9, 1991)89, No. 49,RegionalConference,San Antonio,Aug. &7. 62-65.
19, Noble, S,: ‘The Use of Geologic Description for 31. “HorizontalWell to Tap Thin Troll Oil Pay Zone,” 011Horizontal TechnologyApplicadonin the Chalk and & G@J. (June 12, 1989)87, No. 24,26-28.RelatedFormations,”pmaentedat 1991World Oil 2ndAnnual Horizontal Well Technology Re@nal 32. Islam, M.R. and George, A.E,: “sand Control inConference,San Antonio,Aug. 6-7. HorizontalWeUsin Heavy-oil Reservoir,”WT (July
1991)43, No. 7, 844-S3,20, Crouse, P.C.: ‘Resmoir Analysis Abilities and
Economics for Horizontal We..ls-The Chalk, The 33. Asgmpour,S., Springer, S., Pantella,P., and Singhal,Bakken, and Frontier Econor,lics and Screening of A.: %hanced Heavy-oil ~UdiCll by HorizontalCandidates,” presented at tlw 19M World Oil 2nd Drilling,”paper CIM/SPB9G126 premted at the CIMInternationalHorizontalWell ManagementConference, Petrol Sot,/SPE Int’1Tech. MeetingiCalgary,Canada,Las Vegas,March26-27. June 10-13, 1990;Preprints,v, 3, 1990.
21, Peterson, S,K. and Holditch, S.A,,: “Applicationand 34. Joshi, S,D,: Thermal Oil Recovery with HorizcmtalEconomicsof Horizontal Wells for StimulatingLow- Wells,”JPT (Nov. 1991)43, No. 11, 1302-04.Permeability Gas Reservoirs,’ paper SP13 23446presentedat the 1991 SPE Eastern RegionalM@ing, 35. “ThermalResponseRisesa: Horizontal/SteamTeat,”OilLexington,KY, Oct. 23-2S. & (i@ J. (hhy 8, 1989)87, No. 19, 16-17,
22. Ertekin,T., Sung,W., andSchwerer,F,C.: “Production 36. &lhU~UCt, B.C., @ux, A.M., corm, B., and @ittMd,
PerformanceAnalysisof HorizontalDrainageWells for M,P,: “SteamLr@ctionin a Lmv-permeabilityReaenoirthe Degasificationof Coal Seams,”JPT (May 1988)40, Through a HorizontalWell in Lacq SuperieurField,”No. 5,625-32. Pm., 65th SPE AnnualTech, Conf, & Exhib., New
Orkans (Sept. 23-26, 1990), EOR/GeneralPetroleum23. King,G.R. and Eiekin, T.: “ComparativeEvaluationof EngineeringVohme, 5S9-62;SPE 20S26.
Vertical and Horizontal Drainage Wells for theDegasificationof Coal -S,” SPERE(h@ 1988). 37. ‘CanadiansCombineSteamDrive,HorizontalWell,”01/
& ~ J. (June 1, 1987)8S, NO. 22,22.24. “HorizontalHolesMay UnlockDevonianShale,”Oil d
OarJ. (March9, 1987)8S, No. 10, 1?-18. 38. Dletrich,J,K,: ‘l%e Kern River Horizontal-WellSteamPilot,” SPEW%(Aug. 1988)3, 93S-44.
25, Bcllinger,C. and Koziar, Q,: “HorizontalWell in theDevonianShale, MartinCounty,Kentucky,”paperSPE 39. Butler, R,: ‘Developmentand Dwcription of Steam23447 presented at the 1991 SPE Eastern Regional Assisted Gravity Drainage Using Horizontal Wells,”Meeting,Lexington,KY, Oct. 23-2S, preseatcd at the 3rd International Conferere on
Horizontal Well Technology, Houston, Nov. 12-14,26. Salamy, S,P., Locke, C,D,, and Kipema, G.J,: “Prc- 1991,
and Poststimulation Well Test Data Analysis fromHorizontal Wells in the Devonian Shale,” paper SPE 40, Meszaros,(i, Chakma,A,, Jha, K.N,, andIslam,M.R,:23449 presented at the 1991 SPE Eastern Regional “ScaledModelStudiesandNumericalSimulationof InertMeeting,Lexington,KY, Wt. 23-25, Gas Injectionwith HorizontalWells,” Pmc,, 65th SPE
AnnualTech. Conf. & Exhib., New Orleans(Se@,23-27, Ozkan, E, and Raghavan,R,: “A BreakthroughTime- 26, 1990),L?OR/GcnemlPetroleumEngineering,585-98;
Conflation for ConingTowardHorizontalWells,”paper SP1320529,SPE 20964 presentedat the 1990EuropeanPetroleumConference,The Hague,The Netherlands,Oct. 21-24. 41, Ammer,J,R,, Enick,R.M,, andKlara,S.M.: ‘Modeling
the Performanceof HorimMd it@tion Wellsin Carbon28, Herring, T.R,, Martinsen,R,, and Skjaeveland,S,M,: DioxideMiscibleDisplacementMocases,” paper SPE
“ConeBreakthroughTimeforHorizontalWells,”SPERE 21220 presented ●t the 1991 SPE Symposium on(Aug. 1991)6, No. 3,311-18, ReservoirSimulation,Anaheim,Feb. 17=20,
29, Lien, S,C., Seines, K., Havig, S.0., and Kydlam!,T,: 42. McAlpine,J.L. and Joshi, S,D,: “Horizontal-WellPilot‘The First Long-TermHorizontal-WellTest in the Troll WaterfloodTestsShallow,AbandonedField,”Oil &O@Thin Oil tine,” JPT (Aus, 1991)43, No. 8, 914-17& .J, (Aug.5, 1991)89, No. 31,46-47.970-73,
43. ‘HorizontalWaterfloodSche@led in Texas CwbonateReservoir,wOil& GtwJ, (Sept,2, 1991)89, No, 3S, 34,
. .
10 HORIZONTALlNJECI’IONAND PRODUCTIONWELLSFOR EOROR WATERFLOODIN~ SPE 239X
44. Charles, D.D., Star&man,R,A., and Raghavan, R.: S99“Steamtube Modelingof HorizontalWells in Mixed-PatternWaterfloodS,”papsr SP1323451presentedat the1991 SPB Eastern RegionalMeeting, Lexington,KY,Oct. 23-25. 60.
45. Hansen,X.L.: “Applicationfor a HorizontalWell in aWilcoxWaterflood-A CaseStudy,”presentedat the 3rdInternationalConferenceonHorizontalWellTechnology,HouStOII, Nov. 12-14, 1991. 61.
46, ‘Texaco Completes Horizontal Injector in SoutheastTexasOil Field,” 01/ & OtMJ. (Feb. 24, 1992)90, No.8,44. 62.
47. Henderson, J,H. and Tabcr, J,J.: U,!% Patent No.3,199,586(Aug. 10, 1965), 63,
48. Taber, J.J.: “Dynamicand Static Forces Required toRemovea DiscontinuousOil Phase ffom Porous MediaContainingBothOil and Water,’ SPEl (MarchW69)9, 64*3. I
49. Taber, J.J.: “Researchon IMancd Oil Recovery:Past,at, andFuture,”inSu@ce R%encnwnainEnhanced 659OilRecowry (1981)PlenumPress, New York, 44.
50. Taber, J.J,: “IMvironnwmtalImprovemeataand BetterEconomicsh EOR Opedons,* In Situ(1990)14 (4), 66*34s-404.
51. Tabcr, J.J.: “Injectionof De&gent Slugs in WaterFloods,” 7Fons., AIME (19S8)213, 186. 67.
52. Abrams, A,: ‘Tlw Influence of Fluid Viscosity,InterracialTension, and Flow VelocitycmResidualOilSaturationLeftby Waterflood,”SPI?U(1975)1S,437-47.
68.S3, Wyckoff, R.D,, Botset, H.Cl., and Muskat, M.: “The
Mechanicsof Porous Flow Applied to Waterflcoding-Iems,” 7hamT.,AIME (1933)33, 103.
54, Eastman,J.A.: ‘LateralDIidnHoleDrilling,”Petrolewn 698Engr, (Nov. 19s4)27, B-57;(Dec. 1954)27, B-44;(Jan.1955)27, B-111; (’March19S5)27, B-7-
55. Dickey, P.A. ai~dAndresen, K.H.: “The BehaviorofWater-InputWells,”in Secon&vyRecowy of Oilin the 70*U.S., AmericanPetr, Inst., New York (1950)320.
56, Muakat, M,M,: l%ysicafMtciples #’Oil Production,McOraw-Hill,New York(1949)653-724, 71,
57, Craig, F.F,, Jr.: 7he ltesemolr En@cedng AspectsofW@e@oodlng,MonographSeries,SPR,Richardson,TX(1971)3,48-61,
580 Bargaa,C.L. andYanoaik,J,L.: “TheEftkctsof VerticalFractureson ArealSweepEff&ncy in AdverseMolAlityRdO Fkda,” PWW SPE 17@9preaemtadat the 1988InternationalMeetingonPetroleumEngineering,Tia@n,China, Nov. 1-4,
72,
.-
Crawford,P.B. and Collins, R.E,: “EstimatedEf&t ofVertical Fractwes on Secondq Recoveq,” I’ium.,AIME(S954)201, 192-%.
Crawford, P.B., Pinson, J.M., Simmons, J., andLandrum, B.L,: “Sweep Efficiencies of VerticallyFracturedFive-SpotPatterns,’ 27tePam. Engr. (March1956)B95-Bl~.
Donohue,A,To,Hansford,J.T,, andBurton,R.A,:’“TheEffectof InducedVertically-OrientedFractureson Five-Spot sweep Efficiency”,’sPEl (Sept.1968)260-68.
Haggort,J,: “Waterflood-InducedHydraulicFncturing,”TechnisheHogeschcol,Deft, Netherlands(19S0),
Hansford, J.T, snd Donohue, D.A.: “The RelationofMulti-Well Vertical Fractures to Five-Spot SweepEfficiencyat Breakthrough,”Prod. Montldy(1%7) 2-7.
Husky, W.L. and Crawford, P.B.: “PerformanceofPetroleumResewoirsContainingVerticalFracturesintheMatrix,”SPEJ (1%7) 221-28.
Prats, M,: “Effectof Vertical Fractures on ResenoirBehavior-IncompressibleFluidCase,”SPE?(June1%1)105-18.
Mats, M. and tin, J.S.: “Effectof VerticalFractureson Reaerwir Behavior-Results on Oil and Oaa Flow,’7hnr. , AIME (1963)218, 1119-26.
Prata, M., Harebroek,P., and Strickler,W.R,: “Effectof Vertical Fractures on Ressrvoir Behavior-CompressibleFluid case,” 71uns., AIME (1962) 210,87-94.
Pinson,J., Simmons,J., Landrum,B.L,, and Crawford,P.B.: ‘Effect of Large Elliptical Fractures on SweepEfficiencisa in Watsr Flooding or Fluid IqjectionPrograms,”Prod, Monthly(Nov. 1963)20-22,
Simmons, J., Landrum, B.L., Pinson, J,M,, andCrawford, P.B.: “Swept Areas After Breduough inVertkdy FracturedFive-SpothttHIM,” ~., AIMB(1959)216,73-77.
Dyes, A.B,, Kemp, C. FL, and Caude, B,H,’:“EffectofFractureson SweepOut Pattern,”7hIu, , AIME(1958)213,245-49,
Hartsock, J,H, and Slobod, R.L.: “The Effect ofMobility ~tiO and Vertical Fractures on the SweepEfficiencyof a Five-Spot,”Pd. Monthfy(Sept. 1%1)2-7.
Veatch,R.W., Jr,, MoachovidiqZ.A., andFast, C,R.:“An Overview of Hydraulic Fracturing,” RecwatMvances in f$WuuUc FRtctwfng,MonographSeAa,SPE, RiChUdSOIl, TX (1989), 12, 1-38.
-u &.?T,J& a,a. ● nvum na-s~ -*U. Uwulab.—
.,
73, Von FM&m, R,: “Technology Keeps Pace with 88. Prieditis, J., WoIle, C,R,, and Notz, P,K.: “AHorizontalDrilling,’ PEf (Nov. 1991)63, No. 11, 16. Laboratoryand Fieldl@ctivity Study C~ WAGin the
SanAndresFOfmlttiOIlof WestTexas,”Proc., 1992SPE74, Coudeyre,M., Le Goc, B,, and Lombez,LP,: “Ouiding Annual Technical Conference and Exhibition,
a HorizontalWell Using Gwiogic Monitoring,’ World EOR/GeneralPetroleum Engrg, Vol., 595-61Q SPEOil (July 1991)212, No. 7,59-66. 226S3.
75. Moritis, CL: “C02 and HC Injection Lead EOR 89, Roper, M.K,, Jr,, Pope, (LA,, and Sepehrnoori,K,:ProductionIncrease,”Oil & Gas J. (April23, 1990)49- “Amdyis of Tertiary Injectivityof Carbon Dioxide,”82. paper SPE 23974 to be presentedat the 1992Permian
Basin011and Gas RecoveryConference,Midland,TX,76, Seright, R,S.: “The Effectsof MechanicalDegradation March 18-20.
and Viscoelastic Behavior on Injectivity ofPolyacrylamideSolutions,”SPW (June 1983)475-85. 90, Shyeh-Yung, J-G,J,: “Mechanisms of Mkible Oil
Recovery Effects of pressure on Miscibleand Ncar-77. Moore,T,F, and Sbbod, R,L,: ‘The Effectof Viscosity Miscible Displacementsof Oil by Carbon Dioxide,”
and Capillarityon the IXsplacementof Oil by Water,” Pm,, 1992 SPE Annual Technical Conference andProd, Monthly(Aug. 1956)20, h?O, 10,20-30. Exhibition,EOR/GenemlPetroleumEngrg. Vol., S95-
610; SPE 22561,78, Stegemeier, Ci.L.: in Anprowd 011 ?lecove~ by
Su@cMI od PolymerFlooding, D.O. Shah and R.S. 91, Taber,J.J, andMeyer,W,K.: “Investigadonsof MiscibleSchcchtcr(eds,), AcademicPress, Inc., NewYork, NY, Displacementsof AqueousandOleicPhasesfromPorous64 (1977), Media,”SPE? (1964)4, 37-48.
79. Idke, L.W.: Enhan@ Oil Recmwy, Prentice Hall, 92, Stem, D,: “Mechanismsof Miscible Oil RecoveyEnglwood Cliffs, NJ (1989)69. Effectsof Pore-LevelFluid Distribution,”P?uc., 1992
SPE Annual Technical Conference and Exhibition,80, Foster, W.R.: “ Low-TensionWateMlcoding Process,’ EOR/General Petroleum Engrg, Vol., S95-61w SPB
JPT (M. 1973)20s. 22632*
81, Taber, J.J., Kirby, J.C., and Schrueder,F,U.: “Studies 93, Watkins, R.W,: “A Technique for the IAmratoryon the Dh@acement of Residual Oil:Viscosity and Measurementof Carbon Dioxide Unit DispkmnentPermdMty Effects,”~ymposium Series, AIChE, New Efficiencyin ReservoirRock,”paperSP137474presentedYork City (1973)69, No. 127,53-S6, at the 1978 SPE Annual TechnicaJConference and
Exhibition,Houston,Oct. 1-4.82. Gupta, S.P. and ‘hushenski, S.P.: “Mhllar Flooding-
Compositional Effects on Oil Dhplacement,” SPIU 94. Huang, W.S, and Hight, M.A,: ‘Evaluation of(1979)19, 116-28, Steanmocd Procmses with HorizontalWells,” SPERE
(Feb, 1989)4, 68-76.83. Chatzis,L and Morrow,N,R.: “Correlationof Capillary
NumberRelationshipsfor Sandstone,”SPllf (Oct.1984) 95. “Arco Hopes White Wolf Horizontal Unlocks Heavy555-62. Gil,”kfv. RecovwyWeek(June24, 1991)2, No, 2S, 3.
S4. Martin, F.D. and Taber, J.J.: “Carbon Dioxide 96. Butlw,R,M. andStephens,D.J.: ‘The (MwityDrainageFlooding,”WT (April 1992)-in press, of Steam-HeadedHeavy Oil to Pardd Horizoritd
Wch,” paper CIM 80-31931pR$M&d at the 198085. (hod, P,A. and Downer,D,(3,: “CedarCreekAnticline Petroleum SW. of CIM Annual Techniad Meeting,
Carbon Dioxide Injectivity Test: Design, Calgary,May 25-28,Implementation,and Analysis,”Pm,, SPE/DOESixthSymposiumon EOR (1988)S9-78;SPFMXX?17326, 97, Petxet, G,A,: “JAS Pinpoints Cost of Drilling U,S,
wells,” Oil & G@J, (Dec. 30, 1991)30-32,86. Christman, P.(3, and Gore]], S.B,: “A Comparisonof
L40ratory andField-ObservedC% TertiarylqjeMivity,” 98. Gould, T,L, and Sarem, A,M,S,: “Infill Drilling forPruc,, SPWDOESixthSymposiumon EOR(1988)173- IncrementalRecovery,”W?’(March 1989)229-37.1S6;SPE/DOE17335.
99. VSII EV@l’I@Jt, A.Foand Kriss, H,S.: “A P’IOP@to87, Patel, P.D., Christman, P,G,, and (U+rdrteitJ.W.: ImproveRecoveryEfficiency,”JPT(July19S0)1164-6S,
“InvestigationofUnexpectedlyLowField-ObservedFluidNobilities Durin,gSome C% Tertiary Floods,”SPERE 100, Helm, L,W,: “InflliDrillingVS,Tertiary Oil Recovery(lJOv,19$7)!!07-13, vs. Mom imports,”W2’(July 1980)1169-74,
,—--U7
● ✌✌
12 HORIZONTALINJECTiONAND PRODUCTI
101.
102.
103,
104,
Rcatine, J.L, Wma, W.Q,, and Elias, R., Jr,: “InfillDrilling in a SteunfloodOperation:Kern River Field,”SPERE(May1987)243-48.
Wu, C.H., Laughlin, B.A., and Jar&n, M.: “InfillDrilling Enhances Waterflood kovesy,’ YPT (Oct.1989)1088-9S,
Hurst, w,: “Infill Drilling in the RotherwoodField,HarrisCounty,Texas,”SPERE(Nov. 1991)5024)7.
Sarem, A,M, S,, Fuller, S.M., and Gould, T.L.:“ScreeningWaterflood for Infill Drilling opemtions,’paper SPE 22333 to be presentedat the 1992 PermianBasinOil and Gas RecoveryConference,Midland,TX,March 18-20.
NWELLSFOR EOR OR WATERPIXX3DINC3 SPE 2395
Table 1, The htiO of the LinearRate to the Vertica&WelJFive-SpotRate●t VariousSpacings
Spacing(Acres) w (Feet) 9&l%
1,9 1,0
0,625 16s,0 6,71,000 209,0 7,02s00 330,0 ‘?,55NO0 467,0 8,0;O,ooo 660,0 8,42o1100 935,0 8,940,000 1320,0 9*3
80,000 1859,0 9,7160.000 2640.0 10,2320,000 3734,0 10.6
q&($- 2’93q’[10g:-0’42W = distam betweenlike wells
9s = five-spotmte for verticalwells
rw - well radius = 0.33 t%
Table2, The Displacementof ResidualCMfromSandstoneand AiundumCores
Initial Percentof Maximum critical
core core Porosity Permora:;;ity Residual ResidualOil A?/Lovalue of critical
Number Typcx (%) oil Recoveredat (psi/ft) A~ko Valueof(millidarcies) Saturation Maximum (#l) Nti(dydcm) (dyndCni)(%) A P/Lu
2-269 ..... 11,1 40,8 41.7 17,8 52,3 13*MI 1,23 X 1042-205 ..... 13,1 52,1 42,4 9*3 52,8 20.10 2,33 X 10s
D-8-9 ..... 17,3 9sJf 5S,6 4,8 82.6 23,80 S*O5x lo~J-13 Jclm 17,2 189,0 37,1 6!.3 16,30 6483X 104S-12 ..... 17,6 242,0 40,4 ‘1::; 61,9 11.40 6,13 X 10sON-2 ..... 18,3 320,0 33*O 28*2 86.7 4,03 2,87 X 105
TA Tensleep 18,6 381,0 36,6 22,4 56,2 S*I2 4634x 10sBer-1 Mm 19,1 346,() 40,5 2s. 1 46,2 6,13 4,72 x lo~D.1”8 ...-. 19*7 503,0 a 49,5 17.I 26,3 1,71 1,91x 1(Y$SP-6 St, Peters 22*1 533,0 45*3 24,8 IS*2 2,25 x 10sJK-i Alundum 28,2 585.0 34,1 11,7 20.4 ;:: 4*96x 10sSP-5 St, Peters 23,1 613,0 48,6 27,2 30,9 1,12 1,s3 x lo~DOT ..... 29,9 635,0 32,1 29,8 21,s 2*O1 2,84 x 1(Y521X Berea 21,6 752,0 4s,3 26.4 2,78 X @
80 ...... 14,4 1450,0 49*7 34,1 E:; ;:: 2,35 X 104Mci-6 ..... 16,6 2190,0 49,3 48.2 12,2 0!31 1,51x lo~
,.. , . . .‘All coresexceptAlundum(A120) were sandstone, Many were mm otl.pfoauclngreservoirsand the oil companhsaaskedthat the namesnot be released. ~ata fromRef. 81,
R-Jl====f-T—’—~(or 2r@)
Fig. 1. Five-spot patterri vvith horizontalinjection and production wells.
1 4 ?3 P
~ C*—..
‘,,,A— J~’ /+
\.=. ” “’””’:’‘~ol~..””f ,’ -r@. . w
,, ‘,.,, ”.. .’ .,.$,.” . . . . . . . . . . . . . . . . ...”
,..,.
L[Al--’—.. ——
Fig, 2. Segment of parallel and opposedhorizontal injection and production wells.
,- sy&l320— ——__
80I
10 : -%“Q*..**....* ... ,,.O--=JO ““..,, ‘\ I
8“ --% “,5
‘, \‘+ *.
\ \\\ “.,
\ ‘o\ ‘. \
6 “ \ ‘b\ “,
\\
*0,\ ‘.
4 - \ ‘*\=’0 ●*
2
0 I10
=Constant Injectionpressure
20 30 60 100 20’0300 Sw 1,00Formation thicicne$% ft
Fig. 3, Comparison of injection rates forhorizontalweils reiative to verticai weils
70 - Verticai weiist
\i,
8 \,\
,\l \t
I I I I I \80 ‘
10 20 30 60 100 200300 500 1,0(
Formation thickness, ft
Fig. 4$ Cornp8riaonof sweep efficienciesforverticai weiis versushorizontalweiiai(paralleiand opposed in a five-spot pattern, unhmobiiity t’tNiO)
-—. -L.——.-
14 HORIZONTALINJECTIONANDPRODIJCT1OiUwl?l.1.~ I?OR mn m WATURm nnnmufi @nn 9%1
Olln GRAVITY ‘APIBENEFITS POSSIBLE WITH
PIORIZONTAL WELLS
s, Ps,s, P
WATER
[<<
ALKAL13JE
POLYMER FLOODS
S, P, R?
S, R
s, R
!3, P, R?
s, P
s, P
Fig. 5. Possible benefits from ~ontal we IJsfor EOR methods:- Better areal sweep (S),- Improved recovery from faster flooding rates (R),- Lower injection pressures (P).
1
0.6
0.3
0,2
0! 1
0.06
0.03
0,021’/’ Five-spot pattern
Unit mobility ratioHomogeneousformationConstant injection pressure
0!01~ I I I I I I I 1
10 20 30 50 100 200300 500 1,000
Formation thickness, ft
Fig, 6, Comparison of fluid velocities at thewellbore for horizontal versus vertical wells
Fig, 7, Potential for improved oil recovery fromfaster floc+ding rates with horizontal injectionand production wells, Original figure withcalculated capillary number vaiues (anddefinitions of symbols) from Stegemeier;78data from Refs, 48, 52, and 77,
..-