1、ProppantTransport&ScreenoutBehaviorR.D.Barree2009Inthissession Lookattraditionalproppanttransportanditsassumptions Lookatcommonremediesforearlyscreenout Whatfactorsactuallyaffectproppanttransport?2009ProppantTransport&Settling Simplemodelsassume1Dflowandmodelsingleparticlesettling(StokesLaw) Fluidan
2、dparticlevelocityprofilesaremuchmorecomplexin3Dflow. Slurrydensitygradientscausegravityunderrunning,fluidshear,andnonhomogeneousconcentrationprofiles. Fluidpropertiesandleakoffcausetransverseparticlemigration LateralparticlemotionchangestransportandleadstoscreenoutTraditionalPropTransportFrac height
3、(assumed to beconstant)Suspended proppant slurry(uniform concentration)Clean pad fluid tocreate w=3-6xdFracture half-lengthSettled sand bank20092009CommonAssumptions:FluidLoss/Transport/Screenout Proppantishomogeneouslydistributed Vertically,laterally,transverselySandandfluidtraveltogetherPadisrequi
4、redtoopenwidthforsandPadisdepletedbyleakoffScreenoutscausedbypropbridgingPropconcentrationincreasedbyleakoffFalseassumptionsleadtofailedremediesCommonRemediesforEarlyScreenout(Ifcausedbypaddepletionandbridging)PumpmorepadvolumeIncreasepumprateUsehigherviscosityfluidsUsesmallerproppantsUsefluidlossad
5、ditivesSometimes they work, and sometimes NOT!20092009FactorsAffectingProppantTransport Particlevelocityprofileinfracture Concentrationdistributionacrossfracturewidth Slurryviscosityincreasewithsolidsaddition Singleparticle“Stokes”settlingvelocity “Hindered”particlesettling Convectionfromslurrybulkd
6、ensitygradients Proppantholdup ProppantbridgingCum. Particle Count200910.80.60.40.20VelocityDistributionofParticlesBetweenParallelPlates1.20123456Particle Velocity, cm/secFor a uniform particledistribution, the velocityprofile is given by thecumulative frequencyplot.Cum. Particle Count20090.80.60.40
7、.20ParticleVelocityProfilesNormalizedtoFluidVelocity1.2100.20.40.60.811.21.41.6Relative Particle VelocityCv=0%Cv=10%Cv=25%Cv=35%Cv=55%Velocity, cm/secProppantnotHomogeneouslyDistributed654321000.20.40.60.81Normalized Slot Width2009Particles at highconcentrationParticles at lowconcentration2009Single
8、ParticleSettlingVelocityPredictions Terminalsettlingvelocityforasingleparticleinaninfinitefluidbody: StokesLawforlaminarflow AllensEquationfortransitionflow NewtonsEquationforturbulentflow Terminalvelocitycanbemodifiedformultipleparticleinteractions. Walleffectscanbeconsideredfornarrowchannels.( )vt
9、 =1.74d g(s l) 2009SingleParticleTerminalSettlingVelocitiesStokes laminarflowregime18Allen transitionflowregime0.721.180.20d(g(s l) l)0.45lvt =0.5Newton turbulentflowregime0.5 l Settling Velocity, cm/sec20090.10.010.0010.11Particle Diameter, inchesSingleParticleSettlingRatesina1.0cpNewtonianFluid100
10、101StokesAllenNewtonActual1000.011250 40 30 20Mesh SizeSettling Velocity, cm/sec2009SingleParticleSettlingRatesina55.0cpNewtonianFluid1001010.10.010.0010.11Particle Diameter, inchesStokesAllenNewtonActual1000.011250 40 30 20Mesh Size2009SlurrySettlingExperimentsinaVerticalSlotModel Parallelplatemode
11、l5feetx6inchesx0.25inches 30%and40%PEGsolutions 30/50meshand95meshsilicasandslurries Volumetricconcentrationsfrom055%solids SlurrysettlingvelocitycomparedtoStokesvelocitySettling Velocity, cm/sec2009Solids Concentration, Cv10.1SlurrySettlingRatesControlledbyBulkDensityGradients1001000.10.20.30.40.5V
12、meas(95)Vmeas(40)Stokes(40)Stokes(100)Vcalc0.010.001( ) P g gh f c f + wa 12vs =2009ProppantMovementbyBulkFlowor“Convection” ConvectionPhys:”Transmissionofenergyormassbyamediuminvolvingmovementofthemediumitself.” McGrawHillDictionaryofScientificandTechnicalTerms,FourthEditionz2For fluid flow between
13、 parallel plates.2009ThinFluidTransportisDifferentFromSuspensionTransportProppant drops out of fluid quickly. All solid transportis in a thin “traction carpet”.Bank height builds to an equilibrium based on fluidvelocity. A clear fluid layer is maintained above thesettled bank. The bank advances by “
14、dune building”.2009Video ofSlickWaterSandTransport2009ProppantBridgingandScreenouts Proppantparticlesbridgeinacircularorifice36xtheparticlediameter ParticlesbridgeinaslotwhenthegapequalsthelargestparticlediameterStable BridgeUnstable-FlowDismantles BridgePVariableSlot WidthViewing direction in video
15、2009Slurry Inlet4-12 ppaSlurryOutletFracture ChannelWidth = 0.3”VariableWidthSlotApparatusModel Width=18”SlotBridgingVideoFlowislefttorightBoratexlinkedGuarfluid8ppa20/40OttawasandslurryBlack16/30ceramicmarkersSlotwidthequalsmaximum(16mesh)particlediameter20092009ProppantBridgingVideoCopyright B&A20
16、09SummaryofBridgingStudies Bridgestabilityinholesandslotsisdifferent Slurriesupto16ppawerepumpedthroughaslot1+particlediameterwide Proppantbridgesarepermeableandtransmitfluidpressure Slightopeningoffracturewidthreleasesbridge BridgingaloneisatemporaryandineffectivescreenoutmechanismAnnularFlowAppara
17、tusOuter wallAnnular gapInner wall2009Frac fluid pressureLeakoff pathInternal pressureBottom -Fluid InTop Fluid OutFluidloss2009ProppantTransportinthePresenceofFluidLeakoff Concentrationprofileacrossslot Transversevelocityfromfluidloss Migrationofentrainedparticles Forcebalanceonparticlesatthewall20
18、09ParticlesHeldDynamicallyattheFractureWall Particlesarepulledtotheleakoffsite Transversefluidvelocitygeneratesliftanddrag Leakoffvelocityimposesstabilizinggradient2009Sand“Node”FormationFlowisfrombottomtotopFluidistypicalofcrosslinkedguarVelocityis12fpsPropconcentrationis1.5ppaFluidefficiencyis90%2
19、009Node FormationatFractureLeakoff2009SandAccumulation:LowCvLeakoff Volume * Injected ppaMass of Sand in Node2009Stablechannelflow“Node”growsinlengthFluidvelocityinchannelerodessandChanneldimensionsbecomestableSandheldinplacedynamicallyNoteeffectsofinhibitingleakoff2009DynamicallyStableChannel2009Pr
20、oppantHoldupinFracturedSystems EarlyInjectionSlurry injection at low concentration builds“islands” or “nodes” of packed sand2009ProppantHoldupinFracturedSystems ContinuedInjectionNodesinterconnectandleaveopenchannelsforallinjection minimalpressurerisenotedatinlet2009ProppantHoldupinFracturedSystems
21、IncipientScreenoutEntirefractureispackedexceptfornarrowflowchannel1”.ScreenoutoccurssuddenlywithoutwarningQ=TP/ LW/ PYME2009InteractionofFissureOpeningMechanisms:PDL,HoldupandStorageW/ PPDL andStorageVpVf2009EffectsofProppantHoldup Firstproppantin: accumulatesathighleakoffsites becomesimmobileatthef
22、racwall Laterinjectedfluid: flowsinlocalizedhighvelocitychannels islesssubjecttoheatup,aging,breaking remainsnearinjectedpropconcentrationTracer surveys show first proppant injected remainingat wellbore. Does this indicate localized high leakoff?2009EffectsofHoldup(cont.) Leadstoshortproppedlength Y
23、ieldsanonuniformproppantdistribution Cancausenearwellscreenoutsandperforationplugging Canbelinkedtoproppantinducedpressureincreases CansubstantiallyaffectfinalconductivityandwellperformanceProppant Holdup Factor = 1.22009ExampleofProppantInducedPressureIncreaseModeling00:001/2/197000:10 00:20 00:308
24、00070006000500040003000200010000A1614121086420B6543210CGOHFER Bottom Hole Pressure (psi)GOHFER Slurry Rate (bpm)GOHFER Bottom Hole Pressure (psi)A GOHFER Surface Pressure (psi)AB GOHFER Surface Prop Conc (lb/gal)A GOHFER Surface Pressure (psi)CACustomer:Well Description:TimeJob Date: Ticket #:UWI:00
25、:401/2/1970GohWin v1.3.015-Mar-01 15:212009MitigatingProppantHoldup Highviscositygels Minimizefluidloss Deepinvasionoffracturesystem Possiblesevereproductivitydamage Particulatefluidlossadditives Mustbridgenaturalfractures Requireslowpermeabilitytostopleakoff Canminimizeinvasionoffractures Alteredde
26、signphilosophy Useclean,nondamagingfluids Staybelowcriticalsandinputconcentration2009Leakoff Controlwith100mesh2009FluidRequirementsforTransport Padvolume: Notdeterminedbytipscreenoutcriteriaorfluidefficiency Unnecessaryinwaterfracsandslickwaterjobs Howmuchisenough? Fluidstability Howmuchviscositydo
27、youneed? Howlongshouldthefluidremain“stable” Whattemperatureprofileshouldbeusedinbreaktestdesign? Whataretheimplicationsoncleanupandproduction?2009Stagnant FluidFluid travels through small channelsat high rate with little residence timeor formation contactVariableFluidRheologyLeadstoChannelFlowandPr
28、oppantBypassFluid mobility decreases inareas of high prop conc andlow shearHigh-LeakoffFluid at frac tip is“new” and cold2009ModelingProppantTransport Modeltimedependentfluidrheology,especiallylowshearviscosity,duringbreak Tracklocalshearrates,fluidcomposition,age,andsolidsconcentration Inputproppan
29、tsizeanddensityandtrackslurrybulkdensity Checklocal,timedependentbridgingconstraints Determinepressuredistributionfrommobilitydistribution Feedbackpressuredistributiontofracturewidthprofile,shear,andvelocityprofile2009ClosingComments Whenjobsaredifficulttoplaceitisusuallybecauseofproppanttransportan
30、dimpendingscreenouts Themechanismsthatleadtoscreenoutmustbeunderstoodandevaluated Inmanycasesearlyscreenoutsarenearwelleventscausedbypropholdupandfissureleakoff Thesecasescanbediagnosedandpredictedandappropriatedesignchangescanbemade Makingdesignchangesbasedonincorrectmodelsofthescreenoutprocessfailstoprovidesolutionsandmaymakethingsworse Dontbetoohastytoblamenarrowfracwidthsorlowfluidviscosity
