Fluid Selection Methodology Leads to Implementation ofMethylglucoside Blend for Drilling Low Permeability

Minehead Cardium Sandstone Gas Reservoirs Overbalanced

Jim Masikewich, Newpark Canada Inc.D. Brant Bennion, Hycal Energy Research Laboratories Ltd

Mel Blackbume, Rio Alto Exploration Ltd.

design of the fluid and its implementation are presented Theinitial production results are also included

ABSTRACT

Variow formation damage mechanisms occurring in theCardium sandstone have been documented. Many of thesehave been studied in the context of production enhancement.Recently, increasing attention is being devoted to damage inthis formation from a drilling perspective. Often the damageconcerns and issues incurred while drilling can be eased ifperforations are expected to penetrate beyond the localizednear wellbore damage. In vertical wells where casing isexpected to cover the production zone, .. drilling in» with a

reasonably clean fluid exhibiting low API fluid losscharacteristics is likely adequate in most instances. However,as horizontal wells and slotted-liner completions become morecommon and as operators rely on tighter, less permeable zonesto increase reserves and production that are completed in anon-perforated openhole mode, the damage-while-drillingissue becomes more prevalent.

INTRODUCTION

Some of the most difficult challenges in fluid design areperpetrated by tight, sub-irreducibly saturated, sandstone gasreservoirs. When these reservoirs are to be drilledunderbalanced, the task is less daunting. However, diligence isstill required. Underbalanced fluid design focuses on bothfluid-fluid and fluid-rock compatibility as a contingencyagainst unplanned periods of overbalance or against theoccurrence of spontaneous counter-current imbibition!. Whendrilling overbalanced, the fluid design incorporates oneadditional dimension - the selection of the constituents to createan appropriate filter cake. This is to minimize the loss of wholefluid and mud filtrate into the reservoir. Because the verynature of cake deposition and bridging on pore throats andfractures denotes some degree of spurt loss associated invasionthe same design diligence must be applied to the liquid phaseof the overbalanced drilling fluid. The solid phase ofdte fluid-the bridging particles, must be soluble in some fluid when it'stime to produce dte well. In sandstone reservoirs, even if porethroats are plugged with a portion of insoluble drilled solids, if

The objective of this paper is to present and discuss themethodology and testing results used in selecting a horizontaldrilling fluid to be employed in the Cardium Gas reservoir atMinehead in Central Alberta. Various aspects of both the

some soluble product is also present, bridge integrity will oftenbe degraded so that production can begin.

The problem is that the bridging materials available todayare either oil soluble, acid-soluble or water-soluble. Often (notalways) these solvents are incompau"ble with the sandstone gasreservoir, especially if there is clay in the pore throats or if thecementation matrix is carbonaceous or if the reservoir lackswater. The problem is complicated in that many horizontalwells in formations with questionable competency arecompleted using a slotted liner, making any type of selectivestimulation almost impossible.

Reservoir TypeAcid GasDepthPressure (current)Equivalent Density to Kill (current)Permeability (air)PorosityAverage Pore Throat SizeT~tureBitumen (present)FracturesWettability8wi (average)

GasGas2400 mTVD17.0 mPa670 kg/m30.01-10 mD8-14%0.7-7 J1Dl75oCNoMicrofracmres Present

Water Wet16%

Various fom1ation damage mechanisms occurring in theCardium Sandstone have been documented, most of them in thecontext of production enhancement These include clayswelling, fines migration, paraffm deposition. scale andbacteria2. The design focus for this well centered on swellingclay and another mechanism - phase trapping or water

blocking.

BOREHOLE STABILITY

Rio Alto Exploration Limited chose to place a horizontalwellbore into the Cardium reservoir at the Minehead field inCentral Alberta. Given the geological and reservoir conditionsthe tmst cost effective drilling strategy would likely have beento drill into die reservoir underbalanced. Prior to designing thewell. a borehole stability study was conducted by anindependent contractor. Caliper logs from vertical wells wereanalyzed to identify and locate hole enlargement occurrenceswithin and immediately above and below the CardiumSandstone. The borehole stability model ST ABView was usedto predict the extent of rock yielding for the tmst criticalgeological units in the planned horizontal well. Inputparameters for this modeling were estimated from availablegeological data, reservoir data, drilling data and wireline logdata provided by Rio Alto. Sensitivities of rock yieldingseverity to critical input parameters were investigated over arange of bottom hole pressures including underbalanced andoverbalanced conditions. It was detennined that the risk ofborehole instability from excessive rock yielding in a weakinterbedded sandstone - shale interval was high forunderbalanced conditions. Therefore, the well was designed tobe drilled overbalanced.

GEOLOGY AND RESERVOIR CHARACTERISTICS

The Minehead Cardium "C" gas reservoir is located in the 49-18 W5M area of Alberta and consists of a low permeabilityCardium sandstone at an average depth of 2350-2400 m Initialreservoir pressure ( 1986) had a value of 23670 kPa at atemperature of 75oC. Matrix Cardium quality in Minehead isvery low with ~ilities (to air) ranging from 0.01 to 10mO, but with the majority of the pay exhI'biting peIUJeabilitiesof less than 1 mO and porosity in the 10-14% range. Thematrix consists of a ~tely sorted mid to upper fine-grained quartz cemented quartzose litharinite. Primaryintergranular porosity is lowered by compaction effects andcementation by authigenic quartz. Authigenic chlorite claysincrease micro-porosity in some areas of the sand and reducespermeability. The chlorite coats grains and lines pore throats.In average matrix, pre diameters range from 45-65 microns andpore throat sizes from 0.6 to 7 microns. It is postulated thatso~ degree of natural fracturing is present in Minehead in theCardium sand which will act as a feed source from a largevol~ of the tight rmtrix. Intersection of these fractures bythe proposed horizontal well was another primary objective ofthe program.

FORMATION DAMAGE ISSUES - BACKGROUND

The presence of mixed layer clay in the reservoir was aprimary concern during fluid system selection. The adverseeffects on ~ility by the swelling, smectite componenthave been documented in the literature. In conventional oilproduction, most of the clay related problems occur in the nearwellbore region, and are associated with well operations suchas drilling, completion and workover" Therefore, the operatorsolicited recommendations from selected drilling fluidsuppliers with "Drill in Fluid" systems designed to inbt"bit thetendency of the smectitic clay to swell. An equally importantissue was the possibility that invaded aqueous fluids couldimpose a permeability reduction due to a phenomenon termedaqueous phase trapping.

Previous laboratory analysis conducted on Cardium samplesfonn this vicinity indicated the bulk rock was up to 4.3% acidsoluble with traces of calcium, magnesium, iron and sulfate inthe acid filtrate. An S.E.M. equipped with an energy dispersiveanal~r detected kaolinite, illite, mixed layer illite / smectiteand chlorite. These clays were descn"bed as being dispersedthroughout the core matrix, lining pore throats and coatingdetrital grains. The following properties describe the reservoirat this location:

2

Phase napping, so~ti~s called "water-blocking" may bedescribed as a saturation hysteresis effect associated with theincrease and retention of an immobile liquid phase into aporous media. This effect is exaggerated in reservoirs thathave a low initial immobile liquid saturation. The result ofplacing additional ilDDX>bile fluid into the pore system is a netreduction in pe~bility to the producing hydrocarbon phase.This pheno~non has been described extensively in theliterature4-7. In fact, it was the first fonnation damagemechanism to be recognized8 and, by 1952, at least 28 patentshad been issued to cover water-blocking ren¥>val treatments.9

hydrocarbon or aqueous filtrates are imbibed intosub irreducibly saturated reservoir rock. This can occur duringboth overbalanced and underbalanced operations.

Factors affecting the severity of the phase trapping includethe magnitude of the difference between Swi and Swirr;reservoir quality (because the capillary retention forcescontrolling the magnitude of Swirr are inversely proportional topore throat radii); the configuration of the relative penn curvesat low liquid saturation levels (the more convex the curve, thegreater the permeability reduction for a given increase intrapped liquid saturation); the depth of invasion of trappedphase since deeper invasion results in a reduced effectivedrawdown gradient per unit reservoir length and availabledrawdown pressure.

Fluids moving through reservoir rock IIDlSt travel throughpores connected by narrow channels or throats, which behavelike various sized capillary tubes. Surface tension. wettabilityeffects and tortuosity playa role in determining nature of thisflow. For example, in a water-wet system, water willspontaneously displace gas or oil from the capillary or porethroat A pressure, called the threshold pressure must beapplied to displace the water back out with air or oil. Near theweUbore, where the pressure may be very low, this thresholdpressure may be too great to allow for the initiation of flowback into the wellbore. The pressure required to displace fluidthrough a capillary is inversely proportional to the pore throatradius. In very small pore throats, up to hundreds of poundsper square inch may be required to initiate flow. Figure Iillustrates this. Reservoir rocks may be characterized in part bytheir proportions or saturation of various fluids. Initial watersaturation (Swi) refers to the average proportion of pore spaceoccupied by water initially - when it is first exposed.Irreducible water saturation (Swirr) represents the watersaturation that is forced to exist due to capillary ~hanics. Inso~ reservoirs Swi is lower than Swirr. This situation isconducive to aqueous phase trapping. BennionlO has reviewedthe origin of low Swi conditions in reservoirs. They include,vaporization. changes in pore geometry, adsorption of reservoirwater by anhydrous clays and minerals and irreduciblesabJration hysteresis effects. The Minehead Cardium sand is aclassical example of an undersaturated low pemleability gasreservoir (16% Swi in rock with K.vc < I mD).

It should be noted that additional immobile filtrate does notalways result in a pem:x:ability reduction because someincremental immobile fluid can be imbibed into ineffectiveporosity. However, ~bility effects due to hysteresis dooccur in DK)St cases where initial saturations are belowirreducible. This impairment can be irreversible in the absenceof special stimulation procedures.

Phase trapping in natural or induced fractures is apossibility if the fracture is small enough - usually less than 10microns - to exhIbit a capillary pressure or if damage to thematrix feeding the fracture is such that the available drawdowncannot exceed d1e threshold pressure described previously.(Note that Darcy's law considers pressure, not thresholdpressure).

In order to predict the occurrence, it is imperative to acquireaccurate initial saturation values. Log evaluations of watersaturations may be incorrect if not properly calibrated or ifgood Rw data is not available (often the case for reservoirswhere the water phase is not mobile and no water saturationdata are available). Cores obtained with drilling fluids may beflushed or otherwise altered. Using low invasion coretechniques or sponge coring coupled with an inert fluid such asan all oil fluid or a tracer treated water-based fluid may providethe best saturation infonnation. BennionS.12 has providedmethodologies for predicting phase trapping with tables, graphsand equations. The basic equation can be written:

In multiphase flow through reservoir rocks, the flow pathsof the fluids are controlled by the wettability of the rock. In awater-wet oil reservoir the water flows along the surface of thesand grains and the small capillaries while the oil flows throughthe center of the pores and through the larger capillaries.Figure 2 shows the permeability relationship between gas andwater as a function of water saturation. Note that theintroduction of water causes a marked decrease in ~ility- irreversibly up to the point where Swirr is plotted (Swirr isnot necessarily the point where the water beco~s nX>bile).

(/)

where:APTi = aqueous phase trap indexK. = uncorrected average fomlation air permeability (mD)8wi = initial (not irreducible) water saturation (fraction)

Range of validity:k. = 0 - 5000 mD8wi = 0-1.00

Phase traps may be established in several ways", the mostcommon being, physical displacement of potentially trappingfluids into the reservoir during overbalanced operations andimbibition and countercurrent imbibition. where either

3

If the APT is ~ 1.0, the fonnation is unlikely to exhibitsignificant permanent sensitivity to aqueous phase trapping. Avalue of between 0.80 and 1.00 ~ the fonnation nmyexhIbit sensitivity to aqueous phase trapping, while a value ofless than 0.80 suggests there is likely a significant sensitivity toaqueous phase trapping. A more rigorous evaluation of APTmay consider relative permeability adjustment, invasion profileadjus~nt and reservoir pressure adjustment factorss,u.

Where the potential for water-block exists, preventativemeasures may be taken. For water-blocking in gas reservoirs,air, nitrogen, or pure oil may be considered as drilling fluids.Hydrocarbon based fluid may be successful in oil reservoirs.In all overbalanced scenarios that are intended to be open-holeco~letions, whole fluid or filtrate loss should be minimized.A proper bridging system design and filtrate reducers areadvised. Underbalanced drilling, when possible, provides aprevention alternative, although sub-ineducibly saturated zonesare likely to spontaneously imbibe drilling or misting fluid intothe reservoir.

Low pem'leability may not always be an indicator of phasetrap potential if the low permeability is the result of a low"average frequency of flow channels" rather than the size of thepore throats. This scenario can occur, particularly incarbonates, where relatively few larger pore channels candominate the permeability for a pore system. In this case, it ispossible that the capillary retention of filtrate is minimal, eventhough pe~ability is low.

FLUID SELECTION

Table 1 provides a summary of the routine air permeabilityand porosity measurements conducted on the plug samplesselected for use in the Minehead Cardium study. AU plugs werecleaned with an azeotropic mixture of chloroform and methanolprior to testing to renX>ve any oxidized hydrocarbons or salts.The best ~thod of validating dte prediction is to perform a

phase trap test on a representative core sample. Figure 3illustrates the apparatus used for dte test. To perfonn the test, asmall core plug is restored to its original wettability and initialsaturation and fluid components in a badt containing dtoseconq>onents at tenq>erature. The plug is ox>unted, overburdenstresses are applied and the apparatus is raised to reservoirtemperature. In the case of a water-wet gas reservoir, after dtepemability to gas bas been established, formation brine, oftensimulated, is slowly introduced to dte plug until several porevolumes have passed through it. The test then ~s diepermeability to gas at several pressures including the thresholdpressure where fluid movement is initiated. If, upon re-establishing the flow of gas through the plug, the originalpemleability is not attained, the reduction can be attributed to aphase trapping effect.

Table 2 sununarizes the results of five different phase traptests conducted on Minehead samples using formation water(base case), various concentrations of n-butanol in water (as an1FT reducing agent) and also with an oil based Cutter "D" fluidsystem. The results of the baseline test with formation wateractually showed the most favorable results, but it was laterconcluded that the sampled tested had a small microfracture(see Photograph 1) which co~mised the results of the testThe other samples tested all exhibited moderate to severereductiom in permeability due to phase trapping, even with theinclusion of fairly high concentrations of butanol in the testsystem as an 1FT reducing agent.

After the phase uap test indicated there was so~ damagedue to phase trapping and that oil-based systems did not appearto offer a significant advantage over water-based systems, thelab work centered on water-based systems that had knownsurface tension reducing characteristics. All of the candidatesystems required a breaker treabnent. The trea~t for Mud1, the MEG system, involved a displacement to a fluid carryinga cellulose specific enzyme. It was recognized that thistreabnent could introduce a fluid with an inefficient surfacetension reducing ability to the reservoir. Therefore, a separatestudy was conducted to ascertain the co~atlDility of methanol,a known surface tension reducer on the effectiveness of theenzyme treatment package.

The literature reviews and discusses various methods ofdealing with phase trapping after it has occurred. Penetratingthe zone of damage by perforating or hydraulic fracturing isone method. Numerous removal techniques have also beenattempted and d~nted. Most of these methods involve1FT reduction. This avenue has merit since capillary pressureis a direct linear function of the interfacial tension existingbetween the trapped phase and the bulk producing phase. Ingas reservoirs, injection of mutual solvents such as methanolhave shown to yield better gas production 13. However, it can

be difficult to use surfactants to efficiently partition across dIewater/gas phase boundary. In oil reservoirs, chemicalsurfactants14 as well as higher alcohols have proven effective.Gaseous 1FT reducing additives such as CO2 or solvent treatedCO2 have been used successfully in some situations. Alteringpore geo~try with acid serves to reduce capillary pressure,increasing pernleability in the near wellbore vicinity. Theeffects of the spent acid (i.e. additional water) may however beas adverse as the initial situation. Odler water-block renX>valtechniques may include dehydrated gas injection and heattreatment

This study measured the efficiency of the breaker treatmentby measuring the time taken for a volume of 2% KCI water toflow through Bandera Sandstone disks before and after thedisks were exposed - first to the drilling fluid. then to theenzyme treatment. The tests, summarized in Table 3 indictedthat the addition of methanol to the carrying fluid did adverselyaffect the enzyme performance.

4

The mud systems tested as a portion of this work included drilling fluids. The specific overbalance pressure for each mudtested was calculated based on that fluids specific and totaloverbalance pressured varied between 9000 to as high as 14000kPa depending on the system under consideration. Whole mud,at the specified overbalance pressure, is displaced past the faceof the core ~le on a continuous basis until either a steadystate leakoff or sealing filter cake is established and the totalfluid losses and depth of invasion of filtrate are tracked.Following this procedure a breaker treatIMnt was applied, withsufficient time for reaction to occur, for many of the tests.Regain permeability measurements to gas were then conducted(in the opposite direction to the mud exposure to simu1ateproduction from the reservoir after drilling) by graduallyincreasing the drawdown pressure in incre~nta1s up to themaximum value expected to be attainable in the reservoir. Thiswas conducted while tracking the increase in permeability ateach level to ascertain the minimum threshold pressure for gasflow initiation into the damaged matrix and determine thedegree of permeability impairment as a function of the degreeof increasing drawdown.

Oil Based Mud 1: Gelled oil and breaker: distillate base oil-viscosified chemically as opposed to viscosification with clay.The viscosity of this system as well as the cake may be brokenchemically.

Oil Based Mud 2: Gelled diesel and breaker, similar to aboveusing diesel as the base oil.

Oil Based Mud 3: DieseVsurfactant blend.

Water Based Mud I : System dtat incorporates a blend ofmethyl glucoside and polyglisseride additives with enzymebreaker.

Water Based Mud 2: Blend of cross-linked polymers and claystabilizer with enzyme breaker.

Water Based Mud 3: Acid soluble system that uses aproprietary surfactant with oxidant breaker.

The results of the two series of experiments are summarizedin Tables 4 and 5. The test results clearly indicated that thewater-based Muds #1 and #2 bad the best performance of themud systems tested. Subsequent to this, it was ascertained thata portion of this favorable regain perDl was due, once again, tothe presence of a microscopic fracture in the core samples(visible only under a high magnification SEM examination).Although it was felt that this bad contributed to the goodperformance of the MEG system, it was determined that, due tothe small aperture of the fracture, some benefit was likely stillapparent in co~n to the other system tested which hadless favorable performance, and that the presence of smallfractures such as this were, in fact. representative of the targettype of matrix in the Cardium sand in Minehead. More detailedstudies using variants of the MEG system are summarized inTable 5. The results suggested that high concentrations ofmethanol degraded the perfOrDJance of the breaker system.Based on this, a 12-14% MEG system with no methanol wasselected as the optimum fluid to drill the well.

Water Based Mud 4: Acid soluble system consisting of HEC,Xanvis, Starch, a proprietary surfactant and a proprietary claystabilizer with "crystalline" breaker.

A total of 12 different whole mud leakoff tests wereconducted on samples of restored state Minehead core ~terialat reservoir conditions to attempt to ascertain which systemprovided the best regain permeability performance. Aschematic of the e~ntal equipment used for these studiesappears as Figure 4. The core sample, which had an initialwater saturation of 16% uniformly dispersed within it, is~unted in a uniaxially confined coreholder that appliedreservoir confining overburden pressure (43000 kPa) to thecore ~terial. The confined sample is then heated to thereservoir te~ture of 75 oC and baseline ~i1itymeasurements with humidified gas are conducted in a flowdirection opposite to the subsequent drilling fluid invasion.

Drilling fluid, prepared by the respective fluid supplier, isadded to a high-pressure continuously stirred reactor vessel. Inaddition, 3% by mass of synthetic drill solids, consisting ofpulverized Minehead Cardium core material (to simulate thosecuttings generated by the drill bit and not removed by thesurface solids control equipment) are added to the IInJd toprovide the best possible representation of the fluid which willbe circulating downhole at the time of drilling. The stirredreactor keeps all solids in a state of continual suspension toensure that a uniform mixture is flowing by the face of the coresample on a continual basis.

MEGTHEORY

Alkyl glucosides are ~mbers of the class of compoundsknown generically as glycosides. The tenn glycoside is appliedto a compound where a sugar is combined through its reducinggroup with an organic substance such as a phenol. Many of theknown glycosides occur naturally in plants and anima1s andwere originally isolated from such sources. The sugar portionof most naturally occurring glycosides is glucose andaccordingly, these giycosides are known specifically asglucosides. The product that was used in this project ismanufactured from cornstarch. It has desirable qualitiesincluding lower viscosity, temperature stability, and bacterialresistance. Thermographic analysis showed stability to about3500 F in a nitrogen atmosphere for a 70010 w/w solution. A

Due to the highly pressure depleted condition of theCardium sand in the Minehead reservoir (currently approx.14000 kPa), high overbalance pressures are expected to beencountered when using conventional oil or water based

5

65% w/w solution of methylglucoside remained fluid at -220 F.A 400/0 w/w solution of methylg1ucoside showed no growth ofmicroorganisms even when inoculated with bacteria. n:x>ld oryeast. Simpson et aIlS attested to the superior lubricitycoefficient of the system. The fluid used in their DownholeSimulation Cell (DSC) test bad a lubricity coefficient of 0.06 astested by the API RP-13B procedure. Water-based fluidstypically have a coefficient of 0.2 to 0.3 while oil-basedsystems usually have coefficients of less than 1.

onto surfaces, proDX>ting a tight cake. Photograph 2 shows aSEM photograph of the methylglucoside system filter cakeagainst one of the Cardium Sandstone core plugs used in fluidselection for the Rio Alto well. The Zhang study alsoconcluded that the methylglucoside fluid was also effective inreducing damage caused by filtrate retention due to its lowsurface and interfacial tension.

HSE ISSUES

The product referred to in this paper is from a commercialproduct that is about an equal mixture of the alpha and betaforms to which KOH has been added to provide an alkalinesolution containing 70% w/w methylglucoside. This solutionbas in turn been complimented with an additional product.Polyglycerine that has proven to be synergistic in terms ofinhibiting clay hydration. The properties of the final product.called NDFX ll9oC are outlined in Table 5.

The National Institute for Safety and Health (NIOSH) andthe Occupational Safety and Health Administration U.S.Department of Labor (OSHA) do not list either PolYglycerineor Methylglucoside in their databases. The WorkplaceHazardous Materials Infommtion System (WHMIS)classification is "D-2(B)" ~aning it can be a skin and eyeirritant, and the Transportation of Dangerous (TGD)classification is "Not applicable".

The primary function of NDFX 119C is to provide shalehydration. Simpson has suggested that the nx:thylg1ucosidesolute becomes fixed in the near borehole surface of the shale.This establishes an effective semi permeable membrane thatallows the solvent (water) to move from the shale to the mudunder a chemical (OSIOOtiC) potential that exceeds the hydraulicpotential tending to force water into the shale. The studyindicated that 44% w/w of the neat product had a vapor activityof 0.88.

The two components of the NDFX 119C inhI"bitor arerelatively non-toxic to JOOst test species however, surfactantsdo not fair well in the Microtox bioassay. The Microtox testuses bioluminescent bacteria as the test species and theseorganisms are sensitive to surfactants. A review of the toxicityto various species is included in Table 6.

Biodegradability tests conducted at the University ofCalgary indicate that NDFX 119C is readily biodegradable.Toxicity tests on soil/waste combinations show that thematerial is readily absorbed by soils and that it can bebiodegraded rapidly.

Liquid chromatography analysis of the shale ftomDownhole Simulation Cell (DSC) tests conducted in Simpson'sstudy provided confimJation of the fixing of themethylgtucoside onto the shale. Shale sampled within 0.25inches of the test borehole contained 1.4% w/wmethylglucoside while no methylglucoside was detected deeperinto the core. The study suggested iliat the presence of thehydroxyl groups in the methylglucoside configuration mightaccount for the unique ability to form a semi penneablemembrane just inside the shale. The hydrated DX>nOIMr seemsto be the right size to penetrate the exposed pore spaces whereinteraction with the clay surfaces cause the methylglucoside tobecome fixed, while the water solvent remains free to DX>ve.Hydrogen bonding was suggested as a possible ~thod offixation. Testing also showed iliat shale exposed to this systemremained intact and was actually harder in the vicinity of thesimulated well bore iliat was tested.

Application of the wastes to a biologically active soil is thebest way to biodegrade the active components of the wholenmd system This will relmve any Microtox test determinedtoxicity from the wastes and they will be able to be disposed ofby routine methods. There are two scenarios for the disposal ofthe wastes. For disposal of solids, samples of the solids andwaste are combined in a few ratios and an aqueous extract istaken from each of the mixtures. Depending on the toxicity ofthe extracts, the solids may be mixed, buried and covered or bespread on lease and allowed to biodegrade. Research hasshown that typical field ratios of soil and waste are usuallynon-toxic. If the whole fluid needs to be disposed the best andImst economic alternative is to pump-off the liquids adjacentto the lease and dispose of the solids as above. During winteroperations there are significant issues with respect to freezingwhich may render this option impractical. The results in Table6 were submitted to the EUB and the disposal methods werediscussed. It was agreed that approvals for the disposal of thissystem would be granted on a site-to-site basis.

Yan Zhang et al16 evaluated several characteristics ofmethylglucoside as they related to sandstone reservoir damage.They found the methylglucoside system was less damagingthan other drilling fluid systems it was compared to. The studyconcluded that the adsorption of the ~thylglucoside hydroxylonto the clay surfaces prevented both clay swelling and claymigration. They also postulated that the tight filter cake and"excellent" filtration properties could be attributed to the samemechanism. The hydroxyl group assisted particles in adsorbing

FIELD IMPLEMENTATION

The well was designed as a new drilL to be drilled from alocation at 01-07-49-17 WSM in Central Alberta. Ensign

6

CONCLUSIONDrilling Rig 33E was chosen to drill the well. This is a rotaryrig equipped with a Mid-Continent U-36EA (500HP / 373kW)draw works. After drilling 349.Omm hole in one pass using aBentonite slurry, 244.5nun casing was set at 6OOm. 222nunintemlediate hole was drilled with an all oil fluid to kick offpoint at 2425m, where angle building continued to 90°.177.8uun casing was set at 262lmMD - 24 I 41VD. The156nun horizontal hole was programmed to use the NDFX 119fluid. The inclination remained at approximately 900, whiledrilling proceeded through the Cardium formation to a totaldepth of 3539mMD. The open horizontal disp1ace~t totaled918m Upon completion of drilling, a 144mm liner was set inthe open-hole horizontal interval.

A methodology for the selection of a low damageoverbalanced drilling fluid system for a clay containing lowpermeability sandstone bas been presented. Major damagemechanisms found to be operative in the Cardium sand in theMinehead reservoir included possible fresh water induced claydamage and severe potential for water and hydrocarbon phasetrapping. Accordingly, the best fluid system, with respect toclay stabilization, fluid loss and IFf and regain pe~bilityproperties for phase trapping, was found to be amethylglucidside based water base system. This system wassuccessfully used to drill a 965 m horizontal well in the highlydepleted microfractured Minehead Cardium sand at approx.13~ kPa overbalance widl no significant fluid losses andgood inflow performance in excess of 5,000,000 scf/day fromthe very low quality «0.5 mD) matrix.

The drilling fluid contained 12% v/v NDFX 119 initially.Table 7 shows the chemical constituents of the drilling fluidsystem Table 8 shows the fluid properties that were typical ofthe interval. After drilling COnmJenced. fresh water was addedto the system in 5m3 increments to maintain volume. 3 barrelsofNDFX 119 were added with each 5m3 of water. The NDFX119 concentration was roonitored at the well site duringdrilling. It was increased to 14v/v% at 2900mMD, to improvethe sliding ROP. This addition helped but it also was associatedwith a minor viscosity increase. All bit trips and pipemovement were reported as "good" until 3307mMD, where itwas necessary to ream and clean through the lateral section,with bit #6B to 3307 m Drilling continued with the rheologybeing increased to improve hole cleaning as well as the slidingROP. At total depth, directional tools were laid down and thehole was wiped with a slick string. The well was not displacedto the enzyme treatment, as it was felt that breaking the fluidviscosity as well as the filter cake could contribute to a loss ofborehole integrity and thereby jeopardize the success of theliner running operation. The 114 DUD slotted liner was loweredto total depth without problems. The liner was not cementedinto place.

ACKNOWLEDGMENTS

Cal Carratt and John Delone at Nowsco - enzyme stimulationwork.Ken Zinger of Newpark Fluids Canada - assistance with fluid

system design.GR Petrology - SEM work.Pat McLellan of Advanced Geotechnology - borehole stabilitywork.Vivian Whiting - preparation oftbe figures and manuscript.

REFERENCES

2.

PRODUCTION 3

Production in the Minehead Cardium Pool (C and F)commenced in late 1986. Initial pool pressure was 23670 kPa.Currently, the pool pressure varies with location and~bi1ity. Pool pressure in the vicinity of the 01-7 well wasestimated to be in the vicinity of 15000 kPa to 17000 kPa. Astatic gradient taken before commencement of productionindicated the pressure was 14260 kPa. Well productioncormnenced at approximately 150 lQ3m3/d (5.3 MMscf/d) andhas declined over three months to 80 lQ3m3/d (2.8 MMscf/d).This rate is approximately twice the flow rate of a fracturedvertical well. Cost of the horizontal well is about twice that ofa fractured vertical well. The incremental economic benefit iscurrently being assessed.

4

s

6,

Bennion, D.B. et aI. "Underbalanced Drilling - Praises

and Perils ", SPEDC, December, 1998.King, R W., Adegbesan, K.O., "Resolution of the PrincipalFormation Damage Mechani~~ Causing Injectivity andProductivity lII1)airment in the Pembina CardiumReservoir", SPE Paper 38870, 1997.Zhou, Zhihong (John), Cameron, S., Kadatz, B. andGunter, W.D. "Clay Swelling Diagrams: TheirApplications in Formation Damage Control", SPE Paper31123-P,June 1997.Bennion, D.B., CiDX>lai. M.P., Bietz, RF., Thomas, F.B."Reductions in the Productivity of Oil and Gas ReservoirsDue to Aqueous Phase Trapping", Petroleum Society ofCIM Paper 93-24, May 9-12,1993.Bennion, D.B., Thomas, F.B., Bietz, RF., Bennion, D.W."Water and Hydrocarbon Phase Trapping in Porous Media:Diagnosis, Prevention and Trea~t", Petroleum SocietyofCIM Paper 95-69,1995.Bennion, D.B., Thomas, F.B., Bietz, RF., Bennion, D.W."Remediation of Water and Hydrocarbon Phase TrawingProblems in Low P~ility Gas Reservoirs", PetroleumSociety of CIM and CANMET Paper 96-80, June 10-12,1996.

7

7. Grober, N.G., "Water Block Effects in Low Pemr;abilityGas Reservoirs", Petrolcurn Society of CIM Paper 96-92,June 10-12, 1996.

8. Darley and Gray, "Composition and Properties", GulfPublishing Co., pp 492.

9. Garst, A.W. "A Low Cost MedlOd of ProductionStimulation", SPEPaper 386-G, Nov. 1954.

10. Bennion, D.B.. Bietz. R.F.. Thomas. F.T.. Bennion, D.W.."Fluid Design to Minimize Invasive Damage in HorizontalWells". OI-Mdian SPFJClM/CANMET International PaperHWC94-71, March 20-23,1994.

11. Mcleod. H.O., Coulter, A. W .. "The Use of Alcohol in GasWell Stimulation", SPE of AIME #1663, Nov. 10-11.1966.

12. Jennings, Jr., A.R. "The Effect of Surfactant-BearingFluids on Pemleability Behavior in Oil-ProducingFormations", SPE of AIME #5635, Sept 28-Oct 1, 1975.

8

Table 1. Routine Core Analysis on Small Diameter Test Plugs

SampleNo.

Depth(m)

I 2360.72-61.11

2364.11 - 64.31

I 2358.96 - 59.292360.42 - 60.722358.96 - 59.292359.29 - 59.74235929 - 59.74235929 - 59.742363.21 - 63.542360.42 - 60.72

I 2364.11 - 64.31

~~i~~i~.I 2383:25 ~ 83:57

2383.57 - 83.832385.15 - 85.462385.15 - 85.462383.83 - 84.092386.23 - 86.53

Perm(mD)

Porosity(Fraction)

Grain Density

15B26B9A14A9BIOClOAlOB2314B26A

0.7800.6400.4600.4000.3000.3000.2800.2800.2700.2500.090

0.1410.1290.1500.1290.1380.1400.1390.1390.1340.1220.083

26402650268026402680266026602660266026402650

1213

20A20B1424

0.6200.5800.2400.2300.2000.060

0.1180.1070.1050.1100.0990.083

263026502650265026302650

Table 2. Results of Phase Trap Tests

Table 3. Enzyme Treatment Test Results

Table 4. Summary of Results from Whole Mud System Leakoff Tests

Table S. NDFX 119°C Properties

. JiolImg Point 0(;: I >93 l::olor: Dark Brown .flashPtPMCC-neat I >149°C i, !JU-IIOcps(()U()IpD1-atSOOC) I

1.260-1300 . 10-11 I

! <4 (MM HU)I ..,_:~..~S~~.

I ~OlUU~l~~ ~: I ~r~~ ~~~dI 4; .~.~~

Table 5. Summary of Results from Wbole Mud System Leakoff Tests

Mud Initial Gas I Regain Gas Perms@kPaDrawdown (mD) .Constituents Perm@Swi (% ret»;aill_oforiginall

~-~-I (0)

L-i~--I (5)

I ~~I Viti)

I 10431

0.022

I (19)I 9B

1100/0 ~= ~G)200/0 MeOH I

Mudzyme "C"I Breaker "X"

I w;R~ ';110% Deepdrill (MEG)

Xanvis20% MeOH

M~ "C"Breaker 'X'WBMud#4 I

I O.ll~

j laCI 0.058

I u

(0)

I 0.011(19) ,

I 0.016(27) r---t~::J-,

0.058 rTlrr0.016

~-1~~r-I v(i)"

~r--U~3'--

rl~:;-r-U~9--

I 268100/0 Deepdri11 (MEG)

Mudzyme "CO'

100/0 Deepdril1 (MEG)100/0 MeOH

M~ "CO'Breaker "X"

,I 200/0 ~ ~G)

Mudzyme "C"-

I 0.262

I 9A J 0.153 I 0.028

(18)

I 0.064

(42)

I 0.077

(50)

I 0.083(54) ,

I 13 I 0.073 ~ I-~~~-I O.U2~

~ (34) rl~:~-

Table 6. Toxicity Summary for NDFX 119

~ IDterpre-tatioD-

I Pass I

~

:i::= ~=~;x -::::- . - -, Biousay I Data

ysl neat I 218,000 ppmr: - 0 - 0 ! ~~:= ~ au .. :1

cro x 0 0 ycennc I ~ - 2.0 OXIC

x Y - TOXIC'!

~I Mlcrotox(~%Mud/soll-1nrtIaI) -.I=i~: Mlcrotox (I U"/oMud/soll- InitJal) -. OXICI

I= i~~ M!";~ ~)X 0 so - - y OXIC

-rn- cro x 0 so - ~- [00 I ~~Xi~-'--.,1

IS SIDlU e IOU WI a ves present.Soil is from an Alberta drilling location - 1:4 extract

Table 7. Chemical Constituents of the Drilling Fluid System

Table 8. Typical Drilling Fluid Properties

- OJae . .I GeIO-Pa' 2.~.0" , GellU-Pa 3.0-1.0-0";'

MBT-kg/lD'l NlA'.

I API~~CC'8 :;1 ~:~~~~O':::

:2!

~~Q. -. -/'-"'--11..'

I"/

/

./'

, ,-,,/"

~

II!~

~

~:'~

eJ~aJd UIPII:>

I/)~

~

!~~

ir:

~...e'y

Q..~tV~§~I!i(/):2~co~

..1..,{Jt.1'{i

QI0c-B~~.(/)

~C"

~

~~

VW

I"d~C

IO>

J

)(~~1;~

f~

'\

i

6;~rn'I:'~~~~.2"..J

'D,=u.c'-:c..~E~~;..'.=S~~~,$~~~'t~~~='5.C

o..""~~..-=~...0=0'.=f-~.:=

IM~=,S

JI~

-i 2&~ bserieniM

.

icI!l-

f~~eQ.

II)~

~~coc..c..

«

c..co~

.-

Q)II)co

..c

n-o

t")Q)~~0)

LL

~

I

f J OJ BJ edes

UOI~POJd

aldwes ~~eM &SUB

9idUJeS

pn~/R

rlld

1-J~

n8IJD

~¥:J

-

~

II)~-co...coQ,

Q,

«~0~CO

Q)

-J:E~u..C

)c.--'CCI~~~.~u..