SPE 60325

Formation Damage Processes Reducing Productivity of Low Permeability Gas Reservoirs

D. B. Bennion, F. B. Thomas and T. Ma, Hycal Energy Research Laboratories Ltd.

~ ~. SodIIr ~ PWdM8n ~ ktC.

TI* ~ - pr8P88d for ~ 81 .,. ~ SPE Rocky Mount8in ~;-".."--' ~ ~" ~ h8d ~ c.-. CO. 12-15 I8dI ~.

n8 ~- ~ b Ma; by.. SP£ ~ ~~ r.-. ~~~" ~ ~ .. 8I*m ~ by .. ~.).~ ~ .. ~. -~. ~ rd ~ ~ by the ScdeIy 01 ~ ~ " - MAIjKt kI~bylle8U8a(.). The~-~~rd'---;i~wrypa.-CII~1Ie~~~~"~~~Pap8s~.SP£n - ~ kI ~ r.-. by ~ Cc;.-=-~ ~.. ~ ~ ~E, ~~. ~ ~..~.,,~~.. ~b ~~ wIIhouI lie WI1II8I ~ ~ .. Sod8Iy ~ P8IrdIUn E,.-. . pr,..;II...:.P..-;;-, kI reproduce ~ prW . IWIIIIded kI .. abIk8d ~ rd .-. th8I 300 ~:~ rr8Y rd be ~ The 888d ~ ~ ~ ~ .., ;. -;: ~ "by -- lie ~ - *d. ~ LI.-WI. M'E. P.O. .. ~ ~..., TX

~ U.sA, - Ot-872-M2-8436.

exhibiting low penneability and often low pressure. Oneparticular area which has received increased emphasis in recentyears is the production of gas from extremely low penneabilityformations. Considerable effort has been expended in a nwuberof areas in the Deep Basin area in Canada. the Powder RiverBasin in the Central United States and in the Pennian Basin inthe Texas area in attempting to exploit gas reservoirs withaverage insitu permeability in the 100 micro-Darcy (0.1 mD] orless range.

This paper details a number of issues which center around theability to flISt diagnose whether a particular reservoir will be aneconomic candidate for viable gas production and secondly, toevaluate what are the best drilling. completion, and ultimateproduction practices for such wells to obtain the maxjmum flowrates and ultimate recoverable reserves of natural gas.

AbstractAs drilling and completion technology advances, the trend toexploitation of gas reservoirs exhl"biting ever lowerpermeabilities continues. This paper discusses issues associatedwith the identification of productive, low permeability, gas-producing fonnations and the successful completion andproduction of these reservoirs. For the purposes of this paper, avery low permeability gas reservoir is defmed as a fonnationhaving an in-situ matrix permeability to gas of 0.5 mD or less.Criteria are presented for identifying economic absolutepermeability cutoffs for low permeability gas-bearingformations. Very low permeability gas reservoirs are typically ina state of capillary undersaturation where the initial water (andsometimes oil) saturation is less than would be expected fromconventional capillary mechanics for the pore system underconsideration. These formations are commonly referred to asdehydrated or desiccated formations and have been documentedon a worldwide basis. Retention of fluids (phase trapping) isdiscussed as one of the major mechanisms of reducedproductivity, even in successfully fractured completions in thesetypes of formations. As well, a variety of diagnostic andremedial treatment options are presented.

What Is a Low Permeability Gas Reservoir?The defmition of a low permeability gas reservoir is arbitrarybut, for the purposes of this presentation, we will consider sucha reservoir to be a reservoir matrix which has an effective insituperDleability to gas of less than 0.5 mD and contains apotentially mobile gas saturation. Extensive reserves of naturalgas are present on a worldwide basis in both sandstone andcarbonate formations which exht"bit these properties. Althoughsuch penneabilities would be prohibitive for the economicproduction of conventional crude oil, they may still be feasiblefor gas production. In many situations, due to the low inherentviscosity of natural gas, the high initial pressure of the producingfonnations. if sufficient reservoir penetration and exposure to thelow pemleability producing zone can be obtained in aneconomic fashion, long-term production of significant volumesof gas may be realized

When addressing the potential productivity of the lowpermeability gas reservoir, three major issues are of paramountconcern.

1. Does the reservoir exhibit sufficient initial pem1eabilityand pressure to facilitate economic gas production rates, even inthe presence of a successful large-scale stimulation treabnent?

IntroductionAs the oil and gas industry matures, ongoing reservoir targetsmove towards more challenging applications coJlUDonly

2 D.B.BENNION, F.B.THOMAS. T.MA SPE 60325

2. Are the initial water and liquid hydrocarbon saturationsthat are trapped within the reservoir matrix low enough to createsufficient gas reserves in place and adequate initial permeabilityfor economic gas production?

3. What potential forms of formation damage can occurduring drilling and completion operations or subsequentproduction operations, that may reduce the ultimate productivecapacity of the reservoir?

Initial Permeability. As drilling and completion technologyimproves, we realize we are in a position to produce gas fromeven lower quality formations. Large-scale hydraulic fracturetreatments in reservoirs with average permeabilities as low as 5-10 jJD (0.005 - 0.010 mD) have been successful. In the absence

of an interconnected natural fracture network. which may assistin the drainage of a low permeability matrix system. thispermeability range appears to be a reasonable cutoff value foreconomic gas production rates using the level of currentcompletion technology. The value of this cutoff will. of course,be highly dependant on a number of other factors, most notablythe original reservoir pressure available for production. If asubnormally pressured or pressure depleted reservoir is underconsideration, the effective pemleability cutoff for economic gasproduction rates may be substantially higher than this value of

5-10jJD.Initial Water and Liquid Hydrocarbon Satllranons. The valueof the initial water and liquid hydrocarbon saturation that arecontained within an intercrystalline matrix system in a gas-producing reservoir are generally controlled directly by thecapillary pressure of the porous media under consideration.Capillary pressure can be defmed rigorously as the pressuredifferential existing between the non-wetting phase [gas] and thewetting phase [water and/or hydrocarbon liquid]. Capillarypressure can also be derived using a number of otherformulations, including the classic height above a free watercontact relationship and also the equihDrium summation of meanradii of fluid interfacial curvature. These respective equationsare illustrated below:

cosOfw , ... R \R. R2) , ,

Fig. 1 provides a schematic illustration of the radii ofcurvature refened to in the specific calculation of capillarypressure in porous media. Figures 2 and 3 provide an illustrationof typical irreducible water saturation profiles in both lowpenneability (Fig. 2) porous media and high pem1eability (li1c.3) porous media. In Figure 3 it can be observed that, in a highpenneability rock, the porous media is generally characterizedby large, open pore throats and pore bodies. We can see that thisgeometry leads to relatively large radii of curvature in the gas-liquid interfaces in the porous media. Since capillary pressureis given by the sum of the inverse of these radii of curvature, itcan be seen that the capitlary pressure values tend to berelatively small, even at low water saturations. This is why high

- 6r ...!-+-!-) . CI)Pc = Pn P... = ADm = 20-

permeability porous media with DO or limited micro-porosityusually tends to exhibit favorable capillary pressurecharacteristics with low capillary pressure values over themajority of the mobile saturation range and relatively lowirreducible water saturations.

Figure 2 provides an illustration of an equivalent situation ina low pernleability gas reservoir. In this case, due to the verysmall size of the pore throats and pore bodies, the tortuous natureof the pore system and high degree of micro-porosity, theobserved radii of curvature of the gas-liquid interfaces are verysmall, particularly at low water saturations, which gives rise tothe higher capillary pressure values and irreducible watersaturation values which are conunonly associated with poorquality porous media.

Fig. 4 provides an illustration of typical curves for capillarypressure for different quality porous media. In general, aspermeability and porosity are decreased and the relative fractionof Jni<:ro-porosity increases, both the capillary pressure valueand the value of the irreducible water saturation tend to increase

substantially.Often associated with this increase in trapped initial liquid

saturation is a significant reduction in net effective permeabilityto gas in the porous media caused by the occlusion of a largeportion of the pore space by the irreducible and immobiletrapped initial liquid saturation present in the porous media. Thisphenomena is schematically illustrated as Fig. 5. On a relativepermeability basis (Fig. 6), one can observe that, in general, thegreater the value of the initial trapped fluid saturation, the lessoriginal reserves of gas in place available for production, andalso the lower the initial potential productivity of the matrix.

In reservoir situations where exceptionally low matrixpermeability is present, one finds that, if the reservoir is in anormally saturated condition (that is, if the reservoir is in freecontact and capillary equiltoriurn with mobile water and is at anormal level of capillary saturation for the specific geometry ofthe porous media under consideration), we find that very hightrapped initial liquid saturations tend to be present Fig. 7provides a schematic illustration of "average" irreducible watersaturations expected for porous media of typical permeabilityranges. The permeabilities given in Fig. 7 represent average,uncorrected routine core analysis permeability data for a matrix-dominated reservoir situation (no fracturing present). One canobserve that in reservoir rock of permeability to gas on anabsolute basis of less than 0.1 mD, effective initial watersaturations are often in the (,00/0 plus region. This often resultsin significant reductions of the original reserves of gas in placein the porous media, and may also result in a very low or zeroeffective permeability to gas, as the gas saturation may be at ornear the critical DK)bile value and hence it will exbJoit limited orno mobility when a differential pressure gradient is applied tothe formation during production operations.

Therefore, it can be observed iliat in most cases where verylow permeability gas reservoirs are potentially productive, the

SPE 60325 FORMATION DAMAGE PROCESSES REDUCING PRODlK:TMTY OF LOW PERMEABILITY ~ RESERVOIRS 3

occurring in-situ particulates within the pore system caused byhigh interstitial fluid velocities induced by differential pressuregradients. In general, fmes migration tends to be mostsignificant when the wetting phase of the reservoir is in motion.Typically, under conditions of pure gas flow commonlyassociated with a low permeability gas reservoir, fines migrationproblems tend to be minimized, as the wetting phase (usuallywater) in which the mobile fines are encapsulated does notmove, and hence, insulates the fmes from the majority of theinterstitial shear associated with the motion of the gas in the poresystem. For this reason, in most low permeability gas reservoirs,fmes migration does not tend to be a major issue since, due tothe low initial fluid saturations which are generally present, nomobile liquids are typically present in the matrix. In addition,many of these reservoirs tend to be at significant depth and haveundergone significant compaction and diagenesis which hasoften removed or cemented most reactive or mobile clays. Finesmigration may be potentially problematic in situations wheresignificant volumes of water or hydrocarbon based fluids havebeen lost to the formation at high spurt losses duringoverbalanced operations, or may subsequently be mobilizedduring high drawdown cleanup or production operations whena portion of the invaded fluid is recovered from the matrix.

External Solids Entrainment. This damage mechanismrefers to the invasion of solid particulate material into the matrixsurrounding the wellbore or fracture face during overbalanceddrilling or completion operations. This mechanism of damage,while significant in higher permeability reservoirs (most notablyin cases of open hole completions) is typically a non-issue invery low permeability gas wells, since it is implicit that sometype of fracture or other invasive stimulation b"eatment will berequired to obtain economic production rates (the surface inflowarea in the unstimulated wellbore is too low to permit economicproduction rates of gas, even if the well is drilled and completedin a totally undamaged fashion). Since invading solids(particularly in rocks with a matrix composed primarily ofmicropores where average pore throat aperture is often less dtanI micron in diameter) usually penetrate only a very shortdistance into the reservoir (less than 10 mm or about 3/8" inmost cases), it is understood that shallow, but severe damagemay not be an issue, as long as it is highly localized to the nearwellbore region and does not cause such severe tortuosity that itis difficult to mechanically propagate the subsequent stimulationtreatment. The exception to this statement would be a situationwhere a horizontal or vertical wellbore is used to penetrate a lowpermeability gas reservoir where the dominant mechanism ofproduction is drainage of the tight matrix through a naturalfracture system. In this case, the well would generally beoriented to have the maximum potential for fracture intersectionand mechanical loss of whole mud or solids which may bridgeand plug the fractures at the fracture-wellbore intersection, andrestrict the ability of the fractures to deliver gas to the formationwhich may be highly detrimental to the well's performance.

Relative Penneability Effects-Phase Trapping. This issue

reservoir exists in a situation where the reservoir sediments havebeen isolated from effective continual contact with a free watersource which is capable of establishing an equilt"brium anduniform capillary transition zone. It appears that a combinationof long-tenD regional migration of gas through the isolatedsediments (resulting in an extractive desiccatin~ effect astemperature and pore pressure are increased over geologic time),or an osmotically-motivated suction of connate water into highlyhydrophillic clays or overlying/interbedded sentiments, may beresponsible for the establishment of what is commonly referredto as a "sub-irreducible" initial water saturation condition.

A reservoir having a sub-irreducible initial water saturationis defmed as a reservoir which exhibits an average initial watersaturation less than the irreducible water saturation expected tobe obtained for that porous media at the given column heightpresent in the reservoir above a free water contact (based on aconventional water-gas capillary pressure drainage test).Significant discussion is present in the literature on theestablishment and diagenesis of sub-irreducibly saturated gasreservoir systems (Ref. 1-5). Many reservoir systems of thistype have been extensively characterized in the Deep Basin areaof Canada, the Powder River Basin area of the United States andthe Permian Basin area in Texas.

In situations where exceptionally low matrix penneability ispresent in a gas-producing reservoir, unless a sub-irreduciblysaturated original condition is present, the reservoir will exhibitinsufficient initial reserves/penneability to be a viable gas-producing candidate. Therefore, with few exceptions, the vastmajority of ultra-low penneability gas reservoirs that would beclassified as exhibiting economic gas-producing pay, would fallinto this classification of subnonnally saturated systems. Thisphenomena will be discussed in greater detail further on in thepaper as it gives rise to one of the most severe potential damagemechanisms in low penneability gas reservoirs, that of fluidretention or phase trapping.

Formation Damage Mechanisms in Low PermeabilityGas Reservoirs During Drilling, Completion, andProduction OperationsFonnation damage is an expansive topic which has beendiscussed in detail by many authors in the literature (Ref. 6-10).In this paper, primary attention is given to mechanisms offonnation damage which often tend to be the most prevalentcauses of reduced productivity in low penneability gasreservoirs. These damage mechanisms predominantly fall intothe following three major classifications:- Mechanical fonnation damage mechanisms

- Chemical fonnation damage mechanisms- Biological formation damage mechanisms

Mechanical Fonnation Damage MechanismsThe following problems tend to be the most significant for lowpermeability gas reservoirs.

Fines Mieration. This refers to the motion of naturally

D.B.BENNION, F.B.T}K)MAS, T.M.-. SPE 60325

even relatively small increases in trapped liquid saturation startto impinge upon the ability of gas to flow within the tight matrixand may result in only a modest increase in initial watersaturation causing a large reduction in effective penneability.

Hydrocarbon phase traps may also be problematic in lowpermeability gas reservoirs. In many situations, no pre-existingliquid hydrocarbon saturation exists in these sediments, and thehydrocarbon fluid is introduced either by the direct displacementof an oil-based completion/stimulation fluid into the reservoir,or deposited directly in place in the matrix due to retrogradecondensate dropout effects when a rich gas is produced at aflowing bottomhole pressure less than the dewpoint pressure ofthe gas. Hydrocarbon phase traps may once again severelyreduce the permeability to gas in the pore system, depending onthe pore geometry and the wetting characteristics of the gasunder consideration. Careful evaluation of the retrogradecharacter of rich gases should be made during the screeningprocess when examining the viability of a low permeability gasreservoir. The presence of a retrograde dewpoint system does notnecessarily preclude the economic production of gas from a lowpermeability reservoir, but will generally degrade theperfonnance character and may require some alterations to thecompletion and production practices required in order to obtainoptimum recovery. More detailed descriptions of the mechanismof phase trapping in low permeability porous media can be foundin the literature (Ref. S).

Glazing and Mashing. Refers to mechanical damageinduced by the drill bit or rotating drill string and represents alayer of very thin damage immediately at the wellbore-formationinterface. This damage, once again, is of very limited extent andwould not be a concern except in an un stimulated open holecompletion (which, as previously discussed, would be extremelyrare in a low permeability gas reservoir application).

Chemical Formation Damage MechanismsAdverse Clay Interactions. Certain clay structures may eitherbe susceptible to hydration by fresh or low salinity water contact(such as smectite or mixed layer clays), or deflocculation ordispersion (kaolinite and other fmes) caused by abrupt salinitytransitions or caustic pH. In many cases, low penneability gasreservoirs are characterized by compacted, cemented, fmegrained matrix and often contain little or no reactive/uncementedclay. Exceptions include dirty sands where the low penneabilityis motivated by a high concentration of pore filling andoccluding clay. Inhibitive water-based or oil- based systems orpure gas systems are sometimes considered for drilling andcompletion options in these circumstances.

Various Precipitates and Solids. Compatt"bility of inb'oducedfluids with in-situ formation fluids is always of importance in aneffective drilling, completion and stimulation program. Standardcompatibility protocols for water-water compatt"bility, scaling,and water-oil emulsion tests have been defined by the API. Ingeneral, due to the small volume of initial fluid in place in most

is one of the most severe that often plagues the success of a lowpenneability gas reservoir production operation. Since manyreservoirs of this type fall into the classification ofsubnonnallysaturated or desiccated reservoirs, (as discussed in the precedingsections), there is a tremendous amount of potential capillarypressure energy (capillary suction as it is sometimes referred to)which wants to imbibe and hold a fluid saturation in the porousmedia. Phase trapping effects can occur in gas reservoirs forboth water and hydrocarbon based fluids. In most cases, wateris the wetting phase, which tends to reduce or eliminate theaffinity for spontaneous imbibition of a hydrocarbon basedliquid phase into the mabix surrounding the wellbore (althoughthis fluid may still be displaced into the matrix of the rock byoverbalanced drilling and completion procedures, or depositedin place by the depressurization of rich retrograde condensatetype reservoir gases).

The basic mechanism of a phase trap in a low penneabilitygas saturated matrix is illustrated in Fig. 8. It can be observedthat the pore system of the reservoir is initially at a low liquidsaturation which provides the maximum cross sectional area forflow in the pore system, and therefore the highest level ofpenneability. If a water-based fluid is introduced into the system(middle frame in Fig. 8), we can see that a high water saturationin the flushed zone is generated and results in some trapped gassaturation. Upon reversal of flow to clean up the well,insufficient capillary drawdown gradient is present to overcomethe capillary pressure effects (Fig. 9) which results in a muchhigher trapped liquid saturation being obtained in the porousmedia. The configuration of the gas-water relative penneabilitycurves for the porous media will detennine the amount ofreduction in permeability associated with this increasedsaturation. Figures 10 and II illustrate favorable andunfavorable rock geometries for phase trapping issues withwater. In Fig. 10, it is observed that the pore system containssubstantial microporosity and that the majority of the effectivepermeability in this media is contained in a relatively smallfraction of the pore space which consists of interconnected mesoor macropores or small fractures. In this case, the naturalcapillary imbibition will draw invaded water into the tightestportions of the pore system on a selective basis. Although theseportions of the reservoir can be saturated with water, effectivegas permeability may not be significantly altered as the occludedportion of the pore system represents solely ineffective porosity.It is only when the trapped water saturation increases to the pointwhere it is sufficiently large that it begins to encroach into themeso/macropores and significant reductions in permeability togas are observed Therefore, rocks of this pore geometry may besignificantly less sensitive to water-based phase trapping. assignificant increases in the initial "trapped" water saturation canbe tolerated without severe accompanying reductions in

permeability.In sharp contrast to this would be a pore system as illustrated

in Fig. 11 which is dominated by micropores and a morerandomized and unifonn pore size distribution. In this situation,

FORMATION DAMAGE PROCESSES REDUCING PRODUCTMTY OF LaN PERMEABILITY GAS RESERVOIRSSPE 60325 5

viable low permeability gas reservoirs, fluid-fluid compatloilityissues do not tend to be one of the dominant issues of concern.

Biological Damage MechanismsThese are associated with the introduction of viable bacteria tothe fonnation which may colonize and propagate in either anaerobic or anaerobic fashion. This may create issues withcorrosion, souring of reservoir fluids (sulphate-reducingbacteria), or production of viscous polysaccharide bio-slimeswhich may occlude perDleabiJity. Most bacteria cannot thrive attemperatures above lOO-IIO°C which, in deeper tight gasforDlations, may reduce the severity of this problem. Relativelylow water saturations may also hinder the ability of the bacteriato colonize and propagate; however, if large volumes of water-based fluids are lost to the forDlation during drilling andcompletion and are pennanently retained, this may serve as abreeding ground for viable bacterial populations. More details onbacterial forDlation damage are provided in the literature (Ref.

10).

Proper Evaluation of Initial Permeability. Routine orconventional core analysis is a valuable comparative tool forcontrasting pay zone quality on a sample to sample basis withina given zone. However, it can be misleading when trying to inferactual in-situ permeability. Permeability data presented instandard core reports have generally not been corrected foroverburden compression effects (most analysis are conducted atonly a nominal confining pressure of 1380 kPa [200 psi] unlessodlerwise specified), surface slippage [Klinkenberg effects], clayand mineral hydration effects and relative permeability effectsassociated with the presence of a trapped water/oil saturation inthe reservoir in comparison to the clean, dry core used for theroutine core analysis. This can result, particularly in low qualitymabix, in the observed routine core analysis gas permeabilitiesbeing orders of magnitude higher than the actual effective in-situpermeability to gas (even in totally undamaged conditions). In-situ permeability measurements from pressure transientlbuilduptests (in homogeneous unfractured formations), or reservoircondition in-situ permeability measurements on properly handledand restored and stressed core samples are generally the bestindication that sufficient undamaged 'mabix' permeability ispresent to facilitate economic gas production rates.

One of the primary mechanisms for d1e establishment of dielow initial water saturation which exists in many lowpermeabilty gas reservoirs, as previously discussed, is associatedwith long term gas migration and gradual evaporation effectsthat have reduced the initial water saturation in place in thereservoir. This process can be seen to have a concentration effecton the dissolved solids present in solution in the remainingtrapped water saturation. For example, assume deposition of diesediments in an initial marine seawater environment with asalinity of approx. 30,000 ppm. If the reservoir is subsequentlyisolated from free water contact by tectonic or erosional eventsand desiccation occurs, reducing the water saturation from anaverage initial value of 50% to 10%, one can see that this willresult in a concentration of the soluble salts in the brine into theremaining water saturation (as they will not be extracted into diegas phase with the evaporating water), and that the effectivesalinity of the remaining brine saturation will increase to150,000 ppm. This obviously causes a significant reduction inthe apparent resistivity of the formation water.

However, if a regional water salinity, or a salinity based onthe composition of fresh water of condensation produced at thewells is used, the resistivity of this brine will be much higherthan is actually present in the reservoir resulting in the predictionof a much higher water saturation than is, in reality, present inthe reservoir (using conventional log analysis parameters). Thismay result in a potentially productive zone being diagnosed aswet and may result in ubypassed" pay as these zones may not bedeemed worthy of completion, when, if near wellbore drillinginduced damage is penetrated, they may be capable of economicproduction rates.

10 the absence of good Rw data and log calibrationparameters, direct measurement of the in-situ water saturation issometimes attempted. The most common technique used in thisarea is evaluation from core samples which have been drilledwith an pure oil-based system or with some type of b'aced waterbased system so that filtrate flushing effects can be backed outof the measured water saturations. Pure gas or air as coringfluids (due to heat and dessication effects) and invert emulsiondrilling muds (due to wettability alterations caused by flusheddrilling mud filtrate which may lower the measured watersaturation), are not recommended for this procedure.

Proper Evaluation oflnitialJ1uid Saturations. Evaluation ofinitial water saturation in a low penneability gas reservoir isdifficult due to the fact that the composition of the fonnationwater in the reservoir and hence, the resistivity of the formationwater for water saturation calculation from induction logs, isoften unknown. Due to the very low initial water saturationpresent in the reservoir, mobile formation water is rarelyproduced from viable low penneability gas reservoirs. The littlewater that is produced nom the wells is generally fresh water ofcondensation from the gas, and is not reflective of the true in-situ composition of the water contained in the reservoir matrix.

Formation Damage Issues: How Much Impact Do TheyReally Have?Formation damage during completion operations has alreadybeen discussed as a potential source of reduced productivity ofa low penneability gas reservoir. Analysis of typical fracturetreatments indicates that large treatments can sustain substantialreductions in permeability on the fracture face before causingappreciable reductions in apparent productivity of the fracturetreatment. This is due to the very large surface area which isgenerated with a well designed and executed fracture treatment.In many cases, the ultimate productivity of d1e fracture treabnentis controlled by the ability of the generated fracture to conduct

D.B.BENNION. F .B.1HOMAS, T.MA SPE603258

required for the first point of gas mobilization). Table 1provides a sample set of typical permeability measurements fora low penneability gas reservoir using a water-based 3% KCIcompletion fluid. The data has been plotted and appears as Fig.13.

gas back to the wellbore, rather than the ability of the fonnationto conduct gas to the fracture.

This stated. some obvious factors still emerge:I. The larger the generated fracture treatment, the less

significant the damage effects on the fracture faces tend tobecome and the more important it is to maintain effectivefracture conductivity. Smaller fracture treatments (i.e. less than20 tons), become increasingly sensitive to fracture face damageeffects as the ratio of inflow area to fracture conductivity areabecomes smaller and therefore impacts ultimate flow capacity toa larger extent.

2. The lower the effective "undamaged" permeability of thefonnation, the more significant the damage effects. Thesuppressing effect of damage on a fracture treatment is reliant onthe ability of the fracture face to supply more gas to the fracturethan it can effectively conduct to the wellbore under a given setof draw down conditions. Iffracture face inflow becomes limitedby the ability of the formation face to provide fluids to thefracture, rather than by the fractures ability to conduct the fluids,then reductions in productivity will be more significant withincreasing damage to the fracture face.

3. Even though the fracture area may be large, if a total 1000/0reduction in fracture face permeability occurs during completion,due to a damage mechanism such as phase trapping, (which isone of the few damage mechanisms that is capable of causingdamage to this extent in very low permeability reservoirmatrices), it can be seen that no matter bow large the effectivefrac area, inflow will be compromised and the frac may beineffective. Therefore, the topic of formation damage andappropriate selection of the proper fracture fluids and technologyis of significant import in low permeability gas reservoirsituations. These issues are pictorially illustrated as Fig. 12.

Imbibition Issues. Fig. 14 provides an illustration of a set ofwater-gas capillary pressure curves for a typical lowpermeability gas reservoir. These curves are typified by a high"threshold pressure" for initial gas intrusion into the watersaturated matrix and a high irreducible water saturation. If areservoir which exhibits this pore geometry is in a subirreducibleinitial water saturation condition, as illustrated by point "A" inFig. 14, one can see that there is a large amount of 'potential'capillary energy that exists between the natural equilibriumwater saturation desired to be present in the rock from a capillarymechanics point of view in comparison to the present watersaturation level. In these situations this creates an extremelypowerful hydrophilic suction tendency for water (assumed to bethe wetting fluid in this example) into the matrix. This meansthat considerable invasion, due to capillary suction effects, canoccur when water based fluids are in contact with the formation,even in the absence of a significant overbalance pressure. Aphenomena known as countercurrent capillary imbibition hasbeen well documented in the literature in previous papers andstudies by the authors (Ref. 16) and illustrates how a significantincrease in water saturation in the near wellbore or fracture faceregion can occur in such a situation, even if underbalancedoperations are being used when water based fluids (includingfoams), are circulated in contact with the formation face.

Therefore, one can see that proper evaluation of the initialfluid saturation values, capillary pressure and imbibitioncharacter, as well as the phase trapping characteristics of theporous media, determined by the specific pore geometry andrelative permeability characteristics under consideration, must allbe carefully considered in concert to determine the sensitivity ofa reservoir to phase trapping concerns.

Evaluating If Phase Trapping b a Problem. A number ofdifferent techniques are available for evaluation if phasetrapping of water or hydrocarbon based fluids may be an issuefor a low penneability gas reservoir. If core material is available,a simple procedure known as a "phase trapping test" is oftenconducted on samples of representative average to better qualitypay at downhole conditions to evaluate the sensitivity of themabix to both water and hydrocarbon based fluids and ascertainthe best techniques for drilling and completion of a given well.

A phase trap test (Ref. 11) uses actual core material restoredto proper initial saturation conditions. A baseline series ofpermeability measurements is conducted over the range ofexpected drawdown pressures so that turbulent flow effectswhich may reduce the permeability at high draw down rates canbe accurately distinguished separately from permanent damageeffects. This is followed by invasion of the sample with aspecified volume of potentially damaging water or oil basedfiltrate. This procedure is then followed by a series of regainpenneability measurements, conducted at the same drawdownlevels as the baseline pre-exposure penneabilities, and thedamage effect and threshold pressure (drawdown pressure

Mitigating Phase Trapping Issues in Low Permeability GasReservoirs. Fluid retention may be one of the greatest barriersto the production of low permeability gas reservoirs. Once theoperator understands the potential severity of this issue withrespect to their particular formation, a number of differentapproaches can be taken to minimize its impact.

Avoid Using Trapping Fluidr. If a severe sensitivity to waterbased phase trapping is present. consideration may be given tousing different fluid bases which have less trapping andimbibition affmity. Oil based fluids may be considered in somesituations for low penn water wet gas reservoirs. As the nonwetting fluid in this case, there will be no spontaneous capillaryimbibition effect to draw these fluids into the reservoir matrix(although they may still be displaced into the reservoir rock dueto overbalanced pressure operations). As the non-wetting fluid.if oil-based fluids are lost to the reservoir, they tend to be

SPE 60325 FORMATION DAMAGE PROCESSES REDUCING PROOUCTMTY OF LOW PERMEABIUTY GAS RESERVOIRS

of the water saturation profile as illustrated in Fig. 17. As longas a recharge source of unbound water is removed from thewellbore or fracture, this will obviously result in a gradualreduction in the value of the trapped water saturation in the nearwellbore or fracture face region, which may result in a slow longterm improvement in the permeability to gas in the region whichpreviously exhibited near zero gas permeability. In many cases,a period of static shut in may substantially improve theproductivity of a 'damaged' gas well. Flowing the well may becounterproductive in such a situation, as this imposes a pressuregradient which opposes the capillary imbibition force which mayslow or negate its effect. Also, high drawdown pressure mayresult in additional water of condensation precipitating from thegas in the high drawdown region adjacent to the wellbore orfracture face, which may add to the already existing trappedwater saturation and exacerbate the trapping effect.

Experimental studies of the imbibition process indicate thatthe process seems to work the best in permeabilities which aresomewhat greater than typically observed in low permeabilitygas reservoirs which are dle subject of this paper. In-situ permsin the range of 1-2 mD seem to be the most suited to thistechnique, as the imbibition effect and improvement in gas permcan be significant in relatively short period of time (few days tofew months). As permeability is lowered. although the force ofcapillary pressure increases, the relative speed of imbibitionslows and many years may be required before a substantial effectis observed. which often is not a viable economic option for astimulation treabnent.

trapped in the cenb"al portion of the pore space, rather thanadhering tightly to the matrix walls as the wetting phase.Although this cenb"al pore space occlusion can cause substantialreductions in permeability, in some cases, the total reducingeffect is less than if a water based system had been used in thesame circumstances, as the total volume of pore system occludedis less. This phenomena is schematically illuSb"ated as Fig. IS.A series of phase trapping tests on representative core materialusually can accurately defme for a given reservoir if the use ofoil based fluids may be advantageous over water based systemsin the same situation. Another alternative, if both water andhydrocarbon based systems appear highly damaging from aretention point of view, is to consider the use of highly energiz.ed(N2 or COJ system, to increase near wellbore charge energy,reduce 1FT and reduce the volume of trapping fluid invading therock, or pure gas based systems such as pure CO2 or N/CO2"foam".

Reduce Exposure Time and Depth of Invasion and 1FT. Inmany circumstances where large fracture treatments must bepropagated, or if reservoir temperatures are very high (or both),current technology is highly weighted towards the use of waterbased heavy metal cross linked gels. In a situation where it isexpected that some degree of water invasion will occur, aprophylactic approach must be taken to:

I. Design fluid rheology and breaker properties such that theminimum amount of leakotI of water based filtrate over theperiod of high overbalance exposure occurs.

2. That the fluids are removed from formation contact byrapid backflow and cleanup as quickly as possible to reduceStatic or dynamic capillary countercurrent imbibition effects.

3. Capillary pressure, which is the dominant variablecontrolling fluid retention, is a direct linear function ofinterfacial tension between the water and gas phase. If thisinterfacial tension can be reduced between the invading waterbased filtrate and the in-situ reservoir gas, the magnitude of thecapillary pressure and the degree of observed fluid retention mayalso be lessened. Common 1FT reducing additives to water basedfluids would include various volatile alcohols (methanol beingthe most prevalent), surfactants and liquid or gaseous carbondioxide. Care must be taken with the use of alcohols and CO2when liquid hydrocarbons may be present as a trapped or mobilesaturation in the reservoir, as problems with deasphalting (CO2)or sludging (MeOH) may occur.

Hot and Conventional Dry Gas Injection. The imbibitionprocess may be hastened in some reservoir situations wherewater based phase trapping is known to be a source ofsubstantial reduced productivity by the injection of dehydrateddry gas into the formation. Due to the suppressed dewpoint ofthis gas, it may have exb'active ability to remove b'apped waterfrom the near wellbore or frac face region by dessication effects.The objective of this treatment is to create some paths of lowwater saturation back to the undamaged portion of the reservoirto initiate conduits of higher gas permeability through thedamaged zone. The speed of the process may be hastened insome situations through the application of heat to speed theeXb'active process. The use of downhole beating tools for thispurpose has been documented in the literature (Ref 13, 14).Onemust be careful that the total dissolved solids content of theb'apped brine is not too great if such technology is contemplated,as the desiccation of the in-situ water may resuh in rapid supersaturation of the remaining water saturation with dissolved ionsand the subsequent precipitation of halite or other crystallinesolids within the pore system which may have an equallydamaging effect to the initial water b'ap which was present. Ingeneral, for low permeability porous media, if the value of theillS in the trapped water saturation exceeds 50,000 ppm, thismethod should not be considered for use.

Static Exposure Time. Fig. 16 illusb'ates the situation where avolume of water-based filtrate invades the region surrounding awellbore or fracture face in a low permeability gas reservoir. Inthis situation we can see that initially the trapped watersaturation is high enough to almost totally occlude thepenneability to gas in the near wellbore/frac face region. Natureabhors steep capillary gradients however, and we can see thatnatural capillary imbibition will want to 'wick' or imbibe waterfrom the high water saturation zone (encompassing the originalinvaded area) deeper into the formation, resulting in a 'smearing'

D.B.BENNION, F.B.THOMAS. T.MA SPE 603258

4. A variety of techniques have been presented for bod1evaluating initial fonnation penneability and fluid saturations, aswe" as for detemlining and reducing the impact of fonnationdamage in various reservoir scenarios.

NomenclaturePc = Capillary pressureP - = Pressure in non wetting phase (gas)P w = Pressure in wetting phase (water/liquid hydrocarbon)9 = Contact angle between gas and liquid in rock surfaceR = Pore throat radius(J = Interfacial tension between gas and waterg = Gravitational constanth = Height above free water contact!J.p = Density difference between gas a liquidR1 = Primary radius of curvature of gas-liquid interfaceR2 S Secondary radius of curvature of gas-liquid interface

Use o( IFf Reducing Solvents (or Water and Oil RetentionProblems. The role of 1FT on capillary pressure has alreadybeen discussed. As a post damage stimulation treatment. the useof various 1FT reducing solvents has long been recognized as apotential means for removing trapped fluid and improving theproductivity of damaged low pemteability gas reservoirs. Onceagain, for water based fluids, 1FT reducers such as varioussoluble volatile alcohols (methanol being the most common fordry gases with no associated hydrocarbon liquids in thefonnation), or liquid carbon dioxide are the most common. Fortrapped hydrocarbons, depending on the reservoir conditions andthe composition of the trapped hydrocarbon phase, various typesof miscible solvents such as carbon dioxide and liquid ethane orpropane or lean gas (dry methane) when reservoir pressure issufficiently high, may be contemplated for use. A wide range oflab tests can be used to evaluate the effectiveness of thesetreatments to determine the technology best suited forapplication in a given set of circumstances (Ref. 11).

Inhibitive Base Fluidf. If studies suggest concentrations ofreactive clays are present. or if fluid-fluid interactions areindicated, a variety of inhibitive based fluids can be evaluatedusing both computer models and special core analysis techniquesto select the fluid base most appropriate for use.

AcknowledgmentsThe authors wish to express appreciation to Vivian Whiting forher assistance in the preparation of the manuscript and the

figures.

References1. Masters, J.A., "Elrnworth - Case Study ofa Deep Basin Gas

Reservoir", AAPG Memoir 38, 1984.2. Katz, D.L., et ai, "'Absence of Connate Water in Michigan

Reef Gas Reservoirs - An Analysis", AAPG Bulletin, Vol

66, No.1, January, 1982, pp 91-98.3. Bennion D.B., Cimolai, M.P., Bietz, R.F., and Thomas,

F .B., "Reductions in the Productivity of Oil and GasReservoirs Due to Aqoues Phase Trapping", JCPT ,November, 1994.

4. Bennion, D.B., et ai, "Water and Hydrocarbon PhaseTrapping in Porous Media, Diagnosis,. Prevention andTreatment", CIM Paper 95-69, 461h Petroleum SocietyATM, Banff, Canada, May 14-17, 1995.

5. Bennion, D.B., et ai, "Low Permeability Gas Reservoirs:Problems, Opportunities and Solutions for Drilling,Completion, Stimulation and Production", SPE 35577,Presented at the Gas Technology Conference, Calgary,Canada, April 28 - May I, 1996.

6. Muecke, T. W ., "Formation Fines and Factors ControllingTheir Movement in Porous Media", JPT, Feb, 1979 (SPE

7007).7. Francis, P., "Dominating Effects Controlling the Extent of

Drilling Fluid Induced Damage", SPE 38182, EuropeanFormation Damage Conference, The Hagye, Netherlands,

June2-3,1997.8. Byrom, T.G., et ai, "Some Mechanical Aspects of

Formation Damage and RemovaI in Horizontal Wells", SPE312145, SPE International Symposium on FormationDamage Control, Lafayette, La., Feb 14-16, 1996.

9. Monaghan, G.A., et aI, "Laboratory Studies of Formation

~~

ConclusionsThis paper has discussed the criteria for selection andcompletion of viable low penneability gas reservoirs. Thisincludes:

1. Evaluating if in-situ penneability to gas in an undamagedstate is sufficient for gas production under the current reservoirpressure. Typically cutoffs for higher pressure formations tendto be in the 5-10 micro Darcy range (0.005-0.010 roD) and willbe greater if subnormally pressured or pressure depleted orshallower, lower pressure reservoirs are under consideration.

2. For a very low penneability «0.10 roD) gas reservoir torepresent potential productive pay in general, we find that thereservoir matrix is undersaturated or desiccated with respect tothe initial water saturation. This issue is believed to be due tolong tenD regional gas migration and evaporative effects andcontributes to the increased sensitivity of low permeability gasproducing formations to fluid retention (phase trapping) as amajor damage mechanism occurring during completion

operations.3. Low penneability gas reservoirs can be subject to a

number of different damage mechanisms during drilling,completion and production. In general, due to the character ofthe formations, fluid retention may often be a significant damagemechanism. The impact of fluid retention and trapping arestrongly affected by the interaction of pore geometry, wettability,invasion depth. drawdowo pressure. capillary pressure andrelative penneability and are unique for every reservoir situation.Not all low penn reservoirs may be adversely impacted by phasetrapping effects, although it has been documented to be asignificant damage mechanism in many cases.

SPE 60325 FORMA noN DAMAGE PROCESSES REDUCING PRODUCTMTY OF LOW PERMEABILITY GAS RESERVOIRS 9

Damage in Sands Containing Clays", Transactions ofAIMME, I I 62-G, 1959.

10. Bennion. D.B., "Fonnation Damage - The Impainnent of

the Invisible, by the Inevitable and Uncontrollable,Resulting in an Indeterminate Reduction of theUnquantifiable", JCPT Distinguished Authors Series, JCPT ,Vol 38, No.2, 1999, pp. II.

11. Bennion, D.B., et ai, "Recent Advances in Laboratory TestPnXocols for Evaluating Optimum Drilling, Completion andStimulation Practices for Low Penneability GasReservoirs", Paper SPE 60324 Presented at the 2000 SPERocky Mountain Regional MeetingiLow PenneabilityReservoirs - March 12-15, 2000, Denver, Colorado.

12. Bennion D.B., et ai, "Underba1anced Drilling of HorizontalWells - Does it Really Eliminate Formation Damage", SPE

27352, Presented at the 1994 Formation DamageSymposium, Lafayette, LA., Feb 9-10,1994.

13. Jamaluddin, A.K.M., et ai, "Process for Increasing NearWellbore Permeability of Porous Formations", US patent5,361,845 - 1994.

]4. Jamaluddin, A.K.M., et ai, "Heat Treabnent for ClayRelated Near Wellbore Damage", JCPT, Vol 37, No.1,1998.

Table 1. Typical Regain Permeability Data for Low Permeability Sandstone Reservoirs

Figure 1 - Typical Interfacial Radii ofCurvature in Porous Media

Figure 2 - Typical Radii of Curvature in LowPermeability Porous Media

Figure 3 - Typical Radii of Curvature in HighPermeability Porous Media

Figure 4 - Typical Capillary Pressure Type Curves forDifferent Permeability Ranges of Typicallntercrystalline- - a' ,.

Figure 5 - Illustration of Reduction in Gas PhasePermeability as Trapped Water Saturation Increases -Pore Scale

Initial Swi High - ReducedGas Reserves and InitialPermeability to Gas

Initial Swi Low - IncreasedGas Reserves and InitialPenneability to Gas

Figure 6 - Reduction in Effective Permeability to Gas asSwi Increases - Relative Permeability Basis

Water Saturation - Fraction

Figure 7 - Typical "Irreducible" Normal WaterSaturations for Various "Air" Permeability Ranges

Figure 8 - Basic Phase Trapping Mechanism for a

li

lI

ISw

Low Permeability Gas ReservoirInlaaJ Conditione

Figure 9 - Illustration of Capillary Pressure Effects onPhase Trapping

After Cl88nupInitial Condition. After we FiIll8t8 Invuion

~~fDfD~c..

2:'~'0.to

U

s~w

Figure 10 - Favorable Relative Permeability Curves/RockGeometry Which Minimizes Phase Trapping Concerns

i0..G)i~

~G)

a..

.Jm

~

Sw Sw

Figure 11 - Unfavorable Relative Permeability andRock Geometry for Phase Trapping

-

Sw

Figure 12 -Issues Affecting Inflow/Damage inHydraulically Fractured Wells

Figure 13 - Typical Baseline and Regain PermeabilityCurves for a Low Permeability Gas Reservoir Exposedto 3% KCL Filtrate

0.14

0.12

0.02

07 35 70 140 350 1380 3500 7000 14000 28000

Drawdown Pressure Applied. kPa

Figure 15 - Comparison of Water Based vs Hydrocarbon

-.-0~~OIB888dF81818

~~~CI88IUpF8r8

OIfg8I8ICordIOIw

Figure 16 - Illustration of Water Invasion in a Water WetLow Perm Reservoir - Prior to Time Induced Dispersion

Figure 17 - Illustration of Water Invasion in a Water WetLow Perm Reservoir -After Time Induced Dispersion